Headaches are the fifth leading cause of emergency department visits,1 with the greater occipital nerve (GON) being the most commonly implicated nerve for headache pain.2 Occipital nerve blockade (ONB) is an effective treatment for headaches,2–5 but remains underutilized in the acute care and nonspecialist settings, with opioids and barbiturates, still prescribed more commonly than nonopioid targeted treatment in the emergency department.6 Increasing access to targeted pain procedures such as ONB is key to effectively treating pain while further reducing unnecessary opioid prescribing. The purpose of this study was to evaluate the efficacy and validity of an alternative, single skin insertion, paresthesia-based ONB technique, with the ultimate goal of increasing utilization of this targeted approach to treat occipital headache pain.
Occipital headache is typically described as a shooting, stabbing pain that may start at the base or back of the skull and extend up over the top of the head in the dermatome of the GON.7 The lesser occipital nerve (LON) may also be involved with symptoms extending laterally on the head to the ear and temple. The most common trigger of occipital neuralgia is compression of the GON and LONs.8
Traditional approaches to ONB vary in exact needle insertion point but typically use the landmarks of external occipital protuberance (EOP) and mastoid process to approximate the locations of both the occipital artery (OA) and the GON in the suboccipital region.2 Challenges with the traditional approaches to ONB include variability of the path of the GON in the suboccipital region,2,9 which has led to some authors to recommend “fanning” of the needle or higher volume injection (up to 5 mL for GON block)10 to increase likelihood of blocking the GON.2 In the suboccipital region, the LON is also more lateral, requiring a separate needle insertion,11 with an injection of a separate volume of 3 mL of local anesthetic10 for the successful blockade. In addition, there is no consistent recommended depth of needle insertion for ONB in the suboccipital region, and caution must be taken to prevent catastrophic injection into the cranial vault.2
An alternative location for ONB is at the superior nuchal line, where studies have shown a consistent relationship between the GON and the OA.12,13 The GON is more superficial and lateral at this location13 and lays directly on top of the bony protuberance of the superior nuchal line, thereby eliminating concerns of inserting the needle into deeper structures. Shimizu et al12 evaluated the anatomic relationship of the GON to the OA in 12 cadaveric specimens (24 occipital regions), noting that in all of the specimens the GON always crossed over the OA at the level of the superior nuchal line (Fig. 1). In a separate study evaluating the sonographic anatomy of the GON and OA complex at the superior nuchal line, Shim et al14 used ultrasound measurements to document the distance of the GON from the EOP, noting that the GON was located 23.1±3.4 mm from the EOP on the right and 20.5±2.8 mm from the EOP on the left.13 The LON is also more superficial at the level of the superior nuchal line and joins with the GON above the occiput15 thereby increasing the likelihood of both GON and LON blockade without requiring a separate needle stick.
In this prospective quality improvement project, we evaluated the efficacy and reliability of an alternative ONB technique that uses a single skin insertion at the superior nuchal line to block both the GON and LON. External landmarks of the cervical spinous process and ipsilateral tragus were used to predict the location of the OA pulse (and the cross-over point of the GON) at the superior nuchal line. These external landmarks were selected due to the ease of identification visually, and because of their relationship to both the EOP and the superior nuchal line. The cervical spinous process aligns vertically with the EOP.16 The ipsilateral tragus is a visually identifiable marker of the external auditory canal, which align with the EOP17 and the superior nuchal line.
In this alternative ONB technique, once the predicted location of the OA pulse at the superior nuchal line was identified using the above external landmarks, the recommended needle insertion point was identified 2 mm lateral to the OA pulse. Patient-reported paresthesia in the distribution of the GON and LON upon injection of local anesthetic at the target location was used to confirm contact with the target nerves.
This study was designed to evaluate 2 key measures. First, we assessed the efficacy of the actual procedure in terms of reduction in pain score (Numerical Rating Scale) and reported numbness in the GON and LON distributions. Second, we evaluated the validity of the external landmarks to predict OA pulse location and needle insertion/paresthesia.
MATERIALS AND METHODS
Institutional Review Board waiver was obtained for this quasi-experimental one group pretest, posttest quality improvement design. The study was conducted in a comprehensive outpatient pain management clinic. Patients were considered for the study if they had a complaint of headache and met the following inclusion criteria.
Inclusion and Exclusion Criteria
Four criteria were required in order for patients to be included in this study: (1) 18 years of age or older, (2) report of pain in the distribution of the GON with pain radiating up over occiput to the top of the head or to the eye, (3) optional report of head pain in the distribution of the LON with pain radiating to the ipsilateral ear or temple, (4) confirmed to comprehend the informed consent documents and interviews. The exclusion criterion included previous craniotomy surgery that would alter the posterior cranial anatomy.
After obtaining informed consent, ONB was performed by the attending physicians, nurse practitioners, or resident physicians under the direction of an attending physician (see Video, Supplemental Digital Content 1, http://links.lww.com/CJP/A627, which demonstrates the steps of the alternative greater and lesser ONB technique outlined below).
Patient Positioning and External Landmarks
Each patient was positioned seated with neck flexed and forehead resting on folded arms on a pillow on the examination table (Fig. 2, 1). External landmarks of the cervical spinous process and ipsilateral tragus were identified.
Predicted Location of OA Pulse
A vertical line 1.5 fingerbreadths lateral to the cervical spinous process was then traced and marked over the occiput. A horizontal line perpendicular to the midpoint of the ipsilateral tragus was traced and marked at the occiput, intersecting the first vertical line (Fig. 2, 2). The intersection of these 2 lines marked the predicted location of the OA pulse. The person performing the block then felt for a pulse at this predicted location and recorded whether or not the OA pulse was detected.
Skin Insertion and Greater ONB
A location ∼2 mm lateral to the predicted location of the OA pulse was identified as the planned needle insertion point (Fig. 2, 2). This location was then prepped with alcohol. A 25 G needle connected to the injectate syringe was then inserted at the planned insertion point. Care was taken to maintain the initial needle angle perpendicular to the skull in all planes. The needle was advanced until bony contact was made at the superior nuchal line (Fig. 2, 3). After checking for negative aspiration, 0.1 to 0.2 mL of 0.25% bupivacaine was injected, and the patient asked to report if they perceived paresthesia to the top of their head (indicating contact with the GON). If paresthesia to the top of the head was reported, a total volume of 2 mL of 0.25% bupivacaine was injected without moving the needle. If no paresthesia reported, the needle was withdrawn slightly, within the skin, and the needle tip was redirected 10 to 15 degrees cephalad, inserting again to the bone and checked again for paresthesia (Fig. 2, 3, inset). If no paresthesia reported after redirection, a total volume of 3 mL of 0.25% bupivacaine18 was injected at that location without further moving the needle.
Optional Lesser ONB
LON block was performed after blocking the GON (as above). The needle was withdrawn slightly within the skin and then the needle tip redirected 10 to 15 degrees laterally, along the horizontal line previously marked (Fig. 2, option) and reinserted to the bone. After checking for negative aspiration, a small amount of 0.25% bupivacaine was injected, and the patient was asked to report if they perceived paresthesia to ipsilateral ear (indicating contact with the LON). If paresthesia to the ear was reported, a total volume of 1 mL of 0.25% bupivacaine was injected at the target location without moving the needle. If no paresthesia to the ear was reported, the needle was redirected slightly more laterally, again contacting bone. If no paresthesia to ear reported after redirection of the needle, an additional 1 mL of 0.25% bupivacaine was injected at this location (2 mL total volume to LON if no paresthesia elicited). It is noted that this volume of local anesthetic is less than the volume required for the LON block in the suboccipital region.10 Upon completion of the block, the needle removed from the patient and pressure held until hemostasis appreciated.
Patient Safety and Monitoring
Patients maintained verbal communication with the provider at all times during the procedure. In addition, patients were observed during the postprocedure period by clinical support staff. Preprocedure and postprocedure vital signs were also obtained per clinic protocol.
Procedure efficacy was assessed using the following measures: (1) postprocedure pain reduction and (2) reported postprocedure numbness in the distribution of the GON and LON. We collected preprocedure and 5-minute postprocedure pain scores as reported by the patient on a 0 to 10 Numerical Rating Scale. The postprocedure pain reduction was then calculated as the difference between preprocedure and postprocedure pain scores. Patient report of numbness to the top of the head (GON distribution) and ipsilateral ear/temple (LON distribution) was also recorded at 5 minutes postprocedure.
The validity of the alternative ONB technique was evaluated by (1) OA pulse detection at the predicted location (based on landmarks), and (2) patient report of GON and LON paresthesia during the injection. Patient report of paresthesia to the top of the head signified contact with GON, and paresthesia to the ipsilateral ear signified contact with LON. We also recorded the number of needle redirection attempts required to obtain paresthesia.
A sample size calculation was conducted for purposes of achieving adequate statistical power in the study. The researchers hypothesized a small to medium effect size (dz=0.40) associated with pain intervention.18 Using a 2-tailed hypothesis, an α value of 0.05, a β value of 0.20, and dz=0.40, a total of n=50 participants would be needed to achieve adequate statistical power to detect significant pain reduction in this population.
Data Analysis and Statistical Testing
The demographic characteristics of the study sample were analyzed using descriptive and frequency statistics. The primary analysis was associated with the change in pain scores from preintervention to postintervention. Each pain score observation was tested for the statistical assumption of normality using the Kolmogorov-Smirnov test. When the statistical assumption of normality was met for both observations of the pain scores, repeated-measures t test was used to assess the degree of change in pain from preintervention to postintervention for the study participants. Means and SDs for the preintervention and postintervention pain scores were reported and interpreted to give context to the repeated-measures t test findings.
Frequency, percentage, and 95% confidence interval (CI) statistics were conducted on data collected for (1) pulse palpation for the left and right side of the head, (2) confirmation of GON paresthesia on both sides of the head, (3) confirmation of LON paresthesia on both sides of the head, (4) numbness to the top of the head for both sides, and (5) numbness to the ipsilateral ear for LON only. χ2 statistics were used to compare confirmation of OA pulse palpation (yes/no) to achieving GON paresthesia (yes/no) for both the left and right sides of the head. The same analysis was performed to compare GON paresthesia (yes/no) to achieving numbness to the top of the head (yes/no) for both the left and right sides of the head, and to compare LON paresthesia to successful numbness to the ipsilateral ear. Unadjusted odds ratios with 95% CIs were reported and interpreted for the χ2 analyses. All analyses were conducted using SPSS, version 25 (IBM Corp., Armonk, NY) and statistical significance was assumed at an α value of 0.05.
A total of 50 patients were included in this study. The patients were evaluated and diagnosed with occipital neuralgia in the course of usual clinical care. They had headaches of varying duration but all had the classic occipital neuralgia findings of posterior head pain, plus or minus radiation to the top of the head, eye, or ear.
Table 1 shows the baseline demographic and relevant clinical history and examination findings of the included patients. Ages ranged from 35 to 78 years with a mean of 55.42 years (SD 12.06). Thirty-three of the patients (73.3%) were women. Thirty-six patients (72.0%) reported bilateral posterior headache pain (GON involvement), with 20 patients (40%) also reporting headache to bilateral temples and/or ears (LON involvement). All but 6 (12%) of participants had physical examination findings of tenderness over the occiput at the time of the evaluation.
The laterality and number of nerves blocked per patient depended on their clinical presentation at the time of evaluation. Patients with headache radiating to the top of the head or eye were offered a GON block for the affected side(s). Patients with headache radiating to the temple or ear were offered a LON block for the affected side(s). LON block was not performed without also performing the GON block on any patient. Figure 3 shows the breakdown of procedure type by the patient. Table 2 shows the frequencies and percentages of GON and LON blocks performed in this patient sample. In this study, n=21 of the procedures (42.0%) were performed by the attending physician, n=20 (40.0%) were performed by nurse practitioners, and n=9 were completed by resident physicians (19.0%).
The statistical assumption of normality was met for the preprocedure and postprocedure observations of pain (n=47). There was a statistically significant decrease in pain scores from preprocedure (M=6.04, SD=2.58) to postprocedure (M=2.74, SD=2.19) in the patient sample (t49=10.49, P<0.001, d=1.36, post hoc power=1.00) (Fig. 4). The difference in means across time showed an overall 54.64% decrease in pain scores for all participants.
Clinically significant reduction in pain (defined as 30% or higher reduction in pain postprocedure) was observed in 80.9% of participants.19 Almost two-thirds of participants (61.7%) reported at or above at 50% reduction in pain from preprocedure to postprocedure. Further analysis of pain scores showed that 29.8% of participants experienced a 70% or greater reduction in pain postprocedure.
Incidence and 95% CI for procedural outcomes of postprocedure numbness, OA pulse palpation, and GON and LON paresthesia are reported in Table 3. Numbness to the top of the head after GON block using this alternative approach was reported 80.6% of the time on the left side of the head, and 90% of the time on the right side of the head. Alternative technique landmarks predicted OA pulse location 87.5% on the left side and 72.7% on the right side. GON paresthesia with this procedure technique was confirmed 90.0% of the time on the left side of the head and 90.9% of the time on the right side. LON paresthesia was confirmed 74.2% of the time on the left side of the head and 77.4% of the time on the right side of the head.
The relationship between OA pulse palpation at a predicted location based on landmarks and corresponding GON paresthesia report with GON block was also analyzed. Of those patients for whom OA pulse was palpated at the predicted location on the left side of the head, 88.6% (31/35) also reported corresponding GON paresthesia on the left. In addition, 5/5 (100%) patients for whom OA pulse was not palpated at a predicted location on the left, still reported GON paresthesia with left GON block. Of those patients for whom OA pulse was palpated on the right side of the head, 93.8% (30/32) reported GON paresthesia on the right, with 10/12 (83.3%) of patients for whom OA pulse was not palpated, still reporting right GON paresthesia.
We also analyzed the data on the total volume of local anesthetic injected per patient, and the relationship to paresthesia report at the GON and LON injection sites. Per the procedure protocol, those patients who did not report paresthesia at the GON or LON, received an additional 1 mL of local anesthetic at the relevant injection site. For GON blocks on the left side, n=2 (5.0%) participants did not report paresthesia on with this technique and received 3 mL (instead of 2 mL) at the injection site. For the right side, n=4 (9.1%) required 3 mL (instead of 2 mL) due to not reporting GON paresthesia on the right. For LON blocks, n=3 (9.7%) participants reported no paresthesia on the left side, thereby receiving 2 mL (instead of 1 mL) of local anesthetic. With LON blocks on the right side, n=5 (16.7%) received 2 mL (instead of 1 mL) due to not reporting paresthesia. The mean volume of injectate across all patients was 5.16 mL (SD=1.57) with a range of 6 (2 mL [n=1] to 8 mL [n=5]). The majority of patients received either 4 mL (n=12) or 6 mL (n=16).
The correlation between confirmation of paresthesia and numbness to the top of the head was assessed for both sides of the head. For the left side of the head, there was no association between confirmed GON paresthesia and numbness to the top of the head (P=0.16, n=27, 84.4%), but there was a significant association between confirmed LON paresthesia and numbness to the ipsilateral ear (P=0.007, n=18, 81.8%). On the right side of the head, there was not an association between confirmed GON paresthesia and numbness to the top of the head (P=0.34, n=34, 91.9%), and there was not an association between confirmed LON paresthesia and numbness to the ipsilateral ear (P=0.24, n=19, 79.2%).
The results of this study demonstrate that a single skin insertion, paresthesia-based ONB technique produces a statistically significant reduction in occipital headache pain within 5 minutes of the procedure. The results are also clinically significant,19 with over 80% of patients reporting at least a 30% reduction in pain, and over 60% of patients reporting a >50% reduction in headache pain postprocedure. This compares favorably to other prospective studies documenting the efficacy of ONB for acute headache.2 Additional measures of efficacy include the high percentage of patients that reported numbness in the GON and LON distributions after block, although these numbers did not reach statistical significance.
The study results also show that readily identifiable external landmarks (cervical spinous process and ipsilateral tragus) can be used to predict the location of the OA pulse at the superior nuchal line, and by proximity, the location of the GON as confirmed by paresthesia. In addition, even if the OA pulse was not palpated at the predicted location—which could also be a factor of operator ability to detect the OA pulse—bilateral GON paresthesia was still reported >90% of the time using this procedure technique. This corresponds with anatomic studies showing the relationship of the GON and OA at the superior nuchal line.12,13 Interestingly, contrary to traditional approaches to LON blockade, this study also shows that LON can be effectively blocked (as confirmed by paresthesia) using a single skin insertion at the superior nuchal line.
Limitations of this prospective quality improvement study include a lack of direct comparison to a control group, such as patients undergoing ONB using a traditional suboccipital approach. In addition, the timing and frequency of follow-up visits after the procedure were dictated by the course of usual clinical care, thereby resulting in varied timing of postprocedure follow-up. This is as opposed to a prescribed frequency and duration of follow-up encounters as would be the case with a prospective observational study. Because of this short-term observation of participants, the duration of effect was not consistently measured for the purpose of this study.
Despite these limitations, the present results justify the potential of future studies that focus on measuring the reproducibility of this technique in different clinical settings. It would also be beneficial to measure the incidence of adverse effects using this alternative ONB technique, and directly compare this to the more traditional approach to ONB. Future studies may also focus on the duration of relief from this alternative ONB technique as compared with the traditional ONB approach.
As we continue to combat the dual crises of chronic pain and opioid overprescribing, it is imperative that we increase access to and use of targeted procedures, such as ONB, to effectively treat the anatomic source of pain. In this alternative ONB technique, the easily identifiable external landmarks described above, combined with the single skin insertion and bony endpoint of this approach increase the confidence with which providers may safely perform ONB. These initial data support the interpretation that single skin insertion, paresthesia-based technique to block both the GON and LPNs is an effective, targeted treatment option for headache arising from occipital neuralgia.
1. Minen MT, Tanev K, Friedman BW. Evaluation and treatment of migraine in the emergency department: a review. Headache. 2014;54:1131–1145.
2. Voigt CL, Murphy MO. Occipital nerve blocks in the treatment of headaches: safety and efficacy. J Emerg Med. 2015;48:115–129.
3. Okmen K, Dagistan Y, Dagistan E, et al. Efficacy of the greater occipital nerve
block in recurrent migraine type headaches. Neurol Neurochir Pol. 2016;50:151–154.
4. Zhang H, Yang X, Lin Y, et al. The efficacy of greater occipital nerve
block for the treatment of migraine: a systematic review and meta-analysis. Clin Neurol Neurosurg. 2018;165:129–133.
5. Allen SM, Mookadam F, Cha SS, et al. Greater occipital nerve
block for acute treatment of migraine headache: a large retrospective cohort study. J Am Board Fam Med. 2018;31:211–218.
6. Ahmed ZA, Nacopoulos DA, John S, et al. An algorithm for opioid and barbiturate reduction in the acute management of headache in the emergency department. Headache. 2017;57:71–79.
7. Headache Classification Committee of the International Headache Society (IHS). The International Classification of Headache Disorders, 3rd edition. Cephalalgia. 2018;38:1–211.
8. Choi I, Jeon SR. Neuralgias of the head: occipital neuralgia. J Korean Med Sci. 2016;31:479–488.
9. El Sekily NM, Zedan IH. Surgical anatomy of greater occipital nerve
and its relation to occipital artery. Alexandria J Med. 2015;51:199–206.
10. Lauretti GR, Correa SW, Mattos AL. Efficacy of the greater occipital nerve
block for cervicogenic headache: comparing classical and subcompartmental techniques. Pain Pract. 2015;15:654–661.
11. Cesmebasi A, Muhleman MA, Hulsberg P, et al. Occipital neuralgia: anatomic considerations. Clinical Anat. 2015;28:101–108.
12. Shimizu S, Oka H, Osawa S, et al. Can proximity of the occipital artery to the greater occipital nerve
act as a cause of idiopathic greater occipital neuralgia? An anatomical and histological evaluation of the artery-nerve relationship. Plast Reconstr Surg. 2007;119:2029–2034; discussion 2035–2036.
13. Ducic I, Moriarty M, Al-Attar A. Anatomical variations of the occipital nerves: implications for the treatment of chronic headaches. Plast Reconstr Surg. 2009;123:859–863; discussion 864.
14. Shim JH, Ko SY, Bang MR, et al. Ultrasound-guided greater occipital nerve
block for patients with occipital headache and short term follow up. Korean J Anesthesiol. 2011;61:50–54.
15. Kemp WJ III, Tubbs RS, Cohen-Gadol AA. The innervation of the scalp: a comprehensive review including anatomy, pathology, and neurosurgical correlates. Surg Neurol Int. 2011;2:178.
16. Tubbs RS, Iwanga J, Loukas M, et al. Clinical Anatomy of the Ligaments of the Craniocervical Junction. Newcastle upon Tyne, UK: Cambridge Scholars Publishing; 2019.
17. Mosser SW, Guyuron B, Janis JE, et al. The anatomy of the greater occipital nerve
: implications for the etiology of migraine headaches. Plast Reconstr Surg. 2004;113:693–697; discussion 698–700.
18. Naja ZM, El-Rajab M, Al-Tannir MA, et al. Occipital nerve blockade for cervicogenic headache: a double-blind randomized controlled clinical trial. Pain Pract. 2006;6:89–95.
19. Rowbotham MC. What is a “clinically meaningful” reduction in pain? Pain. 2001;94:131–132.