Long-term follow-up of the TRED-HF trial

Abstract

Abstract Background The TRED-HF trial (1) found that 44% of patients with recovered dilated cardiomyopathy (DCM) relapsed in the short term after withdrawal of medical therapy. The longer-term risk of relapse is uncertain. Purpose This follow-up aims to investigate the longer-term risk of relapse in patients with recovered DCM and any possible predictive factors. Methods TRED-HF trial participants were followed until May 2023. The primary outcome was DCM relapse defined as >10% reduction in left ventricular (LV) ejection fraction to <50%, doubling in N-terminal pro-B-type natriuretic peptide (NT-pro-BNP) to >400ng/L, or clinical recurrence of heart failure (HF). Therapy intensity for HF was calculated using QUAD scores. The incidence of relapse was calculated using Kaplan-Meier estimates. Predictors of relapse were investigated using Cox proportional hazard modelling. Results For all fifty-one patients, median follow-up from enrolment was 6 (IQR: 6-7) years. Overall, thirty-three patients (65%) relapsed with an estimated 5-year relapse rate of 61% (95% CI 45-73). Twenty patients relapsed during the randomised trial, nine of whom had a recurrent relapse during follow-up (Figure 1). Thirteen patients relapsed for the first time after completion of the trial; of those, one patient did not have therapy withdrawn during the trial, 8 had restarted low intensity HF therapy and 4 patients stayed on no therapy. Five patients stayed off HF therapy and did not relapse. The overall mean intensity of HF therapy was lower after the trial compared to enrolment (mean difference in QUAD score -6 [-8 to -4]; p<0.001). A mean reduction in medication intensity was also seen at the point of relapse during the trial and during follow-up (Figure 2). Of relapses during follow-up, seven had identifiable triggers (arrhythmia (n=4), pregnancy (n=1), hypertension (n=1), infection (n=1)). Patients taking more HF medications at baseline or with higher baseline NT-pro-BNP were more likely to relapse. Corrected atrial fibrillation was associated with reduced future risk of relapse. Conclusions In patients with recovered DCM, the risk of relapse after therapy withdrawal increased with longer follow-up. Lower doses of HF medications and recurrent external triggers contributed to relapse. Few patients who stopped medications had sustained remission. Recovery of DCM after correction of AF was associated with a low risk of relapse provided atrial fibrillation did not recur.