ASSESSING AND MITIGATING RISK OF QT INTERVAL PROLONGATION

Abstract

Background: Torsades de pointes (TdP) is associated with the electrocardiogram abnormality known as QT interval prolongation and is associated with sudden cardiac death, which is a catastrophic outcome. TdP is most commonly caused by drugs; more than 200 available medications may cause QT prolongation and TdP. Risk factors are important for the development of QT prolongation and TdP; drug-induced TdP is extremely rare in the absence of risk factors. Significant evidence exists linking older age and female sex to QT interval lengthening and identifying both as independent risk factors for QT interval prolongation and TdP. Mechanisms of lengthening of ventricular repolarization with advancing age are likely, in large part, related to changes in serum sex hormone concentrations. The difference in ventricular repolarization between women and men, and the increased susceptibility to drug-induced lengthening of ventricular repolarization and risk of TdP, is due in large part to the effects of sex hormones. Older age is an independent risk factor for QT interval prolongation and TdP. However, independent risk factors for QT interval prolongation within this high-risk population have not been identified. In a retrospective case-control study using multiple large electronic medical records databases, our first Aim was to identify independent risk factors for QT interval prolongation in older adults. Furthermore, female sex is an independent risk factor for QT interval prolongation and TdP. We have shown previously that administration of oral progesterone attenuates drug-induced QT interval lengthening in healthy premenopausal women during the menses phase of the menstrual cycle, when serum estradiol concentrations are low. However, estradiol lengthens the QT interval and could offset the protective effects of progesterone. For the second Aim, we conducted a randomized, double-blind, placebo-controlled, crossover design study to determine the efficacy of oral progesterone to diminish drug-induced QTc interval lengthening in premenopausal women during the ovulation phase of the menstrual cycle, when serum estradiol concentrations are high; then, for the third Aim, we determined the influence of oral progesterone on drug-induced increases in dispersion of repolarization (represented by Tpeak-Tend) and lengthening of early [represented by heart rate-corrected J-Tpeak (J-Tpeakc)] versus late ventricular repolarization (Tpeak-Tend). Methods: To achieve Aim one, we conducted a retrospective case-control study by obtaining a limited dataset using electronic health records from the Regenstrief Institute Data Core from January 2004 to December 2024. After applying our inclusion and exclusion criteria, 230,670 patients aged 65-105 years were eligible for the final analysis. Then we applied the least absolute shrinkage and selection operator regression (LASSO) model for independent risk factors selection, multiple logistic regression (MLR) model to estimate odds ratios and 95% confidence intervals for the association between the predictors and the outcome, and prediction error assessment such as, accuracy and area under the receiver operating characteristic curve [AUC-ROC (c-statistic)] to evaluate the regression model performance. We defined QT interval prolongation as Bazett’s heart rate-corrected QT (QTcB) interval >500 ms, or >550 ms for patients with a QRS >120 ms. To achieve Aims two and three, eighteen premenopausal female volunteers, aged 21-40 years, were randomized in a double-blind, placebo-controlled crossover-design fashion to receive oral progesterone 400 mg or a matching placebo once daily for 7 days, timed to the ovulation phase of the menstrual cycle (median between-phase washout = 41 days). On day 8, intravenous ibutilide 0.003 mg/kg was infused over 10 minutes. ECG intervals were recorded during and for 8 hours post-infusion. Results: With regard to Aim one, the following independent risk factors for QT interval prolongation were identified for older adults: male sex; age ≥80 years; taking one QT interval-prolonging drug; taking two or more QT interval-prolonging drugs; pharmacokinetic drug-drug interactions (DDI) perpetrated by inhibitors of CYP450 enzymes metabolizing QT interval-prolonging drugs; hypocalcemia; hypokalemia; hypomagnesemia; loop diuretics; acute decompensated heart failure (ADHF); sepsis and septic shock, any-stroke including ischemic stroke and hemorrhagic stroke; acute kidney injury (AKI); acute coronary syndrome (ACS); mood disorder including depression and bipolar disorder; hyperparathyroidism; epilepsy; heart failure with preserved ejection fraction (HFpEF); heart failure with reduced ejection fraction (HFrEF); anorexia; hypothyroidism; diabetes mellitus (DM); hypoxia; cancer; hypertension; arrhythmia; chronic kidney disease (CKD); amyloidosis; rheumatoid arthritis; sleep apnea; systemic lupus erythematosus; liver disease; obesity; schizophrenia; chronic ischemic heart disease (CIHD); bradycardia; tobacco use; and drug abuse. With respect to Aim 2, baseline (pre-ibutilide) Fridericia-corrected QT (QTF) in the progesterone and placebo phases was not significantly different. The maximum ibutilide-associated QTF and area under the effect curve during and for 1-hour post-ibutilide infusion (AUEC0-1.17) was significantly lower during the progesterone phase. For Aim 3, baseline (pre-ibutilide) J-Tpeakc were similar in the progesterone and placebo phases, but maximum J-Tpeakc, maximum % change from pretreatment value, and J-Tpeakc AUEC0-1.17 were significantly lower during progesterone treatment compared to placebo, as were pre-ibutilide Tpeak-Tend, maximum Tpeak-Tend, and Tpeak-Tend AUEC0-1.17. Conclusions: We identified multiple independent risk factors for QT interval prolongation and TdP in older adults. These risk factors can help clinicians identify older adults who are at higher risk for QT interval prolongation, which may guide treatment and monitoring decisions. Future research will include incorporating these independent risk factors into a risk score that can be developed and validated for this high-risk population, who can be targeted for amelioration of these modifiable risk factors and therapeutic substitution, where feasible and appropriate. This will improve medication safety in this vulnerable patient population. In addition, we have shown that oral progesterone attenuates drug-induced lengthening of ventricular repolarization in premenopausal women, regardless of menstrual cycle phase, and in the presence of high serum estradiol concentrations. This research suggests that pre-menopausal women who require therapy with QT interval-prolonging drugs and who have risk factors for drug-induced QT interval prolongation and torsades de pointes could be prescribed oral progesterone to lower their risk, regardless of the phase of the menstrual cycle. This hypothesis requires testing in future studies.