Tuberculosis and diabetes in Nigerian patients with and without HIV.

BACKGROUND: Type 2 diabetes mellitus (DM) and HIV increase the risk of tuberculosis (TB). The frequency of DM among patients with TB with and without HIV is poorly documented in many low- and middle-income countries.

METHODS: This was a cross-sectional hospital-based study performed in Abuja, Nigeria. Adults with presumptive TB were screened consecutively. Sputum culture was used for TB screening and blood was used for HIV screening, as well as fasting plasma glucose (FPG) and glycated haemoglobin (HbA1c) assessment for the diagnosis of DM. HbA1c was measured using the D-10 Haemoglobin Testing System and a point-of-care test (A1C Now+ system) for comparison. Patients were classified as having DM or pre-diabetes using the D-10 reference test.

RESULTS: Four hundred and ten individuals had TB culture, FPG, and HbA1c results. Participants had a mean (±standard deviation) age of 37.8±12.6 years and 217 (54.8%) were male. One hundred and thirteen (27.6%) patients were culture-positive, 62 (15.1%) had DM, and 46 (11.2%) had pre-diabetes. One hundred and eighty-four (53.3%) participants were HIV-positive and 95 (51.6%) were on antiretroviral therapy (ART). Patients with pre-diabetes and DM were more likely to have TB (odds ratio (OR) 1.94, 95% confidence interval (CI) 0.01-3.74, and OR 2.39, 95% CI 1.35-4.24, respectively). After adjustment for HIV, age, and sex, only DM was statistically associated with TB (adjusted OR (AOR) 3.10, 95% CI 1.62-5.94). HIV-negative patients with DM had a higher risk of TB (AOR 4.32, 95% CI 1.57-11.92) than HIV-positive patients with DM (AOR 3.31, 95% CI 1.29-8.54), but the difference was not statistically significant. A1C Now+ HbA1c measurements correlated poorly with the D-10 HbA1c reference test.

CONCLUSION: A high proportion of patients in Abuja have markers of DM and pre-diabetes at the time of TB diagnosis.