Final Diagnosis -- Infection of M. Bovis


The culture results suggested that the patient's infection was due to M. bovis. Since the patient had had BCG therapy for a bladder neoplasm after the implantation of his prosthesis, the most likely cause of the patient's present infection was dissemination of BCG. Primary M. bovis infection, however, is not unheard of and this patient's past medical history included contact with farm animals. Genetic studies (performed at the Wadsworth center of The New York State Department of Health) revealed the isolate to be M. bovis BCG.


BCG is an attenuated strain of Mycobacterium bovis, a member of the Mycobacterium tuberculosis complex. Albert Calmette and Camille Guerin developed BCG from a virulent strain of M. bovis more than 75 years ago as a vaccine for the prevention of tuberculosis [2]. It is now widely used in the treatment of superficial bladder carcinoma [1]. BCG instillation produces chronic granulomatous cystitis. This immunologic response to intravesical instillation is thought to be essential in the anti-tumor efficacy of BCG therapy. When there is no cystitis adequacy of treatment is questionable [5].

Complications from BCG instillation are uncommon but potentially severe and fatal. These include: fever, granulomatous prostatitis, BCG pneumonitis, BCG hepatitis, hematuria, skin rash, skin abscess, ureteral obstruction, epididymo-orchitis, bladder contracture, hypotension and cytopenia [1,5]. However, 95% of patients have no serious side effects [5]. Arthritis complicating BCG instillation is uncommon (incidence of 0.5% to 3%) [1.9] but well recognized [3,4]. It occurs more frequently in patients expressing HLA B27 [4,6].

Systemic BCG infection may result from traumatic catheterization or treatment following extensive tumor resection [1]. BCG related deaths have occurred following immunization for tuberculosis and following intralesional administration in the treatment of malignancy [1]. Intravesical instillation has also been implicated in BCG related deaths with an estimated mortality of 1 per 12,500 patients [5].

Complications arising from intravesical BCG administration respond well to treatment with anti-tuberculous drugs. Having had exposure to farm animals in Greece and household contact with tuberculosis, this patient may have had any of three possible etiological mycobacterial agents: Mycobacterium tuberculosis, wild type M. bovis and BCG. Two recently described experimental PCR based techniques were used to identify our patient's isolate.

The first is a multiplex PCR design based on the RD1 region which has been shown to be present in M. bovis and MTB but not in BCG. This method produces a150 bp amplified fragment in species/strains with the RD1 region but a 200bp fragment in RD1 deleted species/strains. The RD1 region is 9650 bp long and is therefore not amplified by PCR [7]. Our specimen (lane 5) produced a 200-bp fragment.

The other method used is "spoligotyping", or spacer oligotyping, which uses DNA polymorphisms within the chromosomal direct repeat locus of M tuberculosis complex organisms. This region consists of multiple well-conserved 36 bp direct repeat sequences separated by 34 to 41 bp long non-repetitive spacers. The number of direct repeats and presence or absence of certain spacers show strain variation and can identify M. bovis. A single primer pair produces multiple varying sized amplification products that were hybridized to immobilized, labeled oligonucleotides on a membrane (reverse line blot). The immobilized labeled oligonucleotide each corresponded to a specific spacer [8].

M. bovis strains show deletion of the five terminal spacers. Our patient's isolate produced a "spoligotype" that lacked the five terminal spacers and was identical to that of M. bovis BCG strain P3 that was serendipitously run as the BCG control.

This "spoligotyping" method allows for simultaneous detection and strain differentiation of M tuberculosis from clinical samples and can reduce the time for diagnosis and strain differentiation from month(s) to a few days. Previous testing for strain differentiation used DNA-RFLP, but this method is limited by the necessity to culture the organism and is technically demanding.

BCG immunotherapy is the treatment of choice for superficial bladder carcinomas [1,4,5]. Complications arising from its use are numerous and potentially serious.

This case illustrates an example of BCG induced arthritis in a knee prosthesis following immunotherapy for bladder carcinoma. PCR based identification was able to confirm this.


  1. DL Lamm, VD Stogdill, BJ Stogdill and RG. Crispen. Complications of Bacillus Calmette-Guerin immunotherapy in 1278 patients with bladder cancer. J Urol 135:72-274, 1985.
  2. A. Morales. From the 19th to the 21st Centuries: BCG in the Treatment of Superficial Bladder Cancer. Eur Urol, 21(suppl 2):2-6, 1992.
  3. RA Hughes, SA Allard and RN Maini. Arthritis associated with adjuvant mycobacterial treatment for carcinoma of the bladder. Annals of the Rheumatic Diseases, 48:432-434, 1989.
  4. P Goupille, JL Poet, F Jattiot, JP Mattei, V Vedre, I Tonnoli-Serabian, H Roux, JP Valat. Three cases of arthritis after BCG therapy for bladder cancer. Clin Exp Rheumatol 12:195-197, 1994.
  5. DL Lamm, ADPM Vam Der Meijden, A Morales, SA Brosman, WJ Catalona, HW Herr, MS Soloway, A Steg and FMJ Debruyne. Incidence and treatment of complications of bacillus Calmette-Guerin intravesical therapy in superficial bladder cancer. J Urol 147:596-600, 1992.
  6. P Goupille, D Soutif and JP Valat. Arthritis after Calmette-Guerin Bacillus Immunotherapy for Bladder Cancer. J Rheumatol 19:1825-1826, 1992
  7. EA Talbot, DL Williams and R Frothingman. PCR Identification of Mycobacterium bovis BCG. J Clin Microbiol, 35:566-9, 1997.
  8. Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, Kuijper S, Bunschoten A, Molhuizen H, Shaw R, Goyal M, van Embden J. Simultaneous Detection and Strain Differentiation of Mycobacterium tuberculosis for Diagnosis and Epidemiology. J Clin Microbiol 35:907-14, 1997
  9. A. Keat. TB or not TB? That is the question. Br J Rheumatol 32(9):769-71, 1993

Contributed by R Persad, MD, JC Dunn, MD, J Driscoll, MD and W Pasculle, MD


IndexCME Case StudiesFeedbackHome