A multicentre quality assessment study to monitor the performance of HIV-1 PCR

Eleven German laboratories and one Swiss laboratory initiated a quality assessment study to evaluate the specificity and sensitivity of their polymerase chain reaction;(PCR) for detection of HIV-1 DNA. Following its own PCR protocols, each laboratory tested a panel of ten coded samples consisting of cell pellets containing 0, 0.1, 1, 10, 10(2), 10(3) and 10(4) ACH-2 cells per 1.5 x 10(5) uninfected peripheral blood mononuclear cells. Of the twelve participating laboratories, three reported correct results for the dilution series as well as for uninfected specimens. One or more classification errors were recorded for 12% of the samples for which the diagnosis was expected to be positive or negative. Samples containing 10 copies of the target template were correctly reported by eleven of the twelve participants. The average sensitivity was 97%. The results of the study revealed no significant differences between the Amplicor kit and in-house procedures. Most of the classification errors occurred in specimens from HIV-negative samples. Out of 36 negative samples, 5 were reported false positive, showing that contamination remains a problem for some laboratories, regardless of the PCR test performed. Careful laboratory techniques and internal as well as external quality control procedures will help avoiding carryover contamination. (C) 1997 Elsevier Science B.V.