Tuberculosis among HIV-positive patients across Europe: changes over time and risk factors.

OBJECTIVE: To describe temporal changes in the incidence rate of tuberculosis (TB) (pulmonary or extrapulmonary) among HIV-positive patients in western Europe and risk factors of TB across Europe. METHODS: Poisson regression models were used to determine temporal changes in incidence rate of TB among 11,952 patients from western Europe (1994-2010), and to assess risk factors for TB among 12,673 patients from across Europe with follow-up after 2001. RESULTS: Two hundred and seventy-seven TB events occurred during 84,221 person-years of follow-up (PYFU) in western Europe. The incidence rate declined from 1.91 [95% confidence interval (CI) 1.51-2.37)] in 1994-1995 to 0.12 (0.07-0.21)/100 PYFU in 2002-2003, and remained stable thereafter. After January 2001, 159 TB events were diagnosed; 65 cases in western Europe and 94 cases in eastern Europe; resulting in incidence rates of 0.12 (0.09-0.14) and 0.65 (0.52-0.79)/100 PYFU, respectively. In multivariable analysis, incidence rate of TB was approximately four-fold higher in eastern Europe compared with western Europe [incidence rate ratio (IRR) 4.25 (2.78-6.49), P < 0.001]. There were no significant temporal changes after 2001 and risk factors did not differ significantly between eastern Europe and western Europe. Lower CD4 cell counts, higher HIV-RNA levels, male sex, intravenous drug usage and African origin were all associated with higher risk of TB. CONCLUSION: Incidence rates of TB in western Europe remained at a very low and stable level since 2001. After 2001, patients in eastern Europe were at substantially higher risk of TB than in western Europe. TB is of great concern in HIV-positive patients, especially in areas with high TB prevalence, high levels of immigration from TB-endemic regions, and with suboptimal access to combination antiretroviral therapy.

Symptom-based screening of child tuberculosis contacts: improved feasibility in resource-limited settings.

OBJECTIVE: National tuberculosis programs in tuberculosis-endemic countries rarely implement active tracing and screening of child tuberculosis contacts, mainly because of resource constraints. We aimed to evaluate the safety and feasibility of applying a simple symptom-based approach to screen child tuberculosis contacts for active disease. METHODS: We conducted a prospective observational study from January through December 2004 at 3 clinics in Cape Town, South Africa. All of the children <5 years old in household contact with an adult tuberculosis source case were assessed by documenting current symptoms and tuberculin skin test and chest radiograph results. RESULTS: During the study period, 357 adult tuberculosis cases were identified; 195 cases (54.6%) had sputum smear and/or culture positive results and were in household contact with children aged <5 years. Complete information was available for 252 of 278 children; 176 (69.8%) were asymptomatic at the time of screening. Tuberculosis treatment was administered to 33 (13.1%) of 252; 27 were categorized as radiologically "certain tuberculosis," the majority (n = 22) of which had uncomplicated hilar adenopathy. The negative predictive value of symptom-based screening varied according to the case definition used, with 95.5% including all of the children treated for tuberculosis and 97.1% including only those with radiologically "certain tuberculosis." CONCLUSIONS: Our findings support current World Health Organization recommendations, demonstrating that symptom-based screening of child tuberculosis contacts should improve feasibility in resource-limited settings and seems to be safe.