Ascertainment Bias
The range of incident HF could be from 6% (chart review only) to 20% (chart review plus telephone survey). Similar to Huang et al, we performed a prospective study that included a prespecified telephone survey at the end of follow-up. (7) One limitation of previous, retrospective studies is underascertainment of the primary endpoint.
LIMITATIONS
Given this was a single-center prospective study, the results may lack generalizability. The study protocol did not require standardization of ablation strategies. Though procedure time and fluid administration may differ based on operator and ablation strategy, these parameters were not significantly associated with the primary outcome. Some echocardiographic parameters, such as diastolic dysfunction, were not available prior to the procedure. However, diastolic dysfunction aims to estimate left atrial filling pressures, which was measured directly in our study. Finally, the telephone survey ascertained over half of HF episodes, whose clinical relevance is unclear, though in all positive cases diuretic adjustment had been made.
CONCLUSION
The current study reports early incident HF may complicate up to 20% of AF ablation procedures. The majority of adverse events was associated with recurrent AF post-ablation. However, in those without a recurrence of AF post-ablation, elevated RAP and LAP post-ablation was associated incident HF events. Future, prospective randomized trials are needed to elucidate if reducing elevated atrial pressure post-ablation can reduce HF events.