Tang et al. (1) reported that the administration of insulin glargine, but not liraglutide, reduced the liver fat burden in patients with type 2 diabetes, whereas the improvement in the liver fat fraction was not significantly different from that in the liraglutide group in a 12-week randomized trial. This is an interesting clinical study on the effects of antihyperglycemic drugs on liver fat. However, there are some concerns about this study.
First, given the different mechanisms by which insulin glargine and liraglutide therapies may reduce liver fat, the short observation period of 12 weeks may not be long enough to show the actual effects of the two agents, especially for liraglutide. The mechanisms by which insulin therapy may reduce liver fat include inhibition of lipolysis in the short term, whereas liraglutide may decrease liver fat by promoting weight loss and reducing hepatic lipogenesis in the long term. As shown in Fig. 2 of Tang et al. (1), a weak positive correlation was found between changes in body weight and changes in the mean liver proton density fat fraction (PDFF) in the total participants. However, the positive correlation in the liraglutide therapy group was stronger than the insulin glargine intervention (1). Insulin glargine could increase body weight in patients with type 2 diabetes in a long-term study (2), which could attenuate the effect of liver fat reduction. A maximal decrease in body weight by liraglutide was achieved after 36 weeks of therapy in patients with type 2 diabetes (3). In addition, liraglutide significantly improved liver function and histological features in patients with nonalcoholic fatty liver disease with glucose intolerance in a 96-week trial pilot study (4). Therefore, a longer intervention might be optimal to show reduced liver fat using liraglutide therapy.
Second, the author stated that 17 patients in each group are required to detect a 5% absolute clinical difference in fat content before and after the intervention. However, comparison between before and after the liraglutide intervention was only performed in 14 patients (4 patients out of 18 discontinued the treatment). The underestimated drop-out rate of a priori sample size calculation could partially account for no difference in fat content before and after intervention. For comparison of changes between groups, 32 patients in each group would be required to detect a 5% absolute clinical difference (whole-liver MRI-determined PDFF = 13.8 ± 7.1, two-tailed α = 0.05, β = 0.20). Therefore, it is not surprising that no difference in liver fat fraction between the two groups in the small sample study was detected. Furthermore, the mean values of baseline MRI-based biomarkers in the liraglutide group are different between Tables 1 and 2 of Tang et al. (1).
Patients with normal liver fat were not excluded in the study and most patients might have mild to moderate steatosis. The average MRI-determined PDFF in the study was 13.8%, whereas the average MRI-determined PDFF increases significantly with increasing histology-determined steatosis grade: 8.9% at grade 1, 16.3% at grade 2, and 25.0% at grade 3 (5). Liraglutide therapy could reduce more liver fat fraction in patients with moderate to advanced steatosis.
Article Information
Funding. P.-F.S. has received research grants from the National Natural Science Foundation of China (81370968 and 81373391) and the Chinese Society of Endocrinology (13040620447).
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
