Baseline FIB-4 index is associated with persistent tricuspid regurgitation but provides no incremental predictive value beyond age after transcatheter atrial septal defect closure in adults: a longitudinal cohort study

Transcatheter closure is the first-line treatment for secundum atrial septal defect (ASD) and induces right cardiac reverse remodelling; however, residual pulmonary hypertension (residual PH) persists in 10%–30% of patients at 1 year. The fibrosis-4 (FIB-4) index is a validated surrogate marker of hepatic fibrosis with prognostic value in heart failure. This study aimed to characterise the longitudinal evolution of right cardiac parameters after closure and to determine whether FIB-4 provides incremental predictive value beyond established predictors for persistent moderate-to-severe tricuspid regurgitation (TR) and residual PH. This single-centre retrospective cohort study enrolled adults undergoing transcatheter secundum ASD closure (January 2021–April 2025). The primary endpoint was residual PH at 1 year, defined by the 2022 ESC/ERS echocardiographic high-probability criteria; the secondary endpoint was persistent moderate-to-severe TR. Multivariable logistic regression and likelihood-ratio tests assessed associations and the incremental value of FIB-4 in nested models; discrimination (C-statistic), calibration, Brier score and decision curve analysis were reported. Longitudinal trajectories were assessed with linear mixed-effects models. Of 143 patients with 1-year follow-up, 135 were included in the primary analyses after excluding 8 peri-operative pulmonary vasodilator users; residual PH and persistent moderate-to-severe TR occurred in 34 (25.2%) and 23 (17.0%) patients. TRV was the strongest predictor of residual PH (adjusted OR [ aOR ] 11.64, 95% CI 3.39–40.00). FIB-4 was associated with persistent TR adjusted for baseline TR grade ( aOR 3.047, 95% CI 1.165–7.970) but not after additional adjustment for age ( aOR 1.394, 95% CI 0.396–4.909; P = 0.605), and added no incremental predictive value beyond TR grade and age (likelihood-ratio test P = 0.609) or beyond TRV, age and NYHA class ( P = 0.522). FIB-4 correlated weakly with right-sided haemodynamic parameters ( rs = 0.27–0.34), but age correlated more strongly ( rs = 0.36–0.37). In this cohort, FIB-4 was associated with persistent TR and right-sided haemodynamic parameters, but these associations were largely explained by age, and FIB-4 provided no incremental predictive value beyond established predictors. Age-containing composite indices such as FIB-4 should be used with caution as independent risk-stratification tools in adult ASD. Longitudinal characterisation of reverse remodelling remains valuable; multicentre prospective validation is warranted.