Paclitaxel-coated balloons had come under fire in recent years, after a meta-analysis of randomized control trials involving their use for the treatment of peripheral artery disease found an associated with higher mortality when compared to non-drug coated devices. The study led to a letter of caution published by the FDA, leading to hesitance on the part of both physicians and patients to use the devices ever since. In June of 2019, the FDA called for additional long-term safety data to allow Paclitaxel-coated balloons to remain on the market.
Dr. Eric Secemsky of the Beth Israel Deaconess Medical Center, and his team at the Richard A and Susan F. Smith Center for Outcomes Research in Cardiology, were up to the challenge. As new randomized control trials to assess the safety and efficacy of paclitaxel coated balloons were not feasible, they designed a novel study with feedback from the FDA, to provide ongoing evaluation of the safety of drug coated devices using Medicare claims data. The primary objective was to evaluate whether DCDs used for femoropopliteal revascularization were non-inferior to non-drug coated devices (NDCDs) with respect to all-cause mortality. A noninferiority design was adopted, with a consideration of 5% relative increase in the mortality rate as the noninferiority margin.
During the median 3.7-year study period, 168,553 patients treated at 2978 centers across the united states were included in the analysis. The mean age of the cohort was 77 years and almost half of the patients were women. DCDs were not associated with increased mortality compared to NDCDs (51.6% vs. 56.7% respectively, HR 0.84, 95% CI 0.82-0.85, p<0.001). Various sensitivity analyses were performed to confirm the data and account for confounding given the observational nature of the research. Furthermore, the primary endpoint held its own across all pre-specified subgroups, including those with and without critical limb ischemia, inpatients vs. outpatients, and those who received a stent vs. those who did not.
“Not only does this study put the claims that paclitaxel-coated balloons increase mortality to rest,” said Dr. Eric Secemsky in an interview with Cardiology Now News, “it opens the door for similar observational studies in the future. Our prespecified endpoints and sensitivity analyses minimized the issues of confounding seen in observational studies, and allowed for the FDA to assess the safety of this device using real-world data. The 2021 CURES ACT calls for more real-time data to inform clinical practice, and we have shown that this is certainly possible”.
The authors acknowledged the study limitations with respect to its observational design, as they relied on appropriate reporting of claims data for procedural information. Given the reliance on claims data, specific procedural information was not available either. Furthermore, the median length of follow up is shorter than the 4-5 year interval when the harm signal was originally seen.
The study was simultaneously published in the Journal of the American Medical Association Internal Medicine.