Gastrostomy for enteral access - A comparison among placement by laparotomy, laparoscopy, and endoscopy

Background: Access to the stomach for long-term enteral feeding can be achi eved via laparotomy (open GT), laparoscopy (lap CT) or endoscopy (PEG). We compared the three methods of gastrostomy to determine whether any one has an advantage over the others. Methods: A retrospective analysis was done of 356 gastrostomies performed b etween January 1990 and June 1995. Results: Of these 356 gastrostomies, 214 were open GT, 60 were lap GT, and 82 were PEG. The completion rate was high, 98.1% to 100%. The perioperative mortality rates were low and similar among the 3 methods; 4.2% for open GT , 5.3% for lap GT, and 4.9% for PEG (p = 0.87, Chi square test). Cardiac ar rest was the predominant immediate cause of all perioperative deaths (68.8% ). Overall, none of the deaths was directly related to the gastrostomy proc edure. Major complications occurred in 24.9% of patients receiving open GT, in 18.3% of patients with lap GT, and in 17.1% of patients with PEG. Long- term complications developed in 25.9% of open GT, 25.6% of lap GT, and 30.4 % of PEG. The revision rates were similar for all 3 methods, 6.7% for open GT, 10% for lap GT, and 6.1% for PEG. Conclusions: Gastrostomy can be performed safely by all three techniques, w ith similar outcomes. PEG is our method of choice. Lap GT is preferred in p atients with head and neck carcinoma, patients with obstructing esophageal carcinoma, and patients who have problems with overlying liver or colon. Op en GT is reserved for cases with extensive intraabdominal adhesions or thos e where the procedure is done during an ongoing laparotomy.