The primary indication for nerve blocks of the head and neck is for diagnostic and therapeutic purposes in cases of head and neck pain. Some of the more common indications are discussed in detail in this chapter, including the rationale for selecting regional nerve blocking techniques in the care and management of these patients.
Headaches
Differentiating the etiology for headaches is a vexing task, often requiring the skill and experience of multiple clinicians from diverse specialties. The pain practitioner and skilled regional block specialist are well suited to the task by virtue of their respective experiences in performing conduction blockade of the greater and lesser occipital nerves (occipital headache), the cervical medial branches of the facet joints (cervicogenic headache), the cervical epidural nerve block (nonspecific headache), and the atlanto-axial joint (suboccipital headache). Therefore, it is by use of the various modalities of regional block that a diagnostic/therapeutic block of the respective structures implicated in a given patient’s pain may be discernible. The anatomic method is a more scientific method of pursuing the source of a headache than that of solely relying on historical information presented by the patient.
Atlanto-Axial Joint Block
The primary indication for performing atlanto-axial (A-A) nerve block is for the diagnostic and therapeutic evaluation of suboccipital pain that occasionally radiates into the temporomandibular joint (TMJ) area that is exacerbated by head rotation. Whiplash injuries and cervicogenic headaches are two of the more common indications for this block. The A-A joint ( Figs. 52.1 and 52.2 ) lacks posterior articulations and therefore is neither a bona fide facet joint nor a true zygapophyseal joint. Also, there is no intervertebral disk between the atlas (C1) and the axis (C2), nor is there an intervertebral foramen at that level to accommodate an exiting nerve root. At the A-A joint, the head flexes, extends, and rotates in a horizontal plane, up to 60 degrees, giving the joint significant responsibility for both the stability and the mobility of the head and neck. The joint can be injured by seemingly trivial insults and trauma. The resulting pain syndrome can be significant, manifesting as dull, continual and achy pain in the posterior neck and suboccipital region. More severe injuries, such as those caused by motor vehicle accidents, might subject the joint to acceleration-deceleration type injuries, with sequelae exceeding pain and dysfunction. Indeed, paralysis and even death could result from ligamentous disruption, analogous to an odontoid fracture. The fibers of the respective spinal nerves C1 and particularly C2 contribute to the formation of the occipital nerves (greater and lesser), which are frequent targets for the practicing pain physician. An extremely important anatomic concept involves the relationship of the vertebral artery to the A-A joint; whereas the artery is medial to the atlanto-occipital (A-O) joint, it is found lateral to the A-A joint. Therefore, needles directed at the joint during nerve block need to be oriented slightly more medially, cognizant of the danger of aiming toward the interlaminar space, or even toward the foramen magnum.
Technique of Atlanto-Axial Nerve Block (See Figs. 52.3 and 52.4 )
Blockade of the A-A joint requires fluoroscopic assistance to ascertain that the advancing needle does not encroach on the critical anatomic structures such as the vertebral artery and the spinal cord. Patients are placed prone, after obtaining the appropriate medical history, performing a targeted physical examination, and establishing that there are no bleeding problems or infectious issues related to the target for intended needle placement. Baseline vital signs are obtained and recorded. It is recommended that an intravenous access be established for prophylactic purposes. Bolsters are typically placed beneath the chest to elevate the shoulders off the procedure table, permitting the patient to flex the neck and rest the forehead on a neurosurgical donut or pillow. After performing a sterile skin preparation and draping, the fluoroscopic unit is oriented in an anterior-posterior direction to identify the atlas and the foramen magnum, implementing a moderate craniad tilt of the unit until all structures are clearly visualized. The A-A joint is located lateral and inferior to both the foramen magnum and the atlas (see Fig. 52.2 ). A local anesthetic skin wheal is made over the intended injection site using a small gauge, 1.5-inch needle. Next, an 18-gauge skin core is made using a sharp cutting needle to permit the passage of a blunt, 22-gauge, styletted Whitacre type subarachnoid needle. A curve is made at the distal tip of the needle to allow better steering once it has been advanced through the skin and subcutaneous tissues. The needle is advanced under continued fluoroscopic guidance, rotating the beam of the unit until the needle appears in tunnel or “gun barrel” view, which is represented as a dot advancing toward the posterolateral aspect of the A-A joint. The needle should be directed slightly medially to avoid the vertebral artery, which is situated laterally, but not too medially, as this would potentially engage the spinal cord. Occasionally, but not always, a “popping” sensation will be appreciated as the needle traverses the joint and enters it from posterior to anterior (see Fig. 52.3 ). The fluoroscopic unit must then be rotated laterally (see Fig. 52.4 ) to confirm the needle placement at the appropriate depth between the atlas and axis. Once placement has been verified, gentle aspiration of the needle is undertaken to assess for the presence of cerebrospinal fluid (CSF) or blood. If none is present, a small (i.e., 1-mL) volume of radiocontrast media may be incrementally injected under real-time fluoroscopy. If the needle is situated within the confines of the joint, a bilateral concavity will be demonstrated, indicative of an intact joint capsule. If the joint capsule has been ruptured, then the dye may be seen as spreading into the peridural space, and care should be taken not to inject long-acting local anesthetic (bupivacaine, ropivacaine) through the needle. Rapid runoff of the dye may signify vascular injection and is to be guarded against, particularly if it is suspected that the needle may have entered the laterally situated vertebral artery. If this occurs, the needle should be redirected medially, re-aspiration of the needle in four quadrants undertaken, and re-injection of contrast material prior to considering injecting local anesthetics or adjuvants. Even without directly injecting into the vertebral artery (even miniscule volumes of dilute local anesthetics injected here can lead to grand mal seizures), ataxia following A-A block is not uncommon and is likely due to vascular uptake of local anesthetics in this extremely vascular region.
Third Occipital Nerve Block
Diverse pain syndromes result from disorders of the cervical intervertebral disks, the facet joints (zygapophyseal, or Z joints), or both. When confronted by individuals with headache and neck or shoulder pain, the clinician is compelled to seek a diagnosis based largely on history and physical examination. The physical examination is complicated by the finding of significant variability and overlap in the innervation patterns related to the sensory distribution of the cervical medial branch nerves ( Fig. 52.5 ). The third occipital nerve has been implicated in chronic headache pain, primarily as attributed to cervicogenic headache. Syndromes related to dysfunction of the medial branches of the upper cervical spine have at times been misdiagnosed as being due to tension-type and other forms of headache. Patients, having been misdiagnosed as such, have ultimately failed to derive analgesic benefit from medication administration and other conservative treatment measures that were misguided.
The third occipital nerve (TON) is the superficial branch of the C-3 dorsal ramus. It is the only medial branch that innervates the facet joint at C2-C3 ( Figs. 52.6 through 52.8 ). Therefore, successful blockade of the TON, and the resultant block of the facet joint at C2-C3, has been associated with relief of headache pain in individuals suffering from “third occipital headache.”
Anatomic Considerations
The C2-C3 joint is the first joint of the cervical spine possessing a true joint capsule and synovium, and this level is the first wherein an intervertebral disk exists and wherein a foramen exists to accommodate the exiting C3 nerve root. In this regard, the C2-C3 joint is distinct from the atlanto-occipital (A-O) (C0-C1) and atlanto-axial (A-A) (C1-C2) joints superior to it. This level, then, represents a sort of transitional zone between the rotational joint of the neck (AA) and the lower cervical facet joints, which function not in neck and head rotation but in neck flexion and extension.
Facet joints contain free and encapsulated nerve endings. In addition, nerves in the zygapophyseal joint (Z joint) contain substance P and calcitonin gene-related peptide (CGRP). The capsule of the Z joint contains low-threshold mechanoreceptors, mechanically sensitive nociceptors, and silent nociceptors. Each of these may respond to noxious stimulation, including moderate to severe levels of osteoarthritis, with the result being a nociceptive stimulation perceived as headache or neck pain.
The clinical anatomy of the cervical dorsal rami was described by Bogduk, following dissections of five adult cadavers. He pointed out that classic textbook descriptions of the cervical dorsal rami had been limited in scope and detail. While dissecting the medial branches, he was able to place wires superficially and parallel to the nerves and then perform radiographic analyses of the wires in relationship to the cervical vertebral skeleton. He noted that the semispinalis capitis muscle covers the cervical medial branches, whereas the lateral branches of C3-C7 lie superficial to the tendons of origin of that muscle. The TON penetrates the semispinalis capitis. The C3 dorsal ramus, a short nerve, was noted to arise from the C3 spinal nerve in the C2-C3 intervertebral foramen and then to curve dorsally through the intertransverse space. At this point, the C3 dorsal ramus was noted to divide into major branches: the two medial branches, the lateral branch, and a communicating branch. In three out of five of his specimens, the two medial branches of C3 were noted to arise from a common stem, and in the other two specimens they had their own respective origins. The third occipital nerve (TON) was the principal and constant medial branch of the C3 dorsal ramus. After arising from the C3 dorsal ramus, the TON was found to curve dorsally and medially around the superior articular pillar of the C3 vertebral body, crossing the C2-C3 Z joint either just below the joint itself or at the joint level. The TON runs transversally medially through the fibro-adipose tissue below the obliquus inferior and dorsal to the lamina of the C2 vertebra.
A branch of the TON supplies the semispinalis capitis, which lies superficial to it. A communicating branch to the greater occipital nerve (GON) is also given off. This branch arises just above the level of the C2 spinous process. The TON then passes dorsally and pierces the semispinalis capitis and the splenius capitis, turning rostrally to pierce the trapezius, which lies over it. The more medial terminal branch supplies the skin of the rostral neck and the occiput below the external occipital protuberance. More lateral branches travel toward the mastoid process, communicating with the cutaneous rami of the GON and lesser occipital nerves (LONs). Because of the transverse direction of the GON and the TON, these two nerves may be injured by parasagittal incisions made in the upper part of the neck.
The target branch for nerve block, the TON, wraps dorsally and medially around the middle of the waist of the articular pillar of C3 (center of the bony trapezoid). As it moves more medially, the TON invests the multifidus muscle. There is a communicating branch that typically arises from the C3 dorsal ramus, running rostromedially across the posterior part of the C2-C3 Z joint. The anatomic target for locating the nerve percutaneously, then, is to utilize lateral fluoroscopy to identify the waistline of the articular pillars of C2 and C3 and to advance a needle or radiofrequency cannula toward the Z joint at C2-C3. As Bogduk stated,
[N]eedles or electrodes introduced obliquely, ventromedially onto the dorsolateral aspect of the articular pillar, will rest on the medial branch. The relationship of the nerve to bone at this site is constant, because the medial branches are bound to the periosteum by an investing fascia and are held against the articular pillar by tendons of the semispinalis capitis.
This target is removed from any major arterial (vertebral artery) or other vascular structure, as well as from the exiting spinal nerve and spinal cord; it is therefore an appropriate target in terms of affording access to the medial branches as well as one that provides for an approach likely to minimize unwanted trespass into nontargeted tissues.
The Third Occipital Nerve and Headache Pain
Osteoarthritis of the C2-C3 Z joint and trauma related to motor vehicle accidents are causes of persistent occipital or suboccipital headache pain. As there is no clear-cut clinical evaluative technique or diagnostic tool for determining who is suffering from headache due to disease of this joint or from neck trauma (or that mediated by the TON), it is somewhat intuitive and logical that blockade of the TON should be performed in cases of persistent headache that are unresponsive to conventional medication management.
Bogduk and Marsland conducted fluoroscopically guided TON blocks in 10 consecutive patients presenting with occipital or suboccipital headache. They elected to place their needles at the lower half of the silhouette of the C2-C3 Z joint, recognizing the landmark on x-ray as a convexity arising upward from the concavity of the C3 articular pillar, lying horizontally opposite the level of the C2 spinous process and C2-C3 disk space ( Figs. 52.9 through 52.11 ). All procedures were performed with patients in the prone position and with a bolster beneath the upper chest and shoulders to permit head flexion forward toward the procedure table. A solution of 0.5 mL of 0.5% bupivacaine was injected at three distinct sites: in the middle of the Z joint at C2-C3, at the lower end of the joint, and between the first two needle placements (for bilateral blocks the total volume was therefore 3 mL; 1.5 mL per side). Seventy percent of patients manifest total pain relief lasting the duration of the bupivacaine, following unilateral blocks. The 30% failures did not have positive responses to the bilateral blocks performed on their behalf. Bogduk stated that blockade of the medial branch nerves, particularly the TON, are easier for patients to tolerate than intraarticular injections because they are more easily performed and are associated with less pain during their performance than are the articular injections. Furthermore, it is apparent that local anesthetic injected into a joint may not actually stay within the joint space, potentially limiting any diagnostic information that might otherwise be derived from the procedure. Finally, markedly degenerated joints might not readily accommodate an advancing needle tip, rendering the choice of this articular approach potentially useless overall. If the ventral margin of the joint space is penetrated, there is also a distinct possibility that entrance into the epidural or subarachnoid spaces may occur, with potentially devastating consequences.
Although no one knows for certain how likely it is that pain in the occipital and suboccipital area emanates from the C2-C3 Z joint and although various studies show greatly disparate findings, several compelling attempts at determining the incidence (a measure of the risk of developing some new condition within a specified period of time) or prevalence (the total number of cases in the population, divided by the number of individuals in the population) need to be reviewed.
A study involved 24 consecutive patients presenting with undiagnosed neck pain who were evaluated using controlled diagnostic blocks incorporating low-volume local anesthetic injections. For individuals with occipital or suboccipital headache and neck pain, the lower half of the lateral margin of the C2-C3 Z joint was selected as the injection site. Complete pain relief was considered to occur if the patients, having received the bupivacaine blocks and then going home to perform their activities of daily living, noted at least 2 hours of total analgesia. Failure to derive pain relief resulted in patients undergoing additional cervical medial branch blocks at contiguous levels. Nineteen out of 24 patients had positive results, with 9 of these ultimately having pain isolated to the C2-C3 level. Therefore, in 47% of patients who had diagnostic block-proven cervical Z joint pain, the TON was the nerve responsible for mediating the pain. Although this study has not been replicated, it is tempting to use the information as a general rule of thumb when approaching individuals with neck pain resulting from degenerative disease of the cervical spine and when planning nerve blocks based on the correlation of symptoms with joint pain maps. If it is known that in half of cases the level responsible for a given pain problem is at the C2-C3 joint, then it might save time, expense, and patient discomfort to rapidly move toward ruling in or ruling out that level, and the TON, at the outset as part of the diagnostic workup. However, as discussed later, these figures may be overreaching in terms of the actual prevalence of C2-C3–related pain. Although TON block may be performed easily and typically is without significant complications, side effects do occur commonly. Indeed, any patient presenting for this type of procedure must be duly appraised that successful blockade of the TON (or GON) will likely result in the temporary development of ataxia and some gait unsteadiness. For this reason, bilateral blocks at these levels should likely be staggered, or else the benefit-risk ratio must be significant enough that performing bilateral blocks is clearly warranted.
The subject of joint pain maps merits special mention, as noted previously (see Fig. 52.5 ). Multiple attempts have been made at delineating the specific cutaneous area of sensation, and hence pain, invested by each respective facet joint and medial branch. In the report by Fukui and associates, the site of Z joint intraarticular injection and electrical stimulation (selected cases) was chosen based on any focal paraspinal tenderness. They observed that with injections performed at C2-C3 (n = 14), symptoms were referable to the upper posterior cervical region (64%), the occipital region (50%), or the upper posterolateral cervical region (50%). Windsor and colleagues electrically stimulated the cervical medial branches, including the TON in 9 patients suffering from chronic neck pain. They used the lateral midpoint of the C2-C3 Z joint to indicate the location of the TON. Relatively reproducible referral patterns were identified for the TON nerve distribution as well as for the other medial branch nerves studied.
When reviewing these studies, however, considerable overlap in the areas served by each of the respective segments is noted, much more so than that observed with attempts at mapping spinal nerve dermatome distributions. This produces a diagnostic conundrum, as it is entirely conceivable when approaching cervical spine pain that even perfectly performed techniques, using standardized and accepted anatomic landmarks, may result in failures to derive antinociception with the resultant confusion on behalf of the patient and treating physician.
Because both the intervertebral disk and facet joint are fairly equivalently represented as sources of neck pain (approximately 40% each as sources of neck pain, according to two separate studies), a missed diagnosis might lead a pain physician in the wrong direction when seeking to provide pain relief to a given individual. Mapping studies may help to pinpoint discrete areas of pain referral sources that are then amenable to local anesthetic blockade or radiofrequency lesioning techniques. In one such study, joint maps were created by distending the respective joint capsules at segments from C2-C3 to C6-C7. This was done by sequential injections of contrast media, which could be subsequently identified using fluoroscopy. Each joint was noted to produce a characteristic, distinguishable pain pattern from which pain charts could be constructed (see Fig. 52.5 ). Although only four subjects were evaluated, that number is not too disparate from the original small group of subjects selected to develop the original dermatome charts in neurologically impaired individuals; such charts have been mostly unchallenged since the 1950s. In 10 subjects who received injections based on maps such as those derived and described earlier, 90% had complete concordance of the predicted level of pain and the positive response to the blocks. The C3-C4, C4-C5, and C5-C6 levels were most commonly affected, with only 10% (1 of 10) patients having pain that was documented to result from processes occurring at the C2-C3 (TON) level. These percentages are therefore less indicative of the possibility of the C2-C3 level being primarily involved in head and neck pain than they are of alternative levels lower in the cervical spine. However, a study by Barnsley and Bogduk demonstrated that, following cervical medial branch injections performed in 16 patients suffering from chronic neck pain, 33% (3 of 9) patients who had complete pain relief were patients for whom the TON was treated. It remains unclear how likely the TON is to be implicated in individuals suffering from chronic head and neck pain. The closest studies performed to identify the prevalence of TON-induced headache and neck pain may have been those performed by Lord and Barnsley and colleagues. In the first study, 100 consecutive patients underwent double-blind, controlled diagnostic blocks of the TON. On two separate occasions the nerve was blocked using either lidocaine or bupivacaine. The diagnosis of TON nerve involvement was only made if the double diagnostic blocks both relieved the symptoms, with the bupivacaine injection-induced pain relief outlasting that provided by the lidocaine. The prevalence of TON headache among whiplash patients was 27% and among those with dominant headache was 53%.
In the second study, 50 consecutive patients with chronic neck pain and whiplash injury underwent double-blind, controlled diagnostic medial branch blocks using either 0.5-mL lidocaine (2%) or bupivacaine (0.5%) in a random fashion. In 12 of 27 patients (44%), the source of symptoms was identified as being isolated primarily to the C2-C3 Z joint.
These findings corroborated the impression that although there may be substantial overlap in cutaneous innervations of the central and peripheral nervous systems, these joint maps are essential tools in assessing the appropriate approach to a given pain process in any individual. They also demonstrated variability in determining the relative incidence and prevalence of C2-C3 Z joint pain, depending on technique of assessment and provocation maneuvers to assess the joint and the medial branch nerve (TON).
Rationale for Blocking The Ton
The TON block is useful in the diagnostic and therapeutic phases of treating cervicogenic headache. Evidence of efficacy for both local anesthetic blocks as well as for radiofrequency ablation of the cervical medial branches has accrued. However, only recently has the use of medial branch blocks and radiofrequency ablation techniques received scientific support in the literature. Indeed, two separate attempts at meta-analysis have shown somewhat disparate results. The first paper, published in 2001, suggested that radiofrequency neurotomy used for treating pain from the cervical Z joints after flexion-extension injury was only supported by limited scientific data. However, a mere six total studies satisfied the authors’ criteria for inclusion, and so it is entirely possible that the data suffer lack of suitable cohort studies to make a true assessment of its validity. The second review performed in 2007 liberalized the inclusion criteria somewhat and found that the support for medial branch cervical block was moderate (level III) as was the support for cervical medial branch neurotomy. Here the authors relied on the criteria established by the Agency for Healthcare Research and Quality (AHRQ), which included some nonrandomized trials. They also incorporated studies from the Cochrane Musculoskeletal Review Group, which were randomized trials. One of the important points noted was that the evidence for performing cervical intra-articular facet joint injections was limited both for short-term as well as for long-term analgesia efficacy. This corroborates Dr. Bogduk’s earlier assertions that not only are medial branch techniques more likely to target the real source of pain associated with cervical degenerative facet conditions, but also that articular procedures have an inherently high failure rate because there is no guarantee that injected local anesthetic will remain in the confines of the joint.
The rationale for blocking the TON is twofold:
- 1.
To provide a diagnostic evaluation of the C2-C3 facet joint as being the source of pain in the occipital and suboccipital area
- 2.
To provide an indication as to whether or not neuro-ablative techniques applied over that nerve might be fruitful in treating pain on a long-term basis
We suggest performing double-diagnostic blocks using 0.5 mL of local anesthetic solutions, typically 0.5% ropivacaine plain, before undertaking neurotomy of the TON. Ropivacaine is used for two major reasons. It has intrinsic vasoconstrictor properties, and epinephrine does not have to be added to it. It also has a cardiovascular safety profile that is superior to that for bupivacaine. Unintentional injection into vascular structures, including the vertebral artery, is less likely to result in serious morbidity when ropivacaine is used, in contradistinction to bupivacaine. The further rationale for using ropivacaine is to provide duration of analgesia that exceeds that provided by shorter-acting agents such as lidocaine or mepivacaine. In terms of assessing efficacy, it is likely that longer-acting local anesthetics may provide greater information than shorter-acting drugs, as the effect of the short-acting drugs may be so fleeting as to provide confusing or inexact clinical information.
Double-diagnostic blocks have been described for assessing the efficacy of cervical medial branch blocks. Barnsley and associates compared single to double-diagnostic blocks in 55 adult patients who had been having neck pain for longer than 3 months. They used 0.5 mL of either 0.5% bupivacaine or 2% lidocaine, randomly selected. The duration of analgesia was assessed in a double-blind fashion. They found that the false-positive rate (a test that shows evidence of a disease when it not present) of single blocks was 27% (16/60). They also found that the most common positive levels were at C2-C3 and C5-C6. Lord and colleagues found that comparative blocks of the medial branches using lidocaine, bupivacaine, or normal saline in a randomized, double-blind, placebo-controlled fashion in 50 patients had a specificity (the statistical probability that an individual who does not have the particular disease being tested for will be correctly identified as negative, expressed as the proportion of true negative results to the total of true negative and false positive results) (TN/TN + FP) of 88%, but only marginal sensitivity (the proportion of individuals in a population that is correctly identified when administered a test designed to detect a particular disease, calculated as the number of true positive results divided by the number of true positive and false negative results) (TP/TP + FN) (54%). Hence, some 46% of patients who are not placebo responders would incorrectly be labeled as placebo responders if diagnoses were based solely on comparative blocks.
Slipman and colleagues retrospectively reviewed 18 patients with chronic persistent daily headaches lasting a mean of 34 months following whiplash injury who underwent intraarticular C2-C3 facet joint injections. They noted that in 61% of cases, headache frequency diminished from daily to less than three headaches per week. Although this work is retrospective, the results are nevertheless supportive for targeting the TON in chronic headache conditions refractory to conventional conservative management. Furthermore, because the injections were articular and not at the medial branch (TON), the information gleaned from this report may not be useful in determining candidacy for radiofrequency neurotomy.
Technique for Ton Block
We perform C2-C3 medial branch (TON) procedures with the patient in the prone position using continual live fluoroscopic guidance throughout the injection phase (see Figs. 52.9 through 52.11 ). A bolster is placed beneath the shoulders and chest to elevate the thorax off of the table and to permit gentle head and neck flexion forward. An intravenous cannula is placed for purposes of administering resuscitative medications in the unlikely event that it is necessary to treat vasovagal syncope or for supportive medication administration in cases of unintentional vascular injection. Vital signs are monitored using standard American Society of Anesthesiologists (ASA) basic monitors including pulse oximetry and noninvasive blood pressure assessment. After performing a sterile skin preparation, scout films are obtained of the neck in an effort to identify the scalloped margins of the lateral vertebral bodies. Then, skin wheals are raised over the intended target(s) using a hypodermic needle and 1 to 3 mL of lidocaine solution with epinephrine, 1:200,000 (5 mcg/mL). Using short-beveled, 22-gauge, 2.5- to 3.5-inch needles, the targeted lateral scalloped margins of the vertebral bodies are advanced upon until bony contact is made. At this point the fluoroscopy unit is rotated laterally to assess the relationship of the advancing needle tip to the TON target, in the center of the C2-C3 Z joint (discussed previously). The needle tip must be recessed posteriorly away from the C2-C3 intervertebral foramen as well as from the known location of the vertebral artery. If bone has not been contacted and 2 cm of the needle has been inserted into the skin, then a reassessment of the approach and direction is undertaken while still in the A-P fluoroscopy mode. Once the needle is appropriately seated on bone at the target site ( Fig. 5.4 ), the A-P view is once again undertaken to verify that the needle(s) is in the correct position vis-à-vis the lateral vertebral body margin.
Ultrasound-guided approaches are rapidly gaining popularity for all forms of interventional pain procedures, but in our experience they have not yet reached the level of sophistication to supplant fluoroscopy use for this particular procedure. At this point, aspiration tests are performed to verify the absence of blood or cerebrospinal fluid, and the patient is queried to ascertain that no paresthesias have been elicited. If there are none and vital signs remain stable and close to baseline values, 0.5 mL of 0.5% ropivacaine is injected. No glucocorticoid is added to the solution, as Manchikanti and associates, have demonstrated that there is little value, if any, of adding steroid to the blocks. Still, many clinicians continue to add a steroid in their practice. If the risk-benefit ratio favors using steroids in an individual suspected of having an inflammatory component to his or her pain, then there may be little harm in adding a nonparticulate steroid in judicious doses. After injection, the needle(s) is cleared and withdrawn, and a sterile dressing is applied over the injection site. The patient should be observed for at least 30 minutes, as it is common for patients to become ataxic or have unsteady gait following successfully performed procedures. Discharge instructions should be clear and precisely instruct patients to seek emergency medical care if they develop any delayed-onset side effects or complications from the procedure.
Radiofrequency Neurotomy
In selected patients, radiofrequency (RF) of the TON may be considered for long-term therapeutic benefit. In a study published in 1995, Lord and colleagues demonstrated the efficacy of performing radiofrequency ablation (RFA) of the TON in only 40% of patients (4 of 10), with only three having long-lasting pain relief. They used 10-cm needles with either 4-mm or 6-mm exposed tips. No stimulation test was performed to assure concordance of needle placement with the nerve; instead, the authors relied solely on fluoroscopic anatomic cannula placement. The mean duration of the C2-C3 neurotomy procedure was stated to have been 1.5 hours. The rather meager success rate was in sharp contrast to the 70% success that they noted for lower cervical medial branch procedures. The authors stated that radiofrequency of the cervical medial branches “carries a high rate of technical failure. ” In 1996, Lord and associates noted that RFA of the cervical medial branches, performed in 24 patients with a median duration of pain of 34 months following motor vehicle accidents, provided a median duration of analgesia persisting 263 days. This contrasted sharply to the median analgesia of 8 days in a control group of patients. However, they excluded individuals with C2-C3 Z joint pain based on their rather dismal results from the previous study noted earlier. In a subsequent study published by the same group of investigators, complete pain relief was found in 71% of 28 patients following a single RF procedure, persisting a mean duration of 219 days. When failures were excluded from consideration, the mean duration extended out to 422 days (60 weeks). Again, excluded from study were individuals suffering from TON-mediated C2-C3 pain. In 2003, Govind and associates used three large-gauge electrodes to perform RFA procedures of the TON. Following controlled diagnostic blocks, 51 nerves in 40 patients were treated with RFA. Eighty-eight percent (43 of 49) achieved successful outcomes using this approach, with a median duration of 297 days (42 weeks). Side effects included slight ataxia, numbness, and temporary dysesthesias. Their improved success was attributed to use of three needles, as well as use of Ray electrodes instead of SMK electrodes (smaller), as well as the assurance that each of the three lesions was made at a distance no greater than one electrode width from an adjacent lesion.
Cohen found that the only factor predicting the success of RFA, defined as at least 50% pain reduction lasting at least 6 months, was paraspinal tenderness. Although the authors stated that C2-C3 pain was included in this analysis, they never indicated how many individuals were thusly treated or what the success rate was for TON RFA.
Cervical Facet Block; Medial Branch Block ( Figs. 52.6 Through 52.12 )
The cervical facet joints and medial branches are commonly implicated in the etiology of cervicogenic headache. C2 and C3 blocks are undertaken to effectively disrupt neural pathways giving rise to greater and lesser occipital nerve dysfunction. Blockade of the joints via the medial branches is typically undertaken for such headaches as well as for the treatment of nonspecific neck pain, degenerative arthropathies of the cervical spine, degenerative disk disease, cervical sprain, and trauma-related pain. Pain may be localized to the neck or may radiate in a capelike fashion from the neck over the shoulders, with extension to the suboccipital area and the supraclavicular area (see Fig. 52.5 ). The cervical facet joints are present from C2 toC3 caudally, as the atlanto-occipital (A-O) and atlanto-axial (A-A) joints are not true zygapophyseal joints, as described previously. From C2 to C3 caudally, the joints are lined by synovium and possess a true joint capsule, which is generously innervated and which may be a source of neck pain and headache pain in certain individuals. The joints themselves may be injured in acceleration-deceleration types of injuries or may be affected by chronic degenerative arthritic changes. Pain results from synovial joint inflammation caused by irritation from repetitive motion at an injured segment. The cervical facet joints, like the thoracic and lumbar facet joints, receive innervation from two adjacent spinal levels. The dorsal ramus from the level above the joint as well as from the level at the joint provides fibers that invest the joint. This forms the foundation for blocking both the affected level and superior levels to completely denervate the facet joint. Each dorsal ramus innervates two facet joints, and each facet joint receives its innervation from two separate nerves. The medial branch, which wraps around the waist of the articular pillars of the vertebral bodies (see Figs. 52.6 through 52.8 ), is the site for peripheral nerve block as well as for neuromodulation techniques of radiofrequency at the joint. These nerves are held against the bone by a fascial envelope and are anchored there by tendons of the semispinalis capitis. The medial branch is consistently found at this anatomic location between C4 and C7, making needle placement for these levels a relatively simple task to accomplish. For example, at the C4-C5 facet joint, the medial branches of C4 and C5 are blocked at their respective articular pillars to anesthetize the joint. At C2-C3, the joint receives innervation from the third occipital nerve (one of two median branches of the C3 dorsal ramus, as discussed previously), and also somewhat by the C2 dorsal ramus. The medial branch of the C2 dorsal ramus is the greater occipital nerve (discussed later). C2-C3 facet joint innervation is a complex anatomic scheme, and for practical purposes, our technique of blocking the third occipital nerve at the C3 articular process waist is sufficient for denervating or anesthetizing this joint.
Technique of Blocking the Medial Branches for Cervical Facet Joint Pain (See Figs. 52.9 Through 52.11 )
Our technique of blocking the medial branches requires the use of fluoroscopic guidance, although there are increasingly more advocates of performing an ultrasound-guided technique. We believe this is essential to minimize the likelihood of advancing our needles too far ventrally with the patient in the prone position. Advancing too far ventrally could cause injury or irritation to the exiting cervical spinal nerve roots or to the more ventrally situated vertebral artery. Also, if strict adherence to maintaining needle-to-bone contact is not maintained, it could be possible for advancing needles to stray too far medially toward the interlaminar or transforaminal spaces, or too far laterally with a resultant failure of accessing the medial branches, which are secured to the waists of the articular pillars by a fascial sheath and the semispinalis capitis tendons. Appropriately screened candidates (i.e., no coagulation problems, no infection at the intended insertion site, etc.) are placed in the prone position, with a bolster placed under the chest to permit flexion of the head on the neck. The forehead is supported using a ring neurosurgical donut or pillow, with unrestricted breathing and with minimal to no sedation utilized. Vital signs are monitored using standard ASA monitors, and an intravenous cannula is placed in a distal extremity for purposes of administering resuscitative medications or adjuvants, as indicated. A careful sterile skin prep and drape is performed using a generic antiseptic solution. The fluoroscope is oriented anterior to posterior with a moderate-steep cephalad orientation. This is an essential step that if not undertaken inhibits or impairs the ability to visualize the periarticular osseous structures because of the obstruction imposed by the thick mandible. Scout films are obtained using sterile needles placed over the skin of the intended needle entry sites, and local anesthetic skin wheals are raised using a 27-gauge, short (1.5-inch) needle with a short acting local anesthetic agent (i.e., lidocaine 1% without epinephrine), in doses of 2 to 3 mL per site. Next, a 22-gauge, 3.5-inch blunt (i.e., Whitacre type or equivalent) subarachnoid or block needle is advanced completely perpendicular to the skin toward the lateral-most edge of the vertebral body on the side selected. In this position, it may not be practical to gauge the exact site (i.e., C3 versus C4) where the needle is being directed. When 2 cm of needle have been introduced, it is essential to switch to a straight lateral view to observe the depth of needle advancement, as well as to ascertain which anatomic segment is being addressed. This is usually of little concern, as two levels (the level at the joint and the level cephalad to the joint) are being blocked to anesthetize a single level. So even if C4 was chosen, and the needle is seen as advancing toward the trapezoid body articular pillar at C3, the second needle may merely need to be placed caudad to the first to block both medial branches. Once the needle is seated at the appropriate depth, as seen on lateral fluoroscopy (see Fig. 52.9 ) and is seen not to be encroaching on either the visible neural foramen, or too far ventral, toward the anatomic site of the vertebral artery, the needle is checked to ascertain that the tip is touching bone. Using a curve tipped needle helps to realign the needle in cases where the depth appears appropriate, but the needle tip is laterally directed away from bone. In such cases, a mere twisting of the needle hub will usually suffice to turn and steer the tip toward, and subsequently against, the bone where the medial branches are situated in their respective fascial planes and are anchored by the semispinalis capitis tendons. When all needles are in the appropriate anatomic position for a given individual, they should be aligned like a picket fence. The lateral fluoroscopic image should demonstrate almost perfect parallel lines derived from the needles. An alternative technique is to place the patient prone and advance the needle from lateral to medial toward the bone under fluoroscopic guidance. This technique is clearly fraught with greater danger of striking the vertebral artery or an exiting cervical spinal nerve root than the technique described previously as a result of the direction of the needle toward the central neuraxis. Additionally, and potentially more than a mere nuisance, the clinician who subsequently determines that the patient might derive benefit from a radiofrequency or a neurodestructive procedure of the cervical medial branches is then faced with the prospect of performing the procedure with the active tip of any RF needle making minimal and insignificant contact with the target area/nerve of interest. Contrast this with the approach described earlier, where the RF needle’s active tip will be hugging the bone through its entire length, making the likelihood of success inherently much greater. Additionally, there is less muscle in the pathway of the advancing needle using the lateral approach. This increases the likelihood that needles with not be firmly seated against the target and will be easily displaced once the second and possibly third or fourth needles are inserted at a given level.
Whichever technique is chosen, once the needles are situated, and once there has been no paresthesia or blood from the needle during aspiration, a small (typically 0.5 mL to 1 mL) volume of radiocontrast media may be injected merely to verify that no arterial cannulation has occurred. The incidence of intravascular injection following this procedure in the cervical spine is about 3.9%. If this injection proves negative for such a cannulation, the same volume of a short-acting (i.e., lidocaine, mepivacaine) local anesthetic with or without corticosteroid may be added. We prefer to use the nonparticulate steroid dexamethasone acetate for this purpose, in doses of 2 mg per medial branch (up to 12 mg total), because the particulate agents such as methylprednisolone acetate tend to flocculate or clump, with the potential for neurologic compromise if such a phenomenon occurs in an end artery. When assessing response to local anesthetic (LA) blocks, it may be prudent to query patients on more than one occasion to minimize a perception bias associated with single time frame, snapshot data capture.
If RF procedures are anticipated to ameliorate recalcitrant symptoms, we typically perform double-diagnostic blocks as a prelude to RF, accepting an arbitrary 60% or better response prior to doing the actual RF. In the future, pulsed RF techniques may replace conventional RFA lesioning in the neck for the most part. These are performed by applying 42° to 45° C temperatures for 120 seconds to each nerve in question. Success rates of RFA in carefully selected patients may be found in up to 74% of patients for up to 20 months. If performed following cervical spine surgery in patients with ongoing neck pain, success may be found in 59.4% for up to 15 months. However, when data from eight cervical RFA studies were summarized in a systematic review, the average duration of analgesia was found to be more modest, at about 7 to 9 months. Even pulsed RF techniques of the cervical dorsal root ganglia may provide significant analgesia in patients with radicular pain, lasting up to 12 months. Using ultrasound guidance in the performance of the procedure has been associated with similar rates of success for RFA of the medial branches as that identified for fluoroscopic techniques.
Occipital Nerve Block ( Figs. 52.13 and 52.14 )
The greater occipital nerve (GON) arises from the dorsal primary rami of C2, with occasional contributions from C3. The nerve penetrates the fascia inferior to the superior nuchal crest, where it runs alongside the occipital artery for a variable distance. The sensory area innervated by the nerve includes the medial portion of the posterior scalp, with radiation ventrally up to the vertex. The lesser occipital nerve (LON) arises from the ventral primary rami of C2 and C3 and passes superiorly and laterally from the occiput to the lateral edge of the sternocleidomastoid muscle. Here the nerve divides into cutaneous branches that innervate the lateral portion of the posterior scalp and the cephalad surface of the ear pinna. Along with the great auricular nerve, the GON and LON provide the majority of sensory afferent information for the occipital area and are responsible for transmitting information derived from C2-C3 facet joint derangements to the referral area described.
Occipital nerve block has have become an increasingly popular method of managing headaches of diverse etiologies, even though the scientific support for doing so is limited. GON blocks have been shown to be ineffective for chronic tension type headache when prilocaine and dexamethasone were used in a group of 15 patients. They have been used with various degrees of success for postconcussive headaches (80% successful in then patients), atypical orofacial pain, and especially for migraine headaches, when they are often combined with trigger point injections or supraorbital nerve blocks, in both cases affording a very high degree of success for ameliorating pain and brush allodynia. A novel use of GON block has been in the successful treatment of patients with abnormal head movements who also suffered from tinnitus and dizziness associated with a previous history of trauma. Although oftentimes the diagnosis of headache due to occipital neuralgia is not too difficult to make, there are cases of refractory headache that require advanced evaluation techniques. Some suggest performing computed tomography (CT)–guided C2-C3 nerve blocks as a prelude to considering patients who may be candidates for percutaneous rhizotomy ; in our experience, fluoroscopically guided techniques of C2-C3 nerve block have been extremely useful and have precluded the requirement to seek more advanced imaging for guidance (see Figs. 52.8 and 52.9 ).
Technique of Blocking the Greater and Lesser Occipital Nerves
Occipital nerve blocks have been erroneously described by some to be merely a field block of local anesthetic and steroid layered somewhere in the posterior part of the occiput, without regard to anatomic landmarks or consequences of errantly placed medications. Indeed, there are cases of sudden unconsciousness reported following LON block, some resulting from occipital artery injection and at least one the result of unintentionally injecting local anesthetic into a previous bone defect from a craniotomy. Also, it is possible that unconsciousness might occur after GON block has been made too far inferiorly toward the foramen magnum. Additionally, if practitioners are not careful and discriminating in their choices of frequency of GON block, and use judicious doses of agents including corticosteroids, additional complications like the development of Cushing syndrome may result. Our technique of blockade of the occipital nerves is to perform the procedure with the patient in the sitting position, with the forehead forward, resting on either a Mayo stand, the edge of a padded table, or on a gurney. If possible, the occipital artery is palpated at the superior nuchal ridge. Using ultrasound guidance may help to identify the artery, which is often not robust or bounding and, subsequently, not always readily discernible as a distinct structure. As it is virtually impossible to completely disinfect the scalp injection site, we utilize alcohol wipes or a Betadine-soaked pledget to soak the area with disinfectant prior to inserting the needle, without expectations of complete asepsis. Next, we insert a fine-gauge (25-gauge, 1.5-inch) cutting needle immediately medial to the artery and advance it perpendicular to the skin until the needle tip touches periosteum. Once the occipital bone has been thusly contacted, the needle is retracted about 1 mm and is redirected slightly cephalad. After gently aspirating, 5 mL of local anesthetic (0.5% bupivacaine or ropivacaine) with 4 mg of dexamethasone or 6 mg of betamethasone is injected in a fanlike manner, taking care not to direct the needle too far medially toward the foramen magnum. The lesser occipital nerve may be blocked, as can the greater auricular nerve, by removing the needle and placing it into the skin about 3 to 4 cm lateral to the entry point for GON block, while also directing it inferiorly instead of superiorly as for GON block. After gentle aspiration, 5 mL of the same combination of local anesthetic with or without corticosteroid as noted previously may be injected, again in a fanlike distribution, to block the nerve. After completing GON and LON blocks, it is important to gently massage the tissues of the scalp to assist in the spread of the agents while maintaining pressure over the injection site(s). This will minimize ecchymosis or hematoma formation from this highly vascular area. Often, an ice pack is applied for 20 to 30 minutes post-procedure to further minimize swelling and inhibit vascular absorption of the local anesthetic agents, particularly if bilaterally blocks have been performed.
Related posts:
Full access? Get Clinical Tree
