- Email: [email protected]
Utilizing quantitative measures of visceral adiposity in evaluating trauma patient outcomes
International Journal of Surgery 21 (2015) 51e56
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
Utilizing quantitative measures of visceral adiposity in evaluating trauma patient outcomes Salvatore Docimo Jr. *, Brooke Lamparello, Melissa Fay Cohen, Anthony Kopatsis, Fausto Vinces NYU Lutheran Medical Center, Department of Surgery, 150 55th Street, Brooklyn, NY 11220, USA
h i g h l i g h t s Quantitative radiologic measures of visceral adiposity is the gold standard to assess obesity. This is the first attempt to associate V/S ratios with trauma outcome measurements. Clinical association of an increased V/S ratio and complications in trauma patients noted. Further prospective studies are required for further analysis.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 March 2015 Received in revised form 11 May 2015 Accepted 20 June 2015 Available online 9 July 2015
Introduction: Body mass index (BMI) has commonly been used as a parameter to assess obesity in trauma patients. However, the variability of height and weight data in trauma patients limits the use of BMI as an accurate assessment tool in the trauma population. Quantitative radiologic measurements of visceral adiposity is an accurate method for assessing obesity in patients but requires further analysis before it can be accepted as a measurement tool for trauma patients. Methods: A retrospective review of trauma cases with pre-operative CT scan from 2008 to 2015 produced 57 patients for evaluation. Preoperative BMI was calculated using measured height (m2) and weight (kg). Radiologic measurements of adiposity were obtained from preoperative CT scans using OsiriX DICOM viewer software. Visceral fat areas (VFA) and subcutaneous fat areas (SFA) were measured from a single axial slice at the level of L4-L5 intervertebral space. Results: No statistically significant results were found relating visceral fat:subcutaneous fat ratios to length of stay or post-operative complications. Initial clinical observations noting an increased incidence of complications among patients with a V/S 0.4 demonstrates a possible link between obesity and poor outcomes in trauma patients. A statistically significant correlation was noted between length of stay, peri-nephric fat and injury severity score. Discussion and Conclusion: Our pilot study should be viewed as the foundation for a larger prospective study, utilizing quantitative measurements of visceral adiposity to assess outcomes in trauma patients. © 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
Keywords: Visceral adiposity Quantitative measures Adiposity Trauma outcomes
1. Introduction Obesity in the United States has become a significant health problem. In a 2014 JAMA publication, the prevalence of obesity in the United States was estimated to be 34.9% in 2011e2012 [1]. To
* Corresponding author. Present address: Penn State Milton S. Hershey Medical Center, Department of Minimally Invasive & Bariatric Surgery, 500 University Dr, Hershey, PA 17033, USA. E-mail address: [email protected] (S. Docimo).
date, the association between obesity and patient outcomes in the setting of trauma is still under investigation. Previous studies have demonstrated higher rates of in-hospital complications and increased mortality in obese patients suffering blunt traumatic injury [2e4]. Glance et al. [5] performed a retrospective review of nearly 150,000 patients, which after adjusting for injury severity and other factors, found severely obese trauma patients were at least 30% more likely to die and twice as likely to have major complications compared with non-obese patients. Body mass index (BMI) has commonly been used in the literature as a parameter to assess obesity in trauma patients [3e5]. The
http://dx.doi.org/10.1016/j.ijsu.2015.06.069 1743-9191/© 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
52
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
previously published literature demonstrating obesity as an independent risk factor of mortality in trauma patients utilized only BMI as their standard for comparison. However, variability in the definition of obesity in previously published studies presents a difficult problem in standardizing the assessment of obesity and its affect on patient outcomes. The National Institutes of Health [7] define obesity as a BMI 30 kg/m2. Previous studies have applied their own definitions and interpretations of obesity, which can lead to findings that may not be easily extrapolated to other trauma populations. Smith-Choban et al. [8] classified patients as normal (BMI < 27 kg/m2), over-weight (BMI 27e31 kg/m2), and obese (BMI > 31 kg/m2). Byrnes et al. [9] chose to define obesity for their patients as a BMI 35 kg/m2. Such variability in obesity definitions can create a barrier to consistent extrapolation of data. However, quantitative radiologic analysis of visceral adiposity using standard CT scans have been described as being the goldstandard method for assessing visceral adiposity [10,11]. Quantitative analysis of visceral adiposity using CT-based measurements has been evaluated as a more effective means of assessing obesity in patients [10,11]. In some disease processes, such as rectal cancer, visceral adiposity has shown to have stronger associations with cancer recurrence than BMI [12]. Malietzis et al. [13] also utilized CT-based measurements of body composition to demonstrate a link between visceral obesity, reduced skeletal muscle, and poorer short-term recovery, poorer oncological outcomes, and poorer survival. To our knowledge, this study is the first to apply quantitative radiologic analysis of visceral adiposity to trauma patients undergoing surgical exploration to evaluate complication rates, such as anastomotic leaks and pulmonary embolism, and outcomes, such as length of stay. The objective of this pilot study is to introduce quantitative measures of visceral adiposity as a tool to measure the effect of obesity on the post-operative morbidity and mortality of trauma patients undergoing surgical intervention for abdominal trauma. 2. Methods A retrospective review of trauma cases from 2008 to 2015 was performed with institutional review board approval. All trauma patients over the age of 18 years of age with abdominal trauma who underwent surgical repair were examined. Those patients who had a computed tomography scan performed after arriving in the trauma bay was included in the study. Data was collected on patient demographics, medical co-morbidities, preoperative vital signs, operative findings, complications and length of stay. 2.1. Systematic data retrieval Preoperative BMI was calculated using measured height (m2) and weight (kg). Radiologic measurements of adiposity were obtained from preoperative CT scans using OsiriX DICOM viewer software. Visceral fat areas (VFA) and subcutaneous fat areas (SFA) were measured from a single axial slice at the level of L4-L5 intervertebral space. The CT attenuation level to delineate the regions of adipose tissue was set using Hounsfield units of 190 to 30 [12]. The VFA and the SFA were then delineated and measured. Fig. 1 provides an example of the quantitative visceral adiposity measurements taken during the study. The visceral fat area to subcutaneous fat area ratio (V/S) was calculated using the VFA and SFA measurements. Elevated V/S ratio indicates a larger amount of visceral fat compared to subcutaneous fat with a defined obesity threshold of V/S 0.4 [12,14]. Linear perinephric fat (PNF) thickness was measured at the level of the renal veins. Perinephric fat thickness was defined as the shortest distance (in mm) between
the kidney and abdominal wall at the level of the renal vein [12,15]. 2.2. Statistical analysis Adiposity variables (PNF, SFA, VFA) were measured as continuous variables. Linear correlations were calculated using SPSS between various adiposity variables and endpoints (LOS, complications, intra-operative pressors, etc). The chi-square, Fisher's exact tests, and student T-tests were utilized to evaluate the differences in adiposity variables and endpoints. We attempted to eliminate bias in the study by having one individual, who was the not the primary investigator, to gather the visceral and subcutaneous measurements. Furthermore, the individuals who were enrolled in the study were randomly generated from our trauma registry and not specifically chosen by any of the investigators. 3. Results 3.1. Demographics and clinical features The study population consisted of 57 patients, who underwent surgical intervention for abdominal trauma, which was confirmed on preoperative abdominal CT scans. There were more men (n ¼ 49) than women (n ¼ 8) with the mean age of 36.77 ± 16.83 years. The frequency of comorbidities such as diabetes, hypertension, and hypercholesterolemia among this population was 10.5%, 19.3% and 3.5%. Thirty-seven (65%) patients underwent exploratory laparotomy and 20 (35%) underwent laparoscopic surgical repair. Four (9%) patients presented with hypotension in the trauma bay. Eleven (19.3%) patients experienced hypotension in the operating room with seven (12.2%) and 10 (17.5%) patients requiring pressor support and colloids. Table 1 summarizes the demographic and comorbidity findings. 3.2. Adiposity measurements The patients demonstrated a mean BMI of 30.0 kg/m2, a mean perinephric fat thickness of 1.34 cm, a mean subcutaneous fat area of 142.64, a mean visceral fat area of 91.25, and a mean V/S ratio of 0.89. Mean PNF was higher in females than in males (1.56 vs 1.31; p ¼ 0.63), higher in patients without hypertension (1.37 vs 1.26; p ¼ 0.80) and higher in patients without DM (1.37 vs 1.17; p ¼ 0.73). A BMI >30 kg/m2 was noted only in male patients with an average BMI of 37.23 kg/m2 A V/S ratio 0.4 (indicating obesity) was more common in males (n ¼ 35) vs. females (n ¼ 7) who had an average V/S ratio of 0.91 compared to a V/S ratio of 0.75 in females (p ¼ 0.66). There was no statistical significance regarding a V/ S > 0.4 and patients having HTN or no HTN (0.87 vs 1.17; p ¼ 0.46) or DMand no DM (1.10 vs 1.22; p ¼ 0.85). Table 2 summarizes the obesity measurement findings for the 57 patients. 3.3. Overall analysis The mean injury severity score (ISS) 28.4 ± 12.2 and was not correlated with PNF (p ¼ 0.55), V/S ratio (p ¼ 0.86) or BMI (p ¼ 0.60), indicating injuries were well randomized. The mean LOS was 9.2 ± 10.4 days. The LOS did demonstrate a correlation to PNF (p ¼ 0.031) and ISS (p ¼ 0.01). The LOS did not correlate with subcutaneous fat thickness (p ¼ 0.79), visceral fat (p ¼ 0.87), BMI (p ¼ 0.73), and V/S ratio (p ¼ 0.86). The incidence of complications was not correlated with PNF (p ¼ 0.27), subcutaneous fat area (p ¼ 0.10), visceral fat area (p ¼ 0.11), or V/S ratio (p ¼ 0.34). The likelihood of a correlation between V/S ratio and intraoperative hypotension was not statistically significant (p ¼ 0.93). The likelihood of a correlation between V/S ratio and postoperative
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
53
Fig. 1. Quantitative measurements of visceral adiposity. Measurement of pernephric fat thickness at the level of the renal vein (A). Measurement of the subcutaneous fat area (B). Measurement of the visceral fat area (C). The V/S ratio is calculated by dividing the measurement of C by the measurement of B. Subcutaneous and visceral fat area measurements are calculated at the level of the renal vein.
S 0.4 had an incidence of post-operative complication, whereas, two patients with a post-operative complication had a V/S ratio <0.4 (p ¼ 0.29).
Table 1 Demographics and Co-Morbidities. N (%)
3.4. Data stratification
Sex Male Female Race Caucasian Hispanic African-American Asian Unknown Comorbidities Diabetes Hypertension Hypercholesterolemia
49 (86%) 8 (14%) 7 (12.2%) 11 (19.3%) 3 (5.2%) 4 (7.0%) 32 (56.1%) 6 (10.5%) 11 (19.3%) 2 (3.5%)
Table 2 Body mass index and visceral adiposity measurements overall and stratified by male and female.
BMI SFA VFA V/S
Overall (n ¼ 57)
Males (n ¼ 49)
Females (n ¼ 8)
30.0 149.3 99.7 0.75
30.6 155.5 99.8 0.91
21.5 64.0 38.5 0.75
hypotension was not statistically significant (p ¼ 0.49). The likelihood of a correlation between BMI and intraoperative hypotension was not statistically significant (p ¼ 0.93). The likelihood of a correlation between BMI and postoperative hypotension was not statistically significant (p ¼ 0.36). The V/S ratio was not found to correlate with the use of intraoperative pressors (p ¼ 0.57). No significant correlation between V/S ratio >0.4 and postoperative complications was found. Fifteen patients with a V/
The 57 patients were further stratified into two groups: patients with a V/S ratio 0.4 (n ¼ 42, 73.7%) and a V/S ratio 0.4 (n ¼ 15, 26.3%). Student t-tests between the V/S 0.4 and V/S 0.4 groups did not demonstrate any significant findings in terms of LOS (p ¼ 0.38), post-operative complications (p ¼ 0.10), and ISS (p ¼ 0.27). Table 3 summarizes the findings between the V/S ratio groups and the BMI groups. The 57 patients were further stratified into two groups: patients with a BMI 30 kg/m2 (n ¼ 20, 35.0%) and a BMI < 30 kg/m2 (n ¼ 37, 65.0%). Student t-tests between the BMI 30 kg/m2and BMI <30 kg/m2 groups did not demonstrate any significant findings in terms of LOS (p ¼ 0.88), post-operative complications (p ¼ 0.26), and ISS (p ¼ 0.45). Fig. 2 graphically summarizes the findings of LOS, incidence of complications, and average BMI based upon the V/ S ratio and BMI groups. The 57 patients were further stratified into two groups: patients who underwent open surgical repair (n ¼ 36, 63.1%) and those who underwent laparoscopic repairs (n ¼ 21, 36.8%). Comparing the length of stay between the two groups did not present any statistically significant findings (p ¼ 0.10). Comparing the incidence of complications between the two groups did not present any statistically significant findings (p ¼ 0.98). The ISS between the two groups was similar and did not demonstrate any significant differences (p ¼ 0.11). 4. Discussion Using BMI, one-third of the United States population is obese.1
54
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
Table 3 Study characteristics stratified by body mass index and V/S ratio. Body mass index
Age Sex Male Female Comorbidities Hypertension Hypercholesterolemia Diabetes Intra-operative Hypotension Pressor use Colloid use Post-operative Hypotension Pressor use Colloid use Length of stay Complications Injury severity score
V/S ratio
<30 (n ¼ 36)
30 (n ¼ 21)
V/S < 0.4 (n ¼ 15)
V/S 0.4 (n ¼ 42)
38.5
35.1
27.8
39.9
27 (75%) 6 (25%)
21 (100%) 0
14 (93.3%) 1 (6.7%)
35 (83.3%) 7 (16.7%)
7 (19.4) 0 4 (11.1%)
3(14.3%) 2 (9.5%) 1 (4.7%)
2 (13.3%) 0 3 (20.0%)
8 (19%) 2 (4.7%) 3 (7.1%)
6 (16.6%) 5 (13.8%) 6 (16.6%)
3 (14.3%) 1 (4.7%) 3 (14.3%)
3 (20.0%) 1 (6.7%) 4 (26.6%)
7 (16.6%) 5 (11.9%) 6 (14.3%)
6 (16.6%) 5 (13.8%) 3 (8.3%) 9.96 13 (36.1%) 28.91
5 (23.8%) 1 (4.7%) 3 (14.3%) 9.17 4 (19.0%) 29.01
3 (20.0%) 1 (6.7%) 4 (26.6%) 11.26 2 (13.3%) 30.66
8 (19.0%) 6 (14.3%) 7 (16.6%) 8.5 15 (35.7%) 26.68
Fig. 2. Length of stay, incidence of complications, and average body mass index stratified by V/S ratio and body mass index.
In our study, 18 (32%) of the patients were found to be obese using a BMI of 30 kg/m.2 or more. According to the V/S ratio (with a V/ S 0.4 being obese), 43 (79.4%) of patients were classified as obese. This finding may indicate the V/S ratio to be a more sensitive indicator of obesity. Table 1 provides a summary of demographics. Obesity clinically presents as a metabolic syndrome involving dyslipidemia, hypertension, insulin resistance, prothrombotic states, and pro-inflammatory states, which have been linked to worse outcomes following severe injury [16]. The systemic proinflammatory response state in obese trauma patients is due to abundant adipose tissue which releases elevated amounts of inflammatory mediators, including tumor necrosis factor-a (TNF-a), interleukin-6 (IL-6), and leptin [12,17,18]. Shashaty et al. [19] also implicated abdominal adiposity as an independent risk factor for the development of acute kidney injury (AKI) in critically ill trauma patients. Several mechanisms have been theorized to link obesity to AKI, such as, a potentiation of inflammatory mediators (interleukin-6, tumor necrosis factor-a, monocyte chemoattractant protein-1) by adipose tissue and subclinical chronic kidney disease due to an abnormal glomerular architecture seen in morbidly obese patients [19]. However, controversy exists as to whether the cause of increased morbidity and mortality is due to size (a mechanical
disadvantage) rather than an increased pro-inflammatory state [20e22]. The hypothesis of this study stated that obese trauma patients, defined as having a V/S 0.4, would have worse clinical outcomes, such as a longer length of stay (LOS)or incidence of complications. In order to eliminate the concern that injury severity could be cause of increased length of stay or incidence of complications, we calculated the injury severity score among each stratified group of patients before statistical analysis was performed. In order to attempt to control for this variability in injury severity, the ISS was calculated. The ISS was variable and did not correlate statistically with BMI or V/S ratios, demonstrating the randomization of injury among the patients, and eliminating the argument of injury severity as the root cause of any increase in LOS of complication rate. For the sample of 57 patients, the mean LOS was 9.2 ± 10.4 days with an ISS of 28.4 ± 12.2 [20]. The LOS was not correlated with subcutaneous fat thickness (0.79), visceral fat (0.87), BMI ((0.73), and V/S ratio (0.86). The LOS did demonstrate a correlation to PNF (p ¼ 0.031) and ISS (p ¼ 0.01). An elevated ISS signifies more traumatic injuries, which one would expect to translate to a longer LOS. The fact that PNF did correlate with LOS suggests varying distribution of visceral fact can affect the types of traumatic injuries patients' experience. A previous study by Evans et al. [23] did demonstrate a higher BMI to be associated with a higher incidence of torso injury and proximal upper extremity injuries in blunt trauma. The incidence of post-operative complications was also not correlated with perinephric fat thickness (0.27), subcutaneous fat area (0.10), visceral fat area (0.11), or V/S ratio (0.34). Furthermore, the incidence of post-operative hypotension and pressor use was not significantly correlated to PNF (p ¼ 0.812, p ¼ 0.17), and V/S ratio (p ¼ 0.7, p ¼ 0.61). When comparing the LOS in patients with a V/S ratio 0.4 and <0.4, there were no statistically significant differences (p ¼ 0.38). Comparing the incidence of complications, there were no statistically significant findings between these two groups (p ¼ 0.1). It must be noted 15 (37%) patients with a V/S 0.4 had an incidence of post-operative complication, whereas, two (15%) patients with a post-operative complication had a V/S ratio < 0.4 (p ¼ 0.29). Clinically, there may be a link between an elevated V/S ratio (V/S 0.4) and an increased incidence of complications. However, a small sample size likely has hampered our ability to obtain statistically significant results. The ISS between the two groups were not
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
statistically significant and less likely, compared to the increased V/ S ratio, the reason for the differences seen in the complication rates. Further stratification of the data into two groups, patients with a BMI 30 and a BMI <30, did not replicate significant findings that were noted in previous articles. The differences in LOS (p ¼ 0.88) and incidence of complications (p ¼ 0.26) were not found to be statistically significant. The ISS score between the two groups was not statistically significant. Further stratification into open surgical repair and laparoscopic repair also failed to demonstrate any significant differences in the LOS and incidence of complications. Table 3 summarizes the findings of the study characteristics stratified by BMI and V/S Ratio. 4.1. Limitations A significant limitation of this study is the small sample size. Previously published literature also demonstrated the difficulty of capturing BMI parameters, such as height and weight, as exemplified by small sample sizes. Previous studies have reported the need to exclude significant amounts of patients due to height and weight measurements not being available [8,9]. Our small sample size stems from the fact we attempted to obtain pre-operative CT scans in patients with abdominal trauma, majority being penetrating. In the majority of cases with penetrating abdominal trauma, the patient is taken to the operating room for exploration in lieu of a pre-operative CT scan. This fact alone limits our sample size. We also cannot accurately estimate the number of patients who were excluded from the study due to missing height data, weight data and also a preoperatively CT scan. Though our study did not find any statistically significant findings in regards to the V/ S ration, the fact that PNF and ISS did correlate significantly to LOS suggests visceral fat may play a role in altering the types and significance of traumatic injuries. Therefore, the use of visceral adiposity as a measurement of obesity in the setting of trauma needs to be further explored. 4.2. External validity Currently, the results of our study cannot be generalized to the trauma population due to the lack of statistically significant findings. However, the data regarding the increased incidence of complication in patients with higher V/S ratios, as well as the statically significant correlation between LOS, PNF and ISS is promising. Further prospective studies with larger sample sizes are needed in order to obtain statistically significant results. If such findings are discovered, traumatologists and surgical intensivists need to become aware of the increased risk of complications and morbidity associated with obese trauma patients. 5. Conclusions The variability of height and weight data in trauma patients limits the use of BMI as an accurate assessment tool in the trauma population. Quantitative measurements of visceral adiposity is an accurate method for assessing obesity in patients but requires further analysis before it can be accepted as a measurement tool for trauma patients. The initial clinical observations noting an increased incidence of complications among patients with a V/ S 0.4 demonstrates a possibly link between obesity and poor outcomes in trauma patients. The statically significant correlation between LOS, PNF and ISS is promising and should act as a catalyst for further evaluation of the role of obesity in trauma. Our study should be viewed as the foundation for a large prospective study, utilizing quantitative measurements of visceral adiposity to assess outcomes in trauma patients.
55
Ethical approval Not required. Funding No sources of funding were needed for this project. Author contribution Docimo: study design, data analysis, writing. Lamparello: data collection, data analysis. Cohen: data collection. Kopatsis: data analysis. Vinces: data analysis. Conflict of interest statement All authors have no conflicts of interest to state. Guarantor Dr. Salvatore Docimo, Jr. is the guarantor. References [1] C.L. Ogden, M.D. Carroll, B.K. Kit, K.M. Flegal, Prevalence of childhood and adult obesity in the United States, 2011-2012, JAMA 311 (8) (2014) 806e814. [2] B.R. Boulanger, D.P. Milzman, A. Rodriguez, Obesity, Crit. Care Clin. 10 (1994) 613e622. [3] A. Neville, C.V. Brown, J. Weng, D. Demetriades, G. Velmahos, Obesity is an independent risk factor for mortality in severely injured blunt trauma patients, Arch. Surg. 139 (2004) 983e987. [4] Brown CV, Neville A, Rhee P, Salim A, Velmahos G, Demetriades D. The impact of obesity on the [5] L.G. Glance, Y. Li, T.M. Osler, D.B. Mukamel, A.W. Dick, Impact of obesity on mortality and complications in trauma patients, Ann. Surg. 259 (2014) 576e581. [7] National Institutes of Health, Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report, National Institutes of Health, Washington, DC, 1998, pp. 94e4083. Publication. [8] P. Smith-Choban, L.J. Weireter, C. Maynes, Obesity and increase mortality in blunt trauma, J. Trauma 31 (1991) 1253e1257. [9] M.C. Byrnes, M.D. McDaniel, M.B. Moore, S.D. Helmer, R.S. Smith, The effect of obesity on outcomes among injured patients, J. Trauma 58 (2005) 232e237. [10] A.M. Gradmark, A. Rydh, F. Renstrom, E. De Lucia-Rolfe, A. Sleigh, P. Nordstrom, S. Brage, P.W. Franks, Computed tomography-based validation of abdominal adiposity measurements from ultrasonography, dual-energy Xray absorptiometry and anthropometry, Br. J. Nutr. 104 (2010) 582e588. [11] K.M. Pou, J.M. Massaro, U. Hoffman, K. Lieb, R.S. Vasan, C.J. O'Donnell, C.S. Fox, Patterns of abdominal fat distribution: the Framingham Hear Study, Diabetes Care 32 (2009) 481e485. [12] W. Clark, W.M. Siegel, Y.A. Chen, X. Zhao, C.M. Parsons, J.M. Hernandez, J. Weber, S. Thareja, J. Choi, D. Shibata, Quantitative measures of visceral adiposity and body mass index in predicting rectal cancer outcomes after neoadjuvantchemoradiation, J. Am. Coll. Surg. 216 (2013) 1070e1081. [13] G. Malietzis, O. Aziz, N.M. Bagnall, N. Johns, K.C. Fearon, J.T. Jenkins, The role of body composition evaluation by computerized tomography in determining colorectal cancer treatment outcomes: a systemic review, Eur. J. Surg. Oncol. 41 (2) (2015 Feb) 186e189. [14] Y. Matsuzawa, T. Nakamura, I. Shimomura, K. Kotani, Visceral fat accumulation and cardiovascular disease, Obes. Res. 3 (1995) 645Se647S. [15] K. Morris, S. Tuorto, M. Gonen, et al., Simple measurement of intra-abdominal fat for abdominal surgery outcome prediction, Arch. Surg. 145 (2010) 1069e1073. [16] M.E. Falagas, M. Kompoti, Obesity and infection, Lancet Infect. Dis. 6 (2006) 438e446. [17] C.J. Lyon, R.E. Law, W.A. Hsueh, Mini review: adiposity, inflammation and thermogenesis, Endocrinology 144 (2003) 2195e2200. [18] F.A. Moore, A. Sauaia, E.E. Moore, J.B. Haenel, J.M. Burch, D.C. Lezotte, Postinjury multiple organ failure: a bimodal phenomenon, J. Trauma 40 (4) (1996 Apr) 501e510. [19] M.G. Shashaty, E. Kalkan, S.L. Bellamy, J.P. Reilly, D.N. Holena, K. Cummins, P.N. Lanken, H.I. Feldman, M.P. Reilly, J.K. Udupa, J.D. Christie, Computed tomography-defined abdominal adiposity is associated with acute kidney injury in critically ill trauma patients, Crit. Care Med. 42 (7) (2014)
56
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
1619e1628. [20] B. Collier, L. Dossett, J. Shipman, M. Day, G. Lawson, R. Sawyer, A. May, Visceral adiposity is not associated with inflammatory markers in trauma patients, J. Trauma 68 (1) (2010 January) 57e61. [21] P.S. Choban, L.J. Weireter Jr., C. Maynes, Obesity and increased mortality in blunt trauma, J. Trauma 31 (1991) 1253e1257.
[22] B.L. Zarzaur, S.W. Marshall, Motor vehicle crashes obesity and seat belt use: a deadly combination? J. Trauma 64 (2008) 412e419. [23] D.C. Evans, S.P. Stawicki, H.T. Davido, D. Eiferman, Obesity in trauma patients: correlations of body mass index with outcomes, injury patterns, and complications, Am. Surg. 77 (8) (2011 Aug) 1003e1008.
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
Utilizing quantitative measures of visceral adiposity in evaluating trauma patient outcomes Salvatore Docimo Jr. *, Brooke Lamparello, Melissa Fay Cohen, Anthony Kopatsis, Fausto Vinces NYU Lutheran Medical Center, Department of Surgery, 150 55th Street, Brooklyn, NY 11220, USA
h i g h l i g h t s Quantitative radiologic measures of visceral adiposity is the gold standard to assess obesity. This is the first attempt to associate V/S ratios with trauma outcome measurements. Clinical association of an increased V/S ratio and complications in trauma patients noted. Further prospective studies are required for further analysis.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 March 2015 Received in revised form 11 May 2015 Accepted 20 June 2015 Available online 9 July 2015
Introduction: Body mass index (BMI) has commonly been used as a parameter to assess obesity in trauma patients. However, the variability of height and weight data in trauma patients limits the use of BMI as an accurate assessment tool in the trauma population. Quantitative radiologic measurements of visceral adiposity is an accurate method for assessing obesity in patients but requires further analysis before it can be accepted as a measurement tool for trauma patients. Methods: A retrospective review of trauma cases with pre-operative CT scan from 2008 to 2015 produced 57 patients for evaluation. Preoperative BMI was calculated using measured height (m2) and weight (kg). Radiologic measurements of adiposity were obtained from preoperative CT scans using OsiriX DICOM viewer software. Visceral fat areas (VFA) and subcutaneous fat areas (SFA) were measured from a single axial slice at the level of L4-L5 intervertebral space. Results: No statistically significant results were found relating visceral fat:subcutaneous fat ratios to length of stay or post-operative complications. Initial clinical observations noting an increased incidence of complications among patients with a V/S 0.4 demonstrates a possible link between obesity and poor outcomes in trauma patients. A statistically significant correlation was noted between length of stay, peri-nephric fat and injury severity score. Discussion and Conclusion: Our pilot study should be viewed as the foundation for a larger prospective study, utilizing quantitative measurements of visceral adiposity to assess outcomes in trauma patients. © 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
Keywords: Visceral adiposity Quantitative measures Adiposity Trauma outcomes
1. Introduction Obesity in the United States has become a significant health problem. In a 2014 JAMA publication, the prevalence of obesity in the United States was estimated to be 34.9% in 2011e2012 [1]. To
* Corresponding author. Present address: Penn State Milton S. Hershey Medical Center, Department of Minimally Invasive & Bariatric Surgery, 500 University Dr, Hershey, PA 17033, USA. E-mail address: [email protected] (S. Docimo).
date, the association between obesity and patient outcomes in the setting of trauma is still under investigation. Previous studies have demonstrated higher rates of in-hospital complications and increased mortality in obese patients suffering blunt traumatic injury [2e4]. Glance et al. [5] performed a retrospective review of nearly 150,000 patients, which after adjusting for injury severity and other factors, found severely obese trauma patients were at least 30% more likely to die and twice as likely to have major complications compared with non-obese patients. Body mass index (BMI) has commonly been used in the literature as a parameter to assess obesity in trauma patients [3e5]. The
http://dx.doi.org/10.1016/j.ijsu.2015.06.069 1743-9191/© 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
52
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
previously published literature demonstrating obesity as an independent risk factor of mortality in trauma patients utilized only BMI as their standard for comparison. However, variability in the definition of obesity in previously published studies presents a difficult problem in standardizing the assessment of obesity and its affect on patient outcomes. The National Institutes of Health [7] define obesity as a BMI 30 kg/m2. Previous studies have applied their own definitions and interpretations of obesity, which can lead to findings that may not be easily extrapolated to other trauma populations. Smith-Choban et al. [8] classified patients as normal (BMI < 27 kg/m2), over-weight (BMI 27e31 kg/m2), and obese (BMI > 31 kg/m2). Byrnes et al. [9] chose to define obesity for their patients as a BMI 35 kg/m2. Such variability in obesity definitions can create a barrier to consistent extrapolation of data. However, quantitative radiologic analysis of visceral adiposity using standard CT scans have been described as being the goldstandard method for assessing visceral adiposity [10,11]. Quantitative analysis of visceral adiposity using CT-based measurements has been evaluated as a more effective means of assessing obesity in patients [10,11]. In some disease processes, such as rectal cancer, visceral adiposity has shown to have stronger associations with cancer recurrence than BMI [12]. Malietzis et al. [13] also utilized CT-based measurements of body composition to demonstrate a link between visceral obesity, reduced skeletal muscle, and poorer short-term recovery, poorer oncological outcomes, and poorer survival. To our knowledge, this study is the first to apply quantitative radiologic analysis of visceral adiposity to trauma patients undergoing surgical exploration to evaluate complication rates, such as anastomotic leaks and pulmonary embolism, and outcomes, such as length of stay. The objective of this pilot study is to introduce quantitative measures of visceral adiposity as a tool to measure the effect of obesity on the post-operative morbidity and mortality of trauma patients undergoing surgical intervention for abdominal trauma. 2. Methods A retrospective review of trauma cases from 2008 to 2015 was performed with institutional review board approval. All trauma patients over the age of 18 years of age with abdominal trauma who underwent surgical repair were examined. Those patients who had a computed tomography scan performed after arriving in the trauma bay was included in the study. Data was collected on patient demographics, medical co-morbidities, preoperative vital signs, operative findings, complications and length of stay. 2.1. Systematic data retrieval Preoperative BMI was calculated using measured height (m2) and weight (kg). Radiologic measurements of adiposity were obtained from preoperative CT scans using OsiriX DICOM viewer software. Visceral fat areas (VFA) and subcutaneous fat areas (SFA) were measured from a single axial slice at the level of L4-L5 intervertebral space. The CT attenuation level to delineate the regions of adipose tissue was set using Hounsfield units of 190 to 30 [12]. The VFA and the SFA were then delineated and measured. Fig. 1 provides an example of the quantitative visceral adiposity measurements taken during the study. The visceral fat area to subcutaneous fat area ratio (V/S) was calculated using the VFA and SFA measurements. Elevated V/S ratio indicates a larger amount of visceral fat compared to subcutaneous fat with a defined obesity threshold of V/S 0.4 [12,14]. Linear perinephric fat (PNF) thickness was measured at the level of the renal veins. Perinephric fat thickness was defined as the shortest distance (in mm) between
the kidney and abdominal wall at the level of the renal vein [12,15]. 2.2. Statistical analysis Adiposity variables (PNF, SFA, VFA) were measured as continuous variables. Linear correlations were calculated using SPSS between various adiposity variables and endpoints (LOS, complications, intra-operative pressors, etc). The chi-square, Fisher's exact tests, and student T-tests were utilized to evaluate the differences in adiposity variables and endpoints. We attempted to eliminate bias in the study by having one individual, who was the not the primary investigator, to gather the visceral and subcutaneous measurements. Furthermore, the individuals who were enrolled in the study were randomly generated from our trauma registry and not specifically chosen by any of the investigators. 3. Results 3.1. Demographics and clinical features The study population consisted of 57 patients, who underwent surgical intervention for abdominal trauma, which was confirmed on preoperative abdominal CT scans. There were more men (n ¼ 49) than women (n ¼ 8) with the mean age of 36.77 ± 16.83 years. The frequency of comorbidities such as diabetes, hypertension, and hypercholesterolemia among this population was 10.5%, 19.3% and 3.5%. Thirty-seven (65%) patients underwent exploratory laparotomy and 20 (35%) underwent laparoscopic surgical repair. Four (9%) patients presented with hypotension in the trauma bay. Eleven (19.3%) patients experienced hypotension in the operating room with seven (12.2%) and 10 (17.5%) patients requiring pressor support and colloids. Table 1 summarizes the demographic and comorbidity findings. 3.2. Adiposity measurements The patients demonstrated a mean BMI of 30.0 kg/m2, a mean perinephric fat thickness of 1.34 cm, a mean subcutaneous fat area of 142.64, a mean visceral fat area of 91.25, and a mean V/S ratio of 0.89. Mean PNF was higher in females than in males (1.56 vs 1.31; p ¼ 0.63), higher in patients without hypertension (1.37 vs 1.26; p ¼ 0.80) and higher in patients without DM (1.37 vs 1.17; p ¼ 0.73). A BMI >30 kg/m2 was noted only in male patients with an average BMI of 37.23 kg/m2 A V/S ratio 0.4 (indicating obesity) was more common in males (n ¼ 35) vs. females (n ¼ 7) who had an average V/S ratio of 0.91 compared to a V/S ratio of 0.75 in females (p ¼ 0.66). There was no statistical significance regarding a V/ S > 0.4 and patients having HTN or no HTN (0.87 vs 1.17; p ¼ 0.46) or DMand no DM (1.10 vs 1.22; p ¼ 0.85). Table 2 summarizes the obesity measurement findings for the 57 patients. 3.3. Overall analysis The mean injury severity score (ISS) 28.4 ± 12.2 and was not correlated with PNF (p ¼ 0.55), V/S ratio (p ¼ 0.86) or BMI (p ¼ 0.60), indicating injuries were well randomized. The mean LOS was 9.2 ± 10.4 days. The LOS did demonstrate a correlation to PNF (p ¼ 0.031) and ISS (p ¼ 0.01). The LOS did not correlate with subcutaneous fat thickness (p ¼ 0.79), visceral fat (p ¼ 0.87), BMI (p ¼ 0.73), and V/S ratio (p ¼ 0.86). The incidence of complications was not correlated with PNF (p ¼ 0.27), subcutaneous fat area (p ¼ 0.10), visceral fat area (p ¼ 0.11), or V/S ratio (p ¼ 0.34). The likelihood of a correlation between V/S ratio and intraoperative hypotension was not statistically significant (p ¼ 0.93). The likelihood of a correlation between V/S ratio and postoperative
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
53
Fig. 1. Quantitative measurements of visceral adiposity. Measurement of pernephric fat thickness at the level of the renal vein (A). Measurement of the subcutaneous fat area (B). Measurement of the visceral fat area (C). The V/S ratio is calculated by dividing the measurement of C by the measurement of B. Subcutaneous and visceral fat area measurements are calculated at the level of the renal vein.
S 0.4 had an incidence of post-operative complication, whereas, two patients with a post-operative complication had a V/S ratio <0.4 (p ¼ 0.29).
Table 1 Demographics and Co-Morbidities. N (%)
3.4. Data stratification
Sex Male Female Race Caucasian Hispanic African-American Asian Unknown Comorbidities Diabetes Hypertension Hypercholesterolemia
49 (86%) 8 (14%) 7 (12.2%) 11 (19.3%) 3 (5.2%) 4 (7.0%) 32 (56.1%) 6 (10.5%) 11 (19.3%) 2 (3.5%)
Table 2 Body mass index and visceral adiposity measurements overall and stratified by male and female.
BMI SFA VFA V/S
Overall (n ¼ 57)
Males (n ¼ 49)
Females (n ¼ 8)
30.0 149.3 99.7 0.75
30.6 155.5 99.8 0.91
21.5 64.0 38.5 0.75
hypotension was not statistically significant (p ¼ 0.49). The likelihood of a correlation between BMI and intraoperative hypotension was not statistically significant (p ¼ 0.93). The likelihood of a correlation between BMI and postoperative hypotension was not statistically significant (p ¼ 0.36). The V/S ratio was not found to correlate with the use of intraoperative pressors (p ¼ 0.57). No significant correlation between V/S ratio >0.4 and postoperative complications was found. Fifteen patients with a V/
The 57 patients were further stratified into two groups: patients with a V/S ratio 0.4 (n ¼ 42, 73.7%) and a V/S ratio 0.4 (n ¼ 15, 26.3%). Student t-tests between the V/S 0.4 and V/S 0.4 groups did not demonstrate any significant findings in terms of LOS (p ¼ 0.38), post-operative complications (p ¼ 0.10), and ISS (p ¼ 0.27). Table 3 summarizes the findings between the V/S ratio groups and the BMI groups. The 57 patients were further stratified into two groups: patients with a BMI 30 kg/m2 (n ¼ 20, 35.0%) and a BMI < 30 kg/m2 (n ¼ 37, 65.0%). Student t-tests between the BMI 30 kg/m2and BMI <30 kg/m2 groups did not demonstrate any significant findings in terms of LOS (p ¼ 0.88), post-operative complications (p ¼ 0.26), and ISS (p ¼ 0.45). Fig. 2 graphically summarizes the findings of LOS, incidence of complications, and average BMI based upon the V/ S ratio and BMI groups. The 57 patients were further stratified into two groups: patients who underwent open surgical repair (n ¼ 36, 63.1%) and those who underwent laparoscopic repairs (n ¼ 21, 36.8%). Comparing the length of stay between the two groups did not present any statistically significant findings (p ¼ 0.10). Comparing the incidence of complications between the two groups did not present any statistically significant findings (p ¼ 0.98). The ISS between the two groups was similar and did not demonstrate any significant differences (p ¼ 0.11). 4. Discussion Using BMI, one-third of the United States population is obese.1
54
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
Table 3 Study characteristics stratified by body mass index and V/S ratio. Body mass index
Age Sex Male Female Comorbidities Hypertension Hypercholesterolemia Diabetes Intra-operative Hypotension Pressor use Colloid use Post-operative Hypotension Pressor use Colloid use Length of stay Complications Injury severity score
V/S ratio
<30 (n ¼ 36)
30 (n ¼ 21)
V/S < 0.4 (n ¼ 15)
V/S 0.4 (n ¼ 42)
38.5
35.1
27.8
39.9
27 (75%) 6 (25%)
21 (100%) 0
14 (93.3%) 1 (6.7%)
35 (83.3%) 7 (16.7%)
7 (19.4) 0 4 (11.1%)
3(14.3%) 2 (9.5%) 1 (4.7%)
2 (13.3%) 0 3 (20.0%)
8 (19%) 2 (4.7%) 3 (7.1%)
6 (16.6%) 5 (13.8%) 6 (16.6%)
3 (14.3%) 1 (4.7%) 3 (14.3%)
3 (20.0%) 1 (6.7%) 4 (26.6%)
7 (16.6%) 5 (11.9%) 6 (14.3%)
6 (16.6%) 5 (13.8%) 3 (8.3%) 9.96 13 (36.1%) 28.91
5 (23.8%) 1 (4.7%) 3 (14.3%) 9.17 4 (19.0%) 29.01
3 (20.0%) 1 (6.7%) 4 (26.6%) 11.26 2 (13.3%) 30.66
8 (19.0%) 6 (14.3%) 7 (16.6%) 8.5 15 (35.7%) 26.68
Fig. 2. Length of stay, incidence of complications, and average body mass index stratified by V/S ratio and body mass index.
In our study, 18 (32%) of the patients were found to be obese using a BMI of 30 kg/m.2 or more. According to the V/S ratio (with a V/ S 0.4 being obese), 43 (79.4%) of patients were classified as obese. This finding may indicate the V/S ratio to be a more sensitive indicator of obesity. Table 1 provides a summary of demographics. Obesity clinically presents as a metabolic syndrome involving dyslipidemia, hypertension, insulin resistance, prothrombotic states, and pro-inflammatory states, which have been linked to worse outcomes following severe injury [16]. The systemic proinflammatory response state in obese trauma patients is due to abundant adipose tissue which releases elevated amounts of inflammatory mediators, including tumor necrosis factor-a (TNF-a), interleukin-6 (IL-6), and leptin [12,17,18]. Shashaty et al. [19] also implicated abdominal adiposity as an independent risk factor for the development of acute kidney injury (AKI) in critically ill trauma patients. Several mechanisms have been theorized to link obesity to AKI, such as, a potentiation of inflammatory mediators (interleukin-6, tumor necrosis factor-a, monocyte chemoattractant protein-1) by adipose tissue and subclinical chronic kidney disease due to an abnormal glomerular architecture seen in morbidly obese patients [19]. However, controversy exists as to whether the cause of increased morbidity and mortality is due to size (a mechanical
disadvantage) rather than an increased pro-inflammatory state [20e22]. The hypothesis of this study stated that obese trauma patients, defined as having a V/S 0.4, would have worse clinical outcomes, such as a longer length of stay (LOS)or incidence of complications. In order to eliminate the concern that injury severity could be cause of increased length of stay or incidence of complications, we calculated the injury severity score among each stratified group of patients before statistical analysis was performed. In order to attempt to control for this variability in injury severity, the ISS was calculated. The ISS was variable and did not correlate statistically with BMI or V/S ratios, demonstrating the randomization of injury among the patients, and eliminating the argument of injury severity as the root cause of any increase in LOS of complication rate. For the sample of 57 patients, the mean LOS was 9.2 ± 10.4 days with an ISS of 28.4 ± 12.2 [20]. The LOS was not correlated with subcutaneous fat thickness (0.79), visceral fat (0.87), BMI ((0.73), and V/S ratio (0.86). The LOS did demonstrate a correlation to PNF (p ¼ 0.031) and ISS (p ¼ 0.01). An elevated ISS signifies more traumatic injuries, which one would expect to translate to a longer LOS. The fact that PNF did correlate with LOS suggests varying distribution of visceral fact can affect the types of traumatic injuries patients' experience. A previous study by Evans et al. [23] did demonstrate a higher BMI to be associated with a higher incidence of torso injury and proximal upper extremity injuries in blunt trauma. The incidence of post-operative complications was also not correlated with perinephric fat thickness (0.27), subcutaneous fat area (0.10), visceral fat area (0.11), or V/S ratio (0.34). Furthermore, the incidence of post-operative hypotension and pressor use was not significantly correlated to PNF (p ¼ 0.812, p ¼ 0.17), and V/S ratio (p ¼ 0.7, p ¼ 0.61). When comparing the LOS in patients with a V/S ratio 0.4 and <0.4, there were no statistically significant differences (p ¼ 0.38). Comparing the incidence of complications, there were no statistically significant findings between these two groups (p ¼ 0.1). It must be noted 15 (37%) patients with a V/S 0.4 had an incidence of post-operative complication, whereas, two (15%) patients with a post-operative complication had a V/S ratio < 0.4 (p ¼ 0.29). Clinically, there may be a link between an elevated V/S ratio (V/S 0.4) and an increased incidence of complications. However, a small sample size likely has hampered our ability to obtain statistically significant results. The ISS between the two groups were not
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
statistically significant and less likely, compared to the increased V/ S ratio, the reason for the differences seen in the complication rates. Further stratification of the data into two groups, patients with a BMI 30 and a BMI <30, did not replicate significant findings that were noted in previous articles. The differences in LOS (p ¼ 0.88) and incidence of complications (p ¼ 0.26) were not found to be statistically significant. The ISS score between the two groups was not statistically significant. Further stratification into open surgical repair and laparoscopic repair also failed to demonstrate any significant differences in the LOS and incidence of complications. Table 3 summarizes the findings of the study characteristics stratified by BMI and V/S Ratio. 4.1. Limitations A significant limitation of this study is the small sample size. Previously published literature also demonstrated the difficulty of capturing BMI parameters, such as height and weight, as exemplified by small sample sizes. Previous studies have reported the need to exclude significant amounts of patients due to height and weight measurements not being available [8,9]. Our small sample size stems from the fact we attempted to obtain pre-operative CT scans in patients with abdominal trauma, majority being penetrating. In the majority of cases with penetrating abdominal trauma, the patient is taken to the operating room for exploration in lieu of a pre-operative CT scan. This fact alone limits our sample size. We also cannot accurately estimate the number of patients who were excluded from the study due to missing height data, weight data and also a preoperatively CT scan. Though our study did not find any statistically significant findings in regards to the V/ S ration, the fact that PNF and ISS did correlate significantly to LOS suggests visceral fat may play a role in altering the types and significance of traumatic injuries. Therefore, the use of visceral adiposity as a measurement of obesity in the setting of trauma needs to be further explored. 4.2. External validity Currently, the results of our study cannot be generalized to the trauma population due to the lack of statistically significant findings. However, the data regarding the increased incidence of complication in patients with higher V/S ratios, as well as the statically significant correlation between LOS, PNF and ISS is promising. Further prospective studies with larger sample sizes are needed in order to obtain statistically significant results. If such findings are discovered, traumatologists and surgical intensivists need to become aware of the increased risk of complications and morbidity associated with obese trauma patients. 5. Conclusions The variability of height and weight data in trauma patients limits the use of BMI as an accurate assessment tool in the trauma population. Quantitative measurements of visceral adiposity is an accurate method for assessing obesity in patients but requires further analysis before it can be accepted as a measurement tool for trauma patients. The initial clinical observations noting an increased incidence of complications among patients with a V/ S 0.4 demonstrates a possibly link between obesity and poor outcomes in trauma patients. The statically significant correlation between LOS, PNF and ISS is promising and should act as a catalyst for further evaluation of the role of obesity in trauma. Our study should be viewed as the foundation for a large prospective study, utilizing quantitative measurements of visceral adiposity to assess outcomes in trauma patients.
55
Ethical approval Not required. Funding No sources of funding were needed for this project. Author contribution Docimo: study design, data analysis, writing. Lamparello: data collection, data analysis. Cohen: data collection. Kopatsis: data analysis. Vinces: data analysis. Conflict of interest statement All authors have no conflicts of interest to state. Guarantor Dr. Salvatore Docimo, Jr. is the guarantor. References [1] C.L. Ogden, M.D. Carroll, B.K. Kit, K.M. Flegal, Prevalence of childhood and adult obesity in the United States, 2011-2012, JAMA 311 (8) (2014) 806e814. [2] B.R. Boulanger, D.P. Milzman, A. Rodriguez, Obesity, Crit. Care Clin. 10 (1994) 613e622. [3] A. Neville, C.V. Brown, J. Weng, D. Demetriades, G. Velmahos, Obesity is an independent risk factor for mortality in severely injured blunt trauma patients, Arch. Surg. 139 (2004) 983e987. [4] Brown CV, Neville A, Rhee P, Salim A, Velmahos G, Demetriades D. The impact of obesity on the [5] L.G. Glance, Y. Li, T.M. Osler, D.B. Mukamel, A.W. Dick, Impact of obesity on mortality and complications in trauma patients, Ann. Surg. 259 (2014) 576e581. [7] National Institutes of Health, Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report, National Institutes of Health, Washington, DC, 1998, pp. 94e4083. Publication. [8] P. Smith-Choban, L.J. Weireter, C. Maynes, Obesity and increase mortality in blunt trauma, J. Trauma 31 (1991) 1253e1257. [9] M.C. Byrnes, M.D. McDaniel, M.B. Moore, S.D. Helmer, R.S. Smith, The effect of obesity on outcomes among injured patients, J. Trauma 58 (2005) 232e237. [10] A.M. Gradmark, A. Rydh, F. Renstrom, E. De Lucia-Rolfe, A. Sleigh, P. Nordstrom, S. Brage, P.W. Franks, Computed tomography-based validation of abdominal adiposity measurements from ultrasonography, dual-energy Xray absorptiometry and anthropometry, Br. J. Nutr. 104 (2010) 582e588. [11] K.M. Pou, J.M. Massaro, U. Hoffman, K. Lieb, R.S. Vasan, C.J. O'Donnell, C.S. Fox, Patterns of abdominal fat distribution: the Framingham Hear Study, Diabetes Care 32 (2009) 481e485. [12] W. Clark, W.M. Siegel, Y.A. Chen, X. Zhao, C.M. Parsons, J.M. Hernandez, J. Weber, S. Thareja, J. Choi, D. Shibata, Quantitative measures of visceral adiposity and body mass index in predicting rectal cancer outcomes after neoadjuvantchemoradiation, J. Am. Coll. Surg. 216 (2013) 1070e1081. [13] G. Malietzis, O. Aziz, N.M. Bagnall, N. Johns, K.C. Fearon, J.T. Jenkins, The role of body composition evaluation by computerized tomography in determining colorectal cancer treatment outcomes: a systemic review, Eur. J. Surg. Oncol. 41 (2) (2015 Feb) 186e189. [14] Y. Matsuzawa, T. Nakamura, I. Shimomura, K. Kotani, Visceral fat accumulation and cardiovascular disease, Obes. Res. 3 (1995) 645Se647S. [15] K. Morris, S. Tuorto, M. Gonen, et al., Simple measurement of intra-abdominal fat for abdominal surgery outcome prediction, Arch. Surg. 145 (2010) 1069e1073. [16] M.E. Falagas, M. Kompoti, Obesity and infection, Lancet Infect. Dis. 6 (2006) 438e446. [17] C.J. Lyon, R.E. Law, W.A. Hsueh, Mini review: adiposity, inflammation and thermogenesis, Endocrinology 144 (2003) 2195e2200. [18] F.A. Moore, A. Sauaia, E.E. Moore, J.B. Haenel, J.M. Burch, D.C. Lezotte, Postinjury multiple organ failure: a bimodal phenomenon, J. Trauma 40 (4) (1996 Apr) 501e510. [19] M.G. Shashaty, E. Kalkan, S.L. Bellamy, J.P. Reilly, D.N. Holena, K. Cummins, P.N. Lanken, H.I. Feldman, M.P. Reilly, J.K. Udupa, J.D. Christie, Computed tomography-defined abdominal adiposity is associated with acute kidney injury in critically ill trauma patients, Crit. Care Med. 42 (7) (2014)
56
S. Docimo Jr. et al. / International Journal of Surgery 21 (2015) 51e56
1619e1628. [20] B. Collier, L. Dossett, J. Shipman, M. Day, G. Lawson, R. Sawyer, A. May, Visceral adiposity is not associated with inflammatory markers in trauma patients, J. Trauma 68 (1) (2010 January) 57e61. [21] P.S. Choban, L.J. Weireter Jr., C. Maynes, Obesity and increased mortality in blunt trauma, J. Trauma 31 (1991) 1253e1257.
[22] B.L. Zarzaur, S.W. Marshall, Motor vehicle crashes obesity and seat belt use: a deadly combination? J. Trauma 64 (2008) 412e419. [23] D.C. Evans, S.P. Stawicki, H.T. Davido, D. Eiferman, Obesity in trauma patients: correlations of body mass index with outcomes, injury patterns, and complications, Am. Surg. 77 (8) (2011 Aug) 1003e1008.