http://www.wsj.com/articles/bacteria-or-virus-new-tests-may-identify-whats-causing-your-infection-1456768827
Research holds promise for treating respiratory illness and limiting overuse of antibiotics
Duke University researchers are working on an intriguing idea—a blood test to determine whether a respiratory infection is caused by a virus or bacteria.
If it is developed, the test could make prescribing treatment easier and more accurate. Nearly three-fourths of patients with acute respiratory illnesses are prescribed antibiotics even though their infections are usually viral, studies have found. There is no treatment for most viruses.
Antibiotics kill many of the body’s good bacteria and can cause side effects such as gastrointestinal problems. Overuse of antibiotics contributes to the growing global public-health problem of antibiotic-resistant infections as the drugs become less effective.
Like many family physicians, Theodore Ganiats often sees congested, coughing, aching, feverish patients who are desperate for an antibiotic to relieve their ills. The problem is, he can’t usually confirm their respiratory infections are bacterial. Often, they are viral.
“It’s often hard to get a person who doesn’t need an antibiotic to accept that,” said Dr. Ganiats, a family physician and professor at the University of Miami, in Florida. “A test to be able to know when an antibiotic is needed would be tremendously helpful.”
In research published last month, the Duke University researchers said their blood test can determine whether a respiratory illness is caused by a virus or bacteria. “This includes everything from a runny nose all the way down to pneumonia,” said Ephraim Tsalik, an assistant professor of medicine at Duke University Medical Center and the Durham VA Medical Center and lead author of the study.
The test currently is available only as a research tool with an eight- to 10-hour turnaround time. The researchers are exploring development opportunities including collaborating with two biotechnology companies in hopes of developing a one-hour blood test that can be used clinically. Such a test is at least two to three years away and would require Food and Drug Administration approval, Dr. Tsalik said.
The Duke researchers are zeroing in on the interaction between an individual and pathogen. “This is a new paradigm,” said Gregory Storch, professor of pediatrics at Washington University School of Medicine, in St. Louis, who is also researching gene expression as a means of distinguishing between bacterial and viral infections. “Rather than trying to detect the germ that’s causing the disease, you detect something about the host—the person who’s experiencing the infection.”
Currently, when doctors try to identify if an infection is caused by a virus or bacteria, they focus on the pathogen by culturing it to see if it is bacterial or not. That process can take one or more days and has its weaknesses, as not all bacteria can grow in the lab.
There are some faster tests: The rapid-strep test can tell in minutes if the bacteria that causes strep throat is present, but it has high false-negative and false-positive rates, experts say. In addition, a significant percentage of people are carriers, meaning they may have bacteria present but aren’t sick.
In some recent hospital-based polymerase chain reaction (PCR) lab tests, a sample of blood or saliva can be tested for multiple viruses and bacteria at one time. But such tests may include positive results for people who have a virus with no symptoms, said Dr. Tsalik. In such cases, the person may be sick from something other than the virus for which they tested positive.
The Duke researchers have developed what they call gene signatures, or the gene expression of about 120 genes. “The nature of our reaction or response to a viral infection is different than the way our bodies react to a bacterial infection,” said Dr. Tsalik. “We’ve looked at one particular measure of our body’s response—gene expression—and how the genes in our bodies are being turned on and off.”
In a study published in the journal Science Translational Medicine last month, researchers found the test was about 87% accurate in determining whether 273 patients had a viral or bacterial infection, or were sick due to something other than an infection, like heart failure.
The test can’t determine which particular virus or bacteria is causing an infection, but it can identify co-infections, which are both bacterial and viral in nature.
“This need for bacterial versus viral tests is really great and has the ability to fundamentally change the way antibiotics are used,” said Jean Patel, acting director of the Office of Antimicrobial Resistance at the Centers for Disease Control and Prevention. The Duke study, she said, “could really move the testing field forward.”
“There’s a lot more testing that needs to be done to see if a test like this could be applied, but it certainly demonstrates that there’s promise” she said.
Biofire Diagnostics, a Salt Lake City biotech firm, is one of the companies exploring a collaboration with Duke. The company currently has a PCR-based test for a number of respiratory illnesses. The test looks for 20 different upper-respiratory pathogens at the same time—17 viral and three bacterial, said Robert Crisp, Biofire Diagnostics’ senior director of biochemistry. But because the test has been classified as moderately complex, it currently can be used only in hospitals, not doctor’s offices, he said.
Dr. Storch is studying gene expression as a way to distinguish bacterial from viral infections with colleagues at Washington University School of Medicine and Ohio State University. “A lot of times the diagnostic tests done in an office don’t really give the answer,” he said. “Sometimes the right specimen isn’t available or the technology isn’t adequate to detect the microbial agent.”
Gene-expression tests are expected to be able to distinguish between symptomatic infections and nonsymptomatic infections.
Nonsymptomatic infections are quite frequent, Dr. Storch said, noting that studies have found as many as 25% of young children are carrying the rhinovirus, which causes the common cold, and don’t have symptoms. Similarly, as many as 10% of school-age children might have the bacteria that causes strep throat but aren’t sick from it.
Mary Anne Jackson, a professor of pediatrics and infectious diseases at Children’s Mercy in Kansas City, Mo., say a one-hour blood test isn’t practical in an outpatient pediatric population. Blood tests aren’t easy to administer to young children, and an hour is a long time for a patient to wait for results.
More important, she said, is to educate parents on the overuse of antibiotics and share guidelines on how they should be used.