N-Acetylcysteine Enhances Recovery

From Acute Lung Injury in Man*

A Randomized, Double-Blind, Placebo-Controlled
Clinical Study
Peter M. Suter, M.D., F.C.C.P.; Guido Domenighetti, M.D.;
Marie-Denise Schaller, M.D.; Marie-Christine Laverriere, M.D.;
Rudolf Ritz, M.D.; and Claude Perret, M.D., F.C.C.P.

Objective: To detennine the effects of intravenous N- (difference not statistically significant). At admission, 22 of
acetylcysteine (NAC) on the development of severe adult 32 patients (69 percent) in the NAC group were mechani-
respiratory distress syndrome (ARDS) and mortality rate cally ventilated compared with 22 of 29 (76 percent) in the
in patients with mild-to-moderate acute lung injury and placebo group. At the end of the treatment period (day 3),
to analyze the duration of ventilatory support and Flo1 5 of 29 (17 percent) in the NAC group and 12 of 25 (48
required as weD as the evolution of the lung injury score. percent) in the placebo group were stiU receiving ventila-
Setting: Three university hospital ICUs and one regional tory support (p = O.ol), 1be Flo2 was 0.37less than admis-
ICU in Switzerland. sion value (day O) in the NAC group, and 0.20 less in the
Patients: Sixty-one adult patients presenting with mild- placebo group (p < 0.04); the oxygenation index (Pa0plo2)
to-moderate acute lung injury and various predisposing improved significantly (p < 0.05) from day 0 to day 3 only in
factors for ARDS received either NAC, 40 mglkgld, or the NAC-treated group. 1be LIS showed a significant
placebo intravenously for 3 days. regression (p = 0.003) in the NAC-treated group during the
MetJIJUrements: Respiratory dysfunction was assessed first 10 days of treatment: no change was observed in the
daily according to the need for mechanical ventilation placebo group. No adverse effects were observed during
and Flo2 , the evolution of the lung injury score, and the the treatment with NAC.
Pa0plo2 ratio. The cardiovascular state, liver function, Conclusions: Intravenous NAC treatment during 72 h
and kidney function were also monitored. Data were improved systemic oxygenation and reduced the need
collected at admission (day 0), during the Arst 3 days, for ventilatory support in patients presenting with mild-
and on the day of discharge from the ICU. to-moderate acute lung injury subsequent to a variety of
Results: The NAC and placebo groups (32 and 29 underlying diseases. Development of ARDS and mortal-
patients, respectively) were comparable at ICU admis- ity were not reduced signincantly by this therapy.
sion for severity of illness assessed by the simplined (Chest 1994; 105:190-94)
acute physiology score (SAPS) (10.8::!:: 4.6 vs 10.9::!:: 4.8)
and lung injury score (LIS) (1.39::!:: 0.95 vs 1.11 ::!:: 1.08) ARDS =adult respiratory distress syndrome; GSH " ~lu­
(mean::!:: SD). Three patients in each group developed tathione; ICU ,. intensive care unit; LIS =lung inJury
score; NAC = N-acetylcysteine; NS =not significant;
ARDS. The 1-month mortality rate was 22 percent for SAPS = simplified acute physiology score
the NAC group and 35 percent for the placebo group

A cute lung injury characterized by a high-perme- role. 6 ·7 Oxygen metabolites such as superoxide anion
(0 2 ' ), hydrogen peroxide (Hp 2 ), hydroxyl radical
ability, low-pressure pulmonary edema can de-
velop following a number of predisposing diseases ('OH), and hypochlorous acid (HOC!) cause increased
resulting in different degrees of respiratory insuffi- vascular permeability in endothelial cell monolayers,N
ciency. Patients with mild-to-moderate injury could, in isolated lungs, 9 -10 and in vivo . 11 In human ARDS,
within this continuum, represent an interesting subset increased oxidant activity has been observed in ex-
to evaluate a therapeutic approach of lung dysfunc- pired breath or in bronchoalveolar lavage (BAL)
tion . 1-~ The pathogenesis of acute lung injury and its fluid 1 2 - 1 ~ as well as in granulocytes and red blood
severest form, the adult respiratory distress syndrome cells. 15 ·16 In addition, patients with ARDS seem to
(ARDS), involves a number of mechanisms and medi- have a marked deficiency of the tripeptide antioxidant
ators. u .fi Toxic oxygen radicals of intra-cellular as well glutathione (GSH) in the extracellular epithelial lining
as extracellular origin seem to play an important fluid of the lower respiratory tract. 17 Glutathione acts
as an antioxidant for Hp 2• 1N The GSH deficiency
observed in ARDS could favor oxidative stress, thus
*From the Intensive Care Units of the Departme nts of Anesthesia allowing functional and/or morphologic damage in the
and Ml•dieine. Universitv Hospitals of Gt>neva (Drs. Suter and lower respiratory tract, including an overshooting
Lawrrii•rt> ), Lausanne (Drs. SdHtller and Pe rret ), and Basel (Dr.
Hitz). and the Hegional Hospital "La Caritit," Locarno (Dr. inflammatory reaction and epithelial and endothelial
Donw nighetti) , Switzt> rland. lesions resulting in pulmonary edema. N-acetylcyste-
Manuscript received Januarv 7, 1993: rt'vision accepted May 27.
Rq1rint requests: Dr. Suter, Surgical 1CU, Hopital Cantonal ine (NAC) is a GSH agonist,l 9 and previous studies
Uui r.;ersitaire, CH-1200 Genew 14, Srcit;:,er/mul have demonstrated that NAC administration increases

190 Enhanced Recovery From Acute Lung Injury (Suter eta/)

GSH levels in red blood cells, granulocytes, and Respiratory dysfunction was assessed daily by the requirement
plasma of patients with ARDS.15.20 for mechanical ventilatory support, the Fio2 administered, and the
evolution of the lung injury score~! using its first three components
Increasing GSH levels in the early phases of acute
{chest radiograph, Pa01flo2 ratio, and respiratory system compli-
lung injury with NAC could reduce or limit the extent ance). The Flo1 was controlled during mechanical ventilation using
of epithelial and endothelial damage and improve the the built-in oxygen cell of the ventilator. In extubated and sponta-
clinical course. N-acetylcysteine has been used at high neously breathing patients, Flo1 was regulated by using a Ventimask
dosages in the treatment of paracetamol intoxication or a similar device. The Flo1 obtained was checked regularly with
and was well tolerated. 21 a paramagnetic oxygen sensor {Oxydig, Drager, LUbeck, Germany).
Mechanical ventilation was initiated and maintained at the
The purpose of the present study was to assess the discretion of the physician in charge of the patient. The standards
effects of NAC on the development of ARDS and for mechanical ventilatory support, oxygen therapy. weaning crite-
mortality in patients with mild-to-moderate acute lung ria, and other treatment modalities are applied quite uniformly in
injury associated with known risk factors predisposing the four ICUs participating in the trial. Clinical and biochemical
to ARDS. Other important end points, such as dura- assessment was done regularly. Cardiovascular state, central ner-
vous system, kidney, and liver function were monitored closely.
tion of ventilatory support, Flo2 requirements, and
The patients were randomized to receive either NAC, 40 mglkgl
the clinical course of respiratory function were iden- d, or placebo, administered as a continuous intravenous infusion
tified at the end of the investigation and analyzed. over the first 3 days after admission to the ICU. This dose and
duration of treatment were chosen according to previous work on
MATERIALS AND METHODS
NAC in ARDS 15 and in other abnormalities. Adose of 40 mglkg is
one order of magnitude higher than what is normally applied in
Subjects and Protocol regimens of chronic respiratory disease and yet one order lower
During a 12-month period, patients with risk factors known to than the dose for paracetamol intoxication. We believed that this
predispose to the development of ARDS (see below), and presenting dose could offer sufficient efficacy as a preventive therapy in ARDS
with mild-to-moderate acute lung injury, were included in the trial without exposing to drug toxicity.2 1 N-acetylcysteine and placebo
in the four participating intensive care units (ICUs) in Switzerland were packed in similar numbered vials allowing randomization and
(Geneva, Lausanne, Basel, Locamo). double-blind administration.
In order to collect patients with comparable lung dysfunction for Informed consent was obtained from the patient or, if this was
a pharmacologic treatment regimen, we used the expanded defini- not possible because of the clinical condition, from the next of kin.
tion suggested by Murray et al,12 and only considered patients The protocol was submitted to and approved by the committees for
presenting with an initial lung injury score (LIS) between 0.1 and ethics in human research of the university hospital of Basel,
2.5. Patients with cardiogenic pulmonary edema and/or chronic
heart failure were excluded on the basis of medical history, results
Table 1-C~ ofHrtienl Group., SAPS Seal.
of clinical examination, and the use of a pulmonary artery catheter
and l.Mng Injury ~ (US) at Admiaion to the ICU*
in all unclear situations. The following predisposing factors for the
Treatment Group N-acetylcysteine Placebo
development of ARDS were considered.23
Sepsls, defined as proposed by Bone, 24 was clinical evidence of No. of patients 32 29
infection, le, respiratory rate above 20 cycles/min or minute Sex, MIF MIS 2316
ventilation over 10 Umin if mechanically ventilated, heart rate Age,y 46.6±19.7 48.1±21.9
more than 90 beats/min, core or rectal temperature outside the range Rarige 16-76 16-78
of 35.5° to 38.0°C, a white blood cell count above 12,000 or below ~ risk factors (n)
4,000/j11 or 20 percent or more immature cells plus evidence of altered Multiple trauma 9 7
organ perfusion (le, acute change in mental status, Pa01flo2 less than Multiple trauma and 7 5
280, plasma lactate concentration greater than upper limit of normal, aspiration
and urine output below 0.5 nil/kg of body weight for at least 1 h). Sepsis 5 6
Multiple trauma included patients with multiple major fractures Sepsis and aspiration 1 3
(two or more major long bones or !instable pelvic fracture) associated
Aspiration 6 3
with trauma to another region of the body such as craniocerebral or
Near drowning 2 1
abdominal, requiring surgical intervention.
Necrotizing pancreatitis 1 1
Aspiration was defined as recent (during the previous 6 h)
Hemorrhagic shock 1 3
inhalation of gastric contents, documented by suctioning of gastric
Associated diseases and complications
material from the bronchial tree or by fiberoptic bronchoscopy
Coma 17 13
shOwing typical mucosal lesions.
Necrotizing pancreatitis was seen as severe abdominal pain,
Pulmonary infection 11 6
Shock 7 10
vomiting, increased serum amylase levels, circulatory shock, and a
Ranson score of 3 or above.15 Septicemia 8 9
Hemorrhagic shock was defined as requiring administration of Acute reDal failure 3 3
more than 20 U of blood within 24 h. Simplified acute physiology 10.8±4.60 10.9±4.80
Near drowning was defined as an immersion accident requiring score""
endotracheal intubation. us• 1.39±0.95 1.11±1.08
Patients younger than 16 years, pregnant wome.n, immunocom- Outcome
promised patients, and those with a severe lung injury {LIS > 2.5) 22 ICU ~y. days, 11.3±10.5 12.2±9.6
or cardiogenic pulmonary edema were excluded from this trial. mean.±SD
Data were callected at admission to the ICU {baseline), on the first Development of ARDS, n 3 3
3 days after admission, and on the day of discharge from the unit. Mortality rate, n (~) 7 (22) 10 (35)
Severity of illness was assessed by the simplified acute physiological
•Differences between the two patient groups are statistically not
score {SAPS). 26
significant.

CHEST I 105 I 1 I JANUARY, 1994 191

 -- •
Geneva, and Lausanm•.
80 NAC
Statistical Methods
0~
70 ~ PlacebO

- so
Initial homogeneity assessment of the two groups was performed
by I test and X' analysis. The before and after comparisons were U) 60
done with the sign test, the paired I test, and the Wilcoxon-Pratt
test depending on the distribution of the data. Between-group
cCl)
comparisons were performed with the X' test, the Student I test, ;::
ca
and the Wilcoxon-Man-Whitney fU test according to the distribu- CL 40

-
tion of the data. Data are presented as mean values :t standard 'tJ
de,iations unless speeifled otherwise. Cl)
30
.!!
RESULTS ;:: 20
cCl)
Characterization of Patients > 10
A total of 61 patients (47 men and 14 women) were 0
recruited for the study (Table 1). All were considered
evaluable for data analysis. Thirty-two patients received
NAC and 29 received placebo. The two groups were well Days
matched for demographic characteristics on trial entry. FIC:URE l. Percentage of patients requiring mechanical ventilatory
The SAPS for severity of disease26 was 10.8 :!:: 4.6 in the ~upport in the N-ac<'tykysteine (NAC) and placebo group. Evolution
NAC group and 10.9 :!:: 4.8 in the placebo group (x :!:: SD, from days 0 (admission) through day 3 (end of NAC treatment). The
triangle identifies a significant difference between the two groups at
not significant [NS]). On admission, the lung injury day 3 (p =0.01). The asterisks identifY a significant differen(-e in the
score 22 was 1.39 :!:: 0.95 in the NAC group, a value not treated group on day 2 and 3 vs admission (day 0; p = 0.01).
significantly different compared with the placebo group
of 1.11 :!:: 1.08. Risk factors and associated conditions the NAC group from 1.39 :!:: 0.95 to 0.67 :!:: 0.70,
were similar in the two groups. Three patients in each p <0.01, between ICU admission and day 10. No
group developed severe lung injury (ARDS) defined as a significant change was observed in the placebo group.
LIS above 2.5Y Seven of 32 patients (22 percent) in the At ICU admission, the mean chest radiograph score
treatment group and 10 of 29 (3.5 percent) in the placebo was higher in the NAC than in the placebo group
group died (NS). During the study. adverse events due to ( 1.8 :!:: 1.1 vs 1.1 :!:: 1.1; p < 0.05). Up to day 3 this
NAC were not detected. score increased in the placebo group (p < 0.05), whereas
it did not change significantly in the NAC group up to
VentilatonJ Support, Flo 2 , Lung Function day 3, but decreased until the discharge from the ICU
Twenty-two patients in each group (69 percent in (from 1.8 :!:: 1.1 at admission to 1.1 :!:: 1.1, p < 0.01).
the NAC and 76 percent in the placebo group)
DISCUSSION
required ventilatory support at ICU admission (day O);
at day 3, the percentage of patients receiving ventila- Despite major improvements in patient care, ARDS
tory support was significantly reduced in the treated still remains a formidable clinical challenge and carries
group from 69 to 17 percent (p = 0.01 ), whereas in the
placebo group this decrease (from 76 to 48 percent) 400
- - - - - NAC

... d
was not significant, p = 0.01 between groups (Fig 1).
At day 0, there was no significant difference between
•·····• Placebo
**
the two groups \vith regard to Flo 2 administered. At
.
N
day 3, the Flo 2 was significantly decreased in the NAC
group (0.29 :!:: 0.09 vs 0.48 :!:: 0.24 at admission;
0
ir *
- 300
p < 0.01) whereas in the placebo group there was no
~ .. ·f I
significant decrease (0.35:!:: 0.11 vs 0.45:!:: 0.20). The
difference in Flo 2 between the two groups was signifi-
cant on day 3 (p < 0.05). The oxygenation index PaOj
--··· f
Flo 2 was similar at admission for both groups (25.5 :!:: 113
mm Hg in the NAC vs 248 :!:: 99 mm Hg in the 200 --------------------~,~~------.----
0 2 3 DISCHARGE
placebo group. NS) and increased significantly in the
Days
NAC group reaching 294 :!:: 99 mm Hg on day 3
(p<0.05). No significant change was noted in the FIGURE 2. Mean values:tSEM of PaOjFio, (hypoxemia index, 22)
at admission (day 0) through day 3 (end of N-acetylcysteine [NAC]
placebo group (Fig 2). There was no statistical differ- treatment) and the day of discharge from the ICU in treated and
ence in PaOJFio2 between the NAC- and placebo- placebo-treated patients. One ast<•risk indicates p < 0.05 vs day 0;
treated patients. The lung injury score decreased in two asterisks. p = 0.01 vs day 0.

192 Enhanced Recovery From Acute Lung Injury (Suter et at)

a high mortality rate. 27 Recent clinical studies suggest REFERENCES
that patients with ARDS undergo relevant oxidative 1 Asbaugh DC, Bigelow DB, Petty TL, Levine BE. Acute
stress. 12- 14•16•19•28 Toxic oxygen products cause -cellular respiratory distress in adults. Lancet 196i; 2:319-23
2 Rinaldo JE, Rogers RM . Adult respiratory distress syndrome:
and subcellular damage and activate epithelial cells,
changing concepts of lung injury and repair. N Eng! J Med
macrophages, or endothelial cells. Oxygen radicals are 1982; 306:900-09
also recognized as mediators between primary and 3 Rinaldo JE. The prognosis of the adult respiratory distress
secondary effectors, either directly or via cytokine syndrome: inappropriate pessimism? Chest 1986; 90:470-71
release (ie, tumor necrosis fa~tor) with subsequent 4 Artigas A, Carlet J, Chastaing C, Flandre PH, Le Gall JR, and
the European ARDS Collaborative Working Group. Clinical
formation of reactive oxygen. 29 Thus, oxygen-free radi-
presentation, etiology, prognostic factors and outcome of the
cals are important proinflammatmy agents and it seems adult respiratory distress syndrome. Rean Soins Intens Med
desirable to control their activity in states of overshooting t;rg 1988; 4:339-45
inflammation. Endogenous antioxidants and particularly 5 Brigham KL. Mechanisms oflung injury. Clin Chest Med 1982;
glutathione (GSH) normally constitute an effective 3:9-24
6 Repine JE. Neutrophils, oxygen radicals and the adult respira-
control of toxic oxygen. Glutathione is detected in high
tory distress syndrome. In: Said S, ed. The pulmonary circula-
concentrations in the extracellular epithelial lining fluid tion and acute lung injury. New York: Futura Publishing Co,
of the lower respiratory tract of normal subjects30 and 1985; 149-81
could act as a first-time scavenger of toxic oxygen 7 Heffner JE, Repine JE. Pulmonary strategies of antioxydant
intermediates and protect against lung cell damage and defense. Am Rev Respir Dis 1989; 140:531-54
8 Harlan JM, Killen PD, Harker LA, Striker GE. Neutrophil-
injury. 18·31 Patients with severe acute lung injury of the mediated endothelial injury in vitro. J Clin Invest 1981; 68:
ARDS type have a marked decrease of GSH in epithelial 1394-1403
lining fluid, plasma, and blood cells. 1s. 17·20 Glutathione 9 Shasby DM, Vanbenthuysen KM, Tate RM, Shasby SS, Me
depletion in plasma and granulocytes in ARDS is Murtry I, Repine JE. Granulocyte mediate acute edematous
reversible by NAC-mediated glutathione rescue. 20 lung injury in rabbits and in isolated rabbit lungs perfused with
phorbol myristate acetate: role of oxygen radicals. Am Rev
The present trial examines the effects of NAC in Respir Dis 1982; 125:443-47
patients presenting with mild-to-moderate acute lung 10 Tate RM, Vanbenthuysen KM, Shasby DM, Me Murtry IF,
injury. At ICU admission, our two groups were similar Repine JE. Oxygen-radical mediated permeability edema and
for severity of illness and lung injury score. We were vasoconstriction in isolated perfused rabbit lung. Am Rev
unable to find a preventive effect of NAC on the Respir Dis 1982; 126:802-06
11 Johnson KJ, Fantone JC, Kaplan J, Wang PA. In vitro damage
progression to ARDS. Survival had a tendency to be of rat lungs by oxygen metabolites. J Clin Invest 1981; 67:983-
higher in the NAC group. N-acetylcysteine may have 93
positive hemodynamic effects in critically ill patients, 12 Baldwin SR, Simon RH , Grum CM, Ketai LH, Boxer LH,
ie, improvement in cardiac output, systemic oxygen Devall LJ . Oxydant activity in expired breath of patients with
transport, and consumption. 15•32·33 The mechanisms adult respiratory distress syndrome. Lancet 1986; 1:11-4
13 Sznajder JI, Draiman A, Hall JB, Sanders W, Schmidt G,
involved in these changes are not entirely clear. N- Crawford G. Increased hydrogen peroxide in the expired
acetylcysteine possibly restores the decreased activity breath of patients with acute hypoxemic respiratory failure.
of endothelium-derived relaxing factor, 32 •34 thereby Chest 1989; 96:606-12
improving microcirculation. The most interesting find- 14 Cochrane CG, Spragg R, Revak SD. Pathogenesis of the adult
ing of our study is the improvement seen in oxygen- respiratory distress syndrome: evidence of oxidant activity in
bronchoalveolar lavage fluid. J Clin Invest 1983; 71 :754-61
ation index and LIS, which are sensitive markers of the 15 Bernard GR, Swindell BB, Meredith MJ, Carrol FE, Higgins
severity of acute respiratory failure and prognosis.22·35 SB. Glutathione (GSH) repletion by N-acetylcysteine (NAC) in
Fewer patients in the NAC group were receiving patients with the adult respiratory distress syndrome (ARDS).
mechanical ventilatory support at day 2 and 3 and this Am Rev Respir Dis 1989; 139:A221
16 Tagan MC, Markert M, Schaller MD, Feihl F, Chiolero R,
group had a tendency to have a shorter ICU stay
Perret CL. Oxidative metabolism of circulating granulocytes in
(median 7 days vs 10 days for the placebo group). adult respiratory distress syndrome. Am J Med 1991; 9l(suppl
In conclusion, the present study suggests that early 3C):i2-8
intravenous administration of NAC in patients with 1i Pacht ER, Timerman AP, Lykens MG, Merola AJ. Deficiency
mild-to-moderate acute lung injury improves pulmo- of alveolar fluid glutathione in patients with sepsis and the adult
nary gas exchange and may reduce the duration of respiratory distress syndrome. Chest 1991; 100:139i-1403
18 Sagone AL, Husney RM, O'Dorisio MS, Metz EN. Mechanisms
mechanical ventilatory support. Ho~ever, the number for the oxidation of reduced glutathione by stimulated granulo-
of patients included is small and the causes of acute cytes. Blood 1984; 63:96-104
lung injury is heterogeneous. Further and larger trials 19 Repine JE. Scientific perspectives on adult respiratory distress
have to be done to define if this therapy can prevent syndrome. Lancet 1992; 339:466-69
20 Markert M, Schaller MD, Laurent TH, Feihl F, Perret CL.
ARDS and improve outcome.
Oxidant-antioxidant balance in granulocytes during adult respi-
ratory distress syndrome: effect of N-acetylcysteine. Am Rev
ACKNOWLEDGMENTS: We would e~ress our thanks toR. Ruffmann,
M.D., Head of the Medical Division oflnpharzam SA, Zambon Group, Respir Dis 1992; 145:A570
for stimulating discussions and collaboration, and to G. Krejci, Ph.D., 21 Prescott LF, Donovan JW, Jarvie DR, Proudfoot AT. The
Clinical Investigations, Ltd, London, U.K., for the statistical analysis. disposition and kinetics of intravenous N-acetylcysteine in

CHEST I 105 I 1 I JANUARY, 1994 193

 patients with paracetamol overdosage. Eur J Clin Pharmacal Eur J Resplr Dis 1986; 69(suppl147):44-57
1989; 37:501-06 30 Cantin AM, North SL, Hubbard RC, Crystal RG. Normal
22 Murray JF, Matthay NA, Luce JM, Flick MR. An expanded alveolar epithelial lining fluid contains high levels of glu-
definition of the adult respiratory distress syndrome. Am Rev tathione. J Appl Physiol1987; 63:152-57
Respir Dis 1988; 136:720-23 31 Hagan TM, Brown LA, Jones DP: Protection against paraquat-
23 Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico CJ. induced injury by exogenous GSH in pulmonary alveolar type
Clinical predictors of the adult respiratory distress syndrome. II cells. Biochem Pharmacol1986; 35:4537-42
Am J Surg 1982; 144:124-30 32 Harrison PHM, Wendon JA, Crimson AES, Alexander GJM,
24 Bone RC. Let's agree on terminology: definitions of sepsis. Crit Williams R. Improvement by acetylcysteine of haemodynamics
Care Med 1991; i9:973-76 and oxygen transport in fulminant hepatic failure. N Eng! J
25 Ranson JHC, Rifkind KM, Turner JW. Prognostic signs and Med 1991; 324:1852-57
non-operative peritoneal lavage in acute pancreatitis. Surg 33 Keays RT, Gove C, Forbes A, Alexander GJM, Williams R. Use
Gynecol Obstet 1976; 143:209-19 of late N-acetylcysteine in severe paracetamol overdose. Gut
26 Le Gall JR. Loirat PH, Alperovitch AA, Glaser P, Granthil C, 1989; 30:A1512
Mathieu D, et al. Asimplified acute physiology score for ICU 34 Myers PR, Minor RL Jr, Guerra R Jr. Bates JN, Harrison DG.
patients. Crit Care Med 1984; 12:975-77 Vasorelaxant properties of the endothelium-derived relaxing
27 Fowler AA, Hamman RF, Zerbe GO, Banson KN, Hyers TM . factor more closely resemble S-nitrosocystine than nitric oxide.
Adult respiratory distress syndrome: prognosis after onset. Am Nature 1990; 345:161-63 .
Rev Respir Dis 1985; 132:472-78 35 Bone RC, Maunder R, Slotman G, Silverman H, Hyers THM,
28 Zimmerman GA, Renzetti AD, Hill HR. Functional and meta- Kerstein MD, et al. An early test of survival in patients with the
bolic activity of granulocytes from patients with adult respira- adult respiratory distress syndrome. The PaOfFio2 ratio and its
tory distress syndrome. Am Rev Respir Dis 1983; 127:290-300 differential response to conventional therapy. Chest 1989;
29 Raphael GD, Metcalfe DD. Mediators of airway inflammation. 96:849-51

194 Enhanced Rec:Overy From Acute Lung Injury (Suter et s1}