Urgent colonoscopy is recommended during hospitalization for acute lower gastrointestinal bleeding (LGIB), but current guidelines still advocate large-volume polyethylene glycol (PEG) lavage (4–6 L), which is often poorly tolerated in elderly or frail patients. Evidence supporting low- or ultra–low-volume regimens in this setting is scarce. This study aimed to evaluate the feasibility, bowel cleanliness and diagnostic performance of an ultra–low-volume preparation in patients undergoing urgent colonoscopy for acute LGIB. We also assessed whether bowel preparation quality differed across demographic subgroups and analysed the impact of colonoscopy timing (≤24 h vs >24 h) on diagnostic yield.

We conducted a retrospective real-world cohort study including all consecutive patients who underwent urgent colonoscopy for acute LGIB at the Gastroenterology Unit of Santa Maria della Misericordia Hospital, Udine, from September 2024 to September 2025. The bowel preparation consisted of a rapid ultra–low-volume regimen (1-L PEG/ascorbate regimen). Demographics, comorbidities, haemoglobin, Shock Index, Oakland score, timing of colonoscopy, bowel cleanliness (Boston Bowel Preparation Scale, BBPS) and endoscopic outcomes were extracted from electronic records. Bowel preparation adequacy was defined as BBPS ≥6. Statistical analyses included χ²/Fisher’s exact test, Mann–Whitney U test, and multivariable logistic regression to identify predictors of bowel cleanliness and diagnostic yield.

Eighty-nine patients were included (median age 74 years; high comorbidity burden in 87.6%; antithrombotic therapy in 39.3%). Colonoscopy was performed <12 h in 29.2%, 12–24 h in 22.5%, and >24 h in 48.3% of cases.Bowel cleanliness was adequate in 56/85 patients (65.9%), with a median BBPS of 6. Toilettes were comparable across age groups, including patients >80 years (p=0.666), and across BMI categories (p=0.209). Comorbidities were associated with a higher rate of inadequate preparation (adequate 100% in patients without comorbidities vs 61% with comorbidities; Fisher p=0.014), although this association lost significance in multivariate analysis.Bowel cleanliness did not differ significantly between colonoscopies performed ≤24 h and >24 h (BBPS median 6 vs 6; p=0.346).The overall diagnostic yield was 68.5%, and endoscopic therapy was performed in 24.7%. In unadjusted analyses, procedures performed >24 h appeared more frequently positive; however, after adjustment for haemoglobin and clinical severity, colonoscopy ≤24 h was independently associated with higher diagnostic yield (OR ≈ 3.5; p=0.049). Bowel preparation quality did not significantly influence the ability to deliver endoscopic therapy (p=0.432).

Ultra–low-volume bowel preparation was feasible and generally effective in a real-world population of elderly and comorbid patients with acute LGIB, achieving adequate cleanliness in two-thirds of cases and maintaining stable performance across age and BMI subgroups. Importantly, early colonoscopy (≤24 h) improved diagnostic yield without compromising bowel cleanliness. Compared with standard high-volume lavage and conventional low-volume PEG regimens, the ultra–low-volume preparation demonstrated comparable feasibility and acceptable cleanliness in a frail, highly comorbid population, suggesting that reduced-volume strategies may offer a practical and better-tolerated alternative in the urgent LGIB setting without compromising diagnostic performance. Prospective comparative studies against standard and low-volume PEG regimens are warranted.