Electrocardiographic identification of drug effects: QT prolongation, arrhythmias, and clinical monitoring

BACKGROUND: The electrocardiogram (ECG) remains the foremost non-invasive tool for detecting adverse drug effects on cardiac electrophysiology, encompassing QT interval prolongation, torsades de pointes (TdP), atrioventricular (AV) block, and sodium-channel-mediated QRS widening. As polypharmacy becomes ubiquitous - particularly in ageing populations and oncology - clinicians require a structured understanding of drug-ECG interactions to mitigate sudden cardiac death risk. METHODS: A systematic narrative review was conducted using PubMed, Embase, and the Cochrane Library covering January 2006 to April 2026. Priority was given to meta-analyses, systematic reviews, scientific statements, and guidelines addressing drug-induced ECG changes, QTc measurement methodology, TdP risk stratification, and ECG monitoring protocols. Reference lists of key papers were hand-searched. Seventy references meeting quality and relevance criteria are cited. RESULTS: Numerous drug classes - including antiarrhythmics, antibiotics, antipsychotics, antidepressants, antiemetics, and oncology agents - prolong the corrected QT interval primarily through blockade of the cardiac hERG/IKr potassium channel. Modifiable risk factors include hypokalaemia, hypomagnesaemia, bradycardia, and co-prescription of CYP3A4 inhibitors. Female sex, age ≥ 65 years, congenital long QT syndrome carriers, and cardiac failure patients are at disproportionate TdP risk. The validated Tisdale Risk Score enables pre-treatment stratification in hospitalised patients. Drug-specific ECG monitoring protocols with actionable thresholds are summarised. CONCLUSIONS: Drug-induced ECG changes are common, clinically consequential, and often preventable. Evidence-based monitoring protocols, correction of modifiable risk factors, and prompt recognition of ominous ECG patterns substantially reduce adverse outcomes. Prescriber education across all clinical settings is imperative.