Institutional Variation in Patient Radiation Doses During Transcatheter Valve Interventions: A Statewide Experience

JACC Cardiovasc Interv. 2024 Oct 17:S1936-8798(24)01173-7. doi: 10.1016/j.jcin.2024.08.048. Online ahead of print.

Abstract

Background: Little is known about institutional radiation doses during transcatheter valve interventions.

Objectives: The authors sought to evaluate institutional variability in radiation doses during transcatheter valve interventions.

Methods: Using a large statewide registry, transcatheter edge-to-edge mitral valve repair, transcatheter mitral valve replacement, and transcatheter aortic valve replacement procedures between January 1, 2020, and December 31, 2022, with an air kerma (AK) recorded were analyzed. Patient and procedural characteristics were compared between cases with AK ≥2 and <2 Gy. Associations of variables with AK ≥2 Gy were investigated using Bayesian random effects modeling and median ORs for the performing hospital.

Results: Among 9,446 procedures across 30 hospitals, median (Q1-Q3) procedural AK was 0.592 Gy (0.348-0.989 Gy) with AK ≥2 Gy in 533 cases (5.6%). Wide variation in procedural AK was observed, with an institutional frequency of AK ≥2 Gy ranging from 0.0% to 29.5%. Bayesian modeling identified the performing hospital as more strongly associated with the odds of a procedural AK ≥2 Gy than any patient or procedural factors (hospital median OR: 3.54 [95% credible interval: 2.52-16.66]).

Conclusions: In a large, multicenter state-wide registry, there is wide institutional variability in patient-level radiation doses during transcatheter valve interventions, with the performing hospital having a higher odds of an AK ≥2 Gy than any patient or procedural factors. Future interventions are warranted to reduce procedural-related variation in radiation exposure.

Keywords: TAVR; TEER; TMVR; occupational radiation; radiation safety; transcatheter aortic valve replacement; transcatheter edge-to-edge mitral valve repair; transcatheter mitral valve replacement.