Background: Deferring interventions for valvular heart disease (VHD) leads to increased major adverse cardiovascular events (MACE) rates, but the underlying mechanisms are not fully understood. Progressive congestive heart failure (CHF) is associated with both delayed intervention and MACE, suggesting it may serve as a key pathomechanistic mediator.
Methods: Consecutive patients whose non-emergency cardiac intervention for VHD, coronary artery disease (CAD), or rhythmological disorder was postponed during the COVID-19-related lockdown between March 19th and April 30th, 2020 were included (n = 178). Progressive CHF was defined as NT-proBNP > 900 pg/ml at the actual intervention date, in combination with worsening dyspnea (measured by NYHA class), emergency heart failure hospitalization, or declining left ventricular ejection fraction (LVEF) during the prolonged waiting time. The combined endpoint MACE was defined as emergency hospitalization or death during the follow-up period until 36 months post-procedure. Causal mediation analysis using 5,000 nonparametric bootstrap resamples (Hayes PROCESS Model 4) was performed to assess whether CHF progression mediated the relationship between the index disease and MACE. Significant models were adjusted for baseline confounders.
Results: Serial CHF data were available for 39/49 VHD, 71/74 CAD, and 49/55 rhythmological patients. Progressive CHF during the 23 (19, 38) day prolonged waiting time occurred in 61.5%, 24.3%, and 24.4% of these patients. Post-interventional MACE incidence was 46.9%, 33.8%, and 21.8%, respectively. In the VHD cohort, progressive CHF significantly mediated the association between VHD and MACE (indirect effect a×b = 0.112, 95%-CI [0.030–0.221]; p = 0.022), accounting for 20% of the total effect. After adjustment for the confounders of age and pre-existing arrhythmia, mediation proportion increased to 150% (a×b = 0.149, 95%-CI [0.028–0.223]; p = 0.022). In contrast, no significant mediation was observed in the CAD cohort (a×b = –0.232, 95%-CI [–0.539 to 0.031]; p = 0.105) or the rhythmological cohort (a×b = –0.268, 95%-CI [–0.566 to –0.012]; p = 0.050). In both of these cohorts, the index disease was not significantly associated with CHF progression. However, CHF itself remained a significant predictor of MACE in all groups (b-values: VHD 0.736, VHD adjusted 0.430, CAD 1.697, rhythmological disorder 1.171).
Conclusion: Progressive CHF prior to intervention mediates the relationship between VHD and post-interventional MACE and is also independently associated with MACE in patients with CAD and rhythmological disorders. These findings suggest that the progression of CHF is an important pathogenetic mechanism leading to MACE and should therefore be the focus of prevention efforts.
