Could some common antibiotics be inflaming bacterial infections?
The antibiotics most often used to treat sepsis may trigger bacteria to release more inflammation-driving particles than other antibiotic classes, a new study shows.
- 30 April 2026
- 4 min read
- by Priya Joi
At a glance
- Beta-lactam antibiotics (the mainstay of sepsis treatment) made E. coli bacteria release three to five times more inflammatory particles than untreated bacteria, although the drugs remain critical to treating deadly conditions like sepsis.
- The particles, called bacterial extracellular vesicles, can worsen the body’s response to infection.
- Other antibiotic classes caused far less release, suggesting antibiotic choice may shape how sick sepsis patients become.
Antibiotics that disrupt the bacterial cell wall, the class most commonly prescribed in sepsis, caused bacteria to shed up to five times more inflammation-driving particles than untreated controls, according to new laboratory research.
The authors say that while antibiotics are critical to saving lives from conditions such as sepsis, the results point to antibiotic choice as an underexplored factor in how sepsis patients respond to treatment. Researchers at the Rochester Institute of Technology and the University of Rochester Medical Center, New York, USA, publishing in the International Journal of Antimicrobial Agents, compared nine antibiotics routinely used to treat sepsis.
Earlier research had suggested that antibiotics can drive bacteria to release more inflammatory particles, and the team wanted to know whether the effect depended on how each drug killed bacteria.
They found that antibiotics that target the bacterial cell wall triggered a sharp rise in the release of inflammatory material, compared to those that target other bacterial processes.
Co-author Panteha Torabian, PhD candidate at the Rochester Institute of Technology, said, “It is important to emphasize that antibiotics remain one of the most effective and lifesaving tools in modern medicine.” She explains, “This research does not suggest they should be avoided. Instead, it highlights that bacteria are not passive targets.”
Tiny bubbles, big effects
Sepsis kills around 11 million people a year globally, according to World Health Organization estimates, and accounts for roughly one in five deaths worldwide. The burden falls hardest on low- and middle-income countries, which account for around 85% of cases and deaths.
What makes sepsis so deadly is not just the bacterial infection but the body's dysregulated immune response to it, which damages tissue, triggers organ failure and overwhelms the cardiovascular system.
The team knew that bacteria naturally shed nano-sized spheres called bacterial extracellular vesicles, or BEVs. They carry proteins, toxins and genetic material and bacteria use them to communicate with host cells and shed damaged material.
But the research team also knew that when BEVs reach the bloodstream, the molecules they carry can activate the immune system and intensify inflammation, so they tested to see whether this could affect treatment.
The antibiotic test
Some antibiotics, like the beta-lactams used in sepsis, attack the bacterial cell wall until the cell breaks apart. Others interfere with how bacteria replicate or build proteins.
To see how different antibiotics influenced the release of the inflammatory nanospheres, the researchers exposed a clinical strain of Escherichia coli to nine drugs: six beta-lactams (ceftriaxone, piperacillin, imipenem, ertapenem, meropenem and cefepime), two aminoglycosides (tobramycin and amikacin) and one quinolone (ciprofloxacin).
Each antibiotic was tested at two doses: twice the minimum concentration needed to stop bacterial growth, and a dose comparable to what patients receive in the clinic.
They then isolated the nanospheres the bacteria released and found that the pattern was consistent.
Beta-lactams, which kill bacteria by disrupting cell wall synthesis, triggered three to five times more BEVs than untreated bacteria. They also produced roughly four times more vesicles than aminoglycosides and two to three times more than the quinolone tested.
Aminoglycosides, which interfere with protein production, caused almost no increase in vesicle release compared to untreated bacteria.
Beta-lactams destabilise the bacterial cell wall, and previous work has shown that weakening the connections between the outer membrane and the cell wall increases vesicle release.
Under that kind of stress, bacteria appear to shed more membrane material, and with it, more of the molecular signals that activate the human immune system.
Dose matters
The study also found that higher doses drove much higher vesicle release. When ertapenem, meropenem and cefepime were tested at clinical concentrations, which were more than 100 times higher than the minimum effective dose, BEV production rose sharply.
This matters because sepsis patients typically receive broad-spectrum antibiotics at aggressive doses, often before the specific bacterial cause of infection has been identified.
The logic is sound: clinicians need to cover the widest possible range of pathogens, fast. But if higher doses of certain drugs also mean more inflammatory material pouring into the bloodstream, the calculation becomes more complicated.
Cautious conclusions
The authors are cautious about how far their findings can be pushed.
The study was conducted in the laboratory, using a single clinical strain of E. coli. Only one quinolone and two aminoglycosides were tested, compared with six beta-lactams, making direct class comparisons uneven. Whether the same pattern holds across different bacterial species, different strains or different patients, remains to be seen.
The authors are also clear that their results are not a case against antibiotics. The drugs remain essential, and sepsis is fatal without them.
What the study does suggest is that the choice of antibiotic may influence not only how effectively an infection is cleared, but also how inflamed a patient becomes in the process.
For a condition where inflammation is itself the killer, that distinction could matter. Further research, including studies in animal models and eventually in patients, will be needed to determine whether adjusting antibiotic choice in sepsis could reduce the inflammatory burden on patients without compromising bacterial kill rates.
More from Priya Joi
Recommended for you
