Single-Center Experience of Laparoscopically Assisted Percutaneous Endoscopic Gastrostomy Placement

Introduction

Gastrostomy tube (GT) placement is required for a variety of reasons in children, including neurologic disability, nutritional supplementation in children with chronic medical conditions, short-gut syndrome, and food aversion. Percutaneous endoscopic gastrostomy (PEG) insertion was the gold standard for gastrostomy placement in children. The major drawback with this technique is that it is a blind procedure. Complications that have been reported for PEG placement in children include: gastrocolic fistula, colocutaneous fistula, placement of the gastrostomy on the inferior border of the stomach, worsening of gastroesophageal reflux, and gastric separation.^1–3

With the introduction of laparoscopy, there are five main techniques for gastrostomy placement: 1) conventional open (Stamm) gastrostomy; 2) PEG; 3) fluroscopy-guided gastrostomy ⁴ ; 4) laparoscopic gastrostomy without endoscopy ⁵ ; and 5) laparoscopic-assisted PEG. ⁶ In 2006, our institution experienced two episodes of colocutaneous fistula following conventional PEG placement. This complication led to a change in technique. Laparoscopic assistance was introduced in an aim to minimize these complications. On the initial patient, a single port was attempted, but after noting the rotation of the stomach on insufflation, an atraumatic grasper was introduced through a working port to stabilize the stomach. The aim of this study was to present the results from the first 2 years of our prospective audit of laparoscopically assisted PEG placement.

Materials and Methods

Between July 2006 and July 2008, all patients undergoing laparoscopically assisted PEG placement, with and without laparoscopic Boix-Ochoa fundoplication, were prospectively audited. Data presented are supported by a case note review and are presented as medians (range).

Results

There were 29 children (16 male) who underwent a laparoscopically assisted PEG placement during the study period. Five infants had gastrostomy placement as part of a laparoscopic Boix-Ochoa fundoplication. The median weight was 12.3 kg (range, 4.5–76). Median age at operation was 1 year and 8 months (range, 6 months to 16 years and 2 months). Total length of procedure, including anesthetic time, was 80 minutes (range, 45–120).

Feeding was commenced on the same operative day for the laparoscopically assisted PEGs and the following day for those undergoing concomitant laparoscopic fundoplication. Total length of stay was 2 days (range, 2–12). Current follow-up is 6 months (range 1–21). Seven patients have had subsequent low-profile gastrostomy devices fitted.

There was 1 death, unrelated to surgery, in a patient with a progressive neuromuscular disease. There were no complications pertaining to laparoscopy. One patient experienced displacement of the gastrostomy on postoperative day 2. This was due to a faulty gastrostomy, and the manufacturer has accepted liability for the complication. Three patients developed superficial wound infections of the PEG site, and 1 had an overgranulation of the PEG site, which required excision under a general anesthetic.

Discussion

PEG placement was introduced in children in 1980. ⁷ PEG-related complications have been well described,^1,2,8 and major complications and mortalities from this technique have been reported in children. In a series of 130 children undergoing PEG placment, major complications were reported in 17.5%. These included gastrostomy leaks requiring laparotomy in 6%, gastrocolocutaneous fistula in 3%, intestinal obstruction in 3%, and significant hemorrhage in 3%. ⁸

To reduce these complications, there are three main alternatives in children. The first is to perform an open gastrostomy. This has increased morbidity, a larger scar, increased postoperative analgesia requirements, and increased length of hospital stay in patients with potentially complex medical and neurologic pathologies. The second technique is a laparoscopic gastrostomy. This was first described by Rothenberg et al. in 1999 ⁵ and has been reported subsequently in several large series.^9,10 In this technique, the laparoscope is placed in the umbilicus and the working port in the left-upper quadrant at the site where the gastrostomy is intended. The stomach is grasped and pulled out through the working port, where it is secured to the anterior abdominal wall. This technique has been shown to have a lower overall and procedure-specific complication rate than conventional PEG placement. ¹¹ The most recent modification to this technique uses a 4-mm port and a bronchoscopic optical grasper as a single port and has been used in infants as small as 1.4 kg. ¹² Reported complications for this technique are gastric fallaway and perforation of the posterior wall of the stomach.^10,11,13 The third possible alternative, which is the technique adopted by us, is the addition of a laparoscope (with or without working ports) to conventional PEG placement. This was first described by two separate researchers in 1995^14,15 and is the method of choice for gastrostomy placement during laparoscopic fundoplication. ⁶

The techniques described in the literature mainly use a three-port technique,^14,16,17 with some researchers reserving this technique for complex or difficult patients in which conventional PEG placement carries an increased risk of complications.^15,18 The largest series to use this technique placed gastrostomies in 34 children experiencing no major complications. ¹⁶ In a series of 51 mainly undergoing laparoscopic fundoplication, 8 had this performed as a sole procedure and also experienced no complications. ¹⁹ Long-term data from a 5-year follow-up study of 15 children (6 with a fundoplication) having laparoscopically assisted PEG placement reported no complications at the end of the study. ²⁰ In our series of 24 children undergoing this procedure without a fundoplication, we experienced no complications relating to laparoscopy and only 13.7% had minor wound infections. The evidence from our series and the literature would suggest that morbidity is eliminated with laparoscopically assisted PEGs and minor complications are reduced.

Conclusions

PEG placement has the potential for serious complications. Several alternative techniques are available. The addition of a laparoscope with a single working port to conventional PEG placement provides direct visualization of the stomach and aids accurate site placement. Major complications are eliminated and minor complications reduced. Laparoscopically assisted PEG placement is a safer technique than PEG placement and a viable alternative to laparoscopic gastrostomy placement.