Greater occipital nerve block in migraine prophylaxis: Narrative review

Abstract

Objectives

The importance and popularity of peripheral nerve block procedures have increased in the treatment of migraine. Greater occipital nerve (GON) block is a commonly used peripheral nerve block method, and there are numerous researches on its use in migraine treatment.

Materials and methods

A search of PubMed for English-language randomized controlled trials (RCT) and open studies on greater occipital nerve block between 1995 and 2018 was performed using greater occipital nerve, headache, and migraine as keywords.

Results

In total, 242 potentially relevant PubMed studies were found. A sum of 228 of them which were non-English articles and reviews, case reports, letters and meta-analyses were excluded. The remaining articles were reviewed, and 14 clinical trials, seven of which were randomized-controlled on greater occipital nerve block in migraine patients, were identified and reviewed.

Conclusions

Although clinicians commonly use greater occipital nerve block in migraine patients, the procedure has yet to be standardized. The present study reviewed the techniques, drugs and dosages, the frequency of administration, side effects, and efficacy of greater occipital nerve block in migraine patients.

Introduction

Migraine is a common neurological disorder and its treatment can be challenging; discontinuation of preventive medications is common (1). Consequently, peripheral nerve block procedures have been administered for both acute and preventive treatment of migraine (2).

Peripheral nerve block techniques include blockage of the greater and lesser occipital nerves, as well as some branches of the trigeminal nerve, such as the supraorbital, supratrochlear, and auriculotemporal nerves (3). Greater occipital nerve (GON) block is a commonly used peripheral nerve block method in migraine treatment (2).

The effect of GON block is observable in the trigeminovascular system, which plays a vital role in the pathophysiology of migraine. A rat study reported that the excitability of meningeal afferent input increased via the stimulation of the GON and cutaneous C-fiber afferents in response to mustard oil. The study reported that there was a functional process between the caudal trigeminal nucleus and upper cervical segments (4). Piovesan et al. described the anatomic and physiological relationship between the upper cervical spinal segments and the spinal nucleus of the trigeminal nerve in humans by injecting sterile water to GON unilaterally. After injection, patients feel pain in both the GON and trigeminal zone (5). The administration of a local anesthetic injection to the GON (a branch of the upper cervical nerve) blocks afferent stimuli that are coming from the regions that are innervated by the GON. As a result, this situation prevents the sensitization of the C2 dorsal horn convergent neurons by decreasing their input (6).

Although clinicians commonly use GON block in migraine patients, the procedure is yet to be standardized. The members of the American Headache Society Peripheral Nerve Blocks and other Interventional Procedures for Headache Special Interest Section published recommendations for administration of peripheral nerve block, including GON block, in patients with headache disorders (3). There have been several reviews published on occipital nerve block; however, some of them did not focus specifically on migraine (2,7–11). The others included other methods than GON blocks and there have been new studies undertaken after these were published (12). There are two meta-analyses about GON block in migraine, where the researchers concluded that GON block is a valid treatment. The analyzed studies differed in techniques, the drugs used and dosages (13,14). As new studies are being published on GON block in migraine patients, the literature lacks a proper and relative narrative review. The present study aims to review the techniques, drugs and dosages, the frequency of administration, side effects and efficacy of GON block in prophylactic migraine treatment.

Materials and methods

This narrative review of GON block for the treatment of migraine included studies conducted with migraine patients aged ≥18 years that had been treated with GON block. A search of PubMed for English language randomized controlled trials (RCTs), observational studies on GON block published between 1995 and 2018 were conducted using greater occipital nerve, headache and migraine as keywords. The last date of search used was 1 February 2018. Articles were chosen through an independent, unbiased selection of articles using appropriate keywords.

All summaries of articles in English that met the search criteria were reviewed. Figure 1 shows the progression of article selection and the numbers of articles at each step. The review focused on prophylaxis of migraine treatment; studies on acute treatment of migraine using GON block were excluded.

Figure 1.

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram showing the progression of article selection and the numbers of articles at each step.

Results

In this review, 14 trials on GON block for migraine treatment are summarized; however, one of these trials consists of patients with migraine but also cluster headache, new daily persistent headache, and hemicrania continua (15–28). Seven of these trials were RCTs (Table 1). For blockade, bupivacaine was used in 11 trials (15,17–22,24–26,28) (five of these were RCTs). Lidocaine was used in one trial (16). In addition, a mixture of bupivacaine and mepivacaine (27) and a mixture of bupivacaine and lidocaine (15) was used in just one trial. (Table 1 and Table 2).

Table 1.

Characteristics of randomized controlled studies.

First author, year	Number of patients; study design	Side effects	Headache types	Drugs	Number of blocks; block location	Results
Ashkenazi et al., 2008 (15)	Included: 37; randomized single blinded Group A: GON block with local anesthetics and saline, n = 18 Group B: GON block with local anesthetics and 40 mg/ml triamcinolone, n = 19	None reported	Transformed migraines	4.5 ml 2% lidocaine, 4.5 ml 0.5% bupivacaine, 1 ml saline or 40 mg/ml triamcinalone 2 ml to each GON 0.5 ml to 12 trigger points	GON block was performed at medial third to the distance between Po and mastoid process. Bilateral GON block and trigger point injections (TPIs) to 12 areas GON and TPIs blocks applied once.	Headache calendars of 11 patients from group A and 13 patients from group B were evaluated. The mean duration of headache-free days were 2.7 ± 3.8 days and 1.0 ± 1.1 days in group A and B respectively. (p = 0.67) Mean analgesic consumption decreased by 19.3 and 10.9 in group A and B, respectively. Headache severity decreased for 14.3 ± 15.1 days in group A and 5.5 ± 4.9 days in group B (p = 0.60).
Kashipazha et al., 2014 (16)	Included: 52 Completed: 48; randomized double-blind controlled Control group: GON block with local anesthetics and saline, n = 24 Intervention group: GON block with local anesthetics and triamcinolone, n = 24	No serious side effect	Migraine	1.0 ml 2% lidocaine, 0.5 ml of either saline or triamcinalone	GON block was performed bilaterally once. GON block performed at medial third of the distance between Po and mastoid process.	In both groups, pain severity, pain frequency, and analgesic use were lower significantly than pretreatment (p < 0.001) at 2, 4, and 8 weeks after intervention. All of the variables started onwards from the fourth week.
Dilli et al., 2015 (17)	Included: 70 Completed: 63; randomized double-blind placebo controlled Active group: GON block with local anesthetics and triamcinolone, n = 33 Placebo group: 0.25 ml 1% lidocaine (to create numbness over the GON dermatomal region) and saline, n = 30	21 % non-severe adverse effects reported in active group, 25% reported in placebo group. Benign intracranial hypertension in placebo group.	Episodic or chronic migraine	Active group: 2.5 ml 0.5% bupivacaine and 0.5 ml 20 mg methylprednisolone placebo group: 2.75 ml saline and 0.25 ml 1% lidocaine without epinephrine	GON block performed at the medial third of the distance between Po and mastoid process. One block was performed, bilateral or unilateral injection depending on patient's headache.	The percentage of patients with at least 50% reduction in the frequency of moderate or severe headache was 30% for both groups. Migraine frequency and duration did not differ between groups in the next 28 days after intervention.
Inan et al., 2015 (18)	Included: 84 Completed: 72; randomized double-blind placebo controlled Group A: GON block with saline n = 33 Group B: GON block with local anesthetics n = 39	No severe adverse effects Bupivacaine treatment: local pain at the site of injection (6), vertigo (3), nausea (1) Placebo treatment: Pain at the site of injection (2), vertigo (1) back pain (1), cervical neck spasm (2)	Chronic migraine	Treatment group: 1.5 ml 0.5% bupivacaine and 1 ml saline; placebo group: 2.5 ml saline	GON block performed 2 cm lateral, 2 cm inferior to external Po once per week four times; after blinding was removed, block was administered once a month 2 times.	Headache duration decreased in group B Number of headache days and VAS scores were statistically lower in group B compared to group A. After changing placebo group with bupivacaine, the groups showed similar significant decrease due to headache parameters in the third month.
Palamar et al., 2015 (19)	Included: 32 Completed: 23; randomized double-blind placebo controlled Treatment group: GON block with local anesthetics, n = 11; Placebo group: GON block with local anesthetics, n = 12	One patient self-limiting vasovagal syncope	Refractory migraine without aura	Treatment group: 1.5 ml 0.5% bupivacaine; Placebo group: 1.5 ml saline	One block was performed with portable ultrasound, search for occipital artery in the medial one third of the superior nuchal line between the Po and mastoid process.	In both groups VAS decreased until 2 weeks. After the second week, this reduction continued in the treatment group. The monthly pain intensity decreased from 3.93 ± 1.80 to 1.55 ± 1.42 (p = 0.003) on the injected site in the treatment group.
Cuadrado et al., 2016 (20)	Included: 36 women; randomized double blinded placebo-controlled Active group: GON block with local anesthetics, n = 18 Placebo group: GON block with saline, n = 18	Presyncope Placebo group: 2 Active group: 1 Transitory stinging sensation at the puncture site Placebo group: 2 Active group: 1	Chronic migraine	2 mL of bupivacaine 0.5% or 2 mL saline	Injection was applied at the point 3 cm below and 1.5 cm lateral to the inion bilaterally once.	Mean number of reductions in moderate or severe headache days; intragroup differences are −2 and −0.4 in the treatment and placebo group respectively after 1 weeks' intervention (p = 0.027). There was also statistical difference in the reduction of any-intensity headache days, (−1.5 days in treatment group vs. −0.1 in placebo group; p = 0.04). The number of days with acute medication consumption reduction was higher in the treatment group. Mean PPTs one week after the intervention differed statistically. Only variations were at the supraorbital, infraorbital sites.
Gul et al., 2016 (21)	Included:44; Randomized double blinded placebo-controlled Group A: GON block with local anesthetics, n = 22 Group B: GON block with saline, n = 22	No serious adverse effect Local pain at the site of injection, vertigo, and nausea were reported in some patients	Chronic migraine	Active group: 1.5 mL of 0.5% bupivacaine diluted in 1 mL of saline Placebo group: 2.5 mL of saline	GON block applied, four times weekly; 2 cm lateral and 2 cm inferior to the external Po.	The mean number of headache days and VAS scores were statistically lower compared to pretreatment in Group A at the first, second and third months (p < 0.001). In group B, the mean number of headache days and VAS scores were significantly lower after the first month of therapy compared to pretreatment (p < 0.001). Group A showed significant reduction in headache days and VAS scores at the second and third months of treatment.

n: number; Po: protuberancia occipitalis; TPIs: trigger point injection; GON: greater occipital nerve; PPTs: pressure pain thresholds; VAS: Visual Analog Scale.

Table 2.

Characteristics of open studies.

<

First author, year	Number of patients; study design	Side effects	Headache types	Drugs	Number of blocks; block location	Results
Caputi et al., 1997 (22)	Included: 29 Completed: 27; open preliminary	None reported	Unresponsive migraine patients	0.5 – 1.0 mL 0.5% bupivacaine	Repeated block on alternate days up to 10 blocks to GON and supraorbital nerve; GON performed 2 cm lateral, 2 cm inferior to external Po for GON block	23 patients (85%) showed a decrease in the TPI of at least 50% within the first month. This reduction maintained for a 6-month period.
Afridi et al., 2006 (23)	Included: 101 Injections: 116; prospective 54 migraine patients All patients' headache persists ≥ 15 days	Vasovagal syncopal attack:1 Dizziness: 3 Alopecia: 2 Headache:3	Migraine, cluster, new daily persistent headache, hemicrania continua, other	3 ml 2% lidocaine; 80 mg methylprednisolone	GON block was performed 1–2 cm below midpoint between Po and mastoid process unilaterally once	For the migraine group, complete response mean duration was 9 days with a median of 6-day response. The mean and median duration of partial response was 61 and 30 days, respectively.
Takmaz et al., 2008 (24)	Included: 10; prospective open preliminary	Vasovagal syncopal attack: 1	Migraine	1.5 ml 0.5% bupivacaine	GON block performed three times per week, max. five times. 2 cm lateral 2 cm inferior to external Po	At the end of the first month, TPI and headache attack numbers decreased statistically and these reductions maintained for 6 months.
Okmen et al., 2016 (25)	Included: 60; cohort study	None reported	Migraine	2 mL of 0.5% bupivacaine	GON blocks applied four times weekly; medially and laterally to the occipital nerve which is on the 1/3 part an imaginary line between Po and the mastoid,	VAS and the number of attacks were lower during 6 months compared to prior values statistically. MİDAS scores were lower in the third and sixth months compared to prior values.<
Inan et al., 2016 (26)	Included: 78; retrospective Group PGON: Patients with medical prophylaxis and GON block, n = 25 Group GON: Only GON block, n = 53:	No serious side effect	Migraine without aura	2 ml 0.25 bupivacaine,	GON block applied weekly in the first month than monthly for 2 months In total, six blocks were performed; radially 2 cm inferior 2 cm lateral to Po externa	The headache attack frequencies, VAS and headache duration were significantly decreased in the first, second, and third months after GON block in both groups. No significant differences were detected between groups in severity and headache frequency in the first, second, and third months. Headache duration in group PGON was significantly lower in first, second and fourth weeks.
Pinero et al., 2016 (27)	Included: 60 migraine patients; prospective 24 patients (55.8%) also demonstrated symptomatic medication overuse	No serious adverse effect Vasovagal episodes, discomfort near the injection sites	Migraine patients who had received preventive treatment with no relief or poor tolerance of treatment	1:1 mixture of bupivacaine 0.5% and mepivacaine 2%, total 2 ml	Mean 2.12 blocks were applied per patient. GON block applied above the imaginary line between the mastoid process and the Po, GON injections were performed with at least 2 changes of direction to over an area of about 60?, SON block applied according to tenderness.	23 patients (38.3%) showed complete response for at least 15 days, 24 patients (40%) partial response (a reduction of at least 50% in the number of headache days or a reduction in pain severity for a minimum of 2 weeks), and 13 patients (21.7%) showed no response. Response lasted up to 3 months in some cases.
Unal-Artık et al., 2017 (28)	Included: 42, retrospective Completed: 41 Group 1:unilareal GON block, n = 23 Group 2: bilateral GON block, n = 18	Redness at the injection site (1)	Chronic migraine	1.5 mL bupivacaine 0.5 % 1 mL saline	GON blocks were performed for one month weekly, followed by monthly for 2 months. Medial third of the line between Po and the mastoid process.	Number of headache days decreased in both groups from month 1 to month 3 statistically. Headache duration decreased significantly compared to pretreatment during 3 months by bilateral GON block. VAS decreased in both groups for 3 months. Bilateral and unilateral injection do not make a difference between groups.

n: number; Po: protuberancia occipitalis; TPI: Total pain index; GON: greater occipital nerve; SON: supraorbital nerve; VAS: Visual Analog Scale.

An RCT by Ashkenazi et al. compared the effect of local anesthetic mixture (lidocaine 2% + bupivacaine 0.5%) with or without 40 mg triamcinolone (15). Both bilateral GON block and trigger point injections (TPI) to cervical paraspinal and trapezius muscles were performed. Mean duration of headache, mean analgesic consumption, and the mean duration of being headache-free did not differ significantly between the two treatment groups at the end of the fourth week (p < 0.05). GON block and TPIs were effective, the addition of triamcinolone was useless.

Another RCT was conducted on 48 migraine patients to compare the mixture of 1.0 mL 2% lidocaine with either 0.5 mL saline or 0.5 mL triamcinolone (16). GON block using lidocaine with or without triamcinolone reduced pain severity, frequency and analgesic use at 2, 4, and 8 weeks after intervention, and there was no difference between groups (p < 0.001). However, in both groups, pain severity, pain frequency and analgesic use were higher in the eighth week than the second and fourth week of intervention. The limitation of this study was that patients in both groups received propranolol.

In another RCT, Dilli and colleagues compared 0.25 mL 1% lidocaine (placebo) to a mixture of 2.5 mL 0.5% bupivacaine and 20 mg methylprednisolone (17). They reported that there was no difference in response rate, defined as at least a 50% reduction in the frequency of moderate or severe migraine headache days after the intervention. They also stated that migraine duration, acute analgesic consumption, and mean migraine frequency did not differ statistically in each group. The RCT that was undertaken by İnan et al. compared GON block with 1.5 mL bupivacaine 0.5%, and 1 mL saline (treatment) and GON block with 2.5 mL saline (placebo) (18). GON blocks were applied weekly for a month. Mean headache duration reduced in the treatment group after the first month (p < 0.001), while the placebo group experienced no reduction (p = 0.223). After blinding was removed, there was no difference between groups in the second and third months. Mean headache days reduced in both groups, but this reduction was significant in the treatment group. VAS decreased in both groups after the first month, but the VAS decrease was more significant in the treatment group.

The RCT that was undertaken by Cuadro et al. administered GON block bilaterally in a placebo-controlled study. They detected a reduction in the mean number of moderate or severe headache days and a reduction in the mean number of headache days of any intensity in the first week of application (20).

The last RCT, which was conducted by Gül et al., compared 2.5 ml 0.5% bupivacaine with saline. They detected that bupivacaine is superior to placebo according to headache days and VAS (21). The mean number of headache days and VAS reduced after the first month in both groups, but this reduction did not exist in the placebo group in the second and third months.

In open studies, Caputi and colleagues performed repetitive supraorbital nerve and GON blocks up to 10 times with 0.5–1 mL bupivacaine 0.5% in 27 migraine patients (22). Total Pain Index (TPI) decreased one month after therapy (from 347.1 ± SE 73.9 to 106.8 ± SE 33.8, p = 0.0001) and lasting to a significant degree out to 6 months (60.9 ± SE 15.8). They reported that 85% of the patients had shorter headache duration and frequency and lower headache intensity, persisting for six months. However, supraorbital nerve and GON blocks were used together, and none of the patients had an isolated GON block performed. Afridi and colleagues investigated the response of GON block in different headache syndromes (23). They applied a single injection of a mixture of lidocaine and methylprednisolone to 54 migraine patients. The response was grouped as a complete response (pain free), partial response (reduction in severity or frequency of headache by >30%), or no response. Seventeen of them had a partial response, and nine of them had a complete response. Complete response was (number of pain-free days) nine days with a median of 6-day response. Furthermore, Takmaz et al. reported that GON block with 1.5 mL of 0.5% bupivacaine was effective in the prevention of migraine attacks (24). They reported that a three times weekly injection reduced the mean number of migraine attacks from 12.6 ± 4.8 to 4.9 ± 1.8 (p = 0.005) and mean analgesic consumption reduced from 11.0 ± 3.4 to 4.9 ± 1.1 (p = 0.005) during the first month, lasting up to six months.

In a recent cohort study, the effectiveness of GON block was evaluated on 60 migraine patients (25). GON block with bupivacaine was applied weekly for a month. The decrease of MIDAS, VAS score and the number of attacks were the endpoints of this study. GON block was found to be effective in the treatment of migraine for all endpoints. However, all parameters did not decrease significantly after the second month, and there is no statistical difference between the third and sixth months.

İnan et al. compared GON block effectiveness in patients who used GON block alone and GON block with prophylactic medical agents. According to the trial, there were no significant differences between the groups regarding headache duration and headache attack frequencies. They could not find a difference between groups and did not recommend the addition of prophylactic medication to the GON block (26).

Pinero et al. applied the anesthetic block to migraine patients. In this trial, they examined the sensitivity of pressure on four nerves (two greater occipital nerves, two supraorbital nerves) (27). The block was performed on sensitive nerves only. Complete response for at least 15 days was seen in 38.3% of 60 patients, 40% displayed partial response, and 13 patients (21.7%) showed no response. In his trial, details of blocks were as follows: 28 patients' GON and supraorbital nerve (SON) were blocked together, for 18 patients only SON were blocked, and for 14 patients only GON were blocked. The results of the blocks were not given nerve-specifically. Therefore, it is not possible to say that GON blocks alone have no benefit.

The difference between the bilateral and unilateral GON block was examined by Ünal-Artık and colleagues (28). They observed no differences in frequency, severity and duration of headache between the groups in the monthly visits during the 3-month treatment period. They reported that GON block was effective in the management of chronic migraine and bilateral block was not superior to unilateral block.

Discussion

Recently, a systematic meta-analysis on the efficacy of greater occipital nerve block for the treatment of migraine was published (13). In this meta-analysis, seven RCTs were included. These RCTs were also included in this narrative review. They found that GON intervention could significantly reduce pain intensity and analgesic medication consumption. GON block intervention showed no notable influence on headache duration, and there was no significance in adverse events following GON block intervention compared to the control group. In the above-mentioned meta-analysis, the primary outcome was pain intensity. Secondary outcomes included headache duration, analgesic medication consumption, and adverse events. On the other hand, in this narrative review, 14 trials were included. Blocking techniques, drugs and dosages, differences between bilateral and unilateral blocks, predictive factors of effectiveness, the effectiveness of single or recurrent injections, complications and side effects were reviewed and discussed.

Blocking techniques

Different GON block techniques have been described in the literature (15–28). The most applied and recommended technique is injection just over the tender point lateral to the occipital protuberance by one-third of the total distance from the protuberance to the mastoid (15–17,19,25,27). This procedure was applied in four RCTs. Another recommended technique is applying GON to 2 cm lateral and 2 cm inferior to the external occipital protuberance (18,21,22,24,26). This procedure is preferred in two RCTs. In another RCT, Cuadro et al. also detected GON block to the point 3 cm below and 1.5 cm lateral to the inion bilaterally (20).

Ultrasonography (USG) can be used for nerve block. Palamar et al. (19) administered GON block with the help of portable USG to locate the occipital artery. In this RCT, they performed block medial to the occipital artery. Without the use of USG, palpation of the artery and blockage to medial of the artery can be recommended. Single needle insertion and negative aspiration before injection of volume and dispersion of solution by diffusion may be recommended rather than four quadrant needle insertion. This procedure may increase side effects and complications. It is also recommended to test for hypoesthesia and paraesthesia at the area of nerve to show the effectiveness of GON block (3). Weibelt et al. applied occipital nerve blockage to cervicogenic chronic migraine patients. They inserted the needle on the suboccipital area and injected the solution at three immediately adjacent sites. Complications were detected in 24 of 150 patients in this trial (29). As the complication rate was high, this procedure may not be recommended.

Drugs and dosages

The effect of local anesthetic agents is associated with reversible blockade of sodium channels within the nerve fibers. They act on demyelinated C-fibers and myelinated A∂ fibers which prevent transmission of pain signals with disrupting depolarization of the nerve (30). Lidocaine and bupivacaine are the most commonly used agents for GON block.

In the studies reviewed here, as a local anesthetics, 2% lidocaine between 0.25–4.5 mL and 0.5% bupivacaine between 1.5–4.5 mL were used in RCTs (Table 1). In open studies, 2% lidocaine 3 mL; and 0.5% bupivacaine between 0.5–2 mL; 0.25% bupivacaine 2 mL and 1:1 mixture of 0.5% bupivacaine and 2% mepivacaine, total 2mL, were used (Table 2).

Six RCTs and seven open studies found that local anesthetics at different levels were effective. The same dosages (1.5 mL of 0.5% bupivacaine) of local anesthetics were used in three RCTs (18,19,21) and two open studies (24,28). These studies showed effectiveness. Cuadrado et al. used 2 mL of 0.5% bupivacaine, and the response rate was higher than placebo in this RCT (20). A mixture of 4.5 mL 2% lidocaine and 4.5 mL 0.5% bupivacaine showed effectiveness in one RCT (15). Also, the effectiveness of 1 mL 2% lidocaine was shown in an RCT (16). Only Dilli et el. could not find a significant difference between groups. In this RCT, 0.25 mL of 0.5% lidocaine and a mixture of 2.5 mL 0.5% bupivacaine and methylprednisolone were compared (17). However, in both groups, there was a reduction in migraine frequency, severe headache days, analgesic consumption and migraine duration. This may suggest a question, of whether 0.25 mL 0.5% lidocaine may have had a positive effect in the control group.

Corticosteroids decrease inflammation by inhibiting the release and synthesis of proinflammatory substances, directly stabilize membranes, reversibly inhibit nociceptive C-fibers and modulate nociceptive input in substantia gelatinosa (7). The steroid was added to a local anesthetic agent in some trials (15–17,23,27). Ashkenazi et al. (15) conducted an RCT that evaluated the efficacy of local anesthetics with or without triamcinolone in 37 migraine patients who underwent GON block and TPI. Researchers evaluated the variables of 24 migraine patients who returned a four-week symptom calendar and reported that the addition of 40 mg triamcinolone to the local anesthetic for combination therapy did not provide additional benefit to migraine patients. Kashipazha et al. (16) compared the addition of saline and triamcinolone to 1.0 mL lidocaine. They reported that pain severity and frequency were not different among groups, but also that they decreased. Afridi et al. used a mixture of 3 mL of 2% lidocaine and 80 mg of methylprednisolone for GON block in the open study (23). They found it was effective. As a result, we can say that GON block with local anesthetics may have a positive effect in migraine prophylaxis, but steroid addition may not provide extra benefit.

Future studies that are comparing lidocaine and bupivacaine and different dosages of local anesthetics in GON blocks may be helpful.

Bilateral or unilateral injection

Palamar et al. detected that GON block is only effective on the injected side (19). Ünal Artık et al. suggest that headache duration, VAS and headache days did not differ when applying GON block bilaterally or unilaterally (28). The limitation of this study is that it was retrospective and GON tenderness is unspecified. According to this study's results, only unilateral blockade may be recommended so that adverse effects may be minimized.

Predictive factors

Predictive factors positive to GON block were evaluated in three RCTs. Kashipazha et al. detected that GON tenderness does not affect pain frequency, analgesic use and pain severity (16). Dilli et al. detected that triptan and opioid usage did not affect GON block response (17). There are several studies on the relationship between GON tenderness and GON block response (17). In contrast, Cuadrado et al. suggest that patients who had GON hypersensitivity responded 50% less to GON block than their counterparts, which was statistically non-significant (20).

In open studies, Afridi et al. administered GON block and suggested that 20 of 31 migraineurs who were overusing analgesics or triptans had a response to the injection. They suggested that there was no significant relationship between the response to injection and medication overuse (23).

İnan et al. were the first researchers to study the effectiveness of taking or not taking prophylactic drugs in migraine patients. The results of the present study suggest that prophylactic drug treatment in migraine patients did not increase the benefits of GON block in a 3-month follow-up period. (26).

Pinero et al. administered GON and/or SON blocks to patients who showed sensitivity to pressure at the exit point of GON and/or SON and compared patients' demographic and clinical variables between those who responded and those who did not respond to treatment. In total, 60 patients were included in the trial; 23 and 40 patients showed complete and partial response, respectively. Thirteen out of 60 patients showed no response. According to the trial, there was no difference between patients with response and no response in terms of age, progression time (years), days of pain in the previous month, days taking medication in the previous month, days taking triptans in the previous month, gender, and use of preventive treatment when anesthetic block was performed (27). So, we may say there are no predictive findings for the effectiveness of GON block.

Single or recurrent injection

Repeated GON blocks were applied in two RCTs. The mean number of headache days and VAS scores were lower in local anesthetic groups than placebo. Only single GON block was applied in five other RCTs. Four of these showed effectiveness for headache severity, pain frequency and analgesic use (Table 1). Kashipazha et al. (16) detected that headache frequency, severity and the need for analgesic use was significantly lower than the baseline for 8 weeks. In Dilli's study, only one block was applied and the results were not significant (17). In the RCT undertaken by Inan et al., GON blocks were applied four times weekly. There was a significant decrease in the number of headaches and VAS scores. After blinding was opened, GON blocks with bupivacaine were applied to the saline group four times weekly, and a similar significant decrease was found (18).

Palamar et al. performed one block with portable ultrasound. The active group statistically differed from the placebo group in VAS between weeks 2–4 in this RCT (19). Cuadrado et al. applied GON block once. They found a reduction in the number of days with a moderate-to-severe headache or any headache during the week following injection (20).

In another RCT, Gül et al. applied GON block once a week four times and then followed the patients for three months. They detected that headache days and VAS reduced in the first month of the trial in both the placebo and treatment group. This reduction did not continue in the second and third months in the placebo group. In the repeated GON blocks group, reduction of VAS and headache days continued in the second and third months (21).

According to RCTs, the effectiveness of GON blocks on headache frequency and VAS has been found even after the second and third months. In open studies, Takmaz et al. (24) first examined the effectiveness of repetitive GON block in migraine. They performed blockage three times. Afterwards, these were applied for a maximum of five times weekly depending on clinical response. Finally, the frequency of migraine attacks decreased throughout the 6 months. Ökmen et al. (25) administered GON block once a week for 1 month, and its efficacy continued for 6 months. Also, Pinero et al. administered repetitive GON block for three of the 13 patients who did not respond to the first procedure, showing a response in subsequent interventions (27).

Greater occipital nerve block was applied four times, once a week, by Inan et al. (26) and Ünal-Artık et al. (28) and once in the following two months. They found a significant decrease in headache severity, frequency and duration.

Greater occipital nerve block effectiveness can be maintained for several months. Repeated blocks may be necessary when the single block is not sufficient; in order to reach maximum efficacy, repeated blocks may be applied. Further studies may reveal which is better, weekly or monthly repeated GON blocks.

Complications and side effects of GON blockade

Complications of GON block using local anesthetic agents include infection, hematoma and damage to the structures at the site of injection. Vertigo, nausea, and rarely cardiac arrhythmia, seizure, respiratory depression and hypersensitivity reactions related to amide local anesthetics are some systemic side effects of GON block (8). Steroid-related side effects include Cushing's syndrome, focal alopecia at the site of injection, and the development of cutaneous atrophy (31,32). In addition, GON block should not be administrated in patients with a cranial defect or infection at the injection site and allergy to local anesthetic or corticosteroids (33).

Dilli et al. detected that the placebo and treatment group did not differ statistically in adverse effects (17). A total of 600 patients were included in studies of this review and side effects were only reported 39 times. Some of the patients received more than one block. Many of the side effects are minimal, such as presyncope, vertigo, or pain at the injection site, so the GON block procedure seems safe. Application of negative aspiration during the injection should be undertaken in order to avoid injection in an artery. Thus, the risk of development of side effects is minimized. It is not necessary to administer local anesthetics in multiple directions, which increases the risk of injury due to their diffuse spread. Patients should be monitored for the development of side effects for 30 minutes after blockade and should be asked and examined for anesthesia at the GON area. This way, effective block is proved.

Conclusion

Local anesthesia may be one of the useful treatment modalities. Palpation of the occipital artery and block of GON medial to the artery or the most sensitive part at the point lateral to the occipital protuberance by one third of the total distance from the protuberance to the mastoid and/or use of USG are the recommended methods despite anatomic variation of the nerve. However, in most studies USG was not used and complication rates were generally low. USG and other radiographic techniques add to unnecessary patient costs and have not been proven to improve outcomes, based on this investigation.

Single point injection is sufficient, due to the diffusion of the drugs. There is no significant contribution from steroid injection. There may not be a need for bilateral GON block; unilateral blockade seems sufficient. We do not yet know the best efficient dose and frequency of local anesthetics. There is no predictive marker for the effectiveness of GON block. Repeated blockade may be recommended for migraine. Adverse events seem very rare. Separate evaluation can be done according to the patient and course of migraine.

Limitations

All of the studies included in this trial consists of small patient groups and study design varies in all studies. The lack of detail on migraine frequency is also a limitation. Different anesthetics and different block techniques were also detected in studies. Thus, we cannot compare one study's results with those of another study.