The recurrence pattern of esophageal carcinoma after transhiatal resection

Background: There is much controversy about the optimal resection for carci noma of the esophagus. Little is known about the pattern of recurrence afte r transhiatal resection for esophageal carcinoma. Study Design: We retrospectively reviewed the charts of 149 patients who un derwent transhiatal esophagectomy for carcinoma of the mid or distal esopha gus or gastroesophageal junction between June 1993 and June 1997. Recurrenc e was classified as locoregional or distant recurrence. Nine patients with macroscopically evident tumor left after resection and three patients (2.0% ) who died in the hospital were excluded from the analysis. This left 137 p atients; 105 men and 32 women with a median age 65 years (range 37 to 84 ye ars). Results: There were 95 adenocarcinomas (69.3%) and 42 squamous cell carcino mas (30.7%). Overall the median followup was 24.0 months (range 1.4 to 69.2 months). For patients alive at the end of followup without recurrence, the median followup was 36.5 months (range 23.6 to 69.2 months). Seven patient s died of other causes. The median interval between operation and recurrenc e was 11 months (range 1.4 to 62.5 months) for patients who had recurrence, with no significant difference in interval between locoregional and system ic recurrence. Seventy-two of the 137 patients (52.6%) developed recurrent disease. Thirty-two patients (23.4%) developed locoregional recurrence only , 21 patients (15.3%) developed systemic recurrence only and 19 patients (1 3.9%) had a combination of both. In only 8.0% of all patients was there rec urrence in the cervical lymph nodes. The most frequent sites of distant rec urrence were liver (37.5%), bone (25.0%), and lung (17.5%). Recurrence was related to postoperative lymph node status (p < 0.001) and the radicality o f the operation (p < 0.001) in multivariate analysis. Recurrence was not as sociated with localization or histologic type of the tumor. Conclusions: Recurrence after transhiatal resection is an early event. Almo st 40% of patients developed locoregional recurrent disease. For this patie nt group a more extended procedure may be of benefit, especially in the pat ients (23.4%) with locoregional recurrence in whom this is the only site of recurrent disease. But the potential benefit of a more extended procedure has to be balanced against a possible increase in perioperative morbidity a nd mortality. (J Am Coll Surg 2000;191: 143-148. (C) 2000 by the American C ollege of Surgeons).