Validation of continuous QTc measurement in critically ill patients

Abstract

Introduction: Prolongation of the corrected QT interval (QTc) can lead to torsades de pointes. This study is designed to determine the validity of the continuous QTc (cQTc) measurement in critically ill patients.

Methods: In a retrospective cohort study, QTc analysis was performed with manual measurements on a single selected lead from a 12-lead ECG and cQTc measurement obtained at the same time. In addition, automated QTc measurement from the 12-lead ECG was also included in the study. Validation was performed by calculating intraclass correlation coefficient (ICC), Pearson's correlation and Bland-Altman plot.

Results: 119 patients with QRS<120 ms were included with a mean cQTc of 468 ms (standard deviation (SD) 37) and mean manually measured QTc of 449 ms (SD 41) (p<0.001). Pearson's correlation was 0.65 (p<0.01), ICC was 0.65 (95% CI: 0.53-0.74). Bland-Altman plot shows a mean difference of 19.5 ms (limits of agreement (LOA) -44.6 to 83.7). For cQTc compared to automated QTc from the 12-lead ECG the intraclass correlation coefficient was 0.77 (95% CI: 0.68-0.83, p<0.001) and the Bland-Altman plot shows a mean difference of 7.8 ms (LOA -40.2 to 55.8).

Conclusion: cQTc measurement in critically ill patients with a QRS duration shorter than 120 ms shows an acceptable accuracy to be used in routine care.

Keywords: Continuous measurement; Critical care; Intensive care; QTc; Torsades de pointes; Validation.