Maine research raises doubts about how doctors choose antibiotics to treat infections

Researchers at the University of New England are raising some provocative questions about the process doctors rely on to select antibiotics for their patients. 

They just published their findings in a scientific journal. But their work actually began two years ago — with a very sick woman a thousand miles away.

Dr. Meghan May, an associate professor of biomedical sciences at UNE, was contacted by a colleague at Indiana University Hospital who was concerned about the sick woman’s infection. The patient had arrived at the hospital critically ill with with an unusual type of pneumonia, and clinicians suspected a pathogen that’s rarely found in humans called Francisella philomiragia.

Dr. Meghan May

They asked May to confirm their suspicion. Of the few patients who develop the infection, most have experienced a near-drowning and inhaled water contaminated with the pathogen. This patient curiously had not.

“To this day we have no idea how she got it,” May said.

May performed DNA sequencing, which validated the cause as Francisella philomiragia. With no time to waste in the meantime, the patient’s doctors treated her with a cocktail of antibiotics.

Then May and “some very, very diligent medical technicians” in the Indiana hospital’s lab noticed an oddity about the case. The technicians tested the pathogen to see which antibiotics it was sensitive to, a regular practice that helps doctors decide which medications to prescribe to combat infections, May said.

The lab staffers grew the organism in an incubator that mimics the ambient air environment, standard practice with “antibiotic-resistance susceptibility” testing. But because the organism was so uncommon, they took the extra step of testing it in a second incubator that creates an environment with elevated levels of carbon dioxide.

The two results were drastically different. In ambient air, the organism responded to antibiotics commonly used to treat respiratory infections. In high levels of carbon dioxide, it was extremely resistant.

Amoxicillin penicillin antibiotics in the pharmacy at a free clinic in Los Angeles, California, U.S., April 27, 2016. REUTERS | Lucy Nicholson

So the lab technicians ran the test again. They got the same results.

That’s when May jumped back on the case. She zeroed in on antibiotic-resistant genes “that fly around hospitals and create superbugs” by spreading themselves to a variety of bacteria and viruses. May hypothesized that such genes were involved, and behaving differently based on the environment around them.

Further study over the next year with her student team from UNE showed she was right. In ambient air, two antibiotic-resistant genes frequently seen in hospitals remained dormant. But “when you put them in elevated CO2, they get turned on like a light switch,” she said.

Standard testing had indicated that the class of drugs including penicillin should work against the woman’s infection. But elevated carbon dioxide is also found on the surface of the lungs, where her infection had taken hold. In that environment, the antibiotic-resistant genes activated and rendered the usual treatment useless. If her doctors had tried only the medications that standard testing favored, she may not have survived.

The potential implications of their findings reach far beyond one patient in Indiana, however. The work raises doubts about the reliability of standard testing for antibiotic resistance, which doctors across the country count on to quickly and accurately select medications for their patients, for conditions as common as ear infections.

“If we’re always testing in regular air, [antibiotic-resistant genes] are going to be switched off and we’re not going to see them,” May said.

Those genes play a role in an untold number of dangerous infections responsible for a crisis in the nation’s hospitals. Up to 10 percent of hospitalized patients in the U.S. acquire an infection every year, leading to 99,000 deaths and $20 billion in health care costs, according to conservative estimates by the U.S. Centers for Disease Control and Prevention.

A sample of live MRSA bacteria in a laboratory at the University of Chicago. (Zbigniew Bzdak | Chicago Tribune | TNS)

“If we look at causes of death in the U.S., particularly within hospital settings, infection is absurdly high,” May said.

Both of the genes the UNE researchers tested are found among the World Health Organization’s list of the globe’s most dangerous superbugs, May said. But there are many more.

Without accurate testing that equips doctors with the information they need, they may prescribe antibiotics that have no hope of working. Physicians frequently encounter infections resistant to common antibiotics, leading them to try multiple medications in hopes of landing on one that will thwart the illness. That not only puts individual patients’ lives at risk, but also contributes to greater antibiotic resistance among the entire population, experts say.

The UNE team’s research, published in July in mSphere, a journal of the American Society of Microbiology, suggests that testing for antibiotic resistance must also take into account the physiological conditions of the infection, May said. Elevated carbon dioxide, for example, is also found in the blood of sepsis patients. But other conditions that could affect how infections respond to antibiotics include salt concentrations or pH levels, she said. There are also multiple classes of antibiotics to further explore.

“We want to look at other antibiotic-resistant genes and other physiological conditions and see how widespread this phenomenon actually is,” May said.

May acknowledges that the changes she’s suggesting to testing for antibiotic resistance would represent a “sea change in the way things are done.” Clinicians and labs would have to agree upon, validate, and communicate a set of conditions under which to test for antibiotic resistance, depending on the type of infection, severity, and a range of other physiological factors. Now, labs get a swab and run it through a single testing procedure, she said.

This is just one study, and more questions remain, as May said. But it’s fair to say that the research could point to a big blind spot in the process doctors use to choose antibiotics for their patients.

“The potential magnitude of this is kind of overwhelming in a way,” May said. “But the good thing about that is, I think it’s a problem we can solve, with a lot of work.”
Jackie Farwell

About Jackie Farwell

I'm the health editor for the Bangor Daily News, a Bangor native, a UMaine grad, and a weekend crossword warrior. I never get sick of writing about Maine people, geeking out over health care data, and finding new ways to help you stay well. I live in Gorham with my husband Nick and our hound dog Riley.