Structural Cardiology · JACC (TCT-30) · 2025

Keeping the Outflow Open in TMVR

Published - First Author Read in JACC ↗

The Scenario

Replacing the mitral valve through a catheter can block the heart's own outflow tract. Several techniques exist to prevent that, and almost no trials compare them.

Each prevention technique lives in its own small case series, so a conventional two-arm meta-analysis was impossible. The comparison had to be built indirectly, from pooled single-arm proportions, technique by technique.

From Question to Answer

Clinical question

Which LVOT obstruction prevention technique carries the best success and safety profile?

The evidence

Small single-arm series per technique, extreme event proportions, no head-to-head trials

The design

Transformed proportion pooling with technique subgroups, plus paired pre and post hemodynamics

The answer

A structured cross-technique comparison, presented at TCT and published in JACC

The Decisions That Mattered

Freeman-Tukey double arcsine transform for the pooling.

Proportions near 0 or 100 percent misbehave on the raw scale. The transform stabilizes their variance so small series do not distort the pool.

Subgroup pooling by prevention technique.

With no comparative trials, pooling each technique separately and reading them side by side is the honest version of the comparison clinicians wanted.

Paired pre and post analysis of neo-LVOT area and gradients.

Within-patient change in anatomy and hemodynamics shows the mechanism worked, not just that patients survived.

Dual extraction with independent validation.

Every value was extracted twice and cross-checked before it touched the model. In a literature of small series, one transcription error can move a pooled estimate.

Overview

Problem

LVOT obstruction is a feared TMVR complication with several prevention techniques and no comparative trials.

Approach

Freeman-Tukey proportion meta-analysis with technique subgroups, paired hemodynamic deltas, and dual-validated extraction.

Outcome

Published as TCT-30 in the Journal of the American College of Cardiology, October 2025, first author.

Figures

Illustrative figure from simulated data; the real analysis stays with the journal and the clinical team.

Reproducible R Code

Subgrouped proportion pooling

# Pooled proportions, stabilized, grouped by technique
m <- metaprop(event = events, n = total,
              studlab = series,
              sm = "PFT",              # Freeman-Tukey double arcsine
              method.tau = "DL",
              subgroup = technique)

forest(m, xlab = "Event proportion")

# Mechanism check: within-patient change after the technique
delta <- post_gradient - pre_gradient
t.test(delta)                          # paired shift in gradients

Publication

Read in JACC

Read publication ↗