Not every toothache comes from a tooth. Referred muscle pain, joint pain, neuralgias, neuropathic and idiopathic pain, headache-type disorders and sinus disease can all be felt in a tooth — and the irreversible procedures that follow a wrong diagnosis do not relieve them. This guide covers when to suspect a non-dental source, the main mimics, and an examination routine that separates them before you pick up a bur.
Last updated: September 2026
Pain felt in the dentoalveolar region is usually pulpal or periapical — but not invariably. Masticatory muscles, the temporomandibular joint, the trigeminal nerve, central pain processing, headache mechanisms and the maxillary sinus can all project pain onto a tooth that is perfectly healthy or already adequately treated. These patients tend to arrive with a convincing story, sometimes a previous restoration or root canal that "did not help", and a request to do something more. The diagnostic trap is that every dental procedure offered next is irreversible, while none of them treats the actual source.
Before any irreversible treatment, the pain complaint should be reproduced and explained by a dental finding: a pulp that responds abnormally, a tooth that is tender in a way that matches the story, or a lesion on the image. Pain that cannot be provoked from the tooth, that changes with jaw function, that comes in electric jolts, that persists after adequate treatment, or that arrives with numbness, swelling or autonomic signs deserves a non-dental differential first.
Non-odontogenic pain rarely announces itself. It is usually recognised from a mismatch between what the patient describes and what the tooth shows on examination.
None of these clues is diagnostic on its own. A cracked tooth can produce diffuse, intermittent pain that mimics muscle pain, and normal sensibility tests do not exclude an odontogenic source. The clues are a reason to widen the examination, not to close the case.
The conditions below account for most non-dental "toothache". Each has a characteristic pattern and at least one bedside finding that points away from the tooth.
Typical pattern
Deep, dull, aching pain in the jaw, temple or ear region that is modified by jaw function or clenching; the most common non-dental source of persistent "tooth" pain. Masseter and lateral pterygoid commonly refer to the teeth; the temporalis tendon can refer to upper teeth.
Bedside clue
Standardised palpation of the temporalis and masseter (about 1 kg of pressure, held 2 seconds for local pain and 5 seconds to look for referral) or maximal opening reproduces the patient's familiar pain and spreads it beyond the muscle, into the tooth.
Typical pattern
Pain in or in front of the ear, modified by jaw movement, function or parafunction; may accompany clicking, locking or limited opening.
Bedside clue
Palpation around the lateral pole of the condyle or provocation with opening, lateral or protrusive movements reproduces the familiar pain. The tooth itself tests normally.
Typical pattern
Recurrent, unilateral, electric-shock-like paroxysms lasting a fraction of a second to about two minutes, confined to one or more trigeminal divisions and triggered by innocuous stimuli such as light touch, talking, chewing or brushing. Some patients also have a continuous background pain.
Bedside clue
A trigger zone that can be demonstrated, a refractory period after an attack, and no dental finding that explains the paroxysms. Pain of this quality is a reason to refer, not to treat a tooth.
Typical pattern
Constant or mixed pain in a neuroanatomically plausible territory that began within months of a nerve injury — including dental injections, extractions, root canal treatment, implants or surgery — and has persisted beyond three months.
Bedside clue
Altered sensation in the same territory: numbness or reduced sensitivity, or the opposite, hypersensitivity and pain to touch (allodynia). Further surgery in the area tends to worsen rather than relieve it.
Typical pattern
Persistent, usually unilateral, deep, dull, pressure-like pain localised to a tooth or alveolar site, present daily for more than two hours a day and for longer than three months, without a preceding causative event; may be aggravated by stress and may spread over time.
Bedside clue
Normal clinical and radiographic examination once local causes have been excluded, often a trail of restorations, root canals or extractions that changed nothing. Somatosensory testing may or may not show changes.
Typical pattern
Attacks of moderate-to-severe, toothache-like or pulsating intraoral pain lasting hours, without head pain, accompanied by ipsilateral tearing, nasal congestion, cheek swelling, light or sound sensitivity, or nausea.
Bedside clue
Autonomic or migrainous accompaniments with a normal dental examination. A pain diary is often the most useful diagnostic tool here.
Typical pattern
Dull, aching, poorly localised pain across several upper posterior teeth, with pressure under the eyes, worse when bending forward or lying down, and often accompanied by congestion, nasal discharge or a recent cold.
Bedside clue
Several adjacent maxillary teeth are equally tender to percussion, pulp tests are normal, and cheek palpation reproduces the pressure. Note that the reverse also exists: a necrotic upper molar can cause sinusitis, so the pulp tests still matter.
Typical pattern
Pain in one or several teeth, sometimes spreading from tooth to tooth, in a patient with another diagnosed orofacial, neck or widespread pain condition; can range from hypersensitivity to lingering pain and often fluctuates with the other condition.
Bedside clue
Signs of central sensitisation such as pain referral, temporal summation or allodynia, and pain that is not consistently relieved by local anaesthesia or peripherally acting analgesics.
Cardiac pain referred to the jaw, salivary gland disease, and rare intracranial or neoplastic causes are much less common but far more dangerous — they belong in the red-flag section below rather than in this table.
A structured routine finds the source in most cases and produces a defensible record when it does not. The order matters: reproduce the complaint first, then test the tooth, then image.
Ask the patient to describe the pain they came for and then try to provoke exactly that pain. Percuss and palpate the suspected and neighbouring teeth, apply cold or an electric pulp test, and ask each time whether this is the familiar pain or a different sensation. A response that is not the familiar pain is not evidence for the tooth.
Palpate the temporalis and masseter with a standardised, firm pressure and hold it long enough to allow referral; palpate around the condyle; measure opening and note pain on maximal opening, lateral and protrusive movements. Ask specifically whether the provoked pain is the same pain and whether it travels into the tooth. Most patients with referred masticatory pain have not previously been told they have a muscle problem.
Compare sensibility responses with control teeth, check for lingering versus non-lingering responses, probe for an isolated deep pocket, test each cusp for bite-release pain, and transilluminate for cracks. A normal, symmetrical sensibility response with a normal periodontium makes an inflamed or necrotic pulp unlikely — but a cracked tooth can still be present, so bite testing is part of the routine.
Brush the skin and mucosa lightly to look for trigger zones or allodynia and compare touch sensation side to side. Ask about congestion, discharge, a recent cold and positional change; palpate the cheeks; percuss the whole posterior maxillary segment rather than one tooth. Ask about headache history, other chronic pain and previous injuries or surgery in the region.
A periapical radiograph is part of every examination, but a small radiographic irregularity does not explain a pain that behaves neurologically. Reserve CBCT for a specific question — a missed canal, a suspected fracture, a sinus floor relationship — rather than as a search for something to treat. In most patients with non-odontogenic pain the images show nothing that explains the complaint.
A selective infiltration that abolishes the pain supports a local source, and complete relief in a well-localised pain is reassuring. Relief is less specific than it seems, though: sinus and dental innervation overlap, referred pain can be reduced by anaesthetising the target area, and a placebo response is common. Persistent pain despite profound anaesthesia is the more informative result.
The patient whose tooth still hurts months after a technically adequate root canal is the classic presentation of non-odontogenic pain, and also the situation in which the next irreversible step is most tempting.
A tooth that has already been treated and re-treated without relief should be presumed non-odontogenic until a specific odontogenic finding is demonstrated. "Let's try one more thing" is the phrase to be wary of.
A small number of presentations need urgent or specialist evaluation regardless of what the teeth show.
How common is toothache that does not come from a tooth?
Short answer
In general practice most toothache is odontogenic, so the absolute numbers are modest.
The pulp tests are normal but the patient insists it is the tooth. Should I treat it anyway?
Short answer
Generally not on the strength of insistence alone.
Does relief with local anaesthetic prove the tooth is the source?
Short answer
It supports a local source but does not prove it.
How do I tell trigeminal neuralgia from pulpitis?
Short answer
The quality of pain alone is not enough, because pulpitis can also be sharp and spontaneous.
What is persistent idiopathic dentoalveolar pain?
Short answer
It is the current name for what used to be called atypical odontalgia or phantom tooth pain: a persistent, usually unilateral, deep, dull, pressure-like pain in a tooth or alveolar site, present daily for more than two hours and for longer than three months, with normal clinical and radiographic findings once local causes have been excluded.
The patient had a cold last week and now several upper teeth hurt. Sinus or tooth?
Short answer
Several adjacent upper posterior teeth that are equally tender to percussion, with normal sensibility responses, dull pressure under the eye and pain that worsens on bending forward, point toward the sinus.
When should I refer rather than keep investigating?
Short answer
Refer when the pain behaves neurologically (paroxysms, allodynia, altered sensation), when muscle or joint provocation clearly reproduces it, when the dental work-up is complete and negative, when previous dental treatment has not helped, or when any red flag is present.
Continue with the guides that this differential feeds into.
This guide is an educational reference for dental professionals. It is not a substitute for clinical judgment, individual patient assessment, or referral to orofacial pain, neurology or ENT colleagues where indicated.
Reviewed by
Uzm. Dt. Levent Yüksel
DDS · Endodontist
Endodontist — DDS, Hacettepe University; endodontics specialty training, Ankara University. 10+ years of clinical experience. Sole author and reviewer of EndoGuide content.