Molecular and conventional epidemiology of tuberculosis in an inner city district

SETTING: Molecular epidemiology has underlined the importance of recent tuberculosis (TB) transmission and has uncovered notable discrepancies compared with conventional epidemiology. OBJECTIVES: 1) To determine, by RFLP analysis, the percentage of clustered cases in an inner city district with a high incidence of TB (163/100 000) and the groups at risk of being clustered; and 2) to compare the role of conventional contact tracing (CCT) with that of RFLP. DESIGN: RFLP was carried out using the IS6110 and pTBN12 (PGRS) sequences of 165 cultures positive for Mycobacterium tuberculosis in the Ciutat Vella district of Barcelona during 1997-1998. Contact tracing was carried out in 171 of 251 declared cases (68.1%). Associations were assessed by calculating odds ratios (OR) with 95% confidence intervals (CI). Logistic regression was used for multivariate analysis. RESULTS: Using RFLP, 76 (46.0%) strains were found to be clustered by IS6110 and PGRS. From CCT, 30 new patients were found among 858 contacts (3.5%) and 57 patients were linked. In terms of RFLP and CCT, the main risk factor was intravenous drug use (IVDU). In 44 cases who lived alone and were not involved in CCT, 50% were in RFLP clusters. The concordance rate between RFLP and CCT was 8/13 (61.5%); the disagreement corresponded to a cluster of five recent immigrants from Africa. Subsequent to RFLP, an epidemiological connection was found in 15/55 cases (27.2%). CONCLUSIONS: The percentage of clustered cases is very high. CCT was useful for identifying new cases, but it was insufficient for detecting the pathways of transmission. CCT coverage needs to be improved in marginalized individuals, and the results correlated with those of RFLP.