SUMMARY To elucidate inconsistencies in rubella haemagglutinin assays the components of the assay technique were examined. The results of carefully controlled assays of rubella haemagglutinin antigens from different sources ih various plates and diluents with four species of indicator cells are reported. The quality and quantity of gelatin in the dextrose-gelatin-veronal buffered diluent commonly used in rubella haemagglutinin assays had a profound effect on the haemagglutination pattern and antigen titre. The veronal buffered saline used in the complement fixation test offered a valid alternative to the more complex diluents incorporating gelatin currently used in rubella haemagglutinin assays and haemagglutination inhibition tests.Notwithstanding the recent application of passive haemadsorption, immune haemolysis, and immunosorbent techniques to rubella serology, haemagglutination inhibition (HI) remains a widely used procedure in diagnosing rubella and establishing an individual's immunity to the disease. In a limited collaborative study to define the parameters of the rubella HI technique with a view to establishing the procedure of choice, we were surprised to note wide discrepancies in the rubella antigen titre reported by several laboratories following a common protocol. Subsequently, a number of rubella antigens prepared and assayed in the division gave unexpectedly low titres and unstable agglutination patterns; moreover, inconsistencies were found in consecutive titrations of each of the antigens and in simultaneous titrations of an antigen in different batches of dextrose-gelatin-veronal (DGV) buffer supplemented with 02% bovine albumin (DGVA). These antigens gave good results in the radial haemolysis test and were of satisfactory potency in the HI test, though the antigen back titration did not always reflect the haemagglutinin (HA) dose used in the test.In preliminary experiments to elucidate the anomalous HA titrations the effects of plates, diluents, and erythrocyte species on assays of freeze Accepted for publication 17 October 1984 dried antigens from different sources were studied. The investigation was extended in two further experiments in which precautions were taken to exclude subjective bias. The observation that a particular batch of DGV buffer from a commercial source gave consistently satisfactory HA titration results led us to study the effects of the components of DGV buffer on the haemagglutination pattern. The results of all these experiments are reported below. Throughout the investigation the uniformly satisfactory HA titrations in complement fixation test diluent + 02% bovine albumin (CFTA) led us to recommend CFTA as an alternative to DGVA by virtue of consistent performance, ready availability, and low cost. Material and methods SERATwenty seven serum samples were tested for HI antibody. Six of the samples were believed to be without antibody to rubella virus; 21 were shown to have rubella antihaemagglutinin by a sensitive (overnight) rubella HI technique and four were from pat...

No abstract