168

N Hussain¹, AN Barkun^1,2, Y Toubouti², E Rahme², J-F Boivin²

Divisions of Gastroenterology¹ and Epidemiology², McGill University, Montréal, Québec

AIM: Few data exist on the relative safety of performing therapeutic endoscopy in patients on anti-platelet agents. To determine whether the intake of anti-platelet agents is associated with a significant risk of post-procedural hemorrhage.

METHODS: A case-control study was performed where cases unequivocally bled (ascertained using previously published criteria) as a result of an endoscopic polypectomy (P) of the upper or lower gastrointestinal tract, an endoscopic sphincterotomy (ES), or a percutaneous endoscopic gastrostomy (PEG) occurring in a tertiary care centre with experienced endoscopists. Two or 3 controls (patients in whom no bleeding occurred) were selected randomly for each case by consulting an administrative institutional database. Matching was done by gender, age ± 15 years, time of procedure ± 2 years, and procedure-type in patients with available clinical data. Conditional logistic regression modeling was used to assess the association between exposure to anti-platelet medication (within the prior 10 days) and post-procedural bleeding while adjusting for possible confounders. The study was powered to confidently rule out a clinically significant association.

RESULTS: Amongst the 44 cases, 50% bled during the procedure, while 46% bled in the first 14 days thereafter, and 4% bled later yet. 50% of those who bled required blood transfusions (mean: 1.5±1.8 units). The 132 controls and 44 cases were well matched for age (69±16 vs 68±17 yrs), gender (50% female in both), PTT (normal values in 91% vs 82%), and platelet count (over 60,000´10⁹/L in 100% in both). Differences were noted, amongst the ES group, in the prevalence of pre-procedural cholangitis (22% vs 48%), and an abnormal INR (over 1.5 in 2% vs 9%). Overall, 20% of controls and 23% of cases had been exposed to an anti-platelet agent. After adjusting for possible effects attributable to an elevated PTT, INR and the presence of cholangitis (for the ES patients), exposure to an anti-platelet agent in all patients was not associated with post-procedural bleeding (Risk Ratio = 1.04, 95% CI: 0.4–2.5).