Sarcoidosis and tuberculosis: epidemiological similarities and dissimilarities. A review of a series of studies in a Japanese work population (1941-1996) and the general population (1959-1984).

AIM

The aim of this review is to discuss the epidemiological relationship between sarcoidosis and tuberculosis.

METHODS

We have used a series of health surveillance data in a Japanese work population of 460,000 employees including 70,000 working in Tokyo (1941-1996) and the data from a nation-wide sarcoidosis survey in the general population (1959-1991). The work population was annually x-rayed and tuberculin tested. The data of primary tuberculosis were obtained from a 17 year cohort study of tuberculin positive converters and primary pulmonary sarcoidosis data from the registry of the disease (1952-1996) in the same work population. Hilar lymphadenopathy (HL) was observed as a common marker of the two diseases.

RESULTS

1. Sarcoidosis HL was not detected in the work population until tuberculosis HL decreased. 2) BHL was rare in primary tuberculosis, but occurred in 95.5% of sarcoidosis subjects. 3) In both diseases, HL resolved in a few years, though accompanying extra-pulmonary involvements delayed the resolution of sarcoidosis BHL. 4) The grade of tuberculin sensitivity prior to sarcoidosis was not a risk factor for developing sarcoidosis. 5) Several well-documented sarcoidosis cases remained tuberculin-negative before and at the time of diagnosis of the disease and after resolution of pulmonary involvement. 6) Age-specific incidence curves showed a mono-modal curve in tuberculosis and a bimodal curve in sarcoidosis. 7) Tuberculosis prevalence was higher in the South of Japan, while sarcoidosis was higher in the North.

CONCLUSIONS

These epidemiological dissimilarities do not support a tuberculosis etiology of sarcoidosis.