Timothy C. Hain, MD •Page last modified: October 17, 2022
You may also be interested in our many pages on migraine on this site
![]() |
![]() |
Image from https://www.scienceofmassage.com/dnn/som/journal/1204/medical.aspx | Origin of occipital nerves from https://en.wikipedia.org/wiki/File:Gray800.png. View is from back of skull. |
Occipital neuralgia is usually due to trauma to the occipital nerve (ON), often caused by an auto-accident where the head impacts the headrest. Other causes are spondylosis of the upper cervical spine (C1-C2), or rarely focal neuropathies due to diabetes or tumor (Ehni and Benner, 1984). Chiropractic manipulations of the neck are another potential source.
The occipital nerve may become entrapped beneath the attachments of the trapezius and semispinalis capitis muscles to the occipital bone (Loukas et al, 2006).
In occipital neuralgia, there are paroxysms of severe occipital pain, that often resemble severe migraines. The pain may be so severe that blood pressure rises to extreme levels. Some authors report eye pain from occipital neuralgia. (Mason et al, 2004), and even dental pain has been reported (Sulfaro et al, 1995). Experimental injections in humans of the greater occipital nerve cause severe pain in the trigeminal distribution (Piovesan et al, 2001).
Some patients with occipital neuralgia also have dizziness, presumably due to a variant of cervical vertigo. See our cervical vertigo page for more about this complicated subject.
Many patients with occipital neuralgia also experience typical migraine symptoms. This may be due to "convergence" of pain input triggering a common pathway of migraine. (Cady, 2007)
There are two branches to the occipital nerve -- the greater and lesser. Most of the time, the injury is to the greater ON. The ON takes most of its origin from the C2 nerve root. Damage to the C2 nerve root, and possibly also the upper cord, can cause occipital neuralgia.
A comprehensive review of occipital neuralgia can be found in an article by Vanelderen et al. (2010).
Occipital neuralgia is very uncommon, at least as compared to migraine. In our practice, as of the end of 2014, we had 30 patients diagnosed with occipital neuralgia, compared to nearly 3000 patients with migraine. Thus in our otoneurology practice, the ratio of ON to migraine is about 1:100. In our patients, females predominate (25/30), and the average age is 52. This is a similar distribution to our migraine patients.
There is a greater prevalence of litigation in patients with occipital neuralgia than in patients with most other conditions. This is because head or neck trauma is the usual mechanism of injury in occipital neuralgia.
For a person with severe tenderness just below their occiput, a list of the most common sources of pain include:
Less common possibilities
Like many other common health conditions (e.g. psychiatric conditions, migraine), occipital neuralgia is diagnosed solely from symptoms -- there are no blood or imaging studies that can prove that symptoms after a traumatic injury are "real" as opposed to "made up" in an attempt to obtain some benefit, perhaps compensation for an auto accident. Nevertheless, usually an astute clinician can draw a reasonable conclusion - - relief from blocks, and signs of severe pain (such as hypertension) during attacks often allow one to draw a reasonable inference.
An MRI or CT scan of the skull base is the most common test. A CT scan of the cervical spine is probably the most useful, because it visualizes the cervical facet joints. However, a reasonable case might also be made for MRI with soft tissue imaging of the neck, after trauma, looking for objective evidence of damage.
Vascular imaging may be done to look for carotid or vertebral dissection or vascular compression. This is extremely unlikely however and we don't recommend vascular imaging as a routine investigation.
Occipital neuralgia can be extremely painful, and there are several treatment approaches. In general nerve blocks are used. Medications are usually not helpful for occipital neuralgia, but when ON is combined with migraine (which is common), then it makes sense to treat both. (Sahai-Srivastava et al, 2011).
Blocks are injections of medication intended to temporarily deaden pain nerves. They are ordinarily done by anesthesiologists in a pain clinic, or neurologists in a headache clinic. An example is shown above. For occipital neuralgia, if the site of injury is the nerve itself such as when the nerve is bruised on the headrest of a car, the nerves should be blocked. The nerves have a fairly long course and several papers have been written concerning the optimal location to block (e.g. Natsis et al, 2006).
If the site of injury is one of the upper cervical nerve roots, then a more complex C2 cervical nerve block may need to be used. This generally requires X-ray control.
If a block works temporarily, it usually wears off as the anesthetic effect stops. To obtain a more lasting effect, a more permanent procedure is to damage the nerve. Here, partial nerve injury could make the nerve even more irritable, and complete nerve destruction could lead to denervation pain.
If the occipital nerve block doesn't work, it is likely that the pain is coming from somewhere else. We have encountered patients with cervical facet disease who have pain resembling ON. As the cervical facet nerve is closer to the spinal cord than the occipital nerve, blocking the occipital nerve leaves cervical facet pain untouched. Similarly, an injury to the occipital nerves close to the spinal cord prior to emerging into the skull would be untouched by a peripheral occipital nerve block.
![]() |
Xray showing leads for occipital nerve stimulator. |
As a general comment, these all involve doing something fairly long lasting -- usually damaging the occipital nerve.
- Radiofrequency ganglio-neurectomy (RFGN) involves damaging a nerve by heating it with microwaves. It is less invasive than rhizotomy as one just needs a needle rather than open surgery. This procedure is usually performed by pain clinic physicians. According to Vanelderen (2010), pulsed radiofrequency treatment is a promising treatment. It has roughly a 50% success rate, with success being correlated with use of several procedures. (Huang et al, 2012; Choi et al, 2012; Ducic et al, 2014; Hamer et al, 2014). A problem with RFGN is denervation pain -- which may be worse than the original neuralgia.
- Decompression surgery of the occipital nerve can be done at several sites. Ducic et al (2014) reported an 86% response rate in a meta-analysis of 14 studies of nerve decompression. This is an outstanding result compared to the other approaches reviewed here. Dr. Jho also reports success with decompression. https://drjho.com/novel_occipital_neuralgia_surgery.htm. In spite of these reports by neurosurgeons, we do not know of a single patient who has ever had success with this surgery by either of these doctors.
- Rhizotomy surgery means cutting of nerves. Rhizotomy may be used to convert a neuralgia into a numbness. Of course, rhizotomy may also cause denervation pain. Andrychowski et al (2009) discuss surgical approaches. To us, this seems to be similar to RFGN, but less sophisticated.
- Ganglio-neurectomy is a more drastic surgical procedure than a rhizotomy. According to Acar et al (2008), C2 or C3 ganglionectomies are associated with short term pain relief (less than 3 months).
- Occipital nerve stimulator (see above)-- This device is something like an implanted TENS unit, just for the occipital nerve. It has many technical issues (Mcgreevey et al, 2012). Nevertheless they have been reported to provide good durable results (Slavin et al, 2006; Weiner et al, 1999). Ducic et al (2014) reported in a meta-analysis that the success rate was 68%, but the serious complication rate was 31.5%. Serious complications include lead migration.
- Cryo -- This is another way of damaging the nerve. While it may have some advantages over RFGN, including a partial damage rather than complete nerve section, the appropriate probes to treat ON are not available in the USA.
- Botox -- here idea is to block muscle around occipital nerve. We have had no experience with this, and it is also an expensive intervention, usually uncovered by insurance. We like the idea that it is temporary.
- Phenol -- Another denervation method, with danger of phantom pain. There is some danger of phenol induced necrosis of skin. This doesn't seem to us to be a very reasonable idea.
At the present writing (2015), none of these methods seem entirely satisfactory, but decompression surgery and RFGN (done very cautiously) would seem to us most suitable right now. Treatments appear to be in very slow evolution. We find it puzzling that section of the occipital nerve, through RFGN or just rhizotomy is not done more often. As patients in litigation often seem to have symptoms that can persist in spite of reasonable treatments, it is possible that there is some interaction between the often litigious context of ON and the current difficulties with treatments. In other words, it is more difficult to prove that a treatment works in litigating patient populations that are resistant to reporting improvement. There are also formidable barriers to pain physicians to being compensated for treating occipital neuralgia. This likely also contributes to the difficult treatment situation for ON.
Aspirin or acetaminophen, nonsteroidal analgesics such as torodol, and narcotics are frequently used for neuralgia. Usually non-narcotic pain killers are not strong enough to control neuralgia pain, but they are worth a try anyway. Narcotic medications are highly addictive and there is usually an attempt to use other medications first.
In general, topical treatment for nerve pain is a good idea. It avoids many side effects, and avoids addiction.
These are commonly used for trigeminal neuralgia. Tegretol (carbamazepine), Dilantin (phenytoin), and Neurontin (gabapentin) are the most commonly used drugs (Robotham et al, 1998). The author of this review often uses Trileptal (oxcarbamazine). They are given in doses similar to used for epilepsy, but more leeway is given to the patient in adjusting the dose up and down, depending on the amount of activity of the neuralgia. Sodium Valproate has also been used for this purpose.
Regarding oxcarbazepine (Trileptal), although it is not FDA approved for this indication, it behaves similarly to carbamazepine. Adjunctive agents may be used in this situation. These include baclofen and amitryptyline (see following).
These are mainly used for post-herpetic neuralgia rather than ON. Amitriptyline is the most commonly used medication. Nortryptyline, desipramine and others can also be used. These medications have substantial side effects (e.g. sedation, dry mouth, weight gain). Some authors claim that amitriptyline should be started within 3-6 months of onset of shingles to get optimal relief (Bowsher, 1994). SNRI type antidepressants such as Cymbalta are used for neuropathic pain. It has been suggested that for this use, beta-blockers should be avoided (Yalcin et al, 2009). Another SNRI antidepressant, venlafaxine, is a very good drug for migraine.
A middle age woman experienced an automobile accident, and thereafter developed severe headaches with pain behind her right ear, nose bleeds, and loss of smell and taste. There was tenderness and wincing on palpation of the area behind the right ear. A tentative diagnosis of Eagles syndrome was proposed, but X-rays did not bear this out. Diagnostic blocks of the occipital nerve abolished the pain. She subsequently had RF-ganglioneurectomy, with complete relief of headache for 6 months.
Although we presented a successful case here, practically speaking, lack of success is far more common.