Introduction: Post-polypectomy bleeding (PPB) may warrant transfusions, intensive care (ICU) monitoring, and surgery. Studies have suggested that the highest risk is with large and right-sided polyps, coagulopathies, or platelet dysfunction. The role of ICU monitoring, however, has not been established. Aim: To seek factors predictive of significant PPB, and to assess the role of ICU monitoring. Methods: Patients with post-polypectomy bleeding were identified from our prospective GI Bleeding Team Database. Information collected included age, sex, comorbidity, prior PPB, medications (ASA, NSAID, coumadin, or corticosteroids within 3 days; heparin or persantine within 1 day of presentation), polyp features/location, and polypectomy technique. Bleeding severity was assessed by hemodynamic instability (normal, tachycardic, orthostatic, hypotensive), transfusion, and ICU monitoring. The outcomes measured were control of bleeding, transfusions, recurrent bleeding (>1g/dL/day hemoglobin drop, new onset hematochezia), duration of stay, and death. Results: From 4/4/1989-11/ 22/1996, 14,575 colonoscopies with polypectomy were performed. During this time, 84 patients (57 men, 27 women) were diagnosed with PPB. The median age was 72 years (range 18-88). Major comorbidity included cardiovascular (81%), hematological (8.3%), and renal (6%) diseases. Bleeding occurred up to 17 days after polypectomy (median 5.5 days). 32.1% had taken ASA, 10.7% NSAID, 11.9% coumadin, 13.1% corticosteroids, 6% IV heparin, 7.1% SC heparin, and 7.1% persantine. 56% had normal vital signs, 8.3% were tachycardic, and 35.7% orthostatic at the time of presentation. Sessile cecal polyps >10 mm in size snared in toto or piecemeal without submucosal saline injection were more likely the bleeding source. The median hospital stay was 3 days. 43 patients were admitted to the ICU (median stay 2 days) and were more likely to require RBC transfusions (74.4%) vs. medical floor (31.7%). The median number of units transfused was 4 in both groups. 95.2% of post-polypectomy bleeds were managed endoscopically. 8 patients rebled after a median of 5 days after admission (5 medical floor, 3 ICU). Bleeding was trivial in 3 patients. Repeat endoscopic therapy was successful in 2 out of 5 cases. 3 patients continued to have significant bleeding requiring embolization (1), or hemicolectomy (2). Summary: A larger percentage of the bleeding sites corresponded to sessile cecal polyps > 10 mm in size. Medications and comorbid conditions were associated with an increased risk of bleeding. The outcomes measured were not significantly different between the patients admitted to the ICU or the medical floor. Conclusion: Post-polypectomy hemorrhage remains an uncommon but significant complication of colonoscopic polypectomy. Polyp features and location, polypectomy technique, medications, and comorbid conditions may increase the risk of bleeding. Unless indicated because of severe comorbid illness, intensive care monitoring does not appear to affect the outcome.
|Original language||English (US)|
|State||Published - Dec 1 1998|
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging