Introduction TB incidence is rising in the UK, with drug resistance becoming increasingly problematic. Diagnosis with microbiological culture and confirmation of sensitivity is therefore vital. This study investigated how often we are not achieving microbiological diagnosis at our centre, what factors influence this and whether opportunities to obtain microbiological samples were missed.
Methods A retrospective study of all 156 cases (adult and paediatric) diagnosed with TB at Central Manchester Teaching Hospitals in 2009 was carried out. Demographic details, site of disease, types of specimens and results of TB culture were recorded. Cases where there were no specimens or cultures were negative were analyzed in detail.
Results Most disease was pulmonary (n=69). Other disease sites included lymph node (n=34), ocular (n=12) and pleura (n=10). 128 (82%) patients had samples sent for microbiology. 92 (59%) patients were culture positive and 36 (23%) were culture negative. 28 (18%) patients had no specimens sent for culture. Factors which were associated with whether samples were sent for culture included site of disease (p<0.0001), with ocular disease being the least likely to be sampled, and age of patients (p=0.002). 45% patients <17 years did not have samples sent compared with 12.5 % patients 17–64. Ethnicity did not influence the frequency of sampling. A negative culture result was related to the specimen type (p<0.0001) and patient's age (p=0.019), with fewer paediatric samples positive. Site of disease or ethnicity did not affect culture results. In 25/28 cases with no microbiological specimens it was considered reasonable that specimens were not sent, as most of these were either ocular (n=12) or paediatric (n=9). In 3/28 (11%) samples could have been sent and all involved adult patients not born in the UK who had procedures whereby specimens were sent only for histology. 32/36 culture negative cases were considered to have been managed appropriately. 4/36 (11%) culture negative cases were potential missed opportunities for further sampling and were all due to patients with pleural TB not having pleural biopsies.
Conclusion In our centre, despite a microbiology negative rate of 41%, reasonable opportunities to obtain a microbiological diagnosis are seldom missed.