Abstract 9132: Balloon Aortic Valvuloplasty Performed Without Heparin is Associated with a Reduction in Vascular and Bleeding Complications rates

Abstract

Objectives: Bleeding and vascular complications are currently the main adverse events associated with balloon aortic valvuloplasty (BAV). We assessed the hypothesis that BAV performed without heparin may be associated with lower complication rates. Methods We conducted a retrospective analysis of all consecutive patients who had undergone BAV in our center between 2008 and 2011. We evaluated 3 groups: patients included between 2008 and 2009 when BAV was performed with large sheaths (10 to 12 French) and injection of unfractionated heparin (UH) (50IU/kg bolus IV) (group 1); patients included after 2009, for whom smaller size sheaths (8 or 9 French) were used, were divided into 2 groups: group 2 (with UH bolus) and group 3 (without UH bolus). We collected all major in-hospital adverse events, bleeding (≥ BARC 3), vascular access complications (including pseudoanerysm or arterio venous fistula) and acute limb ischemia. Results Overall, 140 patients were included in this study. The 3 groups had similar median age (84 years) and preceding lower extremity artery disease (overall n= 36, 27%, p=0.79). Vascular access and bleeding complications were observed in 16 patients (11.4%) and were lower in group 3 compared to group 1 and group 2 (table 1). The use of heparin was associated with an increased risk of vascular access and bleeding complications [relative risk (IC) 2.85 (1.28 to 9.13] whereas absence of heparin did not increase ischemic complications as acute limb ischemia (p=0.82), stroke (1 patient in group 2) or other major in- hospital adverse events (p=0.5). Vascular and bleeding complications were comparable among patients who received heparin with no influence of the used sheath’ s size (table 1). In conclusion, we showed in this study that balloon aortic valvuloplasty performed without heparin is safe and leads to a dramatic reduction of vascular and bleeding events. Although randomization was not used, this marked difference is difficult to explain by confounding factors. .