EXPLAINER

Do you need antibiotics for a sinus infection? For most people, no

Most acute sinus infections clear without antibiotics — about 46% of people are better within a week regardless. Drugs speed cure in only 5 to 11 per 100, and cause side effects in more.

Number needed to treat with antibiotics for one extra cure, by how sinusitis is diagnosedDiagnosed on symptoms: 19people; Confirmed on X-ray: 10people; Fluid or opacity on CT: 4people0people150people300peopleDiagnosed on symptoms19peopleConfirmed on X-ray10peopleFluid or opacity on CT4people
Number needed to treat with antibiotics for one extra cure, by how sinusitis is diagnosed
GroupValue (people)
Diagnosed on symptoms19 (10 to 205)
Confirmed on X-ray10 (5 to 136)
Fluid or opacity on CT4 (2 to 15)
Number needed to treat with antibiotics for one extra cure, by how sinusitis is diagnosed Cochrane 2018. Lower is better: fewer people treated per additional cure. Bars show the number needed to treat with its 95% confidence interval. Source: Cochrane Database of Systematic Reviews

The honest answer to whether a sinus infection needs antibiotics is that most do not. In pooled trials, about 46% of people with acute rhinosinusitis were cured within a week, and 64% within 14 days, without any antibiotic at all [s1]. Antibiotics can shorten the illness, but only 5 to 11 more people in every 100 are cured faster because of them [s1] — and more people are harmed by side effects than are helped to a quicker cure [s1].

What the trials actually found

A 2018 Cochrane review pooled 15 trials involving 3,057 adults with acute rhinosinusitis, the medical name for a sinus infection lasting less than four weeks [s1]. When sinusitis was diagnosed the ordinary way, on symptoms alone, antibiotics raised the odds of cure only modestly (odds ratio 1.25, 95% confidence interval 1.02 to 1.54) — a number needed to treat of 19 (95% CI 10 to 205), meaning nineteen people take the drug for one extra person to be cured faster [s1].

The picture shifts with how clearly bacterial the infection looks. Where a sinus infection had been confirmed on an X-ray, the effect was larger (OR 1.57, 95% CI 1.03 to 2.39; number needed to treat 10) [s1]. Where a CT scan showed a fluid level or a completely opaque sinus, larger still (OR 4.89, 95% CI 1.75 to 13.72; number needed to treat 4) [s1]. Purulent nasal secretion also resolved faster with antibiotics (OR 1.58, 95% CI 1.13 to 2.22) [s1]. The consistent pattern: the sicker and more clearly bacterial the sinus, the more an antibiotic does — and most everyday sinus infections are neither.

The harm on the other side of the ledger

Against that modest benefit sits a real cost. In the same review, 13 more people out of 100 had side effects on antibiotics than on placebo (OR 2.21, 95% CI 1.74 to 2.82), a number needed to harm of 8 [s1]. Antibiotics did cut outright clinical failure by about 5 fewer people per 100 (Peto OR 0.48, 95% CI 0.36 to 0.63) [s1], but for an illness most people shake off on their own, treating everyone to prevent that is the same stewardship calculus that drives the advice on antibiotics for colds and on sore throats. Every unnecessary course also feeds the resistance that makes antibiotics less useful when they are genuinely needed.

What the guideline recommends

The American Academy of Otolaryngology's 2015 adult sinusitis guideline, built on evidence from 42 systematic reviews, tells clinicians first to separate presumed bacterial sinusitis from the far more common viral kind [s2]. For an uncomplicated bacterial sinus infection it explicitly allows either watchful waiting without antibiotics or an initial prescription — a shared decision, not an automatic one [s2]. If antibiotics are chosen, the first-line drug is amoxicillin, with or without clavulanate, for 5 to 10 days [s2]. The guideline also says to reassess anyone who worsens or fails to improve by 7 days after diagnosis, and not to order X-rays or scans for a straightforward case unless a complication or another diagnosis is suspected [s2].

What actually helps the symptoms

Because the infection will usually clear regardless, the defensible aim is comfort. The guideline lists analgesics, topical intranasal steroids and saline nasal irrigation as options for symptom relief in both viral rhinosinusitis and bacterial rhinosinusitis [s2]. None of these shorten the infection; they make the days more bearable, the same reasoning behind decongestants for a stuffy nose. For chronic sinusitis — symptoms that drag on for months rather than weeks — the guideline recommends saline irrigation, intranasal corticosteroids, or both [s2].

When to see a doctor

The evidence here concerns ordinary, uncomplicated sinusitis in otherwise healthy adults [s2]. Symptoms that warrant prompt medical attention rather than watchful waiting include swelling or redness around the eye, changes in vision or double vision, a severe or worsening headache, confusion or neck stiffness, a high fever that will not settle, or symptoms that improve and then sharply worsen. Sinus symptoms in someone with a weakened immune system, or that persist well beyond ten days without any improvement, are also a reason to seek care. Whether any particular sinus infection is bacterial, and whether it needs an antibiotic, is a clinical judgement this article does not make — the evidence simply shows that, for most people most of the time, the drug is optional.

Sources

  1. Antibiotics for acute rhinosinusitis in adults — Cochrane Database of Systematic Reviews , September 10, 2018
  2. Clinical Practice Guideline (Update): Adult Sinusitis — Otolaryngology–Head and Neck Surgery (American Academy of Otolaryngology–Head and Neck Surgery Foundation) , April 1, 2015

More on

Related coverage