P69 LONG–TERM OUTCOMES WITH DRUG–ELUTING BALLOONS FOR THE TREATMENT OF IN–STENT RESTENOSIS AND DE NOVO LESIONS

Abstract Introduction Drug coated balloons (DCB) have emerged for percutaneous coronary interventions (PCI) and mainly for in–stent restenosis or particular anatomies. However, the indications and the predictors of long–term failure of DCB have been poorly evaluated besides small–sized randomized clinical trials. Therefore, the aim of the present study was to provide a real–world analysis of the prognostic determinants and long–term outcomes among patients treated with DCB for any type of lesion and included in a comprehensive multicenter registry. Methods We included patients undergoing coronary angiography and PCI with DCB for in–stent restenosis or de novo lesions in 3 centers. Quantitative parameters for coronary lesions were calculated by an automatic edge–detection system. The primary study endpoint was the occurrence of major cardiovascular events (a composite of death, MI and target vessel revascularization) at the longest available follow–up. Secondary endpoints were the individual components of the primary endpoint, target lesion failure (TLF) or any acute coronary syndrome. Results Out of 281 patients treated with DCB, the 267 displaying a follow–up > 12 months were included, of whom 196 treated for in–stent restenosis and 71 with de novo lesions. At a median follow–up of 616 [368–1025] days, MACE occurred in 70 (26.2%) of the patients. No difference in clinical, demographic of angiographic features was observed between patients with or without an event, with the exception of a higher rate of in–stent restenosis (p = 0.04), longer and more type C lesions (p = 0.05 and p = 0.04) related with MACE. At multivariate Cox–regression, type C lesions emerged as the only independent predictor of MACE (adjusted OR[95%CI]= 1.83[1.13–2.97], p = 0.014), mainly driven by target vessel revascularization (adjusted OR[95%CI]= 1.78[1.05–2.95], p = 0.03) although not conditioning survival. However, in–stent restenosis emerged as a major determinant of TLF (adjusted OR[95%CI]= 2.59[1.17–5.75], p = 0.02). Conclusion The present registry shows that drug–coated balloons represent a potential treatment strategy even for de–novo lesions, especially in less complex cases. In fact, we observed an increased risk of MACE and target lesion failure in case of type C and restenotic lesions.