Laparoscopic Redo Nissen Fundoplication After Previous Open Antireflux Surgery in Infants and Children

Introduction

Surgical antireflux procedures are some of the most common procedures performed by pediatric surgeons today. Many studies have documented the safety and long-term efficacy of laparoscopic Nissen fundoplication for the treatment of gastroesophageal reflux disease in the pediatric population.^1–3 However, recurrent gastroesophageal reflux is among the most common complications after a fundoplication procedure, and both open and laparoscopic procedures are associated with recurrence rates of 3%–40%. ⁴ Traditionally, this has been managed surgically with revision fundoplication via open laparotomy, with the potential for significant postoperative morbidity. Although there have been multiple studies in the adult literature that have shown that revision laparoscopic fundoplication in the setting of a prior open antireflux procedure is safe, there is a paucity of data in the pediatric literature. ⁵ The purpose of this study was to retrospectively review our experience with laparoscopic redo fundoplication after previous open antireflux surgery in the pediatric population.

Subjects and Methods

The study was approved by the Institutional Review Board of our institution (protocol number 6323). A retrospective review was conducted to identify all patients who underwent a laparoscopic Nissen fundoplication between September 2004 and September 2012. Those patients with a history of a prior open antireflux operation were selected for this study. All operations were performed by three pediatric surgeons whose catchment area includes 3.6 million patients, all of whom are linked by a single electronic healthcare record and central database registry. The outpatient and inpatient electronic medical records were reviewed, and the abstracted data included patient demographics, comorbidities, time since prior open antireflux procedure, presenting symptoms, preoperative workup, operative time, operative complications, time to enteral nutrition postoperatively, postoperative complications, and need for subsequent antireflux procedures. In all cases, gastroesophageal reflux was identified clinically and then confirmed radiographically. Laparoscopic Nissen fundoplications were performed at the discretion of the attending surgeon.

Revision laparoscopic fundoplications were performed via a five-port access technique. The gastrostomy, when present, was taken down laparoscopically to facilitate exposure of the region of the esophageal hiatus. The internal gastrotomies were then closed with intracorporeal knots. Extensive adhesiolysis with complete mobilization of the stomach was then performed. The previous fundoplication was also taken down. All hiatal hernias were repaired over a bougie dilator, with sutures anteriorly and posteriorly, as needed. The esophagus was sutured to the right and left crura in four places. A standard 360° fundoplication was performed over a bougie dilator using nonabsorbable sutures. The length of the fundoplication was 2 cm. The fundoplication was sutured to the right and left crura.

Results for the variables are reported as frequency distribution percentage and mean with standard deviation.

Results

One hundred eighty laparoscopic Nissen fundoplications were performed between September 2004 and September 2012; 23 were redo laparoscopic Nissen fundoplications. Twelve of the 23 patients had a history of prior open fundoplication(s), and 1 of these 12 patients had a history of two prior fundoplications. There were 6 girls and 6 boys. Mean age was 11.6±5.5 years (range, 4–21 years). Mean weight was 30.34±14 kg (range, 13.8–57.4 kg). Eight patients (67%) had a history of cerebral palsy and/or seizure disorder. The average time between initial open antireflux procedure and subsequent revision laparoscopic Nissen fundoplication was 113.7±64 months (range, 19–215 months). Seven patients presented with emesis, 4 with aspiration pneumonia, and 1 with clinical reflux. Eleven patients had further workup with an upper gastrointestinal series test, and 1 patient who presented with emesis and aspiration pneumonia had a computed tomography scan that showed a hiatal hernia. This is summarized in Table 1.

As shown in Table 2, laparoscopic revision was completed in 100% of the patients, with no intraoperative complications. Average operative time was 177.5±86 minutes (range, 45–294 minutes). Nine patients (75%) were noted to have hiatal hernias at the time of the repair. None of the defects in the hiatus was large enough to require repair with synthetic material, such as polytetrafluoroethylene. Half of the patients (n=6) had gastrostomy tubes preoperatively, all of which were replaced during the procedure. Two patients who did not have a gastrostomy tube prior to the surgery had one inserted during the procedure. Seven patients (58%) were able to resume feeds on postoperative Day 1, with an average of 3.4±3.8 days before advancement to full feeds. Median length of stay was 3 days (range, 2–40 days).

Median follow-up was 21 months (range, 6–34 months). Two patients required additional procedures in the immediate perioperative period: 1 returned to the operating room on postoperative Day 5 for repositioning of his gastrostomy to a new site due to inflammation at the original location, and a second patient underwent an esophagogastroduodenoscopy on postoperative Day 8 for food lodged at the gastroesophageal junction. Three patients (25%) required subsequent procedures in the long term: 2 patients underwent lysis of adhesions for small bowel obstructions at 2 and 20 months postoperatively, respectively, and 1 had a laparoscopic takedown of the Nissen fundoplication and creation of a Toupet fundoplication 8 months postoperatively for persistent esophageal dilatation and symptomatic dysphagia. Seven patients (58%) were taking proton pump inhibitors 3 months postoperatively. The patients were kept on proton pump inhibitors at the discretion of their primary care physicians despite no clinical evidence of gastroesophageal reflux. Thus far, none of the patients has required a redo fundoplication for a recurrent hiatal hernia or recurrent gastroesophageal reflux.

Discussion

The most commonly performed surgical procedure to treat pathologic gastroesophageal reflux disease is fundoplication. Recurrent gastroesophageal reflux is among the most common complications after a fundoplication procedure. Prior studies have already demonstrated that redo fundoplications performed laparoscopically have success rates similar to those of open redo surgery, but there are few data to show that laparoscopic redo fundoplication is feasible and safe after a prior open antireflux procedure in the pediatric population.^6,7 In fact, many surgeons avoid complex laparoscopic surgery in the setting of prior open procedures because of concerns of adhesions.

There is robust adult literature to show that laparoscopic redo fundoplication is safe and effective in the management of a failed prior antireflux surgery. Van Beek et al. ⁵ recently completed a meta-analysis of 17 series representing 1167 adult cases of laparoscopic redo fundoplication and showed that although the redo procedure resulted in longer operative times and higher incidence of postoperative complications compared with the primary fundoplication procedure, conversion rate and length of stay were similar to those of primary antireflux procedures. The success rates, although defined variably between studies, were on average comparable to primary antireflux procedures at 81%. Furthermore, they determined that previous open operations should not be considered a contraindication to laparoscopic revision. ⁵

The few studies that have looked at redo laparoscopic fundoplications in pediatric patients tend to group together those who underwent prior open and prior laparoscopic procedures.^4,8,9 No study has been reported that only focused on revision laparoscopic fundoplication performed after an open antireflux procedure. Barsness et al.^10,11 have been able to demonstrate the safety, feasibility, and efficacy of redo Nissen fundoplication, whether done laparoscopically or open, in infants and children after previous open abdominal operations. Although their results were not separated by type of previous open abdominal operation, their study noted no conversions, and similar to our study, all revisions were able to be completed laparoscopically. They also noted earlier return to enteral feeds after laparoscopic fundoplication compared with open fundoplication.^10,11 The current study was able to show that revision laparoscopic fundoplication after a prior open antireflux surgery is feasible, with minimal perioperative morbidity. The longer operative time can be attributed to dense adhesions and the technique of completely taking down the prior fundoplication and repair of large hiatal hernias prior to the revision.

The largest similar study to date by Rothenberg ⁴ consisted of 118 patients, 64 of whom had had previous open fundoplications. All were done at a single center over a 10-year period by one pediatric surgeon. Although the current study is smaller in number, it demonstrates that equivalent results can be obtained by three different surgeons who performed the procedures in different hospitals. The low rate of perioperative complications and the lack of need to convert to open procedures, as demonstrated in both the study of Rothenberg ⁴ and our current study, support the safety and feasibility of a laparoscopic revision after prior open fundoplication.

The limitations of this study should be acknowledged. It was retrospective in nature, and long-term follow-up greater than 2–3 years is not available. Although some have suggested that the percentage of patients still receiving proton pump inhibitors can be used as a measurable and objective indication of surgical success, many patients continue proton pump inhibitor therapy after surgery even when no identifiable reflux is shown in pH studies. ⁵ This is largely due to the preferences of the child's pediatrician and/or gastroenterologist, as risks with its continuation are thought to be low. Instead, this study looked at the number of patients who require subsequent operations as a measure of long-term postoperative outcomes. Only 2 patients required minor additional procedures in the immediate perioperative period. Although 1 of the patients with chronic esophageal dilatation underwent a takedown of the fundoplication to treat his dysphagia, none of the patients has needed an additional antireflux procedure. This series noted an average time between the initial open antireflux procedure and the subsequent revision laparoscopic Nissen fundoplication of almost 10 years. This suggests that a longer follow-up of 10–20 years will be necessary to detect the true rate of recurrence.

In summary, a previous open antireflux procedure is not a contraindication to a minimally invasive approach for surgical treatment of recurrent gastroesophageal reflux disease. Laparoscopic revision Nissen fundoplication after a prior open antireflux procedure is both feasible and safe.