MULTIPLE LABORATORY EVALUATION OF THE REITER PROTEIN COMPLEMENT FIXATION METHOD

Duplicate Reiter protein complement fixation tests were carried out on different days at the UCLA laboratory on 636 specimens entailing diagnostic problems. Results were in agreement in 94.2 per cent of cases.A comparison of the results obtained with the first RPCF test carried out on 639 diagnostic problem specimens submitted to four Los Angeles laboratories showed disagreement ranging from 10 to 13 per cent.Agreement between RPCF and TPI results in each of the four laboratories ranged from 81.2 to 82.7 per cent.A greater correlation was obtained between RPCF and TPI results than between TPI and STS or RPCF and STS. In the presence of nonreactive RPCF tests, from 5.6 to 7.1 per cent of the specimens tested in each laboratory were reactive to VDRL and TPI tests. Nonreactive VDRL and reactive TPI results ranged from 7.2 to 8.3 per cent of the specimens tested. In the presence of reactive TPI and RPCF results, 5.5 to 6.4 per cent of the specimens tested in each laboratory showed nonreactive VDRL tests.Twenty-nine or 4.6 per cent of 625 specimens tested showed nonreactive VDRL, reactive cardiolipin Kolmer and reactive TPI tests.Of the weakly reactive RPCF, VDRL or cardiolipin Kolmer tests, a higher percentage of the RPCF results gave reactive TPI tests than did either the VDRL or cardiolipin Kolmer procedures. The number of specimens showing discrepant RPCF results upon being retested in the four laboratories ranged from 33 to 68. Reactive TPI tests on those specimens ranged from 30.3 per cent to 54.4 per cent.