DETECT: No Long-term Harm From Paclitaxel Devices in PAD

A long-term analysis of French patients who received paclitaxel-coated balloons and stents for the management of peripheral artery disease (PAD) is adding more evidence to refute the existence of a mortality signal with the devices.

Consistent with previous real-word analyses, the nationwide DETECT study, led by Matthieu Wargny, MD, PhD (Nantes Université, France), found no differences in the risk of death between paclitaxel-treated patients and controls treated with plain old balloon angioplasty (POBA), BMS, or covered metal stents without paclitaxel at a median follow-up of 4 years.

“This study presents substantial evidence that the use of paclitaxel-coated devices is safe for the endovascular revascularization of lower limb artery disease,” Wargny and colleagues write in the paper published today in the Journal of the American College of Cardiology.

The study adds to mounting patient-level and real-world data that do not support the findings of a controversial 2018 summary-level meta-analysis that showed a late mortality signal beginning at 2 to 3 years in patients treated with a paclitaxel-based drug-eluting stents or drug-coated balloons (DCB) compared with POBA or plain stent, with an estimated 7% of increased absolute risk at 4-5 years.

Most recently, an updated patient-level meta-analysis of more than 2,600 patients showed no mortality effect or dose response of paclitaxel at 5-year follow-up. Sahil Parikh, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), who led that analysis, told TCTMD that the DETECT data “amplify the efforts that all of us have put forward in this space to demonstrate that there is no increased hazard.”

In addition to using medico-administrative data from the French National Healthcare System representing more than 99% of the population, DETECT is bolstered by inclusion of patient exposure to paclitaxel from other procedures, including paclitaxel-based coronary devices that are commonly used in Europe, Parikh added.

“As we learned in the patient-level meta-analysis that we did, the more crossovers we were able to identify and account for, the closer to unity the hazard ratio,” he said.

DETECT Results

For the DETECT study, Wargny and colleagues included 259,137 PAD patients, 7.7% of whom were treated with one or more paclitaxel-coated devices from 2011 through 2019: 33.9% with DCB only, 62% with DES only, and 4.1% with DCB and DES. The analysis included both polymer and nonpolymer DES and six types of DCBs, including low- and high-dose devices.

Compared with patients not treated with paclitaxel-coated devices, those who were tended to be younger and more likely to present with a history CAD, PAD, and other lower limb artery procedures, but less likely to present with a history of stroke, major lower limb amputation, and diabetes. Those treated with paclitaxel-coated devices also were more frequently hospitalized in a private hospital, with less admission via the emergency department, less lower limb ulceration, and less in-hospital mortality than those in the control group.

There were 7.3 deaths per 100 person-years in the paclitaxel-coated devices group and 10.4 deaths per 100 person-years in the control group. In multivariable analyses, the risk of mortality was lower in the paclitaxel-coated device group compared with the control group (HR 0.86; 95% CI 0.84-0.89). Propensity-score matching based on either nearest-neighbor method or exact matching showed similar results.

Additionally, a landmark analysis excluding patients who died in the first 30 days also showed lower mortality with paclitaxel treatment compared with controls (HR 0.87; 95% CI 0.85-0.90).

Other findings across several multivariable models showed a lower risk of mortality, MACE including amputation, and major lower limb amputation in the paclitaxel group, but more new lower limb artery procedures compared with controls. Wargny and colleagues advise that “these results must be interpreted with caution, because it is not known whether the limb concerned is ipsilateral or contralateral to the index lesion.”

As for the decreased mortality in the paclitaxel-treated group, they urge caution there, too, noting that it is “suggestive of a protective effect” though not as pronounced in the nearest-neighbor propensity-score matching (HR 0.92; 95% CI 0.89-0.96) as it was in exact matching (HR 0.82; 95% CI 0.78-0.87).

To TCTMD, Parikh said he “wouldn’t read too much into” the mortality reductions without further data to support them.

In an accompanying editorial, Peter A. Schneider, MD (University of California-San Francisco), and Jeffrey W. Olin, MD (Mount Sinai Hospital, Icahn School of Medicine, New York, NY), note several limitations to the DETECT study, including indication bias that would explain why some patients were treated with paclitaxel-coated devices while others were not and a lack of data on laterality of the second procedure in patients requiring repeat interventions.

Another thing not taken into account, they add, is that if a patient received a BMS in the index procedure but returned for a second procedure that used a DCB, they were still considered part of the BMS group for the purposes of analyzing long-term outcomes.

Schneider and Olin conclude that “the multifactorial nature of the paclitaxel mortality challenge” has demonstrated the importance of real-world data in assessing long-term outcomes for suggestions of harm signals.