Trying to initially identify a respiratory illness, much less a potential emerging threat, turns out to be much harder than you’d think, even if the doctor listens carefully to your lungs and hears your story. Now scientists at Los Alamos National Laboratory have measured the misdiagnosis rate for upper respiratory tract infections, comparing syndromic diagnostics (doctor’s office observations) with laboratory methods. In a study pitting the results of early identification in an office against polymerase chain reaction (PCR) and RNA sequencing of samples from the same patients, the team found that skilled medical observation without laboratory testing identified only 37 percent of the study’s patients with respiratory tract infections.
“We have tailored, rapid point-of-care diagnostics for only a handful of pathogens,” said study principal investigator Harshini Mukundan, yet many respiratory infections share common symptoms (i.e., flu-like symptoms), making them difficult to diagnose using only the reported symptoms of patients.
The research was undertaken with the goal of understanding the tools needed by physicians, health care workers and first line responders identifying future outbreaks and emerging threats at the point of care.
The team, coordinating with Medical Associates of Northern New Mexico (MANNM, a medical practice in northern New Mexico), found influenza virus to be a major player in respiratory infections, which was as expected. However, they also found coronavirus, respiratory syncytial virus, parainfluenza virus, and human metapneumovirus associated upper respiratory tract infections using PCR and sequencing approaches.
“There is no quick way to discriminate between these viruses at the point of need, and therefore, physicians have to rely on their experience and what is called ‘syndromic diagnosis,’” said Mukundan.
Indeed, syndromic diagnosis appeared to indicate less than half of studied patients to have an upper respiratory tract infection, and the actual type of virus could only be distinguished in one patient.
“Doctors are not given the best tools to work with when it comes to being able to diagnose infections with the same symptoms, as it is virtually impossible to distinguish pathogens that present in very similar ways,” said first author Andrew Bartlow of Los Alamos’ Bioscience Division. “Current syndromic diagnostic approaches failed to accurately identify the specific respiratory pathogens associated with infection, and so when it comes to accurately identifying emerging threats, clearly, they are not suited.”
Bartlow explained that correct diagnosis is naturally critical for knowing how to cure an infection, “So our goal is that of assessing essential gaps in our ability to identify respiratory pathogens,” he said.
“What we found is that there is an urgent need for diagnostics that can track both known and unknown pathogens at the point of care to help prevent and control future outbreaks,” Bartlow said.
In a parallel effort, the team also studied the development of computer modeling strategies to accurately predict respiratory infections. “Both of these are essential components to developing systems that can prepare us against emerging pathogens” said project principal investigator Mukundan.
Paper: "Comparing variability in diagnosis of upper respiratory tract infections in patients using syndromic, next generation sequencing, and PCR-based methods," PLOS Global Public Health, DOI https://doi.org/10.1371/journal.pgph.0000811. Los Alamos authors: Andrew Bartlow, Cheryl Gleasner, Bin Hu, Karen Davenport, Po-E Li, Patrick Chain, Tracy Erkkila, Jeanne Fair, Harshini Mukundan.
Funding: U.S. Department of Homeland Security, Science and Technology Directorate and Director's Funded Postdoctoral Fellowship through the Laboratory Directed Research and Development program