Organ Dysfunction in the ICU: A Clinical Perspective Y. Sakr, C. Sponholz, and K. Reinhart

Y. Sakr, C. Sponholz, and K. Reinhart

Introduction

Multiorgan dysfunction is a major cause of mortality in the intensive care unit (ICU) [1 – 3]. Sequential organ dysfunction syndrome was first described by Tilney et al. [4]

in 1973 in a cohort of 18 patients after repair of ruptured abdominal aortic aneurysm and renal failure. The terms multiple organ failure syndrome (MOFS), multiple organ system failure (MOSF), and multiple organ failure (MOF) have since been used to describe this syndrome [5]. Uncontrolled infections were initially thought to be the main cause of multiorgan dysfunction; however, massive activation of inflammatory mediators following other insults, such as severe trauma, may precipitate a similar condition. In 1992, the American College of Chest Physicians/Society of Critical Med- icine (ACCP/SCCM) consensus conference [6] recommended definitions of sepsis and the proposed systemic inflammatory response syndrome (SIRS). The term multiple organ dysfunction syndrome was also proposed to describe this syndrome; however, firm definitions of organ dysfunction were not established. Several scoring systems have subsequently been developed to quantify organ dysfunction in ICU patients.

In 2001, several North American and European intensive care societies revisited the definitions for sepsis and related conditions [7]. A staging system for sepsis, PIRO, which stratifies patients on the basis of their Predisposing conditions, the nature and extent of the insult (Infection), the nature and magnitude of the host Response, and the degree of Organ dysfunction, was proposed. The use of organ fail- ure scores was encouraged to quantitatively describe organ dysfunction developing over the course of critical illness. The previous definitions of sepsis, severe sepsis, and septic shock were maintained essentially unchanged. Although much debate has focused on the non-specific nature of the SIRS criteria, no clear improvements could be offered. In fact, a number of additional criteria indicative of physiologic derange- ments were added to the four traditional SIRS criteria. These further non-specific alterations consist of clinical abnormalities (altered mental status, ileus) and bio- chemical evidence of involvement (procalcitonin [PCT], C-reactive protein [CRP], creatinine, or cytokines). Standard definitions for organ dysfunction are still lacking and several controversies regarding the best way to quantify organ dysfunction in the ICU remain unresolved [8].

The Epidemiology of Organ Dysfunction in the ICU

Organ dysfunction is common in the ICU [1 – 3]. A large European, multicenter,

observational study, Sepsis Occurrence in Acutely ill Patients (SOAP) [3], reported

that up to 71 % of ICU patients have a considerable degree of organ dysfunction (sequential organ failure assessment [SOFA] score 8 2 for the corresponding organ);

81 % of which was present on ICU admission. Sepsis contributed to 41 % of the reported organ failures. MOF (failure of & 2 organs) occurred more in patients with sepsis (75 vs. 43 %) compared with other ICU patients. The incidence of 2, 3, and &

4 organ failures was higher (38, 24, and 13 vs. 28, 12, and 4 %, respectively), and all forms of organ failure were more common, in patients with sepsis compared with other ICU patients. In contrast, isolated single organ failure occurred more in patients with no sepsis (57 vs. 25 %), with more renal (27 vs. 9 %), respiratory (15 vs.

12 %), and central nervous system (CNS) failure (11 vs. 2 %) compared with patients with sepsis. Patients with no organ dysfunction on admission had ICU mortality rates of 6 % while those with four or more organ failures had mortality rates of 65 %.

Patients with severe sepsis had higher mortality rates (32 vs. 21 %) compared to patients with organ failure without sepsis. Occurring alone or in combination, renal, respiratory, cardiovascular, and hepatic failure were associated with higher ICU mortality rates in patients with severe sepsis (41 vs. 23 %; 35 vs. 26 %; 42 vs. 34 %; 45 vs. 28, respectively) compared to patients with no sepsis.

The results of this multinational study underscore the common occurrence of organ dysfunction/failure in the ICU and its close relation to poor outcome in ICU patients. Sepsis appears to be an important associate of organ dysfunction in the ICU. The incidence of organ dysfunction/failure seems, unfortunately, not to be decreasing overtime, as seen when comparing the results of the SOAP study [3] with those with a similar cohort of ICU patients reported by Vincent et al [9] more than a decade ago.

Quantifying Organ Dysfunction in the ICU

Numerous physiological parameters and therapeutic interventions have been used to define multiple organ dysfunction [5]. The pulmonary, cardiovascular, renal, hepatic, hematologic, and central nervous systems were the organs most commonly considered in describing organ dysfunction/failure in the ICU. But other organs con- sidered in this context included gastrointestinal, metabolic, endocrine, and immuno- logic function. As early as 1980, Fry et al. [10] proposed a system of four organ fail- ures for surgical patients (pulmonary, hepatic, gastrointestinal, and renal failure).

The sepsis severity score proposed by Stevens [11] comprised seven organ systems, each with five severity levels. Knaus et al. [12] defined organ system failure using a dichotomous score for five organ systems (cardiovascular, respiratory, renal, hema- tologic and neurologic failure). Goris et al. [13] proposed a score based on seven organ systems and three categories (0 = normal organ function, 1 = organ dysfunc- tion, and 2 = organ failure). In 1993, Fagon et al. [14] included infection in their organ dysfunction and infection model (ODIN). Other definitions were proposed by Hebert et al. [15], Pine et al. [16], Moore et al. [17] and Bernard et al. [18]

One of the widely used organ dysfunction scores, the multiple organ dysfunction

score (MODS) was proposed in 1995 by Marshall and collaborators [19] as an objec-

tive scale to measure the severity of MOF in critical illness. Descriptors of organ

dysfunction were identified according to a systematic review of the literature and

included PaO

/FiO

ratio for the respiratory system, serum creatinine concentration

for the renal system, serum bilirubin concentration for the liver, platelet count for

the hematologic system, and Glasgow coma scale (GCS) for the CNS. The cardiovas-

cular system was evaluated using the pressure-adjusted heart rate, which is calcu- lated as the product of the heart rate and the ratio of central venous pressure (CVP) to mean arterial pressure (MAP). The relative complexity of the cardiovascular com- ponent may be a drawback and is commonly modified using simple parameters such as heart rate, the use of inotropes, and serum lactate levels [20]. Nevertheless, this score was shown to successfully describe organ dysfunction in the ICU, was corre- lated strongly with the ultimate risk of ICU and hospital mortality rates [19], and reflected organ dysfunction development when measured sequentially in the ICU [19 – 21].

Another commonly used score was developed by a working group of the Euro- pean Society of Intensive Care Medicine [9]: the SOFA score, comprising six organ systems, graded from 0 – 4 according to the degree of dysfunction failure. Organ sys- tems considered in the SOFA score are: respiratory (PaO

/FiO

), cardiovascular (blood pressure, vasoactive drugs), renal (creatinine and diuresis), hematological (platelet count), neurological (GCS) and liver (bilirubin). The reliability and accu- racy of calculating the SOFA score among ICU physicians was shown to be good [22], probably due to its simplicity. It was further validated in various groups of crit- ically ill patients [23 – 26].

Unlike the previous scores, which were established by consensus, the logistic organ dysfunction system (LOD) was developed by Le Gall et al. [27] using logistic regression techniques in a cohort of 13152 adult ICU patients. This score includes physiological variables of six organ systems – GCS, PaO

/FiO

ratio, heart rate, blood pressure, serum urea, creatinine, urine output, white blood cell (WBC) count, biliru- bin, platelet count, and prothrombin. The assignment of points in this score took into account both the relative severity among organ systems and the degree of sever- ity within an organ system. Despite the complexity of this score, its prognostic value was not proven to be superior to the other scoring systems [28].

Oda et al. [29] developed the cellular injury score (CIS) as an index of cellular dysfunction in critically ill patients. CIS is derived from three parameters of intra- cellular metabolism: Arterial ketone body ratio, osmolality gap, and blood lactate.

Each parameter is assigned 0 – 3 points according to arbitrarily defined cut off points. The usefulness of CIS in mortality prediction was reported in 157 patients with MOF [29] and was found to be correlated and comparable to the SOFA score in terms of mortality prediction, in another study by the same authors [30]. This has not been investigated in a large cohort of unselected ICU patients.

Similar scoring systems have been developed and validated in pediatric ICU patients including organ failure index [31] and pediatric logistic organ dysfunction (PELOD) [32] scores, comprising age specific criteria, adapted to the pediatric popu- lation.

The Utility of Organ System Failure Scores

Scoring systems for organ dysfunction/failure have been designed primarily as a

descriptive tool, aimed at establishing standardized definitions to stratify and com-

pare patients’ statuses in the ICU in terms of morbidity rather than mortality. The

LOD is one exception to this, as it was developed using a statistical procedure to

maximize its predictive value in terms of mortality prediction. Accordingly, organ

dysfunction scores can also be used statistically to adjust analyses for baseline char-

acteristics, to control for time dependent changes in matched cohort studies [33], to

define subgroup analyses [34], and to directly compare organ dysfunction between groups as a secondary outcome in randomized trials [35, 36].

Organ dysfunction is a dynamic process; thus, sequential evaluation of organ dys- function during the ICU stay may be helpful in tracing disease progression in the ICU and has been shown to be highly correlated to the subsequent outcome [20, 37].

As early as the 24 hours following ICU admission, changes in organ dysfunction as assessed by the SOFA score were found to predict eventual survival in severe sepsis [38]. The course of organ dysfunction/failure could also be useful in identifying patients who remain unresponsive despite appropriate treatment for several days, where intensive therapy may be considered futile [39]. In addition, as many of the components of the aforementioned scoring systems are readily affected by the vari- ous therapeutic maneuvers, they may be theoretically useful in the determination of therapeutic effects and setting therapeutic targets in the management of critically ill patients.

Indeed, the severity of organ dysfunction is strongly correlated with outcome in critically ill patients [19, 27, 40, 41]. Interventions shown to have improved outcome exhibited similar favorable effects on organ function [42]. Accordingly, organ dys- function scores can be used to predict outcome from critical illness [43]. Because comorbidities and baseline characteristics on ICU admission usually do not contrib- ute to these scoring systems and may play a major role in determining the subse- quent outcome in the ICU, the performance of organ dysfunction scores in terms of outcome prediction may be inferior to that of severity scores such Acute Physiology and Chronic Health Evaluation (APACHE) II [44] and Simplified Acute Physiology Score (SAPS) II [45], that were designed primarily to efficiently predict outcome. It may also be worthy to note that the contribution of each organ dysfunction in asso- ciation with outcome seems not to be equal. The cardiovascular system has persis- tently been shown to have the highest impact followed by the renal, neurological, and respiratory systems [14, 27, 46, 47].

The correlation between organ failure and outcome may justify the use of the for- mer as a surrogate end point for clinical studies. The relatively common occurrence of organ dysfunction in the ICU may permit a decrease in the sample size needed for future interventional studies, if organ dysfunction is used as the primary end- point. However, the Food and Drug Administration (FDA) and the Retirement Medi- cal Benefit Accounts (RMBA) have, until now, not accepted differences in organ dys- function as primary endpoints in sepsis studies.

Which Score is Best?

The aforementioned organ dysfunction scores vary in the strategy of development

(consensus vs. statistical techniques), characteristics of the validation-set, reason for

development (descriptive vs. outcome prediction), data collection (admission values

vs. sequential or summary of values all over the ICU stay), and weights of each

organ system according to the utilized cut-off points. Several studies have evaluated

the comparative prognostic value of the commonly used organ dysfunction scoring

systems (Table 1). According to the current evidence, these scores are quite similar

in terms of outcome prediction.

Table 1. Studies comparing the predictive value of various organ dysfunction scores in terms of mortality prediction.

Author n Setting Evaluated scores Main findings

Oda et al [30]

47 General ICU SOFA and CIS SOFA and CIS similar discriminated poor outcome.

Timsit et al [28]*

1685 General ICU LOD and SOFA LOD and SOFA scores had good accu- racy and internal consistency.

No difference in discrimination between the two scores during the first week in the ICU

Tsai et al [24]

160 Critically ill patients with liver cirrhosis

OSF, SOFA, and Child-Pugh score

OSF and SOFA scores were closely cor- related.

Both OSF and SOFA scores displayed sim- ilarly excellent discriminative power compared to the Child-Pugh score.

Peres Bota et al [47]

949 General ICU APACHE II, MODS, and SOFA

Outcome prediction of the APACHE II score was similar to the initial MODS and SOFA score in all patients and slightly worse in patients with shock.

MODS and SOFA scores were similar in outcome prediction.

Cardiovascular component of SOFA score performed better than that of MODS in outcome prediction.

Petilla et al [48]

520 General ICU APACHE III, SOFA, MODS and LOD

Highest outcome prediction with total maximum scores.

Discriminative power was good and com- parable between all organ dysfunction scores and that of APACHE III.

Hantke et al [26]

874 Surgical ICU APACHE II, MODS, and SOFA

SOFA, MODS and APACHE II scores simi- larly discriminated poor outcome.

* Multicenter study; ICU: Intensive care unit; SOFA: Sequential Organ Failure Assessment; CIS: Cellular Injury Score; LOD: Logistic Organ Dysfunction; MODS: Multiple Organ Dysfunction score; APACHE: Acute Physio- logic and Chronic Health Evaluation; SAPS: Simplified Acute Physiology Score; PELOD: Pediatric Logistic Organ Dysfunction; OSF: Organ System Failure

Several criteria should be taken into consideration when judging the value of any

scoring system in clinical practice. Reliability and validity are important issues that

allow confident use of a scoring system in ICU patients with different case-mixes

and baseline characteristics. Responsiveness is also an important criterion and sig-

nifies the ability of a scoring system to unmask temporal changes in organ dysfunc-

tion if measured sequentially. Dichotomous scores, such as the OSF score, may be

less flexible in detecting such changes during the ICU stay. Simplicity and availabil-

ity of the various components of the scoring system are also crucial. The use of com-

plex criteria, such as with the LOD, without any proven advantage for this complex-

ity, may limit its widespread use in everyday practice. Likewise, using criteria that

may not be available in all ICU patients, as in the cardiovascular component of

MODS, may also be a practical limitation of its use.

Dynamic measures of cellular response to insults such as apoptosis could be con- sidered in the future to describe organ dysfunction at the cellular level [7].This could be particularly useful in developing therapies that target the injurious cellular processes.

Conclusion

Organ dysfunction is common in the ICU and is strongly correlated to outcome.

Sepsis is associated with a significant degree of organ dysfunction and subsequently worse outcome. Despite the variability between various organ dysfunction scoring systems, their prognostic value seems to be similar in terms of mortality prediction.

The presence and use of different scoring systems makes the comparison of different descriptive and mortality predictive studies very difficult. This is of particular importance when these scores are used as entry characteristics, as criteria for strati- fied randomization, or as a surrogate end point for interventional or therapeutic studies in the ICU. Describing organ dysfunction in light of the PIRO system is promising. Further effort is required towards setting standard definitions that could have useful clinical and therapeutic utility.

References

1. Deitch EA (1992) Multiple organ failure. Pathophysiology and potential future therapy. Ann Surg 216:117 – 134

2. de Mendonca A, Vincent JL, Suter PM, et al (2000) Acute renal failure in the ICU: risk factors and outcome evaluated by the SOFA score. Intensive Care Med 26:915 – 921

3. Vincent JL, Sakr Y, Sprung CL, et al (2006) Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 34:344 – 353

4. Tilney NL, Bailey GL, Morgan AP (1973) Sequential system failure after rupture of abdominal aortic aneurysms: an unsolved problem in postoperative care. Ann Surg 178:117 – 122 5. Bertleff MJ, Bruining HA (1997) How should multiple organ dysfunction syndrome be

assessed? A review of the variations in current scoring systems. Eur J Surg 163:405 – 409 6. American College of Chest Physicians/Society of Critical Care Medicine (1992) Consensus

Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 20:864 – 874

7. Levy MM, Fink MP, Marshall JC, et al (2003) 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 31:1250 – 1256

8. Vincent JL, Wendon J, Groeneveld J, Marshall JC, Streat S, Carlet J (2003) The PIRO concept:

O is for organ dysfunction. Crit Care 7:260 – 264

9. Vincent JL, Moreno R, Takala J, et al (1996) The SOFA (Sepsis-related Organ Failure Assess- ment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sep- sis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 22:707 – 710

10. Fry DE, Pearlstein L, Fulton RL, Polk HC Jr (1980) Multiple system organ failure. The role of uncontrolled infection. Arch Surg 115:136 – 140

11. Stevens LE (1983) Gauging the severity of surgical sepsis. Arch Surg 118:1190 – 1192 12. Knaus WA, Draper EA, Wagner DP, Zimmerman JE (1985) Prognosis in acute organ-system

failure. Ann Surg 202:685 – 693

13. Goris RJ, te Boekhorst TP, Nuytinck JK, Gimbrere JS (1985) Multiple-organ failure. General- ized autodestructive inflammation? Arch Surg 120:1109 – 1115

14. Fagon JY, Chastre J, Novara A, Medioni P, Gibert C (1993) Characterization of intensive care unit patients using a model based on the presence or absence of organ dysfunctions and/or infection: the ODIN model. Intensive Care Med 19:137 – 144

15. Hebert PC, Drummond AJ, Singer J, Bernard GR, Russell JA (1993) A simple multiple system

organ failure scoring system predicts mortality of patients who have sepsis syndrome. Chest

104:230 – 235

16. Pine RW, Wertz MJ, Lennard ES, Dellinger EP, Carrico CJ, Minshew BH (1983) Determinants of organ malfunction or death in patients with intra-abdominal sepsis. A discriminant analy- sis. Arch Surg 118:242 – 249

17. Moore FA, Sauaia A, Moore EE, Haenel JB, Burch JM, Lezotte DC (1996) Postinjury multiple organ failure: a bimodal phenomenon. J Trauma 40:501 – 510

18. Bernard GR, Doig BG, Hudson G (1995) Quantification of organ failure for clinical trials and clinical practice. Am J Respir Crit Care Med 151:A323 (abst)

19. Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ (1995) Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 23:1638 – 1652

20. Cook R, Cook D, Tilley J, Lee K, Marshall J (2001) Multiple organ dysfunction: baseline and serial component scores. Crit Care Med 29:2046 – 2050

21. Buckley TA, Gomersall CD, Ramsay SJ (2003) Validation of the multiple organ dysfunction (MOD) score in critically ill medical and surgical patients. Intensive Care Med 29:2216 – 2222 22. Arts DG, de Keizer NF, Vroom MB, de Jonge E (2005) Reliability and accuracy of Sequential

Organ Failure Assessment (SOFA) scoring. Crit Care Med 33:1988 – 1993

23. Ceriani R, Mazzoni M, Bortone F, et al (2003) Application of the sequential organ failure assessment score to cardiac surgical patients. Chest 123:1229 – 1239

24. Tsai MH, Peng YS, Lien JM, et al (2004) Multiple organ system failure in critically ill cirrhotic patients. A comparison of two multiple organ dysfunction/failure scoring systems. Digestion 69:190 – 200

25. Janssens U, Graf C, Graf J, et al (2000) Evaluation of the SOFA score: a single-center experi- ence of a medical intensive care unit in 303 consecutive patients with predominantly cardio- vascular disorders. Sequential Organ Failure Assessment. Intensive Care Med 26:1037 – 1045 26. Hantke M, Holzer K, Thone S, Schmandra T, Hanisch E (2000) [The SOFA score in evaluating

septic illnesses. Correlations with the MOD and APACHE II score]. Chirurg 71:1270 – 1276 27. Le Gall JR, Klar J, Lemeshow S, et al (1996) The Logistic Organ Dysfunction system. A new

way to assess organ dysfunction in the intensive care unit. ICU Scoring Group. JAMA 276:802 – 810

28. Timsit JF, Fosse JP, Troche G, et al (2002) Calibration and discrimination by daily Logistic Organ Dysfunction scoring comparatively with daily Sequential Organ Failure Assessment scoring for predicting hospital mortality in critically ill patients. Crit Care Med 30:2003 – 2013 29. Oda S, Hirasawa H, Sugai T, Shiga H, Matsuda K, Ueno H (1998) Cellular injury score for

multiple organ failure severity scoring system. J Trauma 45:304 – 310

30. Oda S, Hirasawa H, Sugai T, et al (2000) Comparison of Sepsis-related Organ Failure Assess- ment (SOFA) score and CIS (cellular injury score) for scoring of severity for patients with multiple organ dysfunction syndrome (MODS). Intensive Care Med 26:1786 – 1793

31. Doughty LA, Kaplan SS, Carcillo JA (1996) Inflammatory cytokine and nitric oxide responses in pediatric sepsis and organ failure. Crit Care Med 24:1137 – 1143

32. Leteurtre S, Martinot A, Duhamel A, et al (2003) Validation of the paediatric logistic organ dysfunction (PELOD) score: prospective, observational, multicentre study. Lancet 362:192 – 197

33. Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C (1999) The attributable mor- bidity and mortality of ventilator-associated pneumonia in the critically ill patient. The Canadian Critical Trials Group. Am J Respir Crit Care Med 159:1249 – 1256

34. Angus DC, Birmingham MC, Balk RA, et al (2000) E5 murine monoclonal antiendotoxin antibody in gram-negative sepsis: a randomized controlled trial. E5 Study Investigators.

JAMA 283:1723 – 1730

35. Hebert PC, Wells G, Blajchman MA, et al (1999) A multicenter, randomized, controlled clini- cal trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 340:409 – 417 36. Bernard GR, Vincent JL, Laterre PF, et al (2001) Efficacy and safety of recombinant human

activated protein C for severe sepsis. N Engl J Med 344:699 – 709

37. Ferreira FL, Bota DP, Bross A, Melot C, Vincent JL (2001) Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 286:1754 – 1758

38. Levy MM, Macias WL, Vincent JL, et al (2005) Early changes in organ function predict even-

tual survival in severe sepsis. Crit Care Med 33:2194 – 2201

39. Cabre L, Mancebo J, Solsona JF, et al (2005) Multicenter study of the multiple organ dysfunc- tion syndrome in intensive care units: the usefulness of Sequential Organ Failure Assessment scores in decision making. Intensive Care Med 31:927 – 933

40. Moreno R, Miranda DR, Matos R, Fevereiro T (2001) Mortality after discharge from intensive care: the impact of organ system failure and nursing workload use at discharge. Intensive Care Med 27:999 – 1004

41. Vincent JL, de Mendonca A, Cantraine F, et al (1998) Use of the SOFA score to assess the inci- dence of organ dysfunction/failure in intensive care units: results of a multicenter, prospec- tive study. Working group on “sepsis-related problems” of the European Society of Intensive Care Medicine. Crit Care Med 26:1793 – 1800

42. Vincent JL, Angus DC, Artigas A, et al (2003) Effects of drotrecogin alfa (activated) on organ dysfunction in the PROWESS trial. Crit Care Med 31:834 – 840

43. Kajdacsy-Balla Amaral AC, Andrade FM, Moreno R, Artigas A, Cantraine F, Vincent JL (2005) Use of the sequential organ failure assessment score as a severity score. Intensive Care Med 31:243 – 249

44. Knaus WA, Zimmerman JE, Wagner DP, Draper EA, Lawrence DE (1981) APACHE-acute physiology and chronic health evaluation: a physiologically based classification system. Crit Care Med 9:591 – 597

45. Le Gall JR, Lemeshow S, Saulnier F (1993) A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 270:2957 – 2963

46. Vosylius S, Sipylaite J, Ivaskevicius J (2004) Sequential organ failure assessment score as the determinant of outcome for patients with severe sepsis. Croat Med J 45:715 – 720

47. Peres BD, Melot C, Lopes FF, Nguyen B, V, Vincent JL (2002) The Multiple Organ Dysfunction Score (MODS) versus the Sequential Organ Failure Assessment (SOFA) score in outcome pre- diction. Intensive Care Med 28:1619 – 1624

48. Pettila V, Pettila M, Sarna S, Voutilainen P, Takkunen O (2002) Comparison of multiple organ

dysfunction scores in the prediction of hospital mortality in the critically ill. Crit Care Med

30:1705 – 1711