Perspective:

The association between DCDs and long-term mortality reported in the 2018 meta-analysis by Katsanos K, et al. (J Am Heart Assoc 2018;7[24]:e011245) precipitated vigorous world-wide debate about the role of these devices, if any, in treatment of peripheral artery disease (PAD). As mentioned in this article’s introduction, these results were confirmed by an internal analysis by FDA and followed by a letter of caution to clinicians from FDA, and a new trial to answer this long-term mortality controversy would be unlikely due to lack of feasibility. Since then, clinicians treating patients with PAD have been grappling with how to use DCDs (which remained on the market), if at all. In response to the limitations of the meta-analyses and small trials from which these concerns arose, a steady stream of retrospective analyses utilizing a variety of data sources and designs has ensued.

For clinicians disappointed by the relatively nondirective nature of the previous warning from FDA about DCDs, the current study provides welcome evidence from a robust (albeit nonrandomized) dataset that is unencumbered by some of the potential biases related to missing details and loss to follow-up associated within previous meta-analyses. This study also stands apart from the crowd of other retrospective analyses because its sophisticated design was informed by input from FDA and prespecified. Kudos to the FDA for collaborating with extramural researchers and trying to answer these questions with a robust and detailed data source—this collaboration is to be applauded. Although the multi-industry consortium of device manufacturers funding the study was not directly involved in the design, conduct, or analysis, it also is important to note that the consortium includes DCB manufacturers that are also listed among the authors’ conflict of interest disclosures.

Although it is hard to argue with all-cause mortality as an outcome that deserves priority, the incidence of repeat femoropopliteal procedures observed in this cohort underscores the importance of restenosis (and therefore the potential benefit of DCDs) for patients with PAD treated with endovascular interventions. Over 41,000 patients (nearly 25%) had repeat interventions, and nearly 9% of those initially treated with a NDCD were subsequently treated with a DCD. The noninferiority mortality finding and consistency of this observation within the confirmatory and sensitivity analyses may reduce anxiety for providers considering a DCD, especially for high-risk patients and/or lesions. Future analyses of this cohort are planned and will be important to monitor since previous evidence suggested mortality effects increase over longer follow-up. The pendulum may be swinging back toward DCD use, but how far remains to be determined.