Potential Usefulness of a Combination of Inflammatory Markers in Identifying Patients With Sarcoidosis and Monitoring Respiratory Functional Worsening

To the Editor We read with interest the article that appeared in the Journal by De Smet and colleagues,1 who applied likelihood ratios (rather than a dichotomous interpretation) for different results intervals for 3 laboratory tests (the percentage of bronchoalveolar lavage fluid [BALF] lymphocytes, the BALF CD4/CD8 ratio, and the serum angiotensin-converting enzyme activity) to improve clinical interpretation for the diagnosis of pulmonary sarcoidosis. They also examined the allocation power of the 3 diagnostic tests combined together by means of logistic regression analysis. The authors retrospectively evaluated the data for 153 subjects (36 with diagnosed sarcoidosis and 117 control subjects with clinical suspicion of sarcoidosis, but with diagnosis of other pulmonary diseases). The combination of the 3 laboratory tests allowed exclusion of sarcoidosis in 57 (48.7%) of 117 control subjects and confirmation of the diagnosis in 12 (33%) of 36 cases of pulmonary sarcoidosis.1 We strongly agree with the authors that the common practice of applying diagnostic tests by using strict cutoff values may result in a loss of …