Evaluation of the Middlebrook-Dubos Hemagglutination Test in Tuberculosis

¹ The Research and Development Branch, Fitzsimons Army Hospital, Denver 8, Colorado.

The Middlebrook-Dubos hemagglutination test for the diagnosis of tuberculosis was performed on serum samples from 172 patients with active tuberculosis, 257 patients with various non-tuberculous diseases, 54 apparently healthy obstetrical and post-partum cases, and 80 apparently healthy individuals.

1) As performed in this laboratory, the hemagglutination test was neither sufficiently specific nor sufficiently sensitive to warrant its use as a routine diagnostic tool.

2) A definite cross reaction of transient nature was observed with pregnancy.

3) False positive reactions were obtained with approximately one fourth of the serum samples from patients with non-tuberculous disease and persons who were apparently healthy at the time of testing.

4) Only 30 per cent of the cases of minimal tuberculosis studied presented titers in the positive range.

5) The test appeared to be of little prognostic value.

6) Increases in serum hemagglutinins in apparently healthy laboratory workers were apparent two to five months after the individual demonstrated sensitivity to tuberculin.