Mortality after Use of Paclitaxel-Coated Balloons Correlates with Total Cumulative Dosage of Paclitaxel in Real-World Analysis

Abstract

This study used independent, real-world, patient-level data to examine whether the dosage or frequency of paclitaxel exposure correlated with mortality during follow up. We conducted a retrospective analysis of patients treated with a drug-coated balloon (DCB) for an atherosclerotic femoropopliteal lesion from February 2013 to December 2018, excluding patients with non-atherosclerotic lesions or restenosis after DCB treatment in another hospital. We investigated the causes of death, comorbidities (including cancer status), and the initial and total cumulative dosages and frequency of paclitaxel use. To determine whether the dosage or frequency of paclitaxel exposure affected mortality during follow up, we analyzed the risk factors for all-cause death by conducting a time-dependent Cox regression analysis that considered demographics, comorbidities, lesion and procedural characteristics, and paclitaxel exposure data (dosage and frequency). Our analysis examined 225 patients (mean age 71 ± 9 years, range 38-93 years, male 81%). During a mean follow-up duration of 35 months (range 1-89 months), 56 patients (24.9%) died from cardiac disorders (16%, including acute myocardial infarction, heart failure, or sudden cardiac arrest), malignancy (14.3%), respiratory failure with pneumonia (12.5%), septic shock (12.5%), or another cause. Univariable and multivariable Cox regression analyses identified age (hazard ratio, HR, 1.057; 95% confidence interval, CI, 1019-1096; p = 0.0032), critical limb ischemia (CLI) (HR, 4135; 95% CI, 2171-7876; p < 0.0001), and the total dosage of paclitaxel (mg) (HR, 1.040; 95% CI, 1006-1074; p = 0.0210) as predictors of mortality during follow up. The subgroup analysis found that the total dosage of paclitaxel (mg) was also a predictor of mortality during follow up in the CLI group (HR, 1.046; 95% CI, 1007-1087, p = 0.0198). The estimated cut-off value of total cumulative paclitaxel dosage for predicting mortality was 12 mg as evaluated by minimum p value approach. This patient-level analysis identified the total cumulative dosage of paclitaxel as a predictor of mortality after the use of paclitaxel-coated balloons. Our results provide limited information about the potential dose-response relationship underlying paclitaxel-associated mortality concerns.