Predictive Tools for the Presence of Significant Coronary Artery Disease Requiring Intervention in Chronic Hemodialysis Patients.

Abstract

Purpose: Hemodialysis (HD) patients have increased cardiovascular risk. Coronary artery disease (CAD) may not typically manifest in these patients, and the investigational results can be confounded by the fluctuation of body fluid and electrolytes. This study explored the accuracy of these findings in identifying CAD in HD patients.

Methods: Medical record data of all HD patients undergoing coronary angiography (CAG) at a tertiary hospital between January 2017 and March 2021 were retrospectively reviewed as a discovery cohort to identify factors discriminating patients with and without significant CAG-established CAD. HD patients receiving CAG between April 2021 and March 2023 were assigned as a validation cohort.

Results: The discovery and validation cohorts included 108 and 41 patients, respectively. Chest pain out of the dialysis session, but not pulmonary congestion/edema nor other intradialytic events, was associated with significant CAD (odds ratio 3.281, 95% CI 1.313-8.200). The presence of regional wall motion abnormality (RWMA) from echocardiogram, but not other echocardiographic and electrocardiographic findings, significantly predicted significant CAD (odds ratio 2.579, 95% CI 1.052-6.324). Initial high-sensitivity troponin T > 111 ng/L and percent of troponin change > 22% had area under the ROC curve of 0.71 (95% CI 0.52-0.89) and 0.67 (95% CI 0.50-0.85), respectively, for predicting CAD. Applying a combination of these predictors resulted in a specificity of 100%, but low sensitivity for CAD in the validation cohort.

Conclusion: Most intradialytic symptoms possibly result from noncardiac etiologies, while chest pain out of the HD session should alarm the physicians for further management. Applying the suggested cardiac enzyme cut points or the presence of RWMA could prioritize at-risk patients for timed cardiologist referral.