Patient-level and external factors in the use of drug-coated balloons and drug-eluting stents in femoropopliteal endovascular interventions

Objective: Drug-coated balloons (DCB) and drug-eluting stents (DES) have been rapidly adopted for femoropopliteal endovascular interventions due to their favorable patency rates. It is unclear whether choice of using drug coated devices versus bare metal stents (BMS) or plain balloon angioplasty (POBA) as primary treatment in femoropopliteal disease is mostly associated with patient-level factors, safety concerns, or by operator preferences. This study sought to evaluate factors associated with their use in a contemporary dataset.

Methods: All femoropopliteal lesions treated with endovascular interventions between 2016 and 2019 from the Vascular Quality Initiative registry were included. For each procedure, a primary treatment was identified based on the following hierarchy: DES > DCB > BMS > POBA. A hierarchical logistic regression model predicting DCB or DES use included patient-level characteristics, key events (period after Centers for Medicare and Medicaid Services reimbursement change, January 2018 [vs before] and period after Katsanos meta-analysis December 2018 [vs before]), and random effects for site and operator. Operator-level variability for DCB and DES use was summarized with an adjusted median odds ratio (MOR).

Results: A total of 57,753 femoropopliteal endovascular procedures were included. Poor functional status (odds ratio [OR], 0.92; 95% confidence interval [CI], 0.90-0.94), prior anticoagulant use (OR, 0.92; 95% CI, 0.87-0.97), higher Rutherford classification (OR, 0.86; 95% CI, 0.84-0.88), chronic kidney disease stage 4 or 5 (OR, 0.92; 95% CI, 0.86-0.98), and the period after the Katsanos meta-analysis publication (OR, 0.3; 95% CI, 0.29-0.32) were associated with a lower odds of DCB or DES use; whereas female sex (OR, 1.12; 95% CI,1.08-1.17), prior lesion treatment (OR, 1.17; 95% CI, 1.11-1.22), diabetes (OR, 1.07; 95% CI, 1.02-1.12), Trans-Atlantic Inter-Society Consensus class B (OR, 1.16; 95% CI, 1.09-1.24) and C (OR, 1.2; 95% CI, 1.12-1.28), and the period after the Centers for Medicare and Medicaid Services reimbursement change (OR, 1.08; 95% CI, 1.03-1.14) were associated with a higher odds of DCB or DES use. Significant variability in use was found across operators (adjusted MOR, 2.70; 95% CI, 2.55-2.85) and centers (adjusted MOR, 2.89; 95% CI, 2.50-3.27).

Conclusions: DCB or DES use in femoropopliteal disease demonstrates wide variability across operators and is linked strongly with external factors, followed by anatomic lesion characteristics and a history of previous interventions. Future work needs to focus on tailoring DCB or DES use to patient and lesion characteristics and to develop appropriate use guidelines integrating these factors.