Apprising the choice of bariatric surgery in patients with compensated cirrhosis with morbid obesity: results of a national analysis.

Bariatric surgery (BS) was proved safe in carefully selected patients with compensated cirrhosis (CC). However, limited data exist on differential impact of bariatric surgery type on clinical outcomes and health care utilization. This retrospective study utilizes the 2010-2014 Nationwide Readmissions Database. We included obese adults with CC who underwent the two most commonly used BS, Roux-en-Y (RYGB) and laparoscopic sleeve gastrectomy (LSG). Those with decompensation within 6 months of BS were excluded. Rates of hepatic decompensation (new-onset ascites, variceal bleed, encephalopathy, spontaneous bacterial peritonitis, and/or hepatorenal syndrome), surgical complications, health care utilization, and mortality were compared between RYGB and LSG. Multivariable analysis was performed to fit various models. A total of 3032 patients with CC underwent BS, including 1864 (61.5%) RYGB and 1168 (38.5%) LSG. The majority (56%) of BS were performed at large, metropolitan teaching hospitals. There were no significant differences in various decompensations and surgical complications comparing RYGB to LSG. Healthcare utilization including index length of stay (RYGB: 3.4 days vs LSG: 3.0 days), 30-day readmission rate (RYGB: 9.5% vs LSG: 3.7%), and cost of admission (RYGB: $14,006 vs LSG: $12,523) were higher in RYGB (p values < 0.001). Index admission and calendar year mortality could not be analyzed due to the few number of events. Two types of bariatric surgeries in obese patients with compensated cirrhosis have similar rates of decompensated cirrhosis events and surgical complications. However, RYGB procedure incurred increased healthcare utilization. Therefore, LSG may be the preferred BS for patients with CC. Prospective, randomized studies comparing the types of BS are needed to confirm our observations.