Detection of Mycobacterium tuberculosis Complex in Sputum Specimens by the Automated Roche Cobas Amplicor Mycobacterium Tuberculosis Test

Abstract

Three hundred twenty-four sputum specimens from 151 patients with suspected active pulmonary tuberculosis were tested for the presence of the Mycobacterium tuberculosis complex with auramine fluorochrome stain and automated PCR assay (Roche Cobas Amplicor Mycobacterium Tuberculosis Test [MTB]). The results were compared with those of the conventional Löwenstein-Jensen tube culture and the BACTEC radiometer liquid culture. A total of 76 specimens from 32 patients were culture positive for M. tuberculosis. In addition, 37 specimens from 15 patients were smear and culture positive for other Mycobacterium species but negative by the present nucleic acid amplification method and thus were not included in the comparison. Compared with culture, the sensitivities, specificities, and positive and negative predictive values for acid-fast smear were 67, 98, 93, and 91% and those for the Cobas Amplicor MTB were 83, 99, 97, and 95%, respectively. When three consecutive sputum specimens per patient could be obtained, the sensitivity of the Cobas Amplicor MTB improved to 91%, whereas the sensitivity of the acid-fast smear remained unchanged.

Automatized PCR evaluation ofMycobacterium tuberculosiscomplex in respiratory and nonrespiratory specimens

“…Although the COBAS AMPLICOR MTB test is designed to detect M. tuberculosis in respiratory specimens, several studies evaluated the diagnostic value of this test in nonrespiratory samples. These studies stated that results obtained by the PCR with nonrespiratory specimens were in good agreement with those obtained with respiratory specimens (Bodmer et al, 1997;Rajalahti et al, 1998;Reischl et al, 1998;Shah et al, 1998). In our study, the sensitivities of PCR in respiratory and nonrespiratory specimens were 72.7% and 60%, respectively.…”

“…In a number of studies it was stressed that testing with multiple specimens from one patient had increased the sensitivity of the COBAS AMPLICOR MTB PCR assay in smear-negative tuberculosis (Wobeser et al, 1996;Bodmer et al, 1997;Yuen et al, 1997;Rajalahti et al, 1998;Lim et al, 2000;Johnsson & Ridell, 2003;Levidiotou et al, 2003;Kim et al, 2004). By contrast, in smear-positive specimens the rapid diagnosis and/or confirmation of the clinical diagnosis and differentiation between M. tuberculosis and nontuberculous mycobacteria is emphasized (Yuen et al, 1997;Eing et al, 1998).…”

Pulmonary tuberculosis in the West Bank, Palestinian Authority: molecular diagnostic approach

“…Detection of the bacteria among patients with a history of TB and patients having treatment because of active tuberculosis has been described previously (Rajalahti et al 1998). Therefore, the detection of MTB DNA by PCR in sputum smears of three patients 3-13 months after infection and treatment is consistent with previous findings, which showed that non-viable bacteria could be detected by PCR (Rajalahti et al 1998;Lima et al 2008). Thus, it is important to carefully evaluate the medical history of the patient, during the diagnosis procedure especially using the sensitive PCR assay.…”

“…To overcome the low sensitivity from sputum smear-negative samples, a nested PCR amplification has been used to enhance the DNA amplicon and found to be sufficient (Garcia-Quintanilla et al 2000). Detection of the bacteria among patients with a history of TB and patients having treatment because of active tuberculosis has been described previously (Rajalahti et al 1998). Therefore, the detection of MTB DNA by PCR in sputum smears of three patients 3-13 months after infection and treatment is consistent with previous findings, which showed that non-viable bacteria could be detected by PCR (Rajalahti et al 1998;Lima et al 2008).…”

Molecular genetic methods for diagnosis and antibiotic resistance detection of mycobacteria from clinical specimens

“…However, the assay was rather disappointing when tested by different investigators with nonrespiratory specimens such as pleural effusions, cerebral spinal fluids and gastric aspirates (29–34), and smear‐negative respiratory samples. The reported sensitivity and specificity values after resolving discordant results (discrepant analysis) ranged from 79.4–91.9% and 85.7–100% for respiratory specimens and 27.3–98.6% and 85.7–100% for nonrespiratory specimens, respectively (1, 35–50).…”