Supplementary Material for: Specific Electrocardiograph Intervals Predict Hospitalization with Atrial Fibrillation in Those with Chronic Kidney Disease

Introduction: Atrial fibrillation (AF) is common in patients with chronic kidney disease (CKD) and is associated with higher rates of hospitalization compared to those without AF. Whether routine electrocardiographic parameters are predictive of future hospitalizations with AF is not clear. Methods: The present study is an analysis of a prospective cohort of 2,759 patients without baseline AF from the Chronic Renal Insufficiency Cohort, a large prospective multicenter study of patients with nondialysis-dependent CKD. Unadjusted and adjusted Cox regression models were fit to examine the association of baseline categories of QTc, QRS, and PR intervals with time to first hospitalization with AF. Restricted cubic splines were used to display nonlinear associ­ations. Results: The mean age of subjects at baseline was 58 ± 11 years, 55% were male, and 44% were Black. The mean follow-up was 6.6 years during which 224 participants experienced a hospitalization with AF. The association of baseline QTc interval with risk of AF hospitalization was nonlinear, such that the lowest and highest quartiles of QTc (<407 and >431 ms, respectively) had higher adjusted risk of AF hospitalization, compared with the second quartile (407–416 ms) (aHR Q1:Q2 1.58, 95% CI 1.03–2.41; p = 0.03; aHR Q4:Q2 1.84, 95% CI 1.22–2.78; p < 0.01). Longer QRS was associated with a higher risk of hospitalization with AF among the subgroup of patients with a history of heart failure (HF). PR interval was not associated with AF hospitalization. Discussion/Conclusion: The association of QTc with risk for hospitalization with AF among patients with CKD is nonlinear, while the association of longer QRS with AF hospitalization is restricted to patients with baseline HF. Electrocardiography may represent a simple and widely accessible method for risk stratification of future AF in patients with CKD.

引言：心房颤动（Atrial fibrillation, AF）在慢性肾脏病（chronic kidney disease, CKD）患者中较为常见，且相较于无房颤的患者，此类患者的住院率更高。目前尚不明确常规心电图参数是否可预测未来因房颤导致的住院风险。

方法：慢性肾功能不全队列（Chronic Renal Insufficiency Cohort）是一项针对非透析依赖性慢性肾脏病患者的大型多中心前瞻性研究，本研究从中提取了2759名基线无房颤的受试者的前瞻性队列数据展开分析。本研究构建未校正与校正后的Cox回归模型，以探究基线QTc间期（QTc interval）、QRS波群时限（QRS interval）及PR间期（PR interval）的分类指标与首次因房颤住院的时间之间的关联；并采用限制性立方样条可视化非线性关联关系。

结果：受试者基线时的平均年龄为58±11岁，其中男性占比55%，黑人占比44%；本次研究的平均随访时长为6.6年，期间共有224名受试者因房颤住院治疗。基线QTc间期与房颤住院风险的关联呈非线性特征：与第二四分位数组（407~416ms）相比，QTc间期最低与最高的四分位数组（分别为<407ms和>431ms）的校正后风险比（adjusted hazard ratio, aHR）更高，具体为aHR Q1:Q2=1.58，95%置信区间（95% confidence interval, 95% CI）1.03~2.41，P=0.03；aHR Q4:Q2=1.84，95%CI 1.22~2.78，P<0.01。在有心力衰竭（heart failure, HF）病史的亚组患者中，更长的QRS波群时限与更高的房颤住院风险相关；而PR间期与房颤住院风险无显著关联。

讨论与结论：慢性肾脏病患者的QTc间期与房颤住院风险之间的关联呈非线性特征，而更长的QRS波群时限与房颤住院风险的关联仅局限于基线合并心力衰竭的患者。心电图检查可作为一种简便且普及性强的方法，用于慢性肾脏病患者未来发生房颤的风险分层。