Mesenteric angiography for colonic diverticular bleeding: clinical and technical predictors of extravasation and post-procedural outcomes.

objectivesTo predict active extravasation during angiography for colonic diverticular bleeding confirmed by CT angiogram (CTA) and examine the effect of procedural factors on clinical outcomes. MATERIALS AND

methodsMesenteric angiograms performed at three hospitals for colonic diverticular bleeding on CTA between 2016 and 2025 were retrospectively reviewed. Data collection included CTA-to-angiogram time, extravasation and inferior vena cava size, angiogram selectivity, provocative maneuvers, and embolization techniques. Univariate and multivariate analyses were used to find associations between pre-procedural variables and active extravasation on angiography. Clinical outcomes were compared between patients who were or were not embolized.

resultsOne hundred seventeen patients (median age, 76 years (IQR 16), 42 women) underwent 146 angiograms. Active extravasation was found in 40% of angiograms. CTA-to-angiogram time ≤ 4 h (OR = 2.95; CI: 1.17-7.73; p = 0.02), inferior vena cava short axis ≥ 20 mm (OR = 3.63; CI: 1.24-11.6, p = 0.02) and age (OR = 1.06; CI: 1.01-1.12; p = 0.01) were independent predictors for active bleeding on angiography in a multivariate logistic model (AUC: 0.81, CI: 0.72-0.89; p < 0.01). Angiography beyond the named arterial branch revealed more bleeding than main trunk angiography (34% vs 21%). 51/58 (88%) positive and 15 negative angiograms were embolized. Patients with targeted embolization had less rebleeding (10% vs 44%, p < 0.05) and post-procedure colonoscopies (17% vs 51%, p < 0.05).

conclusionMesenteric angiography had a 40% positivity rate for diverticular bleeding after CTA, with a higher yield if performed within 4 h, sub-selectively, and after resuscitation. Targeted embolization decreased rebleeding and post-procedure colonoscopies. KEY POINTS: Question Colonic diverticular bleeding identified on CTA is treated with angiography and embolization, but diagnostic yield and embolization outcome are extrapolated from all-cause lower gastrointestinal bleeding. Findings Extravasation was identified in 40% of angiograms, with better yield if performed ≤ 4 h after positive CTA and sub-selectively. Targeted embolization reduced rebleeding and post-procedural colonoscopies. Clinical relevance Patients with colonic diverticular bleeding with active extravasation on CTA should be vigorously resuscitated and undergo prompt mesenteric angiography. This increases the likelihood of visualizing active extravasation and targeted embolization, leading to less rebleeding and further procedures.