Risk of dyspepsia and anorexia due to glucagon-like peptide-1 receptor agonists in patients with type 2 diabetes mellitus:a Network Meta-analysis
Zhixia Li,Xiantao Zeng,Sanbao Chai,Xiaochi Quan,Shanshan Wu,Zhirong Yang,Siyan Zhan,Feng Sun
DOI: https://doi.org/10.3760/cma.j.issn.1008-5734.2015.03.008
2015-01-01
Abstract:Objective To evaluate the risk of dyspepsia and anorexia due to glucagon-like peptide-1 receptor agonist(GLP-1 RA)in patients with type 2 diabetes mellitus(T2DM). Methods The electronic database of Chinese BioMedical Literature Database( CBMdisc ), Chinese Medical Current Contents( CMCC),Medline,EMbase,the Cochrane Library and web site of ClinicalTrials. gov were searched from inception to May 1st 2014. Those randomized controlled trials whose inclusion criteria including patients with T2DM as the research object,comparisons of GLP-1 RA and placebo or other conventional anti-diabetic drugs as the intervention measures,and dyspepsia and anorexia as the outcomes were collected. A traditional Meta-analysis and a Network Meta-analysis were used and relational graph and rank ordering figure of all the intervention measures were drawn. Results A total of 42 randomized controlled trials were enrolled into this study involving 20 916 patients with T2DM and 13 kinds of intervention measures comprised 7 kinds of GLP-1 RAs( exenatide,exenatide release agent,liraglutide, lixisenatide,taspoglutide, albiglutide, and dulaglutide ), 5 kinds of conventional anti-diabetic drugs ( insulin, metformin, sulfonylureas, sitagliptin, and thiazolidinediones ketones ), and placebo. The traditional Meta-analysis showed that,compared with placebo,the whole of GLP-1 RAs could increase the risks of dyspepsia[odds ratio(OR)=3. 04,95% confidence interval(CI):1. 79-5. 16]and anorexia (OR=2. 57,95%CI:1. 69-3. 91)and there were statistically significant differences(P <0. 05). The Network Meta-analysis showed that,compared with placebo,albiglutide,exenatide,exenatide release agent, liraglutide,lixisenatide,and taspoglutide could increase the risks of dyspepsia with statistically significant differences(all P<0. 05)and,of them,liraglutide was at the greatest risk and the risk was 7. 69(1/0. 13)times as high as that of the placebo. Compared with insulin,metformin,sulfonylureas,sitagliptin, and thiazolidinediones ketones,GLP-1 RAs could also increase the risks of dyspepsia in the patients with T2DM and,of them,liraglutide was at the greatest risk,which was 13. 58,4. 17(1/0. 24),3. 85(1/0. 26),5. 00(1/0. 20),and 3. 70(1/0. 27)as high as that of insulin,metformin,sulfonylureas, sitagliptin,and thiazolidinediones ketones,respectively. Compared with placebo,dulaglutide,exenatide, liraglutide,and taspoglutide could increase the risks of anorexia with statistically significant differences( all P<0. 05)and,of them,dulaglutide was at the greatest risk,5. 53 times as high as that of the placebo. Compared with insulin,sulfonylureas,thiazolidinediones ketones,and sitagliptin,GLP-1 RAs could also increase the risks of anorexia in the patients with T2DM( all OR >1. 00 except for lixisenatide versus sitagliptin )and,of them,dulaglutide was at the greatest risk,48. 91,16. 65,36. 24,and 4. 75 times as high as that of insulin,sulfonylureas,thiazolidinediones ketones,and sitagliptin,respectively. There was no statistically significant difference between two kinds of GLP-1 RAs for risks of dyspepsia and anorexia( all P>0. 05). The risks of dyspepsia and anorexia due to the 13 kinds of intervention measures were ranked by rank ordering figure and liraglutide and dulaglutide were at the greatest risks. Conclusion Both the traditional Meta-analysis and Network Meta-analysis showed that GLP-1 RAs could increase the risk of dyspepsia and anorexia in patients with T2DM.