In large study, antibiotic pairs mostly display ‘concurrent resistance’
Researchers retrospectively analyzed nearly 450,000 antimicrobial susceptibility test results from a couple dozen hospitals to identify antibiotic pairs with disjoint resistance.
“That is, a pathogen can't be resistant to both antibiotics in the pair at the same time,” explained Erik S. Wright, PhD, MS, an assistant professor of biomedical informatics at the University of Pittsburgh. “We called this disjoint resistance because a disjoint set is one that is mutually exclusive.”
According to Wright, the existence of such antibiotic pairs is expected because of a phenomenon known as “collateral sensitivity,” which is when a pathogen adapts to one drug and can become more sensitive to other drugs or more resistant.
“Research had previously shown that collateral sensitivity exists between some antibiotic pairs in vitro,” Wright said. “The question is whether this leads to observing disjoint resistance in the clinic. If it does, then we could potentially use these pairs of antibiotics to avoid multidrug resistance.”
Instead, their review found the opposite.
“Unfortunately, we mainly saw concurrent resistance at the species level among antibiotics used to treat six common bacterial pathogens, which corroborates and extends upon previous findings of concurrent resistance between antibiotics,” Wright said.
For the study, Wright and colleagues retrospectively analyzed 448,563 antimicrobial susceptibility test results acquired between Jan. 1, 2015, and Dec. 31, 2018, from 23 hospitals in the University of Pittsburgh Medical Center hospital system. According to the study, they used a score based on mutual information to identify pairs of antibiotics displaying disjoint resistance.
They then applied this approach to the six most frequently isolated bacterial pathogens Escherichia coli, Staphylococcus aureus, Klebsiella pneumoniae, Enterococcus faecalis, Pseudomonas aeruginosa and Proteus mirabilis and subpopulations of each created by conditioning on resistance to individual antibiotics.
Overall, the researchers identified 69 antibiotic pairs displaying varying degrees of disjoint resistance for subpopulations of the six bacterial species. However, the study demonstrated that disjoint resistance was rarely held at the species level, with only six (0.7%) of 875 antibiotic pairs showing evidence of disjoint resistance.
Instead, they study showed that 53.1% of antibiotic pairs exhibited markers of concurrent resistance when resistance to one antibiotic is associated with resistance to another. This resistance also extended to more than two antibiotics, with observed rates of resistance to three antibiotics being higher than predicted from pairwise information alone.
“We showed how to find antibiotic pairs that are most likely to be successful at mitigating antibiotic resistance. These are good antibiotic candidates for future clinical trials, but they will also require subspecies level classification to be correctly applied,” Wright said. “Our study also revealed which antibiotic pairs display the worst concurrent resistance, and the use of these antibiotics together should be avoided.”
Wright added that one potential strategy to combat antibiotic resistance is to alternate between different antibiotics and that, most often, this strategy has been employed by giving different antibiotics to different patients within the same hospital, either switching antibiotics every other patient or every other month.
“These switching strategies have failed to curb resistance, but this might be because of the wrong choice of antibiotic pairs,” he said.