In the setting of contemporary trimodality care for esophageal cancer, there is ongoing debate on optimal radiotherapy (RT) planning. This multi-institutional review explores associations between RT parameters and cardiopulmonary toxicities in patients treated for locally advanced esophageal cancer. Materials/Methods: We reviewed the clinical and dosimetric parameters of 465 consecutive esophageal cancer patients treated with chemoradiotherapy (CRT) followed by surgery at two institutions between 2007 and 2013. Adverse events (AE) were collected in the time frame during and immediately after CRT as well as perioperatively. These included myocardial infarction, heart failure, arrhythmia, pericardial effusion, pneumonia, pneumonitis, pulmonary fibrosis, and respiratory failure. Cardiac dose-volume histograms (DVH) and lung DVH were compiled for each patient. Associations between cardiac and pulmonary doses with AE were assessed using logistic regression, with odds ratios representing the increased risk of toxicity for a 10% increase in exposed volume. Results: The median age was 61 years (IQR 54-68): 398 patients (85.6%) were male, and 63.7% were current/former smokers. At baseline, 59.8% of patients had known cardiac risk factors, including hypertension, diabetes, coronary artery disease, and atrial fibrillation. Only 9.7% of patients had pulmonary comorbidities, including asthma and chronic obstructive pulmonary disease. Most patients had adenocarcinoma (93.3%), and 95.9% had clinical stage II-III disease. Concurrent CT consisted of platinum doublet therapy in the majority of patients. Ivor-Lewis esophagectomy was the most common surgery type (83.4%). The RT modalities were IMRT (41.3%), 3D (38.3%), and proton therapy (20.4%). The median RT prescription was 50.4 Gy. The mean heart dose was 23.2 Gy (IQR 16.7-29.5); the mean lung dose was 9.3 Gy (IQR 6.6-11.5). On univariate analysis, there was a strong association between heart dose and AE at or above 20 Gy. The V20 Gy OR was 1.20 (95% CI 1.08-1.33, p Z 0.001), the V30 OR was 1.24 (95% CI 1.11-1.38, p < 0.0001), and the V40 OR was 1.18 (95% CI 1.03-1.35, p Z 0.0176). This observation was similar on multivariate analysis adjusting for age, smoking history and previous history of heart disease (OR 1.24, p Z 0.0002). With regard to lung dose and AE, no pulmonary DVH parameter was statistically associated with a risk of pulmonary AE. When associating heart and lung DVH values to any form of AE, the cardiac V30 was associated with an increase in AE on univariate (OR 1.18, 95% CI 1.08-1.29, p Z 0.0003) and multivariable models (OR 1.15, 95% CI 1.04-1.26, p Z 0.0047). Conclusion: This dosimetric review of clinical outcomes in a multi-institutional cohort suggests that heart dose is a significant predictor of cardiac AE in patients undergoing curative treatments for esophageal cancer. Further studies are warranted investigating planning technique for reduction in cardiac dose.