Response to Letter Regarding Article, “Potentially Large yet Uncertain Benefits: A Meta-analysis of Patent Foramen Ovale Closure Trials”

We thank the writers for their comments, which largely underscore our own perspective. We highlight and clarify a few points in response. First, even when individual studies are themselves underpowered, meta-analysis can provide a way to make inferences from the combined results. This was most elegantly shown by Lau et al,1 who demonstrated that small underpowered trials provided estimates of the effects of thrombolytics on acute myocardial infarction that wholly anticipated the subsequent mega-trials—mega-trials that arguably were unnecessary or even unethical. In the case of patent foramen ovale and cryptogenic stroke, even the meta-analysis remains underpowered, as evidenced by the wide confidence interval of the summary result. Second, using transient ischemic attack alone (even …