Time- and dose-dependent high-sensitivity cardiac troponin-T to improve outcome prediction after TAVI: a multicenter...

Pre- and post-procedural high-sensitivity cardiac troponin T (hs-cTnT) detects periprocedural myocardial injury (PPMI) and predicts adverse outcomes following transcatheter aortic valve implantation (TAVI). Current diagnosis of PPMI relies on fixed cutoffs, lacking the integration of time- and dose-dependent effects of hs-cTnT. This limits the precision of risk stratification and subsequent patient management.

To investigate the non-linear and time-dependent effects of pre- and post-procedural hs-cTnT levels on outcomes after TAVI, these findings were compared to the dichotomized definition of PPMI proposed by the Valve Academic Research Consortium-3 (VARC-3).

Consecutive patients undergoing TAVI between 2011 and 2024 at two tertiary university hospitals with available hs-cTnT measurements were enrolled. The primary outcome was all-cause mortality at 1 year. Multivariable Cox proportional hazards models were fitted. To relax the proportional hazards assumption, allowing for hazard ratios (HRs) to vary over time and across hs-cTnT values, a Royston-Parmar model was fitted.

Among 5158 patients, the HR for all-cause mortality at 1 year associated with VARC-3 defined PPMI was not statistically significant. Continuous variable analysis showed that both higher pre- and post-procedural hs-cTnT levels correlated with increased all-cause mortality risk at 1 year. Time-dependent models revealed the hazard to be greatest for higher hs-cTnT levels early post-procedurally and to decline over time.

The dichotomized VARC-3 definition of PPMI showed no prognostic value. Modelling hs-cTnT as continuous and time-dependent revealed a dynamic risk trajectory after TAVI. Incorporating these non-linear and time-dependent effects into risk prediction models may improve clinical decision-making and personalize post-procedural surveillance.