All-cause mortality following atrial fibrillation (AF) has decreased over time. Data regarding temporal trends in causes of death among individuals with AF are scarce. The aim of our study was to analyze temporal trends in cause-specific mortality and predictors for cardiovascular (CVD) and non-CVD deaths among participants with incident AF in the Framingham Heart Study. We categorized all newly diagnosed AF cases according to age at AF diagnosis (< 70, 70 to < 80, and ≥ 80 years) and epoch of AF diagnosis (< 1990, 1990–2002, and ≥ 2003). We followed participants until death or the last follow-up. We categorized death causes into CVD, non-CVD, and unknown causes. For each age group, we tested for trends in the cumulative incidence of cause-specific death across epochs. We fit multivariable Fine-Gray models to assess subdistribution hazard ratios (HR) between clinical risk factors at AF diagnosis and cause-specific mortality. We included 2125 newly diagnosed AF cases (mean age 75.5 years, 47.8% women). During a median follow-up of 4.8 years, 1657 individuals with AF died. There was evidence of decreasing CVD mortality among AF cases diagnosed < 70 years and 70 to < 80 years (ptrend < 0.001) but not ≥ 80 years (p = 0.76). Among the cases diagnosed < 70 years, the cumulative incidence of CVD death at 75 years was 67.7% in epoch 1 and 13.9% in epoch 3; among those 70 to < 80 years, the incidence at 85 years was 58.9% in epoch 1 and 18.9% in epoch 3. Advancing age (HR per 1 SD increase 6.33, 95% CI 5.44 to 7.37), prior heart failure (HR 1.49, 95% CI 1.14–1.94), and prior myocardial infarction (HR 1.44, 95% CI 1.15–1.80) were associated with increased rate of CVD death. In this community-based cohort, CVD mortality among AF cases decreased over time. Most deaths in individuals with AF are no longer CVD-related, regardless of age at AF diagnosis.

中文翻译：

---

弗雷明汉心脏研究中新诊断的房颤患者病因特异性死亡率的时间趋势

随着时间的推移，房颤 (AF) 后的全因死亡率已经下降。关于 AF 患者死因时间趋势的数据很少。我们研究的目的是分析弗雷明汉心脏研究中发生 AF 的参与者的特定原因死亡率以及心血管 (CVD) 和非 CVD 死亡的预测因素的时间趋势。我们根据 AF 诊断年龄（< 70、70 至 < 80 和 ≥ 80 岁）和 AF 诊断时期（< 1990、1990-2002 和 ≥ 2003）对所有新诊断的 AF 病例进行分类。我们跟踪参与者直到死亡或最后一次随访。我们将死亡原因分为 CVD、非 CVD 和未知原因。对于每个年龄组，我们测试了不同时期特定原因死亡累积发生率的趋势。我们拟合多变量 Fine-Gray 模型来评估 AF 诊断的临床危险因素与特定原因死亡率之间的亚分布风险比 (HR)。我们纳入了 2125 例新诊断的 AF 病例（平均年龄 75.5 岁，47.8% 为女性）。在 4.8 年的中位随访期间，1657 名 AF 患者死亡。有证据表明，诊断为 < 70 岁和 70 至 < 80 岁（ptrend < 0.001）但 ≥ 80 岁的 AF 病例的 CVD 死亡率降低（p = 0.76）。在确诊<70岁的病例中，75岁的CVD死亡累积发生率在第1时期为67.7%，在第3时期为13.9%；在 70 至 < 80 岁的人群中，85 岁的发病率在第 1 时期为 58.9%，在第 3 时期为 18.9%。 年龄增长（HR 每 1 SD 增加 6.33，95% CI 5.44 至 7.37），既往心力衰竭（HR 1.49） , 95% CI 1.14–1.94) 和既往心肌梗死 (HR 1.44, 95% CI 1.15–1.80) 与 CVD 死亡率增加有关。在这个基于社区的队列中，AF 病例的 CVD 死亡率随着时间的推移而下降。大多数 AF 患者的死亡不再与 CVD 相关，无论 AF 诊断时的年龄如何。