Background: There is little information about serum phosphate levels among patients with pulmonary tuberculosis (TB) and HIV infection.

Objective: We aimed to assess the role of TB, HIV, inflammation and other correlates on serum phosphate levels.

Methods: A cross-sectional study was conducted among TB patients and age- and sex-matched non-TB controls. Pulmonary TB patients were categorized as sputum -negative and -positive, based on culture. Age- and sex-matched non-TB controls were randomly selected among neighbours to sputum-positive TB patients. Data on age, sex, alcohol and smoking habits were obtained. HIV status, serum phosphate, and the acute phase reactants C-reactive protein (serum CRP) and α₁-acid glycoprotein (serum AGP) were determined. Linear regression analysis was used to identify correlates of serum phosphate.

Results: Of 1605 participants, 355 (22.1%) were controls and 1250 (77.9%) TB patients, of which 9.9% and 50.4% were HIV-infected. Serum phosphate was determined before start of TB treatment in 44%, and 1–14 days after start of treatment in 56%. Serum phosphate was up to 0.10 mmol/L higher 1–3 days after start of TB treatment, and lowest 4 days after treatment, after which it increased. In multivariable analysis, TB patients had 0.09 (95% CI: 0.05; 0.13) mmol/L higher serum phosphate than controls, and those with HIV had 0.05 (95% CI: 0.01; 0.08) mmol/L higher levels than those without. Smoking was also a positive correlate of serum phosphate, whereas male sex and age were negative correlates.

Conclusion: While HIV and TB are associated with higher serum phosphate, our data suggest that TB treatment is followed by transient reductions in serum phosphate, which may reflect hypophosphataemia in some patients.