mean maternal age (33.2 years). Based on the limited available data, no specific risks have been identified with UST exposure during pregnancy. REFERENCES: 1 Martin PL, Sachs C, Imai N, Tsusaki H, et al. Birth Defects Res B Dev Reprod Toxicol. 2010;89(5):351-363. 2 Hoffman MB, Farhangian M, Feldman SR. Expert Rev Clin Immunol. 2015;11(6):709-720. 3 Miscarriage. http://americanpregnancy.org/pregnancy-complications/miscarriage/. Accessed 2017. 4. US Department of Health and Human Services. Reviewer Guidance: Evaluating the Risks of Drug Exposure in Human Pregnancies. April 2005. http://www.fda.gov/downloads/ Drugs/Guidances/ucm071645.pdf. Su1800 EOSINOPHILIC-PREDOMINANT COLONIC INFILTRATION CAN PREDICT STEROID-DEPENDANCY AND DISEASE COURSE IN ULCERATIVE COLITIS Aaron Mendelson, Katie A. Falloon, Maryam Tajamal, Raphael Rothenberger, Reezwana Chowdhury, Mark Lazarev, Brindusa Truta, Sharon Dudley-Brown, Florin Selaru, Joanna Melia, Alyssa M. Parian Background: Colonic eosinophils are involved in ulcerative colitis (UC) pathogenesis, but it is unclear if they predict clinically-relevant outcomes . Objective: To evaluate established UC patients with prominent eosinophils on colonic biopsies, and to compare their clinical course to UC patients without eosinophilic predominance. Methods: This is a retrospective study of adult UC patients with colonoscopic biopsies from 2003-2016. Patients with eosinophilic predominance, defined as colonic biopsies with qualitatively significant eosino- philia as determined by a surgical pathologist, were identified through our Pathology Dataset and were compared to a control group of UC patients without eosinophilic predominance. The groups were matched for age, disease duration, and Mayo endoscopic subscore. Out- comes between groups were compared with student t-test for continuous variables and X 2 - test for categorical variables. Patients with eosinophilic colitis, positive ova and parasites study within 90 days of biopsy, steroid use for non-IBD indications, HIV with CD4 count<200 or an AIDs-defining illness, or colectomy before biopsy were excluded. Included patients were followed for 1 year. Results: 54 eosinophilic-predominant patients and 53 controls were included. Patients with colonic eosinophilic predominance were more likely to have been on 5-ASA [46(85.19%) vs 33(62.26%), p=0.01] and systemic steroids [17(31.48%) vs 7(13.21%), p=0.02] within 90 days of biopsy, but were similar with regards to 6-MP/AZA, TNF alpha agents, or anti-integrin inhibitors exposure. Those with eosinophilic predominance had more frequent hospitalizations [14(25.45%) vs 5(9.43%), p=0.03] and were more likely to be steroid dependent, defined as an inability to taper steroids without clinical relapse or steroid use for >3 months [7(12.96% vs 0(0%), p=0.007]. There was a trend towards significance for more frequent colectomy rates in those with eosinophilic predominance [7(12.96%) vs 2(3.77%), p=0.09]. In the subset of patients on immunomodulators and/or biologics, eosinophilic predominant patients were still more likely to be steroid dependent [5(21.74%) vs 0(0%), p=0.03] and to have more frequent hospitalizations [7(30.43%) vs 0 (0%), p=0.01]. Within this subset, eosinophilic-predominant patients were also more likely to have undergone colectomy [5(21.74%) vs 0(0%), p=0.03] and to have required topical steroids [9(39.13%) vs 1 (5.56%), p=0.01]. Conclusion: When followed for 1 year, patients with eosinophilic-predominant colon biopsies had higher rates of steroid dependency and hospitalization. This remained true even in those previously on immunomodulators and/or biologics. Further investigation is needed to better understand the disease course of eosino- philic-predominant patients and whether therapy can be tailored to them. S-589 AGA Abstracts Su1801 DIFFERENT HISTOLOGICAL INDEXES FOR THE ASSESSMENT OF ULCERATIVE COLITIS ACTIVITY Fernando Magro, Joanne D. Lopes, Paula Borralho, Susana Lopes, Rosa Coelho, José Cotter, Francisca Dias de Castro, Helena T. Sousa, Paula M. Lago, Marta Salgado, Patricia Andrade, Ana Vieira, Pedro Figueiredo, Paulo Caldeira, A Sousa, Maria A Duarte, Filipa Ávila, João Pereira da Silva, Joana Moleiro, Sofia Mendes, Silvia Giestas, Paula Ministro, Paula Sousa, Raquel Gonçalves, Bruno M. Gonçalves, Ana Maria Oliveira, Isadora Rosa, Marta Rodrigues, Cristina Chagas, Joana Torres, Camila Dias, Joana Afonso, Karel Geboes, Fatima Carneiro Background and Aims Histopathological analyses have become an important part of Ulcera- tive Colitis (UC) assessment both in clinical practice and in research trials. Our work is focused on three histological indices that have been at least partially validated - Geboes score (GS), Nancy index (NI) and Robarts histopathology index (RHI). It aims to evaluate the agreement between these three histological classification systems and to relate these scoring systems with the patients' endoscopic outcomes and inflammation burden. Moreover, intends to disclose which (if any) histological features could be predicted from fecal calprotec- tin(FC) levels. Methods Biopsy samples from 377 UC patients were blindly evaluated using GS, NI and RHI. The results were compared with the patients' Mayo endoscopic score and FC levels. Results The concordance between GS, NI and RHI, evaluated in terms of histological remission/activity, was considerably high. In fact, using the GS with a 3.1 cut-off as the reference, 87 and 71% of the patients with histological activity and 89 and 100% of the patients in histological remission, were correctly identified by the NI and RHI, respectively. Regarding endoscopic outcomes, the sensitivity is higher when 1 is used as the Mayo endoscopic score cut-off (96 to 100% of all patients with macroscopic lesions have histological inflammation), and the specificity is higher when o is used as the Mayo endoscopic score cut-off (56 to 93% of all patients considered to be in mucosal remission have no histological activity). Patients that have an FC level below 150 ug/g are in histological remission according to NI, RHI and GS (with a negative predictive value (NPV) of 81, 93 and 88%, respectively). Moreover, the presence of neutrophils in the lamina propria did not associate with FC levels. Conversely, higher FC levels were statistically associated with the presence of neutrophils in the epithelium, as well as with ulceration or erosion of the intestinal mucosa. Conclusion The classical GS is accurate concerning the newly-developed and formally-validated NI and RHI scores. Our results suggest that one can spare a patient from an endoscopic examination based solely on the non-invasive and cheaper FC test. Additionally, FC levels are increased when neutrophils are present in the epithelium and the intestinal mucosa has erosions or ulcers. Su1802 HISTOLOGIC DISEASE ACTIVITY DID NOT PREDICT SYMPTOMS AMONG ULCERATIVE COLITIS PATIENTS IN MUCOSAL REMISSION Kelly C. Cushing, William Tan, Jay Luther, Hamed Khalili, Francis Colizzo, John J. Garber, David H. Alpers, Ramnik Xavier, Ashwin Ananthakrishnan Background: Mucosal remission, defined as an absence of endoscopic disease activity, is associated with better short- and long-term outcomes. However, whether histologic activity in the setting of mucosal remission is associated with patient-reported symptoms is unknown. Aim: To assess the relationship between histologic activity and patient-reported symptoms among patients with ulcerative colitis (UC) in mucosal remission . Methods: The study population included patients enrolled in a prospective registry who had established UC and underwent a lower endoscopy with availability of concurrent histologic as well as clinical disease activity. Endoscopic remission was defined as a modified Baron's score of 0. Clinical disease activity was determined by the simple clinical colitis activity index (SCCAI), remission <2 and symptomatic $ 2. Maximum histologic activity was assessed by retrospective chart review. Histologic activity was defined as: active (cryptitis, crypt abscesses, erosions/ulcera- tion), quiescent (features of chronicity without active inflammation) or normal (no features of chronicity). Comorbidities including anxiety and depression were assessed using PROMIS questionnaires. Students t tests or Wilcoxon rank sum tests were used for continuous variables and chi-square or Fishers exact tests for categorical variables, as appropriate. Binary logistic regression was used to assess for histologic activity as a significant predictor of symptoms as defined by the SCCAI. Results: There were 127 patients included in the study. Of these, 93 (73.2%) were in clinical remission defined as SCCAI <2 and 34 (26.8%) were symptomatic. Those who were symptomatic were older (52 vs 40 years, p =0.03) with increased anxiety (12 vs 9, p =0.02), increased depression (10.5 vs 8, p =0.03) and higher anti-TNF use (44.1% vs 23.7%, p= 0.04). Race, gender, disease duration, disease extent, AGA Abstracts