The following is a summary of “Automated Quantification of QT-Intervals by an Algorithm: A Validation Study in Patients with Chronic Obstructive Pulmonary Disease,” published in the March 2024 issue of Pulmonology by Kohlbrenner et al.
Researchers conducted a retrospective study comparing the diagnostic accuracy of a purpose-designed QTc-scoring algorithm to established hand-scoring in patients with chronic obstructive pulmonary disease (COPD) undergoing sleep studies.
They obtained 62 overnight ECG recordings from 28 patients with COPD. QT-intervals corrected for heart rate (QTc, using Bazett’s formula) were averaged over 1-minute intervals and assessed by both the algorithm and manual cursor-assisted scoring. Manual scoring was performed blindly for the algorithm-derived results. Bland-Altman statistics were calculated, along with confusion matrices for three thresholds (460ms, 480ms, and 500ms).
The results showed 32,944 1-minute periods, and corresponding mean QTc-intervals were assessed manually and via computer. The mean discrepancy between manual and algorithm-based QTc-intervals was -1 ms, with agreement limits from -18 to 16 ms, 8% (2587), 1% (357), and 0 QTc-intervals exceeded the 460, 480, and 500ms thresholds, respectively, as identified through manual scoring. The algorithm consistently identified 2516, 357, and 0. Diagnostic classification accuracy was 0.98 (95% CI 0.98/0.98), 1.00 (1.00/1.00), and 1.00 (1.00/1.00) for 460, 480, and 500ms, respectively. Sensitivity was 0.97, 1.00, and NA for 460, 480, and 500ms, respectively, while specificity was 0.98, 1.00, and 1.00 for the respective thresholds.
Investigators concluded that the automated QTc algorithm effectively detected prolonged QTc in stable COPD patients, potentially enabling sleep labs to identify asymptomatic individuals at risk for arrhythmias
