Short-term effects of anti-thymocyte globulin on cardiac repolarization after hematopoietic stem cell transplantation: a linear mixed-effects analysis

Background Anti-thymocyte globulin (ATG) is widely used in conditioning regimens for hematopoietic stem cell transplantation (HSCT), yet its acute impact on ventricular repolarization and the drivers of QTc prolongation are not well defined. Aim We aimed to evaluate short-term alterations in ECG-derived ventricular repolarization parameters following ATG administration and to identify independent predictors of QTc prolongation. Methods We included consecutive HSCT patients treated with ATG who underwent serial ECGs obtained at baseline and on days 1–4. Assessed parameters included heart rate (HR), QT, Fridericia-corrected QT (QTc), QT dispersion, T-peak–to-end (Tpe), and Tpe/QTc. Confounder-adjusted linear mixed-effects models with patient-level random intercepts were used to assess time-dependent changes, adjusting for baseline QTc, demographic and clinical variables, and time-varying potassium, magnesium, and creatinine. QTc prolongation was further evaluated using multivariable linear regression for continuous peak ΔQTc and multivariable logistic regression for dichotomized ΔQTc ≥30 ms. Results A total of 54 patients receiving ATG were analyzed. The cohort had a mean age of 41.1±13.2 years, and 68.5% were male. Fourteen patients (25.9%) received ATG for 2 days, while the remaining patients received 3 days of ATG. In confounder-adjusted linear mixed-effects models including baseline QTc and time-varying potassium, magnesium, and creatinine, heart rate decreased significantly after ATG, with the greatest reduction on Day 3 (β −11.1 bpm; Holm-adjusted p=9×10−⁶). QT and QTc intervals increased significantly, with peak QT prolongation on Day 3 (β +29.2 ms; p<0.001) and peak QTc prolongation on Day 4 (β +14.7 ms; p=1.42×10−⁴). Repolarization markers Tpe and Tpe/QTc also increased significantly by Day 4, whereas QT dispersion did not change significantly (Figure 1).Peak QTc prolongation (ΔQTc) averaged 24.7±25.9 ms. In multivariable linear regression, older age (per 10 years: β +4.9 ms; p=0.049) and lower baseline QTc (per 10 ms: β −5.9 ms; p<0.001) were independently associated with greater ΔQTc. When dichotomized, ΔQTc ≥30 ms occurred in 44.4% of patients. In multivariable logistic regression, baseline QTc was inversely associated with ΔQTc ≥30 ms (OR 0.60 per 10 ms; 95% CI 0.42–0.86; p=0.005), while age showed a borderline positive association (OR 2.17 per 10 years; p=0.059). ATG dose, treatment duration, electrolytes, and comorbidities were not independent predictors. Conclusion ATG administration is associated with acute, transient prolongation of QTc and ventricular repolarization markers, independent of electrolyte changes. Older age and baseline QTc are key determinants of QTc prolongation, underscoring the importance of early ECG monitoring during ATG therapy.