Automated QT analysis on Holter monitors in pediatric patients can differentiate long QT syndrome from controls

Background Borderline QTc is a common referral to the pediatric cardiology clinic. Evaluation is challenging due to significant overlap of normal and abnormal QTc ranges. We hypothesized that automated QT analysis on Holter could differentiate between patients with long QT syndrome (LQTS) and healthy controls. Methods We conducted a retrospective review of 39 patients with known genotype‐positive, phenotype‐positive LQTS who underwent Holter monitoring between January 2010 and January 2016. They were compared 2:1 to age‐ and sex‐matched controls. Automated QT analysis data were analyzed. Results Significant differences were found in all automated QT and QTc fields, except minimum QTc interval (P = 0.57). Mean QTc interval (LQTS 479 ± 28 ms vs controls 429 ± 16 ms; P ≤ 0.001) and percent QTc intervals (%QTc) >450 ms (LQTS 80 ± 28% vs controls 14 ± 16%; P ≤ 0.001) were selected for further analysis. A receiver operating characteristic curve was generated for each variable demonstrating high area under the curve values of 0.9494 and 0.9540, respectively. Threshold values of ≥461 ms for mean QTc (sensitivity 79.49%, specificity 98.72%) and ≥65% of %QTc >450 ms (sensitivity 79.49%, specificity 98.72%) allowed highly specific discrimination between cohorts (false positive rate 1.09%). Similarly, thresholds of