Contribution of reinfection to recurrent tuberculosis in South African gold miners.

SETTING: A gold mine in South Africa. OBJECTIVE: To investigate incidence and risk factors for tuberculosis (TB) recurrence and the relative contribution of reinfection and relapse to recurrence. DESIGN: Prospective cohort study. METHODS: Employees cured of a first episode of culture-positive TB were followed up for recurrence, which was classified as reinfection or relapse by restriction fragment length polymorphism using an insertion sequence (IS) 6110 probe. RESULTS: Among 609 patients, 57 experienced recurrence during a median follow-up period of 1.02 years, corresponding to a recurrence rate of 7.89 per 100 person-years (py). The culture positive recurrence rate was 5.79/100 py, and was higher in human immunodeficiency virus (HIV) infected patients (8.86/100 py in HIV-infected vs. 3.35/100 py in non-HIV-infected). Among HIV-infected patients, the risk of culture-positive recurrence was higher with decreasing CD4 count (compared with CD4 < 200, hazard ratios for recurrence among individuals with CD4 200-500 and CD4 > 500 were 0.40 [95%CI 0.14-1.09] and 0.14 [95%CI 0.02-1.10], respectively, Ptrend = 0.01). IS6110 genotyping was available on both the initial and subsequent isolate for 16/42 (38%, 14 HIV-infected) patients with culture-positive recurrence, and showed reinfection in 11 (69%). CONCLUSION: HIV-infected gold miners, particularly those who are more immunosuppressed, are at higher risk of TB recurrence. TB control strategies need to take into account reinfection as an important cause of recurrent TB.

Impact of cotrimoxazole on non-susceptibility to antibiotics in Streptococcus pneumoniae carriage isolates among HIV-infected mineworkers in South Africa.

OBJECTIVES: To investigate risk factors for pneumococcal carriage and non-susceptibility among HIV-infected mineworkers in South Africa. METHODS: In a cross-sectional study, HIV clinic attendees were questioned about risk factors for pneumococcal carriage and antimicrobial non-susceptibility. Oropharyngeal and nasopharyngeal swabs were taken for pneumococcal culture, serotyping and susceptibility testing. RESULTS: Among 856 participants (854 male, median age 41.5years, median CD4 290cells/mm(3)), 294 (34.3%) were receiving cotrimoxazole prophylaxis. Overall, 75/856 (8.8%) carried S. pneumoniae; among those taking vs. not taking cotrimoxazole, 8.2% vs. 9.1% were carriers. Risk factors for pneumococcal carriage were living with a child (adjusted OR 2.12, 95% CI 1.06-4.62) and recent hospitalisation (adjusted OR 1.80; 95% CI 0.98-3.30). Among participants not taking cotrimoxazole, the prevalence of carriage was higher in individuals with lower CD4 counts. Comparing participants taking cotrimoxazole vs. not, 60.9% vs. 22.4% (p=0.001) isolates were non-susceptible to cotrimoxazole and 30.4% vs. 8.2% were non-susceptible to penicillin (p=0.014). Thirty three/72 (45.8%) isolates were paediatric serotypes/groups. Nasopharyngeal compared with oropharyngeal swabs had higher sensitivity in detecting carriage (53/75, 70.7% vs. 31/75, 41.3%), and adding oropharyngeal sampling increased detection from 6.2% to 8.8%. CONCLUSIONS: Non-susceptibility to cotrimoxazole and penicillin was more common among isolates from participants taking cotrimoxazole prophylaxis. Surveillance for antimicrobial susceptibility is important where prophylaxis is used. Treatment for pneumococcal disease should take into account a higher risk of non-susceptibility to antibiotics amongst individuals taking cotrimoxazole prophylaxis.