Katsanos et al 1 investigated the risk of drug-eluting technology, showing evidence that the application of paclitaxel-coated devices in the femoropopliteal segment was related to a significantly increased risk of death [relative risk (RR) 1.68, 95% confidence interval (CI) 1.15 to 2.47]. 1 Out of the 28 studies, 790 patients were from 4 RCTs testing a paclitaxel-eluting stent (PES). These PES studies represented a minority of patients (17.82% at 1 year). Thus, the mix of PES and paclitaxel-coated balloon studies is a main bias since many differences exist between the devices in terms of platform, excipient, polymer, drug concentration, and formulation (RR 1.66). 2 Numbers of deaths were similar between the PES and control arms at 1 year, but there was an increased risk of death following application of PES at 2 years. However, only 2 RCTs testing PES were available at 2 years [515 patients (22.2%) of 2316 subjects]. They found 20 deaths among 320 patients in the paclitaxel group (6.25%) and 7 deaths (3.58%) of 195 patients in the control group. Overall, Katsanos et al 1 reported a significant increase of death in the PES subgroup (RR 1.87, 95% CI 1.11 to 3.15). Recently, the prospective BATTLE RCT (Bare Metal Stent vs Paclitaxel Eluting Stent in the Setting of Primary Stenting of Intermediate-Length Femoropopliteal Lesions) was published comparing bare nitinol stents (Misago; Terumo, Tokyo, Japan) and PES (Zilver PTX; Cook Medical, Bloomington, IN, USA) for de novo femoropopliteal occlusive disease. 3 In the BATTLE trial, no difference was found between the groups in terms of in-stent restenosis at 1 and 2 years. After 2 years of follow-up, 7 deaths were noted in the Misago group and 1 in the Zilver PTX group (hazard ratio 7.3, 95% CI 0.9 to 59, p=0.06). Causes of death were pulmonary cancer (n=3), sepsis (n=2), trauma (n=1), and multisystem organ failure (n=1) in the Misago group and hemorrhagic shock (n=1) in the Zilver PTX group. The deaths did not appear to be linked to the devices or the procedures. Interestingly, BATTLE was included in the Katsanos et al 1 meta-analysis, but at that time only 1-year follow-up