Increased perioperative mortality following bariatric surgery among patients with cirrhosis.

BACKGROUND & AIMS: The prevalence of nonalcoholic fatty liver disease and ensuing cirrhosis is expected to increase as a result of the obesity epidemic. These trends might increase the number of bariatric surgeries among patients with cirrhosis. We sought to assess the impact of cirrhosis on perioperative mortality after bariatric procedures.
METHODS: Data on patients who underwent bariatric surgery in the United States between 1998 and 2007 were extracted from the Nationwide Inpatient Sample. In-hospital mortality and length of stay were compared for patients with no cirrhosis, compensated cirrhosis, and decompensated cirrhosis.
RESULTS: Patients without cirrhosis had lower mortality rates than those with compensated and decompensated cirrhosis (0.3% vs 0.9% and 16.3%, respectively, P = .0002). After adjusting for covariates, the adjusted odds ratio of mortality among compensated and decompensated cirrhotic patients compared with noncirrhotic patients was 2.17 (95% confidence interval, 1.03-4.55) and 21.2 (95% confidence interval, 5.39-82.9), respectively. Mortality increased with volume of surgery among centers; those with more than 100 surgeries per year had the lowest mortality rates, compared with those with 50 to 100 surgeries per year and fewer than 50 surgeries per year (0.2% vs 0.4% and 0.7%, respectively; P < .0001). The average length of stay was longer for patients with decompensated and compensated cirrhosis, compared with patients without liver disease (6.7 and 4.4 d vs 3.2 d, respectively; P = .0001 and P = .03).
CONCLUSIONS: Bariatric surgery in patients with cirrhosis should be performed while liver disease is well compensated. Patients with cirrhosis should undergo surgery at centers that perform large numbers of these procedures.