key: cord-0019804-bdf5oyhm
authors: Hakam, Nizar; Nabavizadeh, Behnam; Sadighian, Michael J.; Holler, Jordan; Shibley, Patrick; Li, Kevin D.; Low, Patrick; Amend, Gregory; Stein, Deborah M.; Breyer, Benjamin N.
title: The Impact of Obesity on Renal Trauma Outcome: An Analysis of the National Trauma Data Bank from 2013 to 2016
date: 2021-08-09
journal: World J Surg
DOI: 10.1007/s00268-021-06275-1
sha: 4e8ba0768492aac3e133d88aae22b4385f2465c8
doc_id: 19804
cord_uid: bdf5oyhm

BACKGROUND: The obesity paradox has been recently demonstrated in trauma patients, where improved survival was associated with overweight and obese patients compared to patients with normal weight, despite increased morbidity. Little is known whether this effect is mediated by lower injury severity. We aim to explore the association between body mass index (BMI) and renal trauma injury grade, morbidity, and in-hospital mortality. METHODS: A retrospective cohort of adults with renal trauma was conducted using 2013–2016 National Trauma Data Bank. Multiple regression analyses were used to assess outcomes of interest across BMI categories with normal weight as reference, while adjusting for relevant covariates including kidney injury grade. RESULTS: We analyzed 15181 renal injuries. Increasing BMI above normal progressively decreased the risk of high-grade renal trauma (HGRT). Subgroup analysis showed that this relationship was maintained in blunt injury, but there was no association in penetrating injury. Overweight (OR 1.02, CI 0.83–1.25, p = 0.841), class I (OR 0.92, CI 0.71–1.19, p = 0.524), and class II (OR 1.38, CI 0.99–1.91, p = 0.053) obesity were not protective against mortality, whereas class III obesity (OR 1.46, CI 1.03–2.06, p = 0.034) increased mortality odds. Increasing BMI by category was associated with a stepwise increase in odds of acute kidney injury, cardiovascular events, total hospital length of stay (LOS), intensive care unit LOS, and ventilator days. CONCLUSIONS: Increasing BMI was associated with decreased risk of HGRT in blunt trauma. Overweight and obesity were associated with increased morbidity but not with a protective effect on mortality. The obesity paradox does not exist in kidney trauma when injury grade is accounted for.

The kidney is the third most commonly injured organ in abdominal trauma and the second most commonly injured in penetrating trauma [1] . Traumatic renal injury is associated with in-hospital complications and mortality in 33% and 14% of cases, respectively [2] . Studying the mechanics of renal injury could lead to improved trauma therapeutics, safety equipment, and outcomes.

The obesity paradox is the apparent association of increased survival in overweight and class I obese patients compared to normal weight patients, despite higher morbidity [3] . Originally established in cancer [4] [5] [6] , cardiovascular [7, 8] , and other chronic diseases [9, 10] , the obesity paradox has been recently observed in trauma patients in two large cohorts [3, 11] . Although several hypotheses have been proposed to explain the obesity paradox, it is unknown whether the protective effect of higher body mass index (BMI) is caused by less severe injuries or an increased injury tolerance. This analysis is limited, at least in part, by the difficulty in comparing trauma severity across multiple organs. The American Association for the Surgery of Trauma (AAST) has developed an anatomical-based severity grading system for providers to communicate renal injury which can objectively assess renal trauma severity across potential groups of interest [12] .

We sought to examine the relationship between BMI categories as defined by the World Health Organization (WHO) and outcomes of injury grade and in-hospital mortality in patients with kidney trauma using a large dataset. Understanding the role of BMI in trauma morbidity and mortality, in addition to the mechanism by which its effects are mediated, aids risk stratification and clinical decision making. We hypothesize that being overweight and obese does not improve survival when controlling for AAST kidney injury grade.

Data source and study design

The National Trauma Data BankÒ (NTDB) is a large US and Canadian trauma registry that includes comprehensive injury data [13] . We performed a retrospective cohort using NTDB data between 2013 and 2016, and following the STROBE guidelines for observational studies [14] . Institutional review board exemption was provided given that all data were de-identified.

We identified patients age 18 [12, 15, 16] . We excluded patients that did not map to a specific AAST grade (n = 23,931) and patients who died in the field or in the emergency department (ED) (n = 846). BMI was classified according to the WHO categories (underweight \18.5; normal weight 18.5-24.9; overweight 25.0-29.9; class I obesity 30.0-34.9; class II obesity 35.0-39.9; class III obesity C 40). We excluded patients with incomplete data (n = 2646) and outliers of weight and height using cutoff points from published literature (weight \30 kg or [ 600 kg; height \ 80 cm or [ 250 cm, n = 562) [3] . For each case, we collected: age, sex, race, mechanism of injury (blunt vs penetrating), Injury Severity Score (ISS), pulse at arrival to ED, hypotension (systolic blood pressure \ 90) at arrival to ED, initial Glasgow Coma Scale (GCS), trauma center level, intensive care unit (ICU) admission, ventilator use, blood transfusion, angioembolization use, length of stay (LOS), ICU LOS, ventilator days, comorbidities, complications, and hospital disposition. Low-grade renal trauma (LGRT) was defined as AAST grades I-II, whereas high-grade renal trauma (HGRT) was AAST grades III-V. To identify if a patient underwent surgery, we used ICD-9 codes (55.51; 55.52; 55.53; 55.54; 55.4) and ICD-10 codes (0TT00ZZ; 0TT04ZZ; 0TT10ZZ; 0TT14ZZ; 0TT20ZZ; 0TT24ZZ; 0TT30ZZ; 0TT40ZZ; 0TB10ZZ; 0TB00ZZ; 1PC89LB) for nephrectomy and ICD-9 codes (55.81; 55.86; 55.87; 55.89) and ICD-10 codes (0TQ00ZZ; 0TQ03ZZ; 0TQ04ZZ; 0TQ07ZZ; 0TQ08ZZ, 0TQ10ZZ; 0TQ13ZZ; 0TQ14ZZ; 0TQ17ZZ; 0TQ18ZZ; 0TQ30ZZ; 0TQ33ZZ; 0TQ34ZZ; 0TQ37ZZ; 0TQ38ZZ; 0TQ40ZZ; 0TQ43ZZ; 0TQ44ZZ; 0TQ47ZZ; 0TQ48ZZ) for nephrorrhaphy.

Descriptive statistics were reported as frequencies and percentages. Continuous variables were assessed for normality and reported as medians and interquartile ranges (IQR) or means and standard deviations (SD) accordingly. Mann-Whitney U or student-t test and Chi-square tests were used to compare continuous and categorical variables, respectively.

Multiple regression analysis was done to examine the association between BMI categories and HGRT with a priori adjustment for age, sex, mechanism of injury, and ISS. This was done using Poisson regression with robust SE reporting, since outcome (HGRT) was a common event (52%) and logistic regression would overestimate the risk ratio (RR) [17] . Subgroup analysis by mechanism of injury (blunt vs penetrating) was performed. Penetrating injuries were further classified as gunshot vs stab wounds. A sensitivity analysis performed by defining HGRT as AAST grades IV-V exclusively did not yield any material changes in our findings.

We used logistic regression to assess the adjusted odds ratios (OR) for inpatient mortality across BMI categories. Following published guidelines on studying mortality using NTDB, we controlled for age, sex, anatomical severity (ISS, AAST grade, initial GCS), physiological severity (hypotension, pulse, transfusion), mechanism of injury, presence of traumatic brain injury, and trauma center level [18] . The final model also controlled for comorbidities (bleeding disorder, cirrhosis, and disseminated cancer), and undergoing any intervention (nephrectomy, nephrorrhaphy, or angioembolization), chosen according to their significance levels (p \ 0.1 in backward selection).

For our secondary analysis, we fit multiple logistic regression models using backward stepwise selection to assess the association between BMI categories and the odds of undergoing any intervention, nephrectomy, nephrorrhaphy, or angioembolization. We also used logistic regression to assess the adjusted OR of developing acute kidney injury (AKI) or cardiovascular events (composite outcome defined as myocardial infarction, cardiac arrest, pulmonary embolism, or cerebrovascular accident/stroke due to low outcome rate of individual components) across BMI categories. Covariates for AKI and cardiovascular events were chosen based on literature review and included age, ISS, GCS, hypotension, pulse, mechanism of injury, AAST grade, nephrectomy, sepsis, and comorbidities (alcohol abuse, hypertension requiring medication, congestive heart failure, diabetes mellitus, chronic kidney disease, previous myocardial infarction, angina within the past 30 days, history of stroke) [19] [20] [21] . Finally, we assessed the following continuous variables across BMI categories using multiple linear regression models for patients who survived to discharge: LOS, ICU LOS, and ventilator days. Models' fit was assessed using the Pearson goodness-of-fit test.

Due to limited ability to adjust for other organ injuries while assessing mortality and secondary outcomes, we performed a sensitivity analysis by considering only patients with isolated renal trauma. All statistical analysis was performed using StataÒ version 16.1 and with a p \ 0.05 considered significant.

Our final cohort consisted of 15,181 renal trauma patients. Median age was 33 years (IQR 24-52), and 11,450 (75.4%) were males. The most common BMI category was normal weight (35.2%), followed by overweight (32.7%), class I obesity (17.1%), class II obesity (6.8%), class III obesity (5.6%), and underweight (2.4%) ( Table 1) .

Overweight and obese patients had significantly lower proportions of HGRT (between 40.5 and 52.3%) compared to normal weight (56.9%) ( Table 1 ). In the adjusted analysis, increasing BMI above normal was associated with a progressive decrease in risk of HGRT ( Fig. 1 ) compared to normal weight (overweight: RR 0.94, CI 0.90-0.97, p \ 0.001; class I obesity: RR 0.86, CI 0.82-0.89, p \ 0.001; class II obesity: RR 0.86, CI 0.81-0.92, p \ 0.001; class III obesity: RR 0.75, CI 0.69-0.82, p \ 0.001). Underweight was not associated with risk of HGRT (RR 1.02, CI 0.93-1.11, p = 0.714). In subgroup analysis by mechanism of injury, the relationship between BMI and HGRT remained in blunt trauma but not in penetrating trauma (Table 2 ). Further subclassification of penetrating trauma did not demonstrate any association in gunshot or stab wounds (''Appendix in Table 5 '').

There were 817 (5.38%) in-hospital mortalities within the cohort. Stratified by BMI, mortality was lowest in individuals with normal weight (4.5%) and highest in class III obese (7%) ( Table 1) . After adjusting for covariates (including AAST grade), overweight, class I, and class II (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) 0.157 20.5 (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) 21 (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) 0.024 21 (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) 0.733 20 (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) 0.739 21 (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) World J Surg obesity were not protective against mortality, whereas underweight and class III obesity were associated with increased odds of mortality (Table 3) .

On adjusted analysis, BMI was not associated with the odds of receiving an intervention or undergoing nephrectomy (Table 4 ). There was a stepwise increase in the odds of developing AKI with increasing BMI categories, and all obese groups had elevated odds of developing cardiovascular events compared to normal weight patients (Table 4) . Furthermore, there was a stepwise increase in total hospital LOS, ICU LOS, and ventilator days among patients who survived to discharge with each successive increase in BMI group above normal weight (Fig. 2) 

There were 1323 (8.7%) patients with isolated renal trauma. Sensitivity analysis for in-hospital mortality and secondary outcomes of morbidity was not qualitatively different from our main results when we only considered these patients (''Appendices in Table 6 and Figure 3'') ; however, the low number of patients and outcome events in Table 7 '') limited the power to obtain statistical significance in some instances.

We report on the impact of BMI on renal trauma grade and outcome. We found increasing BMI above normal to progressively decrease the risk of HGRT. This relationship applied only in blunt trauma and not in penetrating trauma. When accounting for covariates, only extremes of BMI (underweight and class III obesity) were associated with increased in-hospital mortality. Increasing BMI was not associated with the risk of undergoing any intervention, or risk of nephrectomy, but was associated with increased AKI, cardiovascular events, longer total hospital LOS, ICU LOS, and ventilator days. This study expands on the role of BMI in trauma outcomes reported in the literature. Arbabi et al. [22] studied crash injuries in relation to biomechanical factors and found that overweight patients had the least severe abdominal injury. The protection was attributed to increased insulating tissue by extra adiposity. Similar findings were demonstrated by Stein et al. where multivariate analysis showed higher BMI to be associated with lower risk of pelvic fracture in near-side lateral motor vehicle collisions [23] . The protective effect was assumed to be due to a cushion effect of increased soft tissue. Wang et al. [24] found increased subcutaneous fat depth to be associated with decreased injury severity to the abdominal region. In a population-based cohort by Schott et al., hip fracture risk was reduced by 40% for a standard deviation increase in fat mass, while changes in lean body mass did not affect fracture rates [25] . Furthermore, Fu et al. found that obese patients with blunt abdominal trauma sustained less gastrointestinal tract injuries and associated surgery [26] . These reports support the hypothesis that fat mass could aid better tolerating mechanical force incurred in trauma.

We postulate that the protective effect of higher BMI depicted in our study is due to extra adipose tissue acting as a shock absorber, thereby reducing damage to vital organs. An important distinction is that the risk of HGRT was not altered in cases of penetrating trauma. It has been hypothesized that acceleration/deceleration injuries from blunt trauma injure the kidney at natural weak points [27] . Thus, blunt trauma force could be less transmitted to those weak points if larger nearby fat surface area or thickness is present. On the other hand, injury in penetrating wounds results from an object's trajectory through renal parenchyma, and not related to diffuse pressure exerted around the kidney [27] .

In this study, patients with class III obesity had increased adjusted mortality odds, which is in line with prior literature [28] . This could be explained by increased morbidity in these patients, who had the longest hospital LOS, ICU LOS, and ventilator days. Underweight was also associated with increased mortality. Similar findings were reported in general trauma patients [3] . ICU admission is associated with a state of malnutrition; thus, depleted energy stores and poor immune function probably contribute to mortality risk in this group [29] .

Studies reporting the obesity paradox in trauma proposed several mechanisms to explain the counterintuitive relation between higher than normal BMI and improved survival. These include: 1. upregulation of cytokines produced by adipose tissue, especially leptin, which alter susceptibility to infection and toxicity of proinflammatory stimuli [30] ; 2. visceral triglyceride saturation interferes with triglyceride interaction and lipolysis, reducing lipotoxic systemic injury and organ failure [31] ; 3. presence of high nutritional reserves enable tolerance to malnutrition states associated with ICU admission [29, 32] ; 4. shortcomings of BMI in differentiating adiposity from muscle mass, highlighted by studies that found that the obesity paradox disappeared when using alternative adiposity measures (like waist-to-hip ratio) [33, 34] . We hypothesize that studies reporting obesity paradox in trauma missed adjusting for an objective and accurate measure comparing injury grade. In our present study, this was mitigated by controlling for AAST renal injury grade.

Increasing BMI demonstrated a linear relationship with AKI, LOS, ICU LOS, and ventilator days. Obese patients were also at higher risk of cardiovascular events. These patients are exposed to a chronic inflammatory state and high level of baseline pro-inflammatory cytokines which could contribute to higher morbidity [35] [36] [37] . The findings are consistent with other studies of trauma patients [3, 28, 38] . Obese patients have several respiratory complications, including reduced lung volumes, compliance, and gas exchange, leading to difficulty weaning from mechanical ventilation [39] .

Conservative management for renal trauma is increasingly popularized aiming to maximize renal salvage [40] [41] [42] . Our findings support this trend in patients with abnormal BMI, who are at greater operative risk due to the increased morbidity demonstrated. The increased mortality observed at BMI extremes does not necessarily warrant a more aggressive management approach, since the cause of death in these patients is not presumed to be directly due to their kidney trauma.

The study limitations merit mention. The NTDB has several inherent limitations as it is not a population based dataset, thus may not be representative of all trauma hospitals. Excluded cases with missing data (mostly due to cases that did not map to any AAST grade) could have introduced some selection bias. Other limitations include missing data and possible coding errors. However, the large sample size of this cohort is likely resistant to these biases. Although we controlled for many covariates in our modeling adhering to published guidelines, there could be unknown or unmeasured confounders not accounted for. The absence of laboratory-specific data hindered our ability to analyze valuable factors such as serum inflammatory markers. There was limited granularity in several variables (such as renal function data and cause of death, and detailed circumstances of injury beyond blunt vs penetrating, such as crash severity if motor vehicle collision and velocity impact) which could have further informed our analysis. Finally, there was no information on the longterm outcomes of renal trauma patients.

Increasing BMI above normal decreases the risk of HGRT in blunt trauma. The obesity paradox does not exist in kidney trauma when injury grade is accounted for. Extremes of weight in both directions increase the risk of mortality in these patients. Further prospective studies are needed to confirm these findings across different types of trauma.

See Fig. 3 .

See Tables 5, 6 Funding None.

Conflict of interest All authors declare no conflict of interests.

Informed consent Formal ethical approval was not individually sought as all data were de-identified. 

American association for the surgery of trauma organ injury scale for kidney injuries predicts nephrectomy, dialysis, and death in patients with blunt injury and nephrectomy for penetrating injuries

Operative and nonoperative management for renal trauma: comparison of outcomes. a systematic review and meta-analysis

The obesity paradox in the trauma patient: normal may not be better

Association between baseline body mass index and overall survival among patients over age 60 with acute myeloid leukemia

Postdiagnosis body mass index and risk of mortality in colorectal cancer survivors: a prospective study and meta-analysis

Biological support to obesity paradox in renal cell carcinoma: a review

Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies

Body mass index and the risk of total and cardiovascular mortality among patients with type 2 diabetes: a large prospective study in Ukraine

Association between BMI measured within a year after diagnosis of type 2 diabetes and mortality

Obesity paradox on the survival of elderly patients with diabetes: an AHAP-based study

Obese trauma patients have increased need for dialysis

Organ injury scaling: spleen, liver, and kidney

American college of surgeons: quality programs

The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies

Organ injury scaling. III: chest wall, abdominal vascular, ureter, bladder, and urethra

Renal and extrarenal predictors of nephrectomy from the national trauma data bank

The relative merits of risk ratios and odds ratios

Influence of the national trauma data bank on the study of trauma outcomes: is it time to set research best practices to further enhance its impact?

African American race, obesity, and blood product transfusion are risk factors for acute kidney injury in critically ill trauma patients

Prevalence and risk factors for acute kidney injury among trauma patients: a multicenter cohort study

Trauma induced acute kidney injury

The cushion effect

Risk factors associated with pelvic fractures sustained in motor vehicle collisions involving newer vehicles

Increased depth of subcutaneous fat is protective against abdominal injuries in motor vehicle collisions

How hip and wholebody bone mineral density predict hip fracture in elderly women: the EPIDOS prospective study

Morbid obesity's silver lining: an armor for hollow viscus in blunt abdominal trauma

Contemporary management of high-grade renal trauma: results from the american association for the surgery of trauma genitourinary trauma study

The effect of obesity on outcomes in trauma patients: a meta-analysis

Malnutrition diagnoses in hospitalized patients: United States

Adipose tissue, adipokines, and inflammation

Adipose saturation reduces lipotoxic systemic inflammation and explains the obesity paradox

Positive influence of being overweight/obese on long term survival in patients hospitalised due to acute heart failure

Abdominal obesity is associated with an increased risk of all-cause mortality in patients with HFpEF

Waist-to-hip ratio and mortality in heart failure

Acute kidney injury is associated with early cytokine changes after trauma

Obesity and recovery from acute kidney injury (Ob AKI): a prospective cohort feasibility study

Role of cytokines in cardiovascular diseases: a focus on endothelial responses to inflammation

Body mass index and outcomes in critically injured blunt trauma patients: weighing the impact

Respiratory management of perioperative obese patients

Nationwide procedural trends for renal trauma management

Successful nonoperative management of the most severe blunt renal injuries: a multicenter study of the research consortium of New England centers for trauma

Review of the current management of upper urinary tract injuries by the EAU trauma guidelines panel

Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations