Sphenoid sinusitis is inflammation or infection of the sphenoid sinus, an air space deep in the center of the skull behind the eyes and between the temples. It produces a headache at the crown of the head or behind the eyes that is often severe enough to wake a person from sleep, and it typically arrives without the features people associate with sinusitis: no facial tenderness, because the sinus cannot be pressed on from outside, and often no congestion or nasal discharge. Diagnosis is confirmed with a CT scan of the sinuses using thin slices, since the sphenoid is poorly seen on plain films. Amoxicillin-clavulanate is the first-line antibiotic; azithromycin is specifically not recommended for bacterial sinusitis because roughly 1 in 3 pneumococcal isolates are resistant to macrolides. Fishtown Medicine treats sphenoid sinusitis as a diagnosis worth naming, because the sinus drains poorly, sits against the optic nerves and cavernous sinus, and has a higher complication rate than the sinuses people think about.
TL;DR: Most sinus infections announce themselves the way people expect, with facial pressure over the cheeks or forehead, congestion, discolored discharge, and tenderness when you press on the bone. The sphenoid sinus produces almost none of that. It is an air space deep in the center of the skull, behind the eyes and between the temples, sitting directly beneath the pituitary gland and against the optic nerves and the large veins draining the brain. Nobody can press on it from outside, so the tenderness exam that reassures everyone comes back normal. What it produces instead is a headache at the crown of the head or behind the eyes, often severe, often waking a person from sleep, sometimes with a prolonged low-grade fever and very little else. It gets missed for weeks, and it gets treated with the wrong antibiotic often enough that this diagnosis deserves to be recognized by name.
Why does sphenoid sinusitis get missed?
Sphenoid sinusitis gets missed because every feature people use to recognize a sinus infection is absent. The paranasal sinuses most people know sit close to the surface: the maxillary sinuses under the cheeks and the frontal sinuses above the eyebrows, both of which hurt when pressed and both of which produce the congestion and discharge that make the diagnosis obvious. The sphenoid sits far back and deep, behind the nasal cavity and the ethmoid air cells, roughly level with the space between the temples. There is no way to reach it with a thumb.
The result is a physical exam that reassures in the wrong direction. A person presses along their cheeks and forehead, feels nothing but ordinary pressure, and concludes their sinuses are fine. So does the clinician. Nasal symptoms are often mild or absent as well, because the sphenoid drains through a small opening high in the back of the nose, so its secretions run down the back of the throat rather than out the front.
What is left is a headache with a fever and not much else, which sends the workup toward other explanations. In a classic review of 30 hospitalized cases, headache was present in nearly every patient, was frequently severe enough to interfere with sleep, and the diagnosis was commonly delayed while other causes were pursued.1 Larger series of isolated sphenoid disease found the same delay pattern, with headache as the dominant and often the only complaint.2
What does a sphenoid headache feel like?
A sphenoid headache is usually described as pressure at the crown of the head or deep behind the eyes, and it often refers to the back of the head or the neck. People describe heaviness and fullness rather than throbbing, and they frequently describe their eyes feeling tired or heavy by evening without any change in their vision.
Two features make it distinctive. It is often severe out of proportion to how mild everything else feels, and it commonly wakes a person from sleep in the small hours of the morning, which is unusual for tension headache and unusual for the sinuses closer to the surface. Someone may have felt reasonably well through the day, then wake at 3 in the morning with a splitting headache that settles into a persistent background ache afterward.
A prolonged fever without a clear source is the other clue that gets underweighted. Fever running for 2 or 3 weeks with a headache, a normal chest image, normal blood counts, and negative viral testing narrows the possibilities considerably, and a sinus that nobody has imaged belongs on the short list.
The important caution is that this same headache description overlaps closely with raised pressure around the brain, which produces crown-of-the-head pressure, heaviness behind the eyes, and waking from sleep. The features that separate them are worth asking about directly: reproducible worsening when you bend, cough, or strain, seconds-long gray-outs of vision, a whooshing sound timed to your heartbeat, and double vision that resolves when either eye is covered belong to the raised-pressure picture rather than the sinus one. Both can be present in the same person, and finding one is not a reason to stop asking about the other.
Why is the sphenoid sinus the dangerous one?
The sphenoid sinus is the dangerous one because of what surrounds it. Its thin walls border the pituitary gland above, the optic nerves and the cavernous sinuses on either side, and the internal carotid arteries running through those cavernous sinuses. The cranial nerves that move the eyes and carry sensation to the face pass through the same narrow space.
That anatomy explains the complications, which are uncommon but serious: cranial nerve palsies producing double vision or facial numbness, meningitis, cavernous sinus thrombosis, and abscess formation inside the skull. The 30-case review that established the modern picture of this condition documented those complications in patients whose diagnosis had been delayed.1 The larger series of isolated sphenoid disease also found that a meaningful share of cases turned out to be something other than simple infection, including fungal disease, mucoceles, and tumors, which is a second reason imaging matters here rather than empiric treatment alone.23
The sinus also drains poorly. Its single small opening is high and posterior, which is good for gravity when a person is upright and unhelpful for clearance once the opening is inflamed and swollen shut. Saline irrigation and nasal steroid sprays reach the maxillary and ethmoid sinuses reasonably well and reach the sphenoid poorly, so the mechanical help that resolves a routine sinus infection contributes less here. That combination, a sinus that clears slowly and a neighborhood full of structures you cannot afford to inflame, is why treatment tends to run longer and why a poor response gets escalated rather than watched.
Which antibiotic covers sphenoid sinusitis?
Amoxicillin-clavulanate is the first-line antibiotic for acute bacterial sinusitis, and it is the right choice here. The clavulanate matters because it extends coverage to the beta-lactamase-producing organisms that plain amoxicillin misses, which include a substantial share of Haemophilus influenzae and Moraxella catarrhalis. Infectious Diseases Society of America guidance recommends amoxicillin-clavulanate rather than amoxicillin alone as empiric therapy in adults, with doxycycline as the alternative for penicillin allergy.4
The more useful point is which antibiotics do not work. Macrolides, meaning azithromycin and clarithromycin, are specifically not recommended for bacterial sinusitis, because roughly 30% of Streptococcus pneumoniae isolates are resistant to them.4 Trimethoprim-sulfamethoxazole falls out for similar reasons, and second- and third-generation oral cephalosporins are no longer recommended as single-agent therapy. Azithromycin remains one of the most commonly prescribed antibiotics for respiratory complaints, which means a great many people with a bacterial sinus infection have completed a course of something that was never going to treat it and have concluded, reasonably, that antibiotics did not help.
On duration, guideline recommendations for uncomplicated adult sinusitis run 5 to 10 days.45 Clinical practice in sphenoid disease commonly extends beyond that, often to 14 or 21 days, on the reasoning that a poorly draining sinus adjacent to critical structures clears slowly and relapses when treatment is cut short. That extension reflects clinical judgment about a specific anatomic situation rather than a randomized trial, and it should be stated as such. What is not a matter of judgment is the escalation threshold: someone who is not improving within 48 to 72 hours of appropriate therapy, or who develops any visual or neurologic symptom, needs imaging and an ENT opinion rather than a longer course.
How is sphenoid sinusitis diagnosed?
Fishtown Medicine
A 90-minute conversation with Dr. Ash. A written plan you can actually follow.
Diagnosis rests on a CT scan of the sinuses with thin slices, because the sphenoid cannot be assessed any other way. Plain sinus films are inadequate for this sinus, and a standard CT of the head, the kind ordered in an emergency department to exclude a bleed, uses slices too thick to characterize it reliably. The order needs to specify sinus imaging with thin cuts. MRI is added when a complication is suspected, when there is any concern for spread inside the skull, or when the CT appearance suggests fungal disease or a mass rather than routine inflammation.
Nasal endoscopy performed by an ENT specialist can visualize the sphenoid opening directly and obtain a culture, which changes management when the first antibiotic has failed or when an unusual organism is likely. In an otherwise healthy adult with a classic story, treating empirically with the right antibiotic and reserving imaging for poor response is a defensible approach, and it is what many people choose when the alternative is a scan they are paying for out of pocket. That approach carries one condition: the escalation threshold has to hold and it has to be short. A headache that has not started improving within 3 days of amoxicillin-clavulanate, or any new visual, eye-movement, or neurologic symptom at any point, means the imaging happens now.
Guidance from the Clinic
How Fishtown Medicine approaches a sphenoid headache
Fishtown Medicine starts by taking seriously the headache that has already been evaluated and dismissed, because a normal facial tenderness exam is the single most misleading finding in this condition. That means asking where the pressure is, whether it woke you from sleep, what has already been prescribed, and whether anything about your vision or eye movements has changed.
When the picture fits, treatment starts with amoxicillin-clavulanate at an appropriate dose, or doxycycline where penicillin allergy makes that necessary, paired with high-volume saline irrigation and intranasal fluticasone. Those adjuncts reach the sphenoid imperfectly and they are still worth using, since the ethmoid cells that sit in the drainage path benefit from both. Fishtown Medicine arranges the CT sinus study when imaging is indicated, including finding a reasonable self-pay facility for patients without coverage, and refers to highly qualified in-network ENT specialists for endoscopy or any procedure, while Dr. Ash stays in the case and compares notes with those colleagues so the next decision gets made quickly rather than after another 3-week wait.
Seek same-day medical care if you develop:
- Double vision, drooping of an eyelid, or any change in vision
- Numbness over the forehead, cheek, or around the eye
- Swelling or redness around the eye, or an eye that bulges forward
- Neck stiffness, confusion, or a fever above 102°F with the headache
- A headache that suddenly becomes much worse
- No improvement at all after 3 days on an appropriate antibiotic
If you are in the Philadelphia area and your headache has not been explained, tell Dr. Ash what's going on at Fishtown Medicine.
Key Takeaways
- The sphenoid sinus is deep in the center of the skull behind the eyes, so it produces no facial tenderness and often no congestion, which makes a reassuring sinus exam meaningless for this diagnosis.
- The characteristic picture is a headache at the crown of the head or behind the eyes, often severe, often waking a person from sleep, sometimes with a prolonged low-grade fever and little else.
- Amoxicillin-clavulanate is first-line, with doxycycline for penicillin allergy. Azithromycin is specifically not recommended for bacterial sinusitis, because roughly 30% of pneumococcal isolates are macrolide-resistant.
- Guideline duration for adult sinusitis is 5 to 10 days, and sphenoid disease is commonly treated for 14 to 21 days in practice because the sinus drains through one small opening and relapses when therapy is short. Improvement should still begin within 48 to 72 hours.
- Diagnosis needs a CT of the sinuses with thin slices. Plain films and routine head CT do not characterize this sinus adequately.
- The sphenoid borders the optic nerves, the cavernous sinuses, the carotid arteries, and the nerves that move the eyes, so any visual, eye-movement, or facial-sensation symptom means imaging and an ENT opinion the same day.
- The same headache description fits raised intracranial pressure, and the 2 can coexist, so treating the sinus does not remove the reason to look at the optic nerves.
Related at Fishtown Medicine
- Idiopathic Intracranial Hypertension - the other cause of crown-of-the-head pressure with heaviness behind the eyes
- Sinus Infection Treatment in Philadelphia - when sinus symptoms warrant antibiotics and when they do not
- Droopy Eyelid and Tired Eyes - what eye-movement and eyelid changes mean, and how to sort them
- Migraine vs. Serious: When to Worry About a Headache - the SNOOP red-flag checklist
- Headache Doctor Philadelphia - the wider workup for chronic and recurring headaches
- Congestion and Respiratory Relief - saline irrigation, nasal steroids, and what each one does
Scientific References
- Lew D, Southwick FS, Montgomery WW, Weber AL, Baker AS. "Sphenoid Sinusitis: A Review of 30 Cases." New England Journal of Medicine. 1983;309(19):1149-1154.
- Lawson W, Reino AJ. "Isolated Sphenoid Sinus Disease: An Analysis of 132 Cases." Laryngoscope. 1997;107(12 Pt 1):1590-1595.
- Wang ZM, Kanoh N, Dai CF, et al. "Isolated Sphenoid Sinus Disease: An Analysis of 122 Cases." Annals of Otology, Rhinology and Laryngology. 2002;111(4):323-327.
- Chow AW, Benninger MS, Brook I, et al. "IDSA Clinical Practice Guideline for Acute Bacterial Rhinosinusitis in Children and Adults." Clinical Infectious Diseases. 2012;54(8):e72-e112.
- Rosenfeld RM, Piccirillo JF, Chandrasekhar SS, et al. "Clinical Practice Guideline (Update): Adult Sinusitis." Otolaryngology-Head and Neck Surgery. 2015;152(2 Suppl):S1-S39.
Frequently Asked Questions
Common Questions
Deep-Dive Questions
Ready when you are
Dr. Ash reads every intake himself, and answers questions personally - usually within a few hours.





