Background: The role of neoadjuvant chemotherapy and radiation therapy before resection in esophageal cancer remains controversial. Operative risks may be increased, but this has not been systematically addressed in published trials or reports.
Methods: This was a prospective, nonrandomized, restricted cohort design of patients (n = 200) from 1997 to 2003 with resectable cancer of the esophagus or esophagogastric junction. A total of 102 patients underwent multimodal therapy with 5-fluorouracil, cisplatin, and radiation therapy before surgery, and 98 patients opted for surgery alone. In-hospital mortality and morbidity were the primary end points, and cancer survival was a secondary end point.
Results: In patient cohorts matched for operative risk factors, the odds ratio for postoperative sepsis (P = .007), respiratory failure (P = .009), and acute respiratory distress syndrome (P = .02) was increased in the multimodal group. There was no significant difference between groups comparing median and 1-, 2-, and 3-year survivals.
Conclusions: Multimodal therapy was associated with increased respiratory and septic complications compared with a surgery-only cohort undergoing the equivalent surgery. Respiratory failure was in most cases idiopathic. The data suggest that efforts should be made to limit radiation lung exposure in multimodal regimens, and to understand and modulate the local and systemic effects of preoperative chemoradiation.