Prevalence of Occipital Neuralgia at a Community Hospital-based Headache Clinic

Abstract

Background

Occipital neuralgia (ON) is a paroxysmal disorder involving lancinating pain that originates in the neck or skull base with superior radiation toward the apex. ON more commonly occurs in patients with other coexisting headache disorders. There are limited data regarding the prevalence of ON. This study aims to demonstrate the prevalence of ON in a community hospital-based headache clinic.

Methods

This IRB-approved retrospective study was conducted at the Cambridge Health Alliance Headache Clinic. Medical records of patients presenting with headache as a chief complaint were reviewed from January 2010 to September 2015.

Results

Of 800 study patients, 81% were females (n = 648). A total of 195 patients were diagnosed with ON, and 146 patients had a positive occipital Tinel sign on examination. Isolated ON was present in 15.38% (n = 30) of patients. Multiple regression analysis demonstrated that the odds of ON were higher in patients with chronic migraine vs episodic migraine (adjusted odds ratio = 2.190 [95% confidence interval: 1.364–3.515]), even when adjusted for significant covariates.

Conclusion

ON occurred in nearly 25% of patients presenting with a chief complaint of headache to a community hospital-based headache clinic. Among patients with ON, 15% presented with ON as the chief complaint without another coexisting headache disorder. As such, up to 85% of ON cases occurred in patients having an additional headache type. Approximately 75% of patients with ON had a positive occipital Tinel sign on examination. Elevated body mass index, higher age at presentation, and chronic migraine increased the odds of having ON. Undiagnosed or inadequate treatment of ON can increase the frequency and intensity of other comorbid headache disorders.

Occipital neuralgia (ON) is a paroxysmal pain disorder that involves lancinating pain that originates in the neck or skull base with superior radiation toward the apex. In cases involving the lesser occipital nerve, the pain can radiate to areas of the scalp around the ear approaching the temple. In some cases, referred pain can also involve fronto-orbital regions, which may be due to trigeminocervical interneuronal connections in the trigeminal spinal nuclei. ON can manifest with unilateral or bilateral attacks. The International Classification of Headache Disorders 3rd Edition (ICHD) diagnostic criteria for ON are listed in table 1.¹

Table 1.

International Classification of Headache Disorders 3rd Edition, Beta Version Occipital Neuralgia Diagnostic Criteria

In clinical practice, isolated ON as the sole chief complaint is uncommon, but ON can more commonly be appreciated in patients who have another primary headache disorder like migraine. In a study conducted at the University of Southern California Headache Clinic, among 35 consecutive patients presenting with ON, 20 patients also had a diagnosis of migraine. There was no significant difference in age, sex, or ethnicity between patients with ON with and without migraine. In patients with both of these diagnoses, they experienced significantly more pain traveling to the scalp, scalp tenderness, and paresthesias than patients with isolated ON.²

Making the diagnosis of ON is critical in patients with migraine, as the treatment of these 2 disorders is quite different. A diagnosis of migraine usually results initially in the prescribing of abortive and preventive oral pharmacologic therapies. On the other hand, a suspected diagnosis of ON should prompt the physician to proceed with occipital nerve blocks, which can be diagnostic and therapeutic according to the ICHD criteria. Adequate treatment of ON can lead to reductions in overall pain, stress, and sleep dysfunction, which can subsequently result in an improvement of the frequency and intensity of coexisting migraine. Commercial insurance companies often reject coverage for occipital nerve blocks as “experimental” in migraine or cluster headache, but if there is coexisting ON, coverage may be more likely to be obtained. There are limited data in the literature regarding the prevalence of ON. The Bruneck study, which looked at primary headache and cranial neuralgia in patients aged 55–94 years, found that the lifetime prevalence of all primary headaches was 51%, and that of cranial neuralgia was 1.6%.³ This study reinforces that coexisting ON is a distinct entity that can be diagnosed commonly in people with migraine.

The objective of this study was to demonstrate the prevalence of ON among patients presenting to a community hospital-based headache clinic.

Methods

Standard Protocol Approvals

The study received IRB approval by the Cambridge Health Alliance Institutional Review Board. This retrospective case series was conducted at the Cambridge Health Alliance Headache Clinic. The Cambridge Health Alliance is composed of 3 community hospitals and a network of outpatient clinics. Medical records of all consecutive patients presenting to the headache clinic were reviewed from January 2010 to September 2015. Data including age, sex, age at headache onset, body mass index (BMI), referral source, headache diagnoses, and presence of occipital Tinel sign were recorded. Greater and lesser occipital nerve blocks were performed to confirm the diagnosis of ON according to strict ICHD-3 diagnostic criteria.¹ Nerve blocks were performed on the side of probable ON, which were bilateral if bilateral ON was suspected. All physical examinations and nerve blocks were performed by a single fellowship-trained, board-certified headache specialist, using 5.5 mL of 0.75% bupivacaine and 0.5 mL of triamcinolone acetonide (20 mg) per side (for both greater and lesser occipital nerves). For bilateral greater and lesser occipital blocks, the total volume of injected solution was 12 mL. Data collection and abstraction was completed in Microsoft Excel (version 2011). All statistical analyses including descriptive statistics, simple logistic regression and models using multiple logistic regression, were conducted in SAS software (version 9.3; SAS Institute Inc., Cary, NC).

Data Availability

Deidentified data not provided in the article because of space limitations are available upon request to the corresponding author for other investigators for purposes of replicating procedures and results.

Results

Data were collected on a total of 800 patients (n = 800), 81% of which were females (n = 648). Baseline study characteristics are reported in table 2. The average age at the time of presentation to the clinic was 39.08 years (range 16–88 years). The average reported age at the time on headache onset was 23.17 years (range 2–86 years). The average BMI score at the time of presentation to the clinic was 28.33 (range 14.96–64.34).

Table 2.

Baseline Characteristics and Diagnoses

The majority of patients seen for evaluation in the headache clinic were referred by their primary care providers (n = 705). Other sources of referrals included general neurologists (n = 89) and pain specialists (n = 6). The diagnoses at the time of initial evaluation are detailed in table 2. Primary headache disorders other than diagnosis number 2–9 (table 2) were grouped together to form “other primary headache disorders,” diagnosis number 10 (table 2). This group included disorders such as primary stabbing headache and nummular headache. All primary headache disorders, which did not fulfill the ICHD-3 diagnostic criteria, for example, probable migraine, were grouped as “other headache disorders,” diagnosis number 13 (table 2). Posttraumatic headache (n = 58), diagnosis number 11, was grouped with “other secondary headache disorders” (n = 8), diagnosis number 12 (e.g., postsurgical headache and postepidural headache), to form “secondary headache disorder” (n = 66) for the purpose of statistical analysis (table 2).

A total of 195 patients were diagnosed with ON. Among the patients diagnosed with ON, 146 had a positive occipital Tinel sign on examination, 46 had a negative occipital Tinel sign on examination, and 3 presented for procedure only visits, so no physical examination was performed. As such, approximately 75% of patients with ON had a positive occipital Tinel sign on examination. Among the 605 patients not diagnosed with ON, none of them had a positive occipital Tinel sign.

Table 3 describes the prevalence of ON in our cohort and association with other migraine and secondary headache disorders. Isolated ON was present in 15.38% (n = 30) of the study cohort.

Table 3.

Prevalence of ON and Association With Other Headache Disorders

Regression analysis was performed to predict the odds of ON with other baseline patient characteristics (e.g., BMI) and with the presence of other headache disorders (table 4). Bivariate analysis using logistic regression demonstrated that the unadjusted odds of ON were 2.295 times higher (odds ratio [OR] = 2295; 95% confidence interval [CI]: 1.438–3.664) among those who had chronic migraine vs episodic migraine. It also demonstrated that the unadjusted odds of ON were 1.820 times higher (OR = 1.820 [95% CI: 1.054–3.141]) among those who had a secondary headache disorder vs primary headache disorder. The unadjusted odds of predicting ON with the continuous BMI score were found to be 1.028 (95% CI: 1.003–1.054). There is a significant but small relationship between a higher score and increased odds of ON. Furthermore, bivariate analysis also showed that the unadjusted odds of predicting ON with age at presentation (continuous variable) is 1.016 (95% CI: 1.004–1.027). For each unit increase in age at presentation, we expect to see about 1.6% increase in the odds of having ON.

Table 4.

Results of the Simple Logistic Regression Analysis for the Association Between ON and Covariates Including “Chronic vs Episodic Migraine” (w/+ w/o aura), Secondary vs Primary Headache Disorders (See Table 1 Legend for Groups), BMI, and Age at Presentation

When controlled for significant covariates using multiple logistic regression (table 5), the association of ON with chronic migraine (with + without aura) showed some attenuation but remained significant with the final model showing that the odds of ON were 2.190 times higher (adjusted OR = 2.190 [95% CI: 1.364–3.515]) among those who had chronic migraine vs episodic migraine. Of interest, in this model, the odds of predicting ON with age at presentation also remained significant, with OR 1.030 (95% CI: 1.014–1.046).

Table 5.

Results of the Multiple Logistic Regression Analysis for the Association Between ON and Chronic vs Episodic Migraine (N = 563) and Between ON and Secondary vs Primary Headache Disorders (N = 695)

Similarly, when adjusted for significant covariates (multiple logistic model), the association of ON with secondary headache disorders did not remain significant with the final model showing that the odds of ON were 1.698 times higher (adjusted OR = 1.698 [95% CI: 0.971–2.970]) among those who had a secondary headache disorder vs primary headache disorder. However, in this model, the odds of predicting ON with age at presentation (OR 1.017; 95% CI: 1.004–1.031) and with the BMI score (OR 1.029; 95% CI: 1.000–1.059) remained significant.

Discussion

ON is highly prevalent, occurring in nearly 25% of patients presenting with a chief complaint of headache to a community hospital-based headache clinic. Among patients with ON, only 15% presented with ON as the chief complaint without another coexisting headache disorder. As such, up to 85% of ON cases occurred in patients having a chief complaint of another headache type. The under diagnosis of ON is likely multifactorial.

A growing problem with our health care system is a push to see more patients in less time. As such, providers may not have adequate time to take a more detailed history of head pain that may be occurring in addition to the primary headache disorder for which the patient is presenting. In addition, both the patient and the provider may believe that all pain in the head is due to a single diagnosis and that elaborating on other details may not lead to additional diagnoses or treatment options.

For example, a patient history that involves many years of throbbing unilateral headache with photophobia, phonophobia, nausea, and vomiting would logically suggest that the patient probably has migraine. As such, additional questions would likely be posed that would further substantiate this diagnosis. One question that can be very revealing is asking whether the patient has any head/neck pain in the absence of a typical headache. The answer can often suggest comorbid diagnoses such as ON, primary stabbing headache, and temporomandibular joint disorders. Typical historical features of ON include contact/pressure to the skull base such as when lying down in the supine position or moving the neck in certain positions can trigger lancinating pain in the occiput.

In addition to an expanded history, a few physical examination maneuvers can be very useful in making a diagnosis of ON. An occipital Tinel evaluation involves percussion along the occiput from mastoid to mastoid. A positive Tinel sign should elicit pain or paresthesias that travel from the skull base toward the apex in the case of the greater occipital nerve or toward the ipsilateral ear in the case of the lesser occipital nerve. Performing a passive range of motion neck assessment including flexion, extension, rotation, and lateral flexion can at times also yield similar pain or paresthesias in a single nerve distribution. A positive Tinel sign should not be confused with percussion causing a worsening of nonspecific, global head pain, which can occur with percussion or movement of the head in someone experiencing a migraine. This response can be even more exaggerated in the setting of cutaneous allodynia. Patients with ON and another primary headache disorder are often diagnosed as having cervicogenic headache in the absence of having any identifiable lesions within the cervical spine or soft tissues of the neck as is required by the ICHD criteria. Unfortunately, a diagnosis of cervicogenic headache rather than ON may not lead a provider to perform a diagnostic/therapeutic occipital nerve block.

Another reason that ON may be underdiagnosed is the misconception that multiple headache types do not typically occur within the same patient. This would be like a patient having upper extremity pain and a diagnosis of cervical radiculopathy. The same patient could also have carpal tunnel syndrome, which can also contribute to upper extremity pain.

The higher odds of having ON in patients with chronic migraine compared with episodic migraine may be due to the presence of ON serving as a factor in the chronification of migraine. We hypothesize that this is secondary to ongoing/prolonged modulation of trigeminal nucleus caudalis. The same underlying mechanism (modulation of trigeminal nucleus caudalis) is hypothesized to be the basis of benefit seen in acute treatment of migraine after performing occipital nerve blocks.^4–6 Studies exploring the utility of occipital nerve blocks in treating chronic migraine have produced mixed results.^7–11 Although there were significant methodological limitations in these trials,¹² it is possible that some responders had underlying/undiagnosed ON, treating which improved the disease severity of their chronic migraine both in terms of headache days and headache intensity. However, some^9–11 argue that ON in migraine is potentially directly related to peripheral or central sensitization, and it may not be a comorbidity but rather part of chronic migraine. It is important to note that the patients diagnosed with ON in our study fulfilled the ICHD-3 criteria for ON¹ and did not merely have occipital tenderness and response to nerve block, which can be seen in migraine. It is also important to note that during examination, 146/195 (∼75%) of the patients with ON had a positive Tinel sign, and none of the 605 patients without ON had a positive Tinel sign, which would distinguish ON from migraine-associated allodynia. Based on patient histories, the vast majority of these patients had lancinating attacks of pain characteristic of ON in the absence of photophobia, phonophobia, nausea, and throbbing pain characteristic of migraine. In addition, patients can experience a remission of their ON lancinating pain for weeks, months, or even years after receiving occipital nerve blocks,¹³ but will usually continue to experience acute attacks of migraine while ON is in remission. A possible mechanism is that large-volume nerve blocks (6 cc/side as performed in this study) are causing an expansion of tight tissue planes, which leads to decompression of the occipital nerves.¹⁴ This unique response to nerve blocks further suggests that ON is a separate clinical entity that can occur in patients with migraine. In addition, patients who have surgical decompression of the occipital nerves can often experience a resolution of their ON, but continue to experience migraine.¹⁵

The higher chances of having ON in patients with elevated BMI and age may have to do with the anatomical changes that can occur with obesity, age progression, chronic poor posture, and occupational exposures. Arthritis, skeletal deformity, soft tissue nerve entrapment, and other alterations of the normal anatomy may play a role in the development of ON.

In general, retrospective studies can have some degree of bias. Regarding this study, one author performed all physical examinations and injections. Although this is a potential source of bias, the performing clinician is a fellowship-trained, board-certified headache specialist, who strictly followed the ICHD-3 criteria for making diagnoses. One weakness of this study is that it was conducted at a community hospital headache clinic. Although this population analysis allows for better generalization of data than if it were conducted at a tertiary headache center, it still may not reflect the prevalence of ON in the general population. Data on the sequence of development of migraine, chronic migraine, and ON were not available for analysis, which is also a study limitation.

With a better understanding of the prevalence, clinical features, and physical examination findings of ON, providers may more consistently make the accurate diagnosis of ON, which should trigger the performance of occipital nerve blocks or referral to a provider who performs this procedure. As such, more emphasis should be made to educate residents and fellows on the diagnosis and treatment of ON, which should include how to perform diagnostic/therapeutic occipital nerve blocks. This need for additional procedural training was reinforced in a survey study involving neurology residency program directors, and it was found that only 27.3% of programs permitted trainees to perform procedures independently.¹⁶ In addition, as the body of literature supporting the long-term efficacy of large-volume occipital nerve blocks for the treatment of ON grows, reimbursement for occipital nerve blocks for this indication should improve over time. If undiagnosed, inadequate treatment of ON can trigger an increase in the frequency and intensity of other comorbid headache disorders such as migraine and sleep dysfunction.

Appendix. Authors

Study Funding

No targeted funding reported.

Disclosure

The authors report no disclosures relevant to the manuscript. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.