Global consumption of common drugs has steadily increased in recent decades. Many people take multiple medicines at the same time, often for prolonged periods, without science having had time to clarify the extent to which these habits influence the intestinal microbiome, a component now considered decisive for our well-being.
A line of research had already shown that numerous drugs – not only antibiotics, but also molecules intended for human targets – modify the bacterial populations of the intestine. However, a big question remained: how long does this influence really last? Some in vitro studies had shown that over 70% of antibiotics and a quarter of human-targeted drugs interfere with essential bacteria, including producers of butyrate and propionate, two key metabolites for the body’s health.
Other work had documented that beta-blockers, SSRIs and statins alter crucial bacterial functions, such as vitamin synthesis. Metagenomic analyzes on patients treated with metformin, proton-pump inhibitors or statins had also confirmed important changes in the composition and functions of the microbiome, often dose-dependent.
However, the temporal duration of these changes remained poorly explored. This is where the study by Aasmets and colleagues, published in, comes into play mSystemswhich integrated gut metagenomics data with a huge amount of information extracted from electronic health records (EHRs). A methodological choice that allows us to precisely track the use of drugs – current and past – up to five years before the collection of the samples.
Common drugs and intestinal memory
The researchers analyzed fecal samples from 2,509 individuals, cross-referencing them with detailed data on medications taken over the previous five years.
The first part of the study focused on the effects during the active use of a drug, comparing subjects on therapy with people who had not used it in the previous five years. The result is surprising: almost 90% of the 186 drugs examined were associated with changes in microbial diversity or changes in the abundance of specific bacteria.
In particular:
Using machine learning models, the team was even able to recognize – starting from the microbial profile alone – not only the use of antibiotics but also that of different human-targeted drugs, highlighting how many molecules share similar microbial signatures. Some bacteria, including Dorea longicatena and Eubacteriumwere particularly sensitive to antibiotics, but not to non-antibiotic drugs, confirming the specificity of the effects.
The next phase of the analysis answered the study’s key question: Do these changes persist over time?
Comparing individuals who had been off a drug for 1, 2, 3, or even 4 years with subjects who had never taken it in the previous five years, the authors found that 42% of the drugs left a detectable imprint on the microbiome even years later.
Among drugs with prolonged effects:
Some antibiotic-related microbial signatures, according to another preprint study cited by the authors, remained observable for several years after treatment.
To strengthen these findings, the researchers analyzed a second fecal sample collected in a subgroup of 328 individuals approximately 4.4 years after the first. It emerged that those who started a new therapy between the two time points showed changes consistent with what was observed in the main cohort, thus confirming the robustness of the conclusions.
The study concludes by underlining a crucial point: limiting ourselves to recording only the drugs taken at the time of sample collection risks producing misleading results in microbiome studies. The authors therefore push for the systematic integration between metagenomics and EHR, considered essential to truly understand how therapies – especially common drugs used by millions of people – influence intestinal balance even long after they have been taken.
The work of Aasmets and colleagues highlights the importance of using electronic medical records as an essential tool in microbiological studies. The EHR reduces bias in self-reported data and allows patients’ medication history to be accurately tracked, opening a more robust avenue for interpreting observed bacterial changes.
Overall, this research shows how common drugs exert an effect that can be immediate, cumulative and, above all, long-lasting. A discovery that forces the scientific community to review its methodologies and consider the complete pharmacological history as one of the main factors to keep under control when investigating the human microbiome.
You might also be interested in:
