Abstract
Introduction Diagnosis in of pulmonary tuberculosis (PTB) in resource-limited settings relies heavily on detection of bacilli on sputum smears however HIV positive patients often have smear-negative disease. A new WHO diagnostic algorithm was developed to improve PTB diagnosis in HIV-prevalent resource-limited settings (integrated, expedited pathway with 2 rather than 3 smears in each set and no diagnostic antibiotic trial). We evaluated performance under operational conditions in a Ugandan district hospital.
Methods A prospective observational cohort study design was used. The existing “UgWHO03” and new “WHO07” diagnostic algorithms were evaluated over the first and second phases of the study respectively. New HIV patients were screened for PTB by a history of two or more weeks of cough. Diagnosis by algorithm was compared to gold standard mycobacterial culture.
Results 90/147 and 132/166 of patients enrolled onto “UgWHO03” and “WHO07” respectively and were able to provide adequate sputum for culture. For UgWHO03: five cases of smear negative PTB and four cases of smear positive PTB diagnosed by algorithm and three cases by culture. For WHO07: 18 cases of smear negative PTB and seven cases of smear positive PTB diagnosed by algorithm and 1 case by culture. Sensitivity for smear-negative PTB increased non-significantly from 62.5% to 94.7% (p=0.065), specificity was maintained (99% vs 98%, not significant).
Conclusion Sample size was limited by the ability to provide sputum but there was a non-significant trend towards increased sensitivity which approached significance. This suggests that in the context of a more rapid pathway this algorithm may be beneficial.
Methods A prospective observational cohort study design was used. The existing “UgWHO03” and new “WHO07” diagnostic algorithms were evaluated over the first and second phases of the study respectively. New HIV patients were screened for PTB by a history of two or more weeks of cough. Diagnosis by algorithm was compared to gold standard mycobacterial culture.
Results 90/147 and 132/166 of patients enrolled onto “UgWHO03” and “WHO07” respectively and were able to provide adequate sputum for culture. For UgWHO03: five cases of smear negative PTB and four cases of smear positive PTB diagnosed by algorithm and three cases by culture. For WHO07: 18 cases of smear negative PTB and seven cases of smear positive PTB diagnosed by algorithm and 1 case by culture. Sensitivity for smear-negative PTB increased non-significantly from 62.5% to 94.7% (p=0.065), specificity was maintained (99% vs 98%, not significant).
Conclusion Sample size was limited by the ability to provide sputum but there was a non-significant trend towards increased sensitivity which approached significance. This suggests that in the context of a more rapid pathway this algorithm may be beneficial.
Original language | English |
---|---|
Pages (from-to) | A354-A355 |
Number of pages | 2 |
Journal | Journal of Epidemiology and Community Health |
Volume | 65 |
Issue number | Suppl. 1 |
DOIs | |
Publication status | Published - Aug 2011 |