Using Mendelian randomization to unveil the truth: does COVID-19 impact QT interval prolongation?

Abstract

Background QT interval, a vital marker of ventricular electrical activity during depolarization and repolarization, garnered significant attention during the COVID-19 pandemic. However, it remains unclear whether COVID-19 directly affects QT interval prolongation. This study leverages Mendelian randomization (MR) to investigate the genetic causation between COVID-19 and QT interval alterations. Methods In over 1000,000 individuals of European ancestry, genetic proxies representing three COVID-19 phenotypes (COVID-19, hospitalized COVID-19, and severe COVID-19), were identified under the primary MR assumption, and serve as instrumental variables (IVs). Genetic causal effects of COVID-19 on QT intervals from 84,630 UK Biobank participants were inferred using univariate two-sample MR (TSMR) and multi-exposure-adjusted multivariate MR (MVMR). MR-RAPS method and radial MR frame were used to provide robustness and outlier variant detection for effect assessment, and sensitivity analysis was further applied to detect the presence of horizontal pleiotropy. Results Independent 15, 33, and 29 IVs were used in COVID-19, hospitalized COVID-19, and severe COVID-19, respectively. Univariate TSMR analyses showed non-significant causal effect estimates between COVID-19 and the QT interval across all COVID-19 phenotypes. MR-RAPS and outlier-corrected radial MR analyses further supported this null causal estimation. In confounder-adjusted MVMR analysis, this nonsignificant causality was independent of BMI, smoking, and alcohol consumption. Sensitivity analyses failed to reveal any evidence of bias arising from horizontal pleiotropy, abnormal data distribution, or weak instruments. Conclusions Genetically, COVID-19 is not causally associated with QT interval prolongation. Inconsistent findings in observational research may be attributed to residual confounding.