Panniculitis

To the Editor.-Titov et al1 used Mycobacterium tuberculosis MPB64 genespecific primers to amplify M tuberculosis genetic sequences in joint and bone tissues from their 14 patients with bone and joint disorders. Conventional polymerase chain reaction (PCR) on the arthroscopically obtained tissue biopsy tissues was diagnostic and resulted in only 2 false positives among patients who had suffered from tuberculosis in the past. Nevertheless, to achieve maximum PCR diagnostic utility for clinicians, it would be imperative to carry out extended investigations among those who had been offered bacille Calmette-Guérin (BCG) vaccine either as a prophylactic or a therapeutic agent. A significant proportion of false positives would vitiate the excellent data obtained by Titov et al.1

An in situ rather than conventional PCR on arthroscopically obtained bone or joint tissues would be intriguing and more informative. That was illustrated with the fixed lung cancer tissues at the First Military Medical University, Guangzhou, China. A sensitive and specific indirect in situ nested PCR was used to identify and localize tubercular DNA in 15 formalinfixed, paraffin-embedded lung cancer tissue specimens, which had been demonstrated to be positive for M tuberculosis DNA by the conventional PCR. Positive, brown granules of M tuberculosis DNA were found mainly in the cytoplasm of the alveolar epithelial cells, pulmonary macrophages, inflammatory cells, and a few tumor cells within lung cancer tissues.2

Funds should be allocated to standardize an in situ PCR format for M tuberculosis on different surgically excised or aspirated tissues. Last but not least, the magnitude of any false positives attributable to BCG or other environmental mycobacterium species should be evaluated in different geographic areas. Mycobacterium ulcerans causing Buruli ulcer is an environmental mycobacterium responsible for an infectious necrotizing panniculitis.3 This disabling disease is strongly linked to the aquatic ecosystem. Occurring mainly in children, it is an emergent public health threat in many humid rural tropical areas. Certainly, it would be important to assess any M tuberculosis PCR false positives on arthroscopically obtained bone and joint tissues.1

SUBHASH C. ARYA, PhD

Department of Clinical Microbiology

NIRMALA AGARWAL, FRCOG

Department of Gynecology and Obstetrics

SHEKHAR AGARWAL, MCH, ORTH

Department of Orthopedics

Sant Parmanand Hospital

Delhi, India 110054

1. Titov AC, Vyshnevskaya EB, Mazurenko SI, et al. Use of polymerase chain reaction to diagnose tuberculous arthritis from joint tissues and synovial fluid. Arch Pathol Lab Med 2004;128:205-209.

2. Song LY, Van WS, Zhao T. Detection of Mycobacterium tuberculosis in lung cancer tissue by indirect in situ nested PCR [in Chinese]. Di Yi Jun Yi Da Xue Xue Bao. 2002;22:992-993.

3. Darie H. Mycobacterium ulcerans infection: epidemiological, clinical and therapeutical aspects [in French]. Bull Soc Pathol Exot. 2003:96:368-371.

In Reply.-Bacillus Calmette-Guérin (BCG) is a live, attenuated Mycobacterium tuberculosis strain, which remains latent in vaccinated individuals. It may later become activated and is a known cause of, for example, osteomyelitis in children.1 Presence of this strain in damaged tissue often means that it has been disseminated and has become a causative agent of the disease.2 The Department of Paediatric Bone and Joint Tuberculosis of the St Petersburg Institute of Physiopulmonology specializes in the treatment of tuberculosis caused by Mycobncteruim bovis BCC. All polymerase chain reaction (PCR)-positive specimens from children are also tested with BCG-specific primers1 to verify the diagnosis. Results have shown that about 16% of tuberculotic osteomyelitis is caused by the BCG strain. False-positive PCR of biopsy specimens from children with nontuberculotic diseases was observed in no more than 1% of cases. Therefore, it can be concluded that M bavis BCG may be a cause of clinically evident tuberculosis.

In principle, M bavis BCG might lead to false-positive reactions in PCR-based diagnosis of tuberculosis. Primers used for PCR diagnosis of M tuberculosis are often designed so that they only recognize the wild-type M tuberculosis, but do not recognize the attenuated strain used in BCG vaccination. In practical work, conventional PCR primers have been found to be good and relatively specific for the diagnosis of tuberculosis. Such primers are used, for example, in the wellknown PCR test Amplicor MTB (Hoffman La Roche, Basel, Switzerland), in which the primers recognize all species of the M tuberculosis complex, including M bovis BCG. The sensitivity and specificity of the Amplitude kit (Litech, Moscow, Russia) used for PCR diagnosis of M tuberculosis in our work are close to those of Amplicor MTB. This kit has been licensed for commercial use in Russia. In accordance with the rules of the Russian Ministry of Health, before a license is granted for any diagnostic test, several independent experts must validate the test using hundreds of known positive and negative samples. The proportion of false positives among patients with nontuberculotic diseases in these tests did not exceed 0.5%.4 As almost the whole population of Russia has been vaccinated with BCG, these test results clearly validate the use of this test.

ALEXEI G. TITOV, MD

ELENA B. VYSHNEVSKAYA, PhD

SERGEI I. MAZURENKO, MD

Saint-Petersburg State Scientific Research Institute of Phthisiopulmonology

Saint-Petersburg, Russia

SEPPO SANTAVIRTA, MD

Department of Orthopaedics and Traumatology

Helsinki University Central Hospital

Helsinki, Finland

YRJÖ T. KONTTINEN, MD

Department of Medicine

ORTON Orthopaedic Hospital of the Invalid Foundation

Helsinki, Finland and

COXA Joint Replacement Hospital

Tampere, Finland

1. Aljada IS, Crane JK, Corriere N, Wagle DG. Mycobacterium bovis BCG causing vertebral osteomyelitis (Pott's disease) following intravesical BCG therapy. ; din Microbiol. 1999;37:2106-2108.

2. Lotte A, Wasz-Hockert O, Poisson N, et al. BCG complications: estimates for the risk among vaccinated subjects and statistical analysis of their main characteristics. Adv Tuherc Res. 1984:21:107193.

3. Talbot EA, Williams DL, Frothingham R. PCR identification of Mycobacterium bovis BCG. J Clin Microbiol. 1997;35:566-569.

4. Bochkarev E, Denisova T, Generozov E, et al. The Genetic Diagnosis in Phthisiatry. Moscow, Russia: 2000.

Copyright College of American Pathologists Dec 2004
Provided by ProQuest Information and Learning Company. All rights Reserved