Pooled patient level data are better suited than study level data to investigate the link between dipeptidyl peptidase-4 inhibitors and the risk of heart failure in type 2 diabetesBMJ 2016; 353 doi: https://doi.org/10.1136/bmj.i2920 (Published 24 May 2016) Cite this as: BMJ 2016;353:i2920
- Tejas Patel, ORISE fellow1,
- Bereket Tesfaldet, ORISE fellow1,
- Charu Gandotra, assistant professor and director, nuclear cardiology2
- On behalf of the Meta-AnalyTical Interagency Group (MATIG) (Keith Burkhart, MD; Henry Chang, MD; Jue Chen, MS, PhD; Sean Coady, MS; Lawton Cooper, MD; Gyorgy Csako, MD; Michelle Fennessy, RN, PhD; Jerome Fleg, MD; Ahmed Hasan, MD, PhD, FACC; Ruth Kirby, BS, RN; Eileen Navarro Almario, MD, MS, FACP; Frank Pucino, PharmD, MPH; Subha Raman, MD; Yves Rosenberg, MD, MPH; George Sopko, MD; Helena Sviglin, MPH; Robert Wesley, PhD.
- 1Food and Drug Administration, 10993 New Hampshire Avenue, Silver Spring, MD 20993, USA
- 2Division of Cardiovascular Medicine, Department of Internal Medicine, Howard University Hospital, Washington, DC, USA
Li and colleagues’ systematic review and meta-analysis included a large number of studies and diverse populations of patients with diabetes.1 However, certain limitations make it difficult to interpret the results. Firstly, durations of exposure and length of follow-up varied between trials, making it difficult to assess the effect of treatment duration on the occurrence of heart failure. Secondly, dipeptidyl peptidase-4 (DPP-4) inhibitors were used both as monotherapy and in combination with other antihyperglycaemic agents, making it difficult to determine a direct causal association between DPP-4 inhibition and heart failure. Thirdly, heart failure events were not always prespecified, well defined, or adjudicated across trials. Finally, study level meta-analyses and systematic reviews lack the ability to identify or …