Latent Tuberculosis in Pregnancy: A Systematic Review

BACKGROUND: In countries with low tuberculosis (TB) incidence, immigrants from higher incidence countries represent the major pool of individuals with latent TB infection (LTBI). The antenatal period represents an opportunity for immigrant women to access the medical system, and hence for potential screening and treatment of LTBI. However, such screening and treatment during pregnancy remains controversial. OBJECTIVES: In order to further understand the prevalence, natural history, screening and management of LTBI in pregnancy, we conducted a systematic literature review addressing the screening and treatment of LTBI, in pregnant women without known HIV infection. METHODS: A systematic review of 4 databases (Embase, Embase Classic, Medline, Cochrane Library) covering articles published from January 1st 1980 to April 30th 2014. Articles in English, French or Spanish with relevant information on prevalence, natural history, screening tools, screening strategies and treatment of LTBI during pregnancy were eligible for inclusion. Articles were excluded if (1) Full text was not available (2) they were case series or case studies (3) they focused exclusively on prevalence, diagnosis and treatment of active TB (4) the study population was exclusively HIV-infected. RESULTS: Of 4,193 titles initially identified, 208 abstracts were eligible for review. Of these, 30 articles qualified for full text review and 22 were retained: 3 cohort studies, 2 case-control studies, and 17 cross-sectional studies. In the USA, the estimated prevalence of LTBI ranged from 14 to 48% in women tested, and tuberculin skin test (TST) positivity was associated with ethnicity. One study suggested that incidence of active TB was significantly increased during the 180 days postpartum (Incidence rate ratio, 1.95 (95% CI 1.24-3.07). There was a high level of adherence with both skin testing (between 90-100%) and chest radiography (93-100%.). In three studies from low incidence settings, concordance between TST and an interferon-gamma release assay was 77, 88 and 91% with kappa values ranging from 0.26 to 0.45. In low incidence settings, an IGRA may be more specific and less sensitive than TST, and results do not appear to be altered by pregnancy. The proportion of women who attended follow-up visits after positive tuberculin tests varied from 14 to 69%, while 5 to 42% of those who attended follow-up visits completed a minimum of 6 months of isoniazid treatment. One study raised the possibility of an association of pregnancy/post-partum state with INH hepatitis (risk ratio 2,5, 95% CI 0.8-8.2) and fatal hepatotoxicity (rate ratio 4.0, 95% CI 0.2-258). One study deemed INH safe during breastfeeding based on peak concentrations in plasma and breast milk after INH administration. CONCLUSION: Pregnancy is an opportunity to screen for LTBI. Interferon-gamma release assays are likely comparable to tuberculin skin tests and may be used during pregnancy. Efforts should be made to improve adherence with follow-up and treatment post-partum. Further data are needed with respect to safety and feasibility of antepartum INH therapy, and with respect to alternative treatment regimens.