Intended for healthcare professionals


Database study of antibiotic resistant tuberculosis in the United Kingdom, 1994-6

BMJ 1999; 318 doi: (Published 20 February 1999) Cite this as: BMJ 1999;318:497
  1. Charles Irish, senior registrar in communicable disease control (charles{at},
  2. Josephine Herbert, Mycobacterial Resistance Network coordinatorb,
  3. Diane Bennett, consultant epidemiologistb,
  4. Clare Gilham, statisticianc,
  5. Francis Drobniewski, directord,
  6. Rhian Williams, clinical scientiste,
  7. E Grace Smith, consultant medical microbiologistf,
  8. John G Magee, head of unitg,
  9. Brian Watt, directorh,
  10. Maureen Chadwick, principal biomedical scientisti,
  11. John M Watson, consultant epidemiologistb
  1. aBrent and Harrow Health Authority, Harrow, Middlesex HA1 3EX
  2. bPublic Health Laboratory Service Communicable Disease Surveillance Centre, London NW9 5EQ
  3. cPublic Health Laboratory Service, Statistics Unit, London NW9 5EQ
  4. dPublic Health Laboratory Service Mycobacterium Reference Unit, King's College Hospital, London SE22 8QF
  5. eRegional Centre for Mycobacteriology, Cardiff Public Health Laboratory, University Hospital of Wales, Cardiff CF4 4XW
  6. fRegional Centre for Mycobacteriology, Birmingham Public Health Laboratory, Birmingham Heartlands Hospital, Birmingham B9 5ST
  7. gRegional Centre for Mycobacteriology, Newcastle Public Health Laboratory, General Hospital, Newcastle upon Tyne NE4 6BE
  8. hScottish Mycobacteria Reference Laboratory, City Hospital, Edinburgh EH10 5SB
  9. iDepartment of Microbiology, Royal Brompton Hospital, London SW3 6NP
  1. Correspondence to: Dr Irish, Brent and Harrow Health Authority, Grace House, Harrow, Middlesex HA1 3EX
  • Accepted 6 October 1998

The global increase in tuberculosis which has occurred in the 1980s and 1990s, and the associated re-emergence of resistance to antituberculous drugs, has focused attention on recent trends in resistance in Europe and the United States.13 In the United Kingdom overall drug resistance levels have been low.4 A surveillance system, the UK Mycobacterial Resistance Network (MYCOBNET), was established in 1994 by the Public Health Laboratory Service to record drug resistance in laboratory isolates of tuberculosis. We used data from this network to examine resistance among people with newly diagnosed tuberculosis.

Subjects, methods, and results

We analysed the data on initial isolates of Mycobacterium tuberculosis complex referred to United Kingdom reference laboratories5 during 1994 to 1996. Initial isolates were defined as the first positive culture from a person from whom no positive culture had been recorded during the past 12 months. Since M bovis isolates are intrinsically resistant to pyrazinamide these were excluded from estimates of pyrazinamide resistance.

Isoniazid and multidrug resistance in isolates from patients with newly diagnosed tuberculosis in United Kingdom, 1994-6

View this table:

We calculated the resistance to each first line antibiotic and multidrug resistance (resistance to isoniazid and rifampicin with or without resistance to other antituberculous drugs) together with 95% confidence intervals. The incidence was assumed to follow the Poisson distribution. A chi2 test for trend was used to investigate changes in isoniazid and multidrug resistance over time.

Of 10 142 isolates recorded for 1994-6, 599 (5.9%; 95% confidence interval 5.5% to 6.4%) were resistant to isoniazid, 174 (1.7%; 1.5% to 2.0%) to rifampicin, 90/7494 (1.2%; 1.0% to 1.5%) to pyrazinamide; and 71 (0.7%; 0.6% to 0.9%) to ethambutol; 152 (1.5%; 1.3% to 1.8%) showed multidrug resistance.

The number and proportion of isolates resistant to isoniazid or with multidrug resistance increased from 1994 to 1996 (table). However, these increases were not significant chi2=0.797, df =1, P=0.372 for isoniazid resistance; chi2 =1.253, df =1, P= 0.263 for multidrug resistance). People aged 15 to 44 had the highest percentage of initial isolates with isoniazid resistance (8.1%) and multidrug resistance (2.0%) (table). A slightly higher percentage of males than females showed isoniazid resistance (6.2% v5.6%) and multidrug resistance (1.8% v1.2%).

In all, 568 (5.6%) patients had a known history of tuberculosis. These patients had a higher percentage of isoniazid resistance (18.8% v5.1%) and higher percentage of multidrug resistance (11.3% v0.9%) than those with no known history (table).

Resistance was higher among patients resident in England than in the rest of the United Kingdom (6.2% v3.8% for isoniazid resistance, and 1.6% v0.9% for multidrug resistance). Furthermore, patients diagnosed in London were more likely to have isolates resistant to isoniazid (8.0% v4.7%) or multidrug resistant (2.3% v1.0%) than those diagnosed outside London.

Resistance to isoniazid and multidrug resistance were observed among 13.5% and 6.1% respectively of the 460 patients known to be infected with HIV compared with 5.5% and 1.3% among the combined group of six HIV negative and 9676 patients whose HIV status was unknown.


This preliminary analysis of resistance in laboratory isolates establishes the importance of drug resistance in the United Kingdom and the need for continuing surveillance. Although overall resistance is low and the small increase was not significant, resistance remains a concern and should be considered in all newly detected cases. Action to prevent the emergence of new resistance by the supervision and completion of treatment and to stop the spread of established resistance is essential.


We thank the following for contributing to data collection, and to developing the surveillance system: B Bannister, S Bex, M Cocksedge, M Connolly, I Farrell, W Ferguson, J Grace-Parker, G Gleissberg, G Harris, A Hayward, E Hemingway, R Henry, D Howitt, P Jenkins, E Kazcmarski, J Leat, R Mallard, F Marais, A Middleton, J Mobray, P Morrell, C Newman, P Ormerod, R Poll, G Ratcliffe, A Rayner, M Roberts, R Schofield, R Shaw, G Stewart, D Tompkins, T Turner, T Wilson, and M Yates.

Contributors:CI performed the main analysis and wrote the paper. JH coordinated data collection and collation and participated in data analysis and writing the paper. DB designed the surveillance system and managed its establishment. CG participated in data analysis. FD, RW, EGS, JGM, BW, and MC managed all laboratory investigations and coordinated management of the surveillance system. JMW coordinated management of the surveillance system, contributed to the design and writing the paper, and is the study guarantor.

Funding:Public Health Laboratory Service.

Conflict of interest:None declared.


View Abstract