Incidence and predictors of gastrointestinal hemorrhage following mesenteric revascularization.

Abstract

Objective: Postoperative gastrointestinal hemorrhage (GIH) following mesenteric revascularization when performed either openly (OR) or endovascularly (ER) has been clinically observed but not reported. The aim of the study is to assess the incidence and predictors of GIH in patients undergoing mesenteric revascularization.

Methods: This was a single-center retrospective review of consecutive patients treated with open or endovascular mesenteric revascularization from 2009 to 2019. Patients with nonocclusive mesenteric ischemia, intraoperative or perioperative death within 24 hours, or no postoperative follow-up were excluded. Primary endpoints were incidence and predictors of clinically significant GIH (CS-GIH) within 30 and 60 days postoperatively. CS-GIH was defined if patients required red blood cell transfusion, hospital readmission, escalation to intensive care, prolonged discontinuation of anticoagulation, or need for endoscopy/colonoscopy.

Results: A total of 260 patients presented with mesenteric ischemia and underwent OR/ER. Two hundred five patients met inclusion criteria (139 female [68%]; mean age of 69.9 years [range, 18-92 years]). Presentation was chronic mesenteric ischemia in 128 patients (62%), acute-on-chronic in 45 (22%) and acute in 32 (16%). Ninety-three patients (45%) underwent OR, 93 (45%) ER, and 19 (9%) hybrid. Fifty patients (24%) presented with GIH, 44 (21%) within 30 days of OR/ER, at a median time of 6.5 days postoperatively. CS-GIH occurred in 37 patients (18%), which led to death in two patients (1%), prolongation of intensive care unit course or transfer to intensive care unit in 28 patients (14%), red blood cell transfusion in 21 (10%), diagnostic/therapeutic endoscopy/colonoscopy in 18 (9%), and hospital readmission in 14 patients (7%). Endoscopy/colonoscopy was diagnostic in nine patients (ulcer in five patients, angioectasia in two, and anastomotic bleeding or colonic necrosis in one each), therapeutic in four, and identifying one patient with diffuse bleeding requiring operative intervention. Factors associated with increased risk of CS-GIH were bowel resection during index hospitalization (odds ratio [OR], 11.29; P < .001), acute presentation (OR, 5.42; P < .001), atrial fibrillation (OR, 3.01; P = .004), first-time initiation of antiplatelet therapy (OR, 2.61; P = .01), and treatment with stenting (OR, 2.31; P = .03).

Conclusions: Patients undergoing mesenteric revascularization are at high risk for postoperative GIH, which increases morbidity and hospitalization resources in nearly 20% of patients. Specific patient groups are at high risk for CS-GI hemorrhage. Postoperative care pathways should consider these risk factors to reduce CS-GIH after mesenteric revascularization to improve outcomes.