Clinically relevant QTc prolongation due to overridden drug–drug interaction alerts: a retrospective cohort study

摘要

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT ? A large number of drugs can prolong the QTc interval on the electrocardiogram (ECG). ? Clinical decision support systems may generate drug safety alerts on QTc prolongation. ? Drug safety alerts are frequently overridden. WHAT THIS STUDY ADDS ? QTc alert overriding does rarely result in ECG recordings. ? ECGs before and after QTc overrides reveal clinically relevant QTc prolongation. AIMS To investigate whether, in patients in whom drug–drug interaction (DDI) alerts on QTc prolongation were overridden, the physician had requested an electrocardiogram (ECG), and if these ECGs showed clinically relevant QTc prolongation. METHODS For all patients with overridden DDI alerts on QTc prolongation during 6 months, data on risk factors for QT prolongation, drug class and ECGs were collected from the medical record. Patients with ventricular pacemakers, patients treated on an outpatient basis, and patients using the low-risk combination of cotrimoxazole and tacrolimus were excluded. The magnitude of the effect on the QTc interval was calculated if ECGs before and after overriding were available. Changes of the QTc interval in these cases were compared with those of a control group using one QTc-prolonging drug. RESULTS In 33% of all patients with overridden QTc alerts an ECG was recorded within 1 month. ECGs were more often recorded in patients with more risk factors for QTc prolongation and with more QTc overrides. ECGs before and after the QTc override were available in 29% of patients. Thirty-one percent of patients in this group showed clinically relevant QTc prolongation with increased risk of torsades de pointes or ventricular arrhythmias. The average change in QTc interval was +31 ms for cases and ?4 ms for controls. CONCLUSIONS Overriding the high-level DDI alerts on QTc prolongation rarely resulted in the preferred approach to subsequently record an ECG. If ECGs were recorded before and after QTc overrides, clinically relevant QTc prolongation was found in one-third of cases. ECG recording after overriding QTc alerts should be encouraged to prevent adverse events.