Clinical Predictors and Accuracy of Empiric Tuberculosis Treatment among Sputum Smear-Negative HIV-Infected Adult TB Suspects in Uganda

• DOI: 10.1371/journal.pone.0074023
• Publication Date: 2013-09-06T21:11:30Z
• Authors: Lydia Nakiyingi; John Mark Bwanika; Bruce Kirenga; Damalie Nakanjako; Catherine Katabira; Gloria Lubega; Joseph Sempa; Barnabas Nyesiga; Heidi Albert; Yukari C. Manabe
• Keywords: Empiric treatment; AIDS-Related Opportunistic Infections

Introduction The existing diagnostic algorithms for sputum smear-negative tuberculosis (TB) are complicated, time-consuming, and often difficult to implement. decision initiate TB treatment in resource-limited countries is largely based on clinical predictors. We sought determine the predictors accuracy of empiric initiation HIV-infected suspects using culture as a reference standard. Setting Out-patient HIV-TB integrated urban clinic Kampala, Uganda. Methods were screened smear microscopy, mycobacterial liquid solid cultures performed. Smear results made available clinician who patients. Clinic records reviewed patients whose smears negative collect data socio-demographics, symptomatology, chest X-ray findings, CD4 cell counts initiation. Results Of 253 suspects, 56% (142/253) females, median age 38 IQR (31–44) years, with count 291 (150–482) cells/mm3. 85 (33.6%) empirically initiated treatment, 35.3% (n = 30) positive compared only 18 (10.7%) 168 untreated (p<0.001). Abnormal [aOR 10.18, 95% CI (3.14–33.00), p<0.001] advanced HIV stage 3.92, (1.20–12.85), p 0.024] significantly associated sensitivity specificity diagnosis after microscopy was 62.5% 73.7% respectively. Conclusion In settings, clinically abnormal predict Empiric correlates poorly cultures. Affordable, accurate rapid point-of-care diagnostics needed settings more accurately which have TB.