Laparoscopic Gastrostomy Is Superior to Percutaneous Endoscopic Gastrostomy Tube Placement in Children Less Than 5 years of Age

Abstract

Purpose:

Minimally invasive procedures for enteral access in children have evolved over the years, resulting in various techniques of gastrostomy tube placement. The two most common techniques are laparoscopic gastrostomy (LG) and percutaneous endoscopic gastrostomy (PEG). Our study compares the outcomes of both procedures exclusively in children under the age of five.

Methods:

All procedures relating to enteral access in children <5 years of age were reviewed retrospectively from July 2009 to July of 2014 as approved by our Institutional Review Board. Demographics, techniques, and complications were collected and analyzed.

Results:

Of 293 patients in our study, 150 patients underwent PEG, 75 LG, and 68 LG with Nissen fundoplication (LNG). The most common indication for enteral tube placement was failure to thrive and feeding intolerance. Operative time was less in the PEG group than in the other two groups (P = .001). Overall complication rate was 60% for LG and LNG and 58% for PEG (P = NS). The major complication rate was 3.3% in the PEG group and 0.7% for the LG and LNG groups. There were two deaths in the PEG group. Sixty-eight patients (45.3%) from the PEG group underwent tube changes under anesthesia, requiring additional trip to the operating room with general anesthesia compared with LG and LNG groups (2%) (P = .001). From the PEG group, 134 patients (89%) required many fluoroscopic interventions for tube dislodgments and conversion to gastrojejunostomy tubes for significant reflux and inability to use the gastrostomy (P = .001).

Conclusion:

PEG tubes had a higher major complication rate than LG tubes with or without fundoplication in children <5 years of age. Despite longer operative time, LG seems to be the procedure of choice for children of this age for enteral access. Elimination of unnecessary tube changes under anesthesia and the fluoroscopic interventions after the PEG would be beneficial.

Introduction

Indications for a gastrostomy tube vary. Gastrostomy is often required for infants and children with neurological disorders, short gut syndrome, nutritional supplementation in children with cancer, and other chronic medical diseases to aid in long-term enteral feeding.¹ Minimally invasive procedures for enteral access in children have evolved significantly over the years, resulting in various techniques of gastrostomy tube placement. Our institution commonly performs laparoscopic gastrostomy (LG) with Nissen fundoplication (LNG) or without Nissen fundoplication and the percutaneous endoscopic gastrostomy (PEG) for enteral access.

The LNG is frequently performed in children with gastroesophageal reflux and microaspiration. Few studies have compared the surgical and clinical outcome of various types of gastrostomy tube placement in children <5 years of age. The purpose of this study is to compare the clinical and surgical outcomes of LG or LNG versus PEG exclusively in children <5 years of age.

Materials and Methods

An Institutional Review Board (IRB) approval was obtained. We conducted a 5-year retrospective analysis from July 2009 to July 2014 at a free-standing children's hospital. Children <5 years old who underwent an LG tube, PEG tube, or LNG tube placement were included in this study. For PEG insertions, a Corpak/Bower or CorFlo Max (CORPAK Medsystem, Buffalo Grove, IL) was used in our patient population. The PEG was performed by a gastroenterologist and/or pediatric surgeon.

For LG the U-stitch laparoscopic technique was performed as developed by Georgeson.² The LG and LNG were performed by pediatric surgeons. All procedures were performed under general anesthesia and all of the patients received prophylactic antibiotic therapy. An upper gastrointestinal (GI) study was obtained postoperatively in most of the patients. The primary outcome points of interest were the primary diagnosis, age at the time of surgery, surgical technique, days to initiation and goal enteral feeds, length of stay (LOS), complications, and mortality rate. Long-term outcomes were fluoroscopic interventions, tube changes, reoperation, morbidity, and mortality rates.

The study population was described using univariate analysis. Bivariate analysis was completed to compare LG versus LNG versus PEG utilizing Pearson's chi square or Fisher's exact test for categorical variables and Student's t-test for continuous data. A P value of <.05 was considered statistically significant.

Results

Patients

There were 293 children <5 years of age included in this study. There were slightly more males (56%) in the study population. A PEG was performed in 150 (51%) and the remaining patients underwent an LG (75) or LNG (68) (49%) (Table 1). The median age at the time of surgery was 11 months (IQR: 0.5–48 months). Patients who underwent an LNG were significantly younger than children who underwent an LG or PEG with a mean age of 7.3 months, P ≤ .001. The mean weight at the time of surgery was 7.9 ± 3.4 kg. The most common indications for a PEG and LG were failure to thrive and feeding intolerance. Gastroesophageal reflux disease (GERD) was the most common indication in children undergoing LNG. More than 35% (105) of the study population had a neurological disorder and 20% (58) had a cardiac disorder.

Table 1.

Demographics

	LG (n = 75)	LNG (n = 68)	PEG (n = 150)
n = 293	Mean (SD)			P
Age (months)	13.9 (12.1)	7.3 (7.1)	17 (10.6)	<.001
Weight (kg)	8 (3.7)	6.4 (3.4)	8.7 (2.9)	<.001
	n (%)
Female	32 (45)	31 (46)	67 (45)	.937
Male	43 (55)	37 (54)	83 (55)
Neurological disorder	26 (35)	27 (40)	52 (35)	.8861
Cardiac disorder	16 (21)	19 (28)	23 (15)	.0699
Indication for G-tube	68% failure to thrive/feeding intolerance	76% GERD	45% failure to thrive/feeding intolerance

GERD, gastroesophageal reflux disease; LG, laparoscopic gastrostomy; LNG, laparoscopic gastrostomy with Nissen fundoplication; PEG, percutaneous endoscopic gastrostomy; SD, standard deviation.

Operative time

Operating time was significantly longer in children undergoing an LNG (mean OR time 98.4 minutes) than those undergoing an LG (47.3 minutes) and PEG (22.4 minutes), P ≤ .001. There was no difference in days before initiating feeds after surgery, P = .806. The LOS was significantly higher in patients undergoing LNG (mean 20.2 days), P ≤ .001. There were no open conversions in the LG and LNG.

Complications

Many patients had multiple complications (Table 2). Granuloma was the most common minor complication requiring silver nitrate cauterization. Other complications included infection, leak, prolapse, and gastrocolonic fistula. Four (2 LG and 2 PEG) patients had cellulitis requiring hospitalization and antibiotic therapy. Two patients who underwent a PEG developed a gastrocolonic fistula, which was subsequently surgically resected in both patients. There were no intraoperative complications in either group. There were no operative related mortalities in patients who underwent an LG or LNG. However, there were two deaths in the PEG group. Both patients developed peritonitis after a dislodged tube was inadvertently placed into the peritoneal cavity that was unrecognized. In addition, one patient died in the LNG group, unrelated to that procedure from congenital cardiac disease.

Table 2.

Complications

n = 293	LG (n = 75)	LNG (n = 68)	PEG (n = 150)	P
Morbidity
Minor, n (%)
Dislodged tube	14 (18.7)	13 (19.1)	21 (14)	.51
Granuloma	25 (33.3)	16 (23.5)	47 (31.3)	.39
Infection/cellulitis	5 (6.7)	1 (1.5)	10 (6.7)	.266
Leak	7 (9.3)	4 (5.9)	6 (4)	.241
Major, n (%)
Prolapse	1 (1.3)	0	1 (0.67)	1.0
Gastrocolonic Fistula	0	0	2 (1.33)	.737
Mortality	0	0	2 (1.33)	.737

LG, laparoscopic gastrostomy; LNG, laparoscopic gastrostomy with Nissen fundoplication; PEG, percutaneous endoscopic gastrostomy.

A significant number of patients from the PEG group (n = 68) underwent planned gastrostomy tube changes to a MIC-KEY (Kimberly Clark, Inc., Dallas, TX) button requiring additional general anesthesia compared with LG/LNG groups, P = .001 (Table 3). In the PEG group, 18 patients had their gastrostomy tube converted to a gastrojejunostomy (GJ) tube for reflux. There was no difference between the number of patients requiring conversion to a GJ in the PEG versus LG versus LNG, P = .409 (Table 3). In the PEG group, 134 patients required on average of 4.8 fluoroscopic interventions for dislodged tubes and conversion to GJ tubes for significant reflux, P = .001. Although more patients in the PEG group required fluoroscopic interventions, there was no difference between the number of fluoroscopic interventions per patient in the three groups, P = .835.

Table 3.

Comparison of Surgical Outcomes

	LG (n = 75)	LNG (n = 68)	PEG (n = 150)
n = 293	Mean (SD)			P
OR time (minutes)	47.3 (36.3)	98.4 (58.7)	22.4 (18)	<.001
Time to feeds (days)	1.7 (1.7)	1.7 (1.8)	2.8 (19.9)	.806
LOS (days)	15.8 (27.9)	20.2 (31.2)	7 (14)	<.001
Number of fluoroscopic interventions per patient	5.2 (3.4)	4.9 (3.7)	4.8 (5.9)	.835
	n (%)
Patients requiring tube change to GJ	14 (19)	9 (13)	18 (12)	.409
Change to MIC-KEY under anesthesia	2 (2.7)	1 (1.5)	68 (45.3)	<.001
Patients requiring fluoroscopic intervention	16 (21.3)	41 (60.2)	134 (89.3)	<.001

GJ, gastrojejunostomy; LG, laparoscopic gastrostomy; LNG, laparoscopic gastrostomy with Nissen fundoplication; LOS, length of stay; PEG, percutaneous endoscopic gastrostomy; SD, standard deviation.

Discussion

We conducted a retrospective chart analysis to determine the surgical outcomes exclusively in children <5 years of age undergoing gastrostomy tube placement. This study is one of the largest reviews evaluating and comparing gastrostomy tube placement techniques in children <5 years of age. The PEG was the most common procedure performed at our institution. We found that children undergoing LNG were significantly younger than children undergoing a PEG or LG. Patients who underwent an LNG had a significantly longer LOS, which could be attributed to other comorbities.

Compilations associated with gastrostomy tube placement regardless of surgical technique are quite common. The most common minor complication was granuloma formation (30%), requiring silver nitrate cauterization in all groups. The second most common complication was dislodged tube (16.3%), requiring replacement and a fluoroscopic study. Although two patients in the LG group were rehospitalized for cellulitis, there were no major complications associated with the LG and LNG requiring additional hospitalization or surgical intervention. There were two gastrocolonic fistulas in the PEG group requiring resection of the fistula. In addition, there were two deaths in the PEG group directly associated with the gastrostomy tube dislodgement/replacement.

Studies have shown that children and infants undergoing a PEG have a high long-term risk of morbidity related to the gastrostomy tube.³ In addition, major complications may occur at the time of gastrostomy tube exchange.³ There is a 10–15% major complication rate in children undergoing PEG placement.^3–8 Children with a ventriculoperitoneal shunt are at a higher risk for complications after PEG placement.^3,8 In our study, we had 3.3% major complication rate with PEG, including prolapse and gastrocolonic fistula. Although our major complication rate was lower than rates in the literature, there were two deaths associated with inadvertent intraperitoneal insertion of a dislodged tube leading to peritonitis, exploratory laparotomy, and subsequent mortality. Since these two deaths, the institution has provided extensive education regarding management of gastrostomy tubes to ancillary staff and physicians across multiple disciplines. In addition, an institutional protocol has been instituted to manage dislodged tubes. As such, we strongly advocate that fluoroscopic studies are completed after reinsertion of a dislodged gastrostomy tube to ensure placement in the stomach.

There was a 0.7% major complication rate in children undergoing an LG/LNG including herniation and prolapse. In the literature, there is a 3%–7% major complication rate in patients undergoing LG. Multiple studies suggest that an LG is superior to PEG in children requiring gastrostomy tube.^9–12 Zamakhshary et al.¹ found a significantly higher complication rate in children undergoing a PEG (14%) than in those undergoing LG (7.7%), although Brewster et al.,⁵ found that PEG complication rates were similar to LG complication rates.

Advantages to the LG are direct observation of the abdominal cavity and attachment of the stomach to the abdominal wall (either temporarily with U-stitches or more permanently with fascial stitches). Although we did not evaluate the cost of gastrostomy placement and subsequent interventions, the LG may be more cost effective than a PEG placement. Although the operating room cost may be the lowest in the PEG group due to a shorter operating room time, it is plausible that the PEG has a higher overall cost due to multiple postoperative interventions. In this study, 45.3% of PEG group required tube change to a MIC-KEY button under general anesthesia compared with LG (2.7%) and LNG (1.5%), increasing overall reintervention rate in the PEG group significantly. Furthermore, there is added risk associated with additional general anesthesia required for tube changes. In the LG and LNG groups, a MIC-KEY low profile button is placed at the time of surgery. In the PEG group, many patients were informed that a low-profile gastrostomy tube may be placed at a subsequent procedure if desired (usually 6 weeks after PEG), but this is done under general anesthesia due to the young age and 45% of our PEG patients elected to have this done (68/150). Fluoroscopic studies were required in 89.3% of the PEG group for malfunctioning, obstructed, and/or dislodged tubes, which definitely increases the radiation exposure in these children as well as likely the cost, although we did not evaluate this objectively. A total of 32 patients (14 LG and 18 PEG) had a conversion to a GJ tube for significant reflux symptoms. It is possible that these children may have benefited from an initial LNG. Based on our results, we recommend the LG (or LNG) over PEG in children <5 years of age when technically feasible because it has less complications and need for reintervention after placement.

This study is limited by the retrospective nature and thus may be limited by accurate documentation in the medical records. To determine a gold standard for gastrostomy placement in children <5 years old, a prospective randomized control trial should be undertaken. Other studies are needed to determine the cost of various gastrostomy placement techniques, including the costs of placement, replacement, and need for reinterventions.

Conclusion

PEGs had a higher major complication rate than LGs with or without fundoplication. Although operative time is longer in the laparoscopic group, LG seems to be the procedure of choice for children under age of five for enteral access. This would eliminate the requisite tube changes under anesthesia for low-profile tube replacement and the fluoroscopic interventions after the PEG.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Zamakhshary

, Jamal

, Blair

, Murphy

, Webber

, Skarsgard

. Laparoscopic vs percutaneous endoscopic gastrostomy tube insertion: A new pediatric gold standard?. J Pediatr Surg, 2005; 40:859–862.

Georgeson

. Laparoscopic gastrostomy and fundoplication. Pediatr Ann, 1993; 22:675–677.

McSweeney

, Jiang

, Deutsch

, Atmadja

, Lightdale

. Long-term outcomes of infants and children undergoing percutaneous endoscopy gastrostomy tube placement. J Pediatr Gastroenterol Nutr, 2013; 57:663–667.

McSweeney

, Kerr

, Jiang

, Lightdale

. Risk factors for complications in infants and children with percutaneous endoscopic gastrostomy tubes. J Pediatr, 2015; 166:1514–1519.

Brewster

, Weil

, Ladd

. Prospective determination of percutaneous endoscopic gastrostomy complication rates in children: Still a safe procedure. Surgery, 2012; 152:714–719; discussion 719–721.

Fox

, Abel

, Malas

, Duggan

, Leichtner

. Complications following percutaneous endoscopic gastrostomy and subsequent catheter replacement in children and young adults. Gastrointest Endosc, 1997; 45:64–71.

Vervloessem

, van Leersum

, Boer

, Hop

, Escher

, Madern

, de Ridder

, Bax

. Percutaneous endoscopic gastrostomy (PEG) in children is not a minor procedure: Risk factors for major complications. Semin Pediatr Surg, 2009; 18:93–97.

Sathesh-Kumar

, Rollins

, Cheslyn-Curtis

. General paediatric surgical provision of percutaneous endoscopic gastrostomy in a district general hospital—A 12–year experience. Ann R Coll Surg Engl, 2009; 91:404–409.

Jones

, La Hei

, Shun

. Laparoscopic gastrostomy: The preferred method of gastrostomy in children. Pediatr Surg Int, 2007; 23:1085–1089.

10.

Akay

, Capizzani

, Lee

, Drongowski

, Geiger

, Hirschl

, Mychaliska

. Gastrostomy tube placement in infants and children: Is there a preferred technique?. J Pediatr Surg, 2010; 45:1147–1152.

11.

Liu

, Jiwane

, Varjavandi

, Kennedy

, Henry

, Dilley

, Currie

, Adams

, Krishnan

. Comparison of percutaneous endoscopic, laparoscopic and open gastrostomy insertion in children. Pediatr Surg Int, 2013; 29:613–621.

12.

Aprahamian

, Morgan

, Harmon

, Georgeson

, Barnhart

. U-stitch laparoscopic gastrostomy technique has a low rate of complications and allows primary button placement: Experience with 461 pediatric procedures. J Laparoendosc Adv Surg Tech, 2006; 16:643–649.