|Year : 2015 | Volume
| Issue : 3 | Page : 294-300
Evaluation and comparison of the three scoring systems at 24 and 48 h of admission for prediction of mortality in an Indian ICU: a prospective cohort study
Mohd Saif Khan1, Pandey Maitree2, Agarwala Radhika2
1 Department of Anaesthesiology, Pondicherry Institute of Medical Sciences, Puducherry, India
2 Department of Anaesthesiology, Lady Hardinge Medical College, New Delhi, India
|Date of Submission||08-Nov-2014|
|Date of Acceptance||29-Apr-2015|
|Date of Web Publication||29-Jul-2015|
Mohd Saif Khan
No. 3-A, D Block, PIMS Staff Quarters, P.I.M.S. Hospital, Kalapet, Puducherry - 605 014
Source of Support: None, Conflict of Interest: None
Predictive accuracy of severity scoring systems in Indian ICUs does not fit well owing to differences in case mix from the west. We hypothesized that 24 h is too early to predict the outcome and that the predictive accuracy of these scores might be better at 48 h of ICU admission.
The aim of this study was to evaluate and compare Acute Physiology and Chronic Health Evaluation (APACHE) II, Simplified Acute Physiology (SAPS) II, and Sequential Organ Failure Assessment (SOFA) as a predictor of outcome in critically ill patients at 24 and at 48 h in a tertiary care hospital ICU.
Patients and methods
A prospective cohort study was conducted on 169 patients over 2 years at a single medical surgical ICU. Eighty-five critically ill patients were included. For each patient, APACHE II, SAPS II, and SOFA scores were calculated at two time frames (first at 24 and then at 48 h) after ICU admission. The accuracies and comparisons of outcome prediction by the three scores were assessed with standardized mortality ratio and area under the receiver operating characteristic (AUROC) curves.
All scoring systems (except SOFA) underestimated the deaths (standardized mortality ratio>1). All scoring systems displayed larger AUROC curves at 48 h than those at 24 h. APACHE II 48 showed the largest AUROC of 0.933.
All scores performed better when calculated at 48 h rather than at 24 h. Overall, APACHE II 48 is the best predictor of 28-day mortality in critically ill patients. A worsening APACHE II score at 48 h after ICU admission may identify the patients at high risk of mortality.
Keywords: 24 h, 28-day mortality, 48 h, APACHE II, Indian ICU, SAPS II, SOFA
|How to cite this article:|
Khan MS, Maitree P, Radhika A. Evaluation and comparison of the three scoring systems at 24 and 48 h of admission for prediction of mortality in an Indian ICU: a prospective cohort study
. Ain-Shams J Anaesthesiol 2015;8:294-300
|How to cite this URL:|
Khan MS, Maitree P, Radhika A. Evaluation and comparison of the three scoring systems at 24 and 48 h of admission for prediction of mortality in an Indian ICU: a prospective cohort study
. Ain-Shams J Anaesthesiol [serial online] 2015 [cited 2022 Oct 5];8:294-300. Available from: http://www.asja.eg.net/text.asp?2015/8/3/294/159003
presented-byThis study was presented in LIVES-2014 hosted by European Society of Intensive Care Medicine at in Barcelona on 30 September 2014.
| Introduction|| |
The ICU is a specialized area in which medical technology and healthcare personnel interact to care for critically ill patients in every major hospital  . The prime goal of ICUs is to provide the highest quality care to achieve the best outcomes for patients  . Prediction of patient outcome is an important component of patient care in ICU. Intensivists are frequently faced with questions regarding outcome and eligibility criteria for ICU admissions. The most sensitive, reliable, and meaningful measurement of outcome for intensive care is hospital mortality ,,, . Other outcome measures that are of great value are morbidity, hours spent on mechanical ventilation, length of stay in ICU, cost of procedures, treatment, and nursing care being delivered to the patient. Outcome depends on the ICU resources (machines, therapeutics, and nursing staff), processes of care (type, skill, and timing of care), and the case mix of the patients  . As a simple subjective evaluation of patients cannot be used to quantify the severity of critical illness, a variety of ICU scoring systems have been proposed to estimate the same on an objective basis  . These scores have been used as a surrogate measure of ICU performance (ICU benchmarking), which helps in resource allocation  . The prioritization of ICU admission on the basis of scoring systems can help optimize the use of limited financial, medical, and human resources in delivering expensive intensive care. This concept of delivering cost-effective intensive care is now gaining popularity in all developed countries and is becoming a major interest of clinicians, hospital administrations, healthcare managers, medical economists, and governmental policy makers  . Scoring systems consist of two parts: a severity score, which is a number (the higher the number the more severe is the condition) and a calculated probability of mortality. Most models such as Acute Physiology and Chronic Health Evaluation (APACHE) II, Simplified Acute Physiology (SAPS) II, and Mortality Prediction Model (MPM) use values taken within the first 24 h of ICU stay (also called as first-day ICU severity scores) and ignore many factors that may influence patient outcome beyond the first 24 h  .
These scoring systems have been developed and evaluated using large population databases in western countries, and the predictive accuracy in Indian ICUs may not fit well because of small study population, lead time bias, and differences in case mix ,, . The APACHE and the SAPS scores are the most widely used scoring systems in the ICU. Limited data are available from India ,,,, . Therefore, we designed a study to evaluate and compare the accuracy of APACHE II, SAPS II, and Sequential Organ Failure Assessment (SOFA) scores for predicting 28-day mortality in our ICU. We hypothesized that 24 h is too early to estimate the mortality and that the predictive accuracy of these scores might be better at 48 h of ICU admission. The main objective of this study was to evaluate and compare APACHE II, SAPS II, and SOFA as a predictor of outcome in critically ill patients at 24 and at 48 h.
| Patients and methods|| |
The study was conducted between November 2009 and March 2011, at a single medical surgical ICU with five beds, at Lady Hardinge Medical College and Smt Sucheta Kriplani Hospital, a tertiary care teaching hospital in New Delhi, India. The ICU is managed and run by a devoted team of anesthesiologists working in day and night shifts. There are eight staff nurses working in day and night shifts. All physiological monitors, mechanical ventilators, and one arterial blood gas (ABG) machine are available in the ICU. The ICU staff were not aware of the study and so there was no possibility of our data affecting patient treatment.
An observational prospective cohort study was conducted after approval from the institutional ethics committee. We prospectively collected data on patients consecutively admitted to the ICU during the study period. Exclusion criteria were as follows: age less than 18 years, ICU stay less than 48 h, acute coronary syndrome, burn, terminal cancer, do not resuscitate (DNR), and end-of-life care orders. Patients operated during the first week before or after ICU admission were identified as surgical patients. Informed consent was obtained from competent patients or their representatives if they were incompetent.
The three severity scoring systems, APACHE II, SAPS II, and SOFA score, were calculated from the physiological, laboratory, and patient characteristics mentioned in the ICU scoring data sheet, at 24 h and subsequently at 48 h.
The APACHE II score varies from zero to 71 points: up to 60 for physiological variables, up to six for age, and up to five for previous health status  . SAPS II model includes 17 variables and varies from zero to 163 points (up to 116 points for physiological variables, up to 17 points for age, and up to 30 points for chronic diagnosis)  . The SOFA score, developed in 1994, is the most commonly used organ dysfunction score. Six organ systems (respiratory, cardiovascular, renal, hepatic, central nervous, coagulation) are taken into account, and the function of each is scored from 0 (normal function) to 4 (most abnormal)  . Serial changes in SOFA score over time are useful in predicting outcome  .
All patients who survived for at least 48 h in the ICU were subsequently followed up until 28 days. For patients who were discharged earlier than 28 days, welfare was maintained through telephonic communication until 28 days. All major events, as well as mortality, within the 28 days' period following admission were recorded. The predictive mortality based on the score used was calculated and compared with the actual outcome to derive the standardized mortality ratio (SMR). The outcomes of the patient were finally classified as survivors and nonsurvivors. Microsoft Access 2007 software database was used for data storage (Microsoft Corporation, Redmond, WA, USA). Length of ICU stay was the duration of care from admission to the discharge from the ICU.
The observations were compiled, tabulated, and analyzed statistically using SPSS, version 18.0 (SPSS Inc., Chicago, Illinois, USA). The Mann-Whitney U-test was used for analysis of continuous variables and the χ2 -test was used for analysis of categorical variables. The SMR was calculated by dividing observed hospital mortality by the predicted hospital mortality. The accuracy of outcome prediction by the APACHE II, SAPS II, and SOFA systems and the comparisons between the three scoring systems for predicting ability was assessed with SMR and area under the receiver operating characteristic (AUROC) curves.
| Results|| |
During the period of study, 85 patients met the inclusion criteria. Demographic data for these patients are summarized in [Table 1]. ICU mortality was 34.1% and 28-day mortality was 40%. In comparison with the survivors, the nonsurvivors were older (P<0.05) and more severely ill (P < 0.001), but did not have longer stays in the ICU (P = 0.149; Mann-Whitney U-test). Obstetric diseases and sepsis were the most common causes of ICU admissions. Respiratory diseases were significantly higher in nonsurvivors than in survivors [Table 2]. The means of all severity scores at both points of time were significantly higher in nonsurvivors compared with survivors [Table 3].
All scoring systems (except SOFA) underestimated the deaths (SMR > 1) [Table 4]. However, APACHE II 24 and APACHE II 48 predicted mortality better among all the scoring systems. The SMR of SOFA could not be calculated because regression formula to calculate predicted death rate is not available. Both APACHE II 48 and SOFA 48 displayed the largest AUROC curve [Table 5]. All scoring systems displayed larger AUROC curves at 48 h than those at 24 h; hence, discrimination also improved at 48 h rather than at 24 h [Figure 1]. A bivariate analysis (Pearson's correlation) showed statistically significant correlation between the three scores at 24 h of ICU admission [Table 6]. These scores also correlated with each other at 48 h of ICU admission [Table 7]. It was observed that correlations were better (i.e. higher coefficients) at 48 h rather than at 24 h of ICU admission.
|Figure 1: AUROC curves of severity scores at 48 and 24 h of ICU admission. APACHE, acute physiology and chronic health evaluation; AUROC, area under the receiver operating characteristic; SAPS, simplified acute physiology score; SOFA, sequential organ failure assessment.|
Click here to view
|Table 4: Standardized mortality ratio of two severity scores at two points of time|
Click here to view
|Table 5: Discrimination, sensitivity, and specifi city of the three severity scores at two points of time|
Click here to view
| Discussion|| |
The three commonly used severity scoring systems, compared in this study, have been developed using large cohorts of critically ill patients admitted to American and European ICUs. Indian hospitals were not included in any of these cohorts; however, selected Indian ICUs contributed in the development of SAPS III model , . Validation is essential before routine application of any predictive model in a group of patients different from the one originally used for model development. In India, performances of these severity models have been tested only on individual basis in a few studies ,,,, . So far, the predictive abilities of these three models have not been compared altogether.
In the present study, the lower mean age of ICU patients (39.1 years) can be explained by the fact that most of the patients belonged to obstetric category (22.3%), and our hospital is tertiary a care referral center for obstetric cases in Delhi and National Capital Region (NCR). In the present study, 28-day mortality was 40%, which is comparable to that reported in previous studies from India (40.3%) and Indonesia (39.8%) , . It is higher compared with those reported in Germany (9%), Australia (16%), and Saudi Arabia (31.6%) ,, . The high mortality can be attributed to shortage of ICU beds in our center (ICU bed: hospital bed ratio is 0.006). It has been documented that physicians tend to be more selective in their ICU admissions during times of bed shortages, with patients having higher severity of illness being admitted  . Critically ill patients usually deteriorate when managed outside ICU while waiting for availability of ICU beds. Poor nutritional status at ICU admission may also be one of the important causes contributing to high mortality rate in India. Poor nutrition leads to decline in immunity and thereby a rise in acquired infection rate. The reason for poor nutrition is that most of the patients, who come to government sector for free treatment, belong to a lower socioeconomic class. Moreover, anemia and infectious diseases (e.g. tuberculosis, malaria, and dengue) are prevalent problems among patients coming to Indian ICU. Therefore, proper daily assessment of nutritional state of each patient and subsequent intervention is imperative.
The percentage of mortality was significantly higher among patients in medical category (61.7%) compared with those in surgical category (38.2%) (P < 0.001). Similar findings were observed in studies conducted by Khwannimit and Geater  , Livingston et al.  , and Arabi et al.  . The high mortality in medical category may be due to the presence of various severe comorbidities (type 2 respiratory failure with cor pulmonale, end-stage renal disease, end-stage liver disease, decompensated congestive cardiac failure, etc.). Longer ICU stay [8.2 (±9.4) days] in the present study could be due to combinations of several direct or indirect factors such as dependency on mechanical ventilation, too delayed weaning from ventilators, ICU-acquired infections, and inadequate nutrition and high glucose variability , .
APACHE II 24 SMR was 1.09, which is higher than those found in various studies , . Similarly, SAPS II 24 SMR was 1.32 in our study, which is higher than those found in studies from other countries ,, . The SMR value expresses two things: first, the performance of ICU, and, second, how well a score is calibrated. The high SMR values (>1) could be due to the underprediction of mortality by APACHE II and SAPS II scores. Several reasons have been proposed. First, as APACHE II and SAPS II award points to only those above 45 years of age  , majority of our ICU patients below cutoff value were left out; this would have resulted in an underprediction of mortality. Second, various communicable diseases such as tuberculosis, malaria, and dengue that have a greater prevalence in this country and are unaccounted in original APACHE II and SAPS II comprise a different case-mix that would contribute to the higher SMR value observed in this part of the world. Medical management before ICU transfer can partly correct physiological derangements without arresting the basic disease process, and such patients are likely to have relatively lower severity scores and mortality probabilities in comparison with the severity of their underlying illness. This phenomenon is labelled as 'lead-time bias' and can partly explain the high mortality rates in patients with relatively lower calculated probability of death  . Some patients were referred from nearby private hospitals and nursing homes in which patients' clinical conditions would have already deteriorated over time because of inadequate treatment or poor quality care in such centers.
This study was undertaken to evaluate the performance of each scoring system on the basis of calibration and discrimination. This study found the APACHE II scoring system to be the best for the prediction of outcome in our ICU. APACHE II has been shown to perform better than SOFA and SAPS II in a few studies ,,,,, . SAPS II could not perform well; it showed poor calibration, although it stood very good in terms of discrimination. Reasons for poor calibration are inherent to model and hence SAPS II needs to be customized before application in Indian population. There are controversies on the relative importance of calibration and discrimination in assessing a model's performance. It is also suggested that discrimination is a meaningless issue in a poorly calibrating model, although not all experts agree on this. Discrimination properties of different severity scoring systems can be compared by statistical comparison of AUROC values; however, no clear guidelines exist for comparing calibration statistics. Although it is logical that models with higher Lemeshow-Hosmer C- (or H-) values are poorer than those with corresponding high values, a numerical expression of the degree of difference that should be considered significant is not described. We have assessed calibration for two scores using SMR. As the SOFA score itself does not give a quantitative estimation of the risk of mortality, calibration and accuracy cannot be assessed for the SOFA score  . Calibration would have been better assessed using Lemeshow-Hosmer statistics if we could get a large sample size. Unfortunately, despite adequate duration of study, we could get only 85 patients meeting inclusion criteria because of the slow turnover of patients and the small capacity of our ICU.
This may be the first study that compares the three different scores at two time frames, first at 24 h and second at 48 h following admission to ICU. This idea is based on the observation that there are various factors or determinants that may probably affect or modify the outcome of a patient who stays in ICU beyond the first 24 h. Some of these time-sensitive factors may be as follows: change of antibiotics, ventilator-induced lung injury, complications of various diagnostic/therapeutic procedures, cardiac arrest and resuscitation, massive blood transfusion, and iatrogenic infections. These factors might develop beyond 24 h of ICU admission and affect the prognostication of a patient and hence the predictive ability of a particular scoring system. With this hypothesis, the scores were compared at 24 and at 48 h, and it was observed that both discrimination power (AUROC) and calibration (SMR) improved over time. Therefore, prediction of mortality improved at 48 h compared with that at 24 h. After a tedious search of EMABSE, MEDLINE, and Opus databases, we could find few studies on the application of scores at two or three time points and their comparison. In one study, the validity of the APACHE II score has been challenged because it does not take into account the medical therapy delivered to the patient or the subsequent course of disease after the first 24 h in the ICU  . Another study conducted in Saudi Arabia comparing the predictive value of APS of APACHE II and SAPS II in 20 consecutive postoperative liver transplant patients on day 1, day 2, and day 3 concluded that day 3 scores better differentiated survivors from nonsurvivors compared with day 1 and day 2 scores and suggested that a mortality prediction system based on day 3 score or that based on serial scores is likely to be more accurate compared with the traditional systems using the first 24 h data  . An additional study, which was conducted retrospectively in 125 acute pancreatitis patients, observed the 48-h APACHE II score as a better predictor of outcome compared with the admission score, but this conclusion was disease specific and hence may not be extrapolated to general ICU patients  . Therefore, it can be suggested that the APACHE II at 48 h score is a very useful risk adjustment tool in ICU. This has resource implications because the APACHE II score only has to be collected once, and is easier to collect compared with daily collection of SOFA score, especially when patients stay in ICU for longer duration.
The present study has some limitations. First, as a single center study, there may be bias due to case mix, ICU care, and admission criteria. Second, the small sample size was a relevant limiting factor in the analysis of calibration of both models. Sample size (too large or too small), heterogeneity (case mix), lead time bias all can affect the validation of severity score. Another limitation was that we have not evaluated newer scores such as SAPS III, MPM III, and APACHE IV. Therefore, results of this study cannot be extrapolated to the heterogeneous population of Indian ICUs. A multicenter study including ICUs of all Indian states is required to validate the scoring systems. Our data demonstrate better performance of scores at 48 h rather than at 24 h of ICU admission. Overall, APACHE II 48 is found to be the best predictor of 28-day mortality and can be used in Indian ICUs for mortality prediction. A worsening APACHE II score at 48 h after admission may identify patients at risk for an adverse outcome.
| Acknowledgements|| |
The authors thank Dr. Homay Vajifdar who inspired the conduct of this project work and Dr. Rajesh Pande who facilitated the presentation in conference.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Shortell SM, Zimmerman JE, Rousseau DM, Gillies RR, Wagner DP, Draper EA, et al
. The performance of intensive care units: does good management make a difference? Med Care 1994; 32:508-525.
Knaus WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE. APACHE - Acute Physiology and Chronic Health Evaluation: a physiologically based classification system. Crit Care Med 1981; 9:591-597.
Le Gall Jr, Loirat P, Alperovitch A, Glaser P, Granthil C, Mathieu D, et al
. A simplified acute physiology score for ICU patients. Crit Care Med 1984; 12:975-977.
Lemeshow S, Teres D, Pastides H, Avrunin JS, Steingrub JS. A method for predicting survival and mortality of ICU patients using objectively derived weights. Crit Care Med 1985; 13:519-525.
Vincent JL, Moreno R, Takala J, et al.
The SOFA (Sepsis related organ failure assessment) score to describe organ dysfunction or failure. On behalf of the working group on sepsis related problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996; 22:707-710.
Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23:1638-1652.
Moreno R, Vincent JL, Matos R, Mendonça A, Cantraine F, Thijs L, et al
. The use of maximum SOFA score to quantify organ dysfunction/failure in intensive care. Results of a prospective, multicentre study. Working Group on Sepsis related Problems of the ESICM. Intensive Care Med 1999; 25:686-696.
Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 2001; 286:1754-1758.
Murphy-Filkins R, Teres D, Lemeshow S, Hosmer DW. Effect of changing patient mix on the performance of an intensive care unit severity-of-illness model: how to distinguish a general from a specialty intensive care unit. Crit Care Med 1996; 24:1968-1973.
Prakash P, Krishna K, Bhatia D. Usefulness of SAPS II scoring system as an early predictor of outcome in ICU patient. JIACM 2006; 7:202-205.
Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13:818-829.
Ratanarat R, Thanakittiwirun M, Vilaichone W, Thongyoo S, Permpikul C. Prediction of mortality by using the standard scoring systems in a medical intensive care unit in Thailand. J Med Assoc Thai 2005; 88:949-955.
Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270:2957-2963.
Apostolopoulou E, Charcharidou M, Karga M, Kaliakmanis A. The relationship of scoring systems and mortality in the intensive care unit. Icus Nurs Web J 2003; 13:1-8.
Khwannimit B, Geater A. A comparison of APACHE II and SAPS II scoring systems in predicting hospital mortality in Thai adult intensive care units. J Med Assoc Thai 2007; 90:643-652.
Marra AR, Bearman GM, Wenzel RP, Edmond MB. Comparison of severity of illness scoring systems for patients with nosocomial bloodstream infection due to Pseudomonas aeruginosa. BMC Infect Dis 2006; 6:132.
Rao MH, Marella P, Kath B. Assessment of severity and outcome of critical illness. IJA 2008; 52:652-662.
Lim SC, Fok AC, Ong YY. Patient outcome and intensive care resource allocation using APACHE II. Singapore Med J 1996; 37:488-491.
Zygun DA, Laupland KB, Fick GH, Sandham JD, Doig CJ. Limited ability of SOFA and MOD scores to discriminate outcome: a prospective evaluation in 1,436 patients. Can J Anaesth 2005; 52:302-308.
Boyd O, Grounds RM. Physiological scoring systems and audit. Lancet 1993; 341:1573-1574.
Lemeshow S, Hosmer DW Jr. A review of goodness of fit statistics for use in the development of logistic regression models. Am J Epidemiol 1982; 115:92-106.
Moreno R, Matos R. The 'new' scores: what problems have been fixed, and what remain? Curr Opin Crit Care 2000; 6:158-165.
Strauss MJ, LoGerfo JP, Yeltatzie JA, Temkin N, Hudson LD. Rationing of intensive care unit services. An everyday occurrence. JAMA 1986; 255:1143-1146.
Livingston BM, MacKirdy FN, Howie JC, Jones R, Norrie JD. Assessment of the performance of five intensive care scoring models within a large Scottish database. Crit Care Med 2000; 28:1820-1827.
Arabi Y, Haddad S, Goraj R, Al-Shimemeri A, Al-Malik S. Assessment of performance of four mortality prediction systems in a Saudi Arabian intensive care unit. Crit Care 2002; 6:166-174.
Suka M, Yoshida K, Takezawa J. Impact of intensive care unit-acquired infection on hospital mortality in Japan: a multicenter cohort study. Environ Health Prev Med 2004; 9:53-57.
Hermanides J, Vriesendorp TM, Bosman RJ, Zandstra DF, Hoekstra JB, Devries JH. Glucose variability is associated with intensive care unit mortality. Crit Care Med 2010; 38:838-842.
Tan IK. APACHE II and SAPS II are poorly calibrated in a Hong Kong intensive care unit. Ann Acad Med Singapore 1998; 27:318-322.
Del Bufalo C, Morelli A, Bassein L, Fasano L, Quarta CC, Pacilli AM, Gunella G. Severity scores in respiratory intensive care: APACHE II predicted mortality better than SAPS II. Respir Care 1995; 40:1042-1047.
Tempe A, Wadhwa L, Gupta S, Bansal S, Satyanarayana L. Prediction of mortality and morbidity by simplified acute physiology score II in obstetric intensive care unit admissions. Indian J Med Sci 2007; 61:179-185.
Ho KM, Lee KY, Williams T, Finn J, Knuiman M, Webb SA. Comparison of Acute Physiology and Chronic Health Evaluation (APACHE) II score with organ failure scores to predict hospital mortality. Anaesthesia 2007; 62:466-473.
Gursel G, Demirtas S. Value of APACHE II, SOFA and CPIS scores in predicting prognosis in patients with ventilator-associated pneumonia. Respiration 2006; 73:503-508.
Khwannimit B, Bhurayanontachai R. The performance of customised APACHE II and SAPS II in predicting mortality of mixed critically ill patients in a Thai medical intensive care unit. Anaesth Intensive Care 2009; 37:784-790.
Minne L, Abu-Hanna A, de Jonge E. Evaluation of SOFA-based models for predicting mortality in the ICU: a systematic review. Crit Care 2008; 12:R161.
Arabi Y, Goraj R, McIntyre R, Al-Abdulkareem A. The predictive value of serial acute physiology score (APS) and simplified acute physiology score II (SAPS II) in post-operative liver transplant patients. Critical Care 2000; 4:240.
Khan AA, Parekh D, Cho Y, Ruiz R, Selby RR, Jabbour N, et al
. Improved prediction of outcome in patients with severe acute pancreatitis by the APACHE II score at 48 hours after hospital admission compared with the APACHE II score at admission. Acute Physiology and Chronic Health Evaluation. Arch Surg 2002; 137:1136-1140.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]