Efficacy of the conventional diagnostic approach to pulmonary tuberculosis.

OBJECTIVE The aim of this study was to evaluate the efficacy of the current clinical approach to the diagnosis of culture positive pulmonary tuberculosis (PTB) in hospitalised adult patients. METHOD We examined the case records and chest X-rays (CXR) of 199 adult patients with culture positive PTB diagnosed from 1993 to 1995. Patients were divided into two groups: early treatment (ET) and delayed treatment (DT) of PTB. DT were patients who received treatment only after respiratory specimens had returned with positive culture results. We also compared the CXR of the DT group with a group of age, sex matched patients who did not have PTB (control). Usual CXR pattern for PTB was defined as: upper lobe acinar shadows, upper lobe cavitation or miliary pattern. RESULTS There were 199 patients with a mean age of 56 (19). One patient was HIV antibody positive and 27% were diabetics. There were 143 (72%) patients in the ET group and 56 (28%) in the DT group. The ET group was significantly younger and more likely to show usual features and cavitary disease on the CXR than the DT group. The diabetic patients had significantly more frequent cavitation than non-diabetics. When compared to a control group without PTB, the DT group was significantly more likely to show usual CXR pattern and less likely to have a clear CXR. CONCLUSIONS The conventional approach to the diagnosis of PTB is reasonably accurate and efficient enough for the majority of patients. The delay in the diagnosis and treatment of PTB was more common among elderly patients, patients with negative smear results and atypical CXR features. If PTB is suspected in elderly patients, an early decision should be made between empiric treatment and further diagnostic testing.