Antituberculosis drug-resistance surveillance as a tool for tuberculosis control programmes: a retrospective study

The aim of this study was to verify the suitability of antituberculosis (antiTB) drug-resistance surveillance as a tool for tuberculosis (TB) control programmes at local level. A retrospective study reviewing laboratory records and medical records of TB patients referred to Udine Hospital between 1981 and 1995 was analysed. The initial susceptibility pattern for each Mycobacterium tuberculosis isolate was recorded. It was found that between 1981 and 1995, 899 M. tuberculosis strains underwent susceptibility testing for four first-line drugs. Over a period of 15 yrs the annual number of M. tuberculosis strains initially decreased and then stabilized. Overall, 15.3% of the 899 strains showed initial resistance to at least one first-line drug, and 2.8% to two or more first-line drugs. Streptomycin-resistant strains were the most commonly observed (10.8%), with resistance to isoniazid, rifampicin and ethambutal shown to be 6.4, 1.0 and 0.4%, respectively. Multidrug resistant (MDR)-TB was observed in only five cases. An additional four cases eventually developed secondary MDR-TB during the follow-up. The proportion of resistant strains did not vary significantly over time. Recurrent TB disease was significantly associated with resistant strains (odds ratio = 3.59, p < 0.01). Only one patient had a documented human immunodeficiency virus (HIV)-positive serology. All six patients who developed MDR-TB during or after treatment, were suffering either from chronic alcoholism or from a psychotic disorder. In the study it was shown that recurrent tuberculosis cases, tuberculosis patients with behavioural problems (i.e. alcoholism, psychiatric disorder) and patients presenting with primary resistant Mycobacterium tuberculosis strains are at risk of multidrug-resistant tuberculosis and may thus benefit from the directly observed treatment approach, which has been proposed as a mainstay in tuberculosis control programme strategy.