P-023 Increased Risk of Opportunistic Infections in Ulcerative Colitis Patients Undergoing Hematopoietic Stem Cell Transplant

Abstract

Background: Previously we showed that ulcerative colitis (UC) patients who underwent hematopoietic stem cell transplant (HSCT) had higher inpatient mortality as compared to patients without UC (Am J Gastroenterol, 2013;108:S555-S555). In this study we sought to investigate whether the higher mortality in UC patients is linked to increased risk of opportunistic infections (OI's). Methods: We did a cross sectional study utilizing data from National Inpatient Sample (NIS) from the year 1993 to 2012. NIS is one of the largest all payer inpatient dataset in the United States. Patients with UC undergoing HSCT were identified by ICD9-CM codes 556x and 41.0x respectively. Discharge weights were used to generate nationally representative sample. Multivariate logistic regression was used to assess the impact of OI's on inpatient mortality controlling for confounders. To study the risk of OI's, we matched patients with UC to those without UC for the following criteria; age (65 years), sex, Charlson co-morbidity index (62), indication for HSCT, type of HSCT, source of stemcells, total body irradiation, and calendar year (62). Rao-scott chi-square test was used to compare the outcomes. The OI's we investigated included bacterial (tuberculosis, nocardiosis, clostridium difficile, pneumococcal, legionellosis, listeriosis, nontuberculous mycobacteria), viral (cytomegalovirus, invasive herpes simplex virus), parasitic (toxoplasmosis), and fungal (pneumocystis jiroveci, invasive fungal infections {IFI}). Results: For the years 1993 to 2012, NIS contained data on 50,330 patients who underwent HSCT (weighted N = 248,878) out of which 98 had UC (weighted N = 473). UC patients undergoing HSCT had higher risk of OI's compared to patients without UC (30.8% versus 13.6%, P < 0.0001). Among UC patients, the patients who developed OI's had higher risk of inpatient mortality compared to patients without OI's (32.6% versus 8%, P = 0.0001). In multivariate analysis, UC patients with OI's had higher odds of inpatient mortality (Odds ratio: 5.6, 95% CI; 1.2-25.7, P = 0.03) compared to UC patients without OI's. After applying matching criteria, 420 UC patients (out of 473) matched with 1277 patients without UC. In matched sample, UC patients continued to have higher risk of OI's (32.7% versus 15.7%, P = 0.0008). The higher risk of OI's in UC patients compared to patients without UC was consistent in both allogeneic (44% versus 22.3%, P = 0.009) and autologous transplants (23.6% versus 10.1%, P = 0.02). In allogeneic transplant, UC patients had higher risk of tuberculosis (2.4% versus 0%, P < 0.001) and cytomegalovirus infections (26.3% versus 2.6%, P < 0.0001) compared to patients without UC. In autologous transplant, UC patients had higher risk of IFI (11.6% versus 1.4%, P = 0.001) compared to patients without UC. There was no difference in risk of other OI's in UC patients as compared to patients without UC. In similar analysis, we found no difference in mortality and risk of OI's in patients with Crohn's disease (CD) compared to patients without CD who underwent HSCT. Conclusions: UC is associated with higher risk of certain OI's in patients undergoing HSCT. These OI's are associated with significantly higher odds of inpatient mortality in UC patients and may explain the increased mortality in UC patients undergoing HSCT