Paclitaxel-coated devices (PCDs) were not found to be associated with a higher risk of mortality for lower limb endovascular revascularization, according to a French national registry study. These data were reported by Matthieu Wargny, MD, PhD, of Nantes Universite, France, and colleagues, in a manuscript published Monday online and in the April 2 issue of the Journal of the American College of Cardiology. In December 2018, a controversial meta-analysis suggested increased all-cause mortality at 2 and 5 years with PCDs compared with noncoated devices used in the femoropopliteal segment. At the time, PCDs had been routinely used for treatment of peripheral artery disease (PAD), particularly at the femoropopliteal level. Paclitaxel is a cytotoxic drug that prevents the occurrence of restenosis and improves the patency of endovascular revascularization. Since 2018, many studies, including patient-level, pooled analyses from randomized controlled trials (RCTs) and large registry data sets have been conducted to investigate whether there is increased risk of mortality with PCD utilization for lower limb endovascular revascularization procedures, including the present study. The DETECT (Drug EluTing dEvices FrenCh safeTy survey), a real-world nationwide retrospective cohort study from the French National Health Data System that covers >99% of the French population, evaluated the cause of death of all patients age ≥18 years who underwent a first infrainguinal endovascular intervention in any French hospital between October 2011 and December 2019. The period began when the first drug-eluting stents (DES) were reimbursed by the French authorities and ended in 2019 to allow a minimum follow-up period of 2 years for all patients. Patients with discordant data regarding age, sex, date of death, inconsistency between information systems (outpatient care/hospital), and inconsistency between procedures and associated device codes (e.g., procedure report mentions a stent, but no associated device code was declared) were excluded. For the main analysis, two groups were compared: the PCD group (drug-coated balloon [DCB] and/or DES) and the control group (plain old balloon angioplasty [POBA] and/or bare-metal stent [BMS] and/or covered metal stent [CMS], without DCB or DES). Cox proportional hazards models were used to compare the between group incidence of the four main outcomes of interest: Mortality; New lower limb artery procedure (including contralateral limb from the first intervention); Major lower limb amputation (including contralateral limb from the first intervention); and Major adverse cardiovascular event including amputation (MACE-A). Time-to-event figures were drawn using the cumulative incidence function, taking into account all-cause mortality as a competing risk for new lower limb artery procedure or major lower limb amputation. Sensitivity analyses were also performed with 1:1 propensity-score matching (PSM-exact) and propensity-score matching nearest neighbor (PSM-NN) methods. The study analyzed 259,137 patients who underwent at least one infrainguinal endovascular intervention between Oct. 1, 2011, and Dec. 31, 2019, of whom 20,083 (7.7%) received treatment with at least one PCD. Of the patients who received PCDs, 6,804 (33.9%) received DCB only, 12,455 (62.0%) received DES only, and 824 (4.1%) received both DCB and DES. Compared to the population that survived, the population that died was more frequently women, older, and more frequently had history of coronary artery disease, heart failure, major lower limb amputation (4.0% vs. 1.3%), and chronic kidney disease. PCD-treated patients were younger (median age 71 years [interquartile range [IQR]: 63-80 years] vs. 74 years [IQR: 64-83 years) than non-PCD-treated patients. Baseline characteristics were unbalanced, with a higher rate of PAD and lower limb artery procedures in the PCD-treated group, although more amputations occurred in the non-PCD treated group. At a median follow-up before death of 3.2 years, 44.5% of patients (115,190 of 259,081) had died. In the multivariable analyses, the PCD-treated group had a lower risk of mortality than the control group (hazard ratio [HR]: 0.86; 95% confidence interval [CI]: 0.84-0.89; P < 0.001). A lower risk of mortality was also observed with PCDs in the propensity-score matching approaches. Treatment with DCB without DES was associated with a lower risk of mortality in a multivariable model (HR: 0.84; 95% CI: 0.80-0.89; P < 0.001), MACE-A (HR: 0.85; 95% CI: 0.81-0.90; P <0.001), and major lower limb amputation (0.73; 95% CI: 0.64-0.84; P <0.001), but an increased risk of new lower limb artery procedure (HR: 1.10; 95% CI: 1.05-1.1.15; P < 0.001). Overall, the real-world DETECT study showed that PCD treatment was associated with a lower risk of late mortality, particularly in the “PSM-exact” analysis, with an HR of 0.82 (95% CI: 0.78-0.87; P <0.001) using a nearly exhaustive national database representing >99% of the French population. These results are consistent with previous real-world registry data analyses that showed the absence of mortality risk with PCD with and without classical adjustment for confounding factors or in the PSM approach. A significant limitation of this analysis is lack of data regarding the laterality of the procedure, which strongly limits relevancy of the “new lower limb artery procedure” outcome, as reintervention may have occurred in the limb not initially treated with a PCD. In an accompanying editorial, Peter A. Schneider, MD, of the University of California-San Francisco, and Jeffrey W. Olin, MD, of Mount Sinai Hospital, New York, described how the controversial 2018 summary-level meta-analysis of RCTs of PCDs had significant limitations, included limited follow-up of most included studies, no proposed biological mechanism, and a novel method of calculating a dose response relationship. Despite the significant limitations of the methodology of this meta-analysis, regulatory agencies, physicians, and hospital systems criticized the routine use of PCDs, leading to a real alteration in clinical practice, with less utilization of PCDs, despite these devices being found to be safe and efficacious compared to BMS and POBA in prior trials. Schneider and Olin commented that despite limitations in granularity of indications and procedural details, the present study by Wargny and colleagues is a well-designed observational study leveraging real-world data with results consistent with pooled patient-level analysis of RCTs. The editorialists concluded that real-world data should be leveraged to evaluate any potential harm signal from a treatment and that a major claim of harm based on summary-level meta-analyses should be scrutinized in the editorial process. Sources: Wargny M, Leux C, Chatellier G, et al. Mortality in a Nationwide Practice-Based Cohort Receiving Paclitaxel-Coated Devices for Lower Limb Peripheral Artery Disease. J Am Coll Cardiol. 2024;13:1207-1221. Schneider PA, Olin JW. Paclitaxel-Mortality Risk Hypothesis Debunked: What We Learned and How It Will Change Future Clinical Trials. J Am Coll Cardiol. 2024;13:1222-1224. Image Credit: bacsica – stock.adobe.com