The association of hyperglycaemia with prevalent tuberculosis: a population-based cross-sectional study

Sarah Lou Bailey, Helen Ayles, Nulda Beyers, Peter Godfrey-Faussett, Monde Muyoyeta, Elizabeth du Toit, John S. Yudkin, Sian Floyd

Systematic reviews suggest that the incidence of diagnosed tuberculosis is two- to- three times higher in those with diabetes mellitus than in those without. Few studies have previously reported the association between diabetes or hyperglycaemia and the prevalence of active tuberculosis and none in a population-based study with microbiologically-defined tuberculosis. Most have instead concentrated on cases of diagnosed tuberculosis that present to health facilities. We had the opportunity to measure glycaemia alongside prevalent tuberculosis. A focus on prevalent tuberculosis enables estimation of the contribution of hyperglycaemia to the population prevalence of tuberculosis. A population-based cross-sectional study was conducted among adults in 24 communities from Zambia and the Western Cape (WC) province of South Africa. Prevalent tuberculosis was defined by the presence of a respiratory sample that was culture positive for M. tuberculosis. Glycaemia was measured by random blood glucose (RBG) concentration. Association with prevalent tuberculosis was explored across the whole spectrum of glycaemia. Among 27,800 Zambian and 11,367 Western Cape participants, 4,431 (15.9%) and 1,835 (16.1%) respectively had a RBG concentration ≥7.0 mmol/L, and 405 (1.5%) and 322 (2.8%) respectively had a RBG concentration ≥11.1 mmol/L. In Zambia, the prevalence of tuberculosis was 0 · 5% (142/27,395) among individuals with RBG concentration <11.1 mmol/L and also ≥11.1 mmol/L (2/405); corresponding figures for WC were 2 · 5% (272/11,045) and 4 · 0% (13/322). There was evidence for a positive linear association between hyperglycaemia and pulmonary prevalent tuberculosis. Taking a RBG cut-off 11.1 mmol/L, a combined analysis of data from Zambian and WC communities found evidence of association between hyperglycaemia and TB (adjusted odds ratio = 2 · 15, 95% CI [1 · 17-3 · 94]). The population attributable fraction of prevalent tuberculosis to hyperglycaemia for Zambia and WC combined was 0.99% (95% CI 0 · 12%-1.85%) for hyperglycaemia with a RBG cut-off of 11.1 mmol/L. This study demonstrates an association between hyperglycaemia and prevalent tuberculosis in a large population-based survey in Zambia and Western Cape. However, assuming causation, this association contributes little to the prevalence of TB in these populations.