Antibiotics don't help fight most sinus infections, although doctors routinely prescribe them for that purpose, researchers said.
The medications are known to fuel the evolution of drug-resistant bacteria and experts have grown increasingly worried about overuse.
That's a particular concern with sinus infections, because doctors can't tell if the disease is caused by bacteria or by a virus, in which case antibiotics would be worthless.
The new study, published in the Journal of the American Medical Association, found antibiotics didn't ease patients' symptoms or get them back to work any sooner than an inactive placebo pill.
Refrain from antibiotics
"There is not much to be gained from antibiotics," said Dr Jane Garbutt of Washington University School of Medicine in St. Louis, who led the work.
"Rather than give everybody an antibiotic hoping to find the ones with bacteria, our findings would suggest refraining from antibiotics and doing what we call watchful waiting," she told Reuters Health.
That involves keeping an eye on patients to see if they get better, but not using other drugs than over-the-counter painkillers.
People with sinus infections, also called acute sinusitis, have lasting and severe cold-like symptoms such as runny nose and pain around the eyes, the nose or the forehead.
Fifth most common reason for antibiotics
"It's the fifth most common reason antibiotics are prescribed for adults," said Garbutt. "It's hard for doctors not to give an antibiotic because patients are so miserable and we don't have anything else to give them."
But it hasn't been clear whether the drugs actually work, and even the studies that have found some benefit see the majority of placebo-treated patients get better by themselves.
Garbutt and her colleagues used official US guidelines to identify patients with sinus infections. They randomly assigned 166 adults to either placebo pills or a 10-day treatment with the antibiotic amoxicillin.
Based on patient ratings on a symptom scale known as the modified Sinonasal Outcome Test-16, or SNOT-16, the researchers found little difference between the two patient groups.
Few sinus infections are bacterial
Using the aptly named scale, where 0 equals "no problem" and 3, a "severe problem," the antibiotic group rated their symptoms at 1.12 after three days, while the placebo group averaged 1.14. After seven days, there were signs of benefit from the antibiotic, but the effect was small and had vanished another three days later.
Neither were there differences in how many work days patients missed, how satisfied they were with their treatment, how often they relapsed or how much additional medical care they got.
After 10 days, 78% of the people on antibiotics and 80% of the placebo-treated people said they felt a lot better or no longer had symptoms.
Dr Anthony Chow, an expert in infectious diseases at the University of British Columbia in Vancouver, Canada, said less than 2% of sinus infections are bacterial.
Antibiotics still have a place
"Most cases are viral, and the vast majority don't require antibiotics," he told Reuters Health. "Antibiotics have been abused, so there is a need to be more cautious in prescribing them and to hold back."
But Chow said antibiotics still have a place.
He recently chaired a committee at the Infectious Diseases Society, which has developed guidelines to help spot infections that are more likely to be bacterial.
Those guidelines, still in press, recommend treating only patients whose symptoms last for at least 10 days and keep worsening; who are severely sick with high fever and other symptoms; or who improve and then get worse again.
(Frederik Joelving, Reuters Health, February 2012)