Risk of infection with Mycobacterium tuberculosis in travellers to areas of high tuberculosis endemicity

Background No data exist on risks of infection with Mycobacterium tuberculosis in travellers. We studied incidences of and risk factors for tuberculin skin-test conversion among Dutch long-term travellers to countries of high tuberculosis endemicity. Methods In a multicentre, prospective cohort study based in travel and tuberculosis clinics in the Netherlands, 1072 BCG-naïve immunocompetent travellers to countries with an estimated annual risk of M tuberculosis infection of at least 1% were skin tested before departure with 1 tuberculin unit purified protein derivative (PPD) of M tuberculosis in Tween-80. Those with results less than 2 mm were retested 2–4 months after their return with simultaneous testing for cross-sensitivity to environmental mycobacteria (1 tuberculin unit PPD of M scrofulaceum in Tween-80). M tuberculosis infection was defined as a post-travel M tuberculosis tuberculin skin-test result of at least 10 mm that was 3 mm or more larger than the M scrofulaceum result. Findings Post-travel skin-test results were available for 656 (66%) of 988 individuals who were eligible for follow-up. Among these, 12 M tuberculosis infections were identified (1·8%). The overall incidence rate was 3·5 per 1000 person-months of travel (95% Cl 2·0–6·2), and 2·8 per 1000 person-months of travel (1·2–5·5) after exclusion of health-care workers. Two had active tuberculosis at the time of testing (incidence rate 0·6 per 1000 person-months of travel [0·3–2·3]). Work in patient care abroad was an independent risk factor (adjusted rate ratio 5·34, p=0·015). Interpretation The risk of M tuberculosis infection in long-term travellers to high-endemicity countries, even if not engaged in health-care work, is substantial and of similar magnitude to the average risk for the local population. BCG vaccination or post-travel tuberculin skin-testing of high-risk travellers should be considered.