It’s long been known that antibiotics can disrupt the balance of bacteria in the gut, typically causing short-term symptoms like diarrhea. Some research suggests the disruption can last for several months.
But a new study led by researchers in Sweden indicates certain antibiotics may have a far longer impact on the gut microbiome.
The study, based on drug registry data and genetic analysis of fecal samples from nearly 15,000 Swedish residents, found that people who took certain oral antibiotics had fewer bacterial species in their gut, and a lower abundance of individual species of gut bacteria, for four to eight years after taking them. The antibiotics most strongly associated with these long-lasting impacts were clindamycin, fluoroquinolones, and flucloxacillin.
The findings were published last week in Nature Medicine.
An author of the study says the findings provide strong evidence on the persistent effects that certain antibiotics can have on the taxonomy of the gut microbiome.
“This is something we imagined already based on previous studies, but I think this is the strongest evidence that we have so far for this effect,” first author Gabriel Baldanzi, PhD, MSc, a postdoctoral researcher at Uppsala University at the time the study was conducted, told CIDRAP News.
Reduced bacterial diversity, abundance
For the study, Baldanzi and colleagues from Uppsala University, Karolinska Institutet, and Lund University combined individual-level data from the Swedish Prescribed Drug Register, which includes information on all drugs prescribed at pharmacies in Sweden, with fecal metagenomic data collected from three population-based studies. Their aim was to assess how oral antibiotics used in the eight years before fecal sampling were associated with changes in gut microbiome composition.
Baldanzi said that some previous studies have indicated that microbiome differences can last as long as 1 or 1.5 years. But having access to years’ worth of individual antibiotic prescription data and genetic profiles of individual gut microbiomes allowed the researchers to see if the effect was longer.
“The question was, ‘does the microbiome actually recover from the antibiotic, or does this effect persist in the microbiome?’” he said. “We thought this was an important question, and we felt we had the data to investigate it.”
The results of their analysis, which controlled for factors such as use of non-antibiotic medications and comorbidities, showed that in 14,979 adults, use of six of 11 antibiotic classes a year before fecal sampling was associated with significant reductions in gut microbiome diversity compared with no antibiotic use.
This is something we imagined already based on previous studies, but I think this is the strongest evidence that we have so far for this effect.
Each course of clindamycin taken the year before fecal sampling was associated with an average of 47 fewer species detected. Each course of fluoroquinolones or flucloxacillin (a type of penicillin approved for use in Europe but not the United States) was associated with 20 and 21 fewer species, respectively. In contrast, penicillin V and amoxicillin were not associated with reduction of species diversity.
When the researchers analyzed the abundance of gut bacteria species following antibiotic use, those three antibiotics again stood out. Clindamycin taken a year before sampling was associated with changes in 296 of the 1,340 species analyzed, while flucloxacillin was associated with changes in 203 species and fluoroquinolones with 172. With most of the species analyzed, there was a decreased abundance, though some species saw an increased abundance.
Gut microbiome diversity recovered most quickly in the two years after antibiotic use, then more slowly in subsequent years. But use of clindamycin, fluoroquinolones, and flucloxacillin one to four years and four to eight years before fecal sampling was still significantly associated with reduced gut microbiome diversity. And use of these antibiotics four to eight years before sampling was associated with altered abundance of 10% to 15% of the gut microbiome species studied.
When the researchers restricted the sample to participants who had received only one antibiotic course or none, they found that a single course of seven of the 11 antibiotics studied was associated with reduced gut microbiome diversity four to eight years later.
“These findings support the notion that the effect of antibiotics on the gut microbiome may persist for several years,” they wrote.
Antibiotics, the gut microbiome, and health
The researchers note that the findings aren’t generalizable, since it’s limited to the gut microbiome profiles of Swedish residents and to the antibiotics commonly used in the country, which has fairly restrictive policies on antibiotic use. The bigger question, which was beyond the scope of this study, is whether the long-term disruption of the gut microbiome has an impact on health.
“That is the big question behind this and all other studies looking at the impact of antibiotics on the microbiome,” Baldanzi said.
As the researchers note, previous observational studies have found associations between long-term antibiotic use and increased risk of obesity, type 2 diabetes, and cardiovascular disease. And their analysis found that use of clindamycin, fluoroquinolones, and flucloxacillin was associated with an increased abundance of certain gut microbiome species that have previously been linked to higher body mass index, higher serum triglyceride levels, and increased risk of type 2 diabetes.
But there isn’t enough evidence yet to say whether antibiotic-induced disruptions of the gut microbiome cause these conditions.
Going forward, Baldanzi said there are discussions to examine the impact of antibiotic use on antibiotic-resistance genes in the gut microbiome, and to investigate whether antibiotic use changes how the gut microbiome functions, or if the gut bacteria eradicated by antibiotics are replaced by microbes that perform similar functions. He’d also like to see similar studies conducted in other countries with different patterns of antibiotic use.
“This study is not the final answer,” he said.
