The choice of the surgical procedure should be tailored to the individual patient by considering a number of factors: reasons for failure of the first operation, esophageal length, peristaltic reserve, presence of Barrett’s esophagus, and concomitant gastric pathology. In most patients, laparoscopic fundoplication revision is feasible, although dealing with adhesions of a previous laparotomy may require extra time and increase the morbidity of the procedure. Esophageal resection should only be considered in patients with multiple previous repairs, extensive fibrosis with stricture refractory to endoscopic dilatation, and evidence of dysplasia on Barrett’s esophagus [30].

Patients with a slipped/misplaced Nissen require an esophageal lengthening procedure combined with re-fundoplication if the esophagus is found to be truly short. Complete takedown of the old repair is a mandatory step before considering any surgical option. A stapled wedge resection of the gastric fundus [31] provides a safe esophageal elongation and is easier to perform and to teach compared to the Steichen “buttonhole” technique, requiring both a circular and linear stapler, and to the transthoracic gastroplasty [32]. It has been proposed that in borderline case vagotomy may represent a safe alternative to the Collis gastroplasty in patients with excessive longitudinal tension [33]. In patients with chest herniation of the wrap attention should be directed to assess the tissue quality of the crura and to consider the opportunity of mesh reinforcement [25] and/or crural relaxing incisions [34].

Impairment of esophageal motility may indicate the opportunity to perform a partial 270° Toupet rather than a 360° Nissen fundoplication. This is the case when more than 30% of the esophageal waves are synchronous or the mean amplitude is less than 30 mmHg, or when the criteria for the diagnosis of ineffective esophageal motility are met at high-resolution manometry based on the Chicago classification. An esophageal myotomy combined with a Dor fundoplication is usually performed in patients with previously misdiagnosed achalasia [35, 36].

In some patients, a re-fundoplication cannot be performed because the fundus is inadequate for any type of repair. An alternative surgical strategy, which is especially useful after multiple previously failed surgical attempts, consists of vagotomy, antrectomy, and Roux-en-Y reconstruction to effectively reduce both acid and alkaline components of the refluxate [15, 37]. Laparoscopic gastric bypass is another option that can be considered in obese patients with recurrent reflux symptoms after failed antireflux surgery [38, 39].

Pyloroplasty, or even a total gastrectomy in extreme cases, may be indicated in the occasional patients who present with severe gastroparesis, especially after inadvertent vagotomy at the time of the index operation [40].

Historically, reoperations for failed antireflux procedures were performed through an open trans-abdominal or trans-thoracic technique [41–44]. As experience with advanced minimally invasive surgery has increased, more redo operations are performed laparoscopically, and it appears that the thoracic approach has now been actually abandoned.

The principles of reoperative laparoscopic surgery are similar to those of the open approach. All redo procedures should be considered complex and should be scheduled as the first case of the day. On table endoscopy is routinely planned after induction of anesthesia, and the scope is left in the esophagus for intraoperative evaluation. Five access ports are used. Initial port placement is generally performed using the Hasson trocar inserted away from any previous incisions. Adhesiolysis between the stomach and the liver and around the hiatus may be long and tedious. Extreme care should be taken to avoid perforations of the stomach and esophagus and injury to the vagal trunks. Full mobilization of the fundoplication and the lower mediastinum is performed by removing the crural sutures and by taking down the short gastric vessels. A linear stapler can help dividing the two halves of the wrap (Fig. 12.2). The fat pad should be routinely excised to identify the true gastroesophageal junction, and a 3 cm tension-free intra-abdominal esophageal segment should be obtained. Care should be taken to minimize tension on the crura repair by clearing the entire surface of the right crus and decrease the insufflation pressure to less than 10 mmHg to facilitate approximation of the muscle bundles. The hiatus can be repaired with interrupted non-absorbable stitches, and placement of a composite or synthetic absorbable mesh should be considered (Fig. 12.3).

If a short esophagus is suspected, a modified Collis wedge gastroplasty procedure can be performed. Once the gastric fundus has been completely freed from posterior and lateral adhesions, a bougie is inserted in the esophagus under direct laparoscopic visualization and placed across the gastroesophageal junction along the lesser curve. The fundus is retracted inferiorly to the patient’s left side, and sequential fires of a linear stapler are directed toward the bougie to a point 3 cm below the gastroesophageal junction. The gastroplasty is then completed by resecting the wedge of fundus with the stapler applied parallel to the bougie toward the angle of His. The operation ends with a Nissen or Toupet procedure around the neo-esophagus (Fig. 12.4).

A systematic review and meta-analysis of laparoscopic revisional antireflux surgery, including 19 case series and one case-control study, reported on 922 patients operated between 1990 and 2010 [45]. The mean surgical duration was 166 minutes and the conversion rate to open revision 7%. The most prevalent indication to reoperation was reflux (61%) followed by dysphagia (31%), gas-bloat syndrome (4%), regurgitation or vomit (3%), and chest pain (2%). The most common anatomic problem found at reoperation was mediastinal migration of the wrap. Nissen fundoplication was performed in 70% of patients. The overall complication rate was 14% (0–44%). A satisfactory to excellent result was reported in 84% of patients, while 5% of patients required further surgery.

Long-term results of the Linx procedure in patients with uncomplicated gastroesophageal reflux disease have shown relief of reflux symptoms, discontinuation of daily therapy with proton pump inhibitors, and objective reduction of esophageal acid exposure. In addition, patients maintain the ability to belch and vomit [46, 47]. A recent case-control study found similar control of reflux symptoms after Nissen fundoplication or Linx implant at 1 year follow up. However, the Nissen group showed a higher rate of patients with inability to belch and vomit, along with more severe gas-bloat symptoms [48]. Concerns regarding the safety of the Linx procedure, especially the fear of erosions, stem from past adverse experience with the Angelchick device and, more recently, with the gastric banding device. However, a recent analysis of the safety profile of the first 1000 worldwide implants in 82 hospitals showed 1.3% hospital readmission rate, 5.6% need of postoperative endoscopic dilations, and 3.4% reoperation rate [49].

Reoperation for removal of the Linx device consists of a one-stage laparoscopic procedure with intraoperative endoscopic assistance. Pneumoperitoneum is established with a Veress needle and the abdomen is entered through the prior port-sites. The scar tissue at the gastroesophageal junction corresponding to the site of the Linx implant is identified. A monopolar electrocautery hook is used to cut the scar tissue and to expose a pair of anterior titanium beads. The independent titanium wire connecting the beads is cut with ultrasonic scissors, and one bead is grasped with an Endoclinch and retracted upward (Fig. 12.5). This allows step by step cutting of the thin fibrous capsule overlying each bead and pulling out of the device, entirely or in two pieces. The total bead count in the explanted device is confirmed and the device removed through a 10 mm port. Intraoperative endoscopic assistance helps to check the integrity of the esophageal mucosa during and after removal, and/or to assist during retrieval of the beads migrated into the esophageal lumen. A concurrent antireflux repair (partial or total fundoplication) is then performed.

A recent study focused on reoperations for Linx removal and reported the long-term results of one-stage laparoscopic removal and fundoplication [50]. In this series, out of 164 patients implanted with a Linx device, 11 (6.7%) were explanted at a later date. The main presenting symptom requiring device removal was recurrence of heartburn or regurgitation in 46%, dysphagia in 37%, and chest pain in 18%. In two patients (1.2%) full-thickness erosion of the esophageal wall with partial endoluminal penetration of the device occurred. The median implant duration was 20 months, with 82% of the patients being explanted between 12 and 24 months after the implant. Device removal was most commonly combined with partial fundoplication. There were no conversions to laparotomy and the postoperative course was uneventful in all patients. At 12–58 months after surgery, the GERD-HRQL score was within normal limits in all patients.

Revisional surgery after failed antireflux repairs is technically demanding and requires careful preoperative and intraoperative assessment to identify the cause of the failure and to tailor the procedure to the individual patient. Appropriate training, expertise, and strict adherence to established surgical principles is necessary to overcome the challenge of redo antireflux surgery. Today, more redo operations are attempted laparoscopically with reported low conversion rates, minimal morbidity, and good success rate. Due to the continuously rising epidemic of gastroesophageal reflux disease, reoperative hiatus surgery remains a challenge whose complexity and volume is likely at least to remain stable in the future.