AI Muscle Scans Predict Heart Valve Surgery Survival

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 the muscle numbers turned out to matter more than most of the usual risk factors.^[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.

That’s a meaningful upgrade over how muscle loss is usually assessed in this setting. The older standard method looks at a single CT slice and estimates from there; full AI-based volumetry measures the muscle directly, across the whole scan.

What the muscle numbers predicted

The study followed 470 TAVI patients for a median of about 4.6 years. 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. But the muscle measurement wasn’t just a marker of frailty in general — it independently predicted survival. 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 an independent predictor of death (hazard ratio 1.589, meaning roughly 59% higher risk).

Muscle quality — the intermuscular fat measurement — did not reach statistical significance as an independent predictor (hazard ratio 1.326, p = 0.196). In this particular population and outcome, how much muscle a patient had mattered 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 another data point in a growing body of evidence that whole-body muscle volume is a hard, independent predictor of survival in older adults facing major medical procedures — not a secondary detail.

• 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 frequently already sitting in a scan taken for another purpose.
• Treat muscle-building as pre-habilitation, not just recovery. If a procedure is scheduled weeks or months out, resistance training in the lead-up is a legitimate way to improve the muscle-volume numbers this research ties to survival — not just a “nice to have.”
• Muscle mass belongs in the same conversation as blood pressure and cholesterol. This study adds heart valve surgery to a growing list of major medical outcomes (alongside cancer treatment, ICU recovery, and general surgery) where muscle volume independently predicts who does well and who doesn’t.

The heart gets the headline in a valve procedure. Increasingly, the muscle everywhere else in the body is turning out to be just as predictive of the outcome.

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, 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.