Risk factors for complications and mortality of percutaneous endoscopic gastrostomy: a multicenter, retrospective study.

Background: Percutaneous endoscopic gastrostomy (PEG) is performed to provide nutrition to patients with swallowing difficulties. A multicenter study was conducted to evaluate the predictors of complications and mortality after PEG placement.
Methods: This study retrospectively analyzed patients who underwent initial PEG placement between January 2004 and December 2011 at seven tertiary hospitals in the Republic of Korea.
Results: All 1,625 patients underwent PEG placement by the pull-string method. The median age of the patients was 66 years, and 1,108 of the patients were men. The median follow-up period was 254 days. The common indications were stroke (31.6%) and malignancy (18.9%). The complication rate was 13.2%. The prophylactic use of antibiotics (odds ratio [OR], 0.58; 95% confidence interval [CI], 0.38-0.88; p = 0.010) reduced the PEG-related infection rate, but the actual usage rate was 81.1%. The use of anticoagulants (OR, 7.26; 95% CI, 2.23-23.68; p = 0.001) and the presence of diabetes mellitus (OR, 4.02; 95% CI, 1.49-10.87; p = 0.006) increased the risk of bleeding, but antiplatelet therapy did not. The procedural, 30-day, and overall mortality rates were 0.2, 2.4 and 14.0%, respectively. Serum albumin levels lower than 31.5 g/L (OR, 8.55; 95% CI, 3.11-23.45; p < 0.001) and C-reactive protein levels higher than 21.5 mg/L (OR, 3.01; 95% CI, 1.27-7.16; p = 0.012) increased the risk of 30-day mortality, and the patients who had both risk factors had a significantly shorter median survival time than those who did not (1,740 vs 3,181 days) (p < 0.001, log-rank).
Conclusions: The findings showed PEG to be a safe and feasible procedure, but the patient's nutritional and inflammatory status should be considered in predicting the outcomes of PEG placement.