Sputum smear negative pulmonary tuberculosis: sensitivity and specificity of diagnostic algorithm

Area of Science:

• Medical Research
• Infectious Diseases
• Public Health

Background:

• Diagnosing pulmonary tuberculosis in Human Immunodeficiency Virus (HIV) patients is challenging due to a high prevalence of sputum smear-negative cases.
• Current diagnostic approaches rely on algorithms involving symptoms, signs, and laboratory results.

Purpose of the Study:

• To evaluate the sensitivity and specificity of the National Tuberculosis and Leprosy Programme's treatment algorithm for diagnosing sputum smear-negative pulmonary tuberculosis.

Main Methods:

• A prospective cross-sectional study in Dar-es-Salaam, Tanzania, enrolled 467 patients.
• Sputum cultures, HIV testing, chest X-rays, Mantoux tests, and Full Blood Pictures were performed.
• Sensitivity, specificity, and predictors of culture-positive results were analyzed.

Main Results:

• The algorithm demonstrated a sensitivity of 38.1% and specificity of 74.5%.
• Of 127 sputum culture-positive patients, 48.1% were missed by the algorithm.
• 37.4% of sputum culture-negative patients were incorrectly treated.

Conclusions:

• The current algorithm is insufficiently sensitive and specific for diagnosing smear-negative pulmonary tuberculosis.
• The algorithm leads to both under-treatment of active tuberculosis and over-treatment of non-tuberculous conditions.