Abstract:
INTRODUCTION
Tuberculosis (TB) is a significant public health problem and the diagnosis in human immunodeficiency
virus (HIV)—infected individuals is challenging. The use of mycobacterial culture
remains an important complementary tool and optimizing it has important benefits. We
sought to determine the effect of an increase in the number of specimens evaluated, addition
of nutritional supplementation to the culture medium, sputum appearance and volume
on diagnostic yield and time to detection of pulmonary TB among smear-negative, HIVinfected
adults.
METHODS
In this prospective study conducted at the Tshwane District Hospital and Academic TB Laboratory,
Pretoria, South Africa we collected three sputum specimens an hour apart from presumptive
TB cases at an antiretroviral treatment site. We analysed specimens from 236
patients. Specimen appearance and volume were recorded. All specimens were processed
for culture using both standard and supplemented media.
RESULTS
A single specimen identified 79% of PTB cases using standard media; the second and third
specimens added 12.5% and 8.3% respectively. Media supplementation, sputum appearance and specimen volume had no effect on culture yield or contamination rates. The mean
time to detection was reduced from 19.8 days in standard cultures to 11.8 days in nutrient
supplemented cultures (p = 0.002). For every 1 ml increase in sputum volume, time to
detection was decreased by a factor of 0.797 (p = 0.011).
CONCLUSION
Use of an inexpensive culture supplement substantially reduced time to detection and could
contribute to reducing treatment delay among HIV-infected cases.
