Pessaux P, Arnaud J, Delattre J, Meyer C, Baulieux J, Mosnier H, Association Française de Chirurgie. Laparoscopic Antireflux SurgeryFive-Year Results and Beyond in 1340 Patients. Arch Surg. 2005;140(10):946-951. doi:10.1001/archsurg.140.10.946
Although the long-term results of open fundoplication for gastroesophageal reflux disease are well documented, few reports exist on the long-term results of laparoscopic fundoplication.
Retrospective study with clinical evaluation or mailed survey for patients unable to return to the hospital center.
Multicenter studies (ie, private medical centers, institutional hospitals, and university hospitals).
Between January 1992 and December 1998, 2684 patients with gastroesophageal reflux disease underwent laparoscopic fundoplication in 31 hospital centers. Outcome data covering a period of 5 or more years after surgery were available for 1340 patients: 711 who underwent complete fundoplication, 559 who underwent partial posterior fundoplication, and 70 who underwent partial anterior fundoplication.
Main Outcome Measures
Evaluation of clinical and quality-of-life actions used to treat the symptoms of gastroesophageal reflux disease.
The overall residual severe dysphagia rate was 5.1% (n = 68). A further surgical procedure was required for 59 patients (4.4%) for a total of 63 interventions. Subsequent operation was performed laparoscopically in 32 cases (50.8%). Twelve of these procedures were for the repair of a paraesophageal hiatus hernia, 11 were for dysphagia (4 because of a tight esophageal hiatus and 7 for conversion of Nissen fundoplication to a posterior partial fundoplication procedure), 31 were for recurrent reflux (wrap undone), 2 were for intestinal obstruction (adhesiolysis), 1 was for incisional hernia, 1 was for abdominal abscess (drainage), and 1 was for gastroparesis (pyloroplasty). The recurrence rate was 10.1% (n = 136), and 122 patients (9.1%) resumed taking antisecretory medication. Gas bloat syndrome was present in 101 patients (7.5%). A total of 93.1% of the patients were satisfied (Visick classification, grades 1 and 2) and 6.9% were unsatisfied, with no difference among the 3 procedures.
After 5 years of experience, laparoscopic fundoplication remains an effective antireflux procedure.
Since its introduction in 1991,1,2 laparoscopic antireflux surgery has been gaining popularity among surgeons as the procedure of choice for the treatment of severe symptomatic gastroesophageal reflux disease (GERD). With a short hospital stay, moderate postoperative pain, and complete relief of most reflux symptoms, the frequency of laparoscopic antireflux surgery is increasing.3 Many procedures that are performed to treat GERD are reproduced by laparoscopy. Before this surgery finds universal acceptance by physicians and patients as a true alternative to medical therapy, its long-term efficacy needs to be established. Few previous studies4- 8 have reported 5-year follow-up for a large group of patients who underwent a laparoscopic antireflux procedure. The aim of this retrospective study was to evaluate surgical outcome after laparoscopic fundoplication in patients followed up for at least 5 years.
Between January 1992 and December 1998, 2684 patients with GERD underwent laparoscopic fundoplication in 31 hospital centers. Sixty-one surgeons participated in this study, and each surgeon performed more than 20 procedures. There were 1648 men (61.4%) and 1036 women (38.6%) (mean ± SD age, 48.9 ± 13.8 years; age range, 18-94 years). Of these patients, 1593 were eligible for 5 or more years of follow-up. We mailed a notification to each patient. To maximize the response rate, we mailed a second copy of the notification after 2 weeks and attempted to call those who had not responded after 6 weeks. There were 1340 respondents (84.1%) to the mailed notification. A complete medical examination was performed in 1116 patients. A physician completed the survey by telephone for 224 patients (16.7%) unable to return to the hospital center. The predominant symptoms were heartburn and retrosternal pain, and the associated symptoms were classic dyspepsia, regurgitation, dysphagia, chronic cough, and respiratory complications with exacerbation of asthma. The indications for surgery were intractable or recurrent symptoms due to GERD after an adequate trial (minimum of 3 months) of conservative treatment that consisted of proton pump inhibitors (n = 1234; 92.1%); complications of reflux esophagitis, stricture, and Barrett esophagus (n = 33; 2.5%); complications of large sliding or paraesophageal hiatal hernia (n = 5; 0.4%); and an associated disease (n = 68; 5%) (symptomatic cholecystolithiasis [n = 40], inguinal hernia [n = 12], incisional hernia [n = 10], genital prolapse [n = 2], ovarian cyst [n = 2], hematologic spleen [n = 1], and symptomatic uterine fibroma [n = 1]).
The preoperative evaluation included the following: complete history and physical examination for all patients, upper gastrointestinal contrast study in 1016 patients (37.8%), and esophagogastroduodenoscopy with biopsy in 1285 patients (95.9%), of whom 1049 (81.6%) had esophagitis, 5 (0.4%) had stenosis, and 203 (15.8%) had healthy mucus because they were healed by medical treatment. There was 28 cases (2.2%) of Barrett esophagus without dysplasia. Short esophagus was present in 134 patients (10%) and hiatal hernia in 304 patients (22.7%). Esophageal manometry was performed in 924 patients (68.9%), and 24-hour pH monitoring was performed in 815 patients (60.8%).
Postoperative clinical follow-up was performed with a standardized questionnaire. The questionnaire covered 3 main domains: gastrointestinal symptoms, actions required to treat gastrointestinal symptoms, and satisfaction with surgery. The presence or absence of heartburn and liquid- and solid-food dysphagia, as well as patient satisfaction with the procedure, was graded as a product of severity (no, moderate, or severe symptoms). The presence or absence of gaseous bloating, the ability to belch, the ability to relieve abdominal distension, the ability to eat a normal diet, and patients’ opinions on whether they would undergo the same procedure again under similar circumstances were also determined. Details of adverse outcomes such as hospital readmission, complications, or surgical revision were recorded. Patients ranked the outcome of surgery using the Visick classification (grade 1, no symptoms; grade 2, minimal symptoms, no lifestyle changes, no need to see a physician; grade 3, significant symptoms that require lifestyle changes with a physician’s help; grade 4, symptoms as bad or worse than preoperatively). All data analysis was performed on an intention-to-treat basis, and patients whose laparoscopic procedure was converted to an open procedure, as well as those who required later surgical revision, remained in the overall patient group for data analysis.
Three operations were performed: Nissen complete 360° fundoplication (n = 711; 53.1%), Toupet partial posterior 180° fundoplication (n = 559; 41.7%), and partial anterior fundoplication (n = 70; 5.2%). In 368 patients (27.5%), short gastric vessels were divided to achieve adequate looseness of the fundoplication wrap (305 patients with complete fundoplication and 63 patients with partial posterior fundoplication). The crura were reapproximated only if a large defect existed (n = 1203; 89.8%). Mesh was placed at the hiatus in 8 patients (0.3%). In 500 patients (37.3%), the fundoplication was constructed around a 50F Maloney bougie catheter within the lumen of the esophagus. In the other cases, the wrap was constructed without calibration. The fundic wrap was fixed to the esophagus in 860 cases (64.2%) and to the right crus of the diaphragm in 653 cases (48.7%). The wrap was smaller than 3 cm in 46 cases (3.4%), between 3 and 5 cm in 1000 cases (74.6%), and larger than 5 cm in 294 cases (22.0%).
Data are expressed as mean ± SD. Statistical analysis was performed with the χ2 test, Fisher exact test, or t test. Data were analyzed using StatView 4.5 statistical software (SAS Institute Inc, Cary, NC). Statistical significance was set at P<.05.
The mean follow-up was 7.1 ± 1.5 years (range, 5-10 years). There were 815 men (60.8%) and 525 women (39.2%), with a mean ± SD age of 49.3 ± 13.6 years (age range, 18-86 years). The mean duration of symptomatic GERD was 6.8 years (range, 3 months to 45 years). All patients had undergone medical therapy with a mean duration of 1.9 years (range, 3 months to 12 years).
No deaths occurred. The conversion rate was 8.3% (n = 112) (Table 1). The mean operative time was 118 ± 18 minutes, including 177 additional procedures (Table 2). The mean postoperative stay was 5.0 ± 3.2 days (range, 1-48 days). There were 29 intraoperative complications (2.1%), but not all needed a conversion. There were 2 hemorrhages from a superficial splenic lesion that required splenectomy, 14 from short gastric vessels, 3 from sectioning of an anomalous left hepatic artery, and 1 from an aortic injury. The postoperative complication rate was 2.6% (n = 35) (Table 3).
A further surgical procedure was required for 59 patients (4.4%) for a total of 63 interventions. Subsequent operation was performed laparoscopically in 32 cases (50.8%). Twelve of these procedures were for the repair of a paraesophageal hiatus hernia, 11 were for dysphagia (4 because of a tight esophageal hiatus and 7 for conversion of the Nissen fundoplication to a posterior partial fundoplication procedure), 31 were for recurrent reflux (wrap undone), 2 were for intestinal obstruction (adhesiolysis), 1 was for incisional hernia, 1 was for abdominal abscess (drainage), and 1 was for gastroparesis (pyloroplasty). The esophagus hiatus was not closed in 10 (83.3%) of the patients with paraesophageal hiatus herniation. The incidence of paraesophageal hernia in the 134 patients with a short esophagus was 0.7% (n = 1) vs 0.9% (n = 11) in other patients. Among the 31 patients operated on for recurrent reflux, 4 had a new recurrence and were treated by a duodenal diversion. Subsequent operation was performed within 3 months of the initial procedure in 14 of the patients (22.2%), between 3 months and 2 years in 32 patients (50.8%), and beyond 2 years in the remaining 17 patients (27.0%).
Transient postoperative dysphagia (a sensation of hindrance to the passage of swallowed food into the stomach) was common and was observed in 575 patients (43.1%) in the first 3 months. In most cases, the dysphagia improved by 6 weeks after surgery and did not require any intervention. The mean duration of dysphagia was 10.4 ± 23.9 weeks (range, 1-104 weeks). Forty patients with severe dysphagia submitted to endoscopic dilation, with success in 35 patients (2.6%); the other 5 patients underwent subsequent operations (2 because of a tight esophageal hiatus and 3 for the conversion of Nissen fundoplication to a posterior partial fundoplication procedure). Residual dysphagia was observed in 68 patients (5.1%) at 5 years. The gas bloat syndrome decreased to 11.8% of the patients at 3 months, 8.7% at 2 years, and 7.5% at 5 years. Paired short-term and longer-term follow-up data were available for 968 patients. Table 4 gives the Visick grades and adverse symptoms or effects after surgery for all patients at short-term and longer-term follow-up points. Of the 1340 patients, 1248 (93.1%) reported that they were satisfied with the result of surgery at 5-year follow-up. Although 260 patients (19.4%) experienced adverse effects of the operation (dysphagia or gas bloat syndrome), most believed that these were minor compared with the improvement in reflux symptoms. Only 92 patients (6.9%) reported adverse effects combined with dissatisfaction with the results of surgery. When asked whether they would have the operation again, 1261 (94.1%) answered affirmatively. Of the 1340 patients evaluated at 5 years, 122 patients (9.1%) resumed taking antisecretory medication for recurrence of reflux. Of 55 patients (45.1%) who underwent an objective 24-hour pH study, 34 patients (61.8%) had abnormal acid reflux.
At 5 years and beyond, the 3 procedures produced no difference in terms of functional results (Table 5). The rate of recurrent reflux was higher after partial anterior fundoplication even though the dysphagia rate was higher after complete fundoplication. Among the 711 complete fundoplications, 404 procedures (56.8%) were performed without division of the short gastric vessels and 305 procedures (42.9%) with division of the short gastric vessels (missing data in 2 subjects). There was no difference in terms of recurrent reflux, dysphagia, gas bloat syndrome, or Visick grades (Table 6).
In the Western world, approximately 10% of adults experience typical or atypical symptoms of GERD daily.9 The aim of antireflux surgery is the elimination of GERD symptoms and the prevention of further complications. Antireflux surgery is entering a new era with the advent of minimally invasive techniques.1,10 Several studies11- 16 have shown that laparoscopy is a safe and effective procedure with good short-term results, but information on longer-term outcome is required to confirm its durability. However, the laparoscopic procedure has its complications, and failure rates range from 4.5% to 12.5% in larger series. Among the different intraoperative or postoperative complications, some are specific to laparoscopy.
In this study, the intraoperative and postoperative complication rates were 2.1% and 2.6%, respectively, with a subsequent operation rate of 4.4% at 5 years. Of interest, overall patient satisfaction following surgery was high, with 93.1% of patients satisfied with their long-term outcome and 94.1% willing to have the surgery again. Although nearly 10% of patients resumed taking antisecretory medication, in most no evidence of reflux recurrence could be found. Our follow-up was clinical only, and objective tests, such as 24-hour pH monitoring, were not performed systematically. A significant proportion of the patients who were taking medication after the operation were using it to relieve nonreflux symptoms, such as bloating and epigastric pain, and had normal acid exposure times and improved reflux symptoms. There is a risk that the resumption of using antacids or antisecretory medication is a sign of recurrence. Objective tests are necessary to evaluate the true outcomes of surgical intervention. Recent studies5,6,8 have also reported using antisecretory medication in up to 15% of patients after laparoscopic antireflux surgery.
Dysphagia was common in the immediate postoperative period, but it decreased by 6 weeks of follow-up. The subjective degree of dysphagia and the perceived impairment can be predicted according to the personality of the patient.17 Excessive tightening of the wrap or the esophageal hiatus, which is the cause of early dysphagia, may be difficult to assess at operation because of reduced tactile feedback during laparoscopic surgery. It has been suggested that failure to completely mobilize the fundus results in greater degrees of dysphagia. Three prospective randomized trials18- 20 compared laparoscopic Nissen fundoplication with and without the division of short gastric vessels. Division of the short gastric vessels does not improve any measured clinical outcome at 2 or 5 years of follow-up and is associated with an increased incidence of gas-related problems. Our study is in accordance with these results. Manometry complements the preoperative workup, primarily by assessing esophageal body function. If motility is normal or minimally ineffective, a total fundoplication can be performed. If motility abnormalities are major, a partial fundoplication is indicated. Other authors21 have suggested that in this situation a fundoplication can be performed with a floppy wrap. However, there is no evidence from controlled trials to support this proposal. Other studies22 suggest that a loose Nissen fundoplication may produce equally good outcomes in patients with poor motility. Finally, for other authors, intraoperative manometry plays an important role in improving the clinical outcome after fundoplication, but according to the literature, its use is still controversial.23 The use of preoperative workup before antireflux surgery is also still controversial. Routine preoperative manometry and pH monitoring have been advocated to prevent misdiagnosis of GERD.24 Other authors25 argue that selective use of preoperative pH monitoring and esophageal manometry in patients with GERD might be cost-effective and safe. These tests might be useful in the patient with atypical symptoms or abnormal findings on upper gastrointestinal tract fluoroscopy but should not be used routinely.
Paraesophageal hiatus herniation is a specific complication of laparoscopic surgery that has not been reported with open procedures.26 Previous series report an incidence of paraesophageal herniation of up to 7%.27- 29 The elevated intra-abdominal pressure due to the pneumoperitoneum, extensive esophageal dissection further into the thorax, and failure to close the esophageal hiatus usually account for this complication. In addition, vomiting or other forms of early exertion can push the stomach into the thorax. In this study, the crura were not closed in most patients who later developed herniation of the wrap into the chest. Routine hiatal repair has been shown to reduce its incidence by 80%,26 and most surgeons would agree that narrowing or reinforcing the hiatus with sutures is mandatory during laparoscopic antireflux surgery. The use of prosthetic material to reinforce a large hiatal repair can also reduce the risk of recurrent herniation.30 However, complications may arise from the use of mesh at the hiatus, in particular, intraluminal migration of the mesh. Hiatal repair and liberal use of antiemetics may reduce the risk of this complication.
In the laparoscopic era, few studies have compared different procedures. The question of which fundoplication technique offers the best outcome for patients undergoing surgery for GERD is still controversial. There are only 3 prospective randomized trials that compared Nissen fundoplication vs partial posterior fundoplication31,32 and Nissen fundoplication vs partial anterior fundoplication.33 In the first trial, a partial or complete wrap after division of the short gastric vessels offered effective therapy for reflux, with more then 90% patient satisfaction. The study showed no clear advantages of one wrap (partial posterior or complete) over the other. In the second study, partial posterior fundoplication had a lower rate of dysphagia and was as effective as the Nissen fundoplication in controlling reflux. The follow-up in that study was short (4 months). In the third study, laparoscopic anterior partial fundoplication achieved equivalent control of reflux but an improved clinical outcome at 6 months. The follow-up was also short (6 months), and continued follow-up is necessary to confirm the long-term results of the partial anterior fundoplication procedure. Hagedorn et al34 compared the efficacy of 2 partial fundoplications. Laparoscopic posterior fundoplication offered advantages in terms of better reflux control compared with anterior fundoplication. Recently, 2 publications35,36 concluded that laparoscopic Toupet fundoplication should not be used as a standard antireflux procedure, particularly in patients with severe or complicated reflux disease. However, in other studies the Toupet procedure was demonstrated to be safe and effective in treating the symptoms of GERD, including patients with severe forms of the disease37 or those with impaired esophageal peristalsis.38 The present study has revealed no difference among the 3 different procedures in terms of functional results. At 5 years, more than 90% of the patients were satisfied (Visick grades 1 and 2), with no difference among the 3 groups.
One of the highly debated issues within GERD surrounds the existence of the short esophagus. Although many esophageal surgeons currently contend that short esophagus exists as a complication of long-standing GERD, other surgeons argue against the existence of such a disease process.39 Intrinsic shortening of the esophagus most commonly occurs in patients with chronic GERD that involves recurring cycles of inflammation and healing, with subsequent fibrosis. The actual incidence of the short esophagus is estimated to be approximately 10% of patients undergoing antireflux surgery.40 Of this group, 7% can be appropriately treated with extensive mediastinal mobilization of the esophagus to achieve the required esophageal length. The remaining 3% require an aggressive surgical approach, including the use of gastroplasty procedures, to create an adequate length of intra-abdominal esophagus to perform a wrap. It has been suggested that problems such as postoperative paraesophageal hiatus herniation after laparoscopic fundoplication arise in patients with unrecognized esophageal shortening41 and that because laparoscopic esophageal lengthening procedures are difficult, most surgeons are ignoring the possibility of esophageal shortening so that they can simplify their laparoscopic technique. In this study, short esophagus was not a risk factor for developing paraesophageal herniation even though antireflux procedures were performed without any esophageal lengthening procedures.
Several critical observations of this study should be emphasized. First, 84.1% of the surgical population supplied follow-up information at a mean of 7.1 years after surgery. The follow-up rate of 84% is better than in most previous reports, in which the rate was between 80% and 60%.5,6,8,42 The remaining 16% may have altered the overall results. The most complete report to date is the study by Lafullarde et al,4 who document a 99% follow-up rate for 178 patients between 5 and 8 years after operation. Second, to date, long-term follow-up studies have included patients from the learning curve. The results of studies of patients operated on beyond this initial learning curve experience can be expected to show further improvements in long-term outcome. Third, because physician factors may be related to outcomes, the multicenter character of this study allows the extrapolation of the results to the daily surgical practice. Fourth, because all patients underwent laparoscopic surgery, we cannot compare surgical with medical outcomes or outcomes for patients undergoing open vs laparoscopic surgery. In conclusion, the results of this study do not differ significantly from the data reported in the literature, suggesting that laparoscopic antireflux surgery is an effective long-term procedure, is well tolerated, and can be properly used in the treatment of GERD.
Correspondence: Jean-Pierre Arnaud, MD, CHU Angers, Department of Visceral Surgery, 4 Rue Larrey, 49033 Angers CEDEX 01, France (firstname.lastname@example.org).
Accepted for Publication: November 26, 2004.
Members of the Association Française de Chirurgie: Surgeons participating in this study include J. F. Ain (Macon), I. Al Naasan (Chambery), J.-P. Arnaud (Angers), J. M. Atah (Amiens), D. Balenbois (Aubagne), J. Baulieux (Lyon), J. P. Bazin (Elbeuf), P. Berard (Lyon), P. Bouchet (Chambery), L. Bresler (Nancy), C. Bunea (Amiens), J. L. Cardin (Laval), M. Carretier (Poitiers), B. Castel (Lille), Y. Chen (Grenoble), D. Collet (Bordeaux), J. M. Cucchi (Bordeaux), A. Dabrowski (Seclin), N. De Manzini (Strasbourg), J.-F. Delattre (Reims), C. Delteil (Chateauroux), A. Dhahri (Lyon), J. Domergue (Montpellier), N. El Zeenni (Altkirch), Y. Fouques (Caen), G. Fourtanier (Toulouse), A. Gainant (Limoges), F. Gayral (Bicêtre), J. F. Gillion (Thiais), C. Gouillat (Lyon), L. Grimaud (Lyon), A. Hamy (Nantes), P. Herbiere (Albi), J. Lagoutte (Arnas), J. M. Leloup (Ermont), C. Letoublon (Grenoble), B. Masson (Bordeaux), M. Mathonnet (Limoges), C. Meyer (Strasbourg), C. Mondésert (Feurs), H. Mosnier (Paris), J. Paineau (Nantes), D. Paterne (Bicêtre), P. Pessaux (Angers), D. Pezet (Clermont-Ferrand), J. P. Porta (Desertines), S. Rohr (Strasbourg), P. Segol (Caen), J. F. Sledzianowski (Toulouse), O. Touchard (Nancy), G. Trébuchet (Rennes), J. P. Triboulet (Lille), J. R. Tubiana (Elbeuf), and P. Verhaeghe (Amiens).
Previous Presentation: This study was previously presented at the European Surgical Association Meeting; April 3, 2004; Barcelona, Spain.