Background: There are few prospective studies that document the histological follow-up after antireflux surgery of patients with Barrett's esophagus (BE) by the recently standardized criteria. We report the clinical, endoscopic (EGD) and histological results of patients with BE, followed postoperatively for at least 2 years.
Methods: Diagnosis of BE required preoperative endoscopic evidence of columnar lined esophagus (CLE), which on biopsy showed specialized intestinal metaplasia (IM) positive for Alcian blue stain. Between April 1993 and November 1998, 104 patients meeting these criteria underwent fundoplication (Laparoscopic (Lap) Nissen, 84; Open Nissen, 6; Lap Toupet, 11; Lap Collis-Nissen, 1; Lap Collis-Toupet, 1; Open Dor, 1). Short segment BE (IM length < 3 cm) was found preoperatively in 32% and low grade dysplasia was found in 4% of patients. All patients were contacted yearly by mail, phone, or clinic visit. Surveillance EGD and biopsy strategy employed the Seattle protocol (Reid, Gastro., 1992).
Results: At a mean follow-up of 4.6 years (range 2-7.5), 81% of patients were off antisecretory medications and 97% were satisfied with their operation. Eight patients have undergone reoperations for recurrence of symptoms. Two patients have died (peri-operative ARDS, pancreatic cancer) and 2 were excluded from EGD biopsy because of portal hypertension. Sixty-six patients complied with the surveillance protocol, and returned the histological results to our center. Ten patients had no CLE on most recent EGD, but of these, 3 had IM on biopsy. Fifty-six patients had CLE, with histological results of IM (32), LGD (1), cardiac-fundic type metaplasia (CFM, 21)- no biopsy was performed in 2. Patients who have had histological regression of IM had shorter segments of BE (2.8±2.4 cm, p<0.01) and longer follow-up (5.6±1.7 yrs, p<0.05) than those patients with persistent IM (5.5±3.7 cm and 4.7±1.3 yrs). Follow-up of the 34 patients who have refused surveillance EGD revealed 2 patients that take medication for reflux symptoms. No patient has developed high grade dysplasia or esophageal carcinoma during surveillance endoscopy (337 total patient-yrs of follow-up).
Conclusions: The incidence of regression of IM to CFM after successful antireflux surgery is greater than previously reported. We suspect that this is a result of longer follow-up and the inclusion of patients with short segment BE. A substantial number of patients with BE, who are asymptomatic after anti-reflux surgery refuse surveillance endoscopy.