A new screening tool could help clinicians predict acute severe colitis patients’ steroid response at hospital admission, eliminating the three-day wait to actual outcomes and expediting access to effective care, gastroenterologists say.

The four-point survey, assessed and validated in 289 patients across the UK, Australia and India, gauges patients’ risk of steroid failure based on C-reactive protein (CRP) and albumin levels, and Ulcerative Colitis Endoscopic Index of Severity (UCEIS) scores.

Patients receive a point for every value ≥ 100 mg/L (CRP), ≤ 25 g/L (albumin), and ≥ 4 and ≥ 7 (UCEIS scores), with those scoring zero most likely to benefit from steroids, and four, least.

“For over 20 years, the accepted management of acute severe colitis has involved 3 days’ treatment with intravenous corticosteroids, followed by use of accepted protocols to assess response,” the authors wrote in Gut (link here).

However, this new tool — the Admission Model for Intensification Therapy in Acute Severe Colitis (ADMIT-ASC) — uses routine blood tests and endoscopic appearances to predict steroid response on admission, prompting early escalation to second-line therapy or surgery while “avoiding potentially futile steroid exposure” and reducing hospitalisation duration for some patients, they said.

It’s the “first replicated prognostic score which does not rely on waiting to assess steroid response after 3 days” and “the only score derived and validated in patients exposed to biologics”, they added.

Tested and validated

The authors tested several models with subsets of age, albumin, CRP, current biological treatment, disease duration, haemoglobin, platelets, sex, stool frequency and UCEIS, with both even and weighted valuation of each parameter in 117 patients admitted to the Translational Gastroenterology Unit, Oxford, UK between 2015–2019 before presenting the final, best-performing model.

It predicted steroid non-response with CRP, albumin and UCEIS score false discovery rates of P = 0.00066, 0.0066 and 0.015.

When validated in 110 patients admitted to the Gold Coast University Hospital, Queensland, Australia in 2015–2020 and 62 to All India Institute of Medical Sciences, New Delhi, India in 2018–2020, those scoring zero, one, two, three and four had steroid response rates of 100%, 75.0%, 54.9%, 18.2% and 0%, respectively.

“Although scoring four was 100% predictive of steroid non-response, only 13 patients (4.3%) scored four,” the authors wrote.

As a result, they recommend applying a treatment escalation threshold of three and above.

This “identified 17.5% of the combined validation cohorts and was highly specific (0.96, 95% CI: 0.93 to 1.00), with a [positive predictive value] of 0.84 (0.70 to 0.98) and [odds ratio] of 11.9 (10.8 to 13.0)”.

“Use of this threshold to advance treatment would prevent one delayed treatment for every 8.2 (95% CI: 5.8 to 13.7) patients assessed,” they wrote.

For biological-naïve patients, a score ≥ 3 “is an indication for early medical rescue therapy”; for those already on biologicals, it “mandates surgical discussion early in admission”.

Meanwhile, patients scoring two or less can be managed with steroids and reassessed for response and need for treatment escalation on day three, they suggested.

“A score of 0 may be particularly reassuring since we found complete concordance with steroid response across the cohorts,” they said.

Better insights

The recommendations came with the caveat that, despite being validated in two cohorts with different genetic backgrounds and environmental exposures, the thresholds may not be appropriate in all settings internationally and should be determined based on perceived implications and costs of avoidable rescue therapy versus risk of delayed treatment.

Still, unlike the “only recent comparable index” ACE (Albumin, CRP and Endoscopy), ADMIT-ASC was developed using data generated during wide biological use for UC and tested across multiple cohorts, potentially offering better insights into steroid response.

Where UCEIS is unavailable on admission, haemoglobin levels may be used instead, with one point for values ≤ 120 g/L. This version showed response rates of 75.6%, 53.7%, 42.1% and 20% for zero, one, two and three scores, respectively, the authors noted.

Both versions of ADMIT-ASC are “simple and implementable without other changes in practice; all parameters are readily and reproducibly defined, without the awkward need for assessment of stool frequency”, they wrote.

Despite its limitations, particularly in predicting steroid response for patients scoring one or two, ADMIT-ASC could help expedite effective care delivery on admission and improve acute severe colitis patient outcomes, they concluded.