Aim The aim of the study was to compare the effect of different dietary interventions on alanine aminotransferase (ALT) in obese patients with diabetes. Methods A post hoc analysis of an open label, parallel design, quasi-randomised (allocation by alternation), controlled trial, conducted in Israel. Obese patients with diabetes (n=259), treated in the community, were centrally allocated to one of three diets: (1) the 2003 recommended American Diabetes Association diet (ADA): 50-55% carbohydrate, 30% fat and 20% protein, n=85; (2) a low glycaemic index (LGI) diet: 50-55% LGI carbohydrate, 30% fat, 15-20% protein, n = 89; or (3) a modified Mediterranean diet (MMD): 35% LGI carbohydrate, 45% fat that was high in monounsaturated fat, 15-20% protein, n=85. ALT was measured at 6 and 12 months. Results ALT levels decreased in all arms; however, the MMD was associated with the lowest ALT levels at month 6 (n=201: ADA n=64, LGI n=73, MMD n=64) and month 12 of follow-up (n=179). At 12 months mean ALT levels were 19.8±1.4 U/l in the ADA diet arm (n=54), 18.0±1.5 U/l in the LGI diet arm (n=64) and 14.4±1.7 in the MMD arm (n=61, p<0.001). Evidence for an effect of diet on ALT levels persisted when controlling for post-randomisation changes in waist to hip ratio, BMI, homeostasis model assessment (HOMA) or triacylglycerol. Conclusions A Mediterranean diet may have a beneficial effect on liver steatosis in obese patients with diabetes. Results of trials assessing the effect of dietary composition on clinical outcomes should be awaited before a decisive conclusion can be reached. In addition to clinical outcomes, such studies should address the issue of primary prevention of steatosis in high-risk and healthy individuals.