By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center

Dr. Zimmet has no relevant financial relationships with ineligible companies to disclose.

SYNOPSIS: Paclitaxel-eluting devices vs. bare metal stents in peripheral arterial disease showed no significant difference in all-cause mortality, contradicting the results of a controversial meta-analysis.

SOURCE: Nordanstig J, James S, Andersson M, et al. Mortality with paclitaxel-coated devices in peripheral artery disease. N Engl J Med 2020;383:2538-2546.

In late 2018, the authors of a meta-analysis of trials of paclitaxel-eluting balloons and stents in peripheral arterial disease (PAD) reported a surprising finding.1 Using pooled data from 28 randomized, controlled trials (RCTs; 4,663 total patients), researchers studied all-cause mortality at one, two, and five years. In the entire cohort, mortality at one year was not different between the paclitaxel and control arms. However, in the subsets of studies with available two-year data (12 RCTs; 2,316 patients) and five-year data (three RCTs; 863 cases), the authors reported an increased risk of all-cause mortality among patients treated with the paclitaxel-coated devices. This news led to a flurry of further investigations and produced a chilling effect on clinical use of these tools. Part of the fallout of this resulted in the halting of ongoing research into the efficacy of these devices.

One of the affected clinical trials was SWEDEPAD, a large, multicenter, randomized study of drug-eluting technology in lower limb PAD in Sweden. Investigators explored whether paclitaxel-coated devices can affect amputation rates in patients with limb-threatening ischemia, and to better quality of life measures in claudicants. The authors began enrolling in late 2014, but temporarily stopped in December 2018 over safety concerns raised in the 2018 meta-analysis. The data and safety monitoring committee recommended the unusual step of performing an unplanned analysis of all-cause mortality with the available data. Up until the point of the temporary hiatus, 2,289 patients had been enrolled in SWEDEPAD. The authors had randomized these patients 1:1 to paclitaxel-coated or uncoated devices. Of these, the authors noted chronic limb-threatening ischemia in 1,480. The remaining 809 had intermittent claudication. The average age of included patients was 75 years, and approximately 55% were men. Twenty-three patients in the drug-coated device group and 25 patients in the uncoated device group had undergone intervention with paclitaxel devices before the index procedure. The interim analysis was carried out through March 2020, at which time the mean follow-up period was 2.5 years, with a range of approximately one to four years.

At the time of analysis, 574 patients had died. This included 293 of 1,149 patients in the paclitaxel group and 281 of 1,140 patients in the uncoated device group (HR, 1.06; 95% CI, 0.92-1.22). Similarly, there was no significant difference in mortality for paclitaxel vs. no-drug patients with critical limb ischemia (33.4% vs. 33.1%; HR, 1.04; 95% CI, 0.90-1.21) or for those with intermittent claudication (10.9% vs. 9.4%; HR, 1.18; 95% CI, 0.72-1.93). The authors concluded their analysis did not show an excess mortality hazard among PAD patients treated with paclitaxel-coated devices. Enrollment of patients in the original randomized trial has resumed.


A signal for increased late mortality after treatment with paclitaxel-eluting balloons and stents was initially reported in a meta-analysis in late 2018. The FDA reiterated this using patient-level data from the same trials. The authors of other meta-analyses have come to different conclusions. To date, the results of observational studies have not suggested such excess mortality. The percutaneous interventional treatment of lower extremity PAD has been hampered by a high rate of restenosis, which is responsible for significant morbidity and repeat procedures. There is no mystery as to why there has been such enthusiasm for drug-eluting devices in this space, when most trials have reported improvements in primary patency and target-lesion revascularization vs. uncoated devices.

Importantly, no clear mechanism has been identified that would explain why paclitaxel, in doses far smaller than what is used systemically to treat cancer, would lead to excess mortality at points longer than one year. Explanations have been proposed, including the seemingly quixotic theory that by decreasing restenosis, paclitaxel-eluting devices decrease contact with the medical system, resulting in poorer downstream outcomes. The 574 deaths reported in this single trial exceeds the combined number reported in 36 prior randomized trials. The completeness of the data in this trial, performed within the Swedish health system, is remarkable, with no patients lost to follow-up. For now, the SWEDEPAD data should be reassuring to those who seek to resume related clinical research. Time will tell whether individual clinicians will be adequately reassured about reincorporating paclitaxel devices in routine practice.


  1. Katsanos K, Spiliopoulos S, Kitrou P, et al. Risk of death following application of paclitaxel-coated balloons and stents in the femoropopliteal artery of the leg: A systematic review and meta-analysis of randomized controlled trials. J Am Heart Assoc 2018;7:e011245.