Comparison of in house polymerase chain reaction with conventional techniques for the detection of Mycobacterium tuberculosis DNA in granulomatous lymphadenopathy
- Krishna K Singh1,
- M Muralidhar2,
- Arvind Kumar2,
- T K Chattopadhyaya5,
- Kusum Kapila3,
- Manoj K Singh3,
- S K Sharma4,
- N K Jain6,
- Jaya S Tyagi1
- 1Department of Biotechnology, All India Institute of Medical Sciences, New Delhi 110029, India
- 2Department of Surgical Disciplines, All India Institute of Medical Sciences
- 3Department of Pathology, All India Institute of Medical Sciences
- 4Department of Medicine, All India Institute of Medical Sciences
- 5Department of Gastrointestinal Surgery, All India Institute of Medical Sciences
- 6New Delhi TB Centre, New Delhi, India
- Dr Tyagi email: jstyagi{at}hotmail.com;jst{at}aiims.ernet.in
- Accepted 13 September 1999
Abstract
Aims—To evaluate the usefulness of the devR based polymerase chain reaction (PCR) in the detection of Mycobacterium tuberculosis in lymph node aspirates and tissues of lymphadenitis and to compare PCR with conventional diagnostic techniques.
Subjects and methods—Coded specimens of fine needle aspirates and biopsies from 22 patients with tuberculous lymphadenitis, 14 patients with non-tubercular lymphadenitis, and nine patients with granulomatous lymphadenitis were processed and subjected to analysis by PCR, smear microscopy, M tuberculosis culture, histology, and cytology.
Results—Tuberculous lymphadenitis was correctly diagnosed by PCR in 18 patients, by culture in five patients, by histology in 13 patients, and by cytology in seven patients. PCR gave two false positive results in 14 patients with non-tubercular lymphadenitis. The sensitivity of the conventional techniques was significantly higher with biopsies (17 of 22 specimens; 77%) than with fine needle aspirates (nine of 22 specimens; 41%). However, the sensitivity of PCR was not significantly higher with biopsies (68%) in comparison with fine needle aspirates (55%). The sensitivity of either biopsy PCR or fine needle aspirate PCR was not significantly different from that of either histology combined with culture or cytology combined with culture. The overall combined specificity of PCR was 86%. Mycobacterium tuberculosis DNA was detected in six of nine patients with granulomatous lymphadenitis.
Conclusion—PCR is the most sensitive single technique available to date for the demonstration of M tuberculosis in specimens derived from patients with a clinical suspicion of tuberculous lymphadenitis. The value of PCR lies in its use as an adjunct test in the diagnosis of tuberculous lymphadenitis, particularly in those patients where conventional methods fail. Because fine needle aspiration is not an invasive procedure, it is the procedure of choice, and PCR should be performed initially on these samples. Excisional biopsy histology and PCR should be recommended only for patients in whom fine needle aspirate PCR is negative or when there is discrepancy with the clinical impression.
