Serum lysozyme is used as a marker of sarcoidosis disease activity. In this study we examined the association between lysozyme levels and the clinical features of sarcoidosis and thus the clinical usability of this parameter in a large population. One hundred ten sarcoidosis patients from central Japan were examined for clinical features and serum lysozyme level at the first visit to our hospital and on a regular basis thereafter. The sensitivity of lysozyme for predicting sarcoidosis was 79.1%, whereas that of serum angiotensin-converting enzyme (ACE) was 59.0%. Even in the cases without an elevated serum ACE level, a value of 72.1% was obtained. The serum lysozyme level demonstrated a significant tendency to increase with the number of organs involved (p < 0.01). There were significant differences among the four radiographic stages (p < 0. 05). The maximum serum lysozyme levels of patients without a disappearance of abnormal shadows on chest radiography within 5 years were significantly greater than those of individuals with a disappearance (p < 0.05). A positive correlation between serum lysozyme and serum ACE levels was observed. Because serum lysozyme is much less specific for sarcoidosis than serum ACE, its diagnostic value may be limited. However, the sensitivity was high even when serum ACE levels were within normal limits and correlated well with clinical features in sarcoidosis. Therefore, this parameter seems suitable for disease monitoring in proven cases.