Background: Failure of antireflux surgery has been difficult to define and usually consists of reports on reoperations. An alternative approach is to consider any dissatisfied patient as a failure as opposed to only those with a correctable functional or anatomic abnormality. This is a more realistic evaluation of the success of antireflux surgery. We hypothesize that failure of laparoscopic Nissen (LN) is more common if patient dissatisfaction, rather than need for reoperation, is used as a marker. Methods: 667 patients had primary LN between 1995-2003. Preop evaluation included video esophagram, endoscopy, manometry and pH testing. Clinical failure was a dissatisfied patient for any reason. Patients who failed were evaluated for functional and/or anatomic abnormalities. Known preop predictors for failure (atypical symptoms, poor response to PPIs, normal pH score, non-reducible hernia, long segment Barrett’s, stricture) were recorded. Results: 42 patients (6%) were dissatisfied with the results of LN and were classified as clinical failures. This occurred between 6-76 months postoperatively. 11 patients (26%) were anatomically and functionally normal, including normal acid exposure. A predictive factor for failure was present in 5/11 (45%), the most common being atypical symptoms and poor response to PPIs (4/5). None had re-operation. 12 patients (29%) were anatomically normal but functionally abnormal. All had normal acid exposure. A predictive factor for failure was present in 10/12 (83%): 7 had atypical symptoms and 3 had typical symptoms but normal acid exposure. Re-operation was performed in 1 patient. 19 patients (45%) had anatomic abnormalities (18 recurrent hernia and/or 7 altered Nissen). Patients with anatomic abnormalities presented later than those without (22 vs. 13 months, p=0.01). A predictive factor for failure was present in 17/19 (89%), the most common was a preop non-reducible hiatal hernia (12/17). Thirteen had postop pH monitoring and abnormal acid exposure was present in 7 (54%). Fifteen had postop manometry and it was abnormal in 8 (53%). Three patients with anatomic abnormalities were functionally normal. Re-operation was performed in 13 patients (12 for recurrent hernia and 1 for a slipped Nissen). Conclusion: Overall failure of LN is 6% and no anatomical or functional abnormality was observed in 26%. These patients commonly have atypical preoperative symptoms questioning the correct diagnosis. Patients with an anatomic abnormality commonly have a functional abnormality and most require reoperation. The majority of these patients have a non-reducible hernia prior to the initial operation suggesting esophageal shortening.