As many as half of the antibiotics prescribed to outpatients in the United States are unnecessary. And many of those scripts come from doctors who aren’t sure whether a patient is infected with bacteria — and needs antibiotics — or whether the illness is caused by a virus.
Now, researchers have developed a new way to tell these kinds of pathogens apart. Instead of looking for the presence of the foreign attacker itself, as existing methods do, the new diagnostic tool uses our body’s response to determine if a patient is feeling lousy because of a bacteria, a virus, or something else entirely.
The technique isn’t ready for the clinic yet, but the researchers who developed it say they hope that will change in the next few years.
“We can put this in a clinical laboratory right now,” said Dr. Geoffrey Ginsburg, director of Duke University’s Center for Applied Genomics and Precision Medicine and one of the authors of the study. “But it will take about 12 hours from the time we get a blood sample to the time we get a result.”
For most clinicians, that’s just too long.
“If it can be turned into a point-of-care test, it could really revolutionize medicine, but that’s a big piece of work to accomplish,” said Dr. Timothy Lahey, an infectious disease specialist at Dartmouth’s Geisel School of Medicine, who was not involved in the study. “Lots of promising technologies fall away because they never make it to the point-of-care.”
This technology is based on a simple idea: In order to properly fight off bacteria and viruses, our bodies need to be able to distinguish between the two. Once our immune cells have identified what kind of attacker is present, they trigger a cascade of communications to coordinate their defense strategy. And because different genes are responsible for different immune responses, measuring the activity of certain genes during infection is a good way to tell what is bothering the body just then.
Ginsburg and his team used blood samples from 273 people who had come to acute care centers with cold or flu-like symptoms and 44 healthy controls.
The researchers knew from traditional methods who was infected by what. That allowed them to compare the gene activity in those blood samples, and to identify which patterns of gene expression clearly showed that the person was fighting off bacteria or viral infections. They also looked for the gene activity in those patients whose symptoms were not caused by an infection at all.
Just as birdwatchers know the exact pattern of feathers that distinguishes one bird family from another, these researchers have narrowed down the genes that are important in identifying what kind of immune response our bodies are having. The researchers hope to use those 130 genes as the basis for a rapid diagnostic test — and that test could be a useful tool in preventing bacteria from becoming drug-resistant.
“We know that antibiotics are prescribed inappropriately the majority of the time,” said Dr. Amesh Adalja, an infectious disease and biosecurity expert at the University of Pittsburgh Medical Center.
He explained that doctors don’t want a patient to go home untreated if he or she might have a dangerous bacterial infection. Even when they suspect that the infection might be caused by a virus, sometimes physicians still prescribe antibiotics, because patients have come to expect them, and might complain if they emerge from the appointment empty-handed.
The extra antibiotics floating around allow bacteria to develop resistance to the drugs, making a whole list of illnesses much harder to treat in the long run.
“Those are all huge downsides that are killing people on a regular basis,” Lahey said, “but they’re less immediate than the person in front of you.”
The results of this new technique, which were published Wednesday in the journal Science Translational Medicine, could stop clinicians from unwittingly perpetuating the epidemic of antibiotic resistance. But they’ll need to perfect the technology for the clinic first.