Study tells why widely used lab test misdiagnoses norovirus

A lab test widely used to determine the cause of a patient’s diarrhea often incorrectly implicates the stomach bug norovirus. Researchers at the University of Washington School of Medicine discovered that the test, made by the biotech company bioMerieux SA, inadvertently interacts with 78 bacteria normally found in the gut. This can lead to misdiagnoses.  

The researchers’ findings were published in the Journal of Clinical Microbiology. 

Norovirus: highly contagious, and not just on cruise ships 

Norovirus is the leading cause of gastroenteritis and foodborne illness worldwide, accounting for roughly 109,000 U.S. hospitalizations annually and 685 million cases globally, according to the Centers for Disease Control and Prevention.  

Typical symptoms are diarrhea, vomiting, nausea and stomach pain. The virus is highly contagious, spreading through contact with sick people as well as through contaminated food and surfaces. People usually recover within one or two days but may continue to spread the virus for several more days. 

Outbreaks are common in long-term care facilities, hospitals, schools and childcare settings, and are often linked to restaurant and catered food. Outbreaks on cruise ships draw widespread news coverage but account for only about 1% of norovirus outbreaks. 

More than a quarter-million false-positive reports 

About five years ago, clinicians around the country began reporting that a lab test, used to diagnose causes of diarrhea, was indicating that patients had norovirus when they did not. This has led to more than a quarter-million false-positive norovirus reports in the years since. 

Norovirus outbreaks typically happen in winter, “so when labs started reporting lots of norovirus cases in the middle of July, people thought it was weird," said senior author Dr. Alexander Greninger, professor of laboratory medicine and pathology at the UW School of Medicine. 

The BioFire FilmArray Gastrointestinal Panel has been adopted by many clinics and hospitals because, from a single sample, it can rapidly detect 22 bacteria and viruses that cause diarrhea. Reports that the panel was incorrectly detecting norovirus led the company to change its instructions on how test results should be interpreted, but did not change the test itself. 

Misdiagnoses hurt on multiple fronts 

"Diagnostic panels are very helpful," said Jonathan Reed, a senior research scientist in the Greninger lab, "but if you're misdiagnosed with norovirus, you may not get the appropriate treatment and the hospital has to implement expensive infection-control procedures." 

Reed and Clarice Mauer, research scientists in the Greninger lab, were co-first authors of the published paper. 

To determine why the BioFire assay was falsely reporting samples as positive for norovirus, the researchers obtained BioFire test cartridges, called pouches, that had been used to analyze stool samples from patients with diarrhea at four Seattle-area healthcare facilities. 

They facilities' false-positive rate ranged from 31% of samples to 74%. "On average across the hospitals, there was about a 50% chance a positive test was false," Greninger said. "It was basically a coin flip whether your patient had norovirus or not." 

Deducing the cause of the misdiagnoses 

Tests like the BioFire panel use a technique called reverse transcription polymerase chain reaction, or RT-PCR, to detect the nucleic acid of specific pathogens in a sample. A short snippet of synthetic DNA, called a primer, is designed to be complementary to a specific segment of the target pathogen's DNA. If the pathogen is present, the primer binds to its nucleic acid and an enzyme called a polymerase begins making multiple copies of DNA, known as amplicons. Acting as a kind of molecular photocopier, PCR can quickly generate billions of amplicons, making it possible to detect minute amounts of nucleic acid in a sample. 

In their study, the researchers opened the BioFire pouches and sequenced amplicons from them to identify what the BioFire primers had amplified in the samples. One of the primers was reacting to DNA sequences found in 78 different species — mostly bacteria commonly found in normal stool. "We were a bit surprised that so many bacterial species were being amplified by the assay," Reed said. 

Reed said the primers appear to be binding to segments of DNA in these other species that only partly resemble the targeted norovirus sequences. This "mispriming" triggered off-target amplification, generating a false signal that norovirus was present, Reed explained.  
 
"Our findings demonstrate the need to redesign this assay to ensure reliable testing and improve patient care,” Mauer said. 

For details about UW Medicine, please see our About page.