Setting: SMIRA (Italian Study on Anti-Tuberculosis Drug Resistance) network, 46 major clinical units and 22 laboratories nationwide in Italy.
Objectives: To determine the main features, adherence to WHO guidelines and the outcomes of multidrug-resistant tuberculosis (MDR-TB) patients enrolled from January 1995 to December 1999.
Design: Observational study, preceded by proficiency testing, according to WHO recommendations. Results were stratified by appropriate and inappropriate regimens (< three active drugs). Analysis of the outcomes was performed according to adequacy of treatment. Analysis of risk factors and factors predicting treatment outcomes was performed using univariate and multivariate analysis (level of significance P < 0.05).
Results: One hundred and twenty-seven MDR patients were diagnosed. The overall success rate was low (39%). Seventy per cent of cases were treated with at least three active drugs. Factors predicting treatment success were new MDR-TB cases (OR 3.45; 95% CI 1.22-9.78; P < 0.05) and treatment for > or = 12 months (OR 5.03; 95% CI 1.65-15. 31; P < 0.05). Immigration and HIV infection were the main risk factors among new MDR-TB cases.
Conclusion: The best available treatment should be provided to all newly diagnosed MDR-TB patients, avoiding the use of poorly modified regimens. MDR-TB patients should be referred to highly specialised centres.