gms | German Medical Science

Background and current state of (inter)national research: Drug-eluting devices (DED) are well established for applications in endovascular revascularization (EVR) of peripheral arteries. Recently, there has been a broad discussion about possible increased long-term mortality. These safety concerns have led to several warnings for the use of paclitaxel-based DED by international regulatory authorities.

Research questions and objectives: Our aim of the study was to analyze the mortality of paclitaxel based DED in a large “real world” dataset for their applications in peripheral endovascular therapy.

Methods or hypothesis: Based on the data of the BARMER Health Insurance, the application of paclitaxel-based drug-eluting stents (DES) and drug-coated balloons (DCB) was assessed from market introduction until present. We indexed all adult (age ≥ 18 years) patients with first peripheral EVR between 01.01.2007 until 31.12.2015 and followed them until 31.12.2017. A multivariable Cox regression model with time-dependent co-variables was fitted to evaluate overall survival (OS) adjusted by patient’s risk profiles. The analysis includes all DES, DCB, bare-metal stents and uncoated balloon devices that were applied at index and during the follow-up (FU) period.

Results: We identified 64 771 patients (mean age 71.5 years, 45.3% female sex) who underwent peripheral index EVRs (5.1% with DED). Lower extremity artery disease (LEAD) was encoded in 94.9% of patients with 42.7% at critical stages of the disease. Adjusted by common cardiovascular risk factors, e.g. sex, age, stage of LEAD, chronic kidney disease, diabetes, heart failure, previous cardiovascular events and procedures, a time-dependent multivariable Cox regression analysis was performed to address OS. During 11 FU years, no statistical indication for a decreased OS related to the paclitaxel-eluting effect of DES could be identified in our dataset (all p>0.05). DCB showed a slightly protective effect on OS during the first year after EVR (HR 0.92, 95%-CI [0.89,0.94], p<0.001), but not in later years (all p>0.05).

Discussion: In our dataset, an association between the application of DED in peripheral EVRs with a decreased OS could not be proven. The result of this study [1] gives a first answer on the serious and recent debate on the sensitive issue of patient safety in the use of DED based on the unexpected results of a meta-analysis [2]. However, compared to this analysis2 based on small-sized selected cohorts of the underlying RCTs without further knowledge about the therapy during FU, our study1 reflects an unselected cohort of 64 771 patients with 107 112 EVRs over a median time period of 92 months.

Health insurance data may provide a favorable opportunity to monitor patient’s safety after regulatory approval. Based on 9.2 million insurants, our study illustrates the current practice in endovascular treatment of LEAD to a high degree.

Practical implications: Our study [1] does not observe an increased mortality after the use of paclitaxel-coated devices. However, it illustrates well that analyzing insurance data is suitable to quickly provide answers to questions about patient’s safety.