Correspondence to: Satoshi Mitarai, MD, PhD, c/o Clinical Research Division, The Research Institute of Tuberculosis, 3-1-24 Matsuyama, Kiyose, Tokyo 204-8533, Japan. Tel.: #81 424 93 3090; E-mail: mitarai@jata.or.jp Accepted: 23 August 2001 Tuberculosis (2001) 81 (5/6), 319}325 ^ 2001 Harcourt Publishers Ltd doi: 10.1054/tube.2001.0305, available online at http://www.idealibrary.com.on Clinical evaluation of Amplicor Mycobacterium detection system for the diagnosis of pulmonary mycobacterial infection using sputum S. Mitarai, A. Kurashima, A. Tamura, H. Nagai, H. Shishido Department of Respiratory Medicine, National Tokyo Hospital, 3-1-1 Takeoka, Kiyose, Tokyo, 204-0023, Japan Summary Setting: The Amplicor Mycobacterium detection kit was evaluated for the diagnosis of active pulmonary mycobacterial infection using sputum. Objective: To assess the clinical usefulness of the Amplicor Mycobacterium kit for the diagnosis of pulmonary tuberculosis and non-tuberculous mycobacterial infection in the country of medium prevalence. Design: All the patients were diagnosed with bacterial, histopathological, and clinical ‘gold standard’. The sensitivity and specificity for diagnosing clinically active pulmonary tuberculosis and Mycobacterium avium and Mycobacterium intracellulare infections were evaluated comparing Amplicor results and clinical diagnosis. Results: A total of 1088 sputum specimens were collected from 780 in and out patients. Mycobacteria were recovered from 339 specimens by culture. The sensitivity and specificity of conventional culture method for the diagnosis of pulmonary tuberculosis were 60.2% and 99.8% respectively based on the number of patients. The figures for Amplicor were 61.8% and 97.4% respectively. There was no statistical significant difference between these methods. In rapidity, the Amplicor was significantly superior to the microscopy method in sensitivity. Conclusion: Patients with Amplicor positive and conventional negative result had mostly mycobacteria related diseases. The Amplicor positive result indicated mostly active mycobacterial infection and was clinically useful for rapid diagnosis. ^ 2001 Harcourt Publishers Ltd INTRODUCTION The polymerase chain reaction (PCR) ampli9cation method for the detection of mycobacteria has recently been utilized in the diagnosis of Mycobacterium tubercu- losis infection. 1}3 Some researchers reported a higher sen- sitivity and speci9city of the PCR diagnosing method compared with conventional smear and culture. 4}7 The former method, however, has not been fully evaluated or established although many clinical specimens are being tested in clinical laboratories using this method. The Amplicor Mycobacterium detection kit is one of the popular genetic diagnostic tools for the diagnosis of mycobacterial diseases. It is currently being used in most areas of the world. Studies to evaluate its performance have been done by several researchers using various ex- perimental conditions. Bergmann et al. examined several types of specimens (i.e. sputum, bronchoalveolar lavages, tracheal aspirates) using this method. 8 Kajiki et al. also evaluated the Amplicor by examining 1515 sputum speci- mens. Both the conventional and PCR positive rate were shown to be low using these methods. 9 The patients enrolled in their study had early-stage tuberculosis or non-mycobacterial disease. Wobeser et al. reported a low rate of M. tuberculosis isolation among patients with posit- ive mycobacterial cultures, while that of non-tuberculous mycobacteria (NTM) was higher considering the normal population. 10 The present study was designed to evaluate the useful- ness of the Amplicor kit for the diagnosis of active pul- monary tuberculosis and non-tuberculous mycobacterial 319