Assessment of the tuberculosis agglutination test

A total of 139 random samples of serum taken at intervals during chemotherapy from 16 patients were analysed using agglutination methods for the serological diagnosis of tuberculosis. The series comprised six cases of far advanced pulmonary tuberculosis, four cases with advanced lesions, two cases with minimal spread and four controls. The samples were analysed simultaneously and independently using two different batches of antigen in two test systems. The test results were comparable in the two series and the diagnostic outcome of the test was not encouraging, the percentages of false negatives being 52 and 60 or 9 and 32, depending on the criteria selected, and those of false positives being 3 and 0 or 59 and 44 respectively. The magnitude of the titres obtained did not reflect the severity of the disease. In addition the test did not prove useful as a method for following up the results of treatment of individual patients in this series.