P384. Predictive factors of response to prolonged intravenous corticosteroids in patients with acute severe ulcerative colitis
D.J. Lee, M.E. Song, S.J. Park, S.P. Hong, T.I. Kim, W.H. Kim, J.H. Cheon, Yonsei University College of Medicine, Department of Internal Medicine and Institute of Gastroenterology, Seoul, South Korea
Severe ulcerative colitis (UC) can still be a life threatening condition despite intensive medical management and early colectomy. Intravenous corticosteroid therapy is used as a bridge therapy for acute severe UC, but it's appropriate duration of administration and timing of response assessment has not yet been defined. The objective of this study was to evaluate the clinical predictive factors after prolonged intravenous steroid therapy in acute severe patients with UC.
We retrospectively evaluated a total of 173 cases among 131 patients, who were hospitalized and received intravenous corticosteroid therapy for acute exacerbations of severe UC between January 2005 and September 2012 from at Severance Hospital, Seoul, South Korea. Patients who achieved remission or had treatment failure were excluded from treatment response assessment at 5 days after intravenous steroid therapy. The remaining 130 case with equivocal response, who were neither early-responders nor non-responders, were classified into late responders (who finally obtained remission; 115 cases) and non-responders (who ultimately underwent rescue therapy; 15 cases). We compared between the 2 groups on the basis of 7 days for the determination of the response assessment time.
On univariate analysis, there was a significant negative correlation of late-responders with initial modified Truelove score, initial Mayo score, initial partial Mayo score, modified Truelove score at 5 days, and partial Mayo score at 5 days. Previous attack count and platelet counts were negatively correlated on the basis of 5 days. On the other hand, delta of partial Mayo score (partial Mayo score at 5 days − initial partial Mayo score), hemoglobin and albumin levels showed a positive correlation at 5 days, hemoglobin level was positively correlated at the time of 7 days.
A multivariable analysis showed that partial Mayo score at 5 days [OR 0.085 (0.013–0.539), p = 0.009], delta of partial Mayo score [OR 5.595 (1.583–19.771), p = 0.008], previous attack count [OR 0.181 (0.039–0.850), p = 0.030] and hemoglobin level [OR 4.227 (1.109–16.112), p = 0.035] were independent predictive values at 5 days. At 7 days, partial Mayo score at 7 days [OR 0.080 (0.012–0.525), p = 0.009], hemoglobin [OR 6.271 (1.437–27.366), p = 0.015] were significantly associated with response to prolonged steroid response.
Partial Mayo score, delta of partial Mayo score, previous attack count and hemoglobin level were the meaningful independent parameters among the predictors for prolonged intravenous steroid treatment response. These factors may help physicians determine whether to continue the steroids for equivocal responders.