SESSION TITLE: Tuberculosis SESSION TYPE: Original Investigation Poster PRESENTED ON: Wednesday, October 26, 2016 at 01:30 PM - 02:30 PM PURPOSE: To evaluate the association between incident and prevalent diabetes mellitus (DM) and clinical presentation, treatment failure, and mortality in adults with active tuberculosis (TB) in Canada. METHODS: A retrospective chart review of all adult patients treated for TB in two large Canadian cities (Calgary, AB and Winnipeg, MB) between 2007-2012 was performed. The primary outcome was all cause mortality during TB treatment. Patients aged ≥18 years with a clinical or microbiologic diagnosis of mycobacterium tuberculosis (pulmonary and/or extrapulmonary) with adequate chart data were included. Determination of DM prevalence at time of TB diagnosis was based on chart diagnosis of DM or documented use of insulin or oral hypoglycemics. An incident diagnosis of DM was defined as new diagnosis within six months of TB diagnosis. Secondary outcomes included: treatment failure (death, relapse, or culture positivity at 4 months), presence of cavitary disease, smear and culture positivity, and presence of drug resistance. Unadjusted odds ratios were used to estimate the association between DM and mortality and treatment failure. RESULTS: 788 patients treated for TB based on a clinical or microbiologic diagnosis were identified. Pre-existing DM was found in 131 patients (17%) and an incident diagnosis of DM was made in 19 patients (2%). The average hemoglobin A1c was 8.7% and 9.4% in patients with pre-existing and new DM, respectively. HIV co-infection was found in zero patients with DM and 41 patients without DM (7%). Compared to patients without a diagnosis of DM, mortality during TB treatment was higher in patients with pre-existing DM (OR 2.62, 95% CI 1.37-5.02), but not in patients with an incident diagnosis of DM (OR 1.13, 95% CI 0.14-8.72). Patients with pre-existing DM were also more likely to have treatment failure (OR 3.02, 95% CI 1.64- 5.56) compared to those without DM. Both death and relapse contributed to treatment failure in pre-exisitng DM (relapse OR 6.49, 95% CI 1.07-39.19). There were no documented culture positive cases at four months in either group. Patients with DM (incident or prevalent) were more likely to have smear positive disease (44% vs. 34%) and cavitary disease (37% vs. 20%) compared to patients without DM. Drug resistance rates between DM and non DM patients were similar (11% versus 10%, respectively). Ethnicity varied by geographic region, with 326 (91%) of TB cases in Calgary, AB occurring in foreign born patients, versus 229 (53%) of TB cases in Winnipeg, MB occurring in Canadian born Aboriginals. Patterns in clinical presentation, treatment failure, and mortality in patients with DM versus without DM were similar between geographic sites. CONCLUSIONS: In our cohort of active TB cases, patients with pre-existing DM had an increased risk of mortality and treatment failure. Consistent with the finding of worse treatment outcomes, patients with pre-existing DM (and incident DM) had more extensive disease at presentation. CLINICAL IMPLICATIONS: Recognition of pre-existing DM in patients with TB is important as it identifies a group of patients more likely to have extensive disease and worse treatment outcomes. Further research is required to determine if these patients may benefit from more intensive TB and DM management. DISCLOSURE: The following authors have nothing to disclose: Leila Barss, Evan Orlikow, Sen Phang, Natasha Sabur, Michael Arget, Julie Jarand, Stephen Field, Martha Ainslie, Dina Fisher No Product/Research Disclosure Information