Medical treatment selection and outcomes for hospitalized patients with severe ulcerative colitis as defined by the Japanese criteria.

Background: Hospitalization for ulcerative colitis (UC) is potentially life-threatening. Severe disease in the Japanese criteria which modifies the Truelove-Witts' criteria might encompass more fulminant cases than the definition for acute severe UC. However, few studies have investigated the predictive factors for clinical remission (CR) after medical treatments for severe hospitalized patients by Japanese criteria.

Methods: Medical treatment selection, CR rates, and factors contributing to CR on day 14 were assessed in severe patients by Japanese criteria. We also investigated whether the reduction rate in patient-reported outcome 2 (PRO2) on day 3 could predict short-term prognosis.

Results: Eighty-five severe hospitalized patients were selected. Corticosteroids, tacrolimus, and infliximab were mainly selected as first-line treatments (76/85; 89.4%). The CR rates on day 14 were 26.8%, 21.4%, and 33.3% in patients receiving corticosteroids, tacrolimus, and infliximab, respectively. Extensive disease (odds ratio [OR] 0.022; 95% confidence interval [CI] 0.002-0.198), higher PRO2 (OR 0.306; 95% CI 0.144-0.821), and higher reduction rate in PRO2 on day 3 (OR 1.047; 95% CI 1.019-1.075) were independent factors predicting CR on day 14. If the cutoff value for the reduction rate in PRO2 on day 3 was 18.3%, sensitivity was 0.714 and specificity was 0.731 to predict CR on day 14. A higher reduction rate in PRO2 on day 3 (OR 0.922; 95% CI 0.853-0.995) was a negative factor to predict surgery within 28 days.

Conclusions: Tacrolimus and infliximab in addition to corticosteroids were used as first-line treatment in severe hospitalized patients. PRO2 on day 3 is a useful marker for switching to second-line therapy or colectomy.