599 Gender-based differences in TAVI outcomes: report from a large contemporary real-world population of self-expandable valves

Abstract

Abstract Aims Small sub-study data derived from randomized clinical trials suggest a gender-based disparity in TAVI outcomes. However, large real-world contemporary data is missing. The aim of this study is to compare the risk factors, procedural characteristics and clinical outcomes of male and female patients who underwent transcatheter aortic valve implantation (TAVI) using two next-generation self-expandable bioprostheses (ACURATE neo and Evolut R/Pro valves). Methods We performed a first unmatched comparison and a propensity score-matched analysis (PSM) to assess the outcomes derived by the sex difference beyond the impact of pre-procedural risk factors in a large, contemporary, real-world, multicentre, international, retrospective registry of 3862 consecutive patients. The primary endpoint was a composite of all-cause death or any stroke (disabling and non-disabling) at 1 year. Results Sixty-four per cent (2162/3353 patients) of the study cohort was female and was older (mean age 82.3 years vs. 81.1 years for men (P<0.001)) had a higher BMI (27.7±5.7 for women vs. 27.2±4.5 for men), and lower prevalence of dyslipidaemia (50.2% vs. 54.7, P=0.037), diabetes (26.8% vs. 33.7, P<0.001), smoking (10.0% vs. 24.3%, P<0.001), COPD (17.4% vs. 21.9%, P=0.002), pacemaker/ICD (9.6% vs. 14.0%, P<0.001), previous cardiac surgery (8.6% vs. 18.8%, P<0.001), previous PCI (23.0% vs. 36.8%, P<0.001). Mean STS score for women was higher 5.2±3.9% vs. 4.5±3.4% (P<0.001). Women had higher mean valve gradients (45.4±17.1 vs. 42.7±14.7 mmHg; P<0.001), smaller valve areas (mean 0.7 cm2 vs. 0.9 cm2, P=0.037) and smaller annular perimeters (56.8±23.0 vs. 62.0±23.8, P<0.001). The primary endpoint was resulted in a rate of 7.9% vs. 6.9% (P=0.337) in the unmatched population and 9.4% vs. 6.0% (P=0.014) after the PSM, respectively for women and for men. Independently, there was no difference in mortality (5.9% vs. 5.6%; P=0.786) and stroke (2.5% vs. 1.8%; P=0.243) rates between women and men in the un-matched groups. Rates of cardiac tamponade (1.5% vs. 0.4%, P=0.008), major vascular complications (7.7% vs. 4.1%, P<0.001), life-threatening bleeding (2.8% vs. 1.4%, P=0.016), major bleeding (5.1% vs. 2.9%, P=0.004), need of transfusion (8.9% vs. 4.6%, P<0.001) and acute kidney injury (8.5% vs. 5.7%, P=0.009), were all significantly higher in women. After PSM, mortality was similar between the two groups (11.3% for women vs. 9.5% for men, P=0.264) but strokes were more prevalent in women (2.8% vs. 1.2%, P<0.024). Furthermore, in the matched population, major vascular complications (6.8% vs. 4.1%, P=0.024), need of transfusion (9.1% vs. 4.6%, P<0.001) and acute kidney injury (8.7% vs. 5.6%, P=0.009) remained significantly different between women and men, respectively. Conclusions In this large real-world contemporary TAVI registry, female gender was associated with higher rates of stroke, vascular complications, major bleeding, and acute kidney injury. Further studies are required to explore the underlying pathophysiological mechanisms for these observations.