Table_6_Hypertension and Atrial Fibrillation: A Study on Epidemiology and Mendelian Randomization Causality.docx

Introduction: Hypertension (HT) and atrial fibrillation (AF) often coexist. However, the causality between these two conditions remains to be determined.Methods: We used individual participant data from the Atherosclerosis Risk in Communities (ARIC) prospective cohort with 9,474 participants. HT was ascertained at visit 1 (1987–1989), and incident AF was identified by ECGs conducted during study examinations at each visit, hospital discharge codes, and death certificates. We used the Kaplan–Meier estimate to compute the cumulative incidence of AF by the HT subgroup. Then we used Cox hazard regression model to assess the association between HT and incident AF. The causality between genetically determined HT and AF was analyzed by the two-sample Mendelian randomization (MR) based on publicly summarized genome-wide association studies (GWASs) data.Results: A total of 1,414 cases (14.9%) of AF were identified during the follow-up period (median 24.1 years). After adjusting for all covariates, the hazard ratio between the participants with HT and incident AF was 1.50 [95% confidence interval (CI) 1.29–1.73]. In the HT → AF MR analysis, we detected a causal correlation between HT and AF (OR: 1.90, 95% CI 1.18–3.04, P = 0.01) with no evidence of heterogeneity from single-nucleotide polymorphisms. Besides, the genetically determined SBP and DBP (10 mmHg) were consistently associated with a higher risk of AF.Conclusions: In the ARIC study, the incident AF increased by 50% in patients with HT. In the MR analysis, our results supported causal inference between HT and AF.

引言：高血压（HT）与心房颤动（AF）常伴发存在。然而，这两种疾病之间的因果关系尚未得到明确。研究方法：本研究利用了来自动脉粥样硬化风险在社区（ARIC）前瞻性队列的个体参与者数据，共计9,474名参与者。高血压在第一次访问（1987-1989年）时被确定，而新发心房颤动通过每次访问的研究期间进行的心电图、医院出院编码和死亡证明来识别。我们使用Kaplan-Meier估计来计算高血压亚组的AF累积发生率。随后，我们采用Cox风险回归模型评估高血压与新发心房颤动之间的关联。通过基于公开发表的基因组关联研究（GWAS）数据的双样本孟德尔随机化（MR）分析，对由遗传因素决定的HT与AF之间的因果关系进行了分析。结果：在随访期间（中位数为24.1年）共识别出1,414例（14.9%）的心房颤动病例。在调整所有协变量后，具有HT和发生心房颤动的参与者的风险比（HR）为1.50 [95%置信区间（CI）1.29–1.73]。在HT→AF的MR分析中，我们检测到HT与AF之间的因果相关性（OR：1.90，95% CI 1.18–3.04，P = 0.01），并且没有来自单核苷酸多态性的异质性证据。此外，由遗传因素决定的收缩压（SBP）和舒张压（DBP）（10毫米汞柱）与AF风险增加呈一致性相关。结论：在ARIC研究中，高血压患者的心房颤动发生率增加了50%。在MR分析中，我们的结果支持了HT与AF之间的因果推断。