SETTING: Butajira, Southern Ethiopia. OBJECTIVE: To compare the diagnostic capacity of the clinical criteria for tuberculous lymphadenitis (TBLN) with histological and/or culture results and to assess the association of human immunodeficiency virus (HIV) with tuberculosis (TB) lymphadenitis. DESIGN: Patients (n=171) were included in the study from October 2005 until July 2006 at Butajira Hospital. Laboratory tests were performed to confirm TBLN. HIV status was identified in TBLN patients and retrospectively in 1608 healthy individuals. RESULT: A total of 136/161 (84.5%) patients were diagnosed with TBLN by histology. TBLN was culture-confirmed in 107/156 (68.6%) patients. The sensitivity, specificity, positive and negative predictive values of histology were respectively 92.5%, 49%, 79.8% and 75% when compared to culture as gold standard. Patients positive for TBLN by cytology and Ziehl-Neelsen (ZN) were also positive by histology and culture. Among the 143 confirmed TBLN patients, nine (6.3%) were HIV-positive. Of the 1608 healthy individuals, 77 (4.8%) were HIV-positive. Younger age (P=0.0001), female sex (P=0.016), not being married (P=0.0001) and illiteracy (P=0.016) showed a strong association with HIV in healthy individuals. CONCLUSION: Clinical criteria alone over-diagnosed TBLN by 15.4% compared to histological and/or bacteriological results. The HIV prevalence in TBLN patients and healthy individuals was the same.
SETTING: Butajira, Southern Ethiopia. OBJECTIVE: To compare the diagnostic capacity of the clinical criteria for tuberculous lymphadenitis (TBLN) with histological and/or culture results and to assess the association of human immunodeficiency virus (HIV) with tuberculosis (TB) lymphadenitis. DESIGN:Patients (n=171) were included in the study from October 2005 until July 2006 at Butajira Hospital. Laboratory tests were performed to confirm TBLN. HIV status was identified in TBLN patients and retrospectively in 1608 healthy individuals. RESULT: A total of 136/161 (84.5%) patients were diagnosed with TBLN by histology. TBLN was culture-confirmed in 107/156 (68.6%) patients. The sensitivity, specificity, positive and negative predictive values of histology were respectively 92.5%, 49%, 79.8% and 75% when compared to culture as gold standard. Patients positive for TBLN by cytology and Ziehl-Neelsen (ZN) were also positive by histology and culture. Among the 143 confirmed TBLN patients, nine (6.3%) were HIV-positive. Of the 1608 healthy individuals, 77 (4.8%) were HIV-positive. Younger age (P=0.0001), female sex (P=0.016), not being married (P=0.0001) and illiteracy (P=0.016) showed a strong association with HIV in healthy individuals. CONCLUSION: Clinical criteria alone over-diagnosed TBLN by 15.4% compared to histological and/or bacteriological results. The HIV prevalence in TBLN patients and healthy individuals was the same.