ICU Outcome Scoring Systems
ICU Outcome Scoring Systems
ICU Outcome Scoring Systems
b.
c.
d.
e.
Requirements for a Good Scoring System a. b. simple, reliable, easily obtainable wide patient applicability - different diagnoses - all age groups - all levels / types of ICU's - ie. should be a good discriminator
c. d. e. f. g. h.
high sensitivity/specificity independent of treatment physiological parameters optimal time is unclear number of criteria is unclear
Potential Problems NB: should not, i. ii. iii. iv. limit treatment of individuals result in nihilistic therapy outweigh clinical judgement depersonalise therapy
anatomical - eg. Injury Severity Score score 0-5 for each anatomical area involved final score sum of 3 highest squared useful for trauma audits & research therapeutic - eg. Therapeutic Intervention Scoring System, TISS sum of weighted scores of 70 therapeutic interventions correlates well with outcome wide applicability limited by available facilities, illness severity, staff enthusiasm and experience physiological eg. "acute physiology and chronic health evaluation", APACHE designed for quality review rather than prognosis extensive, 33 variables & difficult to use clinically simplifications SAPS (13 var) & APACHE II APACHE II = 12 var + age + previous health correlates with hospital mortality limited by subjective scoring
2.
3.
point score based on 12 physiological variables variable selection & weighting was based on 'expert' physician determination
APACHE II
+4 Temperature MAP HR RR Oxygenation1 pH Na K
+ +
+3
39-40.9 130-159 140-179 35-49 350-499 7.6-7.69 160-179 6.6-6.9
+2
+1
38.5-38.9 36-38.4 70-109 70-109 25-34 12-24 < 200 PaO2 > 70 7.5-7.59 7.33-7.49 130-149 3.5-5.4
+1
34-35.9
+2
32-33.9 50-69 55-69
+3
30-31.9
+4
29.9 49
110-129 110-139
40-54
39 5
10-11
6-9
200-349
155-159
150-154 5.5-5.9
50-59.9 20-39.9
46-49.9 15-19.9
30-45.9 3-14.9
20-29.9 1-2.9
< 20 <1
5815 ICU patients from multicentre study, a. b. c. increased score correlated with mortality high specificity (> 98%) but low sensitivity (< 30%) correct prediction only ~ 80%
Chang
Anaesthesia 1987
prospective study of APACHE II for 12 months in Saudi Arabia, 210 patients increased score associated with increased mortality, a. b. sensitivity < 42% specificity > 95%
prospective multicentre study of 5677 ICU patients comparing number of organ systems failing (OSF) and mortality where the organ systems were, 1. 2. 3. 4. 5. CVS respiratory renal CNS haematological
No. of OSF Age < 65 day 1 1 2 3 16% 46% 76% day 7 30% 55% 100%
Mortality Age > 65 day 1 32% 61% 85% day 7 60% 90% 100%
predictive value limited not stratified for disease types nor previous health indicative of mortality during study period only
multiple regression analysis of 775 adult medical & surgical ICU patients measured condition and treatment variables on admission and at 24 hours univariate significant factors a. significant factors on admission i. age ii. BP & HR iii. number of OSF iv. presence of infection v. if CPR used vi. conscious level vii. elective or emergency viii. type of admission (p < 0.001) significant factors at 24 hours = above plus, i. mechanical ventilation ii. number of "lines" iii. use of vasoactive drugs iv. high levels of PEEP v. oliguria vi. use of a S-G catheter vii. PaO2 , FIO2 , arterial pH, creatinine viii. infection ix. shock x. type of admission (p < 0.001)
b.
previously developed weighted scoring system, MPM, mortality predictive model tested on ~ 2000 patients with some modifications admission score better than 24 hour score in mortality prediction with the number of OSF added to criteria, a. b. Chang sensitivity ~ 50% specificity ~ 95% CCM 1988
once only APACHE II versus daily APACHE II in 212 patients later more accurate predictor of mortality, a. b. c. sensitivity ~ 42% specificity ~ 97% prediction ~ 83%
used empirically derived numeric severity index, developed from 220 surgical patients, intensively monitored perioperatively recommends many new "normal ranges" for many vital signs etc. in post-operative patients limited by, a. b. c. d. small study group all elective patients unusual statistical "fiddling" involves invasive monitoring
Nicholas ICM 1985 multicentre study, 792 ICU admissions to 8 institutions, a. b. c. d. e. mortality increased with age < 45 yrs ~ 15% > 65 yrs ~ 37%
mortality similar for all elderly age groups, > 55 yrs there is little increase in mortality from age difference alone high APS score associated with high mortality in all age groups, ie. less difference between ages increased age associated with increased treatment (TISS score) increased age not associated with longer duration in ICU
Jacobs ICM 1988 followed 313 ICU patients, looking at survival, a. b. c. at discharge at 6 months at 12 months ~ 76% ~ 61% ~ 58%
of those discharged, a. b. c. 21% 2% 76% developed decreased physical status and function improved from time of discharge unchanged from time of discharge
NB: health status prior to admission best indicator of "quality of life" after discharge from ICU; age and SAPS scores showed less correlation
if patient has one of more of the following during a 24 hour period, regardless of other values, OSF existed on that day: 1. CVS i. ii. iii. iv. MAP HR VF or VT AGA pH 7.24 with PaCO2 49 mmHg *presence of one or more of the following 49 mmHg 54 bpm
2.
Respiratory *presence of one or more of the following i. RR < 5/min or 49/min. ii. PaCO2 50 mmHg iii. AaDO2 350 mmHg iv. ventilator dependence on the fourth day of OSF ie. not applicable for the initial 72 hr of OSF Renal excluding patients on chronic dialysis before hospital admission presence of one or more of the following, i. urine output 479 ml/24 hr or 158 ml/8 hr ii. urea 36 mmol/l iii. creatinine 270 mol/l Haematological failure presence of one or more of the following, i. WBC 1 x 109/l ii. platelet count 20 x 109/l iii. haematocrit 20%. Neurological failure GCS 6 *in absence of sedation at any point of the day if intubated, use clinical judgement for verbal responses as follows, i. patient unresponsive 1 ii. patient's ability to converse in question 3 iii. patient appears able to converse 5
3.
4.
5.
Respiratory PaO2 / FIO2 without reference to mode of mechanical ventilation and use or level of PEEP Renal creatinine mol/l without reference to use of dialysis Hepatic bilirubin mol/l Pressure adjusted HR PAR = HR RAP/MAP normal, 2.5 to 10 beats/min record the three component variables simultaneously Haematological Neurologic platelet count (platelets/m 10-3) Glasgow Coma Score
5. 6. Scored a. b.
once per day in the morning 0900; missing values entered as normal as worst value, if such is available, after the approach of Knaus et al
this will allow a comparison between the approach of Marshall et al We chose to record these variables at a constant time point (usually the first morning values) to minimize aritifactual variability and to reflect a model of organ dysfunction as a sustained rather than a transient, process. The effect of other approaches merits study as the score evolves.
10
Systemic Inflammatory Response Syndrome : a characteristic clinical response, manifested by two or more of the following, 1. 2. temperature WCC > 38C < 36C > 12,000 < 4,000 > 10% > 90 > 150 > 160 > 20 > 50 > 60 (rectal) /mm3 /mm3 immature band forms adults children infants adults or PaCO2 < 32 mmHg children infants
3.
tachycardia
4.
tachypnoea
Def'n: Sepsis :
Severe SIRS / Severe Sepsis : SIRS / sepsis with associated organ dysfunction, hypoperfusion, or hypotension SIRS with Shock / Septic Shock : SIRS / sepsis with associated organ dysfunction or hypoperfusion, with hypotension not responsive to fluid resuscitation Multiple Organ Dysfunction Syndrome : state characterised by physiologic derrangements in which organ function is not capable of maintaining homeostasis Def'n: Hypotension in the absence of other causes for hypotension, 1. systolic blood pressure < 90 mmHg, or 2. a reduction from baseline > 40 mmHg Def'n: Hypoperfusion and perfusion abnormalities may include, but are not limited to, 1. 2. 3. lactic acidosis oliguria an acute alteration in mental status
patients who are on inotropes / vasopressors need not be hypotensive to fulfill criteria in paediatrics, hypotension is not necessary for diagnosis, as it is a late & ominous sign
11
incidence of sepsis syndrome (US) ~ 176 per 100,000 this has increased 140% from 1979 to 1987
Disease Severity Classification not expected to survive more than 1 year not expected to survive more than 5 years 5 year survival not affected by underlying disease
12
the number of conditions is restricted to nine, 1. 2. smoking habit active smoking of 10 cigarettes/day with 10 UPY-unit pack years alcoholism regular intake > 80g of alcohol per day history of alcoholic pancreatitis or hepatitis portal hypertension and varices cirrhosis at surgery or by biopsy non-cured malignancy diabetes mellitus requiring treatment splenectomy before ICU admission major surgery within 2 months prior to admission previous antibiotic therapy within 2 months prior to admission and for at least 2 months previous cardiogenic shock or cardiopulmonary resuscitation before admission to the ICU Levels
3. 4. 5. 6. 7. 8.
anxious and agitated, or restless, or both cooperative, oriented and tranquil respond to commands only asleep but brisk response to glabellar tap or loud auditory stimulus asleep, sluggish response to glabellar tap or loud auditory stimulus no response
13
ROC area ~ 0.8 SAPSII introduced by LeGall & Lemeshow, JAMA 1993, 17 variables a. b. ROC correlation SAPS / SAPSII ~ 0.86 ~ 0.79
Min 15 2 0.002 1 0 0
(p < 0.05)
14
b.
c.
Generalised Requirements a. b. c. 1-2% of hospital beds patient area ~ 18.2 m2, larger for isolation rooms electrical i. "cardiac protected" electrical area ii. emergency power for ventilators, lighting and equipment iii. 16 power outlets - 8 on emergency circuit required "outlets" per bed, i. 3 oxygen, 2 air, 4 suction ii. 1 non-splash hand washing basin central nurses station with all beds visible adequate storage area adequate equipment with backups and facility for service staff offices, relative waiting area(s) staffing - medical - nursing - physiotherapy - administrative - operating theatres - imaging - laboratories - accident & emergency
d.
e. f. g. h. i.
j.
facility access
15