Outcomes of trans-catheter versus surgical aortic valve replacement for patients with classic low-flow low-gradient aortic stenosis

Abstract Background Classical low-flow low-gradient aortic stenosis (cLFLG-AS) is frequently encountered within clinical practice [1], and is often associated with high mortality [2]. Whilst aortic valve replacement can improve outcomes, there is a lack of real-world data guiding the decision between trans-catheter aortic valve implantation (TAVI) or surgical aortic valve replacement (SAVR) in this specific patient cohort. Purpose To compare outcomes between TAVI and SAVR in patients with cLFLG-AS. Methods cLFLG-AS was defined by an aortic valve area <1cm2, mean gradient <40mmHg or peak velocity <4m/s and left ventricular ejection fraction <50%. All patients who met cLFLG-AS criteria who underwent TAVI or SAVR at our tertiary referral centre between 2015 and 2020 were included. Inverse probability weighting was used to adjust for differences in baseline characteristics and the non-random assignment of treatment modalities. The primary end-point was all-cause mortality. Secondary end-points were procedure-related permanent pacemaker implantation, renal replacement therapy (RRT) and stroke. Results 322 patients were included (220 TAVI and 102 SAVR). Baseline characteristics were: male sex 70%, mean gradient 30mmHg (23–35), aortic valve area 0.8cm2 (0.6–0.9). Patients undergoing TAVI were older than the SAVR group (81.3±8.5 vs 70.6±11.1 years, p<0.0001) with higher Logistic Euroscore (19.3 (11.8–32.3) vs 7.1 (3.7–14.1), p<0.0001) (figure 1). At median follow-up of 2.7 years (1.5–4.1), 99 patients had died; 70 (31.8%) had TAVI and 29 (28.4%) had SAVR. Both unweighted and weighted Kaplan-Meier curve analysis was performed; there was similar survival between SAVR and TAVI (log rank test 0=0.27 and p=0.4 respectively) (figure 2). Adjusted hazard ratio (HR) for mortality with SAVR was 0.78 unweighted (95% CI 0.5–1.21; p=0.27) and 0.66 weighted (95% CI 0.26–1.64; p=0.37). In terms of contributory procedural factors, concomitant multiple valve intervention in the SAVR group independently affected mortality (HR 5.47, 95% CI 2.52–11.51, p<0.001). There was no difference in permanent pacemaker insertion or stroke across the two groups, but rates of RRT were higher in SAVR cohort (13.7% vs 0%, p<0.001). Conclusions Despite the TAVI cohort being both older and at higher risk, there was no observed difference in mortality between TAVI and SAVR in the mid-term. Deciding upon intervention choice can therefore be based upon criteria other than the specific presence of a classic low-flow low-gradient state. Funding Acknowledgement Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Clinical training research fellowship from the British Heart Foundation Baseline characteristics by treatmentKaplan-Meier curves for TAVI vs SAVR