Background: Morbid obesity is associated with non-alcoholic fatty liver disease(NAFLD), which may progress to cirrhosis in up to 25% of patients. Pre-operative liver function tests(LFTs), abdominal ultrasonography(US), and intra-operative assessments(IA) have been proposed to evaluate for NAFLD. We sought to determine the prevalence of NAFLD in patients undergoing roux-en-y gastric bypass(RYGB) and compare the liver histology to the LFTs, US, and IA to determine if a correlation exists.Methods: Patients undergoing RYGB were retrospectively reviewed in our database from January, 2000 to September, 2007. LFTs, US, and IA were compared to liver pathology classified according to the National Institute of Diabetes & Digestive & Kidney Diseases(NIDDK) and National Institute of Health (NIH) guidelines. The student’s t- and Chi-square tests were used to compare continuous and categorical variables, respectively.Results: 251 patients(38 men and 213 women with median age and BMI of 43.4+/-11.1 and 47.9+/-7.8 respectively) had routine liver biopsies performed during RYGB. Histology suggested 3% normal, 24% <5%steatosis, and 73% NAFLD. NASH(NAFLD activity score > or = 5) was present in 18% while fibrosis was present in 34%. NAFLD was equal amongst gender(82% men vs 71% women) and more prevalent in patients > or = 45 years compared to those <45 years(83% vs 65%;p=0.0005). 83% of IA suggesting fatty liver had evidence of NAFLD(p=0.001). 246 patients had LFTs while 112 had ultrasounds. LFTs were elevated in 25.6% with AST/ALT>1 in 23.6%, which did not correlate with predicting NAFLD. 89% of those with NASH correlated with AST/ALT<1(p=0.035). US detected fatty liver in 48.2%. 58% of patients with fatty liver on US had a diagnosis of NAFLD(p<0.001). Conclusion: NAFLD is prevalent among morbidly obese patients, especially those > or = 45 years undergoing RYGB. US and IA are good predictors of NAFLD as compared to pre-operative LFTs and AST/ALT ratio. NASH is associated with an AST/ALT<1.