K. He1, N. Wang2, P. Zhang1,2, S. Holcombe1,2, S. Wang1,2 2International Center For Automotive Medicine,Ann Arbor, MI, USA 1University Of Michigan,Department Of Surgery,Ann Arbor, MI, USA
Introduction: Obesity has become a disease of epidemic proportions nationally and internationally, leading to significant changes in patient morphomic variability. Obese body habitus has been associated with changing motor vehicle collision (MVC) injury patterns due to greater kinetic energy and thicker subcutaneous tissue. In particular, excess posterior subcutaneous adipose tissue (e.g. buttock fat) has been shown in cadaver tests to allow increased forward excursion of the femur and lower extremity, which is hypothesized to result in increased lower extremity (LE) injuries in obese patients. While the clinical significance of LE injuries has been underestimated on injury severity scales due their low mortality risk, LE injuries have been shown to have considerable societal burden approaching that of fatality. Moreover, compared to their normal weight counterparts, overweight patients are at risk for prolonged disability and increased healthcare utilization in the rehabilitation period. In this study, we hypothesize that obese occupants experience greater risk of lower extremity injuries in frontal MVCs.
Methods: Vehicle and demographics data from 1996-2015 were extracted from the University of Michigan International Center for Automotive Medicine crash database. Morphomics data were processed from Computed Tomography scans obtained from the initial trauma evaluation. We fitted logistic regression models using crash, demographic, and morphomic variables for occupants with and without maximum abbreviated injury scale greater than 2 (MAIS2+) LE injuries. The performance of logistic regression models was assessed using the Akaike Information Criterion (AIC), and the area under the receiver operating characteristic curve (AUC). The top 100 models were selected by AIC, and the importance of each variable was calculated using weighted frequencies. Odds ratios and confidence intervals (CI) for obesity-related morphomic factors and lower extremity injury were calculated for a belted male in a 25 miles-per-hour frontal crash.
Results: 243 occupants were included in our logistic regression. We used four vehicle variables, three demographic variables, and six morphomic variables, which resulted in over 8,000 models. The final model predicting MAIS2+ included crash, demographic, and morphomics variables and resulted in an AUC of 0.807. BMI was the most important variable in our final model. The odds ratios for lower extremity injury between posterior top of spine to back skin distance (buttock fat) percentiles were (N=250): Q25 (25th percentile) and Q75: 1.7 (95% CI 1.2, 2.6); Q10 and Q90: 2.8 (95% CI 1.4, 5.9); Q5 and Q95: 4.1 (95% CI 1.5, 11); Q1 and Q99: 6.5 (95% CI 1.7, 25).
Conclusion: Overweight and obese patients are at higher risk for lower extremity injuries in frontal MVCs. Body fat morphomic variables such as the thickness of buttock fat are biomechanically significant and can be used to predict risk for lower extremity injury in frontal MVCs.