AI Muscle Scans Linked to Survival After Heart Valve Surgery

Before a common heart valve procedure, the CT scan a patient already gets may be quietly predicting who’s at higher risk of dying in the years that follow — if anyone bothers to measure the muscle on it. A new Austrian study used AI to do exactly that, and low muscle volume was linked to a higher risk of dying in the years that followed.^[1]

The procedure, and the scan it already requires

Transcatheter aortic valve implantation (TAVI) is a minimally invasive way to replace a diseased heart valve without open-heart surgery — it’s become the standard option for many older patients with aortic stenosis (a narrowed heart valve). Every TAVI patient gets a detailed pre-procedure CT scan for surgical planning. Researchers at Paracelsus Medical University in Salzburg realized that same scan, run through AI-based segmentation software (TotalSegmentator), could also measure something nobody was routinely tracking: the total volume of skeletal muscle and intermuscular fat throughout the body.

The older standard method looks at a single CT slice and estimates from there; full AI-based volumetry measures the muscle across the whole scan. This study did not compare the two methods, so it cannot say whether volumetry is better.

What the muscle numbers predicted

This was a retrospective study at a single center, following 470 TAVI patients for a median of about 4.6 years; 188 of them (40.0%) died during follow-up. Patients were split into sex-specific quartiles by skeletal muscle volume (SMV) and by intermuscular fat (IMF, a marker of muscle quality — how much fat is infiltrated into the muscle tissue itself). Patients in the lowest quarter for muscle volume were, unsurprisingly, older and had lower BMI. The muscle measurement stayed associated with survival after adjustment for other factors, although the study cannot say whether low muscle volume is a cause or simply a marker of frailty. Patients in the lowest SMV quartile had significantly higher mortality (log-rank p = 0.009), and after adjusting for other factors in a multivariable model, low muscle volume remained independently associated with death (hazard ratio 1.589, meaning roughly a 59% higher hazard of death over follow-up in the lowest quartile compared with the other patients).

Muscle quality — the intermuscular fat measurement — did not reach statistical significance as an independent predictor (hazard ratio 1.326, p = 0.196). The confidence interval for that figure was wide (0.865 to 2.035), so this does not show that muscle quality is unimportant; the authors suggest muscle quantity appears to be the dominant factor. In this particular population and outcome, how much muscle a patient had looked to matter more than how “clean” that muscle tissue was.

What this means for you

Most people reading this aren’t facing a heart valve procedure. The relevant takeaway is broader: this is one data point, from a single retrospective cohort of TAVI patients, linking whole-body muscle volume to survival. It is an association, and it does not show that changing muscle volume changes survival.

• If you or a family member is facing any major procedure later in life, ask whether pre-procedure imaging could be used to assess muscle status. As this study shows, the data is often already sitting in a scan taken for another purpose. AI body-composition reporting is not yet a validated clinical standard for TAVI decisions; the authors say it may serve as an imaging biomarker.
• Muscle-building before a procedure is our suggestion, not something this study tested. Nobody in this study changed their muscle volume, and low muscle volume in these patients may reflect frailty or illness rather than a cause you can change, so we can’t say that training improves survival. If you have a heart condition (including aortic stenosis) or a procedure scheduled, get your cardiology team’s clearance before starting resistance training.

The heart gets the headline in a valve procedure. In this one cohort, though, the amount of muscle in the rest of the body was also linked to who survived.

References

1. Clodi N, Knapitsch C, Hecke G, et al. “Artificial intelligence-based whole-body skeletal muscle volume predicts long-term mortality after transcatheter aortic valve implantation.” Experimental Gerontology, 2026;223:113270. Finding used: among 470 TAVI patients followed for a median of 4.56 years, patients in the lowest quartile for AI-measured skeletal muscle volume had significantly higher mortality (log-rank p=0.009), and low skeletal muscle volume remained an independent predictor of all-cause mortality after multivariable adjustment (HR 1.589, 95% CI 1.101-2.294, p=0.013), while intermuscular fat (muscle quality) was not an independent predictor (HR 1.326, 95% CI 0.865-2.035, p=0.196). DOI: 10.1016/j.exger.2026.113270. Source: PubMed 42567492. ↑︎

Kurt Greiner

Kurt is a digital strategist and IT professional blending emerging technology with practical application to help businesses and individuals streamline their digital presence. His current work focuses on the intersection of intentional living and technological resilience, exploring how individuals can leverage modern tools to navigate the second half of life with purpose.