Atrial fibrillation as a prognostic factor for all-cause mortality in patients with transthyretin amyloid cardiomyopathy

Atrial fibrillation (afib) and flutter (aflutter) are among the most common manifestations of transthyretin amyloid cardiomyopathy (ATTR-CM) [1–3]. Studies have suggested that afib does not influence mortality [3–6]. Tafamidis was approved for the treatment of ATTR-CM based on the landmark, placebo-controlled, Tafamidis in Transthyretin Cardiomyopathy Clinical Trial (ATTR-ACT), where all-cause mortality was a primary efficacy outcome [7].

Evaluate whether afib/aflutter was prognostic for all-cause mortality in ATTR-ACT.

ATTR-ACT was a multinational, Phase 3, randomized study of patients with ATTR-CM receiving tafamidis meglumine (80 or 20 mg) or placebo over 30 months. This analysis describes the characteristics of patients with and without current/historical afib/aflutter at the baseline of ATTR-ACT. Current/historical afib/aflutter was then evaluated as an independent prognostic factor for all-cause mortality using Cox proportional hazards modelling.

The 314/441 (71%) patients in ATTR-ACT with current/historical afib/aflutter were older, more commonly male, and White than those without (Table 1). A higher proportion of patients with current/historical afib/aflutter were NYHA class III and had a wild-type genotype than those without. Median 6MWT distance and the proportion of patients with preserved left ventricular ejection fraction (≥50%) were lower, GLS was less negative, and NT-proBNP and BUN were higher among those with current/historical afib/aflutter vs those without.

In ATTR-ACT, treatment, genotype, and NYHA class were included in the pre-specified Cox proportional hazards model assessing all-cause mortality. Once current/historical afib/aflutter was added to this model, all variables were found to be significant independent predictors of all-cause mortality (P<0.05; Table 2). The risk of mortality was 45% higher (hazard ratio = 0.55 [95% CI: 0.37, 0.82]) among patients with current/historical afib/aflutter than in those without. In an expanded stepwise model selection analysis including 23 baseline demographic and clinical covariates, current/historical afib/aflutter was not a significant independent predictor. This model showed BUN and NT-proBNP concentrations, 6MWT distance, genotype, treatment, and GLS to be significant independent prognostic factors for all-cause mortality (P<0.01; Table 2).

Patients in ATTR-ACT with ATTR-CM and current/historical afib/aflutter were older, more likely to be male, have a wild-type genotype, and more advanced heart failure than those without. A Cox model including treatment, genotype, and NYHA class identified current/historical afib/aflutter as a significant independent predictor of all-cause mortality. In an expanded stepwise selection model current/historical afib/aflutter remained important but was less prognostic compared to other covariates. Overall, these results demonstrate the importance of afib/aflutter in patients with ATTR-CM.

Type of funding sources: Private company. Main funding source(s): This study was sponsored by Pfizer.