Compairing the diagnostic performance of two smears versus three smears for the dignosis of pulmonary tuberculosis in selected Health Centers of Lusaka District, Zambia
Date
2013-12-02
Authors
Solo, Eddie
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Current World Health Organization (WHO) and International Union Against Tuberculosis and Lung Diseases (IUATLD) guidelines recommend that patients suspected of pulmonary tuberculosis (PTB) should submit three sputum samples as ‘spot’, ‘morning’ and another ‘spot’. In recent years some TB experts have challenged the policy of examining three sputum samples per patient suspected of PTB. Arguments advanced are that these guidelines are based on old studies; increases cost and are not rewarding or cost effective. These experts have therefore suggested reducing the number of sputum samples required to be examined for diagnosis of PTB from the conventional three smears to two, particularly in resource limited settings. WHO has responded to this debate by recommending country specific studies to evaluate the operational effectiveness of the proposed two smears policy. The objective of this study was to compare the performance of the suggested two smears and the conventional three smears for the diagnosis of PTB in selected health centers of Lusaka district.Between January 2011 and May 2012, we conducted a cross sectional study by reviewing routine Zielh Neelsen (ZN) smear results in four urban health centers laboratory registers and at the same time stored their actual sputum samples after routine use. We stratified the three routine smear results to form two strategies, namely; ‘three smear nstrategy’ (by considering results for all the three samples) and ‘two smear strategy’ (by restricting our analysis to the results for the first two samples). The stored specimens were transported to University Teaching Hospital TB laboratory for culture on Lowenstein Jensen (LJ) slopes. LJ culture results acted as gold standard and were used to compute the sensitivity, specificity, Positive nPredictive Value (PPV), Negative Predictive Value (NPV), test efficiency and likelihood ratios for three smears strategy and two smears strategy respectively. We then compared the performances of the two strategies by testing the differences in the above variables using Yates’s corrected Chi-square test. A result yielding a p value of p<0.05 was statistically significant.
Out of 1030 TB suspects analyzed, 350 were positive on LJ cultures providing a positivity rate of 34.0% (95% CI: 31.1%, 36.9%).The sensitivity for three smears strategy was 61.4% (95% CI: 56.1%, 66.5%), whereas that for two smears strategy was 57.7% (95% CI: 52.5%, 62.9%), p = 0.355. The specificity for both three smear and two smear strategies was 98.1% (95% CI: 97.1%, 98.9%). The test efficiency for three smears strategy was 85.6% (95% CI: 83.4%, 87.7%). while that of two smears was 84.4% (95% CI: 82.2%, 86.6%), p = 0.459. The positive likelihood ratio for three smears strategy was 32.3 (95% CI: 18.6, 55.4) and for two smears strategy was 30.3 (95% CI: 17.5, 52.1), P = 0.96. The negative likelihood ratio for three smears strategy was 0.39 (95%CI: 0.34, 0.44) whereas for two smears strategy was 0.43 (95% CI: 0.38, 0.48), p = 0.882. The p-values for all the variables tested showed no significant differences between the two strategies.The performances of the two strategies were similar therefore this study concludes that the two smears strategy for diagnosis of pulmonary tuberculosis is adopted for Zambia.
Description
Keywords
Tuberculosis