Intestinal Ultrasound, Steroid Failure and Colectomy Risk in Pediatric ACUS

May 5, 2026

Issue 09

Clinical Question

Can intestinal ultrasound predict steroid failure and colectomy risk in pediatric acute severe ulcerative colitis?

Editor’s Bottom Line

Intestinal ultrasound may be a useful and noninvasive bedside tool to help predict treatment failure and decide on rescue therapy among patients admitted for acute severe ulcerative colitis.

Reference

Scarallo L, Alvisi P, Bramuzzo M, et al. Intestinal Ultrasound Scan in Acute Severe Ulcerative Colitis in Children: A Multicenter Prospective Study on Behalf of the Porto Inflammatory Bowel Disease Working Group of European Society for Paediatric Gastroenterology Hepatology and Nutrition. Gastroenterology. Epub ahead of March 16, 2026; doi:10.1053/j.gastro.2026.02.034

Synopsis

The prospective multicentre ASUC-US study evaluated whether serial intestinal ultrasound (IUS) at the time of admission for acute severe ulcerative colitis and five to seven days later could predict short-term outcomes. The study enrolled 60 consecutive biologic-naïve children (61.7% girls; median age 13.5 years; median Pediatric Ulcerative Colitis Activity Index [PUCAI] of 70) with ASUC across 10 European centres between March 2020 and December 2024. Patients underwent IUS within 48 hours of starting intravenous corticosteroids and again 5 to 7 days later. Sonographers had a minimum of eight years of pediatric IUS experience and followed a prespecified quadrant-based protocol. IUS examiners were blinded to the patients’ clinical parameters and treatment decisions.

Key IUS parameters included colonic wall thickness (CWT), colonic wall stratification (CWS), and bowel wall vascularity, as assessed by power Doppler (Limberg’s score). The Milan Ultrasound Score (MUS) – calculated as [CWT×1.4] + [enhanced vascularization × 2] – was derived for each colonic quadrant.

The primary outcome was the need for second-line rescue therapy, while the secondary outcome was failure of medical therapy requiring colectomy within eight weeks.

Corticosteroid failure occurred in 65% of patients, all of whom received infliximab as rescue therapy. At the first IUS, nonresponders had significantly higher CWT in the left lower quadrant (LLQ) (6 vs. 4.2 mm; p <0.001) and left upper quadrant (LUQ) (5 vs. 4 mm; p=.003), and more frequent hypervascularity (Limberg’s score ≥3) in both quadrants. LLQ CWT >5 mm and LLQ MUS >7.8 were the optimal cutoffs for predicting steroid resistance, yielding 82.9% sensitivity and 71.4% specificity. Both LLQ CWT and MUS outperformed PUCAI at first IUS (p=0.036 and p=0.018, respectively) in discriminating steroid responders from nonresponders.

Ten patients (16.7%) required colectomy within eight weeks after failing to respond to all medical therapy. At the second IUS (between days 5 and 7), LLQ CWT >4.8 mm predicted medical therapy failure with 100% sensitivity and 66.7% specificity, and LLQ MUS >8.7 did so with the same sensitivity and specificity. Both parameters again outperformed PUCAI at the second timepoint (p=0.004 and p=0.001, respectively). Among patients who achieved steroid-free clinical remission at week 8, LLQ CWT and MUS on the second IUS were significantly lower than for those patients who did not achieve steroid-free clinical remission at week 8 (3.5 vs. 5 mm, p=0.037; 5.3 vs. 8.7, p <0.001).

Details

Study Design: Prospective blinded longitudinal observational
Funding: None reported
Allocation: Observational
Setting: Multicenter
Level of Evidence: 2b