Are we missing opportunities to confirm the diagnosis of tuberculosis by microbial culture?

Abstract

Setting: Tuberculosis (TB) incidence is rising globally, with drug resistance becoming increasingly problematic. Microbiological confirmation ensures correct anti-tuberculous chemotherapy.

Objective/design: We retrospectively analysed all TB cases diagnosed in Central Manchester in 2009 investigating how often we are not achieving microbiological diagnosis, factors influencing this and whether opportunities to obtain microbiological samples are missed.

Results: 128/156 (82%) cases had samples sent for microbiology. Factors affecting this included disease site, with ocular disease least likely to be sampled (p < 0.0001), and patient age (with children less likely to be sampled p = 0.002). Ethnicity did not affect sampling (n.s.). Overall, 92/156 (59%) cases were culture positive. Negative culture was related to specimen type (p < 0.0001) and patient age (p = 0.019), with children significantly less likely to have a positive culture. Ethnicity and disease site did not affect culture results. There was a trend towards culture positivity being more common in pulmonary (75%) than non-pulmonary (46%) disease (n.s.). In only 7 (4%), could samples have been sent where they were originally absent (3) or further samples obtained where the cultures proved to be negative (4).

Conclusion: Despite an overall culture positive rate of 59%, opportunities to achieve microbiological confirmation are seldom missed. In our centre, which is typical of UK practice, this lack of capacity to increase microbiological confirmation, particularly in an era of increasing importance of extra-pulmonary TB, is concerning. Improvements in sample acquisition and laboratory methods are urgently required.

Keywords: Investigation; Microbiological diagnosis; Mycobacteria.