(189) Risk identification based on prescribed drugs and comorbidities in patients with Torsade de pointes arrhythmia – a population-based Swedish cohort study.

Background: Drug-induced long QT interval is a risk factor for the life-threatening torsade de pointes (TdP) ventricular arrhythmia. Clinical decision support systems (CDSS) are designed to help prescribers to assess the risk for TdP associated with prescription drugs.

Objectives: To retrospectively analyse the predicted risk for TdP based on medication use 90 days prior to TdP diagnosis among the Swedish population using the CDSS Janusmed Risk Profile (https://janusmed.se) and Credible Meds (https://crediblemeds.org/), and to analyse the 5-year prevalence of risk diagnoses for TdP prior to TdP diagnosis.

Methods: All patients in Sweden with a diagnosis of TdP (defined as ICD-10 code I472.C) between the years 2006 and 2018 were included in this retrospective study cohort. Data on medications dispensed within 90 days prior to the TdP occurrence were retrieved from the Swedish Prescribed Drug Register. The patient’s individual risk for TdP based on used medications was calculated using the CDSS Credible Meds and Janusmed Risk Profile. Comorbidities known as risk factors for TdP diagnosed within 5 years prior to TdP were identified using the Swedish Patient Register. Sex differences in medication risk levels for TdP and comorbidities with increased risk were analysed using Fishers exact test.

Results: In total, 762 patients with a diagnosis of TdP between 2006 and 2018 were identified, of which 50% were women. Median age was 72 years (IQR 63-81), and a majority (72%) was 65 years or older. Patients were dispensed in median 6 medications (IQR 2-9) within 90 days prior to the date of TdP diagnosis. In total, 30% of patients had a combination of medications that were identified by Janusmed Risk Profile to be associated with an increased risk of TdP (16% slightly increased, 2% moderately and 13% highly increased risk). According to Credible Meds, 192 (25%) of patients used a drug with known risk for TdP, 7% at least one drug with a possible risk for TdP and 27% a drug with a conditional risk for TdP. The top 5-pharmaceuticals associated with an increased risk for QT prolongation/TdP used were citalopram (n=80), sotalol (n=38), mirtazapine (n=32), hydroxyzine (n=23), and amiodarone (n=22). One or more of the risk comorbidities were present in 338 (44%) individuals during five years before TdP diagnosis. Most patients (n=304, 60%) had neither medication nor comorbidities associated with an increased risk for TdP. More males than females had comorbidities associated with increased risk for TdP (28% vs 22%, p<.0015). We found no significant difference in the medication risk among males and females.

Conclusions: Prior to the occurrence of TdP almost a third of patients had been dispensed medications associated with risk for TdP, according to both CDSSs used.