Background: Because of the inaccessibility of mediastinal nodal metastases, the left thoracoabdominal approach (LTA) has often been used to treat gastric cancer of the cardia or subcardia. In a randomised phase III study, we aimed to compare LTA with the abdominal-transhiatal approach (TH) in the treatment of these tumours.
Methods: Between July, 1995, and December, 2003, 167 patients were enrolled from 27 Japanese hospitals and randomly assigned to TH (n=82) or LTA (n=85). The primary endpoint was overall survival, and secondary endpoints were disease-free survival, postoperative morbidity and hospital mortality, and postoperative symptoms and change of respiratory function. The projected sample size was 302. After the first interim analysis, the predicted probability of LTA having a significantly better overall survival than TH at the final analysis was only 3.65%, and the trial was closed immediately. Analysis was by intention to treat. This study is registered with , number NCT00149266.
Findings: 5-year overall survival was 52.3% (95% CI 40.4-64.1) in the TH group and 37.9% (26.1-49.6) in the LTA group. The hazard ratio of death for LTA compared with TH was 1.36 (0.89-2.08, p=0.92). Three patients died in hospital after LTA but none after TH. Morbidity was worse after LTA than after TH.
Interpretation: Because LTA does not improve survival after TH and leads to increased morbidity in patients with cancer of the cardia or subcardia, LTA cannot be justified to treat these tumours.