[Clinical evaluation of a new non-radiometric automatic system for the rapid diagnosis of tuberculosis]

Tuberculosis (TB) is again a public health problem un many countries and is considered a re-emerging disease. The fastest possible diagnosis in our patients is essential for TB control programs. ESP is a non-radioactive, totally automated, continuously monitored system designed to detect mycobacteria.

Clinical evaluation of this new system for the rapid diagnosis of tuberculosis. During 1997 a total of 1,022 clinical sputum specimens were investigated. Specimens were processed in triplicate for ESP, BACTEC 460 TB and Löwenstein-Jensen systems. The validity, isolates of Mycobacterium tuberculosis and time required for detecting M. tuberculosis by the three systems were determined.

The sensitivity, specificity, positive predictive and negative predictive values of the new systems were 98%, 99.8%, 98% and 99.8%, respectively. No significant differences were found between the recovery rates by the three systems. The mean time for detection was 10 days (range: 7-13 days) for specimens with positive bacilloscopy and 14 days (range: 10-28 days) for specimens with negative bacilloscopy. The difference was statistically significant between ESP and Löwenstein-Jensen, but not between ESP and BACTEC.

The new system proved to have an excellent sensitivity and specificity, which along with its total automation renders it a system of great clinical interest for the rapid diagnosis of TB and an alternative method for radiometric systems.