Hepatitis A diagnosis relies on serology and occasionally on hepatitis A virus (HAV) RNA detection. For timely diagnosis and the avoidance of drawing additional blood, molecular testing is often performed as reflex testing by using blood specimens that were initially sent for anti-HAV serology. Reflex molecular testing is preferably performed from different sample aliquots, but, for limited sample quantities, it uses samples that have been preprocessed in an immunoassay analyzer. In 2012, we first observed sporadic HAV RNA-positive cases that were inconsistent with patients' serological profiles and/or medical histories, suggesting that occasional laboratory contamination was causing false-positive PCR results. Multiple external quality assurance (EQA) and laboratory surface contamination checks were performed, questionable specimens were tested with various HAV RNA tests, and follow-up serum/stool samples were collected. All contamination-check samples and samples from healthy individuals tested HAV RNA-negative, and the laboratory successfully passed all EQAs. The HAV RNA-positive results were reproducible with various HAV RNA assays. No patients seroconverted, and their follow-up samples were consistently HAV RNA-negative. Finally, a detailed review of testing protocols revealed a correlation between HAV RNA false positivity and preceding anti-HAV testing with the Cobas-e411 automated immunoassay analyzer. HAV RNA was detected in the Cobas-e411 anti-HAV reagents, with the HAV sequences matching those from the false-positive samples. Preceding anti-HAV testing using two other immunoassay analyzers did not result in subsequent HAV RNA false positivity during reflex testing. The Cobas-e411 pipetting procedure with a single pipette tip collecting samples and anti-HAV reagents contaminated the original sample with the HAV RNA that was present in the immunoassay's reagents, thereby resulting in HAV RNA false positivity during the reflex testing. IMPORTANCE We present the first report of sporadic HAV PCR false-positive results that have been observed during the reflex testing of serum samples that have previously been tested for anti-HAV antibodies and have been caused by contamination with HAV RNA that is present in the reagents of the commercial anti-HAV immunoassay, with potentially serious clinical consequences. Although HAV RNA was consistently detected in the anti-HAV reagents of all three automated immunoassay analyzers that were in use in our laboratory, only the use of one analyzer and the corresponding commercial anti-HAV immunoassay reagents resulted in contamination that led to false positive HAV RNA results, and this was due to a peculiar pipetting mode of action in which the analyzer uses a single pipette tip to collect both anti-HAV reagents and a sample, which consequently causes the permanent contamination of the original sample with HAV RNA. Manufacturers should strongly consider the occasional need for reflex molecular testing from preprocessed samples and design their analyzers in a way that prevents contamination.
Keywords: HAV RNA; Hepatitis A virus; contamination; diagnostics; reflex testing; serology.