The preoperative predictability of the short esophagus in patients with stricture or paraesophageal hernia.

BACKGROUND: Esophageal shortening is a known complication of advanced gastroesophageal reflux disease that may preclude a tension-free antireflux procedure. A retrospective analysis was performed to test the accuracy of preoperative testing.
METHODS: From September 1993 to December 1998, 39 patients underwent esophageal mobilization with intraoperative length assessment. Patients were selected on the basis of irreducible hiatal hernia, stricture formation, or both. Patients in the upright position with a fixed hiatal hernia larger than 5 cm on an esophagram were considered to have a short esophagus. Manometric length two standard deviations below the mean for height was considered abnormally short.
RESULTS: In 31 patients, intraoperative mobilization was sufficient to allow the gastroesophageal junction to lie 2 cm below the diaphragmatic crus, so no esophageal-lengthening procedure was required. Eight patients with a short esophagus required an esophageal-lengthening procedure after complete mobilization. Two patients subsequently underwent intrathoracic migration of the gastroesophageal junction (GEJ), with recurrence of symptoms and required gastroplasty during the second surgery. An esophagram had a sensitivity of 66% and a positive predictive value of 37%, whereas manometric length had a sensitivity of 43% and a positive predictive value of 25% for the diagnosis of short esophagus. The preoperative endoscopic finding of either a stricture or Barrett's esophagus was the most sensitive test for predicting the need for a lengthening procedure.
CONCLUSIONS: Manometry and esophagraphy are not reliable predictors of the short esophagus. Additional tests and/or tests combined with other parameters are needed.