Some Antibiotics Leave Traces in the Gut Microbiome Up to Eight Years Later

Published in Nature Medicine. Principal investigator Tove Fall, Uppsala University. Study of 14,979 Swedish adults.

A single course of clindamycin, and your gut bacteria may still bear the mark up to eight years later. That is the central finding of a study published in Nature Medicine that tracked antibiotic prescriptions and gut microbiome composition in nearly 15,000 Swedish adults -- the largest analysis of its kind.

The results vary dramatically by antibiotic type. Clindamycin, fluoroquinolones, and flucloxacillin showed the strongest long-term associations with altered gut microbiome composition. Penicillin V, Sweden's most commonly prescribed antibiotic for outpatient infections, was linked to only small and short-lasting changes.

Matching prescriptions to microbiomes at population scale

The study, led by researchers at Uppsala University, exploited two distinctive features of the Swedish healthcare system. First, Sweden maintains a comprehensive prescribed drug register that records every antibiotic dispensed at a pharmacy. Second, Swedish biobanks at Uppsala and Lund University contain detailed gut microbiome data from population-based cohorts.

By linking these two data sources, the research team could compare the gut microbiome composition of participants who had received specific antibiotics over the previous eight years against those who had not received any antibiotics during that period. The 14,979 participants came from four Swedish population-based cohorts: SCAPIS-Uppsala, SIMPLER, SCAPIS-Malmo, and MOS.

Gabriel Baldanzi, the study's first author and a former doctoral student at Uppsala University, described the results as showing that antibiotic use as far back as four to eight years ago is linked to the composition of a person's gut microbiome today. Even a single course of treatment with certain types of antibiotics leaves traces.

Not all antibiotics are equal

The most important practical finding is how strikingly different antibiotics varied in their long-term microbiome footprint. The three with the strongest associations -- clindamycin, fluoroquinolones, and flucloxacillin -- are all used for specific clinical situations. Clindamycin is commonly prescribed for bone and joint infections, dental infections, and as an alternative for penicillin-allergic patients. Fluoroquinolones are broad-spectrum antibiotics used for urinary tract infections, respiratory infections, and other conditions. Flucloxacillin targets staphylococcal infections.

Penicillin V, the workhorse antibiotic for strep throat and other common outpatient infections in Sweden, showed a markedly milder footprint. This is reassuring given its high prescription volume, but it also raises the question of whether clinicians should factor in long-term microbiome impact when choosing between equally effective antibiotics.

Tove Fall, Professor of Molecular Epidemiology at Uppsala University and the study's principal investigator, noted that the strong link between flucloxacillin and the gut microbiome was unexpected. She expressed interest in seeing the finding confirmed by other studies, while suggesting that the results may help inform future recommendations on antibiotic selection, especially when two equally effective options differ in their microbiome impact.

Why the gut matters years later

The gut microbiome is increasingly recognized as a player in conditions far beyond gastrointestinal health. Epidemiological studies have linked high antibiotic use to increased risk of type 2 diabetes, Clostridioides difficile infections, inflammatory bowel disease, and other conditions. The mechanisms are not fully understood, but disruption of the gut microbiome is thought to be a contributing factor.

The finding that certain antibiotics leave detectable traces years after treatment provides a plausible biological pathway for these associations. If a course of clindamycin shifts the gut bacterial community in ways that persist for years, those persistent changes could theoretically influence metabolic, immune, and gastrointestinal function over the same timeframe.

The word "theoretically" deserves emphasis. The study demonstrates an association between past antibiotic use and current microbiome composition. It does not prove that the antibiotic-induced microbiome changes cause downstream health effects. Other factors -- the infections that prompted the antibiotic prescription, for instance, or health behaviors correlated with antibiotic use -- could contribute to both the microbiome changes and any associated health outcomes.

Boundaries of what the data show

The study covered only prescriptions from the previous eight years, limited by the available registry data. Antibiotic exposure earlier in life, including childhood courses that may have the greatest impact on microbiome development, was not captured.

Each participant's gut microbiome was sampled only once. Without serial measurements, the researchers cannot distinguish between antibiotics that cause persistent microbiome changes and antibiotics that happen to be prescribed more often to people whose microbiomes were already different for other reasons.

The population is Swedish, with Sweden's particular patterns of antibiotic prescribing, diet, and healthcare. Antibiotic prescribing practices vary substantially across countries, and the results may not translate directly to populations with different prescribing norms.

Baldanzi emphasized that antibiotic stewardship in Sweden is already strict, and that the findings should not discourage patients from following their doctors' recommendations. The study is about understanding long-term consequences that are rarely considered, not about avoiding antibiotics when they are medically necessary.

A second sample is coming

The research team is currently collecting a second microbiome sample from nearly half the participants. This follow-up will enable a more precise analysis of recovery time and could identify which gut microbiomes are more susceptible to disruption following antibiotic treatment -- potentially pointing toward patient-specific risk factors for long-term microbiome changes.

For clinicians, the immediate takeaway is nuanced: antibiotics remain essential and often life-saving, but the choice of which antibiotic to prescribe may carry consequences that extend years beyond the treatment course. When two drugs are equally effective against a given infection, their long-term microbiome impact may be worth weighing in the decision.