A clinical decision support system for the reduction of the risk of drug-induced QT interval prolongation in hospitalized patients

Drug-induced torsades de pointes (TdP) is a potentially life-threatening adverse drug reaction (ADR). TdP is a polymorphic ventricular tachycardia which occurs in the setting of prolongation of the QT interval on the electrocardiogram (ECG). Although drug-induced TdP is a rare event, it may have lethal consequences since it can result in sudden cardiac death. Many medications commonly used in hospitals, including drugs prescribed for non-cardiac indications, can cause QT interval prolongation and trigger TdP. Previous research has identified certain risk factors which predispose patients to drug-induced QT interval prolongation and TdP. We hypothesize that the risk for drug-induced QT interval prolongation and TdP could be markedly reduced by identifying patients at highest risk for developing this ADE and alerting healthcare professionals regarding this risk. This study describes the evaluation of a computerized clinical decision support system (CDSS) designed to quantify the risk of QT interval prolongation in patients for whom QT interval-prolonging drugs are prescribed, and alert healthcare professionals about susceptible patients, so that risk-reducing measures could be implemented. The system was developed locally and was integrated into the existing pharmacy computer system at two cardiac critical care units at IU Methodist Hospital. A prospective pre-intervention and post-intervention study design was used to assess the impact of the system on several patient and prescribing outcomes relevant to QT interval prolongation. Data collected for each patient included demographics, current medical diagnoses, past medical history, current laboratory data, current medications and daily QTc intervals. Multiple logistic regression was used to assess the impact of the CDSS on study outcomes. A total of 1,240 patient admissions were analyzed. Of these, 900 admissions occurred before the intervention (pre-intervention cohort) and 340 admissions occurred after the intervention (post-intervention cohort). There was no significant difference in the proportion of patients experiencing QTc interval prolongation (Adjusted OR= 1.05 (0.79-1.41), p=0.72). There was no significant difference in the proportion of patients receiving QT interval-prolonging drugs (Adjusted OR=0.88 (0.66-1.17), p=0.37). There was no significant difference in the proportion of patients with two or more risk factors receiving QT interval-prolonging drugs (Adjusted OR=0.88 (0.65-1.20), p=0.42). The mean maximum QTc was 476 ± 39 in the pre-intervention group, and was 475 ± 44 in the post-intervention group (p=0.76). In conclusion, implementation of this computerized CDSS did not have a significant impact on physicians\u27 prescribing behavior of QT interval- prolonging drugs or on patient outcomes related to QT interval prolongation