Back to all papers

Artificial intelligence-based whole-body skeletal muscle volume predicts long-term mortality after transcatheter aortic valve implantation.

August 7, 2026pubmed logopapers

Authors

Clodi N,Knapitsch C,Hecke G,Brandstetter L,Scharinger B,Dinges C,Hammerer M,Hoppe UC,Hergan K,Schörghofer N,Boxhammer E

Affiliations (4)

  • Department of Radiology, Paracelsus Medical University of Salzburg, Salzburg, Austria.
  • Department of Cardiovascular and Endovascular Surgery, Paracelsus Medical University of Salzburg, Salzburg, Austria.
  • Department of Internal Medicine II, Division of Cardiology, Paracelsus Medical University of Salzburg, Salzburg, Austria.
  • Department of Internal Medicine II, Division of Cardiology, Paracelsus Medical University of Salzburg, Salzburg, Austria. Electronic address: [email protected].

Abstract

Frailty and sarcopenia are important determinants of outcomes in patients undergoing transcatheter aortic valve implantation (TAVI). Conventional skeletal muscle depletion assessment relies on single-slice measurements that do not capture total muscle burden. Artificial intelligence (AI)-based CT segmentation enables automated volumetric quantification of skeletal muscle and intermuscular fat (IMF). We investigated the prognostic value of AI-derived skeletal muscle volumetry and IMF in patients undergoing TAVI. This retrospective cohort study analyzed pre-procedural CT scans of patients undergoing TAVI using AI-based segmentation (TotalSegmentator, Basel, Switzerland) to quantify skeletal muscle volume (SMV) and IMF. Patients were stratified by sex-specific quartiles. Skeletal muscle depletion was defined as the lowest SMV quartile (Q1), while muscle quality was assessed by the highest IMF quartile (Q4). The primary endpoint was all-cause mortality. A total of 470 patients were included. During a median follow-up of 4.56 years (IQR 3.57-5.99), 188 patients (40.0%) died. Patients with low SMV were older with lower body mass index and body surface area. Kaplan-Meier analysis demonstrated higher mortality in the lowest SMV quartile (log-rank p = 0.009). In multivariable Cox regression, low SMV remained independently associated with mortality (HR 1.589, 95% CI 1.101-2.294, p = 0.013). High IMF was not associated with mortality (HR 1.326, 95% CI 0.865-2.035, p = 0.196). In a combined model, SMV remained independently associated with mortality, whereas IMF remained non-significant. AI-based CT volumetry identifies low skeletal muscle volume as an independent predictor of long-term mortality after TAVI. Muscle quantity, rather than muscle quality, appears to be the dominant prognostic determinant. AI-driven body composition analysis may serve as an imaging biomarker of biological aging and reduced physiological reserve in older adults undergoing TAVI.

Topics

Journal Article

Ready to Sharpen Your Edge?

Subscribe to join 11k+ peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.