Could a warming climate be one of the factors bringing the world closer to the "post-antibiotic" era that infectious disease experts have been warning about?
That's one of the questions raised by a new paper in Nature Climate Change that explores the role that climate and other factors play in the distribution of antibiotic resistance in the United States. The study, conducted by researchers with the University of Toronto, Harvard Medical School, and Boston Children's Hospital, shows that increasing local temperatures are associated with higher levels of antibiotic resistance in three common bacterial pathogens—Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus.
The researchers also found that higher population densities and antibiotic prescribing rates are associated with higher rates of drug resistance.
The findings of the study do not show that increasing temperatures are causing antibiotic resistance to rise, and the authors of the study note that antibiotic use remains the primary driver for selection of antibiotic resistance. But they say the findings open up an intriguing area of research that could expand the understanding of the forces that affect antibiotic-resistance rates.
"One of the things we're doing with this paper is opening the door to an additional variable that has not been previously thought of as relevant," study co-author Mauricio Santillana, PhD, a professor of pediatrics at Harvard Medical School, said in an interview.
The most recent estimates predict that drug-resistant bacterial infections could kill more than 10 million a year by 2050 if antibiotic resistance levels continue to rise at their current pace. But if the link can be confirmed and a causal relationship established, Santillana and his colleagues warn, it means current predictions of the impact that antibiotic resistance will have on global health in the future could be underestimating the problem, and that a warming climate could promote a more rapid rise in resistance.
Higher minimum temps, higher resistance
To determine the factors that explain regional differences in antibiotic resistance, the researchers created a large database of regional antibiotic resistance patterns using data from hospitals, laboratories, and surveillance units across the country.
Their final data set, restricted to E coli, K pneumoniae, and S aureus, included 1.6 million bacterial isolates collected from 223 facilities in 41 states from 2013 through 2015. They then linked each location by zip code with national climate data, local antibiotic prescribing rates, and population density.
Study co-author Derek MacFadden, MD, an infectious disease specialist and research fellow at Boston Children's Hospital, said the team was interested in the role of climate because of anecdotal evidence from previous research that suggested there may be gradients in antibiotic resistance according to latitude and temperature. Then they started seeing a potential link with temperature as they began looking at their data.
"The more we thought about it, the more it made sense that it may be a contributing factor to the distribution and transmission of antibiotic resistance," said MacFadden. "We then set out to look at it in a more rigorous fashion, and also try to asses a number of other predictors on a population level that might also affect the distribution of antibiotic resistance."
As they explored the relationship between regional patterns of antibiotic resistance and both latitude and temperature, they found that an increase in the minimum temperature across US regions was associated with an increase in resistance to most classes of antibiotics in all three pathogens. They also found that higher antibiotic prescribing rates and an increase in population density of 10,000 people per square mile were likewise associated with increased resistance (but with only K pneumoniae and E coli being affected by increasing density).
The link between prescribing rates and resistance was expected, co-author Sarah McGough, PhD, said, based on existing literature. The association between population density and resistance was also expected, and may reflect easier transmission of pathogens in denser populations, the team concluded. McGough is a researcher with the computational health informatics program at Boston Children's Hospital.
But after adjusting for these and other variables, the association between higher minimum temperatures and increased antibiotic resistance remained. The team calculated that an increase in the average minimum temperature of 10°C (18°F) across regions was associated with an increase in antibiotic resistance of 4.2% for E coli, 2.2% for K pneumoniae, and 2.7% for S aureus. In addition, changes in minimum temperature were associated with larger increases in resistance in each year analyzed.
"The reason we included other predictors, like antibiotic use and population density, is because those are well-known in the literature to be drivers of antibiotic resistance," McGough said. "What was really strong about this study is that, after controlling for those factors, we still find that temperature is a strong predictor of antibiotic resistance."
The strongest associations between temperature and resistance was found in fluoroquinolones and beta-lactam antibiotics, suggesting that warmer temperatures may affect the way bacteria respond to certain drug mechanisms.
Although the reasons higher temperatures are associated with increased antibiotic resistance need to be further explored, the authors suggest several factors may help explain the relationship. One is that higher temperatures may affect bacterial growth rates and drive increased carriage and transmission of drug-resistant strains. Previous studies have shown increased carriage of resistant organisms in warmer seasons.
Another is that temperature may affect horizontal gene transfer—the process through which bacteria share resistance genes. The paper notes that several of the highly mobile genetic resistance mechanisms that have emerged in recent years have originated in central latitudes.
But given the complexity of factors in the emergence and transmission of antibiotic-resistant bacteria, MacFadden said, it's difficult to say exactly what process temperature is affecting.
"We still have a poor understanding of transmission networks; bacteria are being transmitted from human to human, but also from animal to human, from food to human, and between animals and the environment," he said. "It's a very complex web, and it's the prototypical One Health issue."
Evan Snitkin, PhD, a microbiologist at the University of Michigan Medical School who studies antibiotic resistance and was not involved in the study, said that while it's unclear what the biological mechanism behind the association may be, he finds the data in the study compelling. "It will be interesting to see if they are able to tease apart a putative causal pathway underlying this observation in the future," he said.
To that end, Santillana said the next steps will be to analyze data from other parts of the world, to see if the relationship between temperature and resistance is seen elsewhere, and then dissect that link to see what's going on. "We need to understand if temperature is the driver, in some biological way, or if temperature is facilitating other processes or supporting other ecological events that are responsible for antibiotic resistance rates," he said. "We have a lot of homework to do."
May 21 Nat Clim Chang study