Prediabetes and the treatment outcome of tuberculosis [TB0131]

Prediabetes can be diagnosed through impaired fasting glucose, impaired glucose tolerance, or mildly elevated HbA1c. Although the mechanisms linking prediabetes to poor tuberculosis treatment outcomes are not fully clear, they might include chronic low-grade inflammation and immune dysfunction. Literature up to March 31, 2024, was reviewed with the following inclusion criteria:

1. Full-length articles on cohort studies (prospective or retrospective).
2. Patients treated with standard anti-tuberculosis treatment.
3. Prediabetes diagnosed at baseline as per study criteria.
4. Studies reporting outcomes like unfavourable treatment outcomes (treatment modification, failure, recurrence, or death) or all-cause mortality.

Exclusion criteria included:

1. Cross-sectional or case-control studies.
2. Studies not assessing prediabetes as the exposure.
3. Lack of comparison with normoglycemic patients.
4. No reported incidence of outcomes during follow-up.
5. Preclinical studies, reviews, editorials, or prior meta-analyses.

For overlapping study populations, the study with the largest sample size was used.
The association between prediabetes and adverse TB treatment outcomes was evaluated using risk ratios (RR) with 95% confidence intervals (CI), comparing prediabetic to normoglycemic individuals. Hazard ratios (HR) were directly extracted as RR where available. For odds ratios (OR), conversion to RR was performed using the formula: RR = OR/([1 − pRef] + [pRef × OR]), where pRef is the outcome prevalence in the normoglycemic group.

Eight cohort studies with a total of 3001 patients, of which 752 (25.1%) had prediabetes at baseline, were analyzed. These studies, conducted in Georgia, China, India, Peru, Brazil, and Korea from 2015 to 2024, included:

1. Six studies on pulmonary tuberculosis only.
2. Two studies that also covered extrapulmonary tuberculosis.

Patient demographics:

1. Age range: 33.4 to 62.0 years.
2. Gender: 61.8% to 84.5% male.

Follow-up was between six and 24 months post-initiation of TB treatment. Pooled analysis from eight studies indicated that tuberculosis patients with prediabetes at baseline had a higher incidence of unfavourable treatment outcomes compared to normoglycemic patients. However, data from five studies showed no increased risk of all-cause mortality associated with prediabetes.

Source: Liang, L. and Su, Q., 2024. Prediabetes and the treatment outcome of tuberculosis: A meta‐analysis. Tropical Medicine & International Health, 29(9), pp.757-767.