Journals    
Present Issue 
Past Issues 
 

THESMANEWS   
Present Issue 
Past Issues 
 

SMJ Editorial Board 

Letters to the Editor

Instructions to Authors 
 

Singapore Med J 1998; Vol 39(12): 560 - 563

Experience of Percutaneous Endoscopic Gastrostomy at Massachusetts General Hospital - Indications and Complications
J B Y So, F W Ackroyd

INTRODUCTION

Enteral feeding is the preferred route to provide nutrients for patients with inadequate caloric intake but with a functionally intact gastrointestinal tract. Until recently in Singapore, the most common method of feeding for patients who were unable to swallow was through a nasogastric tube. However, nasogastric feeding is associated with well-documented complications(1,2). The nasogastric tube is least tolerated by patients. Local discomfort may cause frequent unintentional displacement of the tube especially in patients with impaired consciousness. This leads to discontinuation of nutrition, and more importantly, an increase risk of pulmonary aspirations.

Since its introduction by Gauderer and Ponsky in 1980(3), percutaneous endoscopic gastrostomy (PEG) has become widely accepted in the United States and United Kingdom, and is gradually becoming the procedure of choice for patients who require prolong enteral nutritional support(4). This technique is simple and can be performed under local anaesthesia. Several large studies have also confirmed the safety of this procedure(5,6). In contrast, traditional surgical gastrostomy commonly requires general anaesthesia, and carries significant morbidity and mortality(7).

METHODS

The method of PEG placement followed the principles as described by Gauderer and Ponsky(3). Two operators were required. The patient was placed in a supine position and the abdomen was cleaned and drapped. Simultaneously, a flexible gastroduodenoscope (GIF 100, Olympus, Lombard, Ill) was placed into the oesophagus. Occasionally, passage of Maloney dilators was required prior to the gastroduodenoscopy for patients with pharyngoesophageal strictures. The posterior pharynx was meticulously suctioned free of secretions under direct vision during every insertion and withdrawal of the endoscope. A diagnostic examination of the oesophagus, stomach, and duodenum was performed to exclude any significant lesions such as gastric outlet obstruction or disease of the anterior stomach wall. Next, the stomach was insufflated with air, and the anterior abdominal wall transilluminated. The site for the gastrostomy tube was selected by finger indentation on the stomach wall by the abdominal operator as viewed through the endoscope (Fig 1). We used the Ponsky Pull PEG kit (C.R. Bard, Inc., Billerica, MA). With local anaesthesia, a 1 cm skin incision was made on the proposed site. Through this incision, a 16-gauge angiocatheter was inserted into the stomach. The trocar was then removed and a wireloop was inserted into the stomach through the catheter. The wire was snared by the endoscopist (Fig 2). The endoscope and the wireloop were then removed through the mouth. A 20F mushroom-tipped catheter was tied to the end of the wireloop, and the catheter was withdrawn back down through the mouth, the stomach, and out through the abdominal wall. The endoscope was re-introduced into the stomach to confirm the position of the tip of the gastrostomy tube and to ensure haemostasis (Fig 3). The tube was then secured with an external bumper and four heavy silk sutures.

RESULTS

DISCUSSION

The results of the study indicate that PEG is a safe and easy method to provide assess for enteral nutrition. Although many patients had head and neck cancer in our series, which might have high oesophageal or posterior pharyngeal stricture, only one patient had failed intubation. The high success rate of this procedure is in agreement with previous reports, in which the failure rate was below 10%(4,8,9). Many of our patients had the procedure performed under local anaesthesia and as outpatients with a low complication rate.

When PEG was first described, it was mainly used for paediatric patients. Over the last decade, the procedure gradually gained acceptance and was applied in all age groups for various indications. Our technique of PEG placement follows the original principles described by Gauderer and Ponsky(3) which is known as the "Pull-string" method. Two other methods have been developed in the last decade. The second method involves pushing a feeding catheter into position over a guidewire from the mouth which is inserted transabdominally into the stomach (push-wire method)(10). Another method involves the use of an introducer(11). An introducer with a sheath is inserted percutaneously into the stomach over a guidewire. A balloon-tip catheter is then inserted through the sheath into the stomach. The balloon is inflated and this catheter serves as the gastrostomy tube. The "introducer" method prevents the contamination of oroesophageal bacteria or malignant cells. Although the last two methods have broad appeal, however published experience data are limited.

There is now strong evidence to suggest that PEG is the preferred method to provide enteral nutrition over surgical gastrostomy, or nasogastric tube. Several reports demonstrated reduced morbidity and mortality with PEG insertion, compared to surgical gastrostomy(12-14). A prospective randomised trial showed that PEG had an economical advantage over surgical gastrostomy, though the difference in operative complications was not obvious(15). Two randomised studies have been performed to compare nasogastric tube feeding with PEG for patients with dysphagic stroke(16-17). In the study by Park et al, enteral feeding was provided for patients who had neurological dysphagia for at least 4 weeks by either PEG or nasogastric tube(16). Gastrostomy patients received a significantly greater proportion of their prescribed feed than nasogastric group, and had more weight gain after seven days of feeding. Another randomised study confirmed the nutrition benefits of PEG over nasogastric tube feeding(17). The study demonstrated that patients with nasogastric feeding had a higher mortality rate due to increased risk of pulmonary aspirations. Several studies have shown that patients on nasogastric tube feeding frequently removed the tube subconsciously and required resiting(1,16-18). The tube displacement and exchange proportionally increase the risk of aspiration. In addition, the frequent discontinuation of tube feeding adversely affects the nutritional benefits of enteral feeding(2,16-17).

Several large series documented that the 30-day mortality and morbidity of PEG feeding to be around 1% and 10% respectively(5,6,12,19). The early mortality rate of this procedure is closely associated with the underlying disease process(9). We only perform PEG for patients who are expected to survive and require enteral feeding for at least four weeks. Common complications of PEG include wound infections, peritonitis, gastric haemorrhage and pulmonary aspiration. Several possible reasons account for the low complication rate (2%) in our series. Prophylactic antibiotics were routinely used in our patients, which can reduce the risk of wound infections(20). Peritonitis is an avoidable but dangerous complication which occurs in 0% to 1.2% of PEG patients(5,6,10,19). Premature removal of the catheter, leakage of gastric contents into the peritoneal cavity due to either insufficient apposition of the button to the stomach wall or pressure necrosis of the wall due to excessive pressure were the common reasons for peritonitis(5,10,19). To avoid these problems, patients or their relatives were carefully educated about the proper care of the catheter. The tube was meticulously secured on the abdominal wall with strong silk ties and tapes. In addition, repeat endoscopy was performed when the catheter was secured to ensure that correct pressure was applied by the button on the stomach wall. Occasionally, traction of the button was applied momentarily to cease any bleeding from the mucosal surface of the gastrostomy. This manoeuvre reduced the possibility of gastrointestinal haemorrhage after the procedure.

One patient in our series developed pulmonary aspiration five days after PEG feeding. Aspirations can occur during PEG placement or any time after feeding starts. Mild sedation, use of local anaesthesia, and short operative time (usually less than 20 minutes) of PEG probably accounts for the low incidence of procedure-related aspiration(5). However, careful precautions are still required to prevent this complication during tube feeding in chronic care settings(9).

PEG has been employed to provide enteral nutrition in a wide range of patient population. Previous prospective studies have shown the use of PEG in patients with neurological deficit, cystic fibrosis, and in elderly(16,17,21,22). Our earlier study demonstrated an excellent result of PEG in patients with head and neck cancer before or after the resectional procedure(8). Apart from providing nutrition, PEG has also been utilised for decompression of gastrointestinal obstruction due to carcinomatosis, and the results were promising(23).

CONCLUSION

In summary, our experience is in agreement with previous reports, in which PEG is a simple and safe procedure. Many of the procedures can be done under local anaesthesia on an outpatient basis. PEG should be the method of choice for patients who require long-term enteral nutrition.

J B Y So, FRCSEd, FRCS (Glas),
FCSHK
Clinical Fellow