Effect of GLP-1 Mimetics on Blood Pressure and Relationship to Weight Loss and Glycemia Lowering: Results of a Systematic Meta-Analysis and Meta-Regression

Original Article Effect of GLP-1 Mimetics on Blood Pressure and Relationship to Weight Loss and Glycemia Lowering: Results of a Systematic Meta-Analysis and Meta-Regression Mohammad Katout,1 Hong Zhu,2 Jessica Rutsky,1 Parthy Shah,1 Robert D. Brook,3 Jixin Zhong,1 and Sanjay Rajagopalan4 background Incretin therapies such as glucagon-like peptide 1 (GLP-1) agonists are commonly used for the treatment of type 2 diabetes mellitus. GLP-1 mimetics, besides improving glycemic control, have been shown to influence multiple pathways regulating blood pressure (BP). We investigated the GLP-1 analogs effects on BP from published randomized studies using a meta-analytic approach. methods Thirty-three trials (12,469 patients) that assessed the efficacy of GLP-1 analogs on glycemic control (HbA1C) over 12–56 weeks that met additional criteria, including the availability of standardized sitting BP assessment and weight parameters, were identified. Comparator therapy included oral antiglycemic drugs or placebo. The weighted mean difference (WMD) in systolic BP (SBP) change was calculated using a random-effects model after performing a test for heterogeneity. results Forty-one percent of patients were treated with liraglutide (0.3–3 mg once daily), whereas 59% were treated with exenatide (5–10 µg twice daily or 2 mg weekly). GLP-1 treatment achieved a greater SBP reduction than comparator therapy (WMD = 2.22 mm Hg; 95% confidence interval (CI) = −2.97 to −1.47). In the pooled analysis, GLP-1 had beneficial effects on weight loss (WMD = −2.56 kg; 95% CI = −3.12 to −2.00), HbA1c reduction (WMD = −0.41%; 95% CI = −0.78 to −0.04) but was associated with a heart rate increase (WMD = 1.30 bpm; 95% CI = 0.26–2.33). In a separate meta-regression analysis, the degree of SBP change was not related to baseline BP, weight loss, or improvement in HbA1C. conclusions This meta-analysis provides evidence that GLP-1 analogs reduce sitting SBP. These findings may support potentially favorable long-term cardiovascular outcomes. Keywords: blood pressure; exenatide; GLP-1; hypertension; liraglutide; meta-analysis. doi:10.1093/ajh/hpt196 Glucagon-like peptide 1 (GLP-1) is a peptide hormone that stimulates insulin and inhibits glucagon secretion in a glucose-dependent manner.1 GLP-1 exerts favorable effects on glycemic control additionally by binding to the GLP-1 receptor expressed in the gastrointestinal tract and central nervous system, exerting effects on gastric emptying and reductions in appetite. Recently it has been recognized that the GLP-1 receptor is widely expressed in the cardiovascular system in a number of different cell types, including endothelial and smooth muscle cells.1 Several GLP-1 receptor (GLP-1R) agonists have been developed as therapeutic agents for the treatment of type 2 diabetes mellitus. GLP-1 agonists are effective in lowering weight and improving glycemic control, with recent meta-analysis suggesting an average lowering of weight by approximately 2 kg and reductions in HbA1C of 1.5%.2 A variety of other pleiotropic effects have also been described, including improvements in endothelial function by activation of phosphoinositide 3 kinase, protein kinase B, and endothelial nitric oxide synthase, that may lead to lowering of blood pressure (BP).1 Both short-term (with some including acute effects) and chronic studies and recent meta-analysis that have included both short-term and long-term studies have suggested that these agents may lower BP. In this meta-analysis, we pooled results from all published studies to date that have reported on BP as part of randomized controlled trials of GLP-1 agonists vs. placebo or non-GLP-1 comparator of at least 12-weeks duration and attempted to understand the relationship between BP lowering and improvements in weight and glycemia using a metaregression approach. Correspondence: Sanjay Rajagopalan (). 1Division of Cardiovascular Medicine, University of Maryland School of Medicine, Baltimore, Maryland; 2Davis Heart and Lung Research Institute, The Ohio State University, Columbus, Ohio; 3Division of Biostatistics Department of Clinical Sciences, UT Southwestern Medical Center, Dallas, Texas; 4Division of Cardiovascular Medicine, University of Maryland, Baltimore, Maryland. Initially submitted May 24, 2013; date of first revision July 12, 2013; accepted for publication September 28, 2013; online publication November 21, 2013. © American Journal of Hypertension, Ltd 2013. All rights reserved. For Permissions, please email: 130 American Journal of Hypertension 27(1) January 2014 Results of a Systematic Meta-Analysis and Meta-Regression METHODS Data sources and literature search Pubmed, Scopus, Cochrane, and Clinicaltrials.gov trials databases were searched for English-language articles published from 1 January 2000 to 1 March 2013. Prespecified search terms were GLP-1, glucagon-like peptide 1, exenatide, liraglutide, albiglutide, taspoglutide, lixisenatide, and BP. Titles and abstracts were searched. Studies included in our analyses were only blinded randomized controlled trials (RCTs) or open-labeled studies of 12- to 56-weeks duration. Review methods and selection criteria The reference lists of relevant papers were reviewed manually by one reviewer, with 1,156 articles included for initial screening. We included trials of adult patients with or without type 2 diabetes mellitus with a GLP-1R agonist treatment arm. Studies were included only if they had a non-GLP-1 comparator (includes placebo and active comparator trials) with sample size mentioned for each comparator group and reported systolic BP (SBP) measurements before and after active treatment with SD or confidence interval (CI). DPP-4 as comparator drugs were allowed, provided there was a nonincretin treatment group and/or placebo arm. All studies had to include a minimum duration of 12 weeks of drug exposure. The control groups were thus either placebo or active comparator arms that included oral antidiabetic drugs, DPP-4 inhibitors (sitagliptin, linagliptin, saxagliptin, alogliptin, and vildagliptin) PPARγ agonist, pioglitazone, sulfonylureas, orlistat, and insulin. We also assessed whether the trials included standardized diet/exercise regimens. If trials included >1 medication strength or >1 study duration, we took the higher titration dose and/or the longer duration group for weight and HbA1c and combined all the strengths in the meta-analysis for SBP effects. The applied exclusion criteria excluded acute dosing studies, duplicate trials, trials with incomplete data, or those without BP measurements. The following article types were excluded: reviews, letters, opinions, or treatment guidelines; abstracts published before January 2000 or duplicate abstracts; experimental nonhuman studies; studies in type 1 diabetes; studies in obesity in the absence of diabetes; and studies of drug mechanism of action, pharmacokinetics, or pharmacodynamics properties. Studies were fur (...truncated)