Printer Friendly

Association of obesity with increased mortality in the critically ill patient.


The impact of obesity on critical care outcomes has been an issue for debate in the literature. Variable data and conflicting results have been reported. The purpose of our study is to examine the impact of obesity on the outcome of patients admitted to a tertiary closed Intensive Care Unit (ICU) in Saudi Arabia.

Data was obtained from a prospectively collected database from September 2001 to May 2004. Patients younger than 18, those with burns, brain death and readmissions were excluded. The study population was stratified into six groups according to their Body Mass Index (BMI). Primary endpoints were ICU and hospital mortality, duration of mechanical ventilation and ICU length of stay.

A total of 1835 patients were included in the analysis. Baseline characteristics were similar among the six groups including severity of illness scores, reflected by Acute Physiology and Chronic Health Evaluation II (APACHE II) scores. The ICU mortality was not statistically different among the groups. Hospital mortality was lower in patients with BMI 35-39.9 kg/[m.sup.2] and BMI >40 kg/[m.sup.2] compared to those with BMI 18.5-24.9 kg/[m.sup.2]. Multivariate analysis showed that a BMI >40 kg/[m.sup.2] was an independent predictor of lower hospital mortality (odds ratio 0.51, 95% confidence interval 0.28-0.92, P 0.025) after adjustment for other confounding factors.

In conclusion, mortality of obese critically ill patients was not higher than patients with normal weight. In fact, the hospital mortality was lower for patients with BMI > 40 kg/[m.sup.2] compared to the normal BMI group despite similar severity of illness. Obesity might have a protective effect, although further studies are needed to substantiate this finding.

Key Words: body mass index, critical illness, mortality, obesity, prognostic index

Obesity is a major health problem worldwide. There is evidence that the prevalence of obesity is increasing worldwide both in developed and developing countries (1). A recent study has indicated that the prevalence of overweight and morbid obesity in the United States of America has increased to 64.5% and to 4.9% respectively (2-4). Overweight and obese patients are at increased risk of morbidity and mortality from many acute and chronic conditions, including hypertension, dyslipidaemia, coronary artery disease, diabetes mellitus, gallbladder disease, gout, arthritis and some forms of cancer (5).

Body Mass Index (BMI), calculated as weight (in kilograms)/height (in metres squared) is an anthropometric index that is a reliable measure of adiposity. It is the most convenient method of quantifying the degree of obesity (6,7), because it correlates strongly with body weight, adjusting for height, and has greater reproducibility than other measures of adiposity such as skin fold'. Severity of illness scoring systems have become clinically and scientifically applicable in critical care research. However, BMI has not been evaluated among the list of co-morbid variables used in these scoring systems (9-12). The data about the relation between body weight and mortality of critically ill patients contain some conflicting results. Some studies demonstrated no association, while others showed a J or U shaped relation or in some instances a protective relationship (8,13-16). With the increasing prevalence of obesity in the general population, and the impact of obesity on many illnesses, it would be expected that an increasing portion of patients admitted to ICU would be obese. There is no published data of the impact of BMI on the mortality and outcome during the course of critical care other than in the USA. The purpose of our study was to examine the impact of obesity on the hospital and ICU mortality and length of stay in addition to duration of mechanical ventilation in a tertiary closed ICU in Saudi Arabia, and in particular to discern whether morbid obesity is associated with increased mortality.


Data on all patients admitted to the Intensive Care Unit between September 2001 and May 2004 were included in the analysis from a prospectively collected ICU database. We excluded patients younger than 18, patients with burns or brain death and readmissions. We collected the following data: baseline demographics including gender, age and the type of admission with prespecified admission diagnoses. Acute Physiology and Chronic Health Evaluation II (APACHE II) (9) scores were recorded. Patients were followed until discharge from the hospital or death, whichever occurred first. The bedside nurse measured patients' height using a measuring tape and weight using the ICU weight-built beds. Body Mass Index (BMI) was calculated by dividing the weight (kg) by the square of height (metres). Patients were classified into six groups according to the classification defined by the National Heart, Lung and Blood Institute (NHLBI) (17).

The groups were as follows: Underweight: BMI < 18.5 kg/[m.sup.2], Normal: BMI between 18.5 and 24.9 kg/[m.sup.2], Overweight: BMI between 25 and 29.9 kg/[m.sup.2], Class I Obesity: BMI between 30 and 34.9 kg/[m.sup.2], Class II Obesity: BMI between 35 and 39.9 kg/[m.sup.2] and Class III Obesity (Morbid): BMI more than 40 kg/[m.sup.2]. ICU length of stay was calculated as number of calendar days of ICU stay. Mechanical ventilation duration (MVD) was calculated as number of calendar days on mechanical ventilation.

Continuous data were expressed as mean [+ or -] standard deviation (SD) and compared using the student t-test or analysis of variance (ANOVA) as appropriate. Categorical data are expressed as percentage and compared using the chi-square test. All groups were compared to the normal BMI group (18.5-24.9). Univariate analysis was performed to examine the association with ICU and hospital mortality of patients with BMI >40 kg/[m.sup.2] compared to patients with normal BMI. Variables with significant association were tested by multivariate analysis. Statistical significance was defined as alpha less than 0.05. Statistical analysis was performed using Statistica package (Version 6, Oklahoma, U.S.A.) and Minitab for windows (Release 12.1, Minitab Inc, College State, PA, USA).


During the study period, there were 2370 admissions to the ICU, of which 1835 met the study inclusion criteria. There were a total of 535 (23%) patients who were excluded; 26 patients (5%) due to burns, 71 patients (13%) due to brain death, 330 patients (63%) due to readmission and 108 patients (20%) due to age less than 18y. Patients' demographics are shown in Table 1.

Sixty-two percent were males. Admission APACHE II was 21[+ or -]9. The distribution of patients by BMI was as follows: 631 (34%) had normal BMI (18.5-24.9), 140 (8%) were underweight, 524 (29%) were overweight, 312 (17%) had BMI 30-34.9 kg/[m.sup.2], 135 (7%) had BMI 35-39.9 kg/[m.sup.2] and 93 (5%) had BMI >40 kg/[m.sup.2].

After stratifying patients according to their BMI, univariate analysis was performed.

Table 1 shows baseline characteristics of these patients according to their BMI group.

Age was comparable to the normal BMI group except for patients with BMI 35-39.9 kg/[m.sup.2] who were slightly older at 55 [+ or -] 15y (P=0.044). There was a trend toward higher percentage of female patients with increased weight since all the overweight groups had statistically higher percentages of females than the normal group. Severity of illness on admission reflected by APACHE II scores was similar among the groups.

Table 2 depicts selected reasons for admission among the six groups. There were no significant differences in the selected reasons of the admission with few exceptions: patients with BMI >40 kg[m.sup.2] were more commonly admitted with Acute Respiratory Distress Syndrome (ARDS), medical reasons and had chronic respiratory illness and less commonly with trauma when compared to patients with normal BMI.

Table 3 shows the outcome data for the six groups. Although hospital LOS as well as mechanical ventilation duration were similar, the ICU LOS was two days shorter for patients with BMI 30-34.9 kg/[m.sup.2] (7[+ or -]9, P=0.039). The ICU mortality ranged between 11 to 17% with no statistical difference. This was not the case for the hospital mortality, which showed lower mortality in patients with BMI 35-39.9 kg/[m.sup.2] and >40 kg/[m.sup.2], and was in fact the lowest for patients with BMI >40 kg/[m.sup.2] (21%, P=0.039).

Table 4 showed the results of multivariate analysis examining the impact of BMI >40 kg/[m.sup.2] on the outcome. The following variables were introduced in the model: BMI > 40 kg/[m.sup.2] (versus normal BMI), chronic respiratory illness, gender, age, medical and trauma admission types. We found that BMI >40 kg/[m.sup.2] and having chronic respiratory illness were associated with better survival [odds ratio 0.51 CI (0.28-0.92) P 0.025; odds ratio 0.39 CI (0.21-0.74) P=0.004 respectively]. On the other hand, age and medical type of admission were associated with higher hospital mortality.


Our study demonstrates that the group with BMI >40 kg/[m.sup.2] is not associated with increased mortality in intensive care. In fact, our data suggests possible protective effect of BMI >40 kg/[m.sup.2]. In our study the prevalence of patients with BMI >30 kg/[m.sup.2] in the ICU population is relatively similar to the pattern of obesity observed among the ICU population in the studies previously reported (7,8).

No major heterogeneity was demonstrated even though the groups were not intentionally matched. Despite the fact that the obese group was slightly older and had more females, the six groups were rather similar in regard to severity of illness on admission to ICU, which is reflected by similar APACHE II scores. The group with BMI >40 kg/[m.sup.2] had fewer cases of trauma which in our opinion can be explained by the predominance of female gender in this category who are by convention less subject to motor vehicle accidents than males. The higher percentage of chronic respiratory illnesses in this particular group is probably due to obesity per se which affects the values of pulmonary function testing and is associated with sleep disorder breathing syndromes.

However, the hospital mortality rate was lower for patients with BMI 30-34.9 kg/[m.sup.2] and BMI >40 kg/[m.sup.2] and especially for patients with BMI >40 kg/[m.sup.2] as it was the lowest among the groups. In multivariate analysis, BMI >40 kg/[m.sup.2] was an independent predictor of lower hospital mortality.

What can be the reason for this surprising and unexpected lower rate of mortality among the morbidly obese? Similar findings were shown previously by Marik and coworkers' who studied outcomes of 48176 patients from 101 ICUs participating in project IMPACT and found that the ICU and hospital mortality rates were less in the morbidly obese (BMI>40) patients (10.4% vs 15.3% respectively P<0.001). Similarly, Galanos et al in a prospective, multicentre study of 4103 critically ill patients found lower mortality rates for the group of BMI>30'$. Our study and many previous reports suggest that obesity may be actually protective. Proponents of this notion base their beliefs on two theories. Both theories are built on the adipose tissue functions. The first theory claims that the excessive adipose tissue in the morbidly obese patients serves as an energy store that is badly needed during the critical care state, which combines low nutritional supply with high catabolic state'. We believe that this theory alone is unlikely to explain the protective role of obesity since many of the morbidly obese are subject to the catabolic state during critical illnesses. The second theory explores the role of many hormones produced by the adipocytes, mainly leptin (8). It has been noticed that leptin increases with increasing body fat mass. Leptin impacts energy expenditure by its effect on the thyroid axis, growth hormone, neuroendocrine apparatus and autonomic system (19). These effects may contribute to the survival benefit attached to obesity in critical illnesses.

However, there are other studies showing contrary results. A BMI higher than 30 was found as an independent risk factor for ICU mortality in the report published by Bercault et al (20) where the mortality rates for the obese were almost double the rates for those considered normal (32 vs 17%, P 0.01). This study was a matched cohort and was performed only on ventilated patients in one medical ICU.

El-solh (21) demonstrated higher mortality rates for the morbidly obese (BMI >40) than the non-obese (BMI <30 kg/[m.sup.2) (30% vs 17%,P=0.019). Mechanical ventilation duration and ICU length of stay were longer for the morbidly obese group. However, this study was retrospective, the two groups were not matched and no severity of illness scoring data was reported.

Flegal and colleagues analysed the data from the National and Nutrition Examination Surveys (NHANS) conducted in three phases (1971-1975, 1976-1980 and 1988-1994) and estimated the relative risk of mortality associated with different levels of BMI. The investigators found that subjects with BMI < 18.5 kg/[m.sup.2] and BMI > 30 kg/[m.sup.2] have higher mortality in comparison to those with BMI of 18.5 to 25 kg/[m.sup.2]. Furthermore, the study found that the impact of obesity on mortality declined over time. The authors attributed this observation to improvement in medical care and public health. In our study severity of illness was evenly matched across the groups. The severity of illness scores utilized in this study have been validated across a wide population of critically ill patients and represent an estimated mortality risk that is independent of the primary diagnosis (23,24].

To our knowledge this is the first study that examines the relationship between BMI and mortality in an ICU outside U.S.A. Our study has several strengths. Our data was collected prospectively by a fulltime dedicated data collector. Our unit is closed and is operated mainly by Critical Care Board certified intensivists, which makes the management rather homogenous. On the other hand our study has some limitations. The accuracy of the weight and the height of the critically ill patients may be a concern 25. Errors in measurements may have an impact on the results. Our ICU beds are weight-built so the chance of errors is minimal. The bedside nurse may have occasionally estimated the heights although we have emphasised strict instructions to measure the height of all patients on admission to the ICU. Despite these concerns, we believe that the large sample size is likely to have minimized the impact of potential systematic errors.

We also did not account for all potential factors that may play a role in morbidity and mortality such as the use of non-invasive ventilation, which may have been used more in the morbidly obese group. Our database did not include the number of patients who received non-invasive ventilation. Despite all the limitations, we think that our data further ignites the debate about the relation between obesity and clinical outcomes and may pave the way for more studies to understand a potentially protective factor for obesity amongst ICU patients. We believe that a larger study with better matching of co-morbid conditions and treatment protocols may give us more insight in to the real impact of obesity on clinically relevant outcomes.


(1.) Adams JP, Murphy PG. Obesity in anesthesia and intensive care. Br J Anaesth 2000; 85:91-108.

(2.) Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adult, 1999-2000. JAMA 2002; 288:1723-1727.

(3.) Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM. Prevalence of overweight and obesity among USA children, adolescents, and adult, 1999-2002. JAMA 2004; 291:2847-2850.

(4.) Centers for Disease Control and Prevention. Update: Prevalence of overweight among children, adolescent and adult-United States 1988-1994. MMWR Morb Mortal Wkly Rep 1997; 46:198-202.

(5.) Pi-Sunyer FX. Health implications of obesity. Am J Clin Nutr 1991; 53 (6 Suppl):1595s-1603s.

(6.) Marik P, Varon J. The obese patient in the ICU. Chest 1998; 113:492-498.

(7.) Tremblay A, Bandi V Impact of body mass index on outcomes following critical care. Chest 2003; 123:1202-1207.

(8.) Marik P, Doyle H, Varon J and the project impact contributors. Is obesity protective during critical illness? An analysis of a national ICU database. Crit Care Shock 2003; 6:156-162.

(9.) Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985;13:818-829.

(10.) Knaus WA, Wagner DP, Draper EA et al. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991; 100:1619-1636.

(11.) Lemeshow S, Teres D, Mar J, Avrunin JS, Gehlbach SH, Rapoport J. Mortality probability models (MPM II) based on an international cohort of intensive care unit patients. JAMA 1993;270:2478-2489.

(12.) Le Gall JR, Lemeshow S, Saulinier E A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270:2957-2963.

(13.) Manson J, Willett W, Stampfer M et al. Body weight and mortality among women. N Engl J Med 1995; 333:677-685.

(14.) Vandenbroucke JP, Mauritz BJ, de Baruin A, Veheesen JHH, Van der Heide-Wessel C, Van der Heiesw RM. Weight, smoking and mortality. JAMA 1984; 252:2859-2860.

(15.) Stevens J, Keil JE, Rust PF, Tyroler HA, Davis CE, Gazes PC. Body mass index and body girth as predictor mortality in black and white women. Arch Intern Med 1992; 152:1257-1262.

(16.) Lee IM, Manson JE, Hennekens CH, Paffenbarger RS Jr. Body weight and mortality: 27-year follow up of middle-age men. JAMA 1993; 270:2823-2828.

(17.) Clinical Guidelines on the Identification, Evaluation and treatment of overweight and obesity in adults-The Evidence Report. National Institutes of Health. Obes Res 1998; Suppl 2:51S-209S.

(18.) Galanos A, Pieper C, Kussin P et al. Relationship of body mass index to subsequent mortality among seriously ill hospitalized patients. Crit Care Med 1997; 25:1962-1968.

(19.) Mantzoros C. The role of leptin in human obesity and disease: A review of current evidence. Ann Intern Med 1999; 130:671680.

(20.) Bercault N, Boulain T, Kuteifan K, Wolf M, Runge I, Fleury J. Obesity-related excess mortality rate in an adult intensive care unit: A risk-adjusted matched cohort study. Crit Care Med 2004;32:998-1003.

(21.) El-solh A, Sikka P, Bozkanat E, Jaafar W, Davies J. Morbid obesity in medical ICU. Chest 2001; 120:1989-1997

(22.) Flegal KM, Graubard BI, Williamson DF, Gail MIT Excess death associated with underweight, overweight, and obesity. JAMA 2005;293:1861-1867.

(23.) Arabi Y, Haddad S, Goraj R, Al-Shimemeri A, Al-Marik S. Assessment of performance of four mortality prediction systems in a Saudi Arabian intensive care unit. Crit Care 2002; 6:166-174.

(24.) Arabi Y, Al Shirawi N, Memish Z, Venkatesh S, Al-Shimemeri A. Assessment of six mortality prediction models in patients admitted with severe sepsis and septic shock to the intensive care unit: a prospective cohort study. Crit Care 2003; 7:R116122.

(25.) Leary TS, Milner QJ, Niblett DJ. The accuracy of the estimation of body weight height in the intensive care unit. Eur J Anaesthesiol 2000; 17:698-703.

A. ALDAWOOD *, Y. ARABI [dagger], O. DABBAGH [double dagger] Department of Intensive Care, King Abdulaziz Medical City, Riyadh, Kingdom of Saudi Arabia

* M.D., F.R.C.P.C., F.C.C.P., Assistant Professor of Medicine, King Saud bin Abdulaziz University for Medical Sciences and Consultant in Pulmonary and Critical Care, Programme Director of Critical Care Fellowship.

[dagger] M.D., F.C.C.P., Assistant Professor of Medicine, King Saud bin Abdulaziz University for Medical Sciences and Consultant in Pulmonary and Critical Care, Deputy Chairman of Intensive Care Department.

[double dagger] M.D., F.C.C.P., Consultant in Pulmonary and Critical Care.

Address for reprints: Dr A. Aldawood, PO Box 241149, Riyadh 11322, Saudi Arabia.

Accepted for publication on July 4, 2006.
Demographic data according to BMI Groups

BMI <18.5 18.5-24.9 25-29.9

n 140 631 524
(%) (8) (34) (29)

Age 48[+ or -] 52[+ or -] 53[+ or -]
 24 21 18

Female n 44 171 181
(%) (31) (27) (35) (b)

APACHE II 22[+ or -] 21[+ or -] 21[+ or -]
 9 9 9

BMI 30-34.9 35-39.9 >40

n 312 135 93
(%) (17) (7) (5)

Age 54.00 55[+ or -] 53[+ or -]
 [+ or -]18 15(c) 17

Female n 167 94 67
(%) (54) (a) (70) (a) (72) (a)

APACHE II 21[+ or -] 23[+ or -] 22[+ or -]
 9 10 9

BMI: Body Mass Index, n: number, APACHE II: Acute Physiology
and Chronic Health Evaluation II.
(a) P<0.001, (b) P=0.006, (c) P=0.044.

Reason for admission according to BMI groups *

BMI <18.5 18.5-24.9 25-29.9 30-34.9

n 140 631 524 312
(%) (8) (34) (29) (17)

Sepsis 8 44 30 22
 (6) (7) (6) (7)

ARDS 20 77 62 39
 (14) (12) (12) (13)

Postoperative 47 257 218 118
 (34) (41) (42) (38)

ICH 14 46 24 17
 (10) (7) (5) (5)

Trauma 25 141 103 55
 (18) (21) (20) (18)

Medical 88 338 294 182
 (63) (54) (56) (58)

CRI 17 44 16 14
 (12) (d) (7) (3) (4)

BMI 35-39.9 40

n 135 93
(%) (7) (5)

Sepsis 10 3
 (7) (3)

ARDS 21 18
 (16) (19) (a)

Postoperative 55 36
 (41) (39)

ICH 6 10
 (4) (11)

Trauma 12 8
 (9) (9) (b)

Medical 76 62
 (56) (67) (c)

CRI 14 14
 (10) (15) (e)

BMI: body mass index, n: number, ARDS: Acute Respiratory
Distress Syndrome, ICH: intracranial hemorrhage, CRI: chronic
respiratory illness. * P value is not significant if not reported.
(a) P=0.056, (b) P=0.002, (c) P=0.017, (d) P=0.043, (e) P=0.007.

Outcome data according to BMI groups

BMI <18.5 18.5-24.9 25-29.9

n 140 631 524
(%) (8) (34) (29)

ICU LOS (d) 9[+ or -]15 8[+ or -] 8[+ or -]
 12 11

Hosp. LOS (d) 41.[+ or -] 42[+ or -] 44[+ or -]
 49 52 54

MVD (d) * 10[+ or -] 9[+ or -] 8[+ or -]
 15 12 12

ICU Mortality 22 109 91
n (%) (16) (17) (17)

Hosp. Mortality 41 197 156
n (%) (30) (32) (30)

BMI 30-34.9 35-39.9 40

n 312 135 93
(%) (17) (7) (5)

ICU LOS (d) 7[+ or -] 7[+ or -] 10[+ or -]
 91 9 18

Hosp. LOS (d) 36[+ or -] 41[+ or -] 39[+ or -]
 49 47 46

MVD (d) * 81[+ or -] 8.[+ or -] 11[+ or -]
 10 10 21

ICU Mortality 52 23 11
n (%) (17) (17) (12)

Hosp. Mortality 79 36 19
n (%) (26) (b) (27) (21) (a)

n: number, LOS: length of stay, MVD: mechanical ventilation
duration, d: days, (a) P=0.039, (b) P=0.044, * Mean calculated for
only mechanically ventilated patients (1186 patients 64.6%).

Result of multivariate analysis of independent predictors of hospital
mortality *

Variable Odds ratio P value
 (95% confidence

Morbid obesity 0.51 (0.28-0.92) 0.025
Chronic respiratory illness 0.39 (0.21-0.74) 0.004
Male gender 1.11 (0.74-1.65) 0.616
Age (in year) 1.03 (1.02-1.05) -0.001
Medical admission 5.14 (3.14-8.41) -0.001
Trauma 1.33 (0.61-2.89) 0.476

* Morbid obesity versus normal body mass index.
COPYRIGHT 2006 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Aldawood, A.; Arabi, Y.; Dabbagh, O.
Publication:Anaesthesia and Intensive Care
Date:Oct 1, 2006
Previous Article:Perfect Medical Presentations--Creating Effective PowerPoint Presentations for the Healthcare Professional.
Next Article:Severe dehydration and acute renal failure associated with external ventricular drainage of cerebrospinal fluid in children.

Related Articles
Many colorectal cancer patients present to ED.
Hyperglycemia is linked to worse ICU mortality in specific disorders.
Obese patients are at increased risk for trauma complications.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters