Percutaneous endoscopic gastrostomy tube placement in patients with malignant bowel obstruction due to ovarian carcinoma.

OBJECTIVES: To analyze the feasibility of using percutaneous endoscopic gastrostomy (PEG) tube placement in ovarian cancer patients with malignant bowel obstruction and to analyze the outcome of these patients.
METHODS: We performed a retrospective review of all patients with ovarian carcinoma who underwent PEG tube placement between 1995 and 2002 at our institution. Abstracted data included patient demographics, procedure information, symptom resolution, diet tolerated, complications, further treatment, and survival.
RESULTS: Ninety-four patients with ovarian carcinoma requiring PEG tube placement for malignant bowel obstruction were identified. The mean age at the time of PEG tube placement was 56 years. The mean interval from the initial cancer diagnosis to the placement of the PEG tube was 3.1 years. Twenty-two of 77 patients who had a computed tomography (CT) scan prior to PEG tube placement had tumor encasing the stomach. Fifty-nine (63%) of 94 patients had ascites, 25 of whom underwent a pre-PEG paracentesis (mean, 2845 ml). Ninety-four patients had a successful PEG tube placement under conscious sedation by the gastroenterology service (92) and/or by interventional radiology (2). Symptomatic relief, defined as no nausea or vomiting within 7 days of PEG tube placement, was noted in 86 (91%) of 94 patients undergoing PEG tube placement. Diets tolerated with and without the PEG tube being clamped were as follows: none, 3; sips, 9; liquids, 40; soft/regular, 40; and unknown, 2. The mean hospital stay after the procedure was 6 days. Eighteen patients had one or more of the following complications: leakage, 8; peristomal infection, 3; obstruction, 3; PEG tube migration, 2; catheter malfunction, 2; hemorrhage, 2; and peritonitis, 1. Nine patients required PEG tube revision due to complications. Chemotherapy after PEG tube placement was administered in 29 (31%) of the 94 patients, with resolution of obstruction and removal of the PEG tube in 4. In addition, 14 (15%) received limited total parental nutrition (TPN) after PEG tube insertion. Seventy-five (85%) of 88 patients died at home or under hospice care. The median overall survival for the 94 patients undergoing PEG tube placement was 8 weeks (95% CI, 6-10). Multivariate survival analysis revealed the presence of liver metastases (P < 0.001) and older age (P = 0.01) to be statistically significant predictors of shorter survival. The use of TPN after PEG tube placement was not a statistically significant prognostic factor in this model (P = 0.09).
CONCLUSIONS: PEG tube placement in ovarian carcinoma is technically feasible and safe in the palliative setting. In addition, PEG tube placement allowed the majority of patients to have end-of-life care either at home or in an inpatient hospice. For the total population, no benefit was seen regarding survival with the use of TPN in this setting. Selected patients (younger age and without liver metastasis) may benefit from chemotherapy after PEG tube placement.