Diagnosis of morbid obesity may not impact healthcare utilization for orthotopic liver transplantation: A propensity matched study

AIM To study mortality, length of stay, and total charges in morbidly obese adults during index hospitalization for orthotopic liver transplantation. METHODS The Nationwide Inpatient Sample was queried to obtain demographics, healthcare utilization, post orthotopic liver transplantation (OLT) complications, and short term outcomes of OLT performed from 2003 to 2011 (n = 46509). We divided patients into those with [body mass index (BMI) >= 40] and without (BMI < 40) morbid obesity. Multivariable logistic regression analysis was performed to characterize differences in in-hospital mortality, length of stay (LOS), and charges for OLT between patients with and without morbid obesity after adjusting for significant confounders. Additionally, propensity matching was performed to further validate the results. RESULTS Of the 46509 patients who underwent OLT during the study period, 818 (1.8%) were morbidly obese. Morbidly obese recipients were more likely to be female (46.8% vs 33.4%, P = 0.002), Caucasian (75.2% vs 67.8%, P = 0.002), in the low national income quartile (32.3% vs 22.5%, P = 0.04), and have = 3 comorbidities (modified Elixhauser index; 83.9% vs 45.0%, P < 0.001). Morbidly obese patient also had an increase in procedure related hemorrhage (P = 0.028) and respiratory complications (P = 0.043). Multivariate and propensity matched analysis showed no difference in mortality (OR: 0.70; 95% CI: 0.27-1.84, P = 0.47), LOS (beta: -4.44; 95% CI: -9.93, 1.05, P = 0.11) and charges for transplantation (beta: $ 15693; 95% CI: -51622-83008, P = 0.64) between the two groups. Morbidly obese patients were more likely to have transplants on weekdays (81.7%) as compared to those without morbid obesity (75.4%, P = 0.029). CONCLUSION Morbid obesity may not impact in-hospital mortality and health care utilization in OLT recipients. However, morbidly obese patients may be selected after careful assessment of co-morbidities.