QUALITY OF PALLIATION AND POSSIBLE BENEFIT OF EXTRAANATOMIC RECONSTRUCTION IN RECURRENT DYSPHAGIA AFTER RESECTION OF CARCINOMA OF THE ESOPHAGUS

BACKGROUND: After ''curative'' resection of carcinoma of the esophagus , late secondary dysphagia almost invariably indicates locoregional tu mor recurrence. The retrosternal reconstruction route is advocated to prevent ingrowth of tumor recurrence in the neoesophagus. STUDY DESIGN : To evaluate the quality of palliation after ''curative'' resection o f carcinoma of the esophagus and the possible benefit of the retroster nal reconstruction route, we retrospectively analyzed the records of p atients who had resection of a malignant tumor of the esophagus, or th e gastroesophageal junction, and a prevertebral reconstruction. The ex traanatomic route would have been only beneficial for patients with in trathoracic tumor recurrence distant from the anastomosis and causing gastrointestinal symptoms. RESULTS: Between 1983 and 1989, 209 patient s (mean age of 61.3 years at the time of operation) had ''curative'' r esection and prevertebral reconstruction in the institution of this st udy. Seventy-three patients (35 percent) had locoregional tumor recurr ence. Univariate and multivariate analysis of various risk factors for locoregional recurrence showed that the presence of positive lymph no des (pN(1)), especially if located at the celiac trunk (pM(1)), and a macroscopically nonradical R(2) resection were the most important risk factors. Forty-six patients (22 percent) had secondary dysphagia as a result of locoregional tumor recurrence, mostly (18 percent) within t wo years postoperatively. Dysphagia lasted on average 5.3 months (rang e of 0.3 to 21.5 months) before the patients died. In 27 patients (13 percent), dysphagia would probably have been prevented by using a retr osternal reconstruction route. CONCLUSIONS: These data are an argument in favor of the extra-anatomic, retrosternal reconstruction route aft er limited transthoracic or transhiatal resection in the presence of p ositive lymph nodes. This method seems especially indicated if the nod es are located at the celiac trunk and in case of a macroscopically no nradical R(2) resection.