Study hints at complex interaction of resistant microbes, antibiotics

Drug-resistant bacteria are a well-documented problem in nursing homes. Residents are frail and prone to infection, and antibiotic use is common. That's one of the reasons public health officials and infectious disease experts believe nursing homes are a critical target for stewardship programs that promote more prudent use of antibiotics.

But a new study from researchers at the University of Michigan suggests that these programs could be hindered by a focus on individual species of multidrug-resistant organism (MDRO) and certain antibiotic classes that may fuel that resistance.

The reality, the researchers say, is that nursing homes are likely filled with multiple MDROs simultaneously interacting with antibiotics and with each other in complex ways that can influence the dynamics of patient colonization and increase the likelihood of infection. Their study scratches at the surface of this drug-resistant ecosystem.

"One of the major ways these organisms are evolving is through resistance to antibiotics, and we've known that," corresponding author Evan Snitkin, PhD, a microbiologist at the University of Michigan Medical School, said in an interview. "But something else they're seeing in addition to antibiotics is each other.

"They're all circulating through these facilities simultaneously, and that opens the potential that they might be interacting with one another, whether by coincidence or by evolved mechanism, and our results hint at that."

MDROs 'ubiquitous' in nursing homes

For the study, Snitkin and his colleagues retrospectively analyzed data from 234 residents at 12 community nursing homes in Michigan. The data, collected from May 2010 through April 2013, came from a larger study on bacterial colonization and antibiotic exposure in nursing home residents over time.

Snitkin and his colleagues focused on patients who had had a urinary catheter for at least 3 days—a population with a high risk of developing urinary tract infections (UTIs). "These are some of the highest risk individuals in nursing homes," Snitkin said.

Focusing on this group allowed the researchers to look at patterns of urinary tract infections and which MDROs might be associated with them.

The MDROs most commonly found to be colonizing these patients were vancomycin-resistant enterococci (VRE), methicillin-resistant Staphylococcus aureus (MRSA), Escherichia coli, Proteus mirabilis, Acinetobacter baumannii, and Pseudomonas aeruginosa. Nearly a quarter of the nursing home residents (23.5%) were colonized with at least one of these organisms, while 20.9% were colonized with two species and 20% were colonized with more than two species.

Although previous studies have shown that drug-resistant pathogens are prevalent in nursing homes, Snitkin said he was surprised by these results. "I was really surprised that, outside the ICU setting in a hospital, antibiotic-resistant organisms would be so ubiquitous, and to find patients colonized with multiple organisms simultaneously was even more intriguing to me," he said.

Colonization patterns emerge

The team then constructed models to figure out what was influencing acquisition of each of these organisms. The high frequency of multi-MDRO colonization, it turns out, was not random.

In fact, the presence of certain MDROs in colonized individuals made it more likely that they would be colonized by another drug-resistant bug. Patients colonized with P mirabilis, for example, had more than two times the risk of also acquiring VRE, MRSA, A baumannii, and E coli, while patients with E coli, A baumannii, and P aeruginosa had an increased risk of acquiring P mirabilis.

Snitkin and his colleagues also found that exposure to individual antibiotic classes was associated with increased risk of acquiring certain pathogens, and in some cases multiple species. Over the course of the study, 146 of the residents received at least one course of treatment with 1 of 50 antibiotics. Patients treated with aminoglycosides were found to have more than five times the risk of being colonized with A baumannii, and more than three times the risk of acquiring E coli. Treatment with nitrofurans increased the risk of acquiring VRE threefold and P mirabilis eightfold. And multiple antibiotics were found to increase the risk of a single MDRO.

From these findings, the researchers were able to create a map of interactions that showed that all MDRO colonization in the patients, except for MRSA, was positively associated with at least one antibiotic and one other MDRO. Antibiotic exposure increased nursing home residents' risk of initially acquiring a primary MDRO, which in turn influenced the risk of colonization with another MDRO. 

This observation led them to ask whether these interactions were influencing risk of infection. Their subsequent analysis of 70 patients with UTIs indicated that co-colonization with certain pairs of species, like P mirabilis and E coli, was linked to increased risk for UTIs, a finding that suggests enhanced pathogenic potential when the two bacteria are both present in a patient. Co-occurrence of E coli and VRE, and P mirabilis and MRSA, was also associated with increased the risk of UTI.

Complex web of interactions

Snitkin said all the findings point to a complex web of interactions among drug-resistant organisms and antibiotics. "It's not just antibiotics but interactions with one another that are driving acquisition and infection," he said.

And what this interconnectivity means, he added, is that stewardship programs in nursing homes and other long-term care facilities will need to take a network-based approach to tackling the underlying MDRO burden, rather than focusing on restricting use of certain antibiotics to reduce individual organisms.

"If the objective is to control the overall burden of resistance in a facility, then maybe taking a single-organism view of stewardship might not fulfill that objective," he said. "The impact of limiting the use of one antibiotic in favor of another will have many downstream influences that could counterintuitively increase the overall burden."

Snitkin notes that for now this is just one study in one set of nursing homes, and that further research will be needed in other nursing homes and other types of healthcare facilities to determine how robust the findings are and understand how these interaction networks might change depending on what organisms are circulating and what antibiotics are being used. "If we see the same patterns over and over again, then we'll have a basis to start influencing stewardship," he said.

In September 2016, the Centers for Medicare and Medicaid Services finalized a rule that requires more than 15,000 long-term care facilities (nursing homes, skilled-nursing facilities, and assisted-living facilities) to have antibiotic stewardship programs. With the lack of antibiotic stewardship expertise at many of these facilities, developing and maintaining these programs is expected to be a challenge.

See also:

Sep 12 Proc Natl Acad Sci USA study

Newsletter Sign-up

Get CIDRAP news and other free newsletters.

Sign up now»


Unrestricted financial support provided by

Bentson Foundation 3MAccelerate DiagnosticsGilead 
Grant support for ASP provided by

  Become an underwriter»