Laparoscopic repair of large paraesophageal (type II) or mixed (type III) hiatal hernias (HH) with prosthetic hiatal closure is now considered feasible and effective with similar results to open surgery. However, in most studies, the outcome analyses are based on symptomatic assessment and relatively short-term follow-up (F-up); concerns remain on possible high recurrence rates in the long-term F-up with objective tests. We retrospectively evaluated the outcomes of laparoscopic repair of large type II-III HH, with or without prosthetic hiatal closure, in the long-term F-up, by means of both symptomatic and radiologic/ endoscopic evaluation. From January 1992 to June 2005, 54 pts (10M:4F, median age 64.5 yrs) with a diagnosis of large type II-III HH (>1/3 radiographic/intraoperative stomach in the chest) underwent laparoscopic repair. Principles of surgical technique included reduction of the hernia, complete excision of the sac, primary closure of the crura (with/without mesh reinforcement) and antireflux procedure (Nissen or Toupet). Simple sutured crural closure was performed in 19 pts (group A), a double mesh was added in 35 pts (group B). Median operative time was 180 min. The operation was completed laparoscopically in 89% of the cases: all but one conversions occurred in the first 13 pts. Intraoperative complications occurred in 2 (3.7%) pts, including 1 spleen lesion and 1 gastric lesion. Postop complication rate was 9%, including 1 cardiorespiratory arrest with apallic status. The median lenght of radiographic/endoscopic F-up was 64 mos (IQR:10-98) group A and 27 mos (IQR:7-47) group B (p=0.05). Recurrences occurred in 11/54 (20%) pts: 8/19 (42.1%) group A pts and 3/35 (8.6%) group B pts (p=0.01). The 3 recurrences in group B (mesh) pts occurred all within the first 12 mos post-op; 5/8 recurrences in group A pts occurred ≥3 yrs after surgery. 5 pts with recurrence underwent reoperation. 4/54 (7%) pts presented a small sliding hernia (≤2cm ‘telescoping’) on post-op barium swallow and did not require reoperation (they are all asymptomatic at median F-up of 93 mos). On multivariate logistic regression analysis, only the absence of mesh significantly predicted hernia recurrence or wrap migration in the chest. Laparoscopic repair of large type II-III hiatal hernias is a safe and effective treatment. Short-term symptomatic results are excellent, but long-term objective radiologic and/or endoscopic evaluation reveals a significant percentage of recurrence: the main reason for failure of hiatal repair is tension. The use of prosthetic mesh effectively reduce the incidence of postoperative hernia recurrence or wrap migration in the chest.