Uncut Collis-Nissen fundoplication: Learning curve and long-term results

Background. Between September 1985 and July 1990, the first 100 consecutive patients (50 female and 50 male) undergoing primary uncut Collis- Nissen fundoplication performed by one surgeon were reviewed. Methods. Median age was 62 years and ranged from 19 to 89 years. Indications for repair included gastroesophageal reflux in 56 patients, obstructive symptoms in 34, and a combination of both in 10. An upper gastrointestinal endoscopy was performed in 99 patients; all were abnormal. Esophagitis was documented in 53 patients, large diaphragmatic hernia in 36, stenosis in 18, 'Cameron's erosions' in 17, Barrett's disease in 13, and other findings in 9 patients. An abnormal upper gastrointestinal series was demonstrated in 96 of 97 patients evaluated. Motility studies were performed in 95 patients, and 11 had abnormal peristalsis. All procedures were performed through a left thoracotomy. Results. Complications occurred in 23 patients and included respiratory failure in 6, atrial fibrillation in 3, atelectasis in 3, pneumonia in 2, myocardial infarction in 2, and chylothorax, severe dysphagia, early breakdown of repair, cardiac tamponade, hematuria, spinal headache, and intraoperative perforation by dilator in 1 each. There were 2 postoperative deaths, both cardiac in origin. Median hospitalization was 8 days (range, 6 to 76 days). The first 25 patients had 10 complications (40%) and 2 deaths (8%). The remaining 75 patients had 13 complications (17%) and no deaths (mortality, p = 0.06; morbidity, p = 0.03). Follow-up was complete in all patients for a median of 100 months (range, 3 to 138 months). Eighty- six of the 98 operative survivors are currently alive. At last follow-up, excellent functional results were observed in 58 patients (59%), good in 24 (25%), fair in 8 (8%), poor in 7 (7%), and unknown in 1 (1%). Conclusion. We conclude that the uncut Collis-Nissen fundoplication provides good to excellent long-term results in 84% of patients. Operative mortality and morbidity is acceptable but is associated with a learning curve.