23 Review on cardiac toxicity grade 3 or higher reported cardiac toxicity in these papers are listed in and marked as the first three papers in table 1. The most frequently observed side effects were cardiac ischemia, pleural and pericardial effusions and heart failure. These “late toxic events” presented occurred relatively soon after treatment, with a median follow up of 26.1 to 57 months. Grade 3 or higher cardiac toxicity, which is considered clinically relevant, was seen in 5.8%- 11.1% of the patients. Given the low cure rates of esophageal cancer in these studies, with 3 years survival rates varying between 22% and 45%, the actuarial rates were not reported, but can be expected to be much higher. Morota et al. also reported on patient and treatment related risk factors for cardiac events.[3] Older age (>75 years) was the only factor significantly associated with late cardiopulmonary toxicity. They reported a crude incidence of 29% in the older patient group vs. 3% for the younger patients. In three other studies, the authors aimed to find clinical and dosimetric factors influencing the risk for pericardial(PCE) or pleural effusion(PE).[5–7] Treatment details are summarized as the next three papers in table 1. Martel et al. was the first to report on pericardial effusion among patients treated with 3D-CRT based on planning-CT and available diagnostic data.[5] Between 1985 and 1991, patients were treated according to 3 different protocols. The only prognostic factor significantly associated with PCE was the dose per fraction (3.5 Gy). After correction, according to the Linear Quadratic model, the mean and maximum heart doses were significant prognostic factors. However, given the relatively small sample size, fitting the data into the Lyman model showed large confidence intervals. Wei et al. performed a retrospective analysis to identify clinical and dosimetric prognostic factors for PCE in 101 patients with inoperable esophageal cancer.[6] The pericardium was contoured as a shell, by extending the actual heart contour with 0.5 cm. PCE was scored using CT scans routinely made during follow-up visits. The mean time to onset of PCE was 5.3 months, leveling off at 16.7 months after treatment. The crude incidence of PCE was 27.7% and the actuarial incidence at 18 months was 48%. No patient or treatment-related factors could be found that were associated with PCE. However, significant associations were found with several dosimetric factors. Pericardial DVH values correlated better with the incidence of PCE as compared to the cardiac DVH parameters. If the mean pericardium dose was 2
RkJQdWJsaXNoZXIy MTk4NDMw