Association between Thoracic Radiation and Heart Rhythm Disorders: Toward a Model for Describing Long-Term Cardiac Risk from Radiotherapy

Abstract

<h3>Purpose/Objective(s)</h3> Radiation-induced heart disease (RIHD) is a major source of morbidity and mortality in patients receiving thoracic radiation. The impact of thoracic radiation on heart rhythm changes remains poorly understood. To further define this association, we retrospectively collected electrocardiogram (ECG) data from patients who received thoracic radiation at our institution and analyzed heart rhythm changes as a function of treatment time and delivery technique. <h3>Materials/Methods</h3> We retrospectively reviewed all patients who received thoracic radiation from 2005-2020 at our institution and identified those who also had digitized ECGs. Abnormal ECGs were defined as ventricular rates (VR) <60 or >100, or incalculable PR intervals as surrogate for composite arrhythmia (CA) of atrial fibrillation/flutter and 2^nd/3^rd degree AV block. We compared atrial rates (AR) and VR via unpaired Wilcoxon rank sum test, paired Wilcoxon test for nearest pre- and post-RT ECGs, and calculated odds ratio (OR) with χ² test. All statistics were performed using statistical software. <h3>Results</h3> We identified 1034 patients with 5332 ECGs, and included 360 patients who had ECGs before and after radiation therapy (RT) with a total of 3328 ECGs available for review. 49.4% (178/360) vs 63.4% (229/360) of patients had an abnormal ECG before RT vs after RT, OR 1.79 (95% CI 1.33 – 2.41, χ² = 14, p < 0.001). Median and interquartile range (IQR) ECGs were obtained 305 days (875) before and 290 days (714) after radiation start date. In all ECGs, pre- vs post-RT AR [77 (27) bpm vs 89 (34) bpm] and VR [77 (24) bpm vs 87 (31) bpm] were statistically greater, (p < 0.001). There were no differences in odds of prolonged PR interval >200 ms or QRS length >120 ms after RT. We analyzed the closest pre- and post-RT ECGs for each patient (720 ECGs total); ECG were obtained 54 (201) days before and 101 (360) days after start of radiation. On paired analysis, AR [76 (24) vs 84 (31) bpm] and VR [76 (23) vs 83 (29) bpm], remained greater, p < 0.001. Of the 182 patients with exclusively normal ECGs before RT, 125 (68.7%) had an abnormal ECG after RT, persistent over multiple ECGs in 70 (56%) patients. 43 patients had new CA after RT (55/360 vs 98/360, OR 2.07, 95% CI 1.44-3.01, χ² = 15, p < 0.001), persistent over multiple ECGs in 27 (63%) patients. 183 patients received SBRT and had no significantly increased odds for CA, however the 177 patients who received conventionally fractionated RT (CFRT) had higher odds for CA after CFRT, OR 2.45 (1.45 – 4.24, χ² = 15, p < 0.001). <h3>Conclusion</h3> Our population had increased AR/VR and increased odds of abnormal ECG and CAs after thoracic radiation. Odds of CAs were highest in patients receiving CFRT. These abnormalities appeared to persist in some patients long after RT. These data strongly suggest that further prospective studies are needed to further define associated risks and to develop mitigation strategies for arrhythmia and conduction abnormalities.