Simon and associates [1] state that their findings "militate strongly" against a role for the immune system in the pathogenesis of multiple chemical sensitivity. Although we commend the authors for the intent and design of their study, we believe that their laboratory data do not militate for or against this conclusion.

First, the use of the term "titers" in the abstract is misleading because only the presence or absence of the five autoantibodies was reported. The importance of these antibodies cannot be assessed without full titration of positive specimens.

Second, fluorescence microscopy, used for detection of surface antigens on lymphocytes, is imprecise and inaccurate compared with flow cytometry. The latter is much more sensitive and objective than the naked eye and can count many more cells (typically 2500 to 10 000 per sample, compared with a few hundred at most by microscopy). The use of flow cytometry is especially important for measurement of scarce, dimly fluorescent cells such as peripheral blood lymphocytes expressing CD25 or CD26. For these reasons, the phenotypic analyses presented are not informative.

Third, the study of immune measures is in no way exhaustive. It remains possible that other unexamined antibodies, lymphocyte populations, or immunologic functions (which can be abnormal without changes in cell numbers) were altered. The authors recognize this, but their argument that the battery of tests "included all those recommended for evaluation of chemical sensitivity by a laboratory prominent in this area" is unconvincing, given the anonymity of this laboratory and the limitations of the tests done. Finally, the article lacks important technical information regarding methods and controls, and mis-states cell concentrations in the peripheral blood.

The authors may be correct that the immune system does not play a role in multiple chemical sensitivity, but the laboratory methods used in this study lack the power to test this hypothesis. What the article does support unequivocally is the need for appropriate laboratory procedures and more critical review of articles submitted for publication.