Pediatric sepsis: Time is of the essence*

Editorials The Nina, the Pinta, and heart rate variability: The search for prognostic indicators after cardiac arrest* M ore than 500 years ago, Christopher Columbus set sail west across the Atlantic searching for better trade routes with East Asia. Although Columbus and his team never found the route to the Orient, they opened an age of European exploration of the Americas and popularized use of the prevailing trade winds for transcontinental travel (1). In this issue of Critical Care Medicine, Tiainen and colleagues (2) set out to study heart rate variability as a potential indicator of prognosis after cardiac arrest. Although one parameter of heart rate variability at 48 hours had prognostic implications for patients treated with therapeutic hypothermia, this report stands out as the first to carefully study electrocardiographic data during therapeutic hypothermia in patients who had already experienced a cardiac arrest. As with Columbus, these secondary discoveries may be the results better remembered. Therapeutic hypothermia has evolved from a novel intervention to a standard treatment for cardiac arrest survivors with encephalopathy (3– 6). Although most deaths are due to hypoxic-ischemic brain injury and not recurrent cardiovascular events (7, 8), concerns regarding the safety or potential adverse events associated with hypothermia may contribute to clinician reluctance to implement this important therapy (9, 10). Two recent reviews guiding efforts to predict outcome after cardiac arrest do not include data from patients treated with hypothermia (11, 12), but do note the potential confounding effect from the 24 hours of hypothermia and the increased use of sedation and neuromuscular *See also p. 403. Key Words: hypothermia; induced/adverse effects; risk assessment/methods; arrhythmias, cardiac/ adverse effects; predictive value of tests The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e3181959abc Crit Care Med 2009 Vol. 37, No. 2 blockade associated with it (13). Tiainen and colleagues provide the best existing data regarding adverse electrocardiographic events during hypothermia and describe a tool that may improve prognostic efforts. As part of the Hypothermia after Cardiac Arrest study, 70 consecutive adult patients resuscitated from out-of-hospital ventricular fibrillation at Helsinki University Hospital were randomly assigned either to therapeutic hypothermia or to normothermia (2). Twenty-four hour electrocardiographic recordings were performed on days 1, 2, and 14, and frequency and time domain assessments of heart rate variability were performed. Premature ventricular beats were increased in the hypothermia-treated group during the first two recordings, but no difference was noted in the occurrence of ventricular tachycardia or fibrillation for patients treated with hypothermia or normothermia. Defibrillation was required by only one patient during hypothermia (compared with two patients in the normothermic group). Two of 36 patients had mean heart rates ⬍50 during hypothermia, and one of these received a temporary pacemaker that fired for ⬍5 minutes. In multivariate analysis, preserved heart rate variability seemed to predict a favorable outcome. By careful documentation of electrocardiographic activity among cardiac arrest survivors treated with hypothermia and normothermia, Tiainen et al have added to previous work in neuroprognostication after cardiac arrest (14, 15). Of perhaps greater applicability, their study expands our understanding of the safety of hypothermia among patients with complex cardiac conditions. Therapeutic hypothermia was recently shown to be safe and beneficial in patients with cardiogenic shock treated with intra-aortic balloon pumps and in conjunction with percutaneous coronary intervention (16, 17). The present study confirms that ventricular arrhythmias are common, but not more prevalent in patients receiving hypothermia, and that bradycardia is common but rarely requires treatment. These findings increase our understanding and are based on data from more reliable monitoring than in existing reports. Just as Europeans provided information facilitating further exploration of the “new world” 500 years ago, Tiainen’s group provides data that advances our understanding of therapeutic hypothermia. We can only hope that many other investigators will travel these waters to discover better and safer ways of providing this important therapy and to furnish us with tools to better predict outcomes. Richard R. Riker, MD, FCCP David B. Seder, MD Gilles L. Fraser, PharmD, FCCM Neurosciences Institute Maine Medical Center Portland, Maine REFERENCES 1. Available at: http://earthguide.ucsd.edu/ virtualmuseum/climatechange1/08_1.shtml. Accessed August 6, 2008 2. Tiainen M, Parikka HJ, Mäkijärvi MA, et al: Arrhythmias and heart rate variability during and after therapeutic hypothermia for cardiac arrest. Crit Care Med 2009: 37:403-409 3. Bernard SA, Gray TW, Buist MD, et al: Treatment of comatose survivors of out-ofhospital cardiac arrest with induced hypothermia. N Engl J Med 2002; 346:557–563 4. HACA Study Group: Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002; 346: 549 –556 5. Anonymous: American Heart Association 2005 guidelines for CPR and ECC. Part 7.5. Postresuscitation support. Circulation 2005; 112 (Suppl 24):IV-84 –IV-88 6. International Liaison Committee on Resuscitation: International consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation 2005; 112:III1–III-136 7. Oddo M, Ribordy V, Feihl F, et al: Early predictors of outcome in comatose survivors of ventricular fibrillation and non-ventricular fibrillation cardiac arrest treated with hypothermia: A prospective study. Crit Care Med 2008; 36:2296 –2301 735 8. Seder DB, Jarrah S, Lord C, et al: Effect of a therapeutic hypothermia protocol on mortality in out-of hospital cardiac arrest: Experiences at a tertiary care hospital. Proc ATS 2007; 4:A792 9. Holzer M, Bernard SA, Hachimi-Idrissi S, et al: Hypothermia for neuroprotection after cardiac arrest: Systematic review and individual patient data meta-analysis. Crit Care Med 2005; 33:414 – 8 10. Brooks SC, Morrison LJ: Implementation of therapeutic hypothermia guidelines for postcardiac arrest syndrome at a glacial pace: Seeking guidance from the knowledge translation literature. Resuscitation 2008; 77:286–292 11. Booth CM, Boone RH, Tomlinson G, et al: Is this patient dead, vegetative, or severely neu- rologically impaired? Assessing outcome for comatose survivors of cardiac arrest. JAMA 2004; 291:870 – 879 12. Wijdicks EFM, Hijdra A, Young GB, et al: Practice parameter: Prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review). Neurology 2006; 67:203–210 13. Sunde K, Dunlop O, Rostrup M, et al: Determination of prognosis after cardiac arrest may be more difficult after introduction of therapeutic hypothermia. Resuscitation 2006; 69:29 –32 14. Tiainen M, Roine RO, Pettilä V, et al: Serum neuron-specific enolase and S-100B protein in cardiac arrest patients treated with hypothermia. Stroke 2003; 34:2881–2886 15. Tiainen M, Kovala TT, Takkunen OS, et al: Somatosensory and brainstem auditory evoked potentials in cardiac arrest patients treated with hypothermia. Crit Care Med 2005; 33:1736 –1740 16. Wolfrum S, Pierau C, Radke PW, et al: Mild therapeutic hypothermia in patients after out-of-hospital cardiac arrest due to acute ST-segment elevation myocardial infarction undergoing immediate percutaneous coronary intervention. Crit Care Med 2008; 36: 1780 –1786 17. Oddo M, Schaller MD, Feihl F, et al: From evidence to clinical practice: Effective implementation of therapeutic hypothermia to improve patient outcome after cardiac arrest. Crit Care Med 2006; 34:1865–1873 Hemodynamic support of shock state: Are we asking the right questions?* S eptic shock is one of the most challenging problems in the critical care. Its mortality toll in the United States ranges between 200,000 and 250,000, a number comparable with myocardial infraction. Diagnosing and treating a septic shock is like looking the stars at night: from single bright spots, you have to reconstitute the constellations to have the complete picture. For the last 40 years, catecholamines have been used routinely in shock state, trying to restore normal or near-normal hemodynamic parameters. The rationale is to maintain a minimal level of blood pressure in septic shock patients (1). From a quasi-empirical use, more and more knowledge has emerged on the mechanisms and effect of these drugs. Our understanding of shock state improves, including myocardial depression in septic shock, the links between inflammation and coagulation, and microcirculation and cellular energetics. Our pharmacologic tool set expanded and include vasopressin and analogs, phosphodiesterase inhibitors, and cal- *See also p. 410. Key Words: septic shock; catecholamines; outcome; sepsis bundle; quality control The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e3181959abc 736 cium sensitizers. As knowledge expand, the picture gets more complex and the clinicians more confused. The Quest for the Magic Bullet In this issue of Critical Care Medicine, Povoa et al (2) adds to the comparison of different catecholamine in shock states. They observed a large multicentered population of septic shock patients, the various catecholamines used, and the patient’s outcome. They show that dopamine, norepinephrine, and dobutamine increase mortality. Should we ban these drugs from our pharmacopoeia? Certainly not. First, the natural catecholamines are part of the acute-phase response to physiologic stress and are essential for human species. These molecules are part of our survival kit (3). Second, using a similar strategy, others (4) observed that that dopamine decreases mortality in septic shock patients and that norepinephrine and dobutamine have no effect. This is contrary to the conclusion in the present study. Other cohorts have investigated this question with variable results (Table 1). One can argue that these are not randomized control trials, but most randomized control studies also failed to point out a consistent effect. Third, most of the studies do not start with patients on no adrenergic support: when a certain level of dopamine or norepinephrine is reached randomize into replace by or add “A” vs. “B.” First, the picture is blurred. Catecholamines at a usual pharmacologic range yield to concentrations 100 times above the physiologic concentrations. Second, there is a significant interindividual variability in catecholamine kinetics: a fixed dose of dopamine can yield to plasma concentrations in a 20-fold range. Third, dopamine is the natural direct precursor of norepinephrine through betahydroxylase. During dopamine infusion (3 ␮g/kg/min), plasma norepinephrine concentration increases (5). Should We Change Our Approach to Shock States Until recently, hemodynamic support of shock state was focused on restoring normal or near-normal physiologic parameters. In the 1980s, there was even a tendency toward supranormal physiologic goals. To easily achieve target hemodynamic parameters, having one single magical drug would make the clinician’s life easier. There is a quest toward the magic bullet applicable in all patients: Is dopamine better than norepinephrine? Is vasopressin better than norepinephrine? What is the optimal target for mean arterial blood pressure? What is the optimal cardiac output/mixed venous saturation? Consensus panel supported the use of catecholamines in septic shock with a grade E evidence (6). Clinical investigations suggest that increasing the target mean arterial pressure (MAP) from 65 to 85 mm Hg does not change oxygenCrit Care Med 2009 Vol. 37, No. 2 Table 1. Some studies investigating the effect of vasopressors in vasodilatory shock in adults Study Cohort studies Goncalves, et al (26) Patients Methods 406 patients requiring NE Main Conclusions Observational prospective cohort Observational prospective cohort Observational prospective cohort Observational multicenter study NE does not facilitate development of MODS NE lower mortality when compared to Dopa Fixed dose AVP is comparable with titrated doses of NE or Dopa Dopa tends to worsen mortality NE and Dobu has no effect 137 septic shock patients all receiving NE, some with AVP 458 septic shock patients Observational prospective cohort Patients receiving AVP plus NE have a worse mortality than patients receiving NE Dopa decreases mortality. NE and Dobu increases mortality Randomized control trials Martin, et al (30) 32 patients in septic shock NE vs. Dopa Ruokonen, et al (31) 10 patients in septic shock Dopa alone or with NE Marik, et al (32) 20 septic shock patients Dopa vs. NE Levy, et al (33) 30 patients with septic shock Epi vs. fixed Dobu plus NE Malay, et al (34) AVP vs. placebo plus NE Seguin, et al (35) 10 trauma patients with vasodilatory shock 22 patients in septic shock Patel, et al (36) 24 patients in septic shock Dunser, et al (37) 48 patients with vasodilatory shock Albanese, et al (38) 20 septic shock patients Lauzier, et al (39) 23 patients in septic shock First-line treatment: NE versus AVP alone, rescue by adding the other drug Schmoelz, et al (40) 61 patients with septic shock NE plus dopexamine or plus Dopa Annane, et al (41) 330 patients with septic shock NE plus Dobu vs. Epi Russell, et al (42) 778 patients in septic shock NE vs. AVP Martin, et al (27) 97 septic shock patients Hall, et al (28) 150 patients receiving either Dopa NE or AVP 462 septic shock patients Sakr, et al (4) Micek, et al (29) Povoa, et al (2) Observational multi center study Titrated Epi vs. fixed Dobu plus titrated NE AVP vs. NE fixed dose plus open label NE AVP 4 U/hr vs. placebo plus norEpi to MAP ⬎70 Terlipressin vs. NE NE provided a better hemodynamic profile, No difference in outcome Discrepancy between global and regional blood flow. No difference in outcome NE improve splanchnic perfusion. No difference in outcome Trend toward less oliguria with NE plus Dobu, No differences in mortality Trend toward less mortality in the AVP group No difference in outcome Better creatinine clearance with AVP, no difference in outcome Better hemodynamic response with AVP but no difference in outcome No differences in MAP, renal function and outcome No difference in outcome Comments Significant international variation in catecholamine use AVP was used as a rescue therapy in NE resistant patients Very small n, 24 hr follow-up 4 hr follow-up Open label. 85% of AVP patients received additional NE due to MAP ⬍70 mm Hg at 1 hr Dopexamine associated with better renal function. No difference in organ failures or mortality No difference in hemodynamic, organ failure, duration of therapy, and mortality Better survival with AVP in less sick patients. No overall mortality difference NE, norepinephrine; AVP, arginine-vasopressin; Dobu, dobutamine; Dopa, dopamine; Epi, epinephrine; norEpi, norepinephrine; MAP, mean arterial pressure; MODS, Multiple Organs Dysfunction Syndrome. ation parameter and skin microcirculation (7) no renal function or outcome (8). Our actual concept is that below a certain MAP, blood flow is linearly dependent on organ perfusion (1). This cutoff is probably not the same in the various organs and probably different between different various capillary beds of the same organ. Crit Care Med 2009 Vol. 37, No. 2 The inlet pressure of physiologic capillaries is 20 –25 mm Hg. There are no data to suggest that 50 mm Hg MAP is more deleterious in term of microcirculatory and organ perfusion than 65 with vasopressors. Sepsis induces a state of nutrient and oxygen deficiency at the cellular level. This cellular energetic failure lead to organ dysfunction, myocardial depression, and microcirculatory dysfunction (Fig. 1). The myocardium is energetically exhausted. Increasing the catecholamines level is like whipping an exhausted horse. With worsening cellular dysfunction, the horse will not respond the whip—the shock will became refractory to cat737 Figure 1. Proposed physiopathology of septic shock. Proposed physiopathologic links between sepsis, mitochondrial energetic failure, and organ dysfunction and death. Therapeutic target is shown in the gray boxes. Innovative therapies and new effects of therapies are shown with a question mark. iNO, inducible nitrous oxide. echolamines. We can change whip—as by using vasopressin or analogs— but the underlying problem will remain and, if not corrected, shock will worsen and be refractory to vasopressin. This may explain why “dopamine sensitive” patients in septic shock have a better outcome than “dopamine resistant” patients (9): In dopamine-resistant shock, the cellular energetic is failing. This may also explain that in the Vasopressin and Septic Shock Trial, the patients who may benefit from the change of whip—switch to vasopressin—are those requiring the lowest doses of norepinephrine—the not too exhausted horses. Researchers have started to explore the link between hemodynamic and microcirculation: Trzeciak et al (10) showed that there is a correlation between survival and microcirculatory disturbances. Sakr et al (11) showed that small vessel perfusion improved over time in septic shock survivors but not in nonsurvivors. Despite similar hemodynamic parameters and amount of support, patients dying after the resolution of shock in multiple organ failure had a lower percentage of perfused small vessels than did survivors. This shows that the cellular dysfunction and microcirculatory disturbances can still be present despite restored hemodynamics. Following these 738 hypothesis, innovative and provocative, strategies have been proposed in septic shock. Spronk et al (12) administered nitroglycerine in septic shock patients and showed an improvement in the microcirculation. De Backer et al (13) demonstrated that dobutamine can improve but not restore microcirculatory perfusion in patients with septic shock, independently of its inotropic effect. Going one step upstream (Fig. 1), Levy et al (14) measured intermediate substrate metabolism in septic shock patients and pointed out a mitochondrial dysfunction. Some have suggested that lactate is not a waste product but a highoctane fuel and an adaptive mechanism during shock states (15). The epinephrine-induced release of glucose and lactate from the muscles during shock may be a survival mechanism aiming to feed a starved horse (16). Recently, Regueira et al (17) showed that norepinephrine was able to increase maximal mitochondrial respiration in the liver during an endotoxin challenge. This suggests that in shock, the so-called cytopathic hypoxia is related to mitochondrial substrate availability (18). Obviously, this view of septic shock is simplistic: There are probably two types of feedback and control sys- tems: First, standard feedback such as improvement of hemodynamic parameter should be associated with improved microcirculation and cell oxygenation. Second, control of hemodynamic parameters may help to avoid vicious circles such as worsening of inflammatory response because of ongoing shock state (19). With this in mind (Fig. 1), hemodynamic support alone has little chances to reverse the complete cascade. It will be hard to find a link between the type of catecholamine and the outcome. This will also require that further randomized trial investigating the effect of hemodynamic support in septic shock patients will have to include a homogeneous patient group and control all these other therapies. Dobutamine and norepinephrine have recently shown unexpected positive side effects in shock state at the microcirculatory and mitochondrial level. Some intensivists dream about the multicentered international randomized control trial comparing norepinephrine vs. dopamine vs. vasopressin in septic shock patients: The design would be study drug “X” vs. “Y” in a concealed bag titrated to MAP of 70 mm Hg. What is the best in terms of MAP? What is the best in terms of outcome? Maybe we should change from “what is the best in term of MAP?” to “what is the best for the mitochondria?” (20). Sepsis Bundle Is a Whole Package: Take it as a Whole Following the fundamental study by Rivers (21), several consensus conferences proposed a structured multisystem approach to patients in septic shock (22). Items of a bundle are inseparable: It is not a flexible contract. The strength of these bundle and outcome improvement come from 1) a standardized approach (23), 2) the check list effect avoiding missed items, and 3) the speed of intervention including time points were goals should be met (6 and 24 hours). Compliance to the surviving sepsis campaign bundle in the SACiUCI study was also published as a preliminary abstract (24). The compliance to selected items of the 6-hour bundle ranges from 30% (antibiotics within 3 hours) to 80% (vasopressors after adequate fluid). Using an allor-none approach, ⬎70% of the patients failed the 6-hour bundle. Failing the 6-hour bundle is associated with an increased mortality in the present cohort (24). There are three strategies to assess Crit Care Med 2009 Vol. 37, No. 2 compliance to a set of interventions/ markers (25): 1) As Povoa et al presented as item-by-item measurement, 2) as a composite variable i.e., four items of six, or 3) using an all-or-none measurement. The outcome improvement does not come from a specific intervention or a specific catecholamine. It is time to raise the bar and assess the surviving sepsis campaign recommendation as an inseparable package. We should go further than the 6-hour limit: In parallel to data from myocardial infarction regarding door-toballoon time, we should focus on a door to sepsis bundle time. Despite an excellent worldwide campaign, endorsement by a dozen of critical care societies and organizations, practice has room for improvement. Sepsis is like myocardial infarction: its an emergency! David Bracco, MD Department of Anesthesia and Critical Care, McGill University Health Center—Anesthesia Montreal General Hospital Montreal, Quebec, Canada REFERENCES 1. Leone M, Martin C: Vasopressor use in septic shock: An update. Curr Opin Anaesthesiol 2008; 21:141–147 2. Póvoa PR, Carneiro AH, Ribeiro OS, et al: Influence of vasopressor agent in septic shock mortality. Results from the Portuguese community-acquired sepsis study (SACiUCI). Crit Care Med 2009; 37:410 – 416 3. Roizen MF: Should we all have a sympathectomy at birth? Or at least preoperatively? Anesthesiology 1988; 68:482– 484 4. Sakr Y, Reinhart K, Vincent JL, et al: Does dopamine administration in shock influence outcome? Results of the Sepsis Occurrence in Acutely Ill Patients (SOAP) Study. Crit Care Med 2006; 34:589 –597 5. Lieverse AG, Lefrandt JD, Girbes AR, et al: The effects of different doses of dopamine and domperidone on increases of plasma norepinephrine induced by cold pressor test in normal man. Hypertens Res 1995; 18(Suppl 1):S221–S224 6. Beale RJ, Hollenberg SM, Vincent JL, et al: Vasopressor and inotropic support in septic shock: An evidence-based review. Crit Care Med 2004; 32:S455–S465 7. LeDoux D, Astiz ME, Carpati CM, et al: Effects of perfusion pressure on tissue perfusion in septic shock. Crit Care Med 2000; 28:2729 –2732 8. Bourgoin A, Leone M, Delmas A, et al: Increasing mean arterial pressure in patients with septic shock: Effects on oxygen variables and renal function. Crit Care Med 2005; 33:780 –786 Crit Care Med 2009 Vol. 37, No. 2 9. Levy B, Dusang B, Annane D, et al: Cardiovascular response to dopamine and early prediction of outcome in septic shock: A prospective multiple-center study. Crit Care Med 2005; 33:2172–2177 10. Trzeciak S, Dellinger RP, Parrillo JE, et al: Early microcirculatory perfusion derangements in patients with severe sepsis and septic shock: Relationship to hemodynamics, oxygen transport, and survival. Ann Emerg Med 2007; 49:88 –98, 98.e1–2 11. Sakr Y, Dubois MJ, De Backer D, et al: Persistent microcirculatory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med 2004; 32:1825–1831 12. Spronk PE, Ince C, Gardien MJ, et al: Nitroglycerin in septic shock after intravascular volume resuscitation. Lancet 2002; 360: 1395–1396 13. De Backer D, Creteur J, Dubois MJ, et al: The effects of dobutamine on microcirculatory alterations in patients with septic shock are independent of its systemic effects. Crit Care Med 2006; 34:403– 408 14. Levy B, Sadoune LO, Gelot AM, et al: Evolution of lactate/pyruvate and arterial ketone body ratios in the early course of catecholamine-treated septic shock. Crit Care Med 2000; 28:114 –119 15. Matejovic M, Radermacher P, Fontaine E: Lactate in shock: A high-octane fuel for the heart? Intensive Care Med 2007; 33: 406 – 408 16. Levy B, Mansart A, Montemont C, et al: Myocardial lactate deprivation is associated with decreased cardiovascular performance, decreased myocardial energetics, and early death in endotoxic shock. Intensive Care Med 2007; 33:495–502 17. Regueira T, Banziger B, Djafarzadeh S, et al: Norepinephrine to increase blood pressure in endotoxaemic pigs is associated with improved hepatic mitochondrial respiration. Crit Care 2008; 12:R88 18. Leverve XM: Mitochondrial function and substrate availability. Crit Care Med 2007; 35:S454 –S460 19. Rivers EP, Kruse JA, Jacobsen G, et al: The influence of early hemodynamic optimization on biomarker patterns of severe sepsis and septic shock. Crit Care Med 2007; 35: 2016 –2024 20. Wagner F, Radermacher P, Georgieff M, et al: Sepsis therapy: What’s the best for the mitochondria? Crit Care 2008; 12:171 21. Rivers E, Nguyen B, Havstad S, et al: Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368 –1377 22. Dellinger RP, Levy MM, Carlet JM, et al: Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med 2008; 36:296 –327 23. Micek ST, Roubinian N, Heuring T, et al: Before-after study of a standardized hospital 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. order set for the management of septic shock. Crit Care Med 2006; 34:2707–2713 Cardoso T, Carneiro A, Ribeiro O, et al: The Portuguese network data: Compliance with the Surviving Sepsis Campaign bundles 关Abstract 0071兴. Intensive Care Med 2006; 32: S23 Nolan T, Berwick DM: All-or-none measurement raises the bar on performance. JAMA 2006; 295:1168 –1170 Goncalves JA Jr, Hydo LJ, Barie PS: Factors influencing outcome of prolonged norepinephrine therapy for shock in critical surgical illness. Shock 1998; 10:231–236 Martin C, Viviand X, Leone M, et al: Effect of norepinephrine on the outcome of septic shock. Crit Care Med 2000; 28:2758 –2765 Hall LG, Oyen LJ, Taner CB, et al: Fixed-dose vasopressin compared with titrated dopamine and norepinephrine as initial vasopressor therapy for septic shock. Pharmacotherapy 2004; 24:1002–1012 Micek ST, Shah P, Hollands JM, et al: Addition of vasopressin to norepinephrine as independent predictor of mortality in patients with refractory septic shock: An observational study. Surg Infect (Larchmt) 2007; 8:189 –200 Martin C, Papazian L, Perrin G, et al: Norepinephrine or dopamine for the treatment of hyperdynamic septic shock? Chest 1993; 103:1826 –1831 Ruokonen E, Takala J, Kari A, et al: Regional blood flow and oxygen transport in septic shock. Crit Care Med 1993; 21:1296 –1303 Marik PE, Mohedin M: The contrasting effects of dopamine and norepinephrine on systemic and splanchnic oxygen utilization in hyperdynamic sepsis. JAMA 1994; 272: 1354 –1357 Levy B, Bollaert PE, Charpentier C, et al: Comparison of norepinephrine and dobutamine to epinephrine for hemodynamics, lactate metabolism, and gastric tonometric variables in septic shock: A prospective, randomized study. Intensive Care Med 1997; 23: 282–287 Malay MB, Ashton RC Jr, Landry DW, et al: Low-dose vasopressin in the treatment of vasodilatory septic shock. J Trauma 1999; 47: 699 –703; discussion 703– 695 Seguin P, Bellissant E, Le Tulzo Y, et al: Effects of epinephrine compared with the combination of dobutamine and norepinephrine on gastric perfusion in septic shock. Clin Pharmacol Ther 2002; 71:381–388 Patel BM, Chittock DR, Russell JA, et al: Beneficial effects of short-term vasopressin infusion during severe septic shock. Anesthesiology 2002; 96:576 –582 Dunser MW, Mayr AJ, Ulmer H, et al: Arginine vasopressin in advanced vasodilatory shock: A prospective, randomized, controlled study. Circulation 2003; 107:2313–2319 Albanese J, Leone M, Delmas A, et al: Terlipressin or norepinephrine in hyperdynamic septic shock: A prospective, randomized study. Crit Care Med 2005; 33:1897–1902 739 39. Lauzier F, Levy B, Lamarre P, et al: Vasopressin or norepinephrine in early hyperdynamic septic shock: A randomized clinical trial. Intensive Care Med 2006; 32:1782–1789 40. Schmoelz M, Schelling G, Dunker M, et al: Comparison of systemic and renal effects of dopexamine and dopamine in norepinephrine-treated septic shock. J Cardiothorac Vasc Anesth 2006; 20:173–178 41. Annane D, Vignon P, Renault A, et al: Norepinephrine plus dobutamine versus epinephrine alone for management of septic shock: A randomised trial. Lancet 2007; 370: 676 – 684 42. Russell JA, Walley KR, Singer J, et al: Vasopressin versus norepinephrine infusion in patients with septic shock. N Engl J Med 2008; 358:877– 887 Treatment of septic renal injury by alkaline phosphatase: An emperor with new clothes* I n this issue of Critical Care Medicine, Heemskerk et al (1) describe a clinical trial on the effect of treatment with purified bovine intestinal alkaline phosphatase (AF) vs. placebo in a small series (n ⫽ 36) of patients with severe sepsis or septic shock from Gram-negative and Gram-positive microorganisms and having (impending or manifest) acute kidney injury and failure. Indeed, AF is capable of detoxifying endotoxin, even at physiologic concentrations, by dephosphorylation of the lipid A moiety of lipopolysaccharide (2), and exogenous administration in animals with Gram-negative sepsis and shock appeared beneficial (3, 4). The Heemskerk et al (1) trial was too small to discern a morbidity or survival benefit of AF treatment, but, nevertheless, the authors observed some effect on renal function parameters. In a small substudy (in patients not on renal replacement therapy), in which the effect of AF seemed even more pronounced, protection appeared associated with less inducible nitric oxide synthase (iNOS)derived nitric oxide (NO) production and release of markers of the (resultant?) injury in the tubular cells and excreted in the urine. These (selected) observations should be regarded as preliminary, and protection by AF of renal function in sepsis and shock should be confirmed in larger studies with stronger end points, such as the need for renal replacement therapy, the speed of recovery of septic renal failure, and alike, *See also p. 417. Key Words: alkaline phosphatase; renal injury; septic shock; nitric oxide synthase The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c12c 740 which have not been studied by Heemskerk et al (1). In contrast, the authors have focused on acute kidney injury criteria including serum creatinine, even in patients already on renal replacement therapy at the start. We cannot formally exclude that AF affected tubular creatinine excretion rather than glomerular filtration. Furthermore, the mechanisms behind this potentially beneficial and relatively specific effect on the kidney remain to be demonstrated, because, among others, it is unclear how AF would also benefit patients with Gram-positive septic shock. Endogenous AF is ubiquitous and abundant in epithelial cells and serves, among others, as an ectonucleotidase, to dephosphorylate and degrade extravascular (high energy, monoester) phosphate compounds. There are various isoenzymes for different tissues, which, for instance, play a role in bone turnover, bile excretion, and placental development and function (5). Expressed in the mucosa of the gastrointestinal tract, AF may help to limit translocation of harmful endotoxin from the lumen (5). The function of the enzyme located in the brush border of the proximal tubule is unclear and may only partly relate to resorption of phosphate. It is shed and excreted in the urine in the course of tubular injury and may be upregulated in the kidney during (experimental) sepsis (6, 7). Extracellular phosphate compounds, released by various cells and tissues particularly on hypoxia or inflammation, may, via nucleotide signaling and purinergic receptors, affect a wide variety of processes, involving innate immunity, epithelial transport, and regulation of blood flow (8 –10). This applies to both adenosine triphosphate and its dephos- phorylation products adenosine diphosphate, adenosine monophosphate, and adenosine, and the kidney, but effects depend on specific receptor stimulations and conditions (11). For instance, the compounds are vasodilators in normal rats but vasoconstrictors (also in the renal bed) in endotoxin-challenged animals (12). Adenosine triphosphate and adenosine, being mediators of tubuloglomerular feedback, may also decrease renal blood flow by afferent vasoconstriction via A1-adenosine receptor stimulation (8, 11, 13, 14). The compounds may have proinflammatory actions so that (specific) adenosine (receptor) antagonists are protective, whereas, in contrast, stimulation of some (A2A) adenosine receptors may have anti-inflammatory actions, even in the kidney (15, 16). We also know that inhibitors of phosphodiesterases, degrading phosphate diesters, such as cyclic adenosine monophosphate to adenosine can be protective in experimental acute kidney injury after a wide variety of challenges, including endotoxemia (17). How AF might interfere with these processes is largely unknown. Indeed, the study by Heemskerk et al (1) does not give insight into the specific effect or the routing of AF in the kidney during sepsis. Nevertheless, it suggests that iNOS upregulation is associated, even perhaps in a causative manner, with proximal tubular damage and, thereby, contributes to acute kidney injury, as observed before (18). Oxidative stress, NO-derived peroxynitrite and subsequent mitochondrial and nuclear DNA damage, and protein nitrosylation may be some of the mechanisms underlying iNOS–NO-derived toxicity. This may also explain, at least in part, the often presumed maintenance of renal blood flow and the observed fall in filtration fraction and glomerular filtration in patients with impending acute renal failCrit Care Med 2009 Vol. 37, No. 2 ure during hyperdynamic sepsis (19). We cannot judge these mechanisms from the data in the study by Heemskerk et al (1), and, therefore, the clinical significance of potential renal AFinduced adenosine and iNOS-derived NO and associated benefits and harms remain to be elucidated. Taken together, the preliminary study by Heemskerk et al (1), done in a difficult-to-study patient population, addresses important and partly novel issues on the role of AF, acute kidney injury, and iNOS in human septic shock. However, many questions remain on the new clothes of the emperor (AF) and, particularly, how they are tied together. The interesting observations need confirmation in a larger trial. A. B. Johan Groeneveld, MD, PhD, FCCP, FCCM Department of Intensive Care, VUMC—Intensive Care Amsterdam, The Netherlands Marc G. Vervloet, MD Department of Nephrology, VUMC—Intensive Care Amsterdam, The Netherlands REFERENCES 1. Heemskerk S, Masereeuw R, Moesker O, et al: Alkaline phosphatase treatment improves renal function in severe sepsis or septic shock patients. Crit Care Med 2009; 37:417– 423 2. Koyama I, Matsunaga T, Harada T, et al: Alkaline phosphatases reduce toxicity of lipopolysaccharides in vivo and in vitro through dephosphorylation. Clin Biochem 2002; 35: 455– 461 3. Verweij WR, Bentala H, Huizinga-van der Vlag A, et al: Protection against an Escherichia coli-induced sepsis by alkaline phosphatase in mice. Shock 2004; 22:174 –179 4. van Veen SQ, van Vliet AK, Wulferink M, et al: Bovine intestinal alkaline phosphatase attenuates the inflammatory response in secondary peritonitis in mice. Infect Immun 2005; 73:4309 – 4314 5. Millán JL: Alkaline phosphatases. Structure, substrate specificity, and functional relatedness to other members of a large superfamily of enzymes. Purinergic Signal 2006; 2:335–341 6. Kapojos JJ, Poelstra K, Borghuis T, et al: Induction of glomerular alkaline phosphatase after challenge with lipopolysaccharide. Int J Exp Path 2003; 84:135–144 7. Trof RJ, Di Maggio F, Leemreis J, et al: Biomarkers of acute renal injury and renal failure. Shock 2006; 26:245–253 8. Komlosi P, Fintha A, Bell PD: Renal cell-tocell communication via extracellular ATP. Physiology 2005; 20:86 –90 9. Bours MJL, Swennen ELR, Di Virgilio F, et al: Adenosine 5⬘-triphosphate and adenosine as endogenous signaling molecules in immunity and inflammation. Pharmacol Ther 2006; 112:358 – 404 10. Vallon V: P2 receptors in the regulation of renal transport mechanisms. Am J Physiol Renal Physiol 2008; 294:F10 –F27 11. Guan Z, Osmond DA, Inscho EW: Purinoceptors in the kidney. Exp Biol Med 2007; 232: 715–726 12. Jolly L, March JE, Kemp PA, et al: Regional haemodynamic responses to adenosine receptor activation vary across time following lipopolysaccharide treatment in conscious rats. Br J Pharmacol 2008; 154: 1600 –1610 13. Vallon V, Mühlbauer B, Osswald H: Adenosine and kidney function. Physiol Rev 2006; 86:901–940 14. Castrop H: Mediators of tubuloglomerular feedback regulation of glomerular filtration: ATP and adenosine. Acta Physiol Scand 2007; 189:3–14 15. Linden J: Molecular approach to adenosine receptors: Receptor-mediated mechanisms of tissue protection. Annu Rev Pharmacol Toxicol 2001; 41:775–787 16. Okusa MD: A2A adenosine receptor: A novel therapeutic target in renal disease. Am J Physiol Renal Physiol 2002; 282:F10 –F18 17. Wang W, Zolty E, Falk S, et al: Pentoxifylline protects against endotoxin-induced acute renal failure in mice. Am J Physiol Renal Physiol 2006; 291:F1090 –F1095 18. Wang W, Zolty E, Falk S, et al: Endotoxemiarelated acute kidney injury in transgenic mice with endothelial overexpression of GTP cyclohydrolase-1. Am J Physiol Renal Physiol 2008; 294:F571–F576 19. Wan L, Bagshaw SM, Langenberg C, et al: Pathophysiology of septic acute kidney injury: What do we really know? Crit Care Med 2008; 36:S198 –S203 Is target population more important than patient location when evaluating tight glycemic control?* C linical trials in the intensive care unit can be divided into two categories. The first category includes disease- or syndrome-specific trials that examine the effect of a drug or practice on a group of patients with that specific disease- or syndrome. Examples of this type of trial *See also p. 424. Key Words: tight glucose control; fibrinolysis; sepsis; insulin therapy; plasminogen activator inhibitor-1 Supported by SCCM Vision Grant (BRG); Supported by K-23 GMO7-1399 (JES). The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318195468e Crit Care Med 2009 Vol. 37, No. 2 would be the use of lung-protective ventilation in patients with acute lung injury (1) or the use of steroids in patients with septic shock (2). The second category of trial enrolls patients with a particular severity of disease that leads to admission to an intensive care unit. This type of treatment-specific trial would include patients with different types of disease that precipitate similar severity of illness and location within an intensive care unit. Examples of this type of trial would include the Saline versus Albumin Fluid Evaluation trial examining the use of specific forms of volume resuscitation in critically ill patients (3) or goal-directed therapy in patients with a broad range of critical illness (4). Such trials in a heterogeneous group of patients may provide improved external validity. However, variation in response to an investigational treatment across different disease processes may obscure a potentially beneficial treatment effect. It is, therefore, important to consider whether the disease process might modify the effect of the treatment being tested when choosing which category of trial to design. Most of the larger trials of tight glycemic control in patients with critical illness fit the second category of studies (5– 8). These trials were performed on a broad range of critically ill patients, including patients with a variety of surgical and medical diagnoses, and the effects of tight glycemic control across these groups of patients were tested. Importantly, the benefits of tight glycemic control seen in 741 primarily cardiovascular surgical patients (6, 9) were not replicated in two large trials in different patient populations (5, 7). In addition, tight glycemic control may also increase episodes of potentially harmful hypoglycemia. Most notably, the Efficacy of Volume Substitution and Insulin Therapy in Severe Sepsis and the Glucontrol studies (7, 8) documented a potential association between hypoglycemic episodes and worsened outcomes in critically ill patients. This tension between tight glycemic control and avoidance of hypoglycemic episodes highlights the importance of selecting the proper patient population in which to test the risk– benefit ratio of tight glycemic control. The largest trial of tight glycemic control, the Normoglycaemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation Study (NCT00220987), which continues to enroll patients, also tests the effects of glucose control in a heterogeneous group of critically ill patients. The study by Savoli et al (10) in this issue of Critical Care Medicine fits the first category of studies, testing tight glycemic control in patients with sepsis. In their prospective, multicentered study of 90 septic patients randomized to tight glycemic control (mean glucose, 112 mg/ dL) vs. conventional glycemic control (mean glucose, 159 mg/dL), blood plasminogen activating factor inhibitor-1 (PAI-1) activity, PAI-1 concentration, and tissue plasminogen activator levels were noted to decrease more rapidly over the course of 28 days in the tight glycemic control group. The treatment group also had lower Sequential Organ Failure Assessment scores over the same time period. Most of the differences between groups were found after 10 –15 days of therapy, when many patients had either left the intensive care unit or died. Fibrinolysis, as measured by PAI-1 activity and concentration, was inhibited in only a subset of their septic patients and this inhibition was associated with higher baseline Sequential Organ Failure Assessment scores and mortality rates (10). Of note, with the exception of interleukin-6, no significant difference in the levels of inflammatory mediators was found between the treatment groups. The findings of Savoli et al (10) are consistent with previous studies showing that 742 insulin dosing may limit levels of PAI-1 in healthy adults. Further, insulin-regulated normoglycemia has been found to prevent immune dysfunction and inappropriate inflammation, endothelial function, and coagulation (11). These data, coupled with the fact that the septic state contributes to microvascular thrombosis, tissue-factormediated thrombin activation, impaired fibrinolysis, and increased PAI-1 levels (12), make it reasonable to propose that a potential mechanism behind the beneficial effect of tight glycemic control via intensive insulin therapy may be due to these coagulation modulatory effects. Of note, neither the findings of Savoli et al nor a recent study by Langouche et al (13) demonstrate differences in inflammatory markers between patients treated with tight glycemic control and conventional glycemic control. Although the study by Savoli et al provides a biologically plausible rationale for improved outcomes in septic patients with tight glycemic control, it does not provide enough evidence to change practice without having the results replicated in a larger study. The lack of information on length of exposure to usual glycemic control before enrollment and the unclear mechanism that might lead to delayed inhibition of fibrinolysis in the control group limit the generalizability of the results. As with the trial by Savoli et al, the selection of a more homogeneous patient population may permit the use of smaller sample sizes to demonstrate a treatment effect. In addition to large multicenter trials such as NICE–SUGAR, there room for additional appropriately designed smaller trials to test biologically plausible treatments in well-defined patient populations with specific syndromes or diseases. B. Robert Gibson, MD Department of Surgery Johns Hopkins Medical Institutions Baltimore, MD Jonathan E. Sevransky, MD, MHS Division of Pulmonary and Critical Care Medicine Johns Hopkins Medical Institutions Baltimore, MD REFERENCES 1. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The acute respiratory distress syndrome network. N Engl J Med 2000; 342: 1301–1308 2. Sprung CL, Annane D, Keh D, et al: Hydrocortisone therapy for patients with septic shock. N Engl J Med 2008; 358:111–124 3. Finfer S, Bellomo R, Boyce N, et al: A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004; 350:2247–2256 4. Gattinoni L, Brazzi L, Pelosi P, et al: A trial of goal-oriented hemodynamic therapy in critically ill patients. N Engl J Med 1995; 333: 1025–1032 5. Van den Berghe G, Wilmer A, Hermans G, et al: Intensive insulin therapy in the medical ICU. N Engl J Med 2006; 354: 449 – 461 6. van den Berghe G, Wouters P, Weekers F, et al: Intensive insulin therapy in the critically ill patients. N Engl J Med 2001; 345: 1359 –1367 7. Brunkhorst FM, Engel C, Bloos F, et al: Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008; 358:125–139 8. Devos P, Preiser J, Melot C, et al: Impact of tight glucose control by intensive insulin therapy on ICU mortality and the rate of hypoglycaemia: Final results of the GluControl study. Intensive Care Med 2007; 33:S189 9. Furnary AP, Zerr KJ, Grunkemeier GL, et al: Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures. Ann Thorac Surg 1999; 67:352–360; discussion 360 –362 10. Savoli M, Cugno M, Polli F, et al: Tight glycemic control may favor fibrinolysis in patients with sepsis. Crit Care Med 2009; 37: 424 – 431 11. Stegenga ME, van der Crabben SN, Levi M, et al: Hyperglycemia stimulates coagulation, whereas hyperinsulinemia impairs fibrinolysis in healthy humans. Diabetes 2006; 55: 1807–1812 12. Levi M, Ten Cate H: Disseminated intravascular coagulation. N Engl J Med 1999; 341: 586 –592 13. Langouche L, Meersseman W, Vander Perre S, et al: Effect of insulin therapy on coagulation and fibrinolysis in medical intensive care patients. Crit Care Med 2008; 36: 1475–1480 Crit Care Med 2009 Vol. 37, No. 2 Myocardial depression/injury in sepsis: Two sides of the same coin?* I n the modern era, the concept of reversible myocardial depression or dysfunction was described by Wiggers (1). He postulated the existence of a myocardial depressing factor responsible for myocardial dysfunction in hemorrhagic shock. During the 1960s and 1970s, experimental studies showed evidence of transient myocardial dysfunction in several forms of critical disease, including hemorrhagic and septic shock (2, 3). Sequential studies have shown that patients in septic shock adequately resuscitated typically displayed a high output and low-systemic resistance hemodynamic circulatory condition, with myocardial depression despite the high output (4 –7). The presence of a normal or even high ejection fraction does not exclude myocardial derangements. In those patients who died, this hemodynamic pattern persisted until death. The initial phase of understanding and the study of cardiovascular manifestations in sepsis and septic shock began with the development of radionuclide cineangiography techniques (radioisotopic ventriculography) and with the application of volumetric echocardiography in managing the critically ill patients. Discussions on the true involvement of the heart in sepsis and septic shock, regardless of hemodynamic conditions, date back to the early 1960s (8) when some studies already used endotoxic shock models in animals. In the 1980s, using nuclear medicine techniques, Parker et al (6) demonstrated the decreased biventricular ejection fraction in these septic patients. The connection between clinical myocardial depression *See also p. 441. KEY WORDS: myocardial depression; myocardial injury; troponin; sepsis; echocardiography; activated protein C The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194de3e Crit Care Med 2009 Vol. 37, No. 2 and the effects of myocardial depressor substances was described by Parillo et al (9), in the late 1980s, by measuring the serum levels of the substances in these patients during the septic phase. This study established a strong tie between in vivo and in vitro observations of cardiac function and the activity of myocardial depressor substances in septic shock. In 1994, we and others (10) published the histopathologic findings of the myocardium in 71 autopsies of patients who met morphologic criteria of sepsis, comparing them with a control group and observing the presence of interstitial myocarditis in 27% of the sample, bacterial colonization in 11%, necrosis of cardiac fibers in 7%, and interstitial edema in 28%, although this last finding did not demonstrate a significant difference relative to controls. Furthermore, the identification of troponin in 1999 as a reliable marker of myocardial injury in sepsis added some new insights to this complex disease (11). In this issue of Critical Care Medicine, Bouhemad et al (12) were auspicious in bringing some light to this matter by identifying two groups of troponinpositive septic patients whose clinical and echocardiographic features were markedly different. Echocardiographic abnormalities showing increased left-ventricular dimensions and reduced ejection fraction were detected in those patients with severe but reversible myocardial dysfunction, whereas those who were not able to dilate, probably because of greater infiltrate of polymorphonuclear cells within the myocardial fibers rendering the myocardium less compliant, were more prone to ultimate death. It is now clear that we cannot manage every septic patient the very same way. As pointed out by the authors, the concept of preload recruitment applies exclusively to the subgroup with systolic left ventricular impairment where ventricular enlargement may represent an adaptive mechanism to main- tain cardiac output, whereas those presenting with ventricular relaxation impairment despite showing no ventricular dilation at all and no drop in ejection fraction exhibit worse prognosis. More recently, John et al (13) were able to show better outcomes in troponin-positive patients submitted to activated protein C administration. Like cardiac patients, we can possibly now stratify the most severe septic patients by determining troponin levels and echocardiographic ventricular dimensions and, therefore, select the most suitable therapeutic choices and probably achieve better outcomes in these critically ill septic patients. Further studies should address these questions to clarify the most controversial issues related to this two-sided coin disease. Constantino Fernandes, MD Hospital Israelita Albert Einstein—Critical Care Department, Avenida Albert Einstein Sao Paulo, Brazil REFERENCES 1. Wiggers CJ: Myocardial depression in shock. A survey of cardiodynamic studies. Am Heart J 1947; 33:633– 650 2. Weil MH, MacLean LD, Visscher MD, et al: Studies on circulatory changes in the dog produced by endotoxin from gram-negative microorganisms. J Clin Invest 1956; 35: 1191–1198 3. Solis RT, Downing SE: Effects of E. coli endotoxemia on ventricular performance. Am J Physiol 1966; 211:307–313 4. Vincent JL, De Backer D: Inotrope/vasopressor support in sepsis-induced organ hypoperfusion. Semin Respir Crit Care Med 2001; 22:61–74 5. Ellrodt AG, Riedinger MS, Kimchi A, et al: Left ventricular performance in septic shock: Reversible segmental and global abnormalities. Am Heart J 1985; 110:402– 409 6. Parker MM, Shelhamer JH, Bacharach SL, et al: Profound but reversible myocardial depression in patients with septic shock. Ann Intern Med 1984; 100:483– 490 743 7. Reilly JM, Cunnion RE, Burch-Whitman C, et al: A circulating myocardial depressant substance is associated with cardiac dysfunction and peripheral hypoperfusion (lactic acidemia) in patients with septic shock. Chest 1989; 95:1072–1080 8. Lefer AM, Martin J: Origin of myocardial depressant factor in shock. Am J Physiol 1970; 218:1423–1427 9. Parillo JE, Burch C, Shelhamer JH, et al: A circulating myocardial depressant substance in humans with septic shock. Septic shock patients with a reduced ejection fraction have a circulating factor that depresses in vitro myocardial cell performance. J Clin Invest 1985; 76:1539 –1553 10. Fernandes CJ Jr, Iervolino M, Neves RA, et al: Interstitial myocarditis in sepsis. Am J Cardiol 1994; 74:958 11. Fernandes CJ Jr, Akamine N, Knobel E: Cardiac troponin: A new serum marker of myo- cardial injury in sepsis. Intensive Care Med 1999; 25:1165–1168 12. Bouhemad B, Nicolas-Robin A, Arbelot C, et al: Acute left ventricular dilatation and shock-induced myocardial dysfunction. Crit Care Med 2009; 37:441– 447 13. John J, Awab A, Norman D, et al: Activated protein C improves survival in severe sepsis patients with elevated troponin. Intensive Care Med 2007; 33:2122–2128 How long does it take to demonstrate the value of an idea?* I n 1991, the Consensus Conference of the American College of Chest Physicians and the Society of Critical Care Medicine introduced the systemic inflammatory response syndrome, defined the presence of both infection and systemic inflammatory response syndrome as sepsis, and stated clear definitions of severe sepsis and septic shock (1). Subsequently, infection and sepsis appeared to have similar outcomes, unaffected by the presence or number of inflammatory response criteria (2). A decade after the previous Consensus Conference, the scientific community of intensivists developed the idea of a staging system for sepsis similar to that used by oncologist to stratify septic patients on the basis of their predisposition, infection, host response, and concomitant organ dysfunction (PIRO) (3). Community-acquired pneumonia (CAP) is one of the most frequent infectious diseases. CAP refers to pneumonia in a previously healthy person who has acquired the infection from outside the hospital, and therefore, it is also responsible for high costs for the society. Furthermore, the mortality associated with CAP leading to organ dysfunction in the patients admitted to intensive care unit (ICU) is high. Taking the PROWESS study as an example of trial on severe sepsis, globally, 602 of the 1690 patients *See also p. 456. Key Words: pneumonia; sepsis; predisposition, infection, host response, and concomitant organ dysfunction; intensive care unit; severity of illness The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194be4f 744 (35.6%) enrolled had CAP, and 160 of them died (26.6%) (4). Therefore, CAP is a frequent cause of severe sepsis and death. To describe and compare CAP patients involved in clinical trials, investigators need both a clear definition of CAP and a tool to assess the severity of illness. The first one was given by the Infectious Diseases Society of America (5), which defined CAP as an acute infection of the pulmonary parenchyma in a patient not hospitalized or residing in a long-term care facility for 14 days or more before onset of symptoms. As far as the assessment of the severity of illness is concerned, mortality rate for CAP ranges from 5% in hospital ward patients to 14% in ICU patients, with a mean Simplified Acute Physiology Score II (SAPS) of 33 (6) to 43% in those with a mean SAPS II of 46 (7). Therefore, the assessment of the degree of the severity of illness in ICU patients with CAP is substantial for any patient comparison. In this issue of Critical Care Medicine, Rello et al (8) present a scoring system specific for CAP based on the PIRO concept, aimed at stratifying critically ill patients in mortality risk groups. The authors compared the performance of the new score, named CAP PIRO, with that of two different scores: one specific, ATSrevised criteria, and the other generic, Acute Physiology and Chronic Health Evaluation II. The CAP PIRO included the presence of the following variables: comorbidities (chronic obstructive pulmonary disease, immunocompromise) and age ⬎70 years, representing the predisposition of the PIRO concept; multilobar opacities in chest x-ray and bacteremia, representing the insult; shock and severe hypoxemia, representing the response; and acute renal failure and acute respiratory distress syndrome, as surrogate of organ dysfunction (8). The CAP PIRO, ranging 0 – 8, allowed stratifying the patients in the following four categories of risk: low, 0 –2 points, with 28-day death rate of 3.6%; mild, 3 points, with 13% 28-day mortality; high, 4 points, with 43% 28-day mortality; and very high, 5– 8 points, having a 28-day mortality of 76.3%. The discriminative ability, measured by the area under the receiver operating curve, was better for CAP PIRO than for ATS-revised criteria and Acute Physiology and Chronic Health Evaluation II. Not only mortality but also ICU length of stay and duration of mechanical ventilation were significantly different in the four levels of risk of CAP PIRO (8). The article by Rello et al (8) has some major strengths. First, the study was performed on data prospectively collected by 33 Spanish ICUs, with a large sample of 529 adults. The wide odds ratio reported for incidence of shock, severe hypoxemia, or acute respiratory distress syndrome, as well as the wide interquartile range of lengths of mechanical ventilation and ICU stay suggest differences in case-mix and/or practices between ICUs, even in the same country, but this finding is common to multicenter studies. Second, the CAP PIRO allows stratifying the patients admitted to ICU with CAP according to the risk of death at 28 days; therefore, it can be useful for researchers involved in clinical trials on this topic (for instance, on antibiotics). Finally, it suggests a relationship between CAP PIRO and healthcare resource use. The study by Rello et al (8) has some limitations. The authors did not collect hospital mortality data and considered patients discharged alive from ICU within 28 days as survivors. Some of those paCrit Care Med 2009 Vol. 37, No. 2 tients may have died in the hospital after ICU discharge, according to a general documented post-ICU mortality of 10.8% (9). Even more relevant, in a group of septic patients, hospital mortality has been reported to be 48.3%, whereas 28day and 60-day mortalities were 44.8% and 47.8%, respectively (10). The finding that hospital mortality is consistently higher than 28-day mortality in such patients suggests that a measure of longterm outcome should be associated with CAP PIRO in clinical trials. Unfortunately, the CAP PIRO does not allow computing a probability of hospital mortality. Furthermore, it was not compared with more recent prognostic models, namely SAPS 3 Admission Score (11, 12) or Acute Physiology and Chronic Health Evaluation IV (13), which were not available at the time of data collection, or SAPS II customized for severe sepsis/septic shock (14). Another article on the application of the PIRO concept has been published few months ago in Intensive Care Medicine (15). The study was performed on the SAPS 3 Admission Score database to develop a model predicting hospital mortality of patients ICU admitted with infection. There is some overlap with the study of Rello et al (8), and not surprisingly, there are predictive variables common to the two studies, even if categorized in a different way. Despite the different target populations of the two scores based on the PIRO concept, namely, CAP and sepsis patients, the publication of those articles (8, 15) in such a short-time interval suggests that 5 years is the time lag required by the scientific com- Crit Care Med 2009 Vol. 37, No. 2 munity to demonstrate the value of an idea like PIRO. Maurizia Capuzzo, MD Department of Surgical, Anesthetic and Radiological Sciences Section of Anesthesiology and Intensive Care, University Hospital of Ferrara Ferrara, Italy 8. 9. REFERENCES 1. Members of the American College of Chest Physicians/Society of Crit Care Med Consensus Conference Committee: American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864 – 874 2. Alberti C, Brun-Buisson C, Goodman SV, et al: Influence of systemic inflammatory response syndrome and sepsis on outcome of critically ill infected patients. Am J Respir Crit Care Med 2003; 168:77– 84 3. Levy MM, Fink MP, Marshall JC, et al: 2001 SCCM/ESICM/ACCP/ATS/SIS International sepsis definitions conference. Crit Care Med 2003; 31:1250 –1256 4. Laterre PF, Garber G, Levy H, et al: Severe community-acquired pneumonia as a cause of severe sepsis: Data from the PROWESS study. Crit Care Med 2005; 33:952–961 5. Mandell LA, Bartlett JG, Dowell SF, et al: Update of practice guidelines for the management of community-acquired pneumonia in immunocompetent adults. Clin Infect Dis 2003; 37:1405–1433 6. Leroy O, Saux P, Bédos J-P, et al: Comparison of levofloxacin and cefotaxime combined with ofloxacin for ICU patients with community-acquired pneumonia who do not require vasopressors. Chest 2005; 128:172–183 7. Paganin F, Lilienthal F, Bourdin A, et al: 10. 11. 12. 13. 14. 15. Severe community-acquired pneumonia: Assessment of microbial aetiology as mortality factor. Eur Respir J 2004; 24:779 –785 Rello J, Rodriguez A, Lisboa T, et al: PIRO score for community-acquired pneumonia: A new prediction rule for assessment of severity in intensive care unit patients with community-acquired pneumonia. Crit Care Med 2009; 27:456 – 462 Azoulay E, Adrie C, De Lassence A, et al: Determinants of postintensive care unit mortality: A prospective multicenter study. Crit Care Med 2003; 31:428 – 432 Garnacho-Monteiro J, Garcia-Garmendia JL, Barrero-Almodovar A, et al: Impact of adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive care unit with sepsis. Crit Care Med 2003; 31:2742–2751 Metnitz PGH, Moreno RP, Almeida E, et al: SAPS 3—From evaluation of the patient to evaluation of the ICU. Part 1. Objectives, methods and cohort description. Intensive Care Med 2005; 31:1336 –1344 Moreno RP, Metnitz PGH, Almeida E, et al: SAPS 3—From evaluation of the patient to evaluation of the ICU. Part 2. Development of a prognostic model for hospital mortality at ICU admission. Intensive Care Med 2005; 31: 1345–1355 Zimmerman JE, Kramer AA, McNair DS, et al: Acute Physiology and Chronic Health Evaluation (APACHE) IV: Hospital mortality assessment for today’s critically ill patients. Crit Care Med 2006; 34:1297–1310 Le Gall JR, Lemeshow S, Leleu G, et al: Customized probability models for early severe sepsis in adult intensive care patients. Intensive Care Unit Scoring Group. JAMA 1995; 273:644 – 650 Moreno RP, Metnitz B, Adler L, et al: Sepsis mortality prediction based on predisposition, infection and response. Intensive Care Med 2008; 34:496 –504 745 Intensive insulin therapy: The swinging pendulum of evidence* S ince 2001, the year of publication of the first intensive insulin trial by Van den Berghe et al (1), hyperglycemia and intensive insulin therapy (IIT) are placed high on the list of critical care controversies. In this trial, Van den Berghe demonstrated a survival benefit for surgical intensive care unit (ICU) patients treated with an intensive insulin protocol to maintain normoglycemia (80⫺110 mg/ dL; 4.6% compared with 8.0% in-hospital mortality). Although this was a singlecenter trial according to evidence base medicine standards is not so high, IIT was adopted by many and even incorporated in guidelines for sepsis patients, such as the Surviving Sepsis Campaign, endorsed by the Society of Critical Care Medicine, and the European Society of Intensive Care Medicine (2, 3). Since the original hallmark study, several observational trials confirmed the positive effects of normoglycemia on outcome in critically ill patients (4, 5). However, enthusiasm weaned after the results came out of three other prospective randomized trials on IIT in critically ill patients. Of these, two trials are published (6, 7), one trial is finished, but only presented as an abstract (http:// clinicaltrials.gov/ct/gui/show/NCT00107601). A fourth trial, the Australian–New Zealand– Canadian “Normoglycemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation” trial, is still recruiting (http://www.controlledtrials.com/ISRCTN04968275). Unfortunately, the results of these trials have made things less clear. The single-center follow-up trial by Van den Berghe in medical ICU patients rendered ambivalent results (6). There was no difference in outcome between patients treated with IIT and conventional insulin therapy. How- *See also p. 463. Key Words: intensive insulin therapy; blood glucose; outcome; critically ill; hypoglycemia; hyperglycemia The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194be79 746 ever, in the subgroup of patients admitted for 3 days or more, IIT was associated with a survival benefit (in-hospital mortality in the conventional treated patients was 53.5% vs. 43% in IIT patients). This subgroup of patients could not be identified beforehand. An important difference with the previous trial of this group was that patients in the medical ICU were more severely ill. Also, there was a much higher incidence of hypoglycemia in patients treated with IIT in the medical ICU, compared with the surgical ICU trial (18.7% vs. 5.1%). The German multicenter Efficacy of Volume Substitution and Insulin Therapy in Severe Sepsis study randomized severe sepsis patients to IIT or conventional insulin therapy and Ringer’s lactate or a starch solution for fluid resuscitation (7). The study was stopped for safety reasons by the data and safety monitoring board of the study, after an interim analysis demonstrated an unexpected high rate of hypoglycemic events (17.0%) in patients randomized to IIT. The positive effects of IIT, as demonstrated in the first study by Van den Berghe et al, could not be confirmed, not even in subanalysis. In the 537 patients eligible for analysis, there was no difference in a series of outcomes including 28- and 90-day mortality between patients treated with conventional insulin therapy and IIT. The European multicenter Glucontrol study was also prematurely stopped after inclusion of 1101 patients because of a higher rate of adverse events in IIT patients. All-cause mortality was not different between patients exposed to conventional and IIT (17% vs. 15%, p ⫽ not significant). In summary, the beneficial effects of IIT were only confirmed in a subanalysis of a more sick patient cohort, by the team that invented the therapy, and could not be reproduced in two, underpowered, multicenter trials. In addition, there are concerns regarding the side effects of IIT. There was a high incidence of hypoglycemia both in the experienced hands of the Leuven team and in the less experienced hands of the centers that participated in the two multicenter studies. These results raise concerns regarding the external validity of IIT. In other words does IIT work in other, less experienced units? It also illustrates that it is not necessarily correct to translate a therapy that works in a specific patient cohort to another cohort. In this issue of Critical Care Medicine, Bagshaw and colleagues present the results from an observational study on a very large database of 66,184 Australian ICU patients (8). Although this is an observational study, it adds a whole new dimension to the IIT controversy. In this study, normoglycemia was associated with better outcome, even after correction for other covariates for increased mortality. In other words, blood glucose levels in the normal range are good, as in the original study by Van den Berghe et al. Interestingly, the “normal” range in this study is higher than that used in the studies on IIT mentioned before (5.6 – 8.69 vs. 4.4 –5.6 mmol/L). This range was determined as the second quartile of all average blood glucose concentrations, and it corresponds remarkably well with the conservative recommendations of the Surviving Sepsis Campaign (blood glucose ⬍150 mg/dL or 8.25 mmol/L). An important aspect in this study is that patients were categorized on an average blood glucose concentration during a 24-hour period, and not on the morning blood glucose, as in the Leuven studies and in the VISEP study. This may be of importance, as it has been demonstrated that glucose control may vary considerably in IIT patients, during a 24-hour observation period (9). Differences in blood glucose control during a 24-hour period, may explain why IIT works in the hands of Van den Berghe’s team, and why this could not be reproduced in the VISEP and Glucontrol trials. One may assume that Van den Berghe’s team is more experienced and dedicated, and therefore is able to maintain a stricter glucose control during a 24-hour period compared with less experienced units. Higher target levels may diminish the beneficial effects of IIT, but more adequate blood glucose control during a 24-hour period may compensate for that. Therefore, average blood glucose may be a more adequate marker of the efficacy of IIT, and should be considered in future studies. In addition, higher target blood glucose concentration will probably lower the inciCrit Care Med 2009 Vol. 37, No. 2 dence of hypoglycemia and make this therapy a more safe one, even in less experienced hands. Another interesting finding is the U-shaped outcome curve. Patients with an average blood glucose at the lower and high end of the spectrum do worse compared with patients in the normal range. Going low is not always good; this supports the hypothesis that less aggressive IIT and blood glucose control during a 24-hour period may improve outcome, even in a multicenter setting. Finally, this study also provides us with an explanation for the different results in the prospective IIT studies published to date. The effects of normal blood glucose were most prominent in the surgical and cardiac surgery subgroups, comparable with the cohort in the first Leuven trial (1). The beneficial effects were less pronounced or even absent in sepsis patients and medical patients, like the patients included in the VISEP study and the second study by the Leuven group (6, 7). Future studies on IIT should probably look at even more specific cohorts, such as acute respiratory distress syndrome patients and patients admitted after trauma. There are several limitations to this trial. It is an observational trial and, therefore, prone to bias, and average blood glucose measured during the first 24 hours was used as a surrogate for blood glucose control during the rest of the ICU stay. The study is, therefore, more hypothesis generating than the absolute proof of evidence. However, the multicenter setting and the large number of patients give us robust data. These data may help us explain results from other studies in this field and design new studies on this topic. Eric A. J. Hoste, MD, PhD Intensive Care Unit, Ghent University Hospital Gent, Belgium 3. 4. 5. 6. 7. 8. REFERENCES 1. Van den Berghe G, Wouters P, Weekers F, et al: Intensive insulin therapy in critically ill patients. N Engl J Med 2001; 345:1359 – 1367 2. Dellinger RP, Carlet JM, Masur H, et al: Surviving Sepsis Campaign guidelines for man- 9. agement of severe sepsis and septic shock. Crit Care Med 2004; 32:858 – 873 Dellinger RP, Levy MM, Carlet JM, et al: Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2008. Intensive Care Med 2008; 34:17– 60 Finney SJ, Zekveld C, Elia A, et al: Glucose control and mortality in critically ill patients. JAMA 2003; 290:2041–2047 Krinsley JS: Effect of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 2004; 79:992–1000 Van den Berghe G, Wilmer A, Hermans G, et al: Intensive insulin therapy in the medical ICU. N Engl J Med 2006; 354:449 – 461 Brunkhorst FM, Engel C, Bloos F, et al: Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008; 358:125–139 Bagshaw SM, Egi M, George C, et al: Early blood glucose control and mortality in critically ill patients in Australia. Crit Care Med 2009; 37:463– 470 Oeyen SG, Hoste EA, Roosens CD, et al: Adherence to and efficacy and safety of an insulin protocol in the critically ill: A Prospective Observational Study. Am J Crit Care 2007; 16: 599 – 608 Cardiocerebral resuscitation: Few answers, more questions* A major concern since the advent of the modern age of cardiac resuscitation is that we might be resuscitating patients but leaving a large group of patients with prolonged neurologic disability. Cobbe et al (1) found that about 40% of initial survivors of out-of-hospital cardiac arrest could be discharged home without major neurologic disability. Adrie et al (2) found that while early death from refractory shock occurred in 42 patients from a group of 130 patients achieving restoration of spontaneous circulation (ROSC) for ⬎1 hour, 60 of the remaining 88 patients died of complications related to neurologic failure later in the hospital course. Concern that survivors from car- *See also p. 471. Key Words: cardiac arrest; heart failure; cerebral oxygenation; ventricular fibrillation; cardiocerebral resuscitation The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c426 Crit Care Med 2009 Vol. 37, No. 2 diac arrest could become a long-term burden on the society was ameliorated by studies like these showing that there was a high rate of mortality in severely neurologically impaired patients in the period immediately after resuscitation. Cole and Corday (3) found that only two patients (7% of the total) requiring more than 4 minutes of advance cardiac life support could be resuscitated and discharged alive, both with permanent brain damage. In a study of resuscitated inhospital intensive care unit and nonintensive care unit victims of cardiac arrest, Bell and Hodgson (4) found that although 30% and 45% of patients, respectively, were comatose after resuscitation, most of these patients died. Only 3% and 4.5% of patients from these groups were discharged alive with brain damage. No-reflow, the concept of achieving ROSC, but subsequently suffering a fall in cerebral blood flow, which, in turn, contributed to poor cerebral outcome, and ultimately, the demise of the patient was postulated by Negovsky (5) and shown to exist in animal and clinical models. Safar (6), challenging the prevailing emphasis on cardiac resuscitation, noted that too many victims achieve ROSC as demonstrated in the studies earlier, only to die of cerebral failure, emphasizing the need for a new type of cardiopulmonary resuscitation (CPR), cardiocerebral CPR, or CPCR. Mullner et al (7) demonstrated that prearrest variables and postarrest neurologic function influenced outcome. Adverse outcome increased with increasing age and in patients with increasing numbers of prearrest diseases, especially congestive heart failure and diabetes mellitus. Again, longterm survival with profound neurologic impairment was not found to be a problem with 106 of the 121 patients (29% of the survivors) with unfavorable neurologic recovery dead within 6 months of the initial cardiac arrest. Age and an ejection fraction ⬍35% were also shown to be associated with unfavorable outcomes by Bunch et al (8) in a study of ischemic vs. nonischemic heart disease-associated cardiac arrests. The influence of left ventricular function was confirmed in a study of in-hospital cardiac arrest. Gonzalez et al (9) found a 19% survival to discharge in patients with a nor747 mal prearrest ejection fraction vs. 8% in those with moderate to severe left ventricular dysfunction. In this issue of Critical Care Medicine, Skhirtladze et al (10) help explain the influence of heart failure on neurologic outcome from cardiac arrest in their fascinating model of reversible humancardiac arrest by studying patients having a cardiodefibrillator tested during implantation. Although physiologic studies have been conducted on victims of cardiac arrest, most have been on patients with long downtimes before instrumentation. In this study, ventricular fibrillation is induced to test the cardiodefibrillator, offering the opportunity to measure cerebral oxygenation before and after ROSC. In this elegant study, the authors showed that despite similar blood pressure, heart rate, and pulse oximetry values, patients with severe heart failure were more likely to have clinically important cerebral desaturation detected by near infrared spectroscopy. This was ev ident at baseline and after brief periods of cardiac arrest. This might explain why patients with heart failure suffering cardiac arrest do so poorly. Cohan et al (11) found seemingly paradoxical results in a xenon study of cerebral blood flow in humans after cardiac arrest. In this study, comatose patients who regained consciousness had relatively normal cerebral blood flow before regaining consciousness, whereas those who died without awakening developed hyperemic cerebral blood flow after ROSC. Inoue et al (12) also found early hyperemia to be associated with a poor prognosis after cardiac arrest. Conversely, Mullner et al (13) found higher cerebral oxygen extraction to be associated with better cerebral recovery. Different models and different techniques do not allow us to directly compare the studies, but the results emphasize that we need to understand more about postresuscitation cerebral blood flow and its relationship to underlying brain ischemia before we can know the correct approach to CPCR in the many different situations that present to us. It does not mean that a hyperemic flush is not helpful after attainment of ROSC, but emphasizes that we really do not know the “formula” for postresuscitation care. Prearrest variables, such as heart failure, vascu- 748 lar disease, diabetes, and intra-arrest variables, such as duration of no-flow and lowflow may all need to be accounted for to determine the appropriate approach to postROSC care for the individual patient. Optimal cerebral resuscitation will probably turn out to be similar to defibrillation— one approach will not fit all patients. Just as we may need to do chest compression before defibrillation in patients with prolonged no-flow times, we may need to tailor our approach differently for patients with premorbid conditions, long no-flow, or lowflow times. Perhaps to achieve better outcomes, we should monitor efficacy of artificial circulation with widely available, easily applied tools, such as end-tidal CO2 (14) while maximizing cardiac output with techniques, such as continuous chest compressionCPR. Kellum et al (15) showed that with continuous chest compression-CPR, survival increased from 19% to 48%, compared with a control group derived from the 3 years preceding the test period. In addition, neurologically intact survival increased from 77% in the first period vs. 84% in the experimental period. Unfortunately, after ⬎50 years of CPR and advanced cardiac life support, the majority of patients still either cannot achieve ROSC or die of postresuscitation disease. Basic physiologic studies in animals and humans, such as the study by Skhirtladze et al in this issue of Critical Care Medicine help answer small parts of the question. In turn, these clues need to be taken back to bench models and cell-based studies to develop better techniques for the key variables involved in resuscitation from cardiac arrest: generation of cardiac output during the arrest phase, generation of an organized rhythm, and finally functional cardiac and cerebral recovery. Robert L. Levine, MD, FACEP, FCCM Department of Emergency Medicine University of Texas School of Medicine at Houston The Department of Medicine The Methodist Hospital Houston, TX REFERENCES 1. Cobbe SM, Dalziel K, Ford I, et al: Survival of 1476 patients initially resuscitated from out 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. of hospital cardiac arrest. BMJ 1996; 312: 1633–1637 Adrie C, Cariou A, Mourvillier B, et al: Predicting survival with good neurological recovery at hospital admission after successful resuscitation of out-of-hospital cardiac arrest: The OHCA score. Eur Heart J 2006; 27:2840 –2845 Cole SL, Corday E: Four-minute limit for cardiac resuscitation. JAMA 1956; 161: 1454 –1458 Bell JA, Hodgson HJF: Coma after cardiac arrest. Brain 1974; 97:361–372 Negovsky VA: The second step in resuscitation- the treatment of the “post-resuscitation disease.” Resuscitation 1972; 1:1–7 Safar P: Resuscitation medicine research: Quo Vadis. Ann Emerg Med 1996; 27: 542–552 Mullner M, Sterz F, Behringer W, et al: The influence of chronic prearrest health conditions on mortality and functional neurological recovery in cardiac arrest survivors. Am J Med 1998; 104:369 –373 Bunch TJ, Kottke TE, Lopez-Jimenez F, et al: A comparative analysis of short- and longterm outcomes after ventricular fibrillation out-of-hospital cardiac arrest in patients with ischemic and nonischemic heart disease. Am J Cardiol 2006; 98:857– 860 Gonzalez MM, Berg RA, Madkarni VM, et al: Left ventricular systolic function and outcome after in-hospital cardiac arrest. Circulation 2008; 117:1864 –1872 Skhirtladze K, Birkenberg B, Mora B, et al: Cerebral desaturation during cardiac arrest: Its relation to arrest duration and left ventricular pump function. Crit Care Med 2009: 37:471– 475 Cohan SL, Mun SK, Petite J, et al: Cerebral blood flow in humans following resuscitation from cardiac arrest. Stroke 1989; 20: 761–765 Inoue Y, Shiozaki T, Irisawa T, et al: Acute cerebral blood flow variations after human cardiac arrest assessed by stable xenon enhanced computed tomography. Curr Nuerovasc Res 2007; 4:49 –54 Mullner M, Sterz F, Domanovits H, et al: Systemic and cerebral oxygen extraction after human cardiac arrest. Eur j Emerg Med 1996; 3:19 –24 Levine RL, Wayne MA, Miller CC: End-tidal carbon dioxide and outcome of out-ofhospital cardiac arrest. N Eng J Med 1997; 337:301–306 Kellum MJ, Kennedy KW, Ewy GA: Cardiocerebral resuscitation improves survival of patients with out-of-hospital cardiac arrest. Am J Med 2006; 119:335–340 Crit Care Med 2009 Vol. 37, No. 2 Vasopressin and its copilot copeptin in sepsis and septic shock* V asopressin levels increase early in septic shock because hypotension is the most potent stimulus of increased synthesis and release of vasopressin. Indeed, if an animal is challenged with conflicting signals to vasopressin release (such as hypotension and hyponatremia), the animal will increase vasopressin because hypotension is a more potent stimulus than hyponatremia is an inhibitor of vasopressin release. After the initial increase of vasopressin levels in septic shock, vasopressin levels then decline rapidly to levels that are inappropriately low (compared with hypotensive patients who have cardiogenic shock) (1–3). Thus, few endocrine systems are so rapidly activated (to increase serum levels) and then are so rapidly exhausted (such that serum levels decrease) as the vasopressin axis in sepsis. Serum levels of vasopressin—a nonapeptide—represent the interactions of the synthesis, release, and metabolism of vasopressin. Synthesis of preprovasopressin occurs in various nuclei of the hypothalamus. Subsequently, there is conversion to provasopressin followed by conversion of provasopressin by subtilisin-like proprotein convertase (SPC3) to vasopressin (4). Vasopressin is metabolized by insulin-regulated aminopeptidase also known as vasopressinase (5). Vasopressin and copeptin levels are altered in sepsis and septic shock. In this issue of Critical Care Medicine, Jochberger et al (6) report an observational cohort study designed to compare serum levels of vasopressin and copeptin of patients who had infection (n ⫽ 10), severe sepsis (n ⫽ 22), and septic shock (n ⫽ 28). Measurements of serum levels of vasopressin and copeptin were made for the first 7 days of admission. Patients with severe sepsis and septic shock had higher *See also p. 476. Key Words: copeptin; vasopressin levels; sepsis; septic shock; cardiac surgery; survival The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d117 Crit Care Med 2009 Vol. 37, No. 2 vasopressin (and copeptin) levels than patients who had infection. However, there was no difference in serum vasopressin levels between patients who had severe sepsis vs. septic shock. This is novel information that requires thoughtful interpretation. Should the levels of vasopressin have been higher in septic shock (than severe sepsis) because of the additional stimulus of hypotension in patients who had septic shock? Alternatively, does this study teach us that serum levels of vasopressin (and copeptin) of patients who have severe sepsis and septic shock are similar and so those patients should be treated similarly? Copeptin is a 39-amino acid glycopeptide that is the C terminal part of provasopressin, very similar to the C peptide of insulin (7). Jochberger et al (6) found that plasma vasopressin levels correlated significantly with copeptin levels. Is the correlation of serum levels of copeptin and vasopressin tight enough so that copeptin could be measured as a surrogate for vasopressin? Alternatively, does this study teach us that serum levels of vasopressin must be measured in patients who have severe sepsis and septic shock (because renal dysfunction was common and disturbed the tight correlation of serum vasopressin and copeptin levels)? This study extends studies of serum levels of vasopressin and copeptin. Lin et al (8) studied patients in the emergency department and found lower vasopressin levels in patients who went on to septic shock (3.6 pg/mL) compared with patients who had sepsis (10.6 pg/mL) and severe sepsis (21.8 pg/mL) (Table 1). Struck et al (9) described the rationale and methods for measurement of copeptin. Dunser et al (10) published a case report showing increased levels of copeptin and vasopressin early (first 36 hours) in septic shock, which decreased during recovery. Jochberger et al (11) found, as in other studies, critically ill patients had higher copeptin levels than controls. Patients who had cardiac surgery had higher copeptin levels than patients who had sepsis or controls. Vasopressin and copeptin levels were highly correlated. Interestingly, the ratio of copeptin to vasopressin was higher in sepsis than after cardiac surgery suggesting altered estimation of vasopressin by copeptin in patients who have sepsis. In another study, Jochberger et al (12) confirmed that critically ill patients had higher vasopressin levels (11.9 pg/mL) than controls (0.9 pg/mL). Patients who had hemodynamic dysfunction had higher vasopressin levels than patients without hemodynamic dysfunction (14.1 vs. 8.7 pg/ mL). As in previous studies, patients who had cardiac surgery had higher vasopressin (1) and copeptin (11, 12) levels than septic patients. Morgentaler et al (13) reported a sandwich assay with two antibodies for copeptin and found higher levels in critically ill patients than healthy controls (79.5 vs. 4.2 pmol/L, respectively) with good correlation between vasopressin and copeptin levels. Morgentaler et al (7) showed that hemorrhagic shock in baboons quickly lead to markedly increased serum copeptin levels (from 7.5 to 269 pM), and resuscitation was associated with declining copeptin (to 27 pM) indicating that copeptin levels are modulated profoundly and rapidly in hemorrhagic shock. Morgentaler et al (7) also evaluated critically ill patients and found increased copeptin levels in sepsis: (healthy controls 4 pM; critically ill without sepsis 27 pM; sepsis 50 pM; severe sepsis 74 pM; and septic shock 171 pM). Muller et al (14) assessed copeptin levels in 545 patients who had lower respiratory tract infection vs. 50 healthy controls. Copeptin levels were higher in patients than healthy controls and increased as severity of pneumonia increased. Both Morgentaler et al (7, 13) and Muller et al found higher levels of copeptin in nonsurvivors compared with survivors (Morgentaler et al [7], 171 vs. 87 pM; Muller et al [14], 70 vs. 24 pM). Russell et al (3) found that vasopressin levels were extremely low in severe septic shock (median 3.2 pmol/L) and increased during low-dose (0.03 U/min) vasopressin infusion to about 74 pM (6 hours) and 98 pM (24 hours). Lodha et al (15) found that serum vasopressin levels in pediatric septic shock are similar to the results of Jochberger et al (6). Vasopressin levels were increased in septic shock (116 pg/mL) and severe sepsis (106 pg/mL), and vasopressin levels did not change over 96 hours of evaluation. As in Morgentaler et al (7) and Muller et al (14), 749 Table 1. Vasopressin and copeptin levels in humans. Studies of septic shock are in bold for comparison 5. Study (Year, Reference) Landry 1997 (1) Lin 2005 (8) Jochberger 2006 (12) Morgentaler 2006 (13) Morgentaler 2007 (7) Lodha 2006 (15) Jochberger 2007 (17) Russell 2008 (3) Jochberger 2008 (6) Condition Vasopressin Levels Copeptin Levels Septic shock Cardiogenic shock Septic shock Sepsis Severe sepsis Sepsis Sepsis Critically ill (no sepsis) Sepsis Severe sepsis Septic shock Pediatric septic shock Pediatric severe sepsis Shock after cardiac surgery Multiple trauma Septic shock Septic shock plus vasopressin infusion (0.03 U/min) at 6 hours Septic shock plus vasopressin infusion (0.03 U/min) at 24 hours Infection Severe sepsis Septic shock 3.1 pg/mL 22.7 pg/mL 3.6 pg/mL 10.6 pg/mL 21.8 pg/mL 11.9 pg/mL 6. 79.5 pM 27 pM 50 pM 74 pM 171 pM 116 pg/mL 106 pg/mL 19.7 pM 44.3 pM 3.2 pM 74 pM 7. 8. 9. 98 pM 10. 3.2 pMa 6.5 pMa 6.4 pMa 25 pMa 60 pMa 70 pMa 11. a Estimated from Figures and text of Jochberger 2008 (6). nonsurvivors had higher vasopressin levels than survivors (118 vs. 76 pg/mL). Can copeptin levels be used as a surrogate for vasopressin levels? I suggest that copeptin levels cannot be used as a surrogate for vasopressin levels without further evaluation because of several aspects presented by Jochberger et al and other studies. Important considerations are that the sample size was relatively small, these are single-center studies, and the correlation between vasopressin and copeptin levels overall was adequate, but was not adequate if there was continuous veno-venous hemofiltration or renal function (statistical correlation with creatinine was p ⫽ 0.05 and with creatinine clearance was p ⫽ 0.06). Thus, I suggest that copeptin and vasopressin levels did not correlate well enough for individual patient evaluation. To make matters worse in patients who have septic shock, there is downregulation of vasopressin receptors in addition to the deficiency of vasopressin levels in sepsis 750 (16) and that further exacerbates the vasopressin deficiency of sepsis. James A. Russell, MD James Hogg iCAPTURE Centre for Cardiovascular and Pulmonary Research St. Paul’s Hospital Vancouver, BC, Canada 12. REFERENCES 15. 1. Landry DW, Levin HR, Gallant EM, et al: Vasopressin deficiency contributes to the vasodilation of septic shock. Circulation 1997; 95:1122–1125 2. Patel B, Holmes C, Russell JA, et al: Beneficial effects of short-term vasopressin infusion during severe septic shock. Anesthesiology 2002; 96:576 –582 3. Russell JA, Walley KR, Singer J, et al: Vasopressin versus norepinephrine infusion in patients with septic shock. N Engl J Med 2008; 358: 877– 887 4. Coates L, Birch N: Differential cleavage of 13. 14. 16. 17. provasopressin by the major molecular forms of SPC3. J Neurochem 1998; 70:1670 –1678 Wallis MG, Lankford MF, Keller SR: Vasopressin is a physiologic substrate for the insulinregulated aminopeptidase IRAP. Am J Physiol Endocrinol Metab 2007; 293:E1092–E1102 Jochberger S, Dörler J, Luckner G, et al: The vasopressin and copeptin response to infection, severe sepsis, and septic shock. Crit Care Med 2009; 37:476 – 482 Morgentaler NG, Muller B, Struck J, et al: Copeptin, a stable peptide of the arginine vasopressin precursor, is elevated in hemorrhagic and septic shock. Shock 2007; 28: 219 –226 Lin IY, Ma HP, Lin AC, et al: Low vasopressin/norepinephrine ratio predicts septic shock. Am J Emerg Med 2005; 23:718 –724 Struck J, Morgentaler NG, Bergmann A: Copeptin, a stable peptide derived from the vasopressin precursor, is elevated in serum of sepsis patients. Peptides 2005; 26:2500 –2504 Dunser M, Mayr VD, Wenzel V, et al: Course of vasopressin and copeptin plasma concentrations in a patient with severe septic shock. Anaesth Intensive Care 2006; 34:498–500 Jochberger J, Morgentaler NG, Mayr VD, et al: Copeptin and arginine vasopressin concentrations in critically ill patients. J Clin Endocrinol Metab 2006; 91:4381– 4386 Jochberger S, Mayr V, Luckner G, et al: Serum vasopressin concentrations in critically ill patients. Crit Care Med 2006; 34:293–299 Morgentaler S, Struck J, Alonso C, et al: Assay for the measurement of copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem 2006; 52:112–119 Muller B, Morgentaler NG, Stolz P, et al: Circulating levels of copeptin, a novel biomarker, in lower respiratory tract infections. Eur J Clin Invest 2007; 37:145–152 Lodha R, Vivekanandhan S, Sarthi M, et al: Serial circulating vasopressin levels in children with septic shock. Pediatr Crit Care Med 2006; 7:220 –224 Schmidt C, Hocherl K, Kurt B, et al: Role of nuclear factor-kappaB-dependent induction of cytokines in the regulation of vasopressin V1A-receptors during cecal ligation and puncture-induced circulatory failure. Crit Care Med 2008; 36:2363–2372 Jochberger J, Mayr VD, Luckner G, et al: Vasopressin plasma concentrations in post-cardiotomy shock: A prospective, controlled trial. Intensive Care Med 2007; 33(Suppl 2): A0763 Crit Care Med 2009 Vol. 37, No. 2 Routine nursing procedures—Take care of the patient and the splanchnic circulation!* C are of the critically ill not only comprises invasive monitoring, complex pharmacologic intervention, and eventually organ replacement, but also requires careful nursing of the patient. Routine nursing procedures consist of mechanical ventilation–associated procedures, e.g., endotracheal suctioning; hygiene measures, e.g., oral care and washing; diagnostics, e.g., chest radiograph and physical examinations; removal/insertion of catheters; and physical treatment, e.g., physiotherapy and patient turning. It has been well established for decades that many of these routine nursing procedures may affect patient homoeostasis (1– 4), in particular, when associated with endotracheal suctioning in mechanically ventilated patients (5–7). In fact, nearly five decades ago, the group of Landmesser already reported a marked fall in arterial oxygen saturation during and after endotracheal suctioning, which could only be partly prevented by preoxygenation with an FIO2 of 1.0 (8). Increased oxygen consumption and CO2 production resulting from enhanced sympathetic tone (9 –11) clearly assumes major importance in this context as a result of increased respiratory muscle activity because of agitation or coughing episodes. In fact, sedation and analgesia with shortacting narcotics markedly attenuated the hemodynamic response to routine nursing procedures (11, 12). Despite this obvious and potentially important role of nursing procedures for patient stability, clinical data dealing with this phenomenon are scarce, and up to, now no study is available on the effects of such procedures on hepatosplanchnic perfusion, a *See also p. 483. Key Words: endotracheal suctioning; chest physiotherapy; sympathetic tone; hepatic venous oxygen saturation The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d157 Crit Care Med 2009 Vol. 37, No. 2 region which has been identified as a high-risk area in the pathogenesis of multiple organ failure, in particular, in the presence of a mismatch between metabolic demands and oxygen supply (13). In the current issue of Critical Care Medicine, Jakob et al (14) present a study exploring the role of routine nursing procedures for decreases of hepatic venous oxygen saturation, a well-accepted marker of splanchnic perfusion and metabolism, and the relationship of these desaturation episodes and outcome. In 36 patients with acute respiratory or circulatory failure, the majority suffering from septic shock (59%), mixed and hepatic venous oxygen saturations were continuously recorded during various nursing procedures that were part of the routine care of the patients during their intensive care unit stay. The main finding was that patients are repeatedly exposed to episodes of hepatic venous desaturation— and thereby possibly to reduced splanchnic perfusion— during routine nursing procedures, and that these episodes were mostly mirrored, in part, only by changes in peripheral arterial or mixed venous oxygen saturation. In addition, the total number of desaturation episodes, regardless of their association with procedures, were directly related to the maximal Sequential Organ Failure Assessment score, but not with length of stay or mortality. What do we learn from this study? Jakob et al confirm the well-established fact that regional, i.e., hepatic, venous oxygen saturation cannot necessarily be derived from the arterial or mixed venous one, in particular in the presence of regional oxygen supply demand dependency (13). Furthermore, most desaturation episodes were due to ventilator-related procedures, and endotracheal suctioning caused the most pronounced fall in hepatic venous oxygen saturation and increase in the hepatic-mixed venous oxygen saturation gradient. As mentioned eariler, endotracheal suctioning is well known to impair pulmonary gas exchange, and given the high proportion of patients with septic shock, it is not surprising that endotracheal suctioning caused the largest decrease in hepatic venous oxygen saturation: these patients often present with a high hepatosplanchnic oxygen extraction, and even a small change in oxygen demand and supply may cause large decreases in regional venous saturation (13). Unfortunately, Jakob et al did not mention whether they used measures to counteract any endotracheal suctioning–induced impairment of pulmonary gas exchange, i.e., application of hyperoxia (6, 8) or alveolar recruitment maneuvers (15), or whether they performed open or closed suction (16, 17). The latter was demonstrated to attenuate the suctioninduced fall in arterial oxygen saturation, albeit its benefit for morbidity remains a matter of debate (18). A very interesting finding of the study by Jakob et al is that more than one quarter (27%) of all observed hepatic venous desaturation episodes occurred independently of any nursing procedure. Although only marginally discussed by the authors, this finding may nevertheless highlight the importance of spontaneous physiologic fluctuations such as heart rate and blood pressure variability (19). As a consequence, regional oxygen saturations are also likely to show spontaneous variations, and this phenomenon may have contributed to the authors’ findings. It should be noted, however, that the degree of heart rate variability, the currently best studied parameter of spontaneous physiologic variation, was inversely related to outcome in several subpopulations of intensive care unit patients (20 –22). By contrast, albeit Jakob et al did not analyze the relation between spontaneous drops of hepatic venous oxygen saturation and the Sequential Organ Failure Assessment score, the authors report that the total— and not only of procedure associated— number of regional venous desaturation episodes was directly related to the severity of the disease. As a logic consequence, they suggest that hepatic venous desaturation 751 episodes, hence, mirror the patients’ compromised capacity if not inability to increase blood flow in response to a rise in regional oxygen demand. In conclusion, Jakob et al elegantly demonstrate that routine nursing procedures may deleteriously affect the hepatosplanchnic circulation and, consequently, visceral organ function. Furthermore, most of it will remain undetected by systemic monitoring parameters, because splanchnic perfusion can only be detected by invasive procedures such as hepatic vein catheterization. Because more than one quarter of all recorded hepatic venous desaturations were not procedure related, the clinical relevance of these findings remains open. Nevertheless, in analogy to the sepsis bundles, nursing procedure guidelines are probably needed to reduce the impact of these measures on patient outcome, and Jakob et al have the merit to highlight this often underestimated issue of critical care medicine. Hendrik Bracht, MD Florian Wagner, MD Sektion Anästhesiologische Pathophysiologie und Verfahrensentwicklung Universitätsklinikum Ulm, Germany Universitätsklinik für Anästhesiologie Universitätsklinikum Ulm, Germany Rainer Meierhenrich, MD, PhD Universitätsklinik für Anästhesiologie Universitätsklinikum Ulm, Germany Peter Radermacher, MD, PhD Sektion Anästhesiologische Pathophysiologie und Verfahrensentwicklung Universitätsklinikum 752 Ulm, Germany Michael Georgieff, MD, PhD Universitätsklinik für Anästhesiologie Universitätsklinikum Ulm, Germany REFERENCES 1. Winslow EH, Clark AP, Whiote KM, et al: Effects of a lateral turn on mixed venous saturation and heart rate in critically ill adults. Heart Lung 1990; 19:557–561 2. Tyler DO, Winslow EH, Clark AP, et al: Effects of a 1-minute back rub on mixed venous oxygen saturation and heart rate in critically ill patients. Heart Lung 1990; 19:562–565 3. Tidwell SL, Ryan WJ, Osguthorpe SG, et al: Effects of position changes on mixed venous oxygen saturation in patients after coronary revascularization. Heart Lung 1990; 19: 574 –578 4. Barker M, Adams S: An evaluation of a single chest physiotherapy treatment on mechanically ventilated patients with acute lung injury. Physiother Res Int 2002; 7:157–169 5. Walsh JM, Vanderwarf C, Hoscheit D, et al: Unsuspected hemodynamic alterations during endotracheal suctioning. Chest 1989; 95: 162–165 6. Clark AP, Winslow EH, Tyler DO, et al: Effects of endotracheal suctioning on mixed venous oxygen saturation and heart rate in critically ill adults. Heart Lung 1990; 19: 552–557 7. Kinloch D: Instillation of normal saline during endotracheal suctioning. Effects on mixed venous oxygen saturation. Am J Crit Care 1999; 8:231–240 8. Boba A, Cincotti JJ, Piazza TE, et al: The effects of apnea, endotracheal suction, and oxygen insufflation, alone and in combination, upon arterial oxygen saturation in anesthetized patients. J Clin Lab Med 1959; 53: 680 – 685 9. Weissman C, Kemper M, Damask MC, et al: Effect of routine intensive care interactions on metabolic rate. Chest 1984; 86:815– 818 10. McCulloch KM, Ji SA, Raju TNK: Skin blood flow changes during routine nursery procedures. Early Hum Dev 1995; 41:147–156 11. Gemma M, Tommasino C, Cerri M, et al: Intracranial effects of endotracheal suctioning in the acute phase of head injury. J Neurosurg Anesth 2002; 14:50 –54 12. Klein P, Kemper M, Weissman C, et al: Attenuation of the hemodynamic responses to chest physical therapy. Chest 1988; 93: 38 – 42 13. De Backer D, Creteur J, Noordally O, et al: Does hepato-splanchnic VO2/DO2 dependency exist in critically ill septic patients? Am J Respir Crit Care Med 1998; 157: 1219 –1225 14. Jakob SM, Parviainen I, Ruokonen E, et al: Increased splanchnic oxygen extraction because of routine nursing procedures. Crit Care Med 2009; 37: 483– 489 15. Maggiore SM, Lellouche F, Pigeot J, et al: Prevention of endotracheal suctioninginduced alveolar drecruitment in acute lung injury. Am J Respir Crit Care Med 2003; 167:1215–1224 16. Johnson KL, Kearney PA, Johnson SB, et al: Closed versus open endotracheal suctioning: Costs and physiologic consequences. Crit Care Med 1994; 22:658 – 666 17. Cereda M, Villa F, Colombo E, et al: Closed system endotracheal suctioning maintains lung volume during volume-controlled mechanical ventilation. Intensive Care Med 2001; 27:648 – 654 18. Jongerden IP, Rovers MM, Grypdonck MH, et al: Open and closed endotracheal suction systems in mechanically ventilated intensive care patients: A meta-analysis. Crit Care Med 2007; 35:260 –270 19. Buchman TG, Stein PK, Goldstein B: Heart rate variability in critical illness and critical care. Curr Opin Crit Care 2002; 8:311–315 20. Goldstein B, Fiser DH, Kelly MM, et al: Decomplexification in critical illness and injury: Relationship between heart rate variability, severity of illness, and outcome. Crit Care Med 1998; 26:352–357 21. Toweill D, Sonnenthal K, Kimberly B, et al: Linear and nonlinear analysis of hemodynamic signals during sepsis and septic shock. Crit Care Med 2000; 28:2051–2057 22. Korach M, Sharshar T, Jarrin I, et al: Cardiac variability in critically ill adults: Influence of sepsis. Crit Care Med 2001; 29:1380 –1385 Crit Care Med 2009 Vol. 37, No. 2 Video instruction for dispatch-assisted cardiopulmonary resuscitation: Two steps forward and one step back!* S urvival from cardiac arrests is dependent on the alignment of the steps outlined in the revised chain of survival: early recognition and call for help, early cardiopulmonary resuscitation (CPR), early defibrillation, and early postresuscitation care (1). Increasing the proportion of the community, willing to provide bystander CPR, would certainly facilitate “early CPR.” Much attention has been recently paid to bystander CPR: its importance and ways of reducing the barriers to its performance (2). This has been accompanied by the reiteration by the American Heart Association that hands-only (compression-only) CPR is an acceptable alternative when bystanders are untrained, unable, or unwilling to provide ventilation as well as compressions (3). Compression-only CPR appears particularly pertinent when applied to dispatch-assisted CPR. The ability to provide instructions over the phone to a bystander is a widely established and effective practice (2, 4), especially if the instructions are simplified (such as removing the additional instructions for ventilation [5, 6]). If the advancing technology associated with telecommunication could be translated into a more sophisticated tool for instruction, then dispatch-assisted bystander CPR could be taken to a new level. Reports of some potential benefits and limitations of such technology are starting to appear (7–9). In this issue of Critical Care Medicine, Yang et al (10) present the results from another such study. *See also p. 490. Key Words: cardiopulmonary resuscitation; dispatch-assist cardiopulmonary resuscitation; bystander cardiopulmonary resuscitation; quality of cardiopulmonary resuscitation; video instruction The author has consulted for AHA and is an Evidence Evacuation Expert for AHA. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d2e1 Crit Care Med 2009 Vol. 37, No. 2 In a randomized study, the authors evaluated the potential benefits of using interactive video instruction to improve the quality of dispatch-assisted, compression-only CPR performed on a manikin by “bystanders” who had only limited (if any) training in CPR. They used an optimally placed video cell phone in hands-free mode and provided either voice-only instruction or “interactive voice and video demonstration and feedback.” The addition of video communication (and, in particular, the realtime feedback) improved the rate of chest compressions (from a median of 63 to 95.5 compressions per minute) and the depth of compressions (from a median of 25 to 36 mm). The published literature suggests that an increase in at least short-term survival would be expected from such improvements in compression rate (11) and depth (12, 13). The downside of the interactive video group was a near 30-second delay in the time till first compression, but given the fact that a median of 0% of the compressions in the control group was of an “appropriate” depth (38 –51 mm [1.5–2 inches]), and that increased to a median of 20% in the interactive video group, this delay becomes much less important. This study provides many messages. Clearly, the ideal positioning of the video cell phone is crucial, and improvements could be made to the instructions and the time required to deliver them. As the technology, coverage, and expertise with these devices are enhanced, we can expect further improvements in their effectiveness. In summary, the study published by Yang et al (10) has many limitations and is obviously only a starting point for the incorporation of some of the more advanced functions of the modern-day communication devices. It appears that more sophisticated interactions (such as the use of video instruction) can improve some aspects of the quality of dispatch-assisted chest compressions (depth and rate), but at this stage, it comes at a cost (delayed commencement of compressions). It may be considered two steps forward and one step back, but at least we appear to be making progress. Peter Morley, MBBS, FRACP, FANZCA, FJFICM Royal Melbourne Hospital University of Melbourne Victoria, Australia REFERENCES 1. Nolan J, Soar J, Eikeland H: The chain of survival. Resuscitation 2006; 71:270 –271 2. Abella BS, Aufderheide TP, Eigel B, et al: Reducing barriers for implementation of bystanderinitiated cardiopulmonary resuscitation: A scientific statement from the American Heart Association for healthcare providers, policymakers, and community leaders regarding the effectiveness of cardiopulmonary resuscitation. Circulation 2008; 117:704–709 3. Sayre MR, Berg RA, Cave DM, et al: Handsonly (compression-only) cardiopulmonary resuscitation: A call to action for bystander response to adults who experience out-ofhospital sudden cardiac arrest: A science advisory for the public from the American Heart Association Emergency Cardiovascular Care Committee. Circulation 2008; 117: 2162–2167 4. Vaillancourt C, Stiell IG, Wells GA: Understanding and improving low bystander CPR rates: A systematic review of the literature. CJEM 2008; 10:51– 65 5. Hallstrom A, Cobb L, Johnson E, et al: Cardiopulmonary resuscitation by chest compression alone or with mouth-to-mouth ventilation. N Engl J Med 2000; 342:1546 –1553 6. Roppolo LP, Pepe PE, Cimon N, et al: Modified cardiopulmonary resuscitation (CPR) instruction protocols for emergency medical dispatchers: Rationale and recommendations. Resuscitation 2005; 65:203–210 7. You JS, Park S, Chung SP, et al: The era of audiovisual dispatch using cellular phones with video telephony. Resuscitation 2008; 76:486 – 487 8. Johnsen E, Bolle SR: To see or not to see— Better dispatcher-assisted CPR with videocalls? A qualitative study based on simulated trials. Resuscitation 2008; 78:320 –326 9. Yang CW, Wang HC, Chiang WC, et al: Impact of adding video communication to dispatch instructions on the quality of rescue breathing in simulated cardiac arrests—A 753 randomized controlled study. Resuscitation 2008; 78:327–332 10. Yang C-W, Wang H-C, Chiang W-C, et al: Interactive video instruction improves the quality of dispatcher-assisted chest compression-only cardiopulmonary resuscitation in simulated cardiac arrests. Crit Care Med 2009; 37: 490–495 11. Abella BS, Sandbo N, Vassilatos P, et al: Chest compression rates during cardiopulmonary resuscitation are suboptimal: A prospective study during in-hospital cardiac arrest. Circulation 2005; 111:428 – 434 12. Edelson DP, Abella BS, Kramer-Johansen J, et al: Effects of compression depth and pre-shock pauses predict defibrillation failure during cardiac arrest. Resuscitation 2006; 71:137–145 13. Kramer-Johansen J, Myklebust H, Wik L, et al: Quality of out-of-hospital cardiopulmonary resuscitation with real time automated feedback: A prospective interventional study. Resuscitation 2006; 71:283–292 Continuous renal replacement therapy circuit contamination: New tale of an old problem?* D ialysis and continuous renal replacement therapies’ (CRRT) technology improved tremendously since the first hemodialysis for chronic renal failure in 1960 and the first CRRT in 1977 (1, 2). As technical challenges of dialysis treatment were overcome and patients began to survive, problems of preparing dialysate from tap water became apparent. Chemical and bacterial contamination of the water used to prepare the dialysate led to the developments of strict standards for water used for hemodialysis by various organizations, with slight differences (3). Although CRRT is the main form of renal support provided in intensive care units worldwide (4), widely recognized standards have not been applied to CRRT as yet. Microbial contamination of intravenous (IV) fluids has been known for long. Episodes of septicemia associated with IV therapy were reported in the 1970s, which were linked to manufactured infusion products; however, extrinsic sources of contamination during use were also identified in a significant proportion of IV therapy infusion systems (5) and are considered to be a more common source of contamination (5–7). Guidelines for the prevention of intravascular catheterrelated infections address how to reduce the microbial contamination of IV infusions and potential hazards to the patients (6). *See also p. 496. Key Words: continuous renal replacement therapy; microbiological; contamination; biofilm; dialysate; intravenous; infusion systems The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d485 754 Furthermore, significant advances during the same time periods led to guidelines for surgical site infection prophylaxis and operating room conditions (8). The source of pathogens in many surgical site infections is the endogenous flora of the patient’s skin, mucous membranes, or hollow viscera; however, exogenous sources include surgical personnel, the operating room environment (including air), instruments, and other materials brought to the sterile field during an operation. Room air may contain microbial-laden dust, lint, skin squames, or respiratory droplets that can also contribute to contamination of IV infusion equipment during use (8). In this issue of Critical Care Medicine, Kanagasundaram et al (9) look at the microbiological integrity of a continuous venovenous hemofiltration (a form of CRRT) circuit using a commercially available replacement solution. They found that of the 24 replacement fluid cultures, 9 breached European Pharmacopoeia standards for ultrapure water. One of 24 endotoxin measurements breached European Pharmacopoeia standards. Internal tubing cultures were negative, but electron microscopy revealed 13 of the 24 collected tubing samples to be contaminated with biofilm. Only 7 of the 24 studied circuits proved to be free from microbial contamination. They found no clear relationship between circuit lifespan and rates of contamination, but this should be interpreted with caution because of their small sample size. Kanagasundaram et al used highly sensitive microbiological techniques to examine replacement fluid, quantitatively assess endotoxin, and examine tubing specimens via electron microscopy for biofilms. They took precautions to mini- mize contamination during sampling; however, they did not separately analyze unused tubing or commercially available solutions directly from the bags before use or other bags of unused fluid with the same lot numbers. They also did not evaluate room air as a potential source of microbial contamination (9). CRRT systems involve a roller blood pump device with additional pumps to provide the circulation of dialysate and replacement fluids, a dialysis filter, dialysate and replacement fluids, and the necessary infusion tubing systems, all of which are usually incorporated in a disposable cartridge or assembled separately depending on the device used (2). The continuous venovenous hemofiltration system that Kanagasundaram et al used was an integrated, single-use circuitry with a commercially available replacement fluid (as usual in the current practice). This CRRT system is essentially a combination of a dialysis filter and a needleless infusion system. Thus, their findings should be interpreted in the same context. Indeed, their results are not surprising and are reminiscent of data on “in use contamination of IV fluids” (5). There are four potential routes of access for the organisms to colonize an intravascular device and subsequently cause blood stream infections: invasion of the skin insertion site, contamination of the catheter hub, hematogenous spread from a distant site of infection, or rarely infusion of contaminated fluid through the device. The first two sources are the most important. Surface colonization of the tubing, probably originating from the skin, predominates in short-term catheters (in place less than 10 days), such as peripheral IV catheters, nontunneled central venous catheters, and arterial catheters. InCrit Care Med 2009 Vol. 37, No. 2 traluminal colonization due to hub contamination increases over time and becomes predominant after 30 days of long-term devices (tunneled central venous catheters, peripherally inserted central catheters, and subcutaneous ports) (6, 7, 10). Stopcocks (used for injection of medications, administration of IV infusions, and collection of blood samples) represent a potential portal of entry for microorganisms into vascular access catheters and IV fluids (6). Catheter material/structure and virulence of microorganisms are important factors along with biofilm formation (6, 7, 10 –12). Biofilm is a living community of microorganisms, embedded in their extracellular polymeric matrix and other components deposited from bodily fluids adherent to the catheter surface. Biofilm can develop within a few hours of device placement (both inside and outside the lumen in the case of a catheter). Biofilms have major medical significance. Biofilms are less susceptible to the effects of antimicrobial agents and host defenses, making the treatment of infections very difficult (7, 10 –12). Where do we go from here? Can we blame the dialysate/replacement fluids for the contamination? This can only be answered by specifically culturing and assaying these fluids before, and during, inline use. More important, we should implement all preventive measures as recommended by current guidelines to prevent catheter-related infections and surgical site infections first. Educating all personnel involved is critical. Achieving operating conditions in intensive care unit is not feasible, but maximum protection during CRRT set up and disconnection of the Crit Care Med 2009 Vol. 37, No. 2 circuit can be used by nursing staff. We may need to use new antiseptic barrier caps with 2% chlorhexidine gluconate in alcohol rather than routine alcohol swabs for disinfection of needleless catheter connectors and access ports (13). We should begin culturing CRRT fluids when the patients develop fever or evidence of sepsis. Because biofilms can form rapidly on the intraluminal surfaces of IV tubing, we should probably adopt shorter changing periods (i.e., 24 hours) for the CRRT circuit filter and tubing, especially because their cost decreased over time, and maintaining patency is also an issue. For the same token, we should perhaps begin routine use of catheter lock solutions with antibiotics and chelating agents that are promising in eliminating biofilms in the catheter lumens (7, 10, 14). Clearly, more work needs to be done in this area, but this is a step forward. 4. 5. 6. 7. 8. 9. ACKNOWLEDGMENTS I thank Dr. Matthew Levison for his valuable input and discussions about the topic. Bulent Cuhaci, MD Department of Medicine, Drexel University College of Medicine Philadelphia, PA REFERENCES 1. Blagg CR: The early history of dialysis for chronic renal failure in the United States: A view from Seattle. Am J Kidney Dis 2007; 49:482– 496 2. Dirkes SM: Continuous renal replacement therapy: Dialytic therapy for acute renal failure in intensive care. Nephrol Nurs J 2000; 27: 581–592 3. Ward R: Worldwide water standards for he- 10. 11. 12. 13. 14. modialysis. Haemodialysis Int 2007; 11: S18 –S25 Uchino S, Bellomo R, Morimatsu H, et al: Continuous renal replacement therapy: A worldwide practice survey. Intensive Care Med 2007; 33:1563–1570 Maki DG, Anderson RL, Shulman JA: In-Use contamination of intravenous infusion fluid. Appl Microbiol 1974; 28:778 –784 Centers for Disease Control and Prevention. Guidelines for the prevention of intravascular catheter-related infections. MMWR Morb Mortal Wkly Rep 2002; 51(RR-10): 1–32 Trautner BW, Darouiche RO: Catheterassociated infections. Pathogenesis affects prevention. Arch Intern Med 2004; 164: 842– 850 Mangram AJ, Horan TC, Pearson ML: Guideline for prevention of surgical site infection, 1999. The Hospital Infection Control Practices Advisory Committee. Infect control Hosp Epidemiol 1999; 20:247–278 Moore I, Bhat R, Hoenich NA, et al: A microbiological survey of bicarbonate-based replacement circuits in continuous venovenous hemofiltration. Crit Care Med 2009; 37:496 –500 Sullivan R, Samuel V, Le C, et al: Hemodialysis vascular catheter-related bacteremia. Am J Med Sci 2007; 334:458 – 465 Passerini L, Lan K, Costerton J, et al: Biofilms on indwelling vascular catheters. Crit Care Med 1992; 20:665– 673 Dolan RM: Biofilm: Microbial life on surfaces. Emerging infectious diseases. 2002; 8:881– 890 Menyhay SZ, Maki DG: Disinfection of needleless catheter connectors and access ports with alcohol may not prevent microbial entry: The promise of a novel antisepticbarrier cap. Infect Control Hosp Epidemiol 2006; 27:23–27 Raad II, Fang X, Keutgen XM, et al: The role of chelators in preventing biofilm formation and catheter-related bloodstream infections. Curr Opin Infect Dis 2008; 21:385–392 755 Arginine pharmacokinetics: Not a new paradigm but the old pharmacology* I n the last 15 yrs, an increasing number of articles have dealt with the hypothesis that critically ill patients in the catabolic state will develop nutrient deficiencies and immune dysfunction. They usually show low plasmatic levels of glutamine and arginine despite adequate nutritional support and are considered conditionally essential in this setting (1, 2). Both have been empirically added, alone or in combination with other substrates, to standard diets trying to demonstrate their beneficial effects on laboratory, immunologic, and clinical parameters in comparison with standard diets. These formulas have been included in the wide concept of immunonutrition with variable success in terms of clinical outcomes but with entertained controversies (3–5). Furthermore, this concept has been questioned, and now the new paradigm of pharmaconutrition has been introduced (6). In fact, what we must do is rethink the research we are doing and, if we will use arginine or glutamine like a drug, we are urged to apply the principles of the pharmacology to get accurate and reproducible results. In this issue of Critical Care Medicine, Loi et al (7) present a good example of pharmacologic study. They administered two different isocaloric and isonitrogenous diets, one standard and the other enriched with arginine, to observe the changes in plasmatic levels of glutamine in the first 90 minutes of diet administration, once the patients were fed enterally and with a washout period of 3 hours. The authors found a significant increase of plasmatic levels of arginine and glutamine in patients treated with the arginine-containing diet. Furthermore, the areas under the curve between arginine *See also p. 501. Key Words: arginine; glutamine; pharmacokinetics; protein metabolism The author has received honoraria from Fresenius Rabl 900 and Novariis 600. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d522 756 and proline and arginine and ornithine showed a good correlation. Interestingly, both groups of patients had similar nitrogen balance or proteolysis. This study confirms in humans that there is a common fate between arginine and glutamine through ornithine in the liver, as the authors suggested in a previous experimental study (8). Nevertheless, this study has some limitations. First, the study diet contains high amounts of omega-3 fatty acids, vitamins C and E, and selenium. We do not know if changes in interleukin production mediated for these fatty acids or a higher level of antioxidants can modify glutamine consumption by immune cells or for glutathione production. Besides, the washout period seems to be very short and a longer fasting period like 8 or 12 hours will be a better approach (9). Second, the study population has in fact two subgroups: one group of patients with cancer underwent an elective surgical procedure and another group of patients with septic shock and pancreatitis underwent an emergent surgical procedure. Although the interaction between the two groups of patients is not significant, it makes the overall population heterogeneous enough to extrapolate these results to different populations without a separate analysis of the data. Nevertheless, Loi et al present a solid pharmacokinetic study that demonstrates that arginine delivered enterally increases the levels of plasmatic glutamine. Arginine and glutamine regulate the immune response through different effects (10, 11). Furthermore, glutamine and arginine share complex metabolic pathways well explained in a recent review of this journal (12). Arginine is metabolized by arginase and produces urea, and ornithine and the last one gives citrulline or glutamine and proline, as demonstrated by Loi et al. Arginine is also transformed to citrulline and nitric oxide by inducible nitric oxide synthase. The temporal switch of arginine as a substrate for the inducible nitric oxide synthase to ornithine–proline axis is regulated by inflam- matory cytokines and by the arginine metabolites themselves (13). On the other hand, glutamine is a primary fuel for enterocytes and for gluconeogenesis in the liver, is a precursor of pyrimidine and purine, is incorporated to glutathione, and serves as a precursor for the de novo production of arginine through the citrulline–arginine intestinal–renal pathway (14). Close to the complexity of these metabolic pathways, we should consider gene expression of different isoenzymes, different patterns of tissue expression, and the route of administration. Glutamine seems to be beneficial in critically ill patients, whereas arginine supplementation is still under debate. Administering high doses of these amino acids needs further research looking at the dose, the way of administration, and the physiologic and clinical effects like other drugs. Maybe this is an unrealistic approach, but it is clear that both amino acids exert regulatory functions different from being only nutrients. And finally, if arginine can produce an excessive and sometimes deleterious amount of nitric oxide, and therefore is an adverse event, why not use glutamine alone? Teodoro Grau, MD, PhD Servicio de Medicina Intensiva Hospital Universitario Doce de Octubre Madrid, Spain REFERENCES 1. Lacey JM, Wilmore DW: Is glutamine a conditionally essential amino acid? Nutr Rev 1990; 48:297–309 2. Luiking YL, Poeze M, Dejong CH, et al: Sepsis: An arginine deficiency state. Crit Care Med 2004; 32:2135–2145 3. Heyland DK, Novak F, Drover JW, et al: Should immunonutrition become routine in critically ill patients? JAMA 2001; 286:22–29 4. Consensus recommendations from the US summit on immune-enhancing enteral therapy. JPEN J Parenter Enteral Nutr 2001; 25(Suppl):S61–S62 5. Montejo JC, Zarazaga A, Lopez-Martinez J, et al; for the Spanish Society of Intensive Care Medicine and Coronary Units: Immunonutrition in the intensive care unit. A systematic Crit Care Med 2009 Vol. 37, No. 2 review and consensus statement. Clin Nutr 2003; 22:221–233 6. Jones NE, Heyland DK: Pharmaconutrition: A new emerging paradigm. Curr Opin Gastroenterol 2008; 24:215–222 7. Löi C, Zazzo J-F, Delpierre E, et al: Increasing plasma glutamine in postoperative patients fed an arginine-rich immune-enhancing diet—A pharmacokinetic randomized controlled study. Crit Care Med 2009; 37:501–509 8. Moynard C, Belabed L, Gupta S, et al: Arginine-enriched diet limits plasma and muscle depletion in head-injured rats. Nutrition 2006; 22:1039 –1044 9. Jackson NC, Carroll PV, Russell-Jones DL, et al: The metabolic consequences of critical illness: Acute effects on glutamine and protein metabolism. Am J Physiol 1999; 276:E163–E170 10. Melis GC, ter Wengel N, Boelens PG, et al: Glutamine: Recent developments in research on the clinical significance of glutamine. Curr Opin Clin Nutr Metab Care 2004; 7:59 –70 11. Popovic PJ, Zeh JH III, Ochoa JB: Arginine and immunity. J Nutr 2007; 137:1681S–1686S 12. Vermeulen MAR, van de Poll MCG, LighartMelis G, et al: Specific amino acids in the critically ill patient-exogenous glutamine/ arginine: A common denominator. Crit Care Med 2007; 35(Suppl):S568 –S576 13. Satriano J: Arginine pathways and the inflammatory response: Interregulation of nitric oxide and polyamines [review]. Amino Acids 2004; 26:321–329 14. Neu J, DeMarco V, Li N: Glutamine: Clinical applications and mechanisms of action. Curr Opin Clin Nutr Metab Care 2002; 5:69 –75 Meaning of pulse pressure variation during cardiac surgery: Everything is open* F luid management is of major importance in critically ill patients and in surgery patients. Static preload indicators, like cardiac filling pressures, are of poor value to assess fluid responsiveness (1–3). Functional hemodynamic parameters, like arterial pulse pressure variation (PPV) and pulse contour stroke volume variation (SVV), have gained wide popularity as predictors of fluid responsiveness in mechanically ventilated patients (4). It must be remembered that the mechanical insufflation decreases preload and increases afterload of the right ventricle. The right ventricular preload reduction is secondary to the decrease of the venous return pressure gradient related to the inspiratory increase in intrathoracic pressure. The increase in right ventricular afterload is related to the inspiratory increase in transpulmonary pressure (alveolar pressure minus intrathoracic pressure). The reduction in preload and the increase in afterload of the right ventricle both lead to a decrease in right ventricular stroke volume, which, in turn, leads to a decrease in left ventricular filling after a phase lag of 2–3 heart beats related to the long blood pulmonary transit time. The resulting left ventricular preload re- *See also p. 510. Key Words: open chest surgery; cardiac surgery; pulse pressure variation; stroke volume variation; fluid responsiveness; cardiac preload The author has consulted for Pulsion Medical Systems. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d575 Crit Care Med 2009 Vol. 37, No. 2 duction may induce a decrease in the left ventricular stroke volume, which is minimal during the expiratory period. Interestingly, the cyclic changes in right ventricular preload induced by mechanical ventilation, result in greater cyclic changes in right ventricular stroke volume when the right ventricle operates on the steep rather than on the flat portion of the Frank–Starling curve. The cyclic changes in right ventricular stroke volume and, hence, in left ventricular preload result in greater cyclic changes in left ventricular stroke volume when the left ventricle operates on the steep portion of the Frank–Starling curve. Thus, the magnitude of the respiratory changes in left ventricular stroke volume should be an indicator of biventricular preload responsiveness and, hence, of fluid responsiveness. In this regard, PPV (5–7) and SVV (7–9) have been demonstrated as good predictors of fluid responsiveness in various clinical settings. Two obvious conditions are required to adequetely interpret such dynamic parameters: controlled positive pressure ventilation with no spontaneous breathing activity (10, 11) and absence of cardiac arrhythmias (10). Because changes in intrathoracic pressure and transpulmonary pressure are the major determinants of the cyclic cardiovascular consequences of positive pressure ventilation, a sufficiently hightidal volume (⬎7 mL/kg) is obviously also required for well interpreting functional dynamic parameters (12). These three conditions are generally met in anesthetized patients undergoing surgery. Nevertheless, under open chest conditions, respiratory changes in intrathoracic pressure are less pronounced so that respiratory changes in stroke volume or in pulse pressure are expected to be low even in fluid responsive patients. In this issue of Critical Care Medicine, de Waal et al (13) have tested this hypothesis by evaluating the ability of PPV and SVV to predict fluid responsiveness under both open and closed chest conditions in patients undergoing coronary artery bypass graft surgery. First, they confirmed that in closed chest conditions (tidal volume: 8 mL/kg), PPV and SVV were better predictors of fluid responsiveness than static measures of preload, such as central venous pressure or global end-diastolic volume. Second, they showed that neither PPV and SVV nor static preload indicators were able to predict accurately fluid responsiveness under open chest conditions, although the number of nonresponders was very low (3 of 18 patients). In half of the fluid responders, PPV was ⬍10%, a value in agreement with the authors’ hypothesis (13). It is likely that the attenuation of cyclic changes in intrathoracic pressure in open chest conditions accounted for these findings. It must be stressed, however, that in the other half of the fluid responders, PPV was higher than 11% (up to 22%). Surprisingly, the authors did not highlight these striking results. One could postulate that under open chest conditions, a high PPV (or SVV) is secondary to the cyclic changes in transpulmonary pressure, which are potentially large in the absence of significant changes in intrathoracic pressure. A marked inspiratory increase in transpulmonary pressure could decrease right ventricular stroke 757 volume at inspiration through an increase in the resistance of intra-alveolar microvessels. Thus, even if the pulmonary vasculature was in West’s zone 3 conditions at expiration in these patients, the marked increase in transpulmonary pressure had probably produced zone 2 conditions at inspiration in some of them (14). It could then be postulated that fluid infusion by increasing the pulmonary venous pressure attenuated the transfer from zone 3 to zone 2 at inspiration and, hence, the inspiratory decrease in right ventricular stroke volume. Although this hypothesis is speculative, it is supported by the fact that PPV (and SVV) decreased while stroke volume increased after fluid infusion in this subgroup of patients. Similar findings were previously reported in open chest surgery patients ventilated with a tidal volume of 10 mL/kg (15). Although they enrolled a limited number of patients, de Waal et al (13) have thus brought evidence that the presence of high PPV (or SVV) is indicative of fluid responsiveness under both closed and open chest conditions. However, under open chest conditions, other tools are still required to diagnose the origin of hemodynamic instability because the presence of low PPV and SVV cannot preclude a positive hemodynamic response to fluid. Jean-Louis Teboul, MD, PhD Service de Réanimation Médicale Centre Hospitalo-Universitaire de Bicêtre Assistance Publique-Hôpitaux de Paris EA 4046 Université Paris-Sud Le Kremlin-Bicêtre, France REFERENCES 1. Kumar A, Anel R, Bunnell, et al: Pulmonary artery occlusion pressure and central venous pressure fail to predict ventricular filling volume, cardiac performance, or the response to volume infusion in normal subjects. Crit Care Med 2004; 32:691– 699 2. Osman D, Ridel C, Ray P, et al: Cardiac filling pressures are not appropriate to predict hemodynamic response to volume challenge. Crit Care Med 2007; 35:64 – 68 3. Marik PE, Baram M, Vahid B: Does central venous pressure predict fluid responsiveness? A systematic review of the literature and the tale of seven mares. Chest 2008; 134:172–178 4. Michard F, Teboul JL: Predicting fluid responsiveness in ICU patients: A critical analysis of the evidence. Chest 2002; 121:2000 –2008 5. Michard F, Boussat S, Chemla D, et al: Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med 2000; 162: 134 –138 6. Kramer A, Zygun D, Hawes H, et al: Pulse pressure variation predicts fluid responsiveness following coronary artery bypass surgery. Chest 2004; 126:1563–1568 7. Preisman S, Kogan S, Berkenstadt H, et al: Predicting fluid responsiveness in patients undergoing cardiac surgery: Functional haemodynamic parameters including the Respi- 8. 9. 10. 11. 12. 13. 14. 15. ratory Systolic Variation Test and static preload indicators. Br J Anaesth 2005; 95: 746 –755 Berkenstadt H, Margalit N, Hadani M, et al: Stroke volume variation as a predictor of fluid responsiveness in patients undergoing brain surgery. Anesth Analg 2001; 92: 984 –989 Reuter DA, Felbinger TW, Schmidt C, et al: Stroke volume variations for assessment of cardiac responsiveness to volume loading in mechanically ventilated patients after cardiac surgery. Intensive Care Med 2002; 28: 392–398 Monnet X, Rienzo M, Osman D, et al: Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med 2006; 34: 1402–1407 Heenen S, De Backer D, Vincent JL: How can the response to volume expansion in patients with spontaneous respiratory movements be predicted? Crit Care 2006; 10:R102 De Backer D, Heenen S, Piagnerelli M, et al: Pulse pressure variations to predict fluid responsiveness: Influence of tidal volume. Intensive Care Med 2005; 31:517–523 de Waal EEC, Rex S, Kruitwagen CLJJ, et al: Dynamic preload indicators fail to predict fluid responsiveness in open-chest conditions. Crit Care Med 2009; 37:510 –515 Jardin F, Delorme G, Hardy A, et al: Reevaluation of hemodynamic consequences of positive pressure ventilation: Emphasis on cyclic right ventricular afterloading by mechanical lung inflation. Anesthesiology 1990; 72:966 –970 Reuter DA, Goepfert MS, Goresch T, et al: Assessing fluid responsiveness during open chest conditions. Br J Anaesth 2005; 94: 318 –323 Antioxidant therapy: Reducing malaria severity?* E ach year, infection with Plasmodium falciparum causes 300 – 600 million illnesses worldwide (1). A significant number of patients will progress to severe malaria with organ dysfunction, which is more common in the immunologically *See also p. 516. Key Words: oxidative stress; cytoadherence; Plasmodium falciparum; endothelium; artesunate; N-acetylcysteine; lactate; glucose-6-phosphate dehydrogenase deficiency The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d5de 758 naïve: young children and travelers or other adults with only sporadic exposure to infected mosquitoes. Mortality from severe malaria ranges from 5% to 40%, and death often occurs shortly after hospital admission. Despite improved survival with artesunate treatment (2), mortality remains high in the first 24 hours of hospitalization, and therefore, adjunctive therapies are still urgently needed. Neither hypovolemia, cardiogenic shock, nor vasodilatory shock is necessary for organ dysfunction in severe malaria (3). The impairment of tissue perfusion occurs within microvessels, where parasitized red blood cells (RBCs) adhere to endothelial cells and circulating immune cells via spe- cific interactions between the parasiteencoded adherence ligand P. falciparum erythrocyte membrane protein 1 on the RBC surface and the host-encoded adhesion receptors cluster of differentiation 36 and intercellular adhesion molecule-1 (4). Adherence is associated with inflammatory cytokine secretion, adhesion molecule expression, tissue factor display, and platelet aggregation (5)—findings that are evident on postmortem examination of patients with severe malaria (6). Microcirculatory dysfunction may be worsened as nitric oxide is scavenged by cell-free hemoglobin released during hemolysis and by reactive oxygen species (7). Crit Care Med 2009 Vol. 37, No. 2 In severe malaria, reactive oxygen species are generated by parasite hemoglobin metabolism in RBCs, nicotinamide adenine dinucleotide phosphate oxidase in phagocytes, and nitric oxide synthase when the substrate arginine or cofactor tetrahydrobiopterin is lacking. Plasma hemoglobin released from lysed RBCs may also catalyze the generation of reactive oxygen species. Patients with severe malaria have increased reactive oxygen species products in urine (8), decreased alpha-tocopherol in RBC membranes (9), and decreased deformability of RBCs under shear stress. Decreased deformability of RBCs is associated with mortality (10); this was the impetus for the study by Charunwatthana et al (8), who hypothesized that decreased deformability impedes the transit of RBCs through capillaries, impairing oxygen delivery. Because N-acetylcysteine (NAC) restores normal deformability of RBCs in vitro by repleting glutathione reserves (11), treatment with NAC was expected to improve RBC deformability and oxygen delivery in vivo. However, oxidant stress in RBCs could offer some compensatory benefits to a patient with severe malaria. First, oxidative stress is a fundamental mechanism for killing phagocytosed pathogens, including P. falciparum. Treatment with a potent antioxidant might conceivably protect parasites from the oxidative burst of phagocytes and could antagonize the oxidation-mediated antimalarial effects of artesunate. In vitro, the parasiticidal effect of artesunate is reduced when given simultaneously with NAC, but is unaffected when NAC is given 2 hours after artesunate (12). In the present study, Charunwatthana et al administered NAC 2 hours after the first dose of artesunate and measured parasitemia every 6 hours to see whether NAC would impair parasite killing by artesunate. Second, oxidative stress in RBCs may play a role in weakening adherence interactions between the parasitized RBCs and the microvascular endothelium. This is best illustrated by the classic protective effect of sickle cell trait against severe malaria. Infected RBCs from patients with sickle cell trait show aberrant display of the parasite-encoded adhesion ligand P. falciparum erythrocyte membrane protein 1 and reduced strength of adhesion to microvascular endothelial cells and blood monocytes in vitro (13). This phenotype may be reproduced in RBCs with glucose-6-phosphate dehydrogenase deficiency or other states of oxidative stress. Crit Care Med 2009 Vol. 37, No. 2 Third, although oxidative stress in infected RBCs might be beneficial, oxidative stress in endothelial cells causes adhesion molecule expression, apoptosis, and capillary leak. Endothelial cell injury by adherent parasitized RBCs is ameliorated by superoxide dismutase, ascorbic acid, or tocopherol in vitro, highlighting the potential therapeutic benefit of antioxidants as endothelial cell protectors (14). So what will be the effect of a systemic antioxidant in severe malaria? In the study by Charunwatthana et al (8), adults with severe malaria were treated with artesunate and randomized to receive NAC (dosed as for acetaminophen overdose) or placebo. NAC had no effect on the primary or secondary end points of the study: lactate clearance time, coma recovery time, parasite clearance time, fever clearance time, or mortality. End products of oxidative stress (F2-isoprostanes) were found to be elevated in the urine of patients with severe malaria compared with uncomplicated malaria, but were unchanged by NAC treatment. Although red cell deformability was lower in fatal cases than in survivors of severe malaria, it was not changed by NAC treatment. Why was NAC ineffective? In a pilot study (15), 30 patients were randomized to NAC or placebo, and at 24 hours a greater proportion of patients in the NAC-treated arm had normal lactate levels compared with the placebo arm (10 of 15 vs. 3 of 15, p ⫽ 0.01); however, the placebo group had higher baseline parasitemia and bilirubin and lower Glasgow coma scores than the treatment group. Baseline differences in disease severity could have overestimated the benefit, if any, of NAC treatment. As a consequence, the current study may have been underpowered to detect the true effect of NAC treatment on severe malaria. Antimalarial treatment was also different in the current study: NAC was given 2 hours after artesunate, a pro-oxidant drug, compared with NAC being given simultaneously with quinine in the pilot study. Furthermore, genetic polymorphisms that influence RBC oxidative stress (such as glucose-6-phosphate dehydrogenase deficiency) were not accounted for, but could have modified the response to antioxidant therapy. In hindsight, it is unclear what effect NAC should have had on severe malaria. Oxidative stress is considered beneficial for parasite killing and for weakening the adherence between infected RBCs and endothelial cells or monocytes. On the other hand, oxidative stress causes endothelial injury and impairs RBC deformability. The results of the current study are neutral. Does this close the book on antioxidants for severe malaria? No, but future strategies might specifically target antioxidants to the endothelium while enhancing oxidative stress in infected RBCs. On second thought, combining a systemic antioxidant NAC with an RBCspecific pro-oxidant artesunate may have accomplished just that. Hans C. Ackerman, MD, DPhil, MSc Laboratory of Malaria and Vector Research National Institute of Allergy and Infectious Diseases Rockville, MD Critical Care Medicine Department National Institutes of Health Clinical Center Bethesda, MD Steven D. Beaudry, BS Rick M. Fairhurst, MD, PhD Laboratory of Malaria and Vector Research National Institute of Allergy and Infectious Diseases Rockville, MD REFERENCES 1. Snow RW, Guerra CA, Noor AM, et al: The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 2005; 434:214 –217 2. Dondorp A, Nosten F, Stepniewska K, et al: Artesunate versus quinine for treatment of severe falciparum malaria: A randomised trial. Lancet 2005; 366:717–725 3. Planche T, Onanga M, Schwenk A, et al: Assessment of volume depletion in children with malaria. PLoS Med 2004; 1:e18 4. Ockenhouse CF, Ho M, Tandon NN, et al: Molecular basis of sequestration in severe and uncomplicated Plasmodium falciparum malaria: Differential adhesion of infected erythrocytes to CD36 and ICAM-1. J Infect Dis 1991; 164:163–169 5. Francischetti IM, Seydel KB, Monteiro RQ: Blood coagulation, inflammation, and malaria. Microcirculation 2008; 15:81–107 6. Taylor TE, Fu WJ, Carr RA, et al: Differentiating the pathologies of cerebral malaria by postmortem parasite counts. Nat Med 2004; 10:143–145 7. Gramaglia I, Sobolewski P, Meays D, et al: Low nitric oxide bioavailability contributes to the genesis of experimental cerebral malaria. Nat Med 2006; 12:1417–1422 8. Charunwatthana P, Faiz MA, Ruangveerayut R, et al: N-acetylcysteine as adjunctive treatment in severe malaria: A randomized double blinded placebo controlled-clinical trial. Crit Care Med 2009; 37:516 –522 759 9. Griffiths MJ, Ndungu F, Baird KL, et al: Oxidative stress and erythrocyte damage in Kenyan children with severe Plasmodium falciparum malaria. Br J Haematol 2001; 113:486 – 491 10. Dondorp A, Angus BJ, Hardeman MR, et al: Prognostic significance of reduced red blood cell deformability in severe falciparum malaria. Am J Trop Med Hyg 1997; 57:507–511 11. Nuchsongsin F, Chotivanich K, Charunwatthana P, et al: Effects of malaria heme prod- ucts on red blood cell deformability. Am J Trop Med Hyg 2007; 77:617– 622 12. Arreesrisom P, Dondorp AM, Looareesuwan S, et al: Suppressive effects of the antioxidant N-acetylcysteine on the antimalarial activity of artesunate. Parasitol Int 2007; 56:221–226 13. Cholera R, Brittain NJ, Gillrie MR, et al: Impaired cytoadherence of Plasmodium falciparum-infected erythrocytes containing sickle hemoglobin. Proc Natl Acad Sci USA 2008; 105:991–996 14. Taoufiq Z, Pino P, Dugas N, et al: Transient supplementation of superoxide dismutase protects endothelial cells against Plasmodium falciparum-induced oxidative stress. Mol Biochem Parasitol 2006; 150:166 –173 15. Watt G, Jongsakul K, Ruangvirayuth R: A pilot study of N-acetylcysteine as adjunctive therapy for severe malaria. QJM 2002; 95:285–290 Is responsiveness to family wishes an expression of professional transcendence?* T he authors are to be praised for proactively exploring the concept of involving family wishes in the decision to admit to the intensive care unit. Participation in a shared decision-making model is endorsed by the Society of Critical Care Medicine (1). In this issue of Critical Care Medicine, Escher et al (2) determine from the results of surveys administered to physicians in Switzerland that older physicians and those with self-determined knowledge of ethics were more likely to admit a hypothetical young woman with uremic syndrome in the intensive care unit. Physicians were also more likely to admit if the family stated that they wanted everything done vs. “spare useless suffering.” As a point of discussion, for those engaged in providing critical care to the ill and injured, the manner in which the phrase “spare useless suffering” was used in this study prompts a gutteral reaction to exclaim that critical care does not uniformly produce “useless suffering”! If a few days of intensive treatment and tests produce a dramatic effect on symptomotology vs. weeks of protracted untreated illness on the ward, which is suffering? One would wonder how lay people would react. Do they know that when they use the phrase “do everything” that the physician interprets that as “admit to the intensive care unit” and when they use the phrase “do not *See also p. 528. Key Words: ethics; end of life; transcendence; decision-making; family; critical care The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e3181959e27 760 let her suffer” that they may actually be pointing the physician in the direction of withholding care or treatment? It would be interesting to reconstruct the scenarios to evaluate whether the terms are congruent with the wishes of the general public, as posited in the basic assumptions in the research design of Escher et al. With that said, the proposition that older physicians and those with greater knowledge in ethics responded in a manner more congruent with family wishes is intriguing. Do the results point more toward the concept of professional-transcendence? Transcendence is defined as, the ability to expand selfboundaries intrapersonally (through increased awareness of one’s beliefs and values), interpersonally (by reaching out to others), and temporally (integrating the past and future into the present), and may be measured by valid and reliable tools (3). Transcendence is associated with experiences that increase awareness of human vulnerability and mortality (4). Do those physicians with more life experience, who have witnessed the suffering of others over longer periods of time, and are in tune to the vulnerabilities of our existence on earth extend beyond themselves and listen more attentively and empathetically to others? As suggested by Weaver et al (5), ethical sensitivity may have an effect on decision making and is related to professional transcendence. Do physicians who are less paternalistic in their views about decision making have greater professional transcendence? If the study by Escher et al (2) were extended to explore a situation where the nurse had suggested that we “spare useless suffering,” would these same older doctors with greater confidence in their skills related to ethics have listened and not admitted to the intensive care unit? The question left to ponder is whether age and experience provide us with a greater self-perceived understanding of ethics and a higher level of professional transcendence. A future study could be one centered on measuring self-transcendence in physicians and the relationship between decisionmaking models used in the decision to admit and/or collaborate with team members. One question could be, “Is there a relationship between degree of self-transcendence in physicians and use of the shared decision making vs. paternalistic model of decision making.” Another could be, “Does the degree of selftranscendence in physicians relate to collaboration in the workplace?” Judy E. Davidson, DNP, RN, FCCM Department of Advanced Practice Nursing and Research Scripps Mercy Hospital San Diego, CA Beth Palmer, DNP, RN Veterans Administration Medical Center San Diego, CA REFERENCES 1. Davidson JE, Powers K, Hedayat KM, et al: Clinical practice guidelines for support of the family in the patient-centered ICU. American College of Critical Care Task Force: 2004 –2005. Crit Care Med 2007; 35:605– 622 2. Escher M, Perneger TV, Heldegger CP, et al: Admission of incompetent patients to intensive care: Doctors’ responsiveness to family wishes. Crit Care Med 2009; 37:528 –532 3. Reed P: Self-transcendence and mental health in oldest-old adults. Nurs Res 1991; 40:1–7 4. Reed P: The theory of self-transcendence. In: Middle Range Theory for Nursing. Smith M, Liehr P (Eds), New York, Springer, 2003 5. Weaver K, Morse J, Mitcham C: Ethical sensitivity in professional practice: Concept analysis. J Adv Nurs 2008; 62:607– 618 Crit Care Med 2009 Vol. 37, No. 2 Diuresis in renal failure: Treat the patient, not the urine output* W hen critically ill patients develop acute kidney injury (AKI), other than treating the underlying cause, there are very limited therapeutic options that have been proven to improve renal recovery. When we observe a drop in urine output associated with AKI, a natural reaction is to administer agents that will increase the urine output, such as loop diuretics (e.g., furosemide or bumetanide) and dopamine. But even if these therapies increase urine output, will they provide any meaningful benefit to the patient in terms of morbidity and mortality? During multidisciplinary rounds in the intensive care unit, I frequently observe that when nurses contact an attending physician because of a low urine output and rising serum creatinine, an almost reflex reaction is to “give furosemide.” This is rather analogous to ordering aspirin for a headache without analyzing the cause of the problem. In the case of diuretics, if the cause of the decrease in urine output is hypovolemia, clearly, diuretics are not an appropriate choice. The literature regarding dopamine in AKI is clear: although it may increase urine output, there is no benefit in any measure of morbidity or mortality, and it should not be used (1–3). The data regarding the use of loop diuretics in AKI were recently summarized by Bagshaw et al (4). In brief, although loop diuretics may have some theoretical benefits in AKI, such as flushing debris from the kidney tubules and maintaining fluid, electrolyte, and acid– base homeostasis, clinical trials have not shown any benefit in terms of renal recovery or any measure of morbidity or mortality. One observational trial actually showed an increase in mortality with loop diuretics (5). As summarized by Bagshaw et al (4), there are many limitations with published clini- *See also p. 533. Key Words: dopamine; furosemide; bumetanide; diuretics; hemofiltration; acute kidney injury; renal replacement therapy The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194de98 Crit Care Med 2009 Vol. 37, No. 2 cal trials, such as lack of generalizability to modern intensive care unit patients, not including intensive care unit patients, confounding cointerventions such as dopamine and mannitol, delayed or late intervention, and bolus diuretic administration without specific therapeutic end points. Despite lack of data showing benefit, a recent survey suggests that the majority of physicians use diuretics in patients with AKI but do not believe that diuretics can reduce mortality or improve renal recovery (6). This survey also found that the majority of physicians would be willing to participate in randomized clinical trials of loop diuretics in AKI. Clearly, more randomized controlled clinical trials are needed. In this issue of Critical Care Medicine, Van Der Voort et al (7) report the results of a randomized, double-blind, placebocontrolled trial in the specific setting of intensive care unit patients with AKI after hemofiltration. The purpose was to determine whether giving furosemide after hemofiltration to increase urine output would provide any benefit on recovery of renal function in terms of creatinine clearance and duration of continuous renal replacement therapy. Not surprisingly, there was no significant effect on creatinine clearance or renal recovery despite the fact that furosemide significantly increased sodium excretion and urine output as compared with placebo. In fact, there was a slight trend favoring placebo in terms of renal recovery (92% in the furosemide group vs. 100% in the placebo group). The limitations of the trial are clearly stated by the authors: it was a small trial (71 patients), the furosemide patients had a higher baseline Sepsis-Related Organ Failure Score and were slightly older than patients in the placebo group, and the fact that measured creatinine clearance was used to assess renal function, which may not accurately reflect glomerular filtration rate in critically ill patients. When considering the main mechanism of action of loop diuretics, the results are not surprising. Loop diuretics inhibit sodium and water resorption from the loop of Henle and do not improve glomerular filtration. In fact, some data suggest that furosemide may actually decrease glomerular filtration rate (8). Conversely, animal models suggest other mechanisms that may be beneficial, such as attenuation of ischemia-related gene expression (9), but the clinical significance of this effect in humans is unknown. Despite the limitations of the trial, the conclusions seem clear. Routine use of furosemide after hemofiltration does not improve renal recovery. Despite the need for more randomized controlled trials, the current literature regarding loop diuretics in AKI does not support routine use. As summarized by Venkataraman and Kellum (10), diuretics and dopamine are ineffective in preventing AKI or improving outcomes once AKI occurs. For now, we need to attempt to prevent renal failure with strategies such as adequate hydration, maintaining an adequate mean arterial pressure, and minimizing nephrotoxin exposure (10). If a patient with AKI has a clinical indication for fluid removal such as pulmonary edema, loop diuretics may be a reasonable therapeutic approach, but the bottom line is to treat the patient not the urine output. Brad E. Cooper, PharmD, FCCM Hamlot Medical Center Erie, PA REFERENCES 1. ANZICS Clinical Trials Group: Low-dose dopamine in patients with early renal dysfunction: A placebo-controlled randomized trial. Lancet 2000; 356:2139 –2143 2. Friedrich JO, Adhikari N, Herridge MS, et al: Meta-analysis: Low-dose dopamine increases urine output but does not prevent renal dysfunction or death. Ann Intern Med 2005; 142:510 –524 3. Kellum JA, Decker JM: Use of dopamine in acute renal failure: A meta-analysis. Crit Care Med 2001; 29:1526 –1531 4. Bagshaw SM, Bellomo R, Kellum JA: Oliguria, volume overload, and loop diuretics. Crit Care Med 2008; 36:S172–S178 5. Meta RL, Pascual MT, Soroko S, et al: Diuretics, mortality, and nonrecovery of renal function in acute renal failure. JAMA 2002; 288: 2547–2553 6. Bagshaw SM, Delaney A, Jones D, et al: Di- 761 uretics in the management of acute kidney injury: A multinational survey. Contrib Nephrol 2007; 156:236 –249 7. van der Voort PHJ, Boerma EC, Koopmans M, et al: Furosemide does not improve renal recovery after hemofiltration for acute renal fail- ure in critically ill patients: A double blind randomized controlled trial. Crit Care Med 2009; 37: 533–538 8. Trivedi H, Dresser T, Aggarwal K: Acute effect of furosemide on glomerular filtration rate in diastolic dysfunction. Ren Fail 2007; 29:985–989 9. Aravindan N, Shaw A: Effect of furosemide infusion on renal hemodynamics and angiogenesis gene expression in acute renal ischemia/ reperfusion. Ren Fail 2006; 28:25–35 10. Venkataraman R, Kellum JA: Prevention of acute renal failure. Chest 2007; 131:300 –308 Prevention of pulmonary dysfunction after cardiac surgery by a vital capacity maneuver: Is it so simple?* R espiratory dysfunction is a frequently encountered complication in postoperative intensive care unit (ICU) patients. After cardiac surgery, pulmonary dysfunction is a well-known encumbering postoperative issue. Hence, this can result in a significantly increased mortality up to 25% and an attributable morbidity with an importantly prolonged stay in the ICU. In this issue of Critical Care Medicine, Shim et al (1) tackled pulmonary impairment after an off-pump cardiac surgery. Respiratory failure after cardiac surgery with cardiopulmonary bypass results in factors leading to major pulmonary complications: atelectasis is present to a much larger extent than induced by solely anesthesia or sternotomy (2), stasis of secretions, due to insufficient coughing capacity, inflammation and infection. In conjunction with advanced age, preexistent lung disease, impaired cardiac performance, and secondary prolonged duration of mechanical ventilation (3, 4), these risk factors intensify respiratory failure. Furthermore, coronary bypass surgery with and without extracorporeal circulation results in a dramatic impairment of respiratory system mechanics (5), not in the least hampered by an intraoperative positive fluid balance (5, 6). All these complications result in hypoxemia and gas exchange impairment, commonly observed after cardiac surgery. Hachenberg et al (7) demonstrated a *See also p. 539. Key Words: pulmonary dysfunction; cardiac surgery; shunting; vital capacity maneuver; respiratory failure The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194dee3 762 nearly doubled intrapulmonary shunting during and after cardiac surgery, with subsequent augmented extravascular lung water and further collapse (8). Although multifactorial in nature, pulmonary dysfunction seems to be mostly related to intrapulmonary right-to-left shunt. Several investigators applied different techniques to eliminate or diminish the impaired gas exchange after cardiac surgery. Recruitment maneuvers, such as continuous positive airway pressure, bilevel positive airway pressure, or vital capacity maneuver (VCM), at the end of the cardiopulmonary bypass have been used, to alleviate atelectasis and shunting. Continuous positive airway pressure with varying levels of airway pressure has lead to conflicting results: either decreased or improved lung function was shown, although moderate continuous positive airway pressure appeared to result in the optimalization of postbypass pulmonary dysfunction (9). Bilevel positive airway pressure was observed to be beneficial in general anesthesia to prevent ventilation-perfusion mismatch and to improve oxygenation (10). With VCM, the lungs are inflated to a peak airway pressure of 40 cm H2O for 15 seconds. With respect to on-pump cardiac surgery, several investigators clearly demonstrated a beneficial effect of VCM on postoperative extubation times. However, the influence on pulmonary gas exchange in the ICU remains controversial. Magnusson et al (11) assessed the effects of VCM to prevent atelectasis after cardiopulmonary bypass in a study on pigs. Control pigs developed extensive atelectasis, whereas VCM-treated pigs showed a significantly smaller proportion of atelectasis, the least in those animals ventilated with lower oxygen concentrations. Repetition within 6 hours seemed to provide no additional ad- vantages based on both laboratory and computed tomographic data (12). In a human study, Murphy et al (13) could not demonstrate any improvement of gas exchange in the ICU after an on-pump cardiac surgery between control and VCM-supported patients, whereas the latter were extubated significantly earlier, after 6.5 ⫾ 2.1 hours vs. 9.4 ⫾ 4.2 hours. In a subset of cardiac surgical patients with cardiopulmonary bypass, Tschernko et al (14) also showed a reduction of intrapulmonary shunting using VCM after termination of cardiopulmonary bypass, although shunting increased considerably after extubation. Finally, Minkovich et al (15) performed VCM twice: both before cessation of cardiopulmonary bypass (35 cm H2O—15 seconds) and after admission in the ICU (30 cm H2O—5 seconds). In contrast to the data on animals presented by Magnusson et al (11), VCM resulted in an improved and prolonged arterial oxygenation for 24 hours in patients. No data, however, were provided concerning the timing of extubation. With respect to off-pump cardiac surgery, less information is available. In the previously referred study, Tschernko et al (14) could not show any effect of VCM on the duration of mechanical ventilation between patients operated on bypass (control), those supported with VCM, and off-pump cardiac surgery patients. In addition, they showed an increased shunting after extubation, although much less than in those patients with cardiopulmonary bypass and with direct impact on length of stay in the ICU and in the hospital, compared with control and onpump cardiac surgical patients. The findings of Shim et al (1) now clearly demonstrate that VCM leads to earlier extubation in off-pump cardiac surgery patients albeit without demonstrable effects on respiratory mechanics, length of stay in the ICU nor in the hospital. Although strict extubation criteria Crit Care Med 2009 Vol. 37, No. 2 were followed, extubation remains a rather arbitrary factor. Many interfering issues, as listed earlier, interfere, not in the least an appropriate sensorium. In this study, patients who were not ready for extubation were sedated with midazolam. At least duration and level of sedation can be a major point of discussion. Also, the value of VCM in this particular study could be questioned as there was no need at all to increase FIO2 throughout the study, even in non-VCMsupported patients, in whom shunting and atelectasis could have been expected throughout a 4-hour-lasting procedure. Another point of discussion is the timing of VCM: Shim et al performed the VCM immediately after sternotomy. As anesthesia may help the development of atelectasis, could there be a supplementary beneficial effect of a VCM before closure of the sternum? Finally, in none of the patients was respiratory infection a cause for prolonged mechanical ventilation, although the importance of this issue cannot be denied (16, 17). From larger studies, it could be expected that ⫾20% of cardiac surgical patients are prone to tracheobronchial or pulmonary infection, associated with prolonged stay in the hospital, multiple organ dysfunction, and increased hospital mortality (16, 18). With respect to prevention of atelectasis and decreased intrapulmonary shunting after cardiac surgery, the last word has not been said. Altering the gas mixture from FIO2 1.0 to 30% oxygen in nitrogen during induction of anesthesia could avoid early atelectasis and shunting (19). In view of the multifactorial nature of this problem, further investigations are needed to ascertain a simple approach of pulmonary dysfunction after cardiac surgery. Jan Poelaert, MD, PhD Department of Anesthesiology and Perioperative Medicine Crit Care Med 2009 Vol. 37, No. 2 Acute and Chronic Paintherapy University Hospital of the Flemish Free University of Brussels Brussels, Belgium Carl Roosens, MD Department of Intensive Care Medicine Ghent University Hospital Ghent, Belgium 10. 11. REFERENCES 1. Shim JK, Chun DH, Choi YS, et al: Effect of early vital capacity maneuver on respiratory variables during multivessel off-pump coronary artery bypass graft surgery. Crit Care Med 2009; 37:539 –544 2. Magnusson L, Zemgulis V, Wicky S, et al: Atelectasis is a major cause of hypoxemia and shunt after cardiopulmonary bypass: An experimental study. Anesthesiology 1997; 87: 1153–1163 3. Nozawa E, Azeka E, Ignez ZM, et al: Factors associated with failure of weaning from longterm mechanical ventilation after cardiac surgery. Int Heart J 2005; 46:819 – 831 4. Paulus S, Lehot JJ, Bastien O, et al: Enoximone and acute left ventricular failure during weaning from mechanical ventilation after cardiac surgery. Crit Care Med 1994; 22: 74 – 80 5. Roosens C, Heerman J, De Somer F, et al: Effects of off-pump coronary surgery on the mechanics of the respiratory system, lung, and chest wall: Comparison with extracorporeal circulation. Crit Care Med 2002; 30: 2430 –2437 6. Ranieri VM, Vitale N, Grasso S, et al: Timecourse of impairment of respiratory mechanics after cardiac surgery and cardiopulmonary bypass. Crit Care Med 1999; 27: 1454 –1460 7. Hachenberg T, Tenling A, Nystrom SO, et al: Ventilation-perfusion inequality in patients undergoing cardiac surgery. Anesthesiology 1994; 80:509 –519 8. Hachenberg T, Tenling A, Rothen HU, et al: Thoracic intravascular and extravascular fluid volumes in cardiac surgical patients. Anesthesiology 1993; 79:976 –984 9. Boldt J, King D, Scheld HH, et al: Lung 12. 13. 14. 15. 16. 17. 18. 19. management during cardiopulmonary bypass: Influence on extravascular lung water. J Cardiothorac Anesth 1990; 4:73–79 Yu G, Yang K, Baker AB, et al: The effect of bi-level positive airway pressure mechanical ventilation on gas exchange during general anaesthesia. Br J Anaesth 2006; 96:522–532 Magnusson L, Zemgulis V, Tenling A, et al: Use of a vital capacity maneuver to prevent atelectasis after cardiopulmonary bypass: An experimental study. Anesthesiology 1998; 88: 134 –142 Magnusson L, Wicky S, Tyden H, et al: Repeated vital capacity maneuvers after cardiopulmonary bypass: Effects on lung function in a pig model. Br J Anaesth 1998; 80: 682– 684 Murphy GS, Szokol JW, Curran RD, et al: Influence of a vital capacity maneuver on pulmonary gas exchange after cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2001; 15:336 –340 Tschernko EM, Bambazek A, Wisser W, et al: Intrapulmonary shunt after cardiopulmonary bypass: The use of vital capacity maneuvers versus off-pump coronary artery bypass grafting. J Thorac Cardiovasc Surg 2002; 124:732–738 Minkovich L, Djaiani G, Katznelson R, et al: Effects of alveolar recruitment on arterial oxygenation in patients after cardiac surgery: A prospective, randomized, controlled clinical trial. J Cardiothorac Vasc Anesth 2007; 21:375–378 Kollef MH, Sharpless L, Vlasnik J, et al: The impact of nosocomial infections on patient outcomes following cardiac surgery. Chest 1997; 112:666 – 675 Kollef MH, Skubas NJ, Sundt TM: A randomized clinical trial of continuous aspiration of subglottic secretions in cardiac surgery patients. Chest 1999; 116:1339 –1346 Bouza E, Perez A, Munoz P, et al: Ventilatorassociated pneumonia after heart surgery: A prospective analysis and the value of surveillance. Crit Care Med 2003; 31:1964–1970 Rothen HU, Sporre B, Engberg G, et al: Atelectasis and pulmonary shunting during induction of general anaesthesia—Can they be avoided? Acta Anaesthesiol Scand 1996; 40: 524 –529 763 And the winner is: Regional citrate anticoagulation* A cute kidney injury requiring renal replacement therapy (RRT) is an independent mortality risk factor in intensive care patients. Most likely, the unwanted consequences of both acute kidney injury and RRT contribute to this increased mortality. Continuous RRT is often the procedure of choice in these patients. One requirement for continuous RRT having potential deleterious consequences is systemic anticoagulation. Although anticoagulation exposes patients to the risk of bleeding, its absence may result in clotting of the circuit and less effective treatment. Worldwide, systemic unfractionated heparin is the most common anticoagulant (1). In the last two decades, various methods of regional anticoagulation using citrate (RCA) have been described (2, 3); however, its use worldwide is limited. Several factors are responsible for this, including concerns about the safety of RCA, the metabolic complexity, the costs and logistics of obtaining custom-made solutions, and the lack of uniformity in therapeutic approaches. Citrate causes anticoagulation in the extracorporeal circuit by chelating ionized calcium (iCa). The citrate– calcium complex is partly filtered and partly diluted by the total blood volume. In the systemic circulation, citrate is rapidly metabolized via the Krebs cycle in the liver and other tissues yielding bicarbonate and releasing iCa. The extracorporeal loss of calcium is substituted. Citrate possibly also inhibits the calciummediated activation of inflammatory cells in the extracorporeal circulation (4, 5). A potential safety risk with RCA is the occurrence of metabolic disturbances, entailing the frequent monitoring of acid– base status, plasma electro- *See also p. 545. Key Words: citrate; heparin; acute kidney injury; continuous renal replacement therapy The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194df2e 764 lytes, and calcium levels. The infusion of hypertonic trisodium citrate may lead to hypernatremia and/or alkalosis, unless adapted replacement solutions are used. Citrate toxicity may occur in patients insufficiently metabolizing citrate (e.g., severe liver failure, decreased muscle perfusion). Systemic citrate levels are not routinely measured in clinical practice, but accumulation is easily detected by an increased total calcium concentration/iCa ratio (6). In this edition of Critical Care Medicine, Oudemans-van Straaten et al compare the safety and efficacy of RCA with the systemic low-molecular weight heparin nadroparin (7). To date, this is the largest randomized controlled trial on RCA. Patients with acute kidney injury requiring RRT and no contraindications for systemic nadroparin and/or RCA were included. Of the 215 randomized patients, 200 patients received continuous veno-venous hemofiltration (CVVH) per protocol (97 RCA and 103 nadroparin). The study was powered on adverse events necessitating discontinuation of study medication. Primary outcomes were safety and efficacy. Discontinuation was required in two patients of the RCA group (citrate accumulation, clotting) and in 20 patients in the nadroparin group (bleeding, thrombocypenia) (p ⬍ 0.001). Unfortunately, anti-Xa levels were not measured and this might have reduced bleeding. Nevertheless, the differences in bleeding and transfusion were not statistically significant between the groups, possibly because nadroparin was discontinued when bleeding occurred, and patients with a high risk of bleeding were excluded. Circuit survival was not significantly different between groups. These results contrast with the findings of previous randomized controlled trials (Table 1). Although previous studies were small and underpowered, some differences among studies are noteworthy. The present study infused fixed doses of nadroparin and citrate, whereas previous studies titrated unfractionated heparin and citrate to achieve a doubling of the activated partial thromboplastin time and a circuit iCa of ⬍0.35 mmol/L. Routine circuit disconnection affects circuit survival. Circuits were routinely disconnected after 72 hours in the studies by Oudemans-van Straaten et al and Betjes et al (8), but not in the studies by Monchi et al (9) and Kutsogiannis et al (10). Notably, in the present study RCA did not result in serious metabolic disturbances confirming previous findings (2, 3, 8 –12). In the present study, RCA even resulted in less metabolic alkalosis (p ⫽ 0.001) compared with nadroparin; however, the RCA protocol aimed at normal pH using two replacement fluids, and this was not the case in the nadroparin group. A remarkable finding in the present study is the survival benefit of the RCA group. Three-month mortality on intention-to-treat was 48% (RCA) and 63% (nadroparin) (p ⫽ 0.03). Three-month mortality per protocol was 45% (RCA) and 62% (nadroparin) (p ⫽ 0.02). There were no significant differences in CVVH-timing and CVVH-dose between groups. Post hoc analysis showed that RCA was particularly beneficial in surgical patients, patients with sepsis, patients with severe organ failure, and relatively younger patients. As suggested by the authors, one can hypothesize that RCA is favorable, nadroparin unfavorable, or both. Although the patients in the nadroparin group tended to bleed more, the differences in bleeding and transfusion between groups were not statistically significant and could not explain the RCA survival benefit. The incidence of metabolic alkalosis was higher in the nadroparin group. Alkalemia has hemodynamic consequences as a result of direct effect and by means of reducing iCa. In the present study, however, iCa was lower in the RCA group. Furthermore, metabolic alkalosis during CVVH was not related to mortality at univariate analysis, supporting the recent findings of Demirjian et al (13). The authors speculate that this reduction in mortality is from RCA blocking inflammation. This is certainly a plausible explanation based on the results of previous studies on patients undergoing chronic dialysis (4, 5, 14). These studies demonstrated that dialysis-induced polymorphonuclear cell degranulation is primarily iCa dependent and is abolished during RCA dialysis (4, 5). In addition, Gabutti et al (14) showed that RCA dialysis had a favorable effect on interCrit Care Med 2009 Vol. 37, No. 2 Table 1. Randomized controlled trials comparing the safety and efficacy of regional citrate anticoagulation and systemic heparins Author Study Design (Number of Patients)a Anticoagulation Monitoring CRRT Modality Filter Oudemans-van RCA 97, LMWH Fixed dose LMWH Postdilution CTA 1.9 m2 Straaten et al (7) 103 and RCA CVVH Monchi et al (9) Betjes et al (8) Kutsogiannis et al (10) Cross-over Circuit iCa ⬍0.3 Postdilution PS 1.9 m2 RCA-UFH 8, mmo/L, APTT CVVH UFH-RCA 12 60–80 sec RCA 21, UFH Circuit iCa 0.25– Postdilution CTA 1.9 m2 27 0.3 mmo/L, CVVH APTT 50–70 sec RCA 16, UFH Circuit iCa 0.25– Predilution AN 69 1.0 m2 14 35 mmo/L, PTT CVVHDF 45–65 sec QB (mL/min) QUF (mL/hr) Citrate (mmol/L QB) 220 2000–4000 3 150 35 mL/kg/hr 4.3 150 1500 125 1000, QD 1000 mL/hr 3 3.3 Median Circuit Survival (hr) Major Bleeding Events Units of PRC per Day of CRRT RCA 27, RCA 6, RCA 0.27, LMWH 26 LMWH 16 LMWH 0.36 (p ⫽ 0.68)b (p ⫽ 0.08) (p ⫽ 0.31) RCA 70, RCA 0, RCA 0.2, UFH 40 UFH 1 UFH 1 (p ⫽ 0.007) (p ⫽ NR) (p ⫽ 0.0008) RCA 36, RCA 0, RCA 0.43, UFH 38.4 UFH 10 UFH 0.88 (p ⫽ NS)b (p ⬍ 0.01) (p ⫽ 0.01) RCA 124.5, RCA 0, UFH 38.3 UFH 7 (p ⫽ 0.001) (p ⫽ NR) RCA 0.17, UFH 0.33 (p ⫽ 0.13) RCA, regional citrate anticoagulation; LMWH, low molecular weight heparin; UFH, unfractionated heparin; iCa, ionized calcium; (A)PTT, (activated) partial thromboplastin time; CVVH, continuous veno-venous hemofiltration; CVVHDF, continuous veno-venous hemodiafiltration; CTA, cellulose triacetate; PS, polysulfone; AN69, polyacrilonitrile; QB, blood flow; QUF, ultrafiltrate flow; QD, dialysate flow; NS, not significant; NR, not reported; PRC, packed red cell; CRRT, continuous renal replacement therapy. a Inclusion criteria, patients without contraindications for citrate or systemic heparin; bRoutine circuit disconnection after 72 hrs. leukin-1␤ release. Unfortunately, the present study does not provide data concerning biocompatibility and inflammatory markers to support the theory. In summary, the study of Oudemansvan Straaten et al provides convincing evidence that RCA is as effective as, yet safer than nadroparin. The study has some limitations, including being a single-center study and using low-molecular weight heparin instead of unfractionated heparin. Furthermore, it is clear that the study was designed to evaluate the safety and efficacy of RCA and not survival. Another large Dutch multicenter study is underway, randomizing patients into CVVH with RCA or unfractionated heparin (www.clinicaltrials. gov/ct2/show/NCT00209378). Primary outcome measures of this study are mortality, circuit survival, and bleeding complications, whereas laboratory markers of inflammation, endothelial dysfunction, and coagulation parameters are secondary end points. If this upcoming study confirms the results of Oudemans-van Straaten et al, RCA will definitely be the number one anticoagulation for continuous RRT. Catherine S. C. Bouman, MD, PhD Department of Intensive Care Academic Medical Center University of Amsterdam Amsterdam, The Netherlands Crit Care Med 2009 Vol. 37, No. 2 REFERENCES 1. Uchino S, Bellomo R, Morimatsu H, et al: Continuous renal replacement therapy: A worldwide practice survey. The beginning and ending supportive therapy for the kidney (B.E.S.T. kidney) investigators. Intensive Care Med 2007; 33:1563–1570 2. Egi M, Naka T, Bellomo R, et al: A comparison of two citrate anticoagulation regimens for continuous veno-venous hemofiltration. Int J Artif Organs 2005; 28:1211–1218 3. Tolwani AJ, Prendergast MB, Speer RR, et al: A practical citrate anticoagulation continuous venovenous hemodiafiltration protocol for metabolic control and high solute clearance. Clin J Am Soc Nephrol 2006; 1:79 – 87 4. Bohler J, Schollmeyer P, Dressel B, et al: Reduction of granulocyte activation during hemodialysis with regional citrate anticoagulation: Dissociation of complement activation and neutropenia from neutrophil degranulation. J Am Soc Nephrol 1996; 7:234 –241 5. Bos JC, Grooteman MP, van Houte AJ, et al: Low polymorphonuclear cell degranulation during citrate anticoagulation: A comparison between citrate and heparin dialysis. Nephrol Dial Transplant 1997; 12:1387–1393 6. Hetzel GR, Taskaya G, Sucker C, et al: Citrate plasma levels in patients under regional anticoagulation in continuous venovenous hemofiltration. Am J Kidney Dis 2006; 48: 806 – 811 7. Oudemans-van Straaten HM, Bosman RJ, Koopmans M, et al: Citrate anticoagulation for continuous venovenous hemofiltration. Crit Care Med 2009; 37:545–552 8. Betjes MG, van Oosterom D, van Agteren M, et al: Regional citrate versus heparin anticoagulation during venovenous hemofiltration in patients at low risk for bleeding: Similar hemofilter survival but significantly less bleeding. J Nephrol 2007; 20:602– 608 9. Monchi M, Berghmans D, Ledoux D, et al: Citrate vs. heparin for anticoagulation in continuous venovenous hemofiltration: A prospective randomized study. Intensive Care Med 2004; 30:260 –265 10. Kutsogiannis DJ, Gibney RT, Stollery D, et al: Regional citrate versus systemic heparin anticoagulation for continuous renal replacement in critically ill patients. Kidney Int 2005; 67:2361–2367 11. Bagshaw SM, Laupland KB, Boiteau PJ, et al: Is regional citrate superior to systemic heparin anticoagulation for continuous renal replacement therapy? A prospective observational study in an adult regional critical care system. J Crit Care 2005; 20: 155–161 12. Mehta RL, McDonald BR, Aguilar MM, et al: Regional citrate anticoagulation for continuous arteriovenous hemodialysis in critically ill patients. Kidney Int 1990; 38:976 –981 13. Demirjian S, Teo BW, Paganini EP: Alkalemia during continuous renal replacement therapy and mortality in critically ill patients. Crit Care Med 2008; 36:1513–1517 14. Gabutti L, Ferrari N, Mombelli G, et al: The favorable effect of regional citrate anticoagulation on interleukin-1beta release is dissociated from both coagulation and complement activation. J Nephrol 2004; 17: 819 – 825 765 New biomarkers of acute kidney injury: Promise for the future but beware the lure of novelty* A t last, a much needed consensus definition of acute kidney injury (AKI) has emerged, which should not only facilitate comparative research and audit but also recognize the adverse impact of even modest disease (1). However, its reliance on rises in serum creatinine—the traditional biomarker of renal dysfunction— does carry the risk of missed therapeutic opportunity because of the time lag between the initiating insult and the diagnostic elevation (2). The interpretation of changes in serum creatinine is also complicated by factors such as diet, muscle mass, and the use of certain drugs and supplements, as well as by the presence of prerenal dysfunction when no cellular injury has actually occurred. It is these limitations that have stimulated an ongoing and intensive evaluation of a variety of alternative early biomarkers of AKI. When examining the evolving literature, it is crucial to bear in mind what the potential candidate biomarker is actually reflecting; cystatin C, for instance, is a measure of renal functional status (a “quick creatinine”) whereas others, such as neutrophil gelatinase–associated lipocalin (NGAL) and urinary interleukin18, are products of the pathophysiologic processes that underlie AKI (3) and indicate active renal damage (a “troponin of the kidney”). In human studies, cystatin C can predict the development of AKI (4) and the requirement for renal replacement therapy (5), although its superiority over serum creatinine has not been a universal finding (6). Serum and/or urine NGAL levels have been shown to be accurate predictors of AKI in settings as diverse as *See also p. 553. Key Words: acute renal failure; acute kidney injury; neutrophil gelatinase–associated lipocalin; cystatin; interleukin 18; biomarkers; creatinine The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194dfd0 766 percutaneous coronary intervention (7), pediatric (8) and adult (9) cardiac surgery, and septic (10) and nonseptic (11) critically ill children. Rising urinary interleukin-18 levels have proved similarly predictive in a nonseptic critically ill pediatric population (12), in critically ill patients with acute respiratory distress syndrome (13), and after adult and pediatric cardiac surgery (14). As well as predicting AKI, there may be an additional prognostic role for novel biomarkers (12, 13). The article by Haase-Fielitz et al (15), published in this issue of Critical Care Medicine, adds to the existing literature. They are the first to study the early use of serum biomarkers in adult cardiac surgical patients (urine NGAL having already been studied in this setting [9]). Compared to the intensive care unit admission creatinine, the contemporaneous serum NGAL and serum cystatin C were found to have good predictive value for the subsequent development of AKI. The accuracy of cystatin C diminished after patients with preexisting renal impairment were excluded from analysis, suggesting that it did not only indicate evolving AKI but was also an independent risk factor for it, being a reflection of the strong, predisposing effects of chronic renal dysfunction. Beyond AKI, both serum NGAL and cystatin C carried excellent prognostic value for the composite outcome of renal replacement therapy or hospital mortality. Taken in the context of the current literature, the study has added further evidence of potential value of the novel biomarkers of AKI but its weaknesses should be borne in mind: it was small (only 23 of 100 subjects developed AKI), confidence intervals of the area under the receiver operating characteristic curves were wide (in part, a reflection of the former), and an absolute intensive care unit admission serum creatinine was used (a more valid comparison might have been a ⌬ creatinine that would have helped correct for the confounding effects of preexisting abnormalities). What seems to be evident from the literature as a whole is that no one biomarker fulfils all the desirable features of early detection of active renal damage, of rapid reflection of changes in renal function, of risk stratification, or of differential diagnosis. It is entirely possible that different panels of biomarkers will be required to meet different needs (16). Exploration of this is currently the focus on significant, multicentered effort. If novel biomarkers of AKI are to be of potential value, can we speculate about how they might be utilized? They have clear roles as research tools. An early diagnosis of AKI may allow timely experimental intervention—a major deficiency of previous human studies (17) in which the initiating insult may have been long over by the time AKI was actually diagnosed. Close linkage between the index insult and time of diagnosis may be particularly important if other, confounding insults are likely, as is clearly the case in the critical care setting. Beyond research, the actual clinical utility of these early biomarkers remains untested with a key question being whether they would add anything to management beyond the information provided using conventional, creatininebased diagnosis. One area of possible use is to identify groups of patients who have undergone a discrete renal “hit” and who require subsequent, augmented monitoring. Highrisk outpatients undergoing contrastenhanced radiography might fall into this category but would require a threshold biomarker level of sufficient negative predictive value to definitively exclude evolving AKI. Conversely, the utility of biomarkers as a less targeted, sequential screening tool, unlinked to a specific renal event, is less clear although the results of largescale population studies are outstanding and definitive therapeutic interventions Crit Care Med 2009 Vol. 37, No. 2 are lacking. Although a case could be made that early detection of AKI could encourage the avoidance of harm (e.g., aminoglycoside antibiotics or premature intensive care unit step down), this approach of risk stratification is also, as yet, untested in the clinical environment. In summary, it is increasingly evident that a variety of novel biomarkers are predictive of overt AKI in a range of different settings. Although holding promise for the future, the lure of their novelty does not remove the need for a clear and guarded appraisal of their potential. Nigel S. Kanagasundaram, MD, FRCP (UK) Department of Renal Medicine Newcastle upon Tyne Hospitals NHS Foundation Trust Newcastle Upon Tyne, United Kingdom REFERENCES 1. Mehta RL, Kellum JA, Shah SV, et al: Acute Kidney Injury Network: Report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007; 11:R31 2. Bagshaw SM, Gibney N: Conventional markers of kidney function. Crit Care Med 2008; 36(Suppl 4):S152–S158 3. Goligorsky MS: Whispers and shouts in the pathogenesis of acute renal ischaemia. Nephrol Dial Transplant 2005; 20:261–266 4. Herget-Rosenthal S, Marggraf G, Husing J, et al: Early detection of acute renal failure by serum cystatin C. Kidney Int 2004; 66: 1115–1122 5. Herget-Rosenthal S, Poppen D, Husing J, et al: Prognostic value of tubular proteinuria and enzymuria in nonoliguric acute tubular necrosis. Clin Chem 2004; 50:552–558 6. Ahlstrom A, Tallgren M, Peltonen S, et al: Evolution and predictive power of serum cystatin C in acute renal failure. Clin Nephrol 2004; 62:344 –350 7. Bachorzewska-Gajewska H, Malyszko J, Sitniewska E, et al: Neutrophil-gelatinaseassociated lipocalin and renal function after percutaneous coronary interventions. Am J Nephrol 2006; 26:287–292 8. Mishra J, Dent C, Tarabishi R, et al: Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renal injury after cardiac surgery. Lancet 2005; 365: 1231–1238 9. Wagener G, Jan M, Kim M, et al: Association between increases in urinary neutrophil gelatinase-associated lipocalin and acute renal dysfunction after adult cardiac surgery. Anesthesiology 2006; 105:485– 491 10. Wheeler DS, Devarajan P, Ma Q, et al: Serum neutrophil gelatinase-associated lipocalin (NGAL) as a marker of acute kidney injury in 11. 12. 13. 14. 15. 16. 17. critically ill children with septic shock. Crit Care Med 2008; 36:1297–1303 Zappitelli M, Washburn KK, Arikan AA, et al: Urine neutrophil gelatinase-associated lipocalin is an early marker of acute kidney injury in critically ill children: A prospective cohort study. Crit Care 2007; 11:R84 Washburn KK, Zappitelli M, Arikan AA, et al: Urinary interleukin-18 is an acute kidney injury biomarker in critically ill children. Nephrol Dial Transplant 2008; 23: 566 –572 Parikh CR, Abraham E, Ancukiewicz M, et al: Urine IL-18 is an early diagnostic marker for acute kidney injury and predicts mortality in the intensive care unit. J Am Soc Nephrol 2005; 16:3046 –3052 Parikh CR, Mishra J, Thiessen-Philbrook H, et al: Urinary IL-18 is an early predictive biomarker of acute kidney injury after cardiac surgery. Kidney Int 2006; 70: 199 –203 Haase-Fielitz A, Bellomo R, Devarajan P, et al: Novel and conventional serum biomarkers predicting acute kidney injury in adult cardiac surgery—A prospective cohort study. Crit Care Med 2009; 37:553–560 Parikh CR, Devarajan P: New biomarkers of acute kidney injury. Crit Care Med 2008; 36(Suppl 4):S159 –S165 Chertow GM: On the design and analysis of multicentre trials in acute renal failure. Am J Kidney Dis 1997; 30(Suppl 4):S96–S101 Improving function following cardiopulmonary bypass in children: Digging deeper than steroids* I n the United States alone, approximately 400,000 cardiac surgical operations are performed using cardiopulmonary bypass (CPB) each year, of which approximately 20,000 are performed on children. This volume of pediatric cardiac surgical procedures is, partly, because of the fact that progress in corrective and palliative surgery for congenital heart disease means that most lesions are now operable. Although overall perioperative mortality *See also p. 577. Key Words: cardiopulmonary bypass; inflammation; ischemia; reperfusion The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194b302 Crit Care Med 2009 Vol. 37, No. 2 has declined considerably in the past two decades, short- and long-term morbidity remains a significant healthcare burden. Mortality and morbidity following cardiac surgery is multifactorial, including changes in the inflammatory response, disordered hemostasis, reduced cardiac output, and multiorgan dysfunction. Most corrective procedures require the use of CPB, which invariably activates the inflammatory system. Frequently, this leads to the development of a systemic inflammatory response syndrome, which is characterized by alterations in cardiopulmonary function, coagulopathy, and multiorgan system dysfunction/failure (1). Although cardiac surgery may be technically successful, children are at risk of harm both in terms of increased risk of death and of long-term disability, as a consequence of these CPB-related dysfunctions. This response is predictably more severe in neonates and small infants because of the relatively high exposure of the child’s blood (small volume) to the relatively large volume of the cardiac bypass circuit. The etiology of the inflammatory response is thought to include activation of blood components by the bypass apparatus, ischemia with subsequent reperfusion injury, and endothelial damage. This leads to reversible contractile dysfunction and impairment to flow at the microvascular level secondary to neutrophil plugging and vasoconstriction. Once inflammation is initiated in response to CPB, it is maintained and amplified by cytokine production. Clinical indicators of organ injury and dysfunction, and worse clinical outcome, are associated with increased levels of certain cytokines (inter767 leukin [IL]-6, IL-8) following CPB. The pattern and magnitude of elevations have been routinely associated with post-CPB morbidity in both animal models and human investigations (2). Nuclear factor-(NF-␬B) is a ubiquitous inducible transcription factor involved in the regulation of transcription of many proinflammatory genes. It is activated by stimuli such as IL-1, tumor necrosis factor-␣, and oxygen-free radicals (3). Normally, NF-␬B is bound to the inhibitory protein of NF-␬B (inhibitory kappa-B alpha [IkBa]) (4). Triggered by ischemia and reperfusion, and proinflammatory cytokines themselves, phosphorylation of IkBa and p38 MAPK promotes gene expression of potential myocardial damaging mediators such as tumor necrosis factor-␣, IL-␤, and IL-6 (5). NF-␬B translocates to the nucleus where, binding to DNA, it is able to induce the expression of several inflammatory mediators. This pathway plays an important role in both the inflammatory response triggered by CPB and the ischemia and reperfusion associated with circulatory arrest. In fact, the proposed mechanisms for the local anti-inflammatory glucocorticoid’s actions inhibition of IkBa degradation resulting in decreased NF-␬B activation and inhibition of p38 MAPK through induction of the mitogen-activated protein kinase phosphatase-1 (6). Understanding the triggers, timing, and pattern of the complex inflammatory cascade associated with CPB is essential for modifying or arresting it. Unlike other triggers associated with whole-body inflammatory reactions such as trauma or sepsis, cardiac surgical teams have the advantage of knowing when the trigger will occur (i.e., during CPB), and hence have an opportunity for preemptive intervention in an effort to attenuate or minimize the response. Current strategies involve modulation of the inflammatory response and include the use of corticosteroids (7, 8), coated circuits, (9), and aprotinin (10). These interventions have thus far focused on large-scale interference with inflammation rather than a specific, tailored therapy. Additionally, specific data examining the efficacy of these therapies in various age populations are not available, leaving open the possibility of tailored therapies based on age. Although steroids have gained widespread use in this setting in pediatric cardiac surgery (11), they have potential detrimental effects on neonates (12). 768 In this issue of Critical Care Medicine, Duffy et al (13) examine NF-␬B and its role in the inflammatory response associated with CPB. Their hypothesis is that partial inhibition of NF-␬B can alleviate cardiopulmonary dysfunction associated with ischemia and reperfusion injury CPB. The article describes the effects of SN50, an inhibitor of NF-␬B translocation on hemodynamic and pulmonary function. Specifically, SN50 was administered 1 hour before CPB with subsequent measurement of NF-␬B activity, endothelin-1, troponin I degradation, and cardiopulmonary function. Their results indicate that the administration of SN50 decreased NF-␬B activity levels in nuclear extracts from left venticular myocardium. Additionally, SN50 treatment maintained IkBa protein at higher levels and lower total troponin I degradation than in untreated animals. This resulted in preservation of myocardial contractile function and prevented the increase in pulmonary vascular resistance. Their findings are consistent with previous reports that NF-␬B inhibition protects the myocardium from ischemic injury. By inhibition of NF-␬B with the compound curcumin, Yeh et al (14) ameliorated the surge of proinflammatory cytokines during CPB and decreased the occurrence of cardiomyocyte apoptosis after global cardiac ischemia and reperfusion. This previous report, however, did not measure a clinical effect. Additionally, work by Schwartz et al (15) examined the relationship among glucocorticoid administration, decreased NF-␬B activity in the heart, and improved cardiopulmonary function. Finally, inhibition of NF-␬B has been demonstrated to play a role in injury following circulatory arrest (16). The findings of Duffy et al are important in two fundamental aspects. First, they give insight into a potential mechanism of action of corticosteroid administration in the modulation of CPBassociated inflammation. As previously stated, the local anti-inflammatory action of glucocorticosteroids maybe, partly, due to the NF-␬B pathway (6). Additionally, Duffy et al have proposed the development of a clinically useful, more specific agent that avoids the potential negative effects found with steroids. Because administration of corticosteroids during CPB remains controversial, the goal of therapies targeted at the beneficial aspects of the mechanism of action of steroids would be welcome. Those of us who work in the field of congenital heart disease and cardiac critical care medicine have a unique opportunity to modify or arrest an inflammatory process that contributes to the morbidity and mortality of children undergoing CPB and cardiac surgery. With clinical investigations, such as the study carried out by Duffy et al, the therapeutic strategies available to treat the postbypass inflammatory response will increase, potentially leading to improved outcomes. As the authors themselves state in their conclusion, targeting a specific activity level for NF-␬B would be clinically challenging; yet, it may provide a mechanism to move beyond steroid adminstration. Paul A. Checchia, MD, FCCM St. Louis Children’s Hospital and Washington University School of Medicine St. Louis, MO Ronald A. Bronicki, MD Children’s Hospital of Orange County, California Orange, CA REFERENCES 1. Kirklin JK, Westaby S, Blackstone EH, et al: Complement and the damaging effects of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1983; 86:845– 857 2. Holmes JHT, Connolly NC, Paull DL, et al: Magnitude of the inflammatory response to cardiopulmonary bypass and its relation to adverse clinical outcomes. Inflamm Res 2002; 51:579 –586 3. Christman JW, Lancaster LH, Blackwell TS: Nuclear factor kappa B: A pivotal role in the systemic inflammatory response syndrome and new target for therapy. Intensive Care Med 1998; 24:1131–1138 4. Baldwin AS Jr: The NF-kappa B and I kappa B proteins: New discoveries and insights. Annu Rev Immunol 1996; 14:649 – 683 5. Qing M, Schumacher K, Heise R, et al: Intramyocardial synthesis of pro- and antiinflammatory cytokines in infants with congenital cardiac defects. J Am Coll Cardiol 2003; 41:2266 –2274 6. Liakopoulos OJ, Schmitto JD, Kazmaier S, et al: Cardiopulmonary and systemic effects of methylprednisolone in patients undergoing cardiac surgery. Ann Thorac Surg 2007; 84: 110 –118; discussion 118 –119 7. Bronicki RA, Backer CL, Baden HP, et al: Dexamethasone reduces the inflammatory response to cardiopulmonary bypass in children. Ann Thorac Surg 2000; 69:1490 –1495 8. Checchia PA, Backer CL, Bronicki RA, et al: Dexamethasone reduces postoperative troponin levels in children undergoing cardiopul- Crit Care Med 2009 Vol. 37, No. 2 monary bypass. Crit Care Med 2003; 31: 1742–1745 9. Moen O, Hogasen K, Fosse E, et al: Attenuation of changes in leukocyte surface markers and complement activation with heparincoated cardiopulmonary bypass. Ann Thorac Surg 1997; 63:105–111 10. Seghaye MC, Duchateau J, Grabitz RG, et al: Complement activation during cardiopulmonary bypass in infants and children. Relation to postoperative multiple system organ failure. J Thorac Cardiovasc Surg 1993; 106:978 –987 11. Checchia PA, Bronicki RA, Costello JM, et al: Steroid use before pediatric cardiac opera- tions using cardiopulmonary bypass: An international survey of 36 centers. Pediatr Crit Care Med 2005; 6:441– 444 12. Yeh TF, Lin YJ, Lin HC, et al: Outcomes at school age after postnatal dexamethasone therapy for lung disease of prematurity. N Engl J Med 2004; 350:1304 –1313 13. Duffy JY, McLean KM, Lyons JM, et al: Modulation of nuclear factor-kappaB improves cardiac dysfunction associated with cardiopulmonary bypass and deep hypothermic circulatory arrest. Crit Care Med 2009; 37:577–583 14. Yeh CH, Chen TP, Wu YC, et al: Inhibition of NFkappaB activation with curcumin attenu- ates plasma inflammatory cytokines surge and cardiomyocytic apoptosis following cardiac ischemia/reperfusion. J Surg Res 2005; 125:109 –116 15. Schwartz SM, Duffy JY, Pearl JM, et al: Glucocorticoids preserve calpastatin and troponin I during cardiopulmonary bypass in immature pigs. Pediatr Res 2003; 54:91–97 16. Yeh CH, Chen TP, Lee CH, et al: Cardioplegia-induced cardiac arrest under cardiopulmonary bypass decreased nitric oxide production which induced cardiomyocytic apoptosis via nuclear factor kappa B activation. Shock 2007; 27:422– 428 “Pas de DEux” for phosphodiesterase-2 in acute lung injury* T he phosphodiesterase (PDE)-2 isozyme was first discovered as a cyclic guanosine monophosphate (cGMP)-stimulated cyclic nucleotide PDE2 capable of hydrolyzing both cyclic adenosine monophosphate (cAMP) and cGMP (1). Three variants of the unique PDE2 gene have been described. All of them are specifically inhibited by 9-(2hydroxy-3-nonyl) adenine monohydrocrochloride (EHNA) when PDE2 is activated by cGMP in vitro. PDE2 is expressed in both endothelial and pulmonary epithelial cells (2) and, beyond membrane receptors, this isozyme regulates cyclic nucleotide levels (cAMP and cGMP) and several highly localized intracellular signaling pathways such as cAMP dependent protein kinase and cGMP dependent protein kinase-dependent phosphorylation. Furthermore, when regulating local cAMP and cGMP, PDE2 controls the (adenosine triphosphate)/(cAMP) energetic ratio. PDEs are ubiquitous enzymes constituted by a multigenic super family (PDE1–11) (3). They have been involved in various feedback processes, and PDE inhibitors have been developed and used as bronchorelaxant (theophylline), cardiotonic (milrinone), anti-inflammatory (rolipram), and vasodilating drugs (such as sildenafil). The putative role of PDE2 *See also p. 584. Key Words: phosphodiesterases; inhibitors; acute lung injury; lung; permeability; pneumonia The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c295 Crit Care Med 2009 Vol. 37, No. 2 in inflammatory processes or infectious diseases has been evoked, although evidences have been somewhat scant. Hence, the classic PDE2 inhibitor EHNA has been mainly used in in vitro assays, because its poor specificity has precluded its use in in vivo studies. PDE2s, which are mainly cytosolic as well as particulate (membrane attached) enzymes, are likely surrounded by a complex microenvironment in vivo, according to intracellular molecular crowding. Consequently, it would not be surprising that PDE2s are associated with specific transmembrane receptors within specific microdomains. Such intracellular location would facilitate the control of local concentrations of cAMP (or cGMP) in the vicinity of the molecular complexes. Conversely, since PDE2 hydrolyzing activity is stimulated by cGMP (5 ␮M) via an allosterically regulated site, the presence of a guanylate cyclase in contiguity with the signaling complex could also represent a hypothesis to be ascertained. Using combined classic methodologic approaches and an array of pharmacologic conditions involving two new PDE2 inhibitors, namely PDP and hydroxylPDP, Witzenrath et al in this issue of Critical Care Medicine (4) demonstrated that prophylactic and systemic infusion of these specific PDE2 inhibitors reduced acute lung injury (ALI) in a mice model of pneumococcal infection. This noteworthy observation was also supported by a constellation of preliminary and original results obtained in endothelial cell monolayers (5). Hence, both PDE inhibitors were able to oppose the electrophysiologic effects of pneumococcal pneumo- lysin in endothelial cell (human umbilical veinous endothelial cells [HUVEC]) monolayers. Thus, PDE2 appears as a key target in this process, since the contribution of other PDE isozymes was elegantly ruled out in control experiments. Despite a relevant set of preliminary data gained on endothelial cell monolayers in which PDE2 expression was stimulated by exogenous tumor necrosis factor-␣ (4), this study partially performed in a murine model and in isolated lungs did not allow to pinpoint the precise location of the inhibited PDE2. In addition, other more specific types of endothelial cell cultures, such as lung microvascularderived cells, should have been favorably compared with HUVEC. Furthermore, although endothelial cells are likely candidates, the effect of PDE2 inhibitors on epithelial cell layers at the infection site (nasal mucosa) or within the respiratory track may have been underestimated (6). In this respect, although the selection of electrical resistance of endothelial cell (HUVEC) and human serum albumin leakage certainly represented a sound choice, other tools were not considered such as small molecular weight molecule (e.g., Evans blue, fluorescein) or microprotein (e.g., Clara cell protein 16) leakages (7), for assessing epithelial damage. Histopathologic examination of lung tissues should have also served for analysis of interstitial edema analysis to further consolidate PDE2 inhibition-induced decreased permeability. Sepsis-induced tissue and lung hyperpermeability is the “underdog” in terms of research on mechanistic events leading to ALI/acute respiratory distress syndrome. 769 Indeed, hyperpermeability is a neglected but important event leading to sepsisinduced organ disability, due to a lack of specific tools to assess this particular phenomenon. However, it is a potentially leading cause of organ dysfunction/failure, either by impairing exchange ability (e.g., lung, kidney) or by compromising organ perfusion, especially for capsule- or membrane-bearing tissues or those exhibiting bone-limited compliances (e.g., kidney, adrenal gland, liver, brain) (8). However, as stated above, microvascular permeability is likely only half of the story, especially with regard to organ exchanges such as in lung or kidney. Indeed, the epithelium should be considered as the other side of the coin, since it is clearly committed in lung barrier permeability (9). Water and solutes (i.e., sodium), through aquaporins and eNac pumps (in conjunction with Na⫹-K⫹adenosine triphosphatase basal pumps), are able to clear excess permeability in lung distal airspaces in ALI (10, 11), and therefore have to be taken into account together with endothelium impairment. Finding inhibitors of tissue permeability is a challenging issue requiring extensive research, which in return could yield invaluable applications at least for combination therapies for ALI at the bedside. This issue should have been explored in the study by Witzenrath et al (4), at least by specific in vitro lung epithelial cultures or optimally by epithelial-endothelial cocultures. In summary, the data described by Witzenrath et al (4) undoubtedly provide an important addition to our current knowledge on the role of PDE2 in host infection by Streptocococcus pneumoniae. This isozyme may indeed regu- 770 late key cell signaling processes, whose activation would be required to facilitate the early steps directly involved in the infectious cascade triggered by S. pneumoniae strains. In an explosion of interest for this concept, a rapid succession of studies will likely identify the missing links, the next step being the identification of the membrane receptor upstream (such as tolllike receptor) and the cyclic nucleotidesensitive effectors, downstream of PDE2 in either endothelial or epithelial cell layers. An alternative issue would be that low levels of cAMP would facilitate interaction of S. pneumoniae exotoxin proteins with specific membrane receptors, whereas highly localized increases in cAMP levels (on PDE2 inhibition) would abolish this early step. Despite the fact that the role of proinflammatory cytokines cannot be ruled out, clinical teams will likely focus on the ability of new specific PDE2 inhibitors to decrease morbidity in ALIrelated conditions to sever S. pneumoniae in curative protocols. Olivier Lesur, MD, PhD MICU, Department of Medicine CHU Sherbrooke-MICU Sherbrooke, Quebec, Canada Eric Rousseau, PhD Department of Physiology and Biophysics, FMSS Université of Sherbrooke Quebec, Canada 3. 4. 5. 6. 7. 8. 9. 10. REFERENCES 1. Beavo JA, Hardman JG, Sutherland EW: Stimulation of adenosine 3⬘,5⬘- monophosphate hydrolysis by guanosine 3⬘,5⬘-monophosphate. J Biol Chem 1971; 246:3841–3846 2. Rousseau E, Gagnon J, Lugnier C: Biochem- 11. ical and pharmacological characterisation of cyclic nucleotide phosphodiesterase in airway epithelium. Mol Cell Biochem 1994; 140: 171–175 Keravis T, Lugnier C: Cyclic nucleotide phosphodiesterase (PDE) superfamily and smooth muscle signaling. In: New Frontier in Smooth Muscle Biology and Physiology. Chap.13. Savineau JP (Ed), Transworld Research Network, Kerala, India, 2007, pp 269 –289 Witzenrath M, Gutbier B, Schmeck B, et al: Phosphodiesterase 2 inhibition diminished acute lung injury in murine pneumococcal pneumonia. Crit Care Med 2009; 37:584 –590 Seybold J, Thomas D, Witzenrath M, et al: Tumor necrosis factor-␣-dependent expression of phosphodiesterase 2: Role in endothelial hyperpermeability. Blood 2005; 105:3569 –3576 Boswell-Smith V, Spina D, Page CP: Phosphodiesterase inhibitors [review]. Br J Pharmacol 2006; 147(Suppl 1):S252–S257 Lesur O, Hermans C, Chalifour JF, et al: Pneumoprotein (CC-16) vascular transfer during mechanical ventilation in rats: Effect of KGF pretreatment. Am J Physiol Lung Cell Mol Physiol 2003; 284:L410 –L419 Farand P, Hamel M, Lauzier F, et al: Effects of norepinephrine and vasopressin on sepsisinduced organ perfusion/permeability [review]. Can J Anaesth 2006; 53:934 –946 Kim K-J, Malik AB: Protein transport across the lung epithelial barrier. Am J Physiol Lung Cell Mol Physiol 2003; 284: L247–L259 Zemans RL, Matthay MA: Bench-to-bedside review: The role of the alveolar epithelium in the resolution of pulmonary edema in ALI. Crit Care 2004; 8:469 – 477 Verkman AS, Matthay MA, Song Y: Aquaporins water channels and lung physiology. Am J Physiol Lung Cell Mol Physiol 2000; 278:L867–L879 Crit Care Med 2009 Vol. 37, No. 2 Hemorrhagic shock and reperfusion injury: The critical interplay of fibrin fragments, leukocytes, and vascular endothelial-cadherin* R eperfusion, the restoration of blood flow after a period of ischemia, can place ischemic organs at risk of further cellular necrosis and thereby limit the recovery of function. In particular, the microvasculature is vulnerable to the deleterious consequences of ischemia and reperfusion (I/R). The underlying pathophysiology of I/R injury is not fully understood, but several mechanisms are involved such as membrane damage inflicted by oxygen radicals, intracellular calcium overload, and tissue damage caused by infiltration and activation of white blood cells (1, 2). Reperfusion of ischemic tissues is often associated with microvascular dysfunction that is manifested as impaired endothelium-dependent dilation in arterioles, enhanced fluid filtration, leukocyte plugging in capillaries, and the trafficking of leukocytes and plasma protein extravasation in postcapillary venules. The resulting imbalance between superoxide and nitric oxide in endothelial cells leads to the production and release of inflammatory mediators (e.g., platelet-activating factor, tumor necrosis factor) and enhances the biosynthesis of adhesion molecules that mediate leukocyte– endothelial cell adhesion. The bioavailability of nitric oxide, an important mediator of vasodilatation, is profoundly decreased during the reperfusion period, resulting in impaired vasodilatation of arterioles. Activated endothelial cells in all segments of the microcirculation produce more oxygen radicals, but less nitric oxide, in the initial period following reperfusion (3). Induction of complement activation and subsequent release of inflammatory mediators as well as increased *See also p. 598. Key Words: B␤15-42; fibrin-related peptide; FX06; hemorrhagic shock; ischemia; organ failure; reperfusion The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194bd9e Crit Care Med 2009 Vol. 37, No. 2 expression of adhesion molecules initiate inflammatory and coagulation cascades that culminate in the occlusion of capillaries, known as the “no-reflow” phenomenon. In postcapillary venules, the recruitment and transmigration of leukocytes further compromise the integrity of the endothelial barrier and increase the oxidative burden, resulting in leakage and tissue edema (4). The trafficking of leukocytes in venules is the rate-limiting determinant of I/R-induced endothelial barrier dysfunction. The magnitude of albumin leakage in postischemic venules is highly correlated with the number of adherent and emigrated leukocytes. Adhesion moleculedirected antibodies that effectively blunt leukocyte adherence/emigration also exert an attenuating action on I/R-induced albumin leakage (5). Leukocytes attach to the blood vessel wall via interaction of selectins and integrins with their respective receptors, and become progressively activated. To accomplish the final step of transmigration through the endothelial layer, leukocytes must cross a multilayered molecular zipper of cell– cell junctions. Leukocyte transmigration is critically controlled by fibrin fragments. These short-lived intermediates engage leukocytes to the endothelial cell junction, which breaks up the VEcadherin interaction between neighboring endothelial cells, allowing the leukocytes to pass through into the tissue (diapedesis). In myocardial infarction, about 50% of overall tissue damage is caused by reperfusion injury. For several decades, it has been known that the inflammatory response during reperfusion is the major cause of myocardial I/R injury. If uncontrolled, the inflammation results in irreversible damage of the heart muscle, limiting the success of reperfusion procedures. The inflammatory reaction can be mitigated by application of B␤1542, also called FX06, concomitant with reperfusion (6 –9). FX06, is a polypeptide, derived from the neo-N-terminus of fibrin, being the natural cleavage product of fibrin after being exposed to plasmin. FX06 targets an endothelial adherens junction protein, VE-cadherin, and reduces leukocyte transmigration across endothelial junctions and the release of proinflammatory cytokines (6 –9). Therefore, FX06 prohibits the emigration of all leukocyte subtypes into myocardial tissue by preventing the critical interaction between fibrin fragments and VE-cadherin, which is rate limiting and irreversible. Previous data on acute and chronic myocardial I/R models were promising and appeared to reduce myocardial inflammation and infarct size (6 –9). In occlusion–reperfusion studies in rats, FX06 (2.4 mg/kg optimal dose) caused 40% reduction in infarct size (6). The positive effects of FX06 on the inflammatory response were reflected by a decrease in cytokines (interleukin [IL]-6, troponin). The safety and tolerability of FX06 has been demonstrated in a phase I trial in 30 healthy male volunteers, as no significant toxicity and adverse events were observed (7). A phase II trial on the treatment with FX06 for reperfusion injury in myocardial infarction (F.I.R.E. trial) has been completed recently and is designed to clinically evaluate infarct size at 5–7 days post-percutaneous coronary intervention (www.clinicaltrial.gov) (10). In this issue of Critical Care Medicine, Roesner et al (11) report on the beneficial effects of this novel endogenous polypeptide FX06 in a hemorrhagic shock model. Hemorrhagic shock followed by volume resuscitation largely resembles the state of myocardial I/R. The key treatment of hemorrhagic shock involves early hemorrhage control, aggressive resuscitation, and the prevention and correction of coagulopathy (12). The current management of hemorrhagic shock relies heavily on transfusion of red blood cells that are associated with the development of multiple organ failure, increased intensive care unit admissions and length of stay, increased hospital length of stay, and mortality (13). Other therapeutic options are limited because of lack of understanding the pathophysiological mechanisms and lack of a specific target. Some novel 771 (experimental) interventions to limit reperfusion damage have some promise such as sodium/hydrogen exchange suppression (14), melatonin (15), IL-11 (16), or selectin inhibition (17), but failed to reach the clinical stage. In this context, the salutary effects of the compound FX06 on hemorrhagic shock-induced organ injury are potentially important and clinically relevant (11). The controlledshock model was characterized by rather severe tissue damage (I/R and laparotomy), a long (5 hours) reperfusion time, and adequate volume resuscitation guided by volumetric preload parameters. FX06-treated animals showed a reduction in myocardial damage and leukocyte infiltration in the heart, as well as less liver/ intestinal damage. The effects on lung damage are less clear; extra vascular lung water and PaO2/FIO2 ratios were only improved at the end of reperfusion. Apparently, IL-6 may be a key factor, as the reduced organ failure and inflammation strongly correlated with a reduction in IL-6 in the FX06 group. These intriguing results are very promising and will certainly lead to further preclinical research in I/R injury. The translation of the findings in the hemorrhagic shock reperfusion model with limitations toward the complex clinical traumatic shock setting with far more tissue injury is too premature. Some questions remain unclear; whether an endogenous HX06 deficit is present during shock or resuscitation or whether multiple (with respect to the very short half-life of 15 minutes) or higher doses of FX06 may be superior. The exact mechanisms involved in the observed improved organ function remain obscure, but similar to previous 772 findings on myocardial I/R damage, the impact of FX06 on neutrophil organ infiltration seems to be most prominent. Albertus Beishuizen, MD Armand R. J. Girbes, MD Department of Intensive Care University Hospital VU Medical Center Amsterdam, The Netherlands 10. REFERENCES 1. Liem DA, Honda HM, Zhang J, et al: Past and present course of cardioprotection against ischemia-reperfusion injury. J Appl Physiol 2007; 103:2129 –2136 2. Carden DL, Granger DN: Pathophysiology of ischaemia-reperfusion injury. J Pathol 2000; 190:255–266 3. Grisham MB, Granger DN, Lefer DJ: Modulation of leukocyte-endothelial interactions by reactive metabolites of oxygen and nitrogen: Relevance to ischemic heart disease. Free Radic Biol Med 1998; 25:404 – 433 4. Granger DN: Ischemia-reperfusion: Mechanisms of microvascular dysfunction and the influence of risk factors for cardiovascular disease. Microcirculation 1999; 6:167–178 5. Kurose I, Anderson DC, Miyasaka M, et al: Molecular determinants of reperfusioninduced leukocyte adhesion and vascular protein leakage. Circ Res 1994; 74:336 –343 6. Petzelbauer P, Zacharowski PA, Miyazaki Y, et al: The fibrin-derived peptide Bbeta15-42 protects the myocardium against ischemiareperfusion injury. Nat Med 2005; 11: 298 –304 7. Roesner JP, Petzelbauer P, Koch A, et al: The fibrin-derived peptide Bbeta15-42 is cardioprotective in a pig model of myocardial ischemia-reperfusion injury. Crit Care Med 2007; 35:1730 –1735 8. Zacharowski K, Zacharowski P, Reingruber S, et al: Fibrin(ogen) and its fragments in the pathophysiology and treatment of myocardial infarction. J Mol Med 2006; 84:469 – 477 9. Zacharowski K, Zacharowski PA, Friedl P, et 11. 12. 13. 14. 15. 16. 17. al: The effects of the fibrin-derived peptide Bbeta(15-42) in acute and chronic rodent models of myocardial ischemia-reperfusion. Shock 2007; 27:631– 637 Atar D, Huber K, Rupprecht HJ, et al: Rationale and design of the ‘F.I.R.E.’ study. A multicenter, double-blind, randomized, placebocontrolled study to measure the effect of FX06 (a fibrin-derived peptide Bbeta(15– 42)) on ischemia-reperfusion injury in patients with acute myocardial infarction undergoing primary percutaneous coronary intervention. Cardiology 2007; 108:117–123 Roesner JP, Petzelbauer P, Koch A, et al: B␤15-42 (FX06) reduces pulmonary, myocardial, liver, and small intestine damage in a pig model of hemorrhagic shock and reperfusion. Crit Care Med 2009; 37:598 – 605 Kauvar DS, Lefering R, Wade CE: Impact of hemorrhage on trauma outcome: An overview of epidemiology, clinical presentations, and therapeutic considerations. J Trauma 2006; 60:S3–S11 Malone DL, Dunne J, Tracy JK, et al: Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma 2003; 54:898 –905 Buerke U, Pruefer D, Carter JM, et al: Sodium/hydrogen exchange inhibition with cariporide reduces leukocyte adhesion via Pselectin suppression during inflammation. Br J Pharmacol 2008; 153:1678 –1685 Mathes AM, Kubulus D, Pradarutti S, et al: Melatonin pretreatment improves liver function and hepatic perfusion after hemorrhagic shock. Shock 2008; 29:112–118 Honma K, Koles NL, Alam HB, et al: Administration of recombinant interleukin-11 improves the hemodynamic functions and decreases third space fluid loss in a porcine model of hemorrhagic shock and resuscitation. Shock 2005; 23:539 –542 Calvey CR, Toledo-Pereyra LH: Selectin inhibitors and their proposed role in ischemia and reperfusion. J Invest Surg 2007; 20: 71– 85 Crit Care Med 2009 Vol. 37, No. 2 Peroxisome proliferator-activated receptor-gamma agonists, control of bacterial outgrowth, and inflammation* I n the 1980s, at a time when few cytokines were described, a new paradigm emerged that sepsis may be a syndrome that consists of an inappropriate and maladaptive systemic inflammatory response induced in response to infection. Numerous attempts were made to treat sepsis in the clinic using immune modulators that block proinflammatory cytokine mediators such as tumor necrosis factor or interleukin-1, or other components of innate immunity. Ultimately, all of these downregulating immune modulators failed in clinical trials to treat sepsis. Despite their failure in the acute setting of infection, many of the same antiinflammatory treatments are now proven to be effective treatments for chronic inflammatory conditions such as rheumatoid arthritis and Crohn’s disease. Ironically, many of the treatments that were initially developed to treat sepsis have, in fact, turned out to predispose to infection and sepsis (1). This finding, coupled with data in animal experiments, suggests that blocking even single components of the innate immune system runs the risk of altering host defense sufficiently to compromise the elimination of microorganisms. For this reason, agents that suppress the inflammatory response in the setting of infection have traditionally been studied for use as adjuvant treatment in addition to primary antimicrobial treatment. Peroxisome proliferator-activated receptor-gamma (PPAR␥) is a member of the nuclear receptor family of transcription factors that mediates transcriptional *See also p. 614. Key Words: peroxisome proliferator-activated receptor-␥; inflammation; bacteria; mouse model; pneumonia The author has received grant support from the National Institute of Health AI059010, GM 59694, and the Shriners Hospital for Crippled Children 8720. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c4f9 Crit Care Med 2009 Vol. 37, No. 2 activation or repression of genes related to lipid metabolism, cell proliferation, angiogenesis, and inflammation. PPAR␥ agonist ligands diminish insulin resistance and are used to help treat type 2 diabetes. PPAR␥ ligands also repress the gene for nitric oxide synthase, as well as tumor necrosis factor, interleukin-6, and interleukin-1 (2, 3). Because of these and other anti-inflammatory activities, it seems reasonable to study their potential use as an anti-inflammatory agent in the setting of infection. In this issue of Critical Care Medicine, Stegenga et al (4) report a study in which a synthetic PPAR␥ ligand, ciglitazone, was evaluated in an established mouse model of pneumonia with Streptococcus pneumoniae. The drug was administered to groups of mice either once at the start of the infectious challenge or twice with the first dose at challenge and a second dose at 24 hrs. The mice were evaluated at 24 and 48 hrs after infectious challenge. Mice that received ciglitazone had decreased bacterial counts in the lungs compared with controls; in addition, there were decreased inflammatory cytokines in lung tissues at 24 hrs and decreased inflammation on histopathology at 48 hrs. There was also a trend toward a decreased mortality in the drug group. It seems remarkable that a drug that possesses anti-inflammatory effects would decrease bacterial outgrowth. Because the innate immune system presumably evolved to control infection, one might expect the opposite effect. Indeed specific inhibition of the interleukin-1 or tumor necrosis factor in a similar model has been reported to lead to increased bacterial loads and decreased survival (5, 6). Therefore, the results reported in the current study in this model with the PPAR␥ ligand ciglitazone seem the best of all situations for an anti-inflammatory drug— decreased inflammation and at the same time, decreased bacteria. The mechanism of the drug’s effect is not clear. There are several possibilities that the authors have addressed. First, there could be some sort of direct antibacterial action if the drug was acting as an antibiotic. There was apparently no antibacterial effect when the authors tested the drug in vitro at 5 ␮g/mL. However, ciglitazone was administered to the mice at 5 mg/kg; depending on the pharmacokinetics, it is possible that drug levels could have exceeded 5 ␮g/mL for some period of time. Second, a decrease in bacterial proliferation could be explained if ciglitazone had stimulated bacterial clearance in some way. PPAR␥ ligands have been reported to increase phagocytosis through increasing expression of CD36 (7), as well as through Fc␥ (8). The authors did not detect an effect of the drug using a murine alveolar cell line to assess phagocytosis of S. pneumoniae labeled with fluorescein isothiocyanate, and did not detect increased killing in these cells using a capsuledeficient strain of S. pneumoniae. It is unclear, however, how well this in vitro system with a cell line reflects the in vivo situation in the mouse model using an encapsulated strain of S. pneumoniae. The results of Stegenga et al with S. pneumoniae are consistent with two other publications that reported that ciglitazone decreased inflammation without an increase in bacterial counts. One of these used the cecal ligation and puncture model (9), and the other a model of brain abscess with Staphylococcus aureus (10). Thus, the results may extend to organisms other than S. pneumoniae. Interpretation of decreased inflammation in the setting of decreased bacterial load is complex. It is not clear from the present study if the antibacterial effect and the decreased inflammation stem from two different activities of the drug, one leading to decreased inflammation and the other to decreased bacterial counts. An alternative and simpler explanation is that the decreased inflammation may follow simply from the presence of fewer bacteria in tissues leading to less of a stimulus to the host. 773 As always, there are many cautions in extrapolating results of animal models to human disease. Mice are much more resistant than humans to most forms of induced inflammation. It will be essential to determine the mechanism of the antibacterial effect and to determine whether there is any additional advantage when ciglitazone is used as adjuvant treatment in addition to simultaneous traditional antibiotics. Regardless, this study is provocative and raises some hope that PPAR␥ ligands could be a new approach to an area that could use a breakthrough in treatment. H. Shaw Warren, MD Infectious Disease Unit Massachusetts General Hospital Boston, MA REFERENCES 1. Bongartz T, Sutton AJ, Sweeting MJ, et al: Anti-TNF antibody therapy in rheumatoid ar- 2. 3. 4. 5. 6. thritis and the risk of serious infections and malignancies: Systematic review and metaanalysis of rare harmful effects in randomized controlled trials. J Am Med Assoc 2006; 295:2275–2285 Ricote M, Li AC, Willson TM, et al: The peroxisome proliferator-activated receptorgamma is a negative regulator of macrophage activation. Nature 1998; 391:79 – 82 Jiang C, Ting AT, Seed B: PPAR-gamma agonists inhibit production of monocyte inflammatory cytokines. Nature 1998;391:82– 86 Stegenga ME, Florquin S, de Vos M, et al: The thiazolidinedione ciglitazone reduces bacterial outgrowth and early inflammation during Streptococcus pneumoniae pneumonia in mice. Crit Care Med 2009; 37:614 – 618 Rijneveld AW, Florquin S, Hartung T, et al: Anti-tumor necrosis factor antibody impairs the therapeutic effect of ceftriaxone in murine pneumococcal pneumonia. J Infect Dis 2003; 188:282–285 Rijneveld AW, Florquin S, Branger J, et al: TNF-alpha compensates for the impaired host defense of IL-1 type I receptor-deficient 7. 8. 9. 10. mice during pneumococcal pneumonia. J Immunol 2001; 167:5240 –5246 Asada K, Sasaki S, Suda T, et al: Antiinflammatory roles of peroxisome proliferatoractivated receptor gamma in human alveolar macrophages. Am J Respir Crit Care Med 2004; 169:195–200 Aronoff DM, Serezani CH, Carstens JK, et al: Stimulatory effects of peroxisome proliferator-activated receptor-gamma on Fcgamma receptor-mediated phagocytosis by alveolar macrophages. PPAR Res 2007; 2007:52546 Zingarelli B, Sheehan M, Hake PW, et al: Peroxisome proliferator activator receptorgamma ligands, 15-deoxy-Delta(12,14)prostaglandin J2 and ciglitazone, reduce systemic inflammation in polymicrobial sepsis by modulation of signal transduction pathways. J Immunol 2003; 171: 6827– 6837 Kielian T, Syed MM, Liu S, et al: The synthetic peroxisome proliferator-activated receptor-gamma agonist ciglitazone attenuates neuroinflammation and accelerates encapsulation in bacterial brain abscesses. J Immunol 2008; 180:5004 –5016 Look on the “air side” in pneumonia* P neumonia is the leading cause of acute lung injury (ALI) (1). The pathophysiology of ALI involves an uncontrolled host defense response, with a malignant alveolar cross talk between inflammation and hemostasis activation (2), which further propagates as the natural anticoagulant axis is disrupted (3). Despite the use of early broad-spectrum antibiotics, the mortality of pneumonia is still high with an “attributable mortality” from hospital-acquired Grampositive and Gram-negative pneumonia of 33% to 50% (4). One explanation is that the mortality is not only attributed to the infection per se but also to the antigens of the bacteria. In this issue of Critical Care Medicine, Hoogerwerf et al (5) shows for the first time *See also p. 619. Key Words: pneumonia; inhaled activated protein C; Gram-positive antigen; Gram-negative antigen; diffuse alveolar damage; pulmonary deposition; alveolar hemostasis activation The author is a CSO x1 in a biotech company that owns the patent on inhaled APC. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194cf47 774 that antigens from cell walls of both Grampositive and Gram-negative bacteria have the same effect in humans. Furthermore, they also showed that alveolar levels of antithrombin and activated protein C (APC) concentrations were reduced. This was meticulously documented by this randomized study of instillation of saline, lipoteichoic acid, or lipopolysaccharide, i.e., applying the antigens from the “air side.” Their findings are not self-explanatory in as much as Grampositive and Gram-negative pathogens actuate immune and procoagulant responses in the lung via different recognition receptors, i.e., via toll-like receptors 2 and 4, respectively. Severe pneumonia in hospitals has still a high mortality despite applying evidence-based early use of broadspectrum antibiotics. As documented by Hoogerwerf et al, the problem is that the pulmonary dysfunction in pneumonia is further aggravated by the antigens of the bacterial cell walls. How do we then offset the changes induced from the air side, and, further does pharmacotherapy with natural anticoagulants and fibrinolytics have a role in the treatment of ALI? It is logical to in- troduce such a therapy using natural anticoagulants because the intra-alveolar hemostasis activation is not sufficiently counterbalanced by natural inhibitors, such as tissue factor pathway inhibitor (6) and APC (7, 8). Tissue factor pathway inhibitor was shown to reduce lung injury and systemic levels of inflammatory cytokines in an animal study (9), and, further, in a subsequent phase II trial, the results were promising. However, this benefit was lost in a subsequent phase III trial (10). At present, we are awaiting the results from a placebo-controlled trial in communityacquired pneumonia using intravenous tissue factor pathway inhibitor (11). Since the introduction of the recombinant APC for the treatment of severe sepsis (Drotrecogin alpha activated) (12), several studies have used APC intravenously to counteract the hemostatic changes in ALI (8). Recently, a controlled trial using intravenous APC in patients with acute lung injury was published (13). The study revealed that APC did not improve outcomes of acute lung injury and concluded that the results “do not support a large clinical trial of APC in acute lung injury.” Furthermore, intraveCrit Care Med 2009 Vol. 37, No. 2 nous APC has been shown to result in noteworthy bleeding as alluded to in two recent articles (14, 15). The adverse effects of systemic bleeding may be avoided if APC is inhaled and provided that there is no spillover from the alveoli to the systemic circulation (16). From a theoretical viewpoint, inhaled APC has all the essential properties to counteract the pathophysiologic changes seen in ALI. In an animal study by Slofstra et al (17), the lipopolysaccharideinduced inflammation was reduced after inhalation of APC. A human experience with inhaled APC has been reported in a patient with ALI (16), with an important improvement in pulmonary gas exchange concomitant with a noteworthy clearing of the pulmonary opacities. There is thus a good reason to “look at the pneumonia from the air side” not only with respect to explaining the acute pathophysiologic changes ensuing pneumonia, but also treating the condition from the same side by inhaling a natural anticoagulant, such as APC. APC may be an important adjunctive intervention in severe pneumonia caused by both Gram-positive and Gram-negative bacteria. There is, however, only sparse information to substantiate such a new treatment paradigm. The next step, therefore, is to conduct controlled trials with inhaled APC because the intravenous intervention has already proved ineffective. Lars Heslet, MD, MSc, DMedSci Department of Intensive Care Rigshospitalet Copenhagen, Denmark Crit Care Med 2009 Vol. 37, No. 2 REFERENCES 1. Gunther A, Mosavi P, Heinemann S, et al: Alveolar fibrin formation caused by enhanced procoagulant and depressed fibrinolytic capacities in severe pneumonia. Comparison with the acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161: 454 – 446 2. Ware LB: Pathophysiology of acute lung injury and the acute respiratory distress syndrome. Semin Respir Crit Care Med 2006; 27:337–349 3. Schultz MJ, Haitsma JJ, Zhang H, et al: Pulmonary coagulopathy as a new target in therapeutic studies of acute lung injury or pneumonia: A review. Crit Care Med 2006; 34: 871– 877 4. Isakow W, Kollef MH: Preventing ventilatorassociated pneumonia: An evidence-based approach of modifiable risk factors. Semin Respir Crit Care Med 2006; 27:5–17 5. Hoogerwerf JJ, de Vos AF, Levi M, et al: Activation of coagulation and inhibition of fibrinolysis in the human lung on bronchial instillation of lipoteichoic acid and lipopolysaccharide. Crit Care Med 2009; 37:619 – 625 6. Gando S, Kameue T, Matsuda N, et al: Imbalance between the levels of tissue factor and tissue factor pathway inhibitor in ARDS patients. Thromb Res 2003; 109: 119 –124 7. Ware LB, Fang X, Matthay MA: Protein C and thrombomodulin in human acute lung injury. Am J Physiol Lung Cell Mol Physiol 2003; 285:L514 –L521 8. van der Poll T, Levi M, Nick JA, et al: Activated protein C inhibits local coagulation after intrapulmonary delivery of endotoxin in humans. Am J Respir Crit Care Med 2005; 171:1125–1128 9. Welty-Wolf KE, Carraway MS, Miller DL, et 10. 11. 12. 13. 14. 15. 16. 17. al: Coagulation blockade prevents sepsisinduced respiratory and renal failure in baboons. Am J Respir Crit Care Med 2001; 164:1988 –1996 Abraham E, Reinhart K, Opal S, et al: Efficacy and safety of tifacogin (recombinant tissue factor pathway inhibitor) in severe sepsis: A randomized controlled trial. JAMA 2003; 290:238 –247 LifeSciencesWorld: Chiron begins phase III trial for tifacogin. May 18, 2004. Available at: www.lifesciencesworld.com/rss/tifacogin/ news. Accessed October 12, 2006 Bernard GR, Vincent JL, Laterre PF, et al; Recombinant human protein C Worldwide Evaluation in Severe Sepsis (PROWESS) study group: Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344: 699 –709 Liu KD, Levitt J, Zhuo H, et al: Randomized clinical trial of activated protein C for the treatment of acute lung injury. Am J Respir Crit Care Med 2008; 178:618 – 623 Martí-Carvajal A, Salanti G, Cardona AF: Human recombinant activated protein C for severe sepsis. Cochrane Database Syst Rev 2008; (1):CD004388 Levi M: Activated protein C in sepsis: A critical review. Curr Opin Hematol 2008; 15: 481– 486 Heslet L, Andersen JS, Sengeløv H, et al: Inhalation of activated protein C: A possible new adjunctive intervention in acute respiratory distress syndrome. Targets Ther 2007; 1:1– 8 Slofstra SH, Groot AP, Maris NA, et al: Inhalation of activated protein C inhibits endotoxin-induced pulmonary inflammation in mice independent of neutrophil recruitment. Br J Pharmacol 2006; 149: 740 –746 775 The role of angiotensin-converting enzyme inhibition in endotoxin-induced lung injury in rats* T he renin-angiotensin system (RAS) plays an important role in the homeostasis of systemic blood pressure, but it can also regulate inflammation. RAS consists of the renin protease that cleaves angiotensinogen into angiotensin I (Ang I). Angiotensin-converting enzyme (ACE) is primarily expressed on the surface of pulmonary microvascular endothelial cells and cleaves Ang I into angiotensin II (Ang II) (1). ACE also inactivates the vasodilator bradykinins. The biological effects of the RAS system are mediated by Ang II on its specific receptors, Ang II receptor type 1 and Ang II receptor type 2. Production of Ang II and stimulation of Ang II receptor type 1 are responsible for vasoconstriction and endothelial dysfunction in models of systemic inflammation with endotoxin. Together with the observation that pulmonary Ang II was up-regulated in patients with acute respiratory distress syndrome (ARDS), inhibition of RAS became a potential treatment for these acute inflammatory disorders (2, 3). The discovery of a new homolog termed angiotensin-converting enzyme 2 (ACE2) revived interest and supported a potential role for RAS in the pathogenesis of lung inflammation. By counterbalancing ACE and reducing Ang II levels, ACE2 can play a protective role in sepsis-induced ARDS (4). These pathways are shown in Figure 1. Polymorphisms of the ACE gene were also the first to be associated with susceptibility and outcome of ARDS (5). In this issue of Critical Care Medicine, Haqiwara et al (6) report a beneficial role for ACE inhibition with enalapril on systemic inflammation and pulmonary in- *See also p. 626. Key Words: angiotensin; angiotensin-converting enzyme; angiotensin-converting enzyme inhibitor; acute lung injury; acute respiratory distress syndrome; sepsis; inflammation The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194cfa6 776 Angiotensinogen Renin Ang-(1-9) Ang I ACE2 ACE-I Ang-(1-7) Ang II ACE AT1R AT2R vasodilation •Inflammatory mediators •cell adhesion •cell growth cytoskeletal rearrangement Vascular permeability vasoconstriction inflammation remoddeling INJURY Figure 1. The pathways by which angiotensinogen system can participate in the regulation of vascular permeability, vasoconstriction, inflammation, remodeling, and vasodilation. As described in the text of the editorial, Ang II receptor type 1 (AT1R) and Ang II receptor type 2 (AT2R) identify the known receptors for angiotensin II (Ang II). ACE, angiotensin-converting enzyme; Ang I, angiotensin I. jury using a rat model of intraperitoneal endotoxin-induced inflammation. Enalapril was administered before endotoxin exposure. The findings in this study agree with the protective role of ACE inhibition reported in models of ventilator-associated and oleic acid-induced lung injury. In these models, ACE inhibition attenuated pulmonary Ang II production with a reduction of 1) inflammatory cytokines; 2) procoagulant effects; and 3) endothelial and epithelial apoptosis (7, 8). After endotoxin administration in the study by Haqiwara et al, the investigators measured increased systemic levels of Ang II that were attenuated by ACE inhibition with enalapril. However, there was no measure of systemic or local pulmonary ACE activity. In the study by Idell et al (9), ACE levels were increased in bronchoalveolar lavage fluid from patients with sepsis-associated ARDS, and plasma ACE activity was decreased in these patients. Animal models (10) of ventilator induced lung injury also demonstrated increased ACE activity in bronchoalveolar lavage fluid. However, in other models of lung injury, pulmonary endothelial ACE activity (assessed by in vivo dilution methods) was reduced as a result of enzyme downregulation caused by reactive oxygen species, as reported by Orfanos et al (11) in patients with ARDS. These findings indicate that it might be important to consider compartmentalization of RAS to understand its pathophysiology. To identify the cellular and molecular mechanisms of ACE inhibition in their model of systemic inflammation, the authors focused on the role of macrophages. Besides endothelial cells, inflammatory cells are equipped with all the components of RAS and can produce Ang II. The key finding that enalapril supCrit Care Med 2009 Vol. 37, No. 2 presses nuclear factor kappa B activation is important, given the central role of nuclear factor kappa B in the regulation of proinflammatory and proapoptic genes. This finding is supported by the systemic reduction of tumor necrosis factor-␣ and interleukin-6 with enalapril treatment. The reduction in high mobility group box protein 1 in serum, lung tissue, and macrophage supernatant was also associated with enalapril treatment. The mechanistic insights in this study are not complete because the investigators did not study the effect of ACE inhibition on other cell types in this model. Others have provided in vitro evidence that Ang II induces apoptosis in alveolar epithelial cells (12). Altered myeloperoxidase activity suggests effects on endothelial cell adhesion molecules and/or direct neutrophil effects. Also, transcription factors such as activation protein-1 and the production of proinflammatory mediators might provide additional insights into the role of RAS during acute inflammatory changes. The primary site of action of Ang II in this model is still unclear. Exploration of specific interactions with Ang II receptor type 1, as has been done in other injury models (13), or receptors of the bradykinin system, might be necessary. A role for Ang II in acute lung injury (ALI) may not be limited to the acute phase. Locally produced Ang II also plays a potential role in the fibroproliferative phase of the disease. Inhibition of the RAS system reduces procollagen production and transforming growth factor-␤ expression in human lung fibroblasts (14). It would be interesting to study the effect of early or late ACE inhibition on the fibroproliferative response in ALI. Additional preclinical animal studies will be necessary to evaluate the potential clinical value of ACE inhibition in sepsis and ALI/ARDS. The potential systemic side effects of vasodilation and systemic hypotension need to be investigated, and dose–response effects and drug admission Crit Care Med 2009 Vol. 37, No. 2 after exposure to the insult will need to be tested. One clinical study that reported outpatient use of ACE inhibitors reduced 30-day mortality for patients with community-acquired pneumonia (15). In summary, the findings of this experimental study should stimulate further preclinical studies to test the potential value of ACE inhibition as a treatment in infectious and noninfectious models of ALI (16, 17). If the results of these experimental studies continue to be promising, then a phase II clinical trial may be warranted with ACE inhibition for early ALI/ ARDS. Arne P. Neyrinck, MD Cardiovascular Research Institute University of California San Francisco, CA Michael A. Matthay, MD Cardiovascular Research Institute University of California San Francisco, CA Departments of Anesthesia and Medicine University of California San Francisco, CA REFERENCES 1. Imai Y, Kuba K, Penninger JM: The discovery of angiotensin-converting enzyme 2 and its role in acute lung injury in mice. Exp Physiol 2008; 93:543–548 2. Lund DD, Brooks RM, Faraci FM, et al: Role of angiotensin II in endothelial dysfunction induced by lipopolysaccharide in mice. Am J Physiol Heart Circ Physiol 2007; 293: H3726 –H3731 3. Wiel E, Pu Q, Leclerc J, et al: Effects of the angiotensin-converting enzyme inhibitor perindopril on endothelial injury and hemostasis in rabbit endotoxic shock. Intensive Care Med 2007; 30:1653–1659 4. Imai Y, Kuba K, Rao S, et al: Angiotensinconverting enzyme 2 protects from severe acute lung failure. Nature 2005; 436: 112–116 5. Marshall RP, Webb S, Bellingan GJ, et al: Angiotensin converting enzyme insertion/ deletion polymorphism is associated with susceptibility and outcome in acute respira- 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. tory distress syndrome. Am J Respir Crit Care Med 2002; 166:646 – 650 Hagiwara S, Iwasaka H, Matumoto S, et al: Effects of an angiotensin-converting enzyme inhibitor on the inflammatory response in in vivo and in vitro models. Crit Care Med 2009; 37:626 – 633 Chen CM, Chou HC, Wang LF, et al: Captopril decreases plasminogen activator inhibitor-1 in rats with ventilator-induced lung injury. Crit Care Med 2008; 36:1880 –1885 He X, Han B, Mura M, et al: Angiotensinconverting enzyme inhibitor captopril prevents oleic acid-induced severe acute lung injury in rats. Shock 2007; 28:106 –111 Idell S, Kueppers F, Lippmann M, et al: Angiotensin converting enzyme in bronchoalveolar lavage in ARDS. Chest 1987; 91:52–56 Wösten-van Asperen RM, Lutter R, Haitsma JJ, et al: ACE mediates ventilator-induced lung injury in rats via angiotensin II but not bradykinin. Eur Respir J 2008; 31:363–371 Orfanos SE, Armaganidis A, Glynos C, et al: Pulmonary capillary endothelium-bound angiotensin-converting enzyme activity in acute lung injury. Circulation 2000; 102: 2011–2018 Wang R, Zagariya A, Ibarra-Sunga O, et al: Angiotensin II induces apoptosis in human and rat alveolar epithelial cells. Am J Physiol Lung Cell Mol Physiol 1999; 276:L885–L889 Chan YC, Leung PS: AT1 receptor antagonism ameliorates acute pancreatitis-associated pulmonary injury. Regul Pept 2006; 134:46 –53 Marshall RP, Gohlke P, Chambers RC, et al: Angiotensin II and the fibroproliferative response to acute lung injury. Am J Physiol Lung Cell Mol Physiol 2004; 286:L156 –L164 Mortensen EM, Restrepo MI, Anzueto A, et al: The impact of prior outpatient ACE inhibitor use on 30-day mortality for patients hospitalized with community-acquired pneumonia. BMC Pulm Med 2005; 13:5–12 Su X, Robriquet L, Folkesson HG, et al: Protective effect of endogenous beta-adrenergic tone on lung fluid balance in acute bacterial pneumonia in mice. Am J Physiol Lung Cell Mol Physiol 2006; 290:L769 –L776 Looney MR, Su X, Van Ziffle JA: Neutrophils and their Fc gamma receptors are essential in a mouse model of transfusion-related acute lung injury. J Clin Invest 2006; 116: 1615–1623 777 A new approach to step on the vagal anti-inflammatory gas pedal* I n response to infection, the central nervous system initiates several anti-inflammatory pathways, designed to prevent the detrimental effects of the uncontrolled release of inflammatory mediators. One of these pathways involves an increased activity in the efferent vagus nerve, called the “nicotinic anti-inflammatory pathway,” which reflexively modifies the inflammatory response (1, 2). The most compelling evidence for role of the cholinergic nervous system in the regulation of inflammation is derived from studies on rodents challenged with endotoxin, the proinflammatory component of the outer membrane of Gram-negative bacteria (3, 4). In these studies, vagotomy led to enhanced systemic tumor necrosis factor-␣ production and accelerated the development of shock; in turn, electrical stimulation of the efferent vagus nerve downregulated tumor necrosis factor-␣ production and protected animals from hypotension (3). These findings were later confirmed and expanded in animal models of sepsis, pneumonia, and pancreatitis (5–9). Further studies showed that the anti-inflammatory properties of the efferent vagus nerve are mediated through its major neurotransmitter acetylcholine (Ach), which interacts with nicotinic Ach receptors on macrophages resulting in inhibition of endotoxininduced responses (3, 4), showing that the nicotinic Ach receptor (and not the muscarinic Ach receptor) and, specifically, the alpha7 subunit are required for this effect (4). The molecular target for this pathway has been identified as it has been shown that the vagal antiinflammatory pathway acts by alpha7 subunit-mediated Jak2-STAT3 activation in macrophages (10). On the basis of these studies, it has been suggested *See also p. 634. Key Words: vagus nerve; sepsis; endotoxin shock; anti-inflammatory; anisodamine The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194d004 778 that stimulation of nicotinic acetylcholine receptors might be beneficial in clinical syndromes involving overshoot inflammation such as septic shock, inflammatory bowel disease, or rheumatoid arthritis. Indeed in several studies, administration of nicotine or specific nicotinic Ach agonists, such as GTS-21, have been shown to reduce inflammation in various animal models (4, 5, 7). In this issue of Critical Care Medicine, Liu et al (11) evaluated the effect of pretreatment with anisodamine, a muscarinic receptor antagonist, in rodent models of endotoxic shock. This is a surprising approach because, as stated above, all reported studies on antiinflammatory compounds derived from the theory of the vagal anti-inflammatory reflex have involved ligands for nicotinic Ach receptors. Anisodamine is a compound that is derived from a Chinese medicinal herb and is chemically related to atropine. The authors state that anisodamine is widely used in the treatment of septic shock in China, although evidence from clinical studies is lacking. However, on the basis of their results there might be some truth and rationale behind the myth. In their report, Liu et al (11) show that blockade of muscarinic receptors using anisodamine results in an anti-inflammatory phenotype in endotoxin shock. Furthermore, they show that this antiinflammatory phenotype can be reversed when nicotinic receptors are blocked concurrently with muscarinic blockade. The authors hypothesize that muscarinic receptor blockade results in rerouting of Ach to nicotinic receptors resulting in increased Ach-mediated activation of nicotinic receptors and activation of the nicotinic anti-inflammatory pathway. These results are strengthened by studies involving vagotomy as well as alpha7 knockout mice because the beneficial effects of anisodamine are absent in vagotomized mice as well as alpha7 knockouts. Using in vitro studies, the authors show that indeed the binding of the selective alpha7 selective agonist bungarotoxin is increased in anisodamine-treated macrophages. Taken together, the authors demonstrate that pretreatment with anidosamine results in an anti-inflammatory phenotype, decreases shock, and increases survival. The observed effects of anisodamine are probably unrelated to muscarinic receptors per se but intimately linked to an alpha7 nicotine receptor-dependent pathway suggesting that a secondary enhanced activation of this pathway is a result of blockade of muscarinic receptors. The exact mechanism, however, can only be suggested based on the presented data and, therefore, remains to be elucidated. The study confirms the potential beneficial effects of anisodamine in the treatment of shock. However, an endotoxin model was used, and in the intensive care setting, septic shock is a different ballgame because it involves systemic circulation of live bacteria. An anti-inflammatory phenotype in genuine sepsis might result in enhanced bacterial load that may be clinically relevant. Although this study suggests that anisodamine might be of use in clinical sepsis, one should be very cautious and remember that no anti-inflammatory intervention has been shown to be beneficial in human studies. Therefore, I strongly feel that the use of this compound in septic shock, which as the authors state, is quite a common practice in China, cannot be recommended. To evaluate the clinical potential in sepsis, I would be interested in the phenotype of anisodamine treatment observed in animal models of sepsis such as cecal ligation and puncture. The most important question in studies involving the vagal anti-inflammatory reflex and pertaining to the location of the “immune synapse” remains. The exact location where the physiologic interaction between vagus nerve, derived acetylcholine and immune cells takes place has not been elucidated so far, although the spleen has been suggested (12, 13). Finally, this study suggests that activation of the vagal immune reflex can not only be pursued by stimulation of the vagus nerve or nicotinic receptors but also by inhibition of muscarinic receptors, which results in secondary activation of the pathway. The auCrit Care Med 2009 Vol. 37, No. 2 thors suggest a new approach to activate the vagal anti-inflammatory pathway, which is very interesting and needs further study on the molecular and preclinical level. Their findings are new and expand our possibilities of using the vagal anti-inflammatory pathway for clinical benefit because we now know that stepping on the muscarinic brake results in stepping on the vagal anti-inflammatory gas pedal. David J. van Westerloo, MD, PhD Department of Intensive Care Medicine Academic Medical Centre University of Amsterdam Amsterdam, The Netherlands REFERENCES 1. Tracey KJ: The inflammatory reflex. Nature 2002; 420:853– 859 2. Blalock JE: Harnessing a neural-immune circuit to control inflammation and shock. J Exp Med 2002; 195:F25–F28 3. Borovikova LV, Ivanova S, Zhang M, et al: Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 2000; 405:458 – 462 4. Wang H, Yu M, Ochani M, et al: Nicotinic acetylcholine receptor alpha7 subunit is an essential regulator of inflammation. Nature 2003; 421:384 –388 5. Giebelen IA, van Westerloo DJ, LaRosa GJ, et al: Local stimulation of alpha7 cholinergic receptors inhibits LPS-induced TNF-alpha release in the mouse lung. Shock 2007; 28: 700 –703 6. Giebelen IA, van Westerloo DJ, LaRosa GJ, et al: Stimulation of alpha 7 cholinergic receptors inhibits lipopolysaccharide-induced neutrophil recruitment by a tumor necrosis factor alpha-independent mechanism. Shock 2007; 27:443– 447 7. van Westerloo DJ, Giebelen IA, Florquin S, et al: The vagus nerve and nicotinic receptors modulate experimental pancreatitis severity in mice. Gastroenterology 2006; 130: 1822–1830 8. van Westerloo DJ, Giebelen IA, Florquin S, et al: The cholinergic anti-inflammatory pathway regulates the host response during sep- 9. 10. 11. 12. 13. tic peritonitis. J Infect Dis 2005; 191: 2138 –2148 van Westerloo DJ, Giebelen IA, Meijers JC, et al: Vagus nerve stimulation inhibits activation of coagulation and fibrinolysis during endotoxemia in rats. J Thromb Haemost 2006; 4:1997–2002 de Jonge WJ, van der Zanden EP, The FO, et al: Stimulation of the vagus nerve attenuates macrophage activation by activating the Jak2-STAT3 signaling pathway. Nat Immunol 2005; 6:844 – 851 Liu C, Shen F-M, Le YY, et al: Antishock effect of anisodamine involves a novel pathway for activating ␣7 nicotinic acetylcholine receptor. Crit Care Med 2009; 37:634 – 641 Rosas-Ballina M, Ochani M, Parrish WR, et al: Splenic nerve is required for cholinergic antiinflammatory pathway control of TNF in endotoxemia. Proc Natl Acad Sci USA 2008; 105:11008 –11013 Huston JM, Ochani M, Rosas-Ballina M, et al: Splenectomy inactivates the cholinergic antiinflammatory pathway during lethal endotoxemia and polymicrobial sepsis. J Exp Med 2006; 203:1623–1628 Role for bacteriophages in the preparedness for fighting against pneumonia in the postantibiotic era* T he discovery of penicillin and the rise of antibiotic drugs changed the nature of medicine. Antibiotics, which actually killed bacteria, were hailed as “miracle drugs” (1), able to strike at the “root” cause of diseases like pneumonia, tuberculosis, malaria, syphilis, and gonorrhea. Unfortunately, bacteria soon demonstrated that they were able to become resistant to different antibiotics and, in turn, to pass their resistance genes onto their descendants, eventually producing populations of bacteria could survive even the strongest drugs. As a consequence, pneumonia that was once thought could be eradicated remained a persistent problem. *See also p. 642. Key Words: pneumococcal pneumonia; bacteriophage; endolysin; Cpl-1; mouse; postantibiotic era The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194fe0d Crit Care Med 2009 Vol. 37, No. 2 Pneumonia is a leading cause of death, particularly in children and elderly. The use of antibiotics and advances in health care lessened the risk; however, pneumonia remained the first cause of death among respiratory and infectious diseases globally. Streptococcus pneumoniae is the commonest etiology of mild, moderate, and severe community-acquired pneumonia. There is an increase in antibiotic resistance of pneumococci limiting the number of antimicrobial agents that produce reliable treatment results for these infections. Other microorganisms producing community-acquired or nosocomial pneumonia show the same resistance acquisition problem. Newer antimicrobials and combination therapy by antimicrobials with different mechanisms of action have been used to treat infections for decades with the goal of producing a wider spectrum of action, preventing the emergence of drugresistant subpopulations, reducing the dose of a single agent, or achieving a synergistic effect; however, this use of antibiotics promoted the development of increasing resistance. This phenomenon has been named the antibiotic paradox— miracle drugs are destroying the miracle (2). The ways to face the antimicrobials resistance challenge include the increase of our efforts to preserve the activity of available antibiotics, or at least expand, as much as possible, the period of their use, while intense research efforts should be focused on the development and introduction of new antimicrobial agents into clinical practice. However, the problem generated with the evolving antimicrobial resistance in Pseudomonas aeruginosa, Acinetobacter baumannii, Staphylococcus aureus, and Klebsiella pneumoniae has led to the emergence of clinical isolates susceptible to only one class of antimicrobial agents and eventually to pandrug-resistant (i.e., resistant to all available antibiotics) isolates. Bacteriophages or phages, so called bacteria-killing viruses, described at the beginning of the 20th century and characterized by Delbrück, Luria, and Hershey who were awarded the novel prize in 1969 for their discoveries (3), have been 779 used successfully in treatments against antibiotic-resistant bacteria in various infections, including pneumonia. These data have come from clinical trials in Eastern Europe, mostly uncontrolled. However, recent findings in well-controlled animal models demonstrating that phages can rescue animals from a variety of fatal infections produced by S. pneumoniae, S. aureus, and P. aeruginosa are encouraging (4 – 6). Therapeutic phages appear to kill their target bacteria by replicating inside and lysing the host cell via a lytic cycle; endolysins are lytic enzymes that ensure phages’ progeny survival, by selecting essential cell wall components as target for destruction to avoid phages’ breed from being locked up inside the bacteria (7, 8). However, lysis of host bacteria by a lytic phage is a complex process consisting of a cascade of events involving several structural and regulatory genes, and it is possible that some therapeutic phages have some unique yet unidentified genes or mechanisms responsible for their ability to effectively lyse their target bacteria (9). The therapeutic and prophylactic application of phages is now experiencing a renaissance of interest. In this issue of Critical Care Medicine, Witzenrath et al (10) explored, in an experimental study on a mouse model of severe pneumococcal pneumonia, the therapeutic potential of Cpl-1, a purified bacteriophage endolysin, which specifically kills pneumococci on contact. They observed that when treatment was commenced 24 hours after experimental infection, 100% Cpl-1-treated mice survived the otherwise fatal pneumonia and showed rapid recovery. When treatment was started 48 hours after infection, mice had developed bacteremia, and three of seven Cpl-1-treated animals (42%) survived. Cpl-1 dramatically reduced pulmonary bacterial counts, and prevented bacteremia, systemic hypotension, and lactate increase when treatment commenced at 24 hours. In vivo, treatment 780 with Cpl-1 effectively reduced counts of penicillin-susceptible pneumococci. The inflammatory response in Cpl-1-treated mice, as determined by multiplex cytokine assay of lung and blood samples, was lower than in untreated mice. They concluded that Cpl-1 may provide a new therapeutic option in the treatment of pneumococcal pneumonia. This study adds new information about the effectiveness of purified recombinant bacteriophage lytic enzymes, and specifically on the role of Cpl-1 for the control of diseases caused by S. pneumoniae. These findings warrant research into further therapeutic uses of this novel technology to advance its application in the clinical setting. Unfortunately, from the microbiological standpoint, pneumonia is a complex problem that cannot be immediately derived from an animal model to the patients’ presentation in the real world. Cpl-1, as most of the lytic enzymes, are targeted to specific pathogenic bacteria. It is difficult to ascertain, based on clinical or rapid microbiological techniques, which is the etiology of pneumonia; additionally, if more than one pathogen is involved, as happen, in about 10% to 50% of pneumonias, it is not clear if in these cases a “cocktail” of different lytic enzymes or bacteriophages could be used (11). The postantibiotic era is more than a hypothesis; actually, pandrug-resistant bacteria are among us, producing severe nosocomial pneumonia and other kinds of infections in the healthcare setting. The strategies to face this problem include the use of novel therapies to fight against bacterial infections including phage therapy, immunomodulation, nonbacterial killing effects of antibiotics, and improved preventive measures. There is still a great lack of formal large, scale clinical-studies on bacteriophages safety and effectiveness; the lack of newer antibiotics requires reevaluating and reinvesting in bacteriophages. Preparedness for the eventual lack of miracle drugs is essential; studies exploring newer ways to struggle against bacterial infections, like the original investigation presented by Witzenrath et al, should be encouraged. Carlos M. Luna, MD, PhD Didier Bruno, MD Department of Medicine Pulmonary Medicine Division Hospital de Clinicas Universidad de Buenos Aires Argentina REFERENCES 1. Parascandola J: From germs to genes: Trends in drug therapy 1852–2002. Pharm Hist 2002, 44:3–11 2. Levy SB: The Antibiotic Paradox: How Miracle Drugs Are Destroying the Miracle. Vol. 1. Plenum, New York, 1992 3. Pennazio S: The origin of phage virology. Rivista di Biologia 2006; 99:103–129 4. Loeffler JM, Djurkovic S, Fischetti VA: Phage lytic enzyme Cpl-1 as a novel antimicrobial for pneumococcal bacteremia. Infect Immun 2003; 71:6199 – 6204 5. Watanabe R, Matsumoto T, Sano G, et al: Efficacy of bacteriophage therapy against gut-derived sepsis caused by Pseudomonas aeruginosa in mice. Antimicrob Agents Chemother 2007; 51:446 – 452 6. Wills QF, Kerrigan C, Soothill JS: Experimental bacteriophage protection against Staphylococcus aureus abscesses in a rabbit model. Antimicrob Agents Chemother 2005; 49:1220 –1221 7. Fischetti VA: Bacteriophage lytic enzymes: Novel anti-infectives. Trends Microbiol 2005; 13:491– 496 8. Young R: Bacteriophage lysis: Mechanism and regulation. Microbiol Rev 1992; 56:430 – 481 9. Sulakvelidze A, Alavidze Z, Morris JG Jr: Bacteriophage therapy. Antimicrob Agents Chemother 2001; 45:649 – 659 10. Witzenrath M, Schmeck B, Doehn JM, et al: Systemic use of the endolysin Cpl-1 rescues mice with fatal pneumococcal pneumonia. Crit Care Med 2009; 37:642– 649 11. Mc Vay CS, Velasquez M, Fralick JA: Phage therapy of Pseudomonas aeruginosa infection in a mouse burn wound model. Antimicrob Agents Chemother 2007; 51:1934 –1938 Crit Care Med 2009 Vol. 37, No. 2 Functional hemodynamics and increased intra-abdominal pressure: Same thresholds for different conditions . . .?* W hat this study tells us . . . In this issue of Critical Care Medicine, Renner et al (1) report the results of a study entitled “Influence of increased intra-abdominal pressure on fluid responsiveness predicted by pulse pressure variation and stroke volume variation (SVV) in a porcine model.” On first sight, it looks as if this study just repeats previously performed studies on SVV and pulse pressure variation (PPV) during intra-abdominal hypertension (IAH) (2–5). Is this really the case? The authors collected prospective data on 14 domestic pigs during loading conditions and increased intra-abdominal pressure (IAP) up to 25 mm Hg via CO2 pneumoperitoneum (PP). The application of PP increased the baseline values for SVV (from 9.6% ⫾ 3.2% to 16.3% ⫾ 5.9%) and PPV (from 12.8% ⫾ 3.5% to 22% ⫾ 8.2%), whereas global end-diastolic volume (GEDV) decreased (from 836 ⫾ 210 mL to 788 ⫾ 198 mL). After fluid loading, PP still increased SVV (from 6.4% ⫾ 2.7% to 13.2% ⫾ 4.7%) and PPV (from 7.4% ⫾ 1.6% to 15.7% ⫾ 4.5%). These changes were correlated with changes in esophageal or intrathoracic pressure (ITP). They also found that changes in PPV, SVV, and GEDV showed significant correlations with changes in SV after volume loading, independent of IAP. The ability of PPV and GEDV to predict fluid responsiveness remained unchanged during PP, whereas SVV lost this ability when IAP was increased up to 25 mm Hg. The bottom line is that in conditions of increased ITP, as *See also p. 650. Key Words: abdominal pressure; abdominal compartment; preload; fluid responsiveness; functional haemodynamics; stroke volume variation; pulse pressure variation Dr. Malbrain has consulted for, received honoraria from, and holds stock in Pulsion Medical Systems. Dr. De laet has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c397 Crit Care Med 2009 Vol. 37, No. 2 is the case with IAH, different SVV and PPV thresholds should be used to predict fluid responsiveness. What previous studies have shown . . . Duperret et al (3) found similar results in a pig model of abdominal banding with IAP up to 30 mm Hg. A dose-related IAP effect was suspected. In their model, the systolic pressure variations (SPV), PPV, and inferior vena cava flow fluctuations were dependent on IAP values, which caused changes in pleural pressure swings, and this dependency was more marked during hypovolemia. Although the application of PP increased both PPV and SPV, a statistical significant effect was only observed for SPV. The increase in SPV was mainly related to the ⌬up component. Duperret et al (3) also looked at left ventricular end-diastolic pressure and left ventricular end-diastolic area and found that the increase in the IAP induced a progressive increase in ITP and, therefore, a relative hypovolemia owing to a redistribution of blood volume. Although this factor is likely to play a role at the highest values of IAP, a moderate value of IAP was associated with an increase in thoracic blood volume, left ventricular end-diastolic area, and transmural left ventricular end-diastolic pressure, suggesting an auto-transfusion effect. Tournadre et al (5) found similar results for SPV in seven pigs undergoing CO2 PP, but this time limited at 12 mm Hg. The SPV increased whereas central venous pressure remained unchanged and left ventricular end-diastolic area decreased. The 30% increase in SPV during elevated IAP was mainly because of an increase in ⌬up component from 3.8% ⫾ 5.2% to 7.3% ⫾ 5.5%. Volume loading with hydroxylethyl starch (10 mL/kg) during PP resulted in a decrease in SPV owing to a decrease in ⌬up to 6% ⫾ 3.1%. In a similar study by Bliacheriene et al (2) in 11 rabbits undergoing CO2 PP with IAP increase up to 10 mm Hg, SPV increased more than PPV. The PP in combination with hypovolemia induced by hemorrhage further increased SPV and PPV, confirming the superiority of PPV over SPV to predict “fluid responsiveness.” The increase in SPV by PP was also caused by an increase in ⌬up from 2% ⫾ 1% to 6.7% ⫾ 2%. This study had some limitations as discussed previously (4). In a last study, Valenza et al (6) looked at the effects of Helium-PP up to 25 mm Hg in 15 rats. The PP increased SVV from 8.7% ⫾ 3% to 17.9% ⫾ 8% together with the central venous pressure whereas GEDV decreased. Table 1 gives an overview of the different studies. What this study adds . . . The changes observed in SVV and PPV for a given IAP are the most important findings of this study. Together with the previously documented increases in SPV that were mainly related to the ⌬up component, possible mechanisms are suggested. First, a change in aortic compliance and an increase in aortic transmural pressure induced by increased IAP (either via direct compression or increased vasomotor tone). Second, errors in the measurement of dynamic indices in conditions of increased IAP, or, if we assume that no measurement errors are induced by IAP then this implies that these indices do not perform well during IAH (since SVV did no longer predict fluid responsiveness). Third, changes in extramural pressure, ITP, or chest wall compliance. This study was one of the first looking at the abdominothoracic index of transmission (7). The authors found a 47% transmission of the ⌬IAP to the lung and a 45% transmission to the thoracic compartment (Table 1). This is in accordance with previous studies showing on average an abdominothoracic index of transmission of 50% and confirms that traditional filling pressures are erroneously increased during increased ITP/IAP (7, 8). Interestingly, the abdominothoracic index of transmission was higher in nonresponders. The authors also looked at receiver operating characteristic curves to identify the best thresholds for predicting fluid responsiveness at baseline and during PP. The results confirmed that GEDV is not 781 Table 1. Overview of studies on functional hemodynamics during intra-abdominal hypertension Stage Animal N Intra-abdominal hypertension IAP-BL (mmHg) IAP-PP ⌬IAP CVP-BL (mmHg) CVP-PP ⌬CVP Preload-BL Preload-PP ⌬preload Paw-BL (cmH2O) Paw-PP ⌬Paw ⌬ITP ATI-thor ATI-car ATI-lung SPV-BL (%) SPV-PP ⌬SPV SVV-BL (%) SVV-PP ⌬SVV PPV-BL (%) PPV-PP ⌬PPV Tournadre et al (5) Bliacherine et al (2) Bliacherine et al (2) Duperret et al (3) Duperret et al (3) Valenza et al (6) Renner et al (1) Renner et al (1) NormoV Pig 7 CO2 NormoV Rabbit 11 CO2 HypoV NormoV Pig 7 Banding HypoV NormoV Rat 15 Helium NormoV Pig 14 CO2 Fluid 0 25 25 0 25 25 2 25 23 7.7 12.7 5 290.4 267.2 –23.2 16.9 28.1 11.2 7 30.4% 21.7% 35.8% 6.5 25.5 19 6.2 11.1 4.9 836 788 –48 21.4 33.6 12.2 8.6 45.3% 25.8% 47.2% 7.2 26 18.8 10.2 14.4 4.2 1069 1034 –35 21.1 34.6 13.5 8.1 43.1% 22.3% 52.8% 8.7 17.9 9.2 (106%) 9.6 16.3 6.7 (70%) 12.8 22 9.2 (72%) 6.4 13.2 6.8 (106%) 7.4 15.7 8.3 (112%) 0 12 12 6b 7 1 9.9 9.5 –0.4 26 37 11 8.3% 67.4% 12.9 16.9 4 (31%) 0 10 10 0 10 10 15 12.5 –2.5 13.4 23.5 10.1 74.3% 8.5 13.3 4.8 (56%) 9.5 11.1 1.6 (17%) 13.4 23.4 10 73.5% 13.3 19.9 6.6 (50%) 11.1 24.9 13.8 (124%) 15 60.0% 14 56.0% 6 17.1 11.1 (185%) 4.1 12.2 8.1 (198%) 3.1 8.1 5 (161%) 12.7 19.7 7 (55%) 7.7 11.8 4.1 (53%) 6.6 9.8 3.2 (48%) Meana 10.8 2.0 19.8 17.9 7.5 11.3 3.8 731.8 696.4 –35.4 18.7 30.0 11.3 10.5 40.5% 19.6% 58.5% 10.7 17.4 6.7 (75.4%) 7.3 14.3 7.0 (106.5%) 8.4 15.3 6.9 (89.2%) ATI-car, abdomino-thoracic index of transmission to the cardiovascular compartment (calculated as ⌬CVP divided by ⌬IAP); ATI-res, abdomino-thoracic index of transmission to the lungs (calculated as ⌬Paw divided by ⌬IAP); ATI-thor, abdomino-thoracic index of transmission to the thoracic compartment (calculated as ⌬ITP divided by ⌬IAP); BL, baseline values; CVP, central venous pressure; IAP, intra-abdominal pressure; ITP, intra thoracic pressure; Paw: peak airway pressure; PP, pneumoperitoneum; PPV, pulse pressure variation; SPV, systolic pressure variation; SVV, stroke volume variation. Preload, left ventricular end diastolic area (in cm2) was used for the studies by Tournadre et al and Duperret et al, while global end diastolic volume (in mL) was in the other studies. a Looking at the pooled data one can see that an increase in IAP (⌬IAP) of 17.9 mm Hg results in a ⌬CVP of 3.8 mm Hg, a ⌬ITP of 10.5 mm Hg, and a ⌬Paw of 11.3 cm H2O. The average abdomino-thoracic index of transmission was 47% to the thorax, 20% to the cardiovascular space, and 58% to the lungs (airway pressure). The bottom line is that dynamic indices like SPV, SVV, or PPV are not exclusively related to volaemia in the presence of increased ITP or IAP; bin the study by Tournadre data was given for pulmonary artery occlusion pressure instead of CVP. only a good static “volumetric” preload parameter (far more superior than the “barometric” indices of preload like central venous pressure or pulmonary artery occlusion pressure) but also a good indicator of fluid responsiveness (equal to PPV), regardless of IAP. In contrast to other studies, transmural filling pressures did not perform better than the central venous pressure and pulmonary artery occlusion pressure values taken at end-expiration (central venous pressureee) (9). Interesting to see was that PPV (but not SVV) kept its ability to predict fluid responsiveness even at IAP levels of 25 mm Hg; however, receiver operating characteristic curve analysis identified 20.5% as the best threshold for fluid responsiveness (instead of the classic 12% and the 9.5% identified in this study at baseline). This is an important 782 clinical message, meaning that we probably cannot use the same thresholds for different conditions. The threshold value will depend on the amount of tidal volume, positive endexpiratory pressure application, or increased ITP and consequent changes in pleural pressure and chest wall elastance, the presence of obesity, heart failure with changes in right and left ventricular preload and afterload, pulmonary hypertension, the use of a PP or increased IAP . . . and may also differ in children or neonates. However, it must be said that because of the limited sample size the cut-off values only give a rough estimation and need to be validated in a larger patient population. This study showed that PPV performed better than SVV derived from pulse contour analysis based on a complex algo- rithm. This is a bit surprising because PPV is a surrogate of SVV and the latter should be less influenced by changes in vasomotor tone. The present study suggests that changes in pulse contour due to increased ITP may be more complex than previously thought. The study also showed that ⌬IAP correlates well with ⌬ITP and thus confirms the rule of thumb to calculate the transmural filling pressure: central venous pressuretm ⫽ central venous pressureee ⫺ IAP/2 (10). What this study does not tell us . . . To play the devil’s advocate one could argue that the questions that the authors tried to answer have been addressed previously. While the study was simple in its concept, it turned out to be complex in its execution and it may leave the average Crit Care Med 2009 Vol. 37, No. 2 reader with more questions than it provides answers . . . First, one could argue that the small number of animals does not allow proper receiver operating characteristic curve analysis to define thresholds. The prediction of fluid responsiveness implies even smaller subgroups of responders vs. nonresponders, proving that dynamic indices work in animals with normal IAP is nothing new. Second, the model for creating IAH, namely a CO2 PP may have affected the results. Gases are likely to have systemic effects. Although helium seems quite inert, argon has been observed to influence hemodynamic parameters, and CO2 can cause a change in vasomotor tone (aortic compliance), partly explaining the observations (11). Third, this study was performed on healthy pigs with normal cardiovascular and respiratory function. Therefore, it remains questionable whether these results can be extrapolated to a pathologic condition with a primary insult, capillary leak, resuscitation, and tissue edema. Finally, the study period and the timecourse between the different stages was quite short (only 15 minutes). The current model therefore is one that mimics “hyperacute” IAH rather than the clinically encountered “acute” or “subacte” IAH (12). What future animal studies should look at . . . Despite the fact that the data raises a lot of questions, this study is important and the authors have to be congratulated. Future studies should look at the long-term effects (24 – 48 hours) of IAH at clinically relevant IAP levels (15–20 mm Hg) on dynamic indices of fluid responsiveness during PP, with and without hemorrhage and before and after fluid loading. Future studies should try to integrate these results with Crit Care Med 2009 Vol. 37, No. 2 global indices of perfusion (lactate, base deficit) and the presence of clinical overt shock in animals and even better in patients who are critically ill. Furthermore, they should also look at a good gold standard for SV measurement by means of an ultrasonic flow probe in the aorta to exclude the possibility for possible mathematical coupling of data. The results of this study confirm the importance of IAH/abdominal compartment syndrome (13). The world society of the abdominal compartment syndrome (www.wsacs.org) invites interested researchers to join the society, to adhere to the consensus definitions posted at the Web site and to submit some prospective data for the next world congress (www. wcacs.org), to be held in Dublin, Ireland, June 24 –27, 2009. Manu L. N. G. Malbrain, MD, PhD Inneke de laet, MD Department of Intensive Care ZiekenhuisNetwerk Antwerpen ZNA Stuivenberg, Belgium REFERENCES 1. Renner J, Gruenewald M, Quaden R, et al: Influence of increased intra-abdominal pressure on fluid responsiveness predicted by pulse pressure variation and stroke volume variation in a porcine model. Crit Care Med 2008; 37:650 – 658 2. Bliacheriene F, Machado SB, Fonseca EB, et al: Pulse pressure variation as a tool to detect hypovolaemia during pneumoperitoneum. Acta Anaesthesiol Scand 2007; 51: 1268 –1272 3. Duperret S, Lhuillier F, Piriou V, et al: Increased intra-abdominal pressure affects respiratory variations in arterial pressure in normovolaemic and hypovolaemic mechanically ventilated healthy pigs. Intensive Care Med 2007; 33:163–171 4. Malbrain ML, De Laet I: Functional haemodynamics during intra-abdominal hypertension: What to use and what not. Acta Anaesthesiol Scand 2008; 52:576 –577 5. Tournadre JP, Allaouchiche B, Cayrel V, et al: Estimation of cardiac preload changes by systolic pressure variation in pigs undergoing pneumoperitoneum. Acta Anaesthesiol Scand 2000; 44:231–235 6. Valenza F, Chevallard G, Porro GA, et al: Static and dynamic components of esophageal and central venous pressure during intra-abdominal hypertension. Crit Care Med 2007; 35:1575–1581 7. Wauters J, Wilmer A, Valenza F: Abdominothoracic transmission during ACS: Facts and figures. Acta Clin Belg Suppl 2007; 62: 200 –205 8. Malbrain ML, Wilmer A: The polycompartment syndrome: Towards an understanding of the interactions between different compartments! Intensive Care Med 2007; 33: 1869 –1872 9. Ridings PC, Bloomfield GL, Blocher CR, et al: Cardiopulmonary effects of raised intraabdominal pressure before and after intravascular volume expansion. J Trauma 1995; 39: 1071–1075 10. Cheatham ML, Malbrain ML: Cardiovascular implications of abdominal compartment syndrome. Acta Clin Belg Suppl 2007; 62: 98 –112 11. Hazebroek EJ, Haitsma JJ, Lachmann B, et al: Impact of carbon dioxide and helium insufflation on cardiorespiratory function during prolonged pneumoperitoneum in an experimental rat model. Surg Endosc 2002; 16:1073–1078 12. Malbrain ML, Cheatham ML, Kirkpatrick A, et al: Results from the International Conference of Experts on intra-abdominal hypertension and abdominal compartment syndrome. I. Definitions. Intensive Care Med 2006; 32:1722–1732 13. Malbrain ML, De Laet I: AIDS is coming to your ICU: Be prepared for acute bowel injury and acute intestinal distress syndrome. Intensive Care Med 2008; 34: 1565–1569 783 Designing clinical trials to improve neurobehavioral outcome after traumatic brain injury: From bench to bedside* T raumatic brain injury (TBI), a physiologic disruption of brain structure and function resulting from the application of external physical force, is a problem of increasing concern within the medical community, among healthcare policymakers, and to the general public. Between one and two million persons in the United States suffer from TBI each year, of which ⬃230,000 are hospitalized. Improvements in prehospital care (1) and refinements in acute care management strategies (2) make survival after TBI more likely today than ever before. Despite improvements in survival after TBI, however, the Centers for Disease Control and Prevention estimate that ⬎80,000 Americans develop long-term TBI-related disability each year and that about 3.2 million persons in the United States are living with such disabilities (3, 4). Mitigating the neurologic, neurobehavioral, and functional sequelae of TBI, therefore, is an essential treatment goal at all stages of postinjury care. Unfortunately, the development of therapies that accomplish this goal has not kept pace with those that improve TBI survival. At least 21 multicentered randomized controlled trials of agents with potential neuroprotective properties, including N-methyl-D aspartate receptor antagonists, calcium channel blockers, magnesium sulfate, cannabinoids, corticosteroids, aminosteroids, progesterone, cyclosporine, and inflammatory modulators, among others, have been conducted in the TBI population (5, 6). None of these studies convincingly demonstrated benefit. *See also p. 659. Key Words: traumatic brain injury; neurobehavioral function; memory; outcome; clinical trials Supported, in part, by HealthONE Spalding Rehabilitation Hospital. The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e3181959e46 784 Interpreting the failures of these randomized controlled trials is a complicated endeavor (6). The proportion of the variance in clinical outcome accounted for by any general neuroprotective intervention is likely to be relatively modest. When investigated in groups that are heterogeneous with respect to TBI characteristics—particularly the presence, type, severity of intracranial abnormalities, clinical severity, and baseline prognostic risk—the likelihood of effecting change through a single intervention is made more modest still. Additionally, when the outcome by which that intervention is assessed is a global measure (such as the Glasgow Outcome Scale [7]) dichotomized into “favorable” or “unfavorable” categories, statistical power to detect clinical neuroprotective effects is attenuated further. The design of such randomized controlled trials and the selection of an outcome measure for use therein is more than a matter of statistics. Several decades of improvements in survival after TBI necessitate a shift, or at least an expansion, of focus from general measures of TBI outcome to the most enduring and difficult problem experienced by the majority of TBI survivors and their families: neurobehavioral disturbances (8, 9). These problems contribute substantially to postinjury disability (10, 11) and present major challenges for TBI survivors and their families throughout the postinjury period (12, 13). Despite the prevalence and functional relevance of posttraumatic neurobehavioral disturbances, their incorporation into the design of randomized controlled trials in TBI and stroke (14) is limited, if and when such problems are considered at all. In this issue of Critical Care Medicine, Longhi et al (15) present findings from an experimental injury study in which neurobehavioral and neuropathological outcomes are afforded equal consideration. Recognizing the complement system as a contributor to the pathobiology of TBI, they investigated the effects of C1inhibitor (C1-INH)—an endogenous serine-protease inhibitor of complement activation within the classic, alternative, and lectin pathways— given 10 minutes or 1 hour postinjury on neurobehavioral sequelae and histologic damage in a controlled cortical impact mouse model of TBI. The selection of both this agent and this particular model of TBI are important initial considerations: the activation of complement and associated inflammatory responses are substantial contributors to brain injury in the setting of cerebral contusion, but are less relevant in the absence of such (i.e., in the setting of diffuse axonal injury alone). In addition to defining a subcategory of TBI in which the agent selected might be useful, the agent itself affects several elements of the complement cascade. Its multiplicity of effects increases the likelihood of attenuating injury with this single agent. Additionally, the authors anchor their investigation to outcomes that more readily model important clinical parameters than do histologic measures alone: they evaluate the effect of C1-INH on neurobehavioral function, including motor performance (as assessed by the composite neuroscore) and cognitive ability (as assessed by the Morris Water Maze). In this study, a single administration of C1-INH at 10 minutes following controlled cortical impact attenuated posttraumatic motor and cognitive impairments, and reduced histologic damage when assessed 4 weeks after injury. When this agent was administered 1 hour postinjury, motor deficits, but not cognitive function or histologic injury, were attenuated significantly. The efficacy of this agent and the timing of administration required to optimize its effects, if such are replicated in additional studies, remain uncertain. Nonetheless, the ability of a single administration of C1INH—when given at any point postinjury—to attenuate motor, cognitive, and histologic abnormalities is noteworthy. Crit Care Med 2009 Vol. 37, No. 2 Also intriguing is the dissociation between motor, cognitive, and histologic outcomes following delayed C1-INH administration. The mechanism by which this dissociation developed is not clear. Nonetheless, its occurrence echoes the importance of expanding the concept of outcome beyond those used conventionally in experimental injury research: if only one of these outcomes was the sole subject of this investigation, then a very different set of the conclusions regarding the possible promise of this agent would be drawn. Given the history of failed clinical investigations of neuroprotection in TBI, skepticism regarding the future of the line of investigation described by Longhi et al would not be unexpected. The most of which we can be certain is that replicating at the bedside the same types and magnitudes of neuroprotective effects observed in the laboratory is an uncertain endeavor, at best. Independent of the future of the neuroprotective strategy described in this report, however, these investigators have provided the field with a model for future translational research in TBI. At the earliest stage of research, a subtype of injury is identified in the service of reducing heterogeneity and of developing a population-targeted therapy. An attempt is made to identify an intervention that might reasonably be expected to intervene on one or more aspects of the pathophysiology of that TBI subtype. If a single agent is used, then one with multiple mechanisms of action is selected to maximize the proportion of variance in out- come for which it might be expected account. Finally, neurobehavioral function, including both motor and cognitive performance, is incorporated into the research program at its earliest stages and is afforded the same level of attention and scrutiny as histopathology—an essential first step in the improvement of neurobehavioral and functional outcomes of persons with TBI and their families. David B. Arciniegas, MD Brain Injury Rehabilitation Unit HealthONE Spalding Rehabilitation Hospital Aurora, Colorado Department of Psychiatry Neurobehavioral Disorders Program University of Colorado School of Medicine Aurora, Colorado 6. 7. 8. 9. 10. 11. REFERENCES 1. Gabriel EJ, Ghajar J, Jagoda A, et al: Brain Trauma Foundation: Guidelines for prehospital managment of traumatic brain injury. J Neurotrauma 2002; 19:111–174 2. Marion DW: Evidenced-based guidelines for traumatic brain injuries. Prog Neurol Surg 2006; 19:171–196 3. Thurman D, Guerrero J: Trends in hospitalization associated with traumatic brain injury. JAMA 1999; 282:954 –957 4. Zaloshnja E, Miller E, Langlois JA, Selassie AW: Prevalence of long-term disability from traumatic brain injury in the civilian population of the United States, 2005. J Head Trauma Rehabil 2008; 23:394 – 400 5. Beauchamp K, Mutlak H, Smith WR, et al: Pharmacology of traumatic brain injury: 12. 13. 14. 15. Where is the “golden bullet”? Mol Med 2008; 14:731–740 Maas AI, Marmarou A, Murray GD, et al: Prognosis and clinical trial design in traumatic brain injury: The IMPACT study. J Neurotrauma 2007; 24:232–238 Jennett B, Bond M: Assessment of outcome after severe brain damage. Lancet 1975; 1:480 – 484 Levin HS: Neurobehavioral outcome of closed head injury: Implications for clinical trials. J Neurotrauma 1995; 12:601– 610 Levin HS, Gary HE Jr, Eisenberg HM, et al: Neurobehavioral outcome 1 year after severe head injury. Experience of the Traumatic Coma Data Bank. J Neurosurg 1990; 73: 699 –709 Dikmen S, Machamer J, Miller B, et al: Functional status examination: A new instrument for assessing outcome in traumatic brain injury. J Neurotrauma 2001; 18:127–140 Hall KM, Bushnik T, Lakisic-Kazazic B, et al: Assessing traumatic brain injury outcome measures for long-term follow-up of community-based individuals. Arch Phys Med Rehabil 2001; 82:367–374 Dikmen SS, Machamer JE, Powell JM, et al: Outcome 3 to 5 years after moderate to severe traumatic brain injury. Arch Phys Med Rehabil 2003; 84:1449 –1457 Groom KN, Shaw TG, O’Connor ME, et al: Neurobehavioral symptoms and family functioning in traumatically brain-injured adults. Arch Clin Neuropsychol 1998; 13:695–711 Anderson CA, Arciniegas DB, Filley CM: Treatment of acute ischemic stroke: Does it impact neuropsychiatric outcome? J Neuropsychiatry Clin Neurosci 2005; 17:486 – 488 Longhi L, Perego C, Ortolano F, et al: C1inhibitor attenuates neurobehavioral deficits and reduces contusion volume after controlled cortical impact brain injury in mice. Crit Care Med 2009; 37:659 – 665 Pediatric sepsis: Time is of the essence* C linical practice guidelines have been widely accepted as a tool to standardize care and improve outcomes in critically ill patients. Periodic revision of *See also p. 666. Key Words: sepsis; pediatrics; guidelines; septic shock; hemodynamic support The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e3181931210 Crit Care Med 2009 Vol. 37, No. 2 guidelines, especially those addressing clinical conditions for which a large number of new studies are published each year, is critical to keeping such guidelines up to date and relevant. Septic shock is one of the most common disorders managed in both adult and pediatric intensive care units, and one in which there is a rapidly growing body of literature. In 2002, the American College of Critical Care Medicine developed clinical practice parameters for hemodynamic support of children and neonates with septic shock (1). In this issue of Critical Care Medicine, the American College of Critical Care Medicine presents the 2007 revision of these clinical practice parameters, chaired by Dr. Joseph Carcillo (2). This revision of the 2002 guidelines incorporates data from new publications over the 5-yr interval. Thirty interested parties, many but not all of whom are recognized experts in the field, participated in the revision process. Recommendations included in the guidelines required that 90% of the committee members agreed with the recommendation. As with the 2002 guidelines, many of the recommendations have limited support in the literature, but the 785 90% agreement requirement demonstrates that there is broad clinical support for these recommendations. A growing body of literature has reported improved outcomes with the implementation of the 2002 guidelines (3, 4). This finding is consistent with the current emphasis on standardization of care to improve outcomes in the critically ill. The 2007 revision includes new trials relevant to both the initial resuscitation and the ongoing hemodynamic management of children with septic shock. It also includes information on new and developing technologies for hemodynamic monitoring of these children. The rapid development of new therapeutics and technologies makes periodic revision of these guidelines essential. The Task Force revising the 2002 guidelines completed a thorough literature review. Overall, there are relatively minor changes in the recommendations, but there is increasing emphasis on the importance of early rapid management. There is continued emphasis on the first hour fluid resuscitation and treatment with vasopressors/inotropes. Antibiotic administration within the first hour has been added to the revised guidelines. It is increasingly recognized that rapid early fluid resuscitation and antibiotic administration are critical elements in the management of the patient with septic shock. The Surviving Sepsis Campaign guidelines emphasize these points as well (5). Initial resuscitation goals for children and infants are clinical goals, aiming for normalization of the heart rate, blood pressure, capillary refill, and mental status. The age-old question of colloids vs. crystalloids for initial resuscitation remains unresolved; the clinician is free to select the fluid of his/her preference. The 2007 revision of the pediatric sepsis guidelines does depart from the conventional wisdom that catecholamines 786 should only be infused through central venous catheters. As a practical matter, it can sometimes take a prolonged period of time to establish central venous access in a small child with shock. The 2007 guidelines include the recommendation from the American Heart Association Pediatric Advanced Life Support guidelines to administer vasoactive drugs through a peripheral intravenous catheter until central access is attained, monitoring the peripheral access site closely for infiltration (6). The revised guidelines continue to recommend monitoring of cardiac index in children with catecholamine-resistant shock, and titrating therapy to a goal of 3.3– 6 L/min/m2. A new addition is that there are several new techniques that can be used instead of the pulmonary artery catheter, including Doppler echocardiography, the PICCO catheter, or femoral artery thermodilution catheter. Several new techniques are under investigation. None of these techniques, however, is possible in neonates and small infants, so the managing physician must continue to rely on clinical goals. The role of corticosteroid therapy remains uncertain. Hydrocortisone is recommended for children with absolute adrenal insufficiency and catecholamine-resistant shock, but the doses used in the literature have varied enormously from 2 to 50 mg/ kg/day; no specific dose recommendations can be made at this time. There continues to be clinical equipoise regarding the use of adjunctive steroid therapy for sepsis, outside the use in children with classic adrenal or known hypothalamic-pituitary-adrenal axis insufficiency. The 2007 guidelines include stepwise algorithms for children and for neonates that will help the clinician at the bedside. Like many guidelines, the guidelines may not be as specific as some clinicians might like, but that is a reflection of the state of our knowledge, and not the quality of the guidelines. As we strive to decrease the mortality in children and infants with septic shock, we need to ensure that they receive early aggressive management with fluids, vasoactive agents, and antibiotics. This study provides a framework that each institution can use to develop its own specific guidelines to ensure this level of care. Margaret M. Parker, MD, FCCM Department of Pediatrics Stony Brook University Stony Brook, NY REFERENCES 1. Carcillo JA, Fields AI, American College of Critical Care Medicine Task Force Committee members: Clinical practice parameters for hemodynamic support of pediatric and neonatal patients in septic shock. Crit Care Med 2002; 30:1365–1378 2. Brierley J, Carcillo JA, Choong K, et al: Clinical practice parameters for hemodynamic support of pediatric and neonatal septic shock: 2007 update from the American College of Critical Care Medicine. Crit Care Med 2009; 37:666 – 688 3. Han YY, Carcillo JA, Dragotta MA, et al: Early reversal of pediatric-neonatal septic shock by community physicians is associated with improved outcome. Pediatrics 2003; 112:793–799 4. de Oliveira CF, de Oliveira DS, Dottschald AF, et al: ACCM/PALS haemodynamic guidelines for paediatric septic shock: An outcomes comparison with and without monitoring central venous oxygen saturation. Intensive Care Med 2008; 34:1065–1075 5. Dellinger RP, Levy MM, Carlet JM, et al: Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med 2008; 36: 296 –327 6. International Liaison Committee on Resuscitation: The International Liaison Committee on Resuscitation (ILCOR) consensus on science with treatment recommendations for pediatric and neonatal patients: Pediatric basic and advanced life support. Pediatrics 2006; 117:e955– e977 Crit Care Med 2009 Vol. 37, No. 2 Brain oxidative stress after traumatic brain injury . . . cool it?* H ypothermia has been demonstrated to reduce intracranial pressure, decrease excitotoxicity, attenuate oxidative stress, and limit the consumption of protective antioxidant agents. Moderate hypothermia has demonstrated benefit and has been specifically recommended in adults after cardiac arrest and in neonates who have sustained hypoxicischemic encephalopathy (1, 2). In children, moderate hypothermia has demonstrated efficacy in the control of intracranial pressure after traumatic brain injury (TBI), yet evidence of a survival or morbidity benefit has remained elusive (3, 4). The failure of a recent, large, multinational, randomized controlled trial to demonstrate the clinical benefit of hypothermia after TBI in children (5) has further diluted confidence that hypothermia may improve survival and neurologic outcome. The brain is particularly sensitive to oxidative injury due to metabolic demands and high rate of oxygen consumption. Evidence of oxidative stress after TBI has been demonstrated via cerebrospinal fluid (CSF) analysis in animals (6), children (7, 8), and adults (9). Evidence of increased oxidative stress and disordered energy metabolism after TBI has preceded refractory increases in intracranial pressure (10). This latter finding suggests a role for the analysis of CSF biomarkers as potential predictors of clinical evolution and outcome and, in particular, elevated CSF markers of oxidative damage (11). The article by Bayir et al (12) in this issue of Critical Care Medicine explored the effect of moderate hypothermia on markers of oxidative stress after TBI. A cohort of pediatric patients were re- *See also p. 689. Key Words: traumatic brain injury; hypothermia; therapy; oxidative stress; children The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194be10 Crit Care Med 2009 Vol. 37, No. 2 cruited in a single center for inclusion in two larger multicenter studies investigating the role of therapeutic hypothermia after TBI. CSF was obtained via external ventricular drains and assayed for oxidative stress with measurements of antioxidant status, protein oxidation, and lipid peroxidation. Children who had been randomized to moderate hypothermia were shown to significantly preserve CSF antioxidant reserve, compared with children in the normothermic group. Analysis of the markers of free radical attack and lipid peroxidation did not demonstrate a similar beneficial profile in the hypothermic patients, although the authors do suggest a trend to decreased protein oxidation in the hypothermia group. The authors conclude that hypothermia attenuates brain oxidative stress after severe TBI in children. This reported investigation is intriguing because it demonstrates a potential mechanism for neuroprotection by therapeutic hypothermia. This study also demonstrates the utility of CSF analysis for TBI prognosis and provides measurable factors that could aid in identifying beneficial therapeutic interventions. Importantly, their work indicates that moderate hypothermia in the context of pediatric TBI is not yet a therapeutic dead end. The study by Bayir et al (12) is, however, limited by a small sample size, a heterogeneous patient population, and limited study protocol information. The data presented also lack fidelity in a time course critical for understanding the pathophysiologic changes in oxidative state after TBI and for determining a potential optimal therapeutic window. It is also unclear if the preservation in oxidative state by therapeutic hypothermia reflects changes in compromised tissue, compared with areas of relatively uninjured brain. Finally, although antioxidant reserve is preserved by hypothermia, the significance of the improved oxidant profile is unclear. A causal relationship between oxidative stress and outcome has not been established. Despite these caveats, Bayir et al (12) should be commended for their reported investigation. Enrolling children with TBI in therapeutic trials is challenging, particularly when combined with external ventricular drain placement and regular CSF sampling for laboratory analysis (13). This report indicates that regular laboratory analysis of CSF obtained via external ventricular drains may offer valuable prognostic information and potentially guide future therapeutic interventions. Furthermore, this study is the first to demonstrate preserved antioxidant reserve in children with therapeutic hypothermia. Hypothermia-induced preservation of the antioxidant profile may be beneficial in the prevention of delayed brain injury after TBI. Excessive oxidative stress damages lipids, proteins, and nucleic acids. Cytoskeleton and mitochondria disruption are consequences of oxidative stress, and, eventually, synaptic plasticity and cognitive function are impaired. Similar to the action of therapeutic hypothermia reported by Bayir et al (12), anesthetic agents commonly administered after TBI may also provide protection against oxidative stress. Barbiturates and propofol limit oxidative stress by reducing the cerebral metabolic rate and decreasing glucose and oxygen consumption (14). Anesthetics also induce direct scavenging of reactive oxygen species and inhibit lipid peroxidation and excitotoxicity. In conclusion, Bayir et al (12) who have previously identified the increased oxidative stress in the context of TBI now report a beneficial effect of therapeutic hypothermia on brain oxidative stress. Their work indicates a likely mode of secondary brain injury, and suggests that hypothermia may offer therapeutic benefit. Further, they offer CSF biomarker analyses as worthy of further evaluation not only as an aid in prognostication, but as an indicator of therapeutic efficacy. This work by Bayir et al (12) and others will fuel the emerging interest in drugs that have potent antioxidative effects, such as agonists of the nuclear peroxisome proliferator-activated receptors (15), and suggest that if therapeutic hypothermia has a role to play in the treatment of TBI, it is not as the magic bullet, but as one of a combina787 tion of therapies that act beneficially on brain oxidation status. Douglas D. Fraser, MD, PhD, FRCPC Critical Care Medicine and Paediatrics, University of Western Ontario London, Ontario, Canada Physiology and Pharmacology, University of Western Ontario London, Ontario, Canada Children’s Health Research Institute London, Ontario, Canada Gavin Morrison, MRCP, DCH Critical Care Medicine and Paediatrics, University of Western Ontario London, Ontario, Canada 4. 5. 6. 7. REFERENCES 1. Abate MG, Cadore B, Citerio G: Hypothermia in adult neurocritical patients: A very “hot” strategy not to be hibernated yet! Minerva Anestesiol 2008; 74:425– 430 2. Polderman KH: Induced hypothermia and fever control for prevention and treatment of neurological injuries. Lancet 2008; 371: 1955–1969 3. Adelson PD, Ragheb J, Kanev P, et al: Phase 8. 9. II clinical trial of moderate hypothermia after severe traumatic brain injury in children. Neurosurgery 2005; 56:740 –754; discussion 740 –754 Biswas AK, Bruce DA, Sklar FH, et al: Treatment of acute traumatic brain injury in children with moderate hypothermia improves intracranial hypertension. Crit Care Med 2002; 30:2742–2751 Hutchison JS, Ward RE, Lacroix J, et al: Hypothermia therapy after traumatic brain injury in children. N Engl J Med 2008; 358: 2447–2456 Ansari MA, Roberts KN, Scheff SW: Oxidative stress and modification of synaptic proteins in hippocampus after traumatic brain injury. Free Radic Biol Med 2008; 45:443– 452 Bayir H, Kagan VE, Tyurina YY, et al: Assessment of antioxidant reserves and oxidative stress in cerebrospinal fluid after severe traumatic brain injury in infants and children. Pediatr Res 2002; 51:571–578 Wagner AK, Bayir H, Ren D, et al: Relationships between cerebrospinal fluid markers of excitotoxicity, ischemia, and oxidative damage after severe TBI: The impact of gender, age, and hypothermia. J Neurotrauma 2004; 21:125–136 Bayir H, Marion DW, Puccio AM, et al: Marked gender effect on lipid peroxidation 10. 11. 12. 13. 14. 15. after severe traumatic brain injury in adult patients. J Neurotrauma 2004; 21:1– 8 Cristofori L, Tavazzi B, Gambin R, et al: Biochemical analysis of the cerebrospinal fluid: Evidence for catastrophic energy failure and oxidative damage preceding brain death in severe head injury: A case report. Clin Biochem 2005; 38:97–100 Darwish RS, Amiridze N, Aarabi B: Nitrotyrosine as an oxidative stress marker: Evidence for involvement in neurologic outcome in human traumatic brain injury. J Trauma 2007; 63:439 – 442 Bayir H, Adelson PD, Wisniewski SR, et al: Therapeutic hypothermia preserves antioxidant defenses after severe traumatic brain injury in infants and children. Crit Care Med 2009; 37: 689–695 Kochanek PM, Berger RP, Bayir H, et al: Biomarkers of primary and evolving damage in traumatic and ischemic brain injury: Diagnosis, prognosis, probing mechanisms, and therapeutic decision making. Curr Opin Crit Care 2008; 14:135–141 Wilson JX, Gelb AW: Free radicals, antioxidants, and neurologic injury: Possible relationship to cerebral protection by anesthetics. J Neurosurg Anesthesiol 2002; 14:66 –79 Jennings JS, Gerber AM, Vallano ML: Pharmacological strategies for neuroprotection in traumatic brain injury. Mini Rev Med Chem 2008; 8:689 –701 How possibly are we to choose albumin or hydroxyethyl starch?* A lbumin treatment of hypovolemia has been shown to be superior to crystalloid solutions in virtually every pathologic condition, including its use as part of the pump prime for pediatric cardiopulmonary support. No known cases of infectious transmission have been reported and the incidence of either urticaria or anaphylactic shock is estimated at ⬍1 per10,000 (1). Unit cost is high and shortages could occur. It is with these facts in mind that we evaluate the prospective, randomized, and quasi-blinded trial of albumin vs. hydroxyethyl starch (HES) 130/0.4 in the perioperative volume ex- *See also p. 696. Key Words: albumin; hydroxyethyl starch; pediatric; volume expansion; cardiac surgery; editorial The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194de57 788 pansion during pediatric cardiac surgery requiring cardiopulmonary bypass (2). Hanart et al from Brussels, Belgium painstakingly evaluated perioperative fluid balance and blood loss in children who received 4% albumin (n ⫽ 59) or 6% HES 130/0.4 (n ⫽ 60). The volume of colloid used and recorded blood loss was similar between groups. Intraoperative fluid balance was more positive in the albumin group and the difference reached significance. In addition, more children in the albumin group received blood transfusions but less fresh frozen plasma. Taking the results at face value, equivalence of albumin and HES 130/0.4 was the conclusion. Factoring in the differential treatment cost, a recommendation to replace albumin with HES 130/0.4 for pediatric cardiac surgery seemed warranted. However, I believe this was premature without further study. The authors are commended for a well-controlled study. There are minor flaws and bias that deserve discussion. In the randomization scheme, the population was blocked for presence or absence of cyanotic cardiac lesions and the type of bypass circuit as determined by subject weight. This led to a reasonable representation of age, diagnose, and similar intraoperative times in the groups. However, the blinding did not include the intraoperative anesthesia team. Since the intraoperative team directed fluid treatment, an element of unintentional bias was possible. The authors were precise in the blinded quantification of fluid balance and the types of blood products used. However, there was less precision in the postoperative care protocols, in example, determining the need for ventilatory and inotropic support. I do not think this affects evaluation of the primary end point, but evaluation of safety is difficult. Hanart et al, do discuss the possible complications of starch-based colloid solutions and discuss an evidence that the newer formulations, of which HES Crit Care Med 2009 Vol. 37, No. 2 130/0.4 is thought to be the safest, with less severe clinical bleeding and renal dysfunction. However, in their study, the range in subject ages and diagnoses coupled with only minimal short-term safety data make it impossible to determine even an estimate of the risks in comparison with albumin. With respect to the altered coagulation caused by HES 130/0.4, this study reports equivalence in blood loss; however, I do note that the HES 130/0.4 group received more fresh-frozen plasma. Fresh-frozen plasma exposure was protocol driven based on clinical bleeding and prothrombin times so it is probable that HES 130/0.4 did indeed make a clinically significant, but controlled deterioration in coagulation. There are reports of clinical bleeding after use of HES 130/0.4 over a wide range of uses (3). Clear signals exist in the literature that long lasting, osmotic renal damage occurs in the face of prior renal disease and exposure to colloid substitutes. There is accumulation of the colloid within renal cells with effects that are both short and long term (4). Renal failure does occur with HES 130/0.4 and it does not seem to be 100% predicable based on pretreatment renal function (5, 6). In addition, long lasting and severe pruritus can occur with HES 130/0.4 (7), and there is 41⁄2 times the incidence of potentially life-threatening anaphylactic reactions compared with albumin (3). HES colloids, in general, are relatively contraindicated in the face of sepsis where renal dysfunction and mortality is higher than with other colloids (8). Hanart et al report solid, but preliminary, data that HES 130/0.4 seems equivalent to albumin within the first 24 hours after uncomplicated pediatric cardiac surgery. Now we need an evaluation of long-term efficacy and safety in a multicentered, randomized block design with true blinding at all stages. We are now challenged to take on the question, “Is HES 130/0.4 equivalent in efficacy and safety to albumin for pediatric congenital cardiac surgery?” before a fiscal decision voids a decision based on quality of care. Brian D. Hanna, MD, PhD Division of Cardiology, Children’s Hospital of Philadelphia Pennsylvania, PA Department of Pediatrics, School of Medicine, University of Pennsylvania Pennsylvania, PA REFERENCES 1. Albumin Product Insert, Octapharma, October 2006 2. Hanart C, Khalife M, De Villé A, et al: Perioperative volume replacement in children undergoing cardiac surgery: Albumin versus hydroxyethyl starch 130/0.4. Crit Care Med 2009; 37:696–701 3. Wiedermann CJ: Hydroxyethyl starch—Can the safety problems be ignored? Wien Klin Wochenschr 2004; 116:583–594 4. Dickenmann M, Oettl T, Mihatsch MJ: Osmotic nephrosis: Acute kidney injury with accumulation of proximal tubular lysosomes due to administration of exogenous solutes. Am J Kidney Dis 2008; 51:491–503 5. Boldt J: New light on intravascular volume replacement regimens: What did we learn from the past three years? Anesth Analg 2003; 97:1595–1604 6. Brunkorst FM, Oppert M: Nephrotoxicity on hydroxyethyl starch solution. Br J Anaesth 2008; 100:856 – 857 7. Bork K: Pruritus precipitated by hydroxyethyl starch: A review. Br J Dermatol 2005; 152:3–12 8. Weidermann CJ: Systematic review of randomized clinical trials on the use of hydroxyethyl starch for fluid management in sepsis. BMC Emerg Med 2008; 8:1 Making catheter-related bloodstream infections history: From the slogan to the serious strategy* C entral venous catheters (CVC) have become an essential and necessary component of the modern management of critically ill patients. However, the benefits of the CVC are often offset by the fact that such devices have now been recognized as the leading source of bloodstream infections in this critically ill patient population, which is associated with high morbidity and mortality (1). *See also p. 702. Key Words: catheter-related bloodstream infection; critically ill patients; anti-infective central venous catheters The author has received honoraria from Cook and royalties from Cook, AMX, Tyrx, and Horizon. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c3dd Crit Care Med 2009 Vol. 37, No. 2 For years, it has been recognized that catheter-related bloodstream infection (CRBSI) in critically ill patients is a preventable serious complication. In 1988, Maki et al (2) predicted that “binding an antimicrobial to the entire catheter surface may ultimately prove to be the most effective technological innovation for reducing the risk of device-related infections.” Subsequently, Maki et al have demonstrated that CVC impregnated with antimicrobial agents are the “most intensively studied technology for the prevention of CRBSI over the past 30 years” and have also shown that such anti-infective CVC (AI-CVC) are highly cost-effective, safe, and do not appear to select for resistance (3). In the systematic review and metaanalysis by Hockenhull et al (4) published in this issue of Critical Care Medicine, 38 prospective randomized controlled trials of AI-CVC were evaluated. Meta-analysis data from 27 trials have demonstrated a strong treatment effect in favor of AI-CVC (odds ratio: 0.49; 95% confidence interval: 0.3– 0.64). However, further subgroup analysis of different types of AICVC showed that the direction of the treatment effect favored the antimicrobial-coated catheters in all subgroups, except for the benzalkonium chloride– treated CVC. Further subanalysis demonstrated that the minocycline/ rifampin coating was associated with the most significant treatment effect (odds ratio: 0.26; 95% confidence interval: 0.15– 0.47), whereas other antimicrobial coatings varied as far as their efficacy with the second-generation chlorhexidine/silver sulfadiazine (CHSS⫹) “just failing to achieve statistical significance.” 789 The review and meta-analysis by Hockenhull et al also demonstrated that the AI-CVC is associated with a decrease in medical costs. They have estimated a cost savings in the United Kingdom for every patient who receives an AI-CVC to be equal to £138.2. This cost benefit is consistent with various analyses that have been demonstrated in multiple reviews conducted in the United States (5, 6). During the last two decades, various infection control interventions have also been shown to be effective in reducing CRBSI, including the use of maximal sterile barrier precautions during insertion of the CVC and applying chlorhexidine at the insertion site (7, 8). More recently, Pronovost et al (9) have shown that when these effective infection control interventions are used concurrently as part of a “bundle” (that includes maximal sterile barrier precautions, hand washing, cleaning the insertion site with chlorhexidine, and avoidance of femoral vein insertion and unnecessary prolonged use of the catheter), a significant decrease in CRBSI is observed in critically ill patients. Despite the fact that the mean rate of CRBSI in the Pronovost et al study was reduced from 7.7 per 1000 catheter days to 1.4 per 1000 catheter days over a 6 –18-month study with a reported median of zero, CRBSI continued to occur, despite the implementation of such infection control interventions. Although the Pronovost et al study included a large number of intensive care unit and critically ill patients, it demonstrated a significant and prolonged reduction in CRBSI that persisted for an 18month period. However, there were several limitations to the study. Among the limitations are the lack of assessment of compliance as far as the implementation of the infection control bundle, the crossover design of the study, the poor definitions of CRBSI included in the study, and the lack of assessment of confounding variables (such as the introduction of AI-CVC into the units being studied during the study period). Like Pronovost et al, other investigators have demonstrated that infection control interventions (such as the use of maximal sterile barriers) are associated with a decrease in CRBSI (7, 10). However, such measures on their own do not eliminate CRBSI completely. Although it is now well established that the infection control bundle is the mainstay of preventing CRBSI, it is also well recognized that 790 these measures are often associated with high cost and poor compliance, are not very durable, and do not completely eliminate or prevent infections (10). Hockenhull et al highlight the fact that “it is important to establish whether the strong treatment effect of AI-CVC remains after effective infection control bundles are established.” Two recent prospective randomized trials that are cited by Hockenhull et al could shed light on this issue (11, 12). In a multicenter prospective randomized study by Rupp et al (11), the second-generation CHSS⫹ was compared with uncoated CVC. During the trial, the infection control bundle was implemented into both arms of the study. In the uncoated CVC control arm, where such infection control bundle measures were implemented, the rate of CRBSI was 1.24 per 1000 catheter days. However, in the test arm, whereby the infection control bundles were implemented in addition to the use of AI-CVC (in this case CHSS⫹), the risk of CRBSI was decreased by more than three-fold to a level as low as 0.4 per 1000 catheter days (11). Another prospective randomized trial by Hanna et al (12) compared the use of an AI-CVC coated with minocycline and rifampin with uncoated CVC. Again, in this study, the elements of the infection control bundle were implemented, including the maximal sterile barrier precautions. In the uncoated CVC arm, the rate of CRBSI was 1.28 per 1000 catheter days, which was further reduced significantly with the use of the AI-CVC (coated with minocycline/rifampin) to a low level of 0.25 per 1000 catheter days (12). Hence, the AI-CVC could complement the infection control bundle measures in further bringing the rate of CRBSI to a very low level approaching zero and, therefore, making CRBSI a very rare entity in highrisk patients. Hockenhull et al also referred to as AI-CVC as a “safety net to prevent contaminating microorganisms from developing into CRBSI.” The referral of AICVC as a “safety net” is appropriate from several view points. First, it is well recognized that the absolute and complete compliance with all elements of the infection control bundle is never at a 100% level. The fact that the CVC is transformed into an anti-infective device that will prevent the microbial adherence of resistant pathogens represents another major barrier against biofilm colonization and, ultimately, CRBSI independent of human behavior and lack of compliance with infection control measures. Furthermore, the AI-CVC indeed serves as a safety net in that the infection control bundle, including maximal sterile barrier precautions and sterilization of the skin insertion site with chlorhexidine, do prevent contamination of the CVC during insertion. However, the AICVC do prevent biofilm colonization of the external and the internal surface not only during insertion but also subsequently during the dwell time of the catheter, where organisms could migrate from the external skin insertion site surface or from the hub into the lumen of the catheter. In addition, impregnating the CVC with antimicrobial agents is very much similar to the concept of inoculating the patient with a modified attenuated microbial organism through vaccination. Whereas good infection control measures, including good hygiene, are important in preventing the transmission of various infections, such as polio, measles, and mumps, the use of effective vaccines does, indeed, represent an important safety net to further eliminate such infections (13). Appropriate infection control precautions do not substitute, but rather complement, vaccination as an intervention. The same is true for the AI-CVC. In conclusion, the review and metaanalysis of Hockenhull et al does demonstrate the strong treatment effect that favors AI-CVC in the prevention of CRBSI, particularly in critically ill patients. Furthermore, the review demonstrates the cost effectiveness of such intervention, showing that such antimicrobial technology represents a safety net in the prevention of CRBSI. Given the new directives by the Central Medical Service System in the United States, including Medicare and Medicaid in not reimbursing hospitals for CRBSI, it is important to further press forward toward the elimination of such infections. To achieve such a zero end point and to realistically eliminate CRBSI, the use of AI-CVC in addition to effective infection control bundle should become the standard of care. Slogans such as “zero tolerance” will not achieve a zero end point of CRBSI and could not eliminate such preventable, serious infections unless they are associated with the implementation of a serious strategy that relies on evidence-based medicine. Using the combination of effective infection control measures with highly efficacious techCrit Care Med 2009 Vol. 37, No. 2 nologies, such as AI-CVC, is a solid foundation for this serious strategy. Issam Raad, MD, FACP, FIDSA Department of Infectious Diseases, Infection Control and Employee Health The University of Texas M. D. Anderson Cancer Center Houston, TX REFERENCES 1. Richards MJ, Edwards JR, Culver DH, et al: Nosocomial infections in medical intensive care units in the United States. National Nosocomial Infections Surveillance System. Crit Care Med 1999; 27:887– 892 2. Maki DG, Cobb L, Garman JK, et al: An attachable silver-impregnated cuff for prevention of infection with central venous catheters: A prospective randomized multicenter trial. Am J Med 1988; 85:307–314 3. Crnich CJ, Maki DG: Are antimicrobialimpregnated catheters effective? Don’t throw out the baby with the bathwater. Clin Infect Dis 2004; 38:1287–1292 4. Hockenhull JC, Dwan KM, Smith GW, et al: 5. 6. 7. 8. 9. The clinical effectiveness of central venous catheters treated with anti-infective agents in preventing catheter-related bloodstream infections: A systematic review. Crit Care Med 2009; 37:702–712 Marciante KD, Veenstra DL, Lipsky BA, et al: Which antimicrobial impregnated central venous catheter should we use? Modeling the costs and outcomes of antimicrobial catheter use. Am J Infect Control 2003; 31:1– 8 Shorr AF, Humphreys CW, Helman DL: New choices for central venous catheters: Potential financial implications Chest 2003; 4:275–284 Raad II, Hohn DC, Gilbreath BJ, et al: Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertions. Infect Control Hosp Epidemiol 1994; 15(4 Pt 1):231–238 Maki DG, Ringer M, Alvarado CJ: Prospective randomized trial of povidone-iodine, alcohol, and chlorhexidine for prevention of infection associated with central venous and arterial catheters. Lancet 1991; 338:339 –343 Pronovost P, Needham D, Berenholtz S, et al: An intervention to decrease catheter-related bloodstream infections in the ICU (published 10. 11. 12. 13. corrections appears in N Engl J Med 2007; 356:2660). N Engl J Med 2006; 355: 2725–2732 Sherertz RJ, Ely EW, Westbrook DM, et al: Education of physicians-in-training can decrease the risk for vascular catheter infection. Ann Intern Med 2000; 132:641– 648 Rupp ME, Lisco SJ, Lipsett PA, et al: Effect of second generation venous catheter impregnated with chlorhexidine and silver sulfadiazine on central catheter-related infections: A randomized controlled trial. Ann Intern Med 2005; 143:570 –580 Hanna HA, Benjamin R, Chatzinikolaou I, et al: Long-term silicone central venous catheters impregnated with minocycline and rifampin decrease rates of catheter-related bloodstream infection in cancer patients: A prospective randomized clinical trial (published correction appears in J Clin Oncol 2005; 23:3652). J Clin Oncol 2004; 22: 3163–3171 Centers for Disease Control and Prevention (CDC): Progress toward interruption of wild poliovirus transmission—Worldwide, January 2007-April 2008. MMWR Morb Mortal Wkly Rep 2008; 57:489 – 494 The ethics of quality improvement research* I n December 2006, an article in the New England Journal of Medicine showed how a simple intervention virtually eliminated central venous catheter-related blood stream infections at hospitals participating in the Michigan Keystone ICU Project (1). Lay and medical authors hailed the findings that suggested opportunities to save thousands of lives and billions of dollars (1–3). Soon afterward, the Office for Human Research Protections (OHRP), responding to an anonymous complaint, launched an investigation culminating in Keystone’s suspension, citing the failure to obtain institutional review board (IRB) approval at participating hospitals and informed consent from patients (4). At first blush, the action seems absurd. The intervention *See also p. 725. Key Words: quality improvement; human subjects research; research ethics; institutional review board; informed consent; expedited review The authors have not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194c4d6 Crit Care Med 2009 Vol. 37, No. 2 studied was harmless—simply formal implementation of standard practice (1, 5). Outrage followed, fueled by concern that the OHRP’s action would discourage similar projects (2, 6 – 8). Leaders of professional organizations protested (6), whereas authors in the lay press labeled the action “bizarre and dangerous” (2). As one ethicist suggested, the government office responsible for protecting human subjects seemed to have taken action that would only increase harm (7). In this issue of Critical Care Medicine, Savel et al (9) illuminate the factors precipitating the OHRP’s actions, particularly the lack of consensus regarding the relationship between quality improvement (QI) and human subjects research (HuSR). In advance of the project, the IRB at the lead investigator’s institution, Johns Hopkins, had exempted Keystone from review and waived the informed consent requirement (1). The OHRP disagreed with this exemption, believing the study constituted HuSR, thus mandating oversight from each participating hospital and informed consent (4). Months later, the OHRP seemed to modify its findings, indicating that research like Keystone that entailed negligible risk would qualify for expedited review and waiver of consent (10). Projects that were QI only would not need IRB oversight. Although the OHRP’s final opinion may have closed this particular case, Savel et al (9) argue persuasively for more clarity regarding the oversight needed for work combining QI and HuSR. To prevent future uncertainty, they offer three recommendations. First, they suggest streamlining approval for QI/HuSR and clarifying the rules guiding the use of central IRBs and waiver of informed consent. Second, they suggest developing ways to make IRB approval less onerous. Third, they suggest that hospitals too small to have IRBs could use IRBs from “nearby regional centers of excellence.” Several others have contributed to this debate (7, 8, 11, 12). Miller and Emanuel (12) believe three questions are key to evaluating projects like Keystone. First, does the project involve HuSR? Second, if yes, is expedited review appropriate? Third, is informed consent needed? At present, there is little controversy regarding questions two and three: expedited review and waived consent 791 would be appropriate if the intervention entails minimal risk, obtaining formal consent is unfeasible, and the waiver will not adversely affect the rights and welfare of subjects (8 –10, 12). However, the first question remains unanswered. Was Keystone HuSR or not? The question is important because HuSR, by definition, requires IRB review, and institutions engaged in federally funded research must file a Federalwide Assurance with the U.S. government (13). However, requiring IRB oversight for studies not involving human subjects could impose unnecessary barriers (7). Many small hospitals do not have IRBs (9). Central IRBs, which could streamline approval (8, 9), are not always available. Furthermore, institutions committed to using a central IRB still have to file a Federalwide Assurance, assuring compliance with HuSR regulations (13). Such documentation could overwhelm institutions lacking research infrastructures and discourage worthy projects. In contrast, if QI projects like Keystone are not HuSR, then IRB oversight is unnecessary. The Code of Federal Regulations Common Rule, which governs human research activity in the United States, defines research as a “systematic investigation. . . designed to develop or contribute to generalizable knowledge (13).” By this definition, Keystone, and work like it, seems to constitute research. The Code of Federal Regulations goes on to define a human subject as “a living individual about whom an investigator . . . conducting research obtains 1) data through intervention or interaction with the individual or 2) identifiable private information.” So was Keystone HuSR or something else? The technique used to reduce central venous catheter infections had already been proven effective; its value in individual patients was not being questioned (5). Rather, Keystone asked if the technique could reduce institutional infection rates (1). Thus, we agree with Baily (7) that the research did not meet the regulatory definition of HuSR. Instead, it seems to have been an example of QI coupled with research on organizations and did not require IRB oversight. In this view, the full implications of what is often perceived as an inflexible and burdensome system (IRB application, review, and approval; negotiation of a Federalwide Assurance for those without 792 one; and continuing review and review of all changes before their implementation) need not be invoked. Another ethical issue surrounds the notion of informed consent as it relates to this project. Savel et al state that the purpose of informed consent is to protect the subjects from whatever risks may be inherent in the research project. But, we would argue that informed consent flows more from the ethical principle of Respect for Persons, as described in the Belmont report (14). This principle relates to a respect for autonomy, allowing individuals to make their own decisions and accept whatever risks are inherent in the project. It reflects the concept of “voluntariness” in deciding to participate in research. This is in contrast to QI implementation, which explicitly is not done on a voluntary basis, but rather is an operational implementation on the part of healthcare organizations. Thus, it could be argued that the Keystone project did not invoke the need for informed consent for participation in research. Research like the Keystone project offers useful guidance to healthcare institutions required to find ways to successfully implement recommended practices (1). Although unfortunate in many ways, the Keystone/OHRP saga has offered an opportunity to consider the ethical framework on which the nascent field of QI research is being built. The resulting discussion, including the contribution by Savel et al, should ensure that future work is performed both effectively and ethically. Mark D. Siegel, MD, FCCP Pulmonary and Critical Care Section Department of Internal Medicine Yale University School of Medicine New Haven, CT Sandra L. Alfano, PharmD, FASHP, CIP Chair, Human Investigation Committees Department of Internal Medicine Yale University School of Medicine New Haven, CT REFERENCES 1. Pronovost P, Needham D, Berenholtz S, et al: An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med 2006; 355:2725–2732 2. Gawande A: A lifesaving checklist. New York Times. December 30, 2007. Available at: http://www.nytimes.com/2007/12/30/opinion/ 30gawande.html. Accessed October 22, 2008 3. Wenzel RP, Edmond MB: Team-based prevention of catheter-related infections. N Engl J Med 2006; 355:2781–2783 4. Borror KC: Human research protections under Federalwide Assurances FWA-5752, FWA-287, and FWA-3834. July 19, 2007. Available at: http://www.dhhs.gov/ohrp/detrm_letrs/YR07/ jul07d.pdf. Accessed October 5, 2008 5. Berenholtz SM, Pronovost PJ, Lipsett PA, et al: Eliminating catheter-related bloodstream infections in the intensive care unit. Crit Care Med 2004; 32:2014 –2020 6. Hanson MD, Thomas A, Ingbar DH, et al: Letter to the Honorable Michael Leavitt. February 5, 2008. Available at: http:// www.hospitalmedicine.org/AM/Template. cfm?Section⫽Advocacy_Policy&Template⫽/ CM/ContentDisplay.cfm&ContentID⫽16830. Accessed September 28, 2008 7. Baily MA: Harming through protection? N Engl J Med 2008; 358:768 –769 8. Kass N, Pronovost PJ, Sugarman J, et al: Controversy and quality improvement: Lingering questions about ethics, oversight, and patient safety research. Joint Comm J Qual Patient Saf 2008; 34:349 –353 9. Savel RH, Goldstein EB, Gropper MA: Critical care checklists, the Keystone Project, and the Office for Human Research Protections: A case for streamlining the approval process in quality-improvement research. Crit Care Med 2009; 37:725–734 10. Borror KC: Human subject research protections under Federalwide Assurances FWA5752, FWA-287, and FWA-3834. February 14, 2008. Available at: http://www.hhs.gov/ ohrp/detrm_letrs/YR08/feb08b.pdf. Accessed October 22, 2008 11. Lynn J, Baily MA, Bottrell M, et al: The ethics of using quality improvement methods in health care. Ann Intern Med 2007; 146:666 – 673 12. Miller FG, Emanuel EJ: Quality-improvement research and informed consent. N Engl J Med 2008; 358:765–767 13. Protection of human subjects: Basic HHS policy for protection of human research subjects. June 23, 2005. Available at: http:// www.hhs.gov/ohrp/humansubjects/guidance/ 45cfr46.htm#46.102. Accessed October 19, 2008 14. The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research: The Belmont report, ethical principles and guidelines for the protection of human subjects of research. April 18, 1979. Available at: http://www.hhs.gov/ohrp/ humansubjects/guidance/belmont.htm. Accessed August 15, 2008 Crit Care Med 2009 Vol. 37, No. 2 Refractory shock in the intensive care unit—Don’t fail to spot obstruction of the left ventricular outflow tract!* L eft ventricular outflow tract obstruction (LVOTO) is a wellrecognized feature of hypertrophic cardiomyopathy (1). The condition of LVOTO, in the absence of hypertrophic cardiomyopathy, has also been described in stress echocardiography with dobutamine (2– 4) or exercise (5, 6) and after cardiac valve surgery (7– 10). The main mechanisms behind the development of LVOTO, also referred to as dynamic LVOTO, are inotropic stimulation of left ventricles (LV) with higher than normal LV wall to lumen ratios such as in hypertensive heart disease. Thus, forceful contractions, particularly of the basal part of an LV with small dimensions and increased wall thickness in combination with peripheral vasodilation, may precipitate subaortic flow obstruction and a decline in cardiac output and hypotension. If the LVOTO is severe enough, blood will be ejected at a high velocity, which may cause the anterior mitral valve leaflet to be drawn to the septum by a Venturi effect, a so-called systolic anterior motion (SAM). Thus, parts of the anteriorly displaced mitral valve leaflet will extend beyond their coaptation point and protrude into the rapid flow velocity of the left ventricular outflow tract. This may result in a further LVOTO and a more or less pronounced mitral valve regurgitation and eventually cardiogenic shock. In cardiac surgery, dynamic LVOTO is a well-recognized phenomenon, which may develop early after aortic valve replacement (7, 8) or mitral valve repair (9, 10) and may be totally reversible after correction of the precipitating factors such as high inotropic state, tachycardia, hypovolemia, and systemic vasodilation. *See also p. 729. Key Words: left ventricle outflow tract obstruction; cardiogenic shock; left ventricle hypertrophy; hypovolemia; cardiac inotropism The author has not disclosed any potential conflicts of interest. Copyright © 2009 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0b013e318194fe38 Crit Care Med 2009 Vol. 37, No. 2 LVOTO with SAM may also occur in the perioperative setting of noncardiac surgery. Luckner et al (11) described two patients undergoing orthopedic surgery who developed cardiovascular collapse early after induction of anesthesia unresponsive to catecholamines and a third patient with severe intraoperative bleeding. LVOTO and SAM were diagnosed by transesophageal echocardiography, and hypotension was resolved by colloid infusion, phenylephrine, and ␤-blocker therapy. In these patients, LVOTO and SAM were triggered by hypovolemia and anesthetic drug–mediated vaso- and venodilation in conjunction with increased catecholamine concentrations. In this issue of Critical Care Medicine, Chockalingam et al (12) describe the clinical recognition and management options of five patients (all women) presenting with unexplained hypotension and systolic murmur in the emergency/critical care setting. The underlying cause of their critical condition was dynamic LVOTO, with peak left ventricular outflow tract pressure gradients ranging from 60 to 150 mm Hg, diagnosed by echocardiography. The common clinical feature of all patients was hypertensive heart disease with more or less pronounced LV hypertrophy. Three patients presented in the emergency room with chest discomfort, anterior T-wave abnormalities, suggestive of myocardial ischemia, and mild to moderate troponin-I elevations. Two of these patients had a normal coronary angiogram, and one had a significant stenosis of the posterior (not anterior) descending artery. Dobutamine treatment was initiated in one patient because of shock. Severe LV apical dysfunction (apical ballooning syndrome) was confirmed in these three patients in association with LVOTO and SAM of the anterior mitral valve leaflet. Echocardiogram also revealed hypercontractility of the basal portions of the LV as a compensatory mechanism for the apical ballooning syndrome–related severe apical dysfunction, which resulted in LVOTO and SAM. The apical ballooning syndrome, also known as Takatsubo cardiomyopathy, is a reversible cardiomyopathy, which is triggered by severe psychological stress occurring in older women with normal coronary arteries, and may mimic evolving acute myocardial infarction and coronary syndrome (13). This condition is associated with a severe elevation of plasma catecholamines (14), which further strengthens the importance of cardiac sympathetic stimulation for the development of LVOTO and SAM in patients with aortic ballooning syndrome. Mild to moderate troponin elevations were observed in these patients with virtually normal coronary arteries. According to the authors, this may be explained by high levels of peak systolic subendocardial wall stress, which may considerably impair the oxygen demand–supply relationship of the subendocardial layers. These patients were successfully treated with ␤-blockers, and there was no evidence of LVOTO or SAM on repeat echocardiograms. The authors also review two mechanically ventilated patients in the intensive care unit presenting with hypotension, tachycardia, systolic murmur, and mild to moderate troponin elevations. In one of the patients, the clinical condition was not improved by dopamine treatment. Bedside echocardiograms revealed hyperdynamic LVs with hypertrophy and LVOTO in both cases, with or without SAM, which was fully resolved by vigorous fluid treatment in combination with a ␤-blocker. These two cases highlight the clinical problem one may encounter when inotropic treatment is instituted in volume-depleted, iatrogenically or not, patients in the intensive care unit with hypertrophied LVs, particularly when inotropic agents with vaso- and venodilating properties are used, e.g., dobutamine, low-dose dopamine, phosphodiesterase inhibitors, or levosimendan. Furthermore, this report emphasizes the importance of the use of echocardiography for early diagnosis of LVOTO in 793 patients in the intensive care unit with refractory shock not responsive to “fluid challenge” and inotropic therapy. An echocardiogram in this situation may reveal that the fluid challenge was not vigorous enough and that inotropic support of the hypertrophied, hypovolemic, and hyperdynamic LV generated an LVOTO and caused a functional aortic stenosis in an LV not used to such high levels of afterload. The echocardiogram will also serve as a guide in the treatment of this critical condition consisting of blood volume expansion, ␤-blockers, and normalization of a low systemic vascular resistance with a vasoconstrictor, if necessary. This treatment strategy may completely resolve this clinical problem. The authors of this interesting report conclude that dynamic LVOTO may occur in the emergency room and the critical care units more often than is recognized. To date, the true incidence of dynamic LVOTO as a cause of hypotension in the intensive care unit is not known, and most likely, this condition is underdiagnosed. The good news is that early recognition and appropriate medical management may dramatically improve the condition of these patients. Sven-Erik Ricksten, MD, PhD Department of Cardiothoracic Anesthesia and Intensive Care 794 Sahlgrens University Hospital Gothenburg, Sweden 8. REFERENCES 1. Wigle ED, Rakowski H, Kimball BP, et al: Hypertrophic cardiomyopathy: Clinical spectrum and treatment. Circulation 1995; 92:1680 –1692 2. Pellika PA, Oh JK, Bailey KR, et al: Dynamic intraventricular obstruction during dobutamine stress echocardiography. A new observation. Circulation 1996; 86: 1429 –1432 3. Sorrentino MJ, Marcus RH, Lang RM, et al: Left ventricular outflow tract obstruction as a cause for hypotension and symptoms during dobutamine stress echocardiography. Clin Cardiol 1996; 19:225–230 4. Scandura S, Arcidiacono S, Felis S, et al: Dynamic obstruction to left ventricular outflow during dobutamine stress echocardiography: The probable mechanisms and clinical implications. Cardiologia 1998; 43: 1201–1208 5. Peteiro J, Montserrat L, Castro-Beiras A: Labile subaortic obstruction during exercise stress echocardiography. Am J Cardiol 1999; 84:1119 –1123 6. Cabrera-Bueno F, Garcia-Pinilla JM, GomezDoblas JJ, et al: Beta-blocker therapy for dynamic left ventricular outflow induced by exercise. Int J Cardiol 2007; 117:222–226 7. Aurigemma G, Battista S, Orsinelli, et al: Abnormal left intracavitary flow acceleration in patients undergoing aortic valve replacement for aortic stenosis. A marker for high 9. 10. 11. 12. 13. 14. postoperative morbidity and mortality. Circulation 1992; 86:926 –936 Bartunek J, Sys SU, Rodrigues AC, et al: Abnormal left ventricular intracavitary flow acceleration after valve replacement for aortic stenosis. Circulation 1996; 93: 712–719 Freeman WK, Schaff HV, Khanderia BK, et al: Intraoperative evaluation of mitral valve regurgitation and repair by transesophageal echocardiography: Incidence and significance of systolic anterior motion. J Am Coll Cardiol 1992; 20:599 – 609 Kreindel MS, Schiavone WA, Lever HM, et al: Systolic anterior motion of the mitral valve after Carpentier ring valvuloplasty for mitral valve prolapse. Am J Cardiol 1986; 57:408 – 412 Luckner G, Margreiter J, Jochberger S, et al: Systolic anterior motion of the mitral valve with left ventricular outflow tract obstruction: Three cases of acute perioperative hypotension in noncardiac surgery. Anest Analg 2005; 100:1594 –1598 Chockalingam A, Dorairajan S, Bhalla M, et al: Unexplained hypotension: The spectrum of dynamic left ventricular outflow obstruction in critical care settings. Crit Care Med 2009; 37: 729–734 Sharkey SW, Lesser JR, Zenovich AG, et al: Acute and reversible cardiomyopathy provoked by stress in women from the United States. Circulation 2005; 111: 472– 479 Wittstein IS, Thiemann DR, Lima JA: Neurohormonal features of myocardial stunning due to sudden emotional stress. N Engl J Med 2005; 352:539 –548 Crit Care Med 2009 Vol. 37, No. 2