Tuberculosis and diabetes mellitus: A dose-response relationship between the odds of tuberculosis and HbA1c

SETTING: The main hospital tuberculosis (TB) clinic in South Tarawa, the capital of the Republic of Kiribati, a Pacific nation located in the central Pacific Ocean. OBJECTIVE: To determine if higher levels of HbA1c were associated with greater odds of TB. DESIGN: A case-control study to assess the association between TB and diabetes (DM). We recruited 275 TB cases and 498 controls (persons without signs and symptoms of TB), aged 18 years and above. A standardised questionnaire was administered and HbA1c was measured in all participants. RESULTS: The median HbA1c among cases was 6.0%; among controls it was 5.6% (P, 0.001). Comparing cases to controls, the odds ratio for TB was 2.8 (95%CI 2.0–4.1). Adjusted odds ratios for TB associated with HbA1c groupings of 5.7–6.4%, 6.5–8.5% and .8.5% were 1.5, 2.7 and 4.3, respectively (P≤0.001 for trend). CONCLUSIONS: The median HbA1c was higher among TB patients than controls. As the HbA1c rose so did the odds of TB. This demonstrates the importance of targeted TB screening of people with DM in TB-endemic settings. Optimal glycaemic control is also crucial as this reduces DM-related end organ damage and may also reduce TB risk.