Non-alcoholic fatty liver disease (NAFLD) is a common cause of raised transaminases. Among patients with NAFLD, one-fifth develop non-alcoholic steato-hepatitis (NASH) with histopathological signs of necroinflammation and fibrosis, which may progress to cirrhosis with terminal liver failure or hepatocellular carcinoma. In patients with alcoholic fatty liver, sudden death has been reported to be associated with prolongation of the Q-T interval and possibly with autonomic dysfunction. However, in patients with NAFLD or NASH, the question of prolonged Q-T interval and sudden death has not been addressed.

Decreased sensitivity to circulating insulin is a central feature in the development of NASH. Drugs that stimulate nuclear transcription factor, which facilitates increased insulin responsiveness, are important in the treatment of type 2 diabetes and are promising second generation drugs for the treatment of NASH, combined with weight reduction and increased physical exercise.

Hepatotoxicity is one of the concerns of glitazone therapy, because of the impaired liver function associated with the now withdrawn troglitazone. Recent reviews and large scale clinical controlled trials have concluded that rosiglitazone does not share the hepatotoxic profile of troglitazone. Several studies have shown that necroinflammation and fibrosis may regress in patients with NASH during rosiglitazone therapy. However, side effects have been reported with treatment of NASH with glitazones such as allergic reactions, a further rise in serum transaminases, weight gain, and fluid retention. Therefore, the use of second generation glitazones is contra-indicated in the presence of active liver disease with a transaminase level 2.5 times above the normal upper limit.

A meta-analysis and an interim analysis have recently been published on the potential effects of rosiglitazone to raise mortality from myocardial infarction and produce an increased prevalence of heart failure.

Th meta-analysis of 42 studies on rosiglitazone in patients with type 2 diabetes found that out of more than 15,000 patients receiving rosiglitazone for more than 24 weeks, 86 patients in the rosiglitazone group had myocardial infarctions versus 72 in the control group. Death from other cardiovascular causes was 39 in the rosiglitazone group versus 22 in the control group. Only one of the 42 studies treated patients with liver disease. Another finding was that rosiglitazone may precipitate congestive heart failure in susceptible patients.

An interim analysis of rosiglitazone assessed on cardiovascular outcome and regulation of glycemia in diabetics (RECORD trial) was recently published. In this study rosiglitazone with either metformin or sulfonylurea was compared to metformin and sulfonylurea. All patients had type 2 diabetes and the mean follow-up time was 3.75 years. There were 74 deaths in 2220 patients in the rosiglitazone group and 80/2227 in the control group. There was no evidence of an increased risk of death, either from any cause or from cardiovascular causes in the rosiglitazone group. There was no statistical difference in acute myocardial infarction (43 had acute myocardial infarction in the rosiglitazone group versus 37 in the control group). There was a statistically significant increase in the risk of heart failure with rosiglitazone treatement: 47 in the rosiglitazone group versus 22 in the control group.

The question of whether or not rosiglitazone increases the risk of acute myocardial infarction is still open, although increased risk of congestive heart failure is well documented in patients with type 2 diabetes. This should be borne in mind when considering the treatment of NASH patients with rosiglitazone.