- Email: [email protected]
Obesity Increases Risk of Organ Failure after Severe Trauma
Obesity Increases Risk of Organ Failure after Severe Trauma David J Ciesla, MD, Ernest E Moore, MD, FACS, Jeffery L Johnson, MD, FACS, Jon M Burch, MD, FACS, C Clay Cothren, MD, FACS, Angela Sauaia, MD, PhD Obesity is an independent risk factor for a variety of diseases, including postinjury morbidity and mortality. Obesity is associated with a proinflammatory state that could affect the postinjury inflammatory response and increase risk of organ dysfunction. The purpose of this study was to determine the relationship between obesity and postinjury multiple organ failure (MOF). STUDY DESIGN: A prospective observational study of patients at risk for postinjury MOF. Inclusion criteria were age older than 15 years, Injury Severity Score ⬎ 15, ICU admission within 24 hours of injury, and survival longer than 48 hours after injury. Isolated head injuries were excluded. Organ dysfunction was assessed using the Denver multiple organ failure score. RESULTS: Data were collected on 716 severely injured patients, 70% were men and 83% were victims of blunt trauma. There was no relationship between body mass index and injury severity or the amount of blood transfused within 12 hours of injury. Postinjury MOF was observed in 123 of 564 (22%) nonobese patients and 56 of 152 (37%) obese patients. Obesity was independently associated with MOF (odds ratio, 1.8; 95% CI, 1.2–2.7) after adjusting for patient age, injury severity, and amount of blood transfused during resuscitation. In this study population, obesity was also associated with increased length of ICU and hospital stay but not death. CONCLUSIONS: Obese patients are at increased risk of postinjury MOF. Study of the obesity-related inflammatory profile could provide additional insight into the pathogenesis of organ dysfunction and identify therapeutic targets for both obese and nonobese patients. Increased morbidity and length of stay in obese trauma patients implies greater resource allocation for this population. (J Am Coll Surg 2006;203:539–545. © 2006 by the American College of Surgeons) BACKGROUND:
The majority of Americans are overweight or obese.1 The consequences of obesity include increased risk of cardiovascular disease, pulmonary dysfunction, diabetes, several types of cancer, fatty liver disease, chronic renal failure, and metabolic syndrome. Obese patients are also at increased risk of death compared with nonobese patients.2 In addition to the mechanical stresses associated with increased body mass, obesity alters nor-
mal metabolic and physiologic pathways. Increased adiposity is associated with an overall proinflammatory state that is thought to be responsible for insulin resistance, endothelial cell dysfunction, atherogenesis, and, ultimately, arteriosclerosis.3 This suggests that the obese patient has a potentially altered immune state that can impact other aspects of the inflammatory response. The realization that obesity is associated with an altered inflammatory profile has important implications in the pathogenesis of postinjury organ dysfunction. The current pathogenic model proposes that organ failure is a result of unbridled postinjury hyperinflammation.4 Obesity could be associated with an altered host inflammatory potential and response to injury and ischemia-reperfusion. Adipose tissue is an active endocrine organ that secretes a variety of inflammatory mediators, including tumor necrosis factor-␣, leptin, and interleukin-6.3 Circulating levels of such inflammatory mediators are elevated in obese humans and animals and
Competing Interests Declared: None. Supported in part by NIH grants P50GM49222, T32GM08315, U546M62119, Jourdan Block Trauma Foundation. Presented at the 36th Meeting of the Western Trauma Association, Big Sky, MT, February 2006. Received March 27, 2006; Revised June 26, 2006; Accepted June 28, 2006. From the Department of Surgery, Washington Hospital Center, Washington, DC (Ciesla); and the Department of Surgery, Denver Health Medical Center, and the University of Colorado Health Sciences Center, Denver, CO (Moore, Johnson, Burch, Cothren, Sauaia). Correspondence address: David J Ciesla, MD, Department of Surgery, Washington Hospital Center, 110 Irving St NW Suite 4B-39, Washington, DC 20005. email: [email protected]
© 2006 by the American College of Surgeons Published by Elsevier Inc.
539
ISSN 1072-7515/06/$32.00 doi:10.1016/j.jamcollsurg.2006.06.029
540
Ciesla et al
Obesity and Organ Failure after Trauma
J Am Coll Surg
Table 1. Obesity Classification Abbreviations and Acronyms
BMI ISS MOF SICU
⫽ ⫽ ⫽ ⫽
body mass index Injury Severity Score multiple organ failure surgical ICU
are thought to create a proinflammatory environment that might be responsible for increased atherogenesis. Additionally, impaired glucose metabolism and insulin resistance associated with obesity alters normal neutrophil function and increases the potential for infection.5 These differences in the inflammatory potential suggest a rich area of clinical and basic science investigation that could uncover therapeutic targets to control the postinjury inflammatory response and give insight into other comorbidities associated with obesity. Severe injury provokes a generalized inflammatory response through direct tissue damage, ischemiareperfusion, and release of systemic inflammatory mediators. Extreme cases of postinjury hyperinflammation can lead to secondary tissue damage and overt organ dysfunction through the action of proinflammatory cytokines and indiscriminate tissue destruction by activated neutrophils. Although undefined, it is plausible that the postinjury inflammatory response is amplified in obese patients. Ultimately, this could manifest as an increase in postinjury complications, such as ARDS and multiple organ failure (MOF). A link between obesity and postinjury immune-mediated complications could give valuable insight into inflammatory pathophysiology and identify potential therapeutic targets. In this study, we sought to define the relationship between obesity and postinjury organ dysfunction. We hypothesized that obesity is associated with an increase in postinjury MOF. METHODS Trauma patients admitted to the Rocky Mountain Regional Trauma Center’s surgical ICU (SICU) at Denver Health Medical Center were studied prospectively. Denver Health Medical Center is a state-designated Level I trauma center verified by the American College of Surgeons Committee on Trauma. Inclusion criteria were Injury Severity Score (ISS) ⬎ 15, survival longer than 48 hours from injury, admission to the SICU within 24 hours of injury, and age older than 15 years. Patients
Obesity class
Underweight Normal Overweight Obese I Obese II Morbidly obese
BMI
⬍ 18.5 18.5–24.9 25–29.9 30–34.9 35–39.9 ⱖ 40
BMI, body mass index (calculated as kg/m2).
with isolated head injuries and head injuries with an extracranial Abbreviated Injury Score ⬍ 2 were excluded. We have maintained a prospective clinical database to study patients at risk for postinjury MOF since 1992. This database is supported by a National Institutes of Health Trauma Center Grant as part of the project’s human subjects’ core. Patient height and weight were added to the dataset in 1998 and were recorded at the time of hospital admission. Body mass index (BMI) was calculated as weight (kg)/height (m2). Patients without recorded height and weight entered before 1998 were excluded from this study. Patients were classified using the system adopted by the National Institutes of Health and the World Health Organization (Table 1). Patients were considered to be nonobese if the BMI was ⬍ 30 and obese if the BMI was ⱖ 30. Daily physiologic and laboratory data were collected through SICU day 28 and clinical events were recorded for all patients thereafter until death or hospital discharge. Data collection and storage processes are in compliance with Health Insurance Portability and Accountability Act regulations and have been approved by our Institutional Review Board. Organ dysfunction is defined using the Denver MOF Scale.6 In brief, four organ systems (pulmonary, hepatic, renal, and cardiac) are evaluated daily throughout the patient’s ICU stay and organ dysfunction is graded on a scale from 0 to 3 (Table 2). The values that determine the division points for respiratory dysfunction have been adjusted for altitude by multiplication of the value by the ratio of atmospheric pressure in Denver to that at sea level (630 mmHg/760 mmHg). MOF score is calculated as the sum of the simultaneously obtained individual organ scores on each hospital day. Single organ dysfunction is defined as an organ failure grade ⬎ 0. MOF is defined as a daily MOF score ⬎ 3 occurring more than 48 hours postinjury.7 Statistical analyses were performed using SAS for Windows version 8.0 (SAS Institute). Categorical vari-
Vol. 203, No. 4, October 2006
Ciesla et al
Obesity and Organ Failure after Trauma
541
Table 2. Denver Multiple Organ Failure Scale Dysfunction*
A Pulmonary† P/F ratio B Renal Creatinine (mg/dL) C Hepatic Total bilirubin (mg/dL) D Cardiac‡
Grade 0
⬎ 250 ⬍ 1.8 ⱕ 2.0 No inotropes
Grade 1
Grade 2
(250, 200)
(200, 100)
(1.8, 2.5)
(2.5, 5.0)
(2.0, 4.0) Minimal inotropes
(4.0, 8.0) Moderate inotropes
Grade 3
ⱕ 100 ⬎ 5.0 ⬎ 8.0 High inotropes
*Multiple Organ Failure Scale score ⫽ A ⫹ B ⫹ C ⫹ D, not due to chronic disease measured daily. † Pulmonary score is based on PaO2-to- FiO2 (P/F) ratio. ‡ Cardiac score: example—inotropes: minimal ⫽ dopamine ⬍ 5 g/kg/min; moderate inotropes ⫽ dopamine 5–15 g/kg/min; high inotropes ⫽ dopamine ⬎15 g /kg/min.
ables were analyzed using chi-square test with Yates’ correction for continuity or Fisher’s exact test when expected cell values were ⬍ 5. For continuous variables with normal distribution, ANOVA, or Student’s t-tests (with the appropriate Welch modification when the assumption of equal variances did not hold) were used. Spearman’s rank correlation was used for comparison of ordinal categorical values. Multivariable analyses were performed using logistic regression for categorical outcomes variables and standard linear regression for continuous outcomes variables. Model goodness-of-fit was assessed using the c statistic. Data are represented as mean ⫾ SE unless otherwise noted. A p value ⬍ 0.05 was considered significant. In summary, we found that obesity is a strong independent risk factor for postinjury organ dysfunction and MOF. Although we have no biochemical data on these patients at this time, future study will include collecting patient samples for inflammatory profiling as part of our NIH-funded Trauma Research Center grant (P50GM49222). The present findings, in conjunction with evidence for an altered inflammatory potential in obese patients, might point to therapeutic targets to improve outcomes in both obese and nonobese patients. Finally, the shift toward a more obese population could affect use of trauma and critical care resources because of the higher risk of postinjury MOF. RESULTS Data were collected on 716 severely injured patients over a 6.5-year period ending in June 2004. The majority (n ⫽ 508; 71%) were men, with a mean (⫾SD) age of 38.6 (⫾17) years and mean BMI (⫾SD) of 26.6 (⫾5.1). Blunt, penetrating, and mixed mechanisms accounted for 572 (80%), 73 (10%), and 71 (10%) injuries, respec-
tively, with an overall mean ISS (⫾SD) of 31.0 (⫾11.6). We did not detect a difference in the prevalence of obesity-related comorbidities before injury. The BMI distribution in the study group is shown in Figure 1. Two hundred eighty-six patients (40%) were normal or underweight (BMI ⬍ 25), 278 patients (39%) were overweight (BMI 25 to 29.9), and 152 patients (21%) were obese (BMI ⱖ 30). The mean BMI and proportion of obese patients did not change over the study period (p ⫽ 0.43 and p ⫽ 0.48, respectively). The mean BMI and proportion of obese patients increased considerably as a function of patient age in models using age as a continuous variable and models using age grouped as in Figure 2. The peak BMI of 29.2 ⫾ 0.8 (43% BMI ⬎ 30) occurring in the 55- to 65-year-old age group (p ⬍ 0.001, Fig. 2). Mean ISS and proportion of severely injured patients (ISS ⬎ 25) did not vary as a function of BMI in this study population (p ⫽ 0.26 and p ⫽ 0.41). We next examined the effect of obesity on the use of blood transfusion during the resuscitation period, which was defined as the first 12 hours after injury. Because blood transfusion is influenced by injury severity, the need for urgent or emergent operation, and patient age, we constructed models to adjust for these factors. Linear models demonstrated no substantial relationship between the number of RBC units transfused during resuscitation and BMI after adjusting for age, injury severity, and need for emergent operation. Results were similar when the independent variables were converted to categorical variables (BMI ⱖ 30, age older than 55 years, and ISS ⬎ 25). Our previous work identified transfusion of ⬎ 6 U RBC during resuscitation as a strong independent risk factor for subsequent development of postinjury MOF.8 Using logistic regression
Obesity and Organ Failure after Trauma
Ciesla et al
45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
J Am Coll Surg
%Patients BMI > 30
% of Patien ts
542
<18.5
18.5-24.9
25-29.9
30-34.9
35-39.9
≥40
50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
<25
25.1-35
BMI (kg/m2)
35.1-45 45.1-55 Age (yrs)
55.1-65
≥65
Figure 1. Distribution of body mass index (BMI) in study population.
Figure 2. Prevalence of obesity according to patient age. BMI, body mass index.
models, we found that obese patients were not more likely to receive ⬎ 6 U blood during resuscitation than nonobese patients after adjusting for advanced age (age older than 55 years), severe injury (ISS ⬎ 25), and need for urgent or emergent operation (Table 3). Results were similar in models that used BMI, age, and ISS as continuous independent variables. The incidence of individual organ dysfunction in obese and nonobese patients is shown in Table 4. Obese patients were observed to have a higher unadjusted incidence of lung, heart, liver, and kidney dysfunction and an increased incidence of ARDS in univariate analyses. But statistical significance was observed in logistic models adjusting for age, ISS, and RBC transfused during resuscitation in only heart and liver dysfunction (although a nonsignificant trend was observed in lung dysfunction [p ⫽ 0.057] and ARDS [p ⫽ 0.073]). Incidence of postinjury MOF in patients stratified by weight is shown in Figure 3. Postinjury MOF was observed in 123 of 564 (22%) nonobese patients and 56 of 152 (37%) obese patients. Logistic models demonstrated that postinjury MOF is 1.81 (95% CI, 1.20– 2.71) times more likely to develop in obese patients than nonobese patients after adjusting for advanced age, se-
vere injury, and transfusion of ⬎ 6 U blood during resuscitation (Table 5). Advanced age, severe injury, and transfusion of ⬎ 6 U blood during resuscitation were also notably associated with postinjury MOF in agreement with earlier studies.9 Logistic models using continuous variables for age, ISS, and units of blood transfused showed similar results. MOF developed within 72 hours of injury (early MOF) in 60 of 126 (48%) nonobese patients and 41 of 56 (73%) obese patients (p ⫽ 0.002). Of the patients who met MOF criteria, the spectrum of organ failure did not differ between obese and nonobese patients in both univariate and multivariable analyses (Table 6). Hospital and SICU length of stay increased with BMI (Fig. 4). Overall hospital length of stay was 20.1 ⫾ 1.6 days and 25.2 ⫾ 1.4 (p ⫽ 0.02) days for nonobese and obese patients, respectively. Similarly, SICU length of stay was 16.1 ⫾ 0.6 and 21.3 ⫾ 1.4 days (p ⬍ 0.001) for nonobese and obese patients, respectively. Of the patients that survived, 320 (61%) of 522 nonobese patients and 99 (74%) of 134 obese patients were transferred to a longterm acute-care facility on discharge (p ⫽ 0.009), the remainder were discharged to home.
Table 3. Association of Body Mass Index, Age, Injury Severity, and Operation Within 48 Hours With Blood Transfusion During Resuscitation RBC > 6 U*
Intercept BMI ⱖ 30 Age (y) ⬎ 55 ISS ⬎ 25 OR48
Odds ratio (95% CI)
1.35 (0.86–2.115) 0.82 (0.48–1.43) 2.84 (1.73–4.67) 7.62 (4.61–12.58) c ⫽ 0.75
p Value
0.188 0.50 ⬍ 0.01 ⬍ 0.01
*Outcomes variable is transfusion of ⬎ 6 U blood within 12 h of injury. BMI, body mass index (calculated as kg/m2); ISS, injury severity score; OR48, operation within 48 h of injury.
Table 4. Incidence of Organ Dysfunction in Obese and Nonobese Patients Organ dysfunction
n Lung (%) ARDS (%) Heart (%) Liver (%) Kidney (%)
Incidence p Value BMI BMI c < 30 > 30 Univariate Multivariate* Statistic
564 78 29 26 24 9
162 89 41 41 38 15
0.003 0.006 ⬍ 0.001 0.001 0.042
0.057 0.073 0.024 0.025 0.435
0.79 0.64 0.69 0.70 0.72
*Multivariate adjusted for age, injury severity, and blood transfusion during resuscitation. BMI, body mass index (calculated as kg/m2).
Vol. 203, No. 4, October 2006
Ciesla et al
Days (m ean±S E )
% o f Pat ient s w it h MO F
35% 30% 25% 20% 15% 10%
25.0 20.0 15.0 10.0
5%
<25
0% <25
25-29.9 BMI (kg/m 2)
Fifty-seven (8%) patients died. Unadjusted mortality rates for obese and nonobese patients were 7% and 11%, respectively, which were not different by univariate analysis (p ⫽ 0.14). There was no difference in mortality between obese and nonobese patients after adjusting for age, injury severity, blood transfused during resuscitation, and postinjury MOF (Table 7). Inclusion of the interaction between obesity and MOF did not improve the logistic model. DISCUSSION The increasing obesity rate threatens to alter current medical and surgical practice.10 Obese patients are more likely to have substantial medical comorbidities and are at increased risk of postsurgical and posttraumatic complications. It is not hard to postulate that the “obesity epidemic” has the potential to alter current practice patterns to the same degree as the “ageing of America.” Although the population of trauma victims does not necessarily mirror the American population as a whole, the general trend toward increased obesity across age groups and gender can be expected to shift the populaTable 5. Influence of Obesity, Age, Injury Severity, and Transfusion of Blood During Resuscitation on the Incidence of Postinjury Multiple Organ Failure Odds ratio (95% CI)
1.81 (1.21–2.71) 2.05 (1.32–3.18) 1.78 (1.16–2.76) 2.41 (1.63–3.56) c ⫽ 0.67
25-29.9
≥30
BM I (kg/m2)
≥30
Figure 3. Percentage with postinjury multiple organ failure (MOF) according to body mass index (BMI).
Intercept BMI ⱖ 30 Age (y) ⬎ 55 ISS ⬎ 25 RBC ⬎ 6*
543
30.0
40%
MOF
Obesity and Organ Failure after Trauma
p Value
0.004 0.001 0.009 ⬍ 0.001
Outcomes variable is the presence of postinjury MOF. *Transfusion of ⬎ 6 U blood within 12 h of injury. BMI, body mass index (calculated as kg/m2); ISS, Injury Severity Score; MOF, multiple organ failure.
Figure 4. Mean ⫾ SE intensive care unit (white) and hospital (black) lengths of stay in nonobese and obese patients. BMI, body mass index.
tion of injured patients toward a more obese demographic. An understanding of obesity-related complications is vital to those involved in trauma care. The prevalence of obesity in this study population was 21%, which is consistent with the overall prevalence of obesity in Colorado but is somewhat lower than the national average of 31%.1 We did not find a substantial increase in the proportion of obese patients in our population during the study period. Although this is in contrast to national trends, it is possible that our sample size could not detect small changes in the prevalence of obesity over this time period. The historic increase in obesity in Colorado has lagged behind other states and national trends. In agreement with others, we found that the prevalence of obesity increased with patient age, and the highest proportion of obese patients were found in the 55- to 65-year-old age group.11 We found no relationship between BMI and injury severity. Although this is consistent with recent reports, others have found different injury patterns in obese patients compared with lean patients. The group from the Los Angeles County Medical Center found that obese patients had a higher frequency of chest injuries and lower extremity fractures, but a lower frequency of head injuries.12 Arbabi and colleagues13 also found a higher Table 6. Incidence of Organ Failure after Postinjury Multiple Organ Failure Developed Incidence (%) Organ system
Lung ARDS Heart Liver Kidney
BMI < 30
BMI > 30
100 64 72 67 29
100 63 82 71 34
BMI, body mass index (calculated as kg/m2).
544
Ciesla et al
Obesity and Organ Failure after Trauma
Table 7. Influence of Obesity, Age, Injury Severity, Transfusion of Blood During Resuscitation, and Multiple Organ Failure on Postinjury Mortality Death
Intercept BMI ⱖ 30 Age ISS RBC MOF
Odds ratio (95% CI)
0.79 (0.39–1.58) 1.05 (1.03–1.07) 1.04 (1.01–1.06) 1.03 (1.00–1.06) 7.66 (3.86–15.20) c ⫽ 0.87
p Value
0.499 ⬍ 0.001 0.002 0.070 ⬍ 0.001
Outcomes variable is death within 28 d of injury. Age, ISS, and units of blood transfused within 12 h of injury (RBC) were treated as continuous variables. BMI, body mass index (calculated as kg/m2); ISS, Injury Severity Score; MOF, multiple organ failure.
frequency of lower extremity fractures, but a lower frequency of abdominal injuries, invoking the “cushion effect” of the obese abdomen that protects the abdominal viscera from injury. Conversely, we found no difference in injury severity or injury pattern between obese and nonobese patients.14 These disparate results make any conclusions about obesity and injury pattern somewhat tentative. Transfusion of stored RBC during resuscitation is independently associated with posttraumatic complications and organ dysfunction.8 The amount of blood transfused during the resuscitation period is influenced by patient age, injury severity, and need for urgent or emergent operation. In the present study, we found no relationship between RBC transfusions during resuscitation after adjusting for these factors. This is important for allocating blood bank resources and for evaluating the effect of obesity on posttraumatic complications independent of the effects of blood transfusion. Previous reports have identified obese patients to be at risk for postinjury organ dysfunction. The Los Angeles County group found that obese patients had a higher rate of ARDS, acute renal failure requiring dialysis, and myocardial infarction.12 Byrnes and colleagues14 also found a higher rate of renal failure, but no difference in respiratory failure or cardiovascular complications in obese patients compared with nonobese patients. These studies did not control for other risk factors for postinjury organ dysfunction, such as patient age, injury severity, and blood transfusion. In the present study, obese patients were found to have higher unadjusted rates of lung, heart, liver, and kidney dysfunction. Multiple logistic regression models demonstrated that obesity was independently associated with only heart and liver dys-
J Am Coll Surg
function, although there was a trend toward increased lung dysfunction and ARDS that did not reach statistical significance. When multiple systems were evaluated, we found that postinjury MOF was 1.8 times more likely to develop in obese patients than nonobese patients after adjusting for the previously established risk factors. Obesity was found to confer an increased risk of postinjury MOF greater than that of severe injury. MOF was more likely to develop within 72 hours of injury in obese patients than nonobese patients. Because early MOF is thought to reflect the effects of unbridled postinjury hyperinflammation as opposed to uncontrolled infection, this observation supports the proinflammatory nature of the obese state. Alternatively, early organ dysfunction could represent decreased physiologic reserve of these patients. Of those in whom MOF developed, there were no differences in the spectrum of involved organs between obese and nonobese patients. The lung was the most frequent organ involved, followed by the heart, liver, and kidney. It is interesting that once the threshold for MOF is reached, obese patients display the same organ dysfunction pattern as nonobese patients. This might be evidence that the final common pathway to postinjury organ dysfunction is not different between obese and nonobese patients. We conclude from these observations and previous research that obesity is associated with a proinflammatory state that influences the postinjury inflammatory response and resultant organ dysfunction. The association between obesity and several chronic diseases is well-known. Obese patients are also more likely to require inpatient and outpatient medical resources.15 In the present study, we found that obese patients were more likely to have longer SICU and hospital lengths of stay and more likely to be transferred to a longterm acute-care facility than to be discharged to home. Increased use of medical resources impacts not only health care costs, but also the allocation of often scarce resources to the injured population.16 Increased length of stay and increased use of post acute-care facilities have the potential to require more care in acute-care hospitals and strain bed availability at the trauma center. Although some authors have reported that obese patients are at increased risk of postinjury death, the issue remains largely controversial. The Los Angeles County group found that obese victims of blunt trauma admitted to the ICU were 5.6 times more likely to die than nonobese patients in multivariable analyses of 242 pa-
Vol. 203, No. 4, October 2006
tients.17 Head injury was the most frequent cause of death. A followup study of 1,153 patients found that obese patients were only 1.6 times more likely to die than nonobese patients using a different multivariable analysis.12 This same group also reported that there was no increased risk of death in 316 obese children, suggesting that the observed effect can be influenced by patient age.18 Others have also observed increased mortality in obese patients stratified by injury severity.13,14,19 We found no relationship between obesity and mortality after adjusting for age, injury severity, blood transfusion during resuscitation, and development of postinjury MOF. Our study population was limited to severely injured patients (ISS ⬎ 15) admitted to the SICU and included both blunt and penetrating mechanisms. Perhaps more importantly, we excluded patients who died within 48 hours of injury and patients with isolated head injuries and head injuries with minor extremity injuries from study. It is possible that a different result would have been observed had these patients been included. The lack of consistent results to date leaves the relationship between obesity and postinjury mortality undefined. Future study using a large inclusive trauma database, such as the National Trauma Data Bank, which includes height and weight data, might ultimately provide an answer. In summary, we found that obesity is a strong independent risk factor for postinjury organ dysfunction and MOF. These findings, in conjunction with evidence for an altered inflammatory potential in obese patients, might point to therapeutic targets to improve outcomes in both obese and nonobese patients. Finally, the shift toward a more obese population could affect use of trauma and critical care resources because of the higher risk of postinjury MOF. Author Contributions
Study conception and design: Ciesla, Moore, Johnson, Sauaia Acquisition of data: Ciesla, Moore, Johnson Analysis and interpretation of data: Ciesla, Moore, Johnson, Burch, Cothren, Sauaia Drafting of manuscript: Ciesla, Moore, Johnson, Sauaia
Ciesla et al
Obesity and Organ Failure after Trauma
545
Critical revision: Ciesla, Moore, Johnson, Burch, Cothren, Sauaia REFERENCES 1. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA 2003;289:76–79. 2. Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005;293:1861–1867. 3. Lyon CJ, Law RE, Hsueh WA. Minireview: adiposity, inflammation, and atherogenesis. Endocrinology 2003;144:2195– 2200. 4. Moore EE, Moore FA. Postinjury multiple organ failure. In: Moore EE, Fleciano DV, Mattox KL, eds. Trauma. 5th ed. New York: McGraw-Hill; 2004:1397–1423. 5. Calvet HM, Yoshikawa TT. Infections in diabetes. Infect Dis Clin North Am 2001;15:407–421, viii. 6. Ciesla DJ, Moore EE, Johnson JL, et al. A 12-year prospective study of postinjury multiple organ failure: has anything changed? Arch Surg 2005;140:432–438; discussion 438–440. 7. Ciesla DJ, Moore EE, Johnson JL, et al. Multiple organ dysfunction during resuscitation is not postinjury multiple organ failure. Arch Surg 2004;139:590–594; discussion 594–595. 8. Moore FA, Moore EE, Sauaia A. Blood transfusion. An independent risk factor for postinjury multiple organ failure. Arch Surg 1997;132:620–624; discussion 624–625. 9. Sauaia A, Moore FA, Moore EE, et al. Early predictors of postinjury multiple organ failure. Arch Surg 1994;129:39–45. 10. Quesenberry CP Jr, Caan B, Jacobson A. Obesity, health services use, and health care costs among members of a health maintenance organization. Arch Intern Med 1998;158:466–472. 11. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA 2002;288:1723–1727. 12. Brown CV, Neville AL, Rhee P, et al. The impact of obesity on the outcomes of 1,153 critically injured blunt trauma patients. J Trauma 2005;59:1048–1051; discussion 1051. 13. Arbabi S, Wahl WL, Hemmila MR, et al. The cushion effect. J Trauma 2003;54:1090–1093. 14. Byrnes MC, McDaniel MD, Moore MB, et al. The effect of obesity on outcomes among injured patients. J Trauma 2005; 58:232–237. 15. Bertakis KD, Azari R. Obesity and the use of health care services. Obes Res 2005;13:372–379. 16. Buntin MB, Garten AD, Paddock S, et al. How much is postacute care use affected by its availability? Health Serv Res 2005; 40:413–434. 17. Neville AL, Brown CV, Weng J, et al. Obesity is an independent risk factor of mortality in severely injured blunt trauma patients. Arch Surg 2004;139:983–987. 18. Brown CV, Neville AL, Salim A, et al. The impact of obesity on severely injured children and adolescents. J Pediatr Surg 2006; 41:88–91; discussion 88–91. 19. Choban PS, Weireter LJ Jr, Maynes C. Obesity and increased mortality in blunt trauma. J Trauma 1991;31:1253–1257.
The majority of Americans are overweight or obese.1 The consequences of obesity include increased risk of cardiovascular disease, pulmonary dysfunction, diabetes, several types of cancer, fatty liver disease, chronic renal failure, and metabolic syndrome. Obese patients are also at increased risk of death compared with nonobese patients.2 In addition to the mechanical stresses associated with increased body mass, obesity alters nor-
mal metabolic and physiologic pathways. Increased adiposity is associated with an overall proinflammatory state that is thought to be responsible for insulin resistance, endothelial cell dysfunction, atherogenesis, and, ultimately, arteriosclerosis.3 This suggests that the obese patient has a potentially altered immune state that can impact other aspects of the inflammatory response. The realization that obesity is associated with an altered inflammatory profile has important implications in the pathogenesis of postinjury organ dysfunction. The current pathogenic model proposes that organ failure is a result of unbridled postinjury hyperinflammation.4 Obesity could be associated with an altered host inflammatory potential and response to injury and ischemia-reperfusion. Adipose tissue is an active endocrine organ that secretes a variety of inflammatory mediators, including tumor necrosis factor-␣, leptin, and interleukin-6.3 Circulating levels of such inflammatory mediators are elevated in obese humans and animals and
Competing Interests Declared: None. Supported in part by NIH grants P50GM49222, T32GM08315, U546M62119, Jourdan Block Trauma Foundation. Presented at the 36th Meeting of the Western Trauma Association, Big Sky, MT, February 2006. Received March 27, 2006; Revised June 26, 2006; Accepted June 28, 2006. From the Department of Surgery, Washington Hospital Center, Washington, DC (Ciesla); and the Department of Surgery, Denver Health Medical Center, and the University of Colorado Health Sciences Center, Denver, CO (Moore, Johnson, Burch, Cothren, Sauaia). Correspondence address: David J Ciesla, MD, Department of Surgery, Washington Hospital Center, 110 Irving St NW Suite 4B-39, Washington, DC 20005. email: [email protected]
© 2006 by the American College of Surgeons Published by Elsevier Inc.
539
ISSN 1072-7515/06/$32.00 doi:10.1016/j.jamcollsurg.2006.06.029
540
Ciesla et al
Obesity and Organ Failure after Trauma
J Am Coll Surg
Table 1. Obesity Classification Abbreviations and Acronyms
BMI ISS MOF SICU
⫽ ⫽ ⫽ ⫽
body mass index Injury Severity Score multiple organ failure surgical ICU
are thought to create a proinflammatory environment that might be responsible for increased atherogenesis. Additionally, impaired glucose metabolism and insulin resistance associated with obesity alters normal neutrophil function and increases the potential for infection.5 These differences in the inflammatory potential suggest a rich area of clinical and basic science investigation that could uncover therapeutic targets to control the postinjury inflammatory response and give insight into other comorbidities associated with obesity. Severe injury provokes a generalized inflammatory response through direct tissue damage, ischemiareperfusion, and release of systemic inflammatory mediators. Extreme cases of postinjury hyperinflammation can lead to secondary tissue damage and overt organ dysfunction through the action of proinflammatory cytokines and indiscriminate tissue destruction by activated neutrophils. Although undefined, it is plausible that the postinjury inflammatory response is amplified in obese patients. Ultimately, this could manifest as an increase in postinjury complications, such as ARDS and multiple organ failure (MOF). A link between obesity and postinjury immune-mediated complications could give valuable insight into inflammatory pathophysiology and identify potential therapeutic targets. In this study, we sought to define the relationship between obesity and postinjury organ dysfunction. We hypothesized that obesity is associated with an increase in postinjury MOF. METHODS Trauma patients admitted to the Rocky Mountain Regional Trauma Center’s surgical ICU (SICU) at Denver Health Medical Center were studied prospectively. Denver Health Medical Center is a state-designated Level I trauma center verified by the American College of Surgeons Committee on Trauma. Inclusion criteria were Injury Severity Score (ISS) ⬎ 15, survival longer than 48 hours from injury, admission to the SICU within 24 hours of injury, and age older than 15 years. Patients
Obesity class
Underweight Normal Overweight Obese I Obese II Morbidly obese
BMI
⬍ 18.5 18.5–24.9 25–29.9 30–34.9 35–39.9 ⱖ 40
BMI, body mass index (calculated as kg/m2).
with isolated head injuries and head injuries with an extracranial Abbreviated Injury Score ⬍ 2 were excluded. We have maintained a prospective clinical database to study patients at risk for postinjury MOF since 1992. This database is supported by a National Institutes of Health Trauma Center Grant as part of the project’s human subjects’ core. Patient height and weight were added to the dataset in 1998 and were recorded at the time of hospital admission. Body mass index (BMI) was calculated as weight (kg)/height (m2). Patients without recorded height and weight entered before 1998 were excluded from this study. Patients were classified using the system adopted by the National Institutes of Health and the World Health Organization (Table 1). Patients were considered to be nonobese if the BMI was ⬍ 30 and obese if the BMI was ⱖ 30. Daily physiologic and laboratory data were collected through SICU day 28 and clinical events were recorded for all patients thereafter until death or hospital discharge. Data collection and storage processes are in compliance with Health Insurance Portability and Accountability Act regulations and have been approved by our Institutional Review Board. Organ dysfunction is defined using the Denver MOF Scale.6 In brief, four organ systems (pulmonary, hepatic, renal, and cardiac) are evaluated daily throughout the patient’s ICU stay and organ dysfunction is graded on a scale from 0 to 3 (Table 2). The values that determine the division points for respiratory dysfunction have been adjusted for altitude by multiplication of the value by the ratio of atmospheric pressure in Denver to that at sea level (630 mmHg/760 mmHg). MOF score is calculated as the sum of the simultaneously obtained individual organ scores on each hospital day. Single organ dysfunction is defined as an organ failure grade ⬎ 0. MOF is defined as a daily MOF score ⬎ 3 occurring more than 48 hours postinjury.7 Statistical analyses were performed using SAS for Windows version 8.0 (SAS Institute). Categorical vari-
Vol. 203, No. 4, October 2006
Ciesla et al
Obesity and Organ Failure after Trauma
541
Table 2. Denver Multiple Organ Failure Scale Dysfunction*
A Pulmonary† P/F ratio B Renal Creatinine (mg/dL) C Hepatic Total bilirubin (mg/dL) D Cardiac‡
Grade 0
⬎ 250 ⬍ 1.8 ⱕ 2.0 No inotropes
Grade 1
Grade 2
(250, 200)
(200, 100)
(1.8, 2.5)
(2.5, 5.0)
(2.0, 4.0) Minimal inotropes
(4.0, 8.0) Moderate inotropes
Grade 3
ⱕ 100 ⬎ 5.0 ⬎ 8.0 High inotropes
*Multiple Organ Failure Scale score ⫽ A ⫹ B ⫹ C ⫹ D, not due to chronic disease measured daily. † Pulmonary score is based on PaO2-to- FiO2 (P/F) ratio. ‡ Cardiac score: example—inotropes: minimal ⫽ dopamine ⬍ 5 g/kg/min; moderate inotropes ⫽ dopamine 5–15 g/kg/min; high inotropes ⫽ dopamine ⬎15 g /kg/min.
ables were analyzed using chi-square test with Yates’ correction for continuity or Fisher’s exact test when expected cell values were ⬍ 5. For continuous variables with normal distribution, ANOVA, or Student’s t-tests (with the appropriate Welch modification when the assumption of equal variances did not hold) were used. Spearman’s rank correlation was used for comparison of ordinal categorical values. Multivariable analyses were performed using logistic regression for categorical outcomes variables and standard linear regression for continuous outcomes variables. Model goodness-of-fit was assessed using the c statistic. Data are represented as mean ⫾ SE unless otherwise noted. A p value ⬍ 0.05 was considered significant. In summary, we found that obesity is a strong independent risk factor for postinjury organ dysfunction and MOF. Although we have no biochemical data on these patients at this time, future study will include collecting patient samples for inflammatory profiling as part of our NIH-funded Trauma Research Center grant (P50GM49222). The present findings, in conjunction with evidence for an altered inflammatory potential in obese patients, might point to therapeutic targets to improve outcomes in both obese and nonobese patients. Finally, the shift toward a more obese population could affect use of trauma and critical care resources because of the higher risk of postinjury MOF. RESULTS Data were collected on 716 severely injured patients over a 6.5-year period ending in June 2004. The majority (n ⫽ 508; 71%) were men, with a mean (⫾SD) age of 38.6 (⫾17) years and mean BMI (⫾SD) of 26.6 (⫾5.1). Blunt, penetrating, and mixed mechanisms accounted for 572 (80%), 73 (10%), and 71 (10%) injuries, respec-
tively, with an overall mean ISS (⫾SD) of 31.0 (⫾11.6). We did not detect a difference in the prevalence of obesity-related comorbidities before injury. The BMI distribution in the study group is shown in Figure 1. Two hundred eighty-six patients (40%) were normal or underweight (BMI ⬍ 25), 278 patients (39%) were overweight (BMI 25 to 29.9), and 152 patients (21%) were obese (BMI ⱖ 30). The mean BMI and proportion of obese patients did not change over the study period (p ⫽ 0.43 and p ⫽ 0.48, respectively). The mean BMI and proportion of obese patients increased considerably as a function of patient age in models using age as a continuous variable and models using age grouped as in Figure 2. The peak BMI of 29.2 ⫾ 0.8 (43% BMI ⬎ 30) occurring in the 55- to 65-year-old age group (p ⬍ 0.001, Fig. 2). Mean ISS and proportion of severely injured patients (ISS ⬎ 25) did not vary as a function of BMI in this study population (p ⫽ 0.26 and p ⫽ 0.41). We next examined the effect of obesity on the use of blood transfusion during the resuscitation period, which was defined as the first 12 hours after injury. Because blood transfusion is influenced by injury severity, the need for urgent or emergent operation, and patient age, we constructed models to adjust for these factors. Linear models demonstrated no substantial relationship between the number of RBC units transfused during resuscitation and BMI after adjusting for age, injury severity, and need for emergent operation. Results were similar when the independent variables were converted to categorical variables (BMI ⱖ 30, age older than 55 years, and ISS ⬎ 25). Our previous work identified transfusion of ⬎ 6 U RBC during resuscitation as a strong independent risk factor for subsequent development of postinjury MOF.8 Using logistic regression
Obesity and Organ Failure after Trauma
Ciesla et al
45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
J Am Coll Surg
%Patients BMI > 30
% of Patien ts
542
<18.5
18.5-24.9
25-29.9
30-34.9
35-39.9
≥40
50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
<25
25.1-35
BMI (kg/m2)
35.1-45 45.1-55 Age (yrs)
55.1-65
≥65
Figure 1. Distribution of body mass index (BMI) in study population.
Figure 2. Prevalence of obesity according to patient age. BMI, body mass index.
models, we found that obese patients were not more likely to receive ⬎ 6 U blood during resuscitation than nonobese patients after adjusting for advanced age (age older than 55 years), severe injury (ISS ⬎ 25), and need for urgent or emergent operation (Table 3). Results were similar in models that used BMI, age, and ISS as continuous independent variables. The incidence of individual organ dysfunction in obese and nonobese patients is shown in Table 4. Obese patients were observed to have a higher unadjusted incidence of lung, heart, liver, and kidney dysfunction and an increased incidence of ARDS in univariate analyses. But statistical significance was observed in logistic models adjusting for age, ISS, and RBC transfused during resuscitation in only heart and liver dysfunction (although a nonsignificant trend was observed in lung dysfunction [p ⫽ 0.057] and ARDS [p ⫽ 0.073]). Incidence of postinjury MOF in patients stratified by weight is shown in Figure 3. Postinjury MOF was observed in 123 of 564 (22%) nonobese patients and 56 of 152 (37%) obese patients. Logistic models demonstrated that postinjury MOF is 1.81 (95% CI, 1.20– 2.71) times more likely to develop in obese patients than nonobese patients after adjusting for advanced age, se-
vere injury, and transfusion of ⬎ 6 U blood during resuscitation (Table 5). Advanced age, severe injury, and transfusion of ⬎ 6 U blood during resuscitation were also notably associated with postinjury MOF in agreement with earlier studies.9 Logistic models using continuous variables for age, ISS, and units of blood transfused showed similar results. MOF developed within 72 hours of injury (early MOF) in 60 of 126 (48%) nonobese patients and 41 of 56 (73%) obese patients (p ⫽ 0.002). Of the patients who met MOF criteria, the spectrum of organ failure did not differ between obese and nonobese patients in both univariate and multivariable analyses (Table 6). Hospital and SICU length of stay increased with BMI (Fig. 4). Overall hospital length of stay was 20.1 ⫾ 1.6 days and 25.2 ⫾ 1.4 (p ⫽ 0.02) days for nonobese and obese patients, respectively. Similarly, SICU length of stay was 16.1 ⫾ 0.6 and 21.3 ⫾ 1.4 days (p ⬍ 0.001) for nonobese and obese patients, respectively. Of the patients that survived, 320 (61%) of 522 nonobese patients and 99 (74%) of 134 obese patients were transferred to a longterm acute-care facility on discharge (p ⫽ 0.009), the remainder were discharged to home.
Table 3. Association of Body Mass Index, Age, Injury Severity, and Operation Within 48 Hours With Blood Transfusion During Resuscitation RBC > 6 U*
Intercept BMI ⱖ 30 Age (y) ⬎ 55 ISS ⬎ 25 OR48
Odds ratio (95% CI)
1.35 (0.86–2.115) 0.82 (0.48–1.43) 2.84 (1.73–4.67) 7.62 (4.61–12.58) c ⫽ 0.75
p Value
0.188 0.50 ⬍ 0.01 ⬍ 0.01
*Outcomes variable is transfusion of ⬎ 6 U blood within 12 h of injury. BMI, body mass index (calculated as kg/m2); ISS, injury severity score; OR48, operation within 48 h of injury.
Table 4. Incidence of Organ Dysfunction in Obese and Nonobese Patients Organ dysfunction
n Lung (%) ARDS (%) Heart (%) Liver (%) Kidney (%)
Incidence p Value BMI BMI c < 30 > 30 Univariate Multivariate* Statistic
564 78 29 26 24 9
162 89 41 41 38 15
0.003 0.006 ⬍ 0.001 0.001 0.042
0.057 0.073 0.024 0.025 0.435
0.79 0.64 0.69 0.70 0.72
*Multivariate adjusted for age, injury severity, and blood transfusion during resuscitation. BMI, body mass index (calculated as kg/m2).
Vol. 203, No. 4, October 2006
Ciesla et al
Days (m ean±S E )
% o f Pat ient s w it h MO F
35% 30% 25% 20% 15% 10%
25.0 20.0 15.0 10.0
5%
<25
0% <25
25-29.9 BMI (kg/m 2)
Fifty-seven (8%) patients died. Unadjusted mortality rates for obese and nonobese patients were 7% and 11%, respectively, which were not different by univariate analysis (p ⫽ 0.14). There was no difference in mortality between obese and nonobese patients after adjusting for age, injury severity, blood transfused during resuscitation, and postinjury MOF (Table 7). Inclusion of the interaction between obesity and MOF did not improve the logistic model. DISCUSSION The increasing obesity rate threatens to alter current medical and surgical practice.10 Obese patients are more likely to have substantial medical comorbidities and are at increased risk of postsurgical and posttraumatic complications. It is not hard to postulate that the “obesity epidemic” has the potential to alter current practice patterns to the same degree as the “ageing of America.” Although the population of trauma victims does not necessarily mirror the American population as a whole, the general trend toward increased obesity across age groups and gender can be expected to shift the populaTable 5. Influence of Obesity, Age, Injury Severity, and Transfusion of Blood During Resuscitation on the Incidence of Postinjury Multiple Organ Failure Odds ratio (95% CI)
1.81 (1.21–2.71) 2.05 (1.32–3.18) 1.78 (1.16–2.76) 2.41 (1.63–3.56) c ⫽ 0.67
25-29.9
≥30
BM I (kg/m2)
≥30
Figure 3. Percentage with postinjury multiple organ failure (MOF) according to body mass index (BMI).
Intercept BMI ⱖ 30 Age (y) ⬎ 55 ISS ⬎ 25 RBC ⬎ 6*
543
30.0
40%
MOF
Obesity and Organ Failure after Trauma
p Value
0.004 0.001 0.009 ⬍ 0.001
Outcomes variable is the presence of postinjury MOF. *Transfusion of ⬎ 6 U blood within 12 h of injury. BMI, body mass index (calculated as kg/m2); ISS, Injury Severity Score; MOF, multiple organ failure.
Figure 4. Mean ⫾ SE intensive care unit (white) and hospital (black) lengths of stay in nonobese and obese patients. BMI, body mass index.
tion of injured patients toward a more obese demographic. An understanding of obesity-related complications is vital to those involved in trauma care. The prevalence of obesity in this study population was 21%, which is consistent with the overall prevalence of obesity in Colorado but is somewhat lower than the national average of 31%.1 We did not find a substantial increase in the proportion of obese patients in our population during the study period. Although this is in contrast to national trends, it is possible that our sample size could not detect small changes in the prevalence of obesity over this time period. The historic increase in obesity in Colorado has lagged behind other states and national trends. In agreement with others, we found that the prevalence of obesity increased with patient age, and the highest proportion of obese patients were found in the 55- to 65-year-old age group.11 We found no relationship between BMI and injury severity. Although this is consistent with recent reports, others have found different injury patterns in obese patients compared with lean patients. The group from the Los Angeles County Medical Center found that obese patients had a higher frequency of chest injuries and lower extremity fractures, but a lower frequency of head injuries.12 Arbabi and colleagues13 also found a higher Table 6. Incidence of Organ Failure after Postinjury Multiple Organ Failure Developed Incidence (%) Organ system
Lung ARDS Heart Liver Kidney
BMI < 30
BMI > 30
100 64 72 67 29
100 63 82 71 34
BMI, body mass index (calculated as kg/m2).
544
Ciesla et al
Obesity and Organ Failure after Trauma
Table 7. Influence of Obesity, Age, Injury Severity, Transfusion of Blood During Resuscitation, and Multiple Organ Failure on Postinjury Mortality Death
Intercept BMI ⱖ 30 Age ISS RBC MOF
Odds ratio (95% CI)
0.79 (0.39–1.58) 1.05 (1.03–1.07) 1.04 (1.01–1.06) 1.03 (1.00–1.06) 7.66 (3.86–15.20) c ⫽ 0.87
p Value
0.499 ⬍ 0.001 0.002 0.070 ⬍ 0.001
Outcomes variable is death within 28 d of injury. Age, ISS, and units of blood transfused within 12 h of injury (RBC) were treated as continuous variables. BMI, body mass index (calculated as kg/m2); ISS, Injury Severity Score; MOF, multiple organ failure.
frequency of lower extremity fractures, but a lower frequency of abdominal injuries, invoking the “cushion effect” of the obese abdomen that protects the abdominal viscera from injury. Conversely, we found no difference in injury severity or injury pattern between obese and nonobese patients.14 These disparate results make any conclusions about obesity and injury pattern somewhat tentative. Transfusion of stored RBC during resuscitation is independently associated with posttraumatic complications and organ dysfunction.8 The amount of blood transfused during the resuscitation period is influenced by patient age, injury severity, and need for urgent or emergent operation. In the present study, we found no relationship between RBC transfusions during resuscitation after adjusting for these factors. This is important for allocating blood bank resources and for evaluating the effect of obesity on posttraumatic complications independent of the effects of blood transfusion. Previous reports have identified obese patients to be at risk for postinjury organ dysfunction. The Los Angeles County group found that obese patients had a higher rate of ARDS, acute renal failure requiring dialysis, and myocardial infarction.12 Byrnes and colleagues14 also found a higher rate of renal failure, but no difference in respiratory failure or cardiovascular complications in obese patients compared with nonobese patients. These studies did not control for other risk factors for postinjury organ dysfunction, such as patient age, injury severity, and blood transfusion. In the present study, obese patients were found to have higher unadjusted rates of lung, heart, liver, and kidney dysfunction. Multiple logistic regression models demonstrated that obesity was independently associated with only heart and liver dys-
J Am Coll Surg
function, although there was a trend toward increased lung dysfunction and ARDS that did not reach statistical significance. When multiple systems were evaluated, we found that postinjury MOF was 1.8 times more likely to develop in obese patients than nonobese patients after adjusting for the previously established risk factors. Obesity was found to confer an increased risk of postinjury MOF greater than that of severe injury. MOF was more likely to develop within 72 hours of injury in obese patients than nonobese patients. Because early MOF is thought to reflect the effects of unbridled postinjury hyperinflammation as opposed to uncontrolled infection, this observation supports the proinflammatory nature of the obese state. Alternatively, early organ dysfunction could represent decreased physiologic reserve of these patients. Of those in whom MOF developed, there were no differences in the spectrum of involved organs between obese and nonobese patients. The lung was the most frequent organ involved, followed by the heart, liver, and kidney. It is interesting that once the threshold for MOF is reached, obese patients display the same organ dysfunction pattern as nonobese patients. This might be evidence that the final common pathway to postinjury organ dysfunction is not different between obese and nonobese patients. We conclude from these observations and previous research that obesity is associated with a proinflammatory state that influences the postinjury inflammatory response and resultant organ dysfunction. The association between obesity and several chronic diseases is well-known. Obese patients are also more likely to require inpatient and outpatient medical resources.15 In the present study, we found that obese patients were more likely to have longer SICU and hospital lengths of stay and more likely to be transferred to a longterm acute-care facility than to be discharged to home. Increased use of medical resources impacts not only health care costs, but also the allocation of often scarce resources to the injured population.16 Increased length of stay and increased use of post acute-care facilities have the potential to require more care in acute-care hospitals and strain bed availability at the trauma center. Although some authors have reported that obese patients are at increased risk of postinjury death, the issue remains largely controversial. The Los Angeles County group found that obese victims of blunt trauma admitted to the ICU were 5.6 times more likely to die than nonobese patients in multivariable analyses of 242 pa-
Vol. 203, No. 4, October 2006
tients.17 Head injury was the most frequent cause of death. A followup study of 1,153 patients found that obese patients were only 1.6 times more likely to die than nonobese patients using a different multivariable analysis.12 This same group also reported that there was no increased risk of death in 316 obese children, suggesting that the observed effect can be influenced by patient age.18 Others have also observed increased mortality in obese patients stratified by injury severity.13,14,19 We found no relationship between obesity and mortality after adjusting for age, injury severity, blood transfusion during resuscitation, and development of postinjury MOF. Our study population was limited to severely injured patients (ISS ⬎ 15) admitted to the SICU and included both blunt and penetrating mechanisms. Perhaps more importantly, we excluded patients who died within 48 hours of injury and patients with isolated head injuries and head injuries with minor extremity injuries from study. It is possible that a different result would have been observed had these patients been included. The lack of consistent results to date leaves the relationship between obesity and postinjury mortality undefined. Future study using a large inclusive trauma database, such as the National Trauma Data Bank, which includes height and weight data, might ultimately provide an answer. In summary, we found that obesity is a strong independent risk factor for postinjury organ dysfunction and MOF. These findings, in conjunction with evidence for an altered inflammatory potential in obese patients, might point to therapeutic targets to improve outcomes in both obese and nonobese patients. Finally, the shift toward a more obese population could affect use of trauma and critical care resources because of the higher risk of postinjury MOF. Author Contributions
Study conception and design: Ciesla, Moore, Johnson, Sauaia Acquisition of data: Ciesla, Moore, Johnson Analysis and interpretation of data: Ciesla, Moore, Johnson, Burch, Cothren, Sauaia Drafting of manuscript: Ciesla, Moore, Johnson, Sauaia
Ciesla et al
Obesity and Organ Failure after Trauma
545
Critical revision: Ciesla, Moore, Johnson, Burch, Cothren, Sauaia REFERENCES 1. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA 2003;289:76–79. 2. Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005;293:1861–1867. 3. Lyon CJ, Law RE, Hsueh WA. Minireview: adiposity, inflammation, and atherogenesis. Endocrinology 2003;144:2195– 2200. 4. Moore EE, Moore FA. Postinjury multiple organ failure. In: Moore EE, Fleciano DV, Mattox KL, eds. Trauma. 5th ed. New York: McGraw-Hill; 2004:1397–1423. 5. Calvet HM, Yoshikawa TT. Infections in diabetes. Infect Dis Clin North Am 2001;15:407–421, viii. 6. Ciesla DJ, Moore EE, Johnson JL, et al. A 12-year prospective study of postinjury multiple organ failure: has anything changed? Arch Surg 2005;140:432–438; discussion 438–440. 7. Ciesla DJ, Moore EE, Johnson JL, et al. Multiple organ dysfunction during resuscitation is not postinjury multiple organ failure. Arch Surg 2004;139:590–594; discussion 594–595. 8. Moore FA, Moore EE, Sauaia A. Blood transfusion. An independent risk factor for postinjury multiple organ failure. Arch Surg 1997;132:620–624; discussion 624–625. 9. Sauaia A, Moore FA, Moore EE, et al. Early predictors of postinjury multiple organ failure. Arch Surg 1994;129:39–45. 10. Quesenberry CP Jr, Caan B, Jacobson A. Obesity, health services use, and health care costs among members of a health maintenance organization. Arch Intern Med 1998;158:466–472. 11. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA 2002;288:1723–1727. 12. Brown CV, Neville AL, Rhee P, et al. The impact of obesity on the outcomes of 1,153 critically injured blunt trauma patients. J Trauma 2005;59:1048–1051; discussion 1051. 13. Arbabi S, Wahl WL, Hemmila MR, et al. The cushion effect. J Trauma 2003;54:1090–1093. 14. Byrnes MC, McDaniel MD, Moore MB, et al. The effect of obesity on outcomes among injured patients. J Trauma 2005; 58:232–237. 15. Bertakis KD, Azari R. Obesity and the use of health care services. Obes Res 2005;13:372–379. 16. Buntin MB, Garten AD, Paddock S, et al. How much is postacute care use affected by its availability? Health Serv Res 2005; 40:413–434. 17. Neville AL, Brown CV, Weng J, et al. Obesity is an independent risk factor of mortality in severely injured blunt trauma patients. Arch Surg 2004;139:983–987. 18. Brown CV, Neville AL, Salim A, et al. The impact of obesity on severely injured children and adolescents. J Pediatr Surg 2006; 41:88–91; discussion 88–91. 19. Choban PS, Weireter LJ Jr, Maynes C. Obesity and increased mortality in blunt trauma. J Trauma 1991;31:1253–1257.