REGULAR ARTICLES Childhood drug-resistant tuberculosis in the Western Cape Province of South Africa H. SIMON SCHAAF 1,2 , BEN J. MARAIS 1,2 , ANNEKE C. HESSELING 2,3 , ROBERT P. GIE 1,2 , NULDA BEYERS 1,2 & PETER R. DONALD 1,2 1 Department of Paediatrics and Child Health, Tygerberg Children’s Hospital and Stellenbosch University, Tygerberg, South Africa, 2 Desmond Tutu Tuberculosis Centre, Tygerberg, South Africa, and 3 Department of Infectious Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom Abstract Aim: Continued surveillance of drug resistance amongst children presenting with culture-confirmed tuberculosis to a tertiary care hospital and to community clinics. Methods: Drug susceptibility testing for isoniazid and rifampicin was prospectively done on all children presenting with culture-confirmed tuberculosis from three study populations: children diagnosed at Tygerberg Children’s Hospital in the Western Cape Province, South Africa, between March 2003 and February 2005 (recent survey, hospital-based group), and August 1994 and April 1998 (previous hospital-based survey), and a community-based group from five clinics during the recent survey. Results: Isoniazid resistance increased significantly from 6.9% (21/306 children) in the previous survey to 12.8% (40/313 children) in the recent survey (odds ratio (OR) 1.99, 95% confidence interval (CI) 1.11 /3.59). Resistance to isoniazid and rifampicin (multidrug resistance) did not increase significantly between the surveys (7/306 in previous survey vs 17/313 in recent survey; OR 2.45, 95% CI 0.94 /6.62). All resistance (isoniazid and/or rifampicin) was 18/127 (14.2%) in the community-based group compared to 23/192 (12.0%) in the hospital-based group (OR 0.82, 95% CI 0.41  /1.68). Conclusion: The prevalence of drug-resistant tuberculosis in this setting is increasing, reflecting ongoing transmission of drug-resistant Mycobacterium tuberculosis . Key Words: Children, drug resistance, survey, tuberculosis Introduction Drug-resistant tuberculosis (TB), and especially mul- tidrug-resistant (MDR) TB, poses a major threat to individual patients and tuberculosis control program- mes. The prevalence of drug-resistant TB gives an indication of how well a TB programme is function- ing. Trends in acquired drug resistance reflect the ability of tuberculosis control programmes to treat patients infected with susceptible strains of Mycobac- terium tuberculosis effectively, and are an important indicator of the effectiveness of a tuberculosis control programme. Trends in transmitted (primary) drug resistance reflect the ability of programmes to limit the spread of resistant organisms within communities [1]. Determining whether adult patients have drug resistance in developing countries is difficult, as they are mainly diagnosed by direct sputum microscopy for acid-fast bacilli and culture, and drug susceptibility testing is not routinely done [2,3]. There are two main mechanisms for the development of drug resistance: in the past, this distinction was made on a history of previous TB treatment and defined as primary drug resistance (no previous TB treatment, implying trans- mitted resistance) or acquired drug resistance (implying development of drug resistance in a previously treated patient due to inappropriate treatment). However, these definitions assume acquisition of drug resistance in previously treated patients; this cannot be proven in the absence of genotyping of M. tuberculosis isolates (Received 25 February 2006; revised 6 March 2006; accepted 7 March 2006) ISSN 0803-5253 print/ISSN 1651-2227 online # 2006 Taylor & Francis DOI: 10.1080/08035250600675741 Correspondence: H. Simon Schaaf, Department of Paediatrics and Child Health, Faculty of Health Sciences, Stellenbosch University, PO Box 19063, Tygerberg 7505, South Africa. Tel:  /27 21 9389112. Fax:  /27 21 9389138. E-mail: hss@sun.ac.za Acta Pædiatrica, 2006; 95: 523  /528