LETTERS TO THE EDITOR Multidrug resistant tuberculosis

We read with interest the report by White and Moore-Gillon on the resource implications of multidrug resistant tuberculosis (MDRTB) in the UK. We studied the outcome of 44 HIV negative patients with MDR-TB admitted to two tertiary care tuberculosis units in the Netherlands between 1985 and 1998. 3 Most (38 patients) had pulmonary tuberculosis. The mean admission period was 164 days and all patients received an individually tailored combination of antituberculosis drugs for a mean period of 608 days. We estimated the cost of treatment per patient to be US$60 000 which included admission fee, costs for outpatient visits, and the costs of drug provision. Although we did not include cost of toxicity monitoring and additional procedures, our costs compare favourably with those of White et al (mean £60 000) and Mahmoudi (mean US$180 000). In the Netherlands the number of patients with tuberculosis resistant to any antituberculous drugs is limited to 11%, and only 0.6% of the bacilli are MDR. Between 1993 and 1997 only 43 cases of MDR-TB were identified, of which 28% had received previous treatment for tuberculosis (Index Tuberculosis 1998, Royal Netherlands Tuberculosis Association, The Hague, 2000). This suggests that transmission of MDR bacilli rather than inadequate treatment contributes to the resistance problem in the Netherlands. In poor resource countries, directly observed short course chemotherapy of tuberculosis is generally recommended to prevent the occurrence of MDR-TB. We feel that new rapid molecular methods for detecting resistance should be developed to limit the period of potential (nosocomial and community) transmission of MDR bacilli and thus prevent the emergence of MDR-TB. Such tests should then be made available to poor resource countries at an aVordable price.