Comparative analysis of multidrug-resistant tuberculosis and extensively drug-resistant tuberculosis - epidemiology and predictive factors.

Introduction: Extensively drug-resistant tuberculosis (XDR-TB) is defined as a form of multidrug-resistant tuberculosis (MDR-TB) with additional resistance to fluoroquinolones and at least one of the injectable drugs used in tuberculosis treatment: amikacin, kanamycin and capreomycin. It was classified by WHO as a serious threat to tuberculosis (TB) control, with world-wide consequences, taking on the proportions of a real pandemic in some regions.

Aim: To compare patients with XDR-TB versus other MDR-TB profiles with regard to epidemiological and demographic characteristics, aetiopathogenic factors and inhospital outcomes.

Methods: Patients admitted to Pulido Valente Hospital (Pulmonology Service III) in the period ranging from April 1999 to June 2007 with MDR-TB diagnosis microbiologically confirmed. The following variables were evaluated: gender, age, race, forms of TB presentation, treatment groups, resistance profile, immigrant status, number and duration of previous treatments, WHO classification, HIV co-infection, alcoholism and/or drug addiction, average length of hospital stay and inhospital mortality. Statistical analysis was performed using the SPSS (Statistical Package for the Social Sciences), version 15.0. In categorical variables, the statistical differences between groups were evaluated by the Chi-square test and numeric variables using the T-test. Logistical regression analysis was used to build the predictive model of XDR-TB existence (dependent variable), which included the following independent variables: WHO classification, HIV co-infection, immigrant status, alcoholism and/or drug addiction and number and duration of previous treatments.

Results: We recorded 132 patients with MDR-TB, of which 69 (52.3%) were XDR-TB. Statistically significant differences were observed in the following variables: race (black race was associated with XDRTB in 74% of cases versus 46% of the Caucasian race); WHO classification (patients with retreatment for therapeutic failure, stopping treatment or relapse were 69.5% of XDR-TB cases versus 44.5% of Not XDR-TB cases; average duration of previous treatments (4.2 months for XDR-TB cases versus 2.8 months for Not XDR-TB cases); HIV co-infection (patients with HIV co-infection constituted 65.2% of XDR-TB cases versus 42.9% of Not XDR-TB cases), mortality (33.3% in patients with XDR-TB versus 14.3% in Not XDR-TB patients).

Conclusions: The variables with predictive value for the diagnosis of XDR-TB vs. Not XDR-TB were presence of HIV co-infection (odds ratio [OR] for XDRTB 2.5; 95% confidence interval [CI], 1.24 - 5.05) and increased average duration of previous treatments ([OR] for XDR-TB 1.2; 95% [CI], 1.11 - 2.30).