# Right-Sided Extravalvular Damage: An Overlooked Driver of Risk Model Failure in Low-Flow, Low-Gradient Aortic Stenosis

## TL;DR

This study shows that right-sided heart damage improves risk prediction for aortic stenosis patients undergoing TAVR, especially in moderate-risk groups.

## Contribution

Identifies right-sided extravalvular damage as a novel factor improving mortality prediction in low-flow aortic stenosis patients.

## Key findings

- Right-sided extravalvular damage is independently linked to higher 1-year mortality after TAVR.
- Adding these markers improved model fit and discrimination, particularly in moderate-risk patients.
- The STS-PROM score overestimates mortality, but this can be corrected with right-sided damage markers.

## Abstract

The Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) score is guideline-endorsed for estimating procedural risk in aortic stenosis (AS) but tends to overestimate long-term mortality after transcatheter aortic valve replacement (TAVR). This study aimed to evaluate whether markers of right-sided extravalvular damage improve STS-PROM’s performance in predicting 1-year mortality post-TAVR for patients with low-flow, low-gradient AS.

Four hundred ten patients with valve area ≤1.0 cm2, stroke volume index ≤35 mL/m2, and mean gradient <40 mm Hg who underwent TAVR were retrospectively stratified into low (<4%), moderate (4%-8%), and high (>8%) STS-PROM risk groups. Performance of an enhanced logistic regression model incorporating both STS-PROM and right-sided extravalvular damage (defined as moderate+ pulmonary hypertension, moderate+ tricuspid regurgitation, or right ventricular systolic dysfunction) was assessed using receiver operating characteristic curves, net reclassification index (NRI), and integrated discrimination index (IDI).

Right-sided extravalvular damage was independently associated with increased 1-year mortality (OR, 2.45; 95% CI, 1.34-4.46; P < .01). Model fit improved with its addition (Δχ2 = 8.68, P < .01), and discrimination increased (IDI = 0.02, P = .01). NRI analysis showed improved survivor classification (non-event NRI = 0.10, P < .01). The largest gain was in moderate-risk patients (area under the curve, 0.46-0.67; P = .06).

Markers of right-sided extravalvular damage conferred meaningful prognostic value in low-flow, low-gradient AS by correcting the STS-PROM score’s tendency to overestimate post-TAVR mortality. The added value of these markers was most pronounced in moderate-risk patients, for whom existing tools appear least reliable. These markers may improve patient selection for TAVR and merit evaluation in other severe AS phenotypes.