Meta-Analysis Design and Results in Real Life: Problem Solvers or Detour to Maze. A Critical Review of Meta-Analysis of DAPT Randomized Controlled Trials.

Abstract

Therapeutic strategies - such as duration of dual antiplatelet therapy after coronary artery stenting - usually generate a large quantity of meta-analyses. The meta-analyses that include the same randomized clinical trials should produce similar results. Our aim in the study is to analyze the quality and to compare the results of meta-analyses focused on a controversial topic such as dual antiplatelet therapy after percutaneous coronary intervention. We searched all published meta-analyses published up to November 2015 (near DAPT trial publication) selecting those that included the same randomized clinical trials comparing patterns of briefer versus longer-term double antiplatelet therapy. Seventeen meta-analyses achieved our selection criteria. Of the seventeen analyzed, we identified seven (41.1%) based on the same ten randomized clinical trials (RCTs), yet their results varied widely. Many of the meta-analyses differed in only some minor aspect of the design (i.e. eligible studies, length of comparators and statistical methods used). Some authors differed in the number of patients participating in RCTs and even, despite reviewing the same underlying trials, only 2 of the 7 meta-analyses included the same number of patients. Meta-analyses around cardiovascular, all-cause or non-cardiovascular death differ frequently. In the DAPT duration setting, several meta-analyses have been recently published based on the same data, presenting several issues making it difficult to determine clear recommendations on certain points.