Between January 2012 and July 2015, 97 consecutive patients who underwent nCRT followed by transthoracic esophagectomy were included in this single-center cohort study. The gastric fundus was contoured on the pretreatment planning computed tomography. Within this contour, dose-volume histogram variables were calculated, and logistic regression analysis was used to determine their influence on the risk of anastomotic leakage.
In 25 of 97 patients (26%) anastomotic leakage occurred. The mean radiation dose to the gastric fundus was significantly higher in patients with than without leakage (median 35.6 Gy versus 24.9 Gy, respectively, p = 0.047). A mean dose more than versus less than 31.4 Gy was associated with leakage rates of 43% versus 15%, respectively. Adjusted for tumor location, clinical T stage, and radiation method, the mean radiation dose to the gastric fundus remained significantly and independently associated with an increased risk of anastomotic leakage (adjusted odds ratio 1.05 per 1-Gy increase, 95% confidence interval: 1.002 to 1.10, p = 0.043).
Efforts should be made to minimize the radiation dose to the gastric fundus when planning nCRT for esophageal cancer, because higher dose levels to the gastric fundus are associated with an increased risk of anastomotic leakage after subsequent transthoracic esophagectomy and cervical anastomosis.