|Home | About | Journals | Submit | Contact Us | Français|
The susceptibility to mycobacterial infections of immunosuppressed subjects, for example, those treated actively for rheumatic diseases, is widely recognized.1,2 However, the co‐existence of mycobacterial infection (caused either by Mycobacterium tuberculosis or other mycobacterial infections) and polymyositis (PM) has been reported only in small series of patients 2,3,4,5 predominantly from countries where tuberculosis (TB) is endemic 2,3 or in immigrants in a Western country.4,5
We assessed the frequency of mycobacterial infections in patients with PM and dermatomyositis (DM) in Finland, a non‐endemic country for TB that has kept detailed registry for TB. All patients with an ICD‐8 code for PM (716.00) and DM (716.10) who had been discharged from any Finnish hospital during the years 1969 to 1985 were identified by using the information from the National Discharge Registry. During this period, diagnostic studies for PM and DM were usually performed on inpatients. Therefore, the present series of patients represents the vast majority of PM and DM cases in Finland during this period.
We found 176 patients with PM and 72 with DM who fulfilled the diagnostic criteria of Bohan and Peter. 6,7 We followed their progress by studying the records until they died or until the end of August 1995. At the time of the diagnosis of myositis, 26 (10%; 19 with PM and 7 with DM) of the whole series of 248 patients had a history of TB. In addition, 13 out of 176 patients (7.4%) with PM and 2 out of 72 patients (2.8%) with DM were documented to have a mycobacterial infection after a median time (range) of 20 (3–111) months from the diagnosis of myositis. Only one of them (who had PM) was in the group with a history of TB (57 years earlier).
Cumulative probability of mycobacterial infections among patients with PM was 4.8% (95% CI 2.4 to 9.4%) in 5 years and 9.3% (95% CI 5.4 to 15.6%) in 10 years (fig 11).). On the basis of acid fast stain and mycobacterial culture, we determined that 11 out of the 15 mycobacterial infections were caused by M. tuberculosis (nine of them in PM patients) and four by other mycobacteria (M. avium intracellulare or mycobacteria group IV). At the time of the diagnosis of the mycobacterial infection, all the patients used corticosteroids, and seven PM patients also used or had used azathioprine. The risk factors for developing mycobacterial infections did not differ between patients with and without the infection.
Altogether, in 40 out of 248 (16%) patients with myositis — in 31 out of 176 (18%) with PM and 9 out of 72 (13%) with DM — there was a former or diagnosed mycobacterial infection noticed during the follow up. To the best of our knowledge, the present study is the first one reporting a high frequency of mycobacterial infections in a large, systematically collected population of subjects with PM and DM in a Western non‐endemic area for TB .
It may be supposed that, at least in part, the high incidence of mycobacterial infections during the treatment of myositis is caused by the reactivation of a former infection. The association between mycobacterial infections and PM is interesting, especially if there was a latent infection present in most of the cases, although this cannot be confirmed from their records. A weak causal relationship with ‘reactive myositis' caused by a former or latent mycobacterial infection cannot be ruled out.
We conclude that the possibility of mycobacterial infection has to be kept in mind when diagnosing and treating patients with myositis.