The major findings of the present study were: 1) liraglutide treatment effectively reduced body weight up to 6
months after discharge compared to conventional insulin therapy in Japanese obese patients with type 2 diabetes; 2) liraglutide-induced reduction in HbA1c was similar to that observed with insulin treatment; 3) maintained improvement in eating behavior up to 6
months after discharge in patients on liraglutide treatment with various scales gradually moving to those recorded in non-obese normal subjects.
The LEAD-5 study demonstrated that liraglutide decreased HbA1c similar to ultra-long acting insulin analogue, glargine, and also significantly reduced body weight compared with glargine [9
]. In the present retrospective study, we compared the effects of liraglutide and insulin on body weight and HbA1c. Almost all patients of the insulin group received intensive insulin therapy at discharge, and the TDD of insulin was relatively high (0.59 unit/kg/day). Significant body weight reduction was observed in both liraglutide and insulin groups during hospitalization and this was due to the in-hospital diet therapy. However, weight reduction was persistently observed after discharge only in the liraglutide group. Liraglutide treatment may be distinguished from the intensive insulin therapy by effective reductions in body weight and HbA1c. Further studies are needed to determine the differences in the therapeutic effects of liraglutide and intensive insulin therapy. The combination treatment of GLP-1 receptor agonist and basal insulin has been also approved in the USA and EU [10
], but is yet to be approved in Japan. The effect of such combination therapy on body weight needs to be examined in the future.
GLP-1 promotes satiety and reduces food intake [11
], but the effect of GLP-1 treatment on eating behavior has not been fully examined in human subjects with type 2 diabetes. We reported recently that short-term treatment with liraglutide did not only reduce food intake and visceral fat adiposity but also decreased the scores of certain parameters of eating behavior during hospitalization [8
]. However, we did not provide evidence for the long-term effects of liraglutide on eating behavior. Interestingly, the scores of obese type 2 diabetic patients at admission were higher in all major scales than non-obese subjects (Figure ). Liraglutide administration decreased these scores and maintained such improvement in eating behavior, with the exception of the scale of emotional eating behavior, until 6
months after discharge, and the total score of eating behavior of some liraglutide-treated patients moved closer to that of non-obese healthy subjects. GLP-1 delays gastric emptying and induces satiety, which is probably related to the combined effect of GLP-1 on the gastrointestinal tract and the brain, leading to decreased energy intake and weight reduction [13
]. Furthermore, the improvement in the scale for eating style by liraglutide may be partly accounted for by GLP-1-induced suppression of gastrointestinal peristalsis. The extent of weight reduction at 6
months after discharge correlated significantly with the reduction in the scores for sense of hunger and eating style, suggesting that the liraglutide-induced weight reduction is mediated partly through improvement in eating behavior.
A recent meta-analysis of weight reduction by GLP-1 receptor agonists showed that administration of these agents resulted in 3% weight reduction at 6
]. The LEAD-5 trial showed 2.1% weight reduction under treatment with liraglutide at 1.8
]. Further, another study, which used liraglutide at up to 3.0
mg/day for 20
weeks for treatment of obesity (BMI; 30–40
), reported 7.4% weight loss [16
]. In present study, liraglutide administered at the maximum dose of 0.9
mg/day achieved about 10% body weight reduction at 6
months after discharge. This result far exceeds those of previous studies [9
]. What are the reasons for the excellent results noted in the present study? 1) Patients treated with liraglutide were also placed on strict diet therapy and received a special educational program for weight reduction during hospitalization over 30.8
days. The importance of these two factors is evident from the rate of weight reduction in the liraglutide group achieved in the present study; weight reduction of −6.5% during hospitalization and only −3.5% between discharge and 6
months after discharge. 2) It is possible that GLP-1 analogue exhibited its utmost effectiveness on weight reduction, because glucose toxicity was corrected mainly by insulin therapy before induction of liraglutide. A poor response to GLP-1 has been described in patients with poorly controlled diabetes but this is reversed following normalization of glycemic control [17
]. High levels of serum GLP-1 were also reported in patients with the metabolic syndrome [19
], suggesting the existence of “GLP-1 resistance”. Our group has also demonstrated low expression levels of GLP-1 receptor mRNA in the islets of obese mice, which were significantly reversed after glucose-lowering medications [20
GLP-1 receptor agonists do not only have glucose-lowering effect but also improve other metabolic parameters such as lipid profile and blood pressure following 5-10% weight reduction [21
]. Long-term treatment with exenatide, a GLP-1 receptor agonist, significantly improved cardiovascular risk factors, such as LDL-C, HDL-C, and triglyceride, and biomarkers of liver function [22
]. Furthermore, treatment with GLP-1 improved postprandial hyperlipidemia, suggesting that GLP-1 could reduce cardiovascular disease risk in type 2 diabetes [23
]. In present study, liraglutide significantly decreased LDL-C at 6
months after discharge (admission: 113.1
36.5, at 6
months after discharge: 95.3
=0.03) although 66% of the patients were treated with statins at admission. As shown in Figure G, the score for requirement for fat decreased after liraglutide, suggesting that reduction of dietary fat intake partly contributed to the observed decrease in LDL-C. However, the questionnaire did not contain questions about the types of fat consumed, e.g. saturated fat or non-saturated fat, and daily frequency or quantity of fat intake. Questions for food preference would be further improved in the future. Several studies have described the extrapancreatic actions of GLP-1, especially its beneficial effects on the cardiovascular system [2
]. Treatment with GLP-1 improved endothelial dysfunction in type 2 diabetics with ischemic heart diseases [24
]. We recently showed that short-term liraglutide treatment decreased serum high-sensitivity C-reactive protein (hsCRP) and soluble intracellular adhesion molecule-1 (sICAM-1) levels [8
]. Long-term prospective studies are needed to determine whether the observed changes in obesity, eating behavior, and various metabolic parameters induced by treatment with liraglutide translate into protection against cardiovascular events in obese type 2 diabetics.
The present study has several limitations. The study was not a randomized clinical trial (RCT) and not prospective in design. The baseline characteristics of the patients of the liraglutide and insulin groups were not identical. In addition, the present study included a relatively small population. Furthermore, assessment of eating behavior could not been performed in the insulin group.
In summary, treatment with liraglutide effectively reduced body weight and improved eating behavior in obese Japanese patients with type 2 diabetes for up to 6
months. Liraglutide is potentially useful for the treatment of obese patients with type 2 diabetes. The weight-lowering effects of liraglutide should be examined in more detail in the future in a double-blind placebo-controlled clinical trial.