Risk of bleeding after percutaneous endoscopic gastrostomy (PEG)

Abstract

Background: Patients who undergo percutaneous endoscopic gastrostomy (PEG) placement are often on anticoagulation and/or antiplatelet therapy with a potential thromboembolic risk if these medications are discontinued. Data on the safety of peri-procedural use of these drugs is limited.

Aims: To assess the risk and to identify any predictive factors for post-PEG bleeding, and to determine if clopidogrel increases the risk of bleeding following PEG.

Methods: A retrospective chart audit was conducted from January 1, 2002 to June 30, 2011.

Results: A total of 1,541 patients underwent PEG placement during this period. Gastrointestinal bleeding after PEG placement occurred in 51 cases (3.3%) and bleeding directly attributed to PEG was noted in six patients (0.4%). Multivariate logistic regression analysis of variables (age, gender, length of hospitalization, indication for PEG, antiplatelet or anticoagulant medications) showed that heparin infusion (P = 0.018) and length of hospitalization (P = 0.029) were statistically significant predictors of bleeding. The mean period for cessation and resumption of clopidogrel with PEG placement were 2.2 and 1.3 days, respectively.

Conclusion: Although PEG is classified as a high-risk endoscopic procedure, bleeding with PEG placement was rare, even with use of anticoagulation and antiplatelet medications. In selected patients on heparin infusion undergoing PEG, delaying the procedure, alternative use of low-molecular-weight heparin or close monitoring and frequent assessments should be considered. Clopidogrel did not contribute to an increase in bleeding risk, despite being held for a much shorter peri-procedural period as recommended by expert consensus.