Fusion Surgery for Axial Neck Pain

Published on 23/05/2015 by admin

Filed under Physical Medicine and Rehabilitation

Last modified 22/04/2025

Print this page

This article have been viewed 1825 times

CHAPTER 68 Fusion Surgery for Axial Neck Pain

Timothy A. Garvey

INTRODUCTION

For those suffering from cervical radiculopathy or myelopathy, there is little debate that an anterior cervical discectomy and fusion (ACDF) is a viable treatment option.¹^–²⁰ However, patients who present with their chief complaint being ‘neck pain,’ with or without referred or radicular symptoms, often are told that there is nothing that can be done surgically. This is wrong! The purpose of this chapter is to enhance the reader’s understanding of the surgical peer-reviewed literature, which leads to a rational decision-making process (Fig. 68.1). For many patients, that decision will be that surgical intervention is a reasonable option with a predictable statistical chance of success. The time-honored adage ‘that it is the decision, not the incision, which is most important,’ cannot be overemphasized.

Fig. 68.1 Algorithmic approach.

INDICATIONS – PATIENT SELECTION

While covered in previous chapters, a brief review of the epidemiology and natural history of neck pain is in order. Prevalence studies have noted 35% of the general Norwegian population to have neck pain complaints within the preceding year, with 14% of respondents reporting those symptoms to have lasted longer than 6 months.²¹ In the Saskatchewan Health and Back Pain Survey, reported in 2000, it was documented that 54% of respondents had experienced neck pain in the preceding 6 months, with almost 5% perceiving themselves to be ‘highly disabled’ by their pain.²² When one looks at the literature of whiplash-associated disorders, it is apparent that while the majority of injured individuals do have spontaneous resolution of their symptoms, a small percentage, yet a significant number of human beings, do develop chronic neck pain.^23,³⁴

As with all medical ailments, it is important to appreciate the natural history of a condition. One approaches the patient with a ‘common cold’ with a vastly different sense of urgency than in comparison to an individual with suspected acute meningitis. We must then realize that in studies of natural history many with neck pain never seek out evaluation. Therefore, there is an inherent selection bias to more self-perceived impairment in those who seek out medical care. In an average 15.5-year follow-up study, Gore et al. reported on those presenting with neck pain.³⁵ Seventy-nine percent of patients noted improvement with nonoperative care, with 43% reporting pain-free status and 32% continuing to report modest to severe pain. The severity of initial presenting symptoms, and the report of a significant injury, were more indicative of those with long-term complaints.

In a series of patients with neck-only, or neck and arm pain, DePalma et al. noted 45% of those treated nonoperatively to have satisfactory long-term outcome.^36,³⁷ In a surgical series report, DePalma et al. reported that at 3-month follow-up those presenting with dominant neck pain had 21% complete relief and 22% no relief with nonoperative care.³⁶ At a 5-year follow-up time, Rothman noted 23% of patients continued to be partially or totally disabled due to significant cervical symptoms stemming from disc degeneration.^36,³⁸ In a review of those presenting with cervical radiculopathy, it was noted that while patients do not typically progress to myelopathy with nonoperative care, they often (two-thirds of the group) will have persistent symptomatology of some degree.^39,⁴⁰

The whiplash literature reveals a common pattern of a small group reporting persistent intrusive symptoms.^23,³⁴ In a textbook chapter, McNab and McCulloch summarized ‘about those who had suffered a whiplash injury, approximately 10–20% are left with discomfort of sufficient severity to interfere with their ability to do work and to enjoy themselves in leisure hours.’⁴¹ Thus, it must be acknowledged that while many who seek out medical care for neck pain will improve with nonoperative management, a finite low percentage of patients will still have symptoms of sufficient magnitude to cause them to seek out a surgical solution. It is for this group of patients that we need to have a rational surgical decision-making process.

OFFICE EVALUATION

History

When a patient presents for initial evaluation, we ask for their chief complaint, as this will focus our line of questioning and examination. In this chapter, we are concerned with those who complain of ‘axial neck pain.’ However, the typical patient does not say, ‘I have dominant axial pain;’ they state rather that they have neck pain, with or without arm pain. Patients who yield an ultimate diagnosis of myelopathy or radiculopathy will often present with a complaint of neck pain.^8,^40,^42,⁴³ It is up to the medical evaluation to elicit the historical features that will lead to a specific diagnosis.

Several critical decisions must be made at the initial evaluation. Is this mechanical or non-mechanical pain? Mechanical pain is typically worse with certain activities, i.e. flexion usually increases discogenic pain, while extension increases the pain from facet arthrosis, and the patient can usually find a position of comfort. Non-mechanical pain suggests tumor, infection, or other etiologies which require a different investigation. This often is the patient with rest pain, night pain, no position of comfort, and they may present with constitutional symptoms of weight loss, fever, chills, malaise, etc. One needs to answer, is this pain local, referred, or radicular? Local pain is just that, focal, localized neck pain. Referred pain, that is pain felt away from the tissue of site of origin, is often inter- or periscapular, upper trapezial, or basioccipital. Radicular pain is the classic radiating pain along the course of the nerve root. This typically is into the brachium with radiation distally such as to the dorsal radial forearm, thumb, and index finger to suggest a C6-type distribution. It is often forgotten, however, that while this classic distribution is along the course of the ventral rami, there also does exist the dorsal rami along which path patients can also perceive pain. This ‘dorsal radicular’ pain can/does often get lumped together with axial symptomatology.

One needs to assess whether the patient is intact neurologically or has deficits. From the historical perspective, we ask about paresthesias, numbness, weakness, ataxia, and bowel and bladder dysfunction. For those with significant myelopathy or progressive loss of neurologic function, an urgent surgical consultation is recommended. We finally need to answer whether there are significant psychosocial contraindications to elective surgical treatment for pain management. This is based on depression, frank psychosis, secondary gain issues, or unrealistic expectation, all of which need to be considered.

Thus, this chapter is focusing on the individual who has dominant axial mechanical neck pain that is primarily local or referred, who is intact neurologically, and who has no obvious psychosocial contraindication to proceed with a surgical work-up. We must realize that patients present with signs and symptoms on a continuum. Some will have 100% arm pain in a C6 distribution while others will present with 100% neck pain with no periscapular or arm pain. Most, however, will have some combination of both. I believe it is vital to have the patient decide if the neck pain is greater than, the same as, or less than the arm pain. I query specifically about a percentage, i.e. 70% neck pain and 30% arm pain. While this is not exact, it does help us understand what the patient wants addressed. Simplistically, we think the more arm pain there is, the more nerve root decompression is required; the greater the neck pain, the more the patient has a joint problem which can be addressed with a fusion of the affected joint (in the near future, potential arthroplasty). I typically ask about fevers, chills, night pain, etc. If tumor or infection is suspected, the patient needs an urgent laboratory evaluation and a magnetic resonance imaging (MRI) diagnostic work-up. Questions about the timing of onset, that being acute and/or traumatic, i.e. whiplash, versus gradual or progressive onset help in the differential diagnosis. Provocative or ameliorative motions can suggest a source, i.e. worsening pain with flexion activities with progressive pain throughout the day that is relieved with rest, tends to suggest a discogenic etiology. Patients who cannot turn to one side and describe unilateral pain when looking over their shoulder (i.e. Spurling’s maneuver) suggest a foraminal radicular etiology.

I question about neurologic function. For myelopathy, I ask about generalized weakness, decreased fine motor dexterity, stumbling or ataxia, sphincter dysfunction such as urinary urgency, and paresthesias which may be nondermatomal. For radiculopathy, often a specific root pattern of numbness, pain, or specific weakness can be elicited. If these symptoms are present, they can help focus the differential to a specific cervical diagnosis. For psychosocial issues, we review the patient intake forms for depression, look for expectations of treatment, and solicit information about worker’s compensation status and litigation as these all may have a bearing on the patient-perceived pain. Specific psychologic treatment may be recommended. It is of interest that our last cervical fusion study did not correlate the presence of worker’s compensation or litigation with poor outcome.⁹

Physical examination

The physical examination should be straightforward. Range of motion is documented. If flexion hurts more than extension, think discogenic; if extension hurts more than flexion, think facet arthrosis or stenosis. If rotation or lateral flexion is preferentially limited to one side, it suggests a unilaterality of pathology. Spurling’s maneuver, i.e. lateral flexion, extension, and rotation which causes provocation of symptoms, suggests foraminal nerve root compression. If this reproduces the patient’s ‘neck pain,’ particularly into the upper trapezial or periscapular region, think nerve root pain. Assess the reflexes: hyperactive suggests upper motor neuron pathology, hypoactive suggests lower motor neuron pathology; and compare right side versus left side. Check for strength and sensation. If the examination documents pathology to suggest myelopathy or dense radiculopathy, the evaluation focuses on the neurologic deficit. An MRI or myelogram computed tomography (CT) scan should then be sought out. Often, those with dominant neck pain will have some component of arm symptomatology, just as those with ‘radiculopathy’ often have substantial complaints of neck pain.^2,^7,^8,^19,^44,⁴⁵

IMAGING

After a history is obtained and a physical examination is performed, one should start with basic radiographs. I prefer anteroposterior (AP), lateral, and flexion extension cervical spine radiographs. I do not routinely obtain an open mouth view or oblique views. One looks to exclude the obvious destruction of tumor and/or infection. We check for dynamic instability by the measurement of listhesis and angular deformity.^43,⁴⁶ We note obvious spondylosis by assessing disc space height, sclerosis, and osteophyte formation.⁴⁷^–⁴⁹ While it is recognized that with increasing age there is an increasing presence of radiographic spondylosis, the films are helpful. If a young patient has multilevel spondylosis, we think less of a surgical approach; if a single level is involved, we think more of a surgical solution. If gross instability is seen, the patient needs to be counseled about the risks of nontreatment.

The MRI is now my next recommendation. Exquisite detail about tumors, infection, neurologic compression, and disc morphology is obtained. However, for this group of patients with axial neck pain, the history and physical examination did not suggest tumor or infection. The disc spaces are assessed for morphology, annular tears, and frank herniations. If the patient has a contraindication to MRI, i.e. pacemaker, metal in the eye, etc., a myelogram CT scan would be recommended and should yield similar diagnostic value.

ADVANCED INVASIVE DIAGNOSTICS

Based upon the evaluation, when one has a patient with chief complaint of neck pain, who is grossly normal neurologically, who has failed a prolonged course of nonoperative management, who has no psychosocial contraindication, and who believes the symptoms are severe enough to contemplate surgical intervention, advanced invasive diagnostic testing can be critical in objectifying the pathologic source of the patient’s symptoms. Surgeons cannot ‘cut out pain.’ They can decompress nerves or spinal cords, stabilize unstable joints, and fuse (replace) bad joints. It is the objectification process which allows us to educate a patient about the statistical chance of success for a given procedure.

In some patients, the history, physical examination, radiographs, and MRI will suffice. For example, a patient may present with a chief complaint of neck pain, which came on acutely, with initially severe arm pain, that now has settled into chronic axial neck pain, 80%, with 20% arm pain into the thumb and index finger. Flexion activities typically increase the pain, and the patient can find positions of comfort at rest. The patient has mild weakness of wrist extension with mild decrease of the brachioradialis reflex. The MRI shows mild to moderate degeneration of C5–C6 with a central to right-sided herniated nucleus pulposus, and all other segments are pristine appearing. In this case, if nonoperative treatment has failed (which may have included a C6 selective nerve root block), I would not typically do more testing. A single-level ACDF may be recommended. However, this clean a scenario is uncommon.

Most often, the patient with axial pain will have several levels of degenerative change on an MRI study or noted on the X-rays. In those whose history and examination suggest a discogenic source, a discogram can be a very valuable tool. While the selection of this ‘older technology’ of discography still emotes controversy in some, I find it to be particularly useful in the surgical evaluation of those with axial pain. Data from our center has shown a statistical association with patient-perceived outcome.⁹ Numerous other authors have cited the utility of discography in patient selection.^9,^17,^20,^45,⁵⁰^–⁶⁰

Most often, the MRI alone will not suffice. There are a substantial number of asymptomatic individuals who have MRI-documented pathology, which increases with age.⁶⁰^–⁶³ In a prospective correlation of MRI and discography, in asymptomatic subjects and pain sufferers, Schellhas et al. concluded that MRI often misses annular tears and cannot reliably identify the source of discogenic pain.⁶² While many asymptomatic individuals had degenerative changes, only 3/40 discs studied in this group elicited a pain response, suggesting a high specificity and positive predictive value.⁶² Pragmatically speaking, the issue is, does the discogram yield information that reliably predicts outcome, and do published reports document this? The answer is yes, Table 68.1 documents such. What is hard to report upon is when the discogram keeps the patient out of surgery. For example, when every level hurts and there is no control, surgical intervention will not typically yield good results. The clean discographic surgical patient will have significant (greater than 6/10) concordant reproduction of pain at the affected level/levels, with little or no pain and normal appearance at the control level.⁹

Table 68.1 Outcomes of surgical treatment

Author	Number of patients	Reported outcome
Zheng⁶⁰	55	76% good or excellent, 18% fair, 6% poor
Garvey⁹	87	82% good or excellent, 16% fair, 2% poor
Ratliff⁵⁴	20	85% satisfaction
Motimaya⁵²	14	78.6% satisfaction
Palit⁵³	38	79% satisfactory, 21% not satisfactory
Whitecloud¹⁹	34	70% good or excellent, 12% fair, 18% poor
Roth⁵⁶	71	93% good or excellent, 1% fair, 6% poor
White⁴⁶	28	62% good or excellent, 23% fair, 23% poor
Riley⁴⁵	93	72% good or excellent, 18% fair, 10% poor
Simmons⁵⁷	30 neck pain	78% good or excellent, 15% fair, 7% poor
Simmons⁵⁷	51 neck and arm	78% good or excellent, 15% fair, 7% poor
William²⁰	15	7% excellent, 20% good, 33% fair, 40% poor
Dohn⁵⁰	34	62% good or excellent, 24% fair, 15% poor
Robinson¹⁷	56	73% good or excellent, 22% fair, 5% poor

A selective nerve block can be useful in determining a surgical level. Such a patient may have neck, upper trapezial, and periscapular pain that increases with Spurling’s maneuver, has foraminal stenosis on MRI, and appears to have symptoms that are more often unilateral. In this patient, I often proceed to a selective nerve block. The selective nerve root block can be both diagnostic and potentially therapeutic. If in the first 30–60 minutes, during the lidocaine anesthetic phase, the patient reports significant diminution, i.e. 70–100% relief of typical pain, that specific nerve root is indicated as a primary pain generator. If steroids give long-term relief, all will be happy.⁶⁴ If the patient has great relief with the selective nerve root block, unilateral pain, foraminal stenosis, and no instability, then a posterior laminoforaminotomy may be a favored option. If there is bilateral pain, and an additional ‘discogenic component,’ then one may proceed with an ACDF at the affected level.

True facet blocks or medial rami branch blocks are not uncommon in the work-up of patients with neck pain, particularly post-traumatic pain.^24,⁶⁵^–⁶⁸ To date, I know of no study to base fusion surgery on this diagnostic tool. Conceptually, one would think, if the patient had excellent temporary relief of the pain with this injection at a specific joint, that fusion of that joint may relieve the pain. If one were to use this intuitive reasoning to base the surgical decision, I would recommend that it be done in a study fashion with attendant IRB approval.

SURGICAL TREATMENT FOR SPECIFIC ANATOMIC DIAGNOSES

When the ultimate clinical diagnosis for a patient with chronic axial mechanical pain is that of discogenic pain, an anterior cervical discectomy and fusion is a very reasonable treatment option (Fig. 68.1). The patient will have symptoms of dominant neck pain with or without referred symptoms to the basioccipital region, upper trapezial region, or interscapular region. The patient may also have a component of arm pain with mild neurologic signs of dysfunction. The X-rays will often show spondylosis with no gross instability. The MRI and discography will pinpoint specific joints as the ‘painful levels.’ The more clean the diagnosis, i.e. one or two levels of concordant reproduction of pain, which matches the MRI appearance of degeneration, with a normal nonpainful adjacent segment, the higher is the likelihood that the patient will favorably respond to an anterior cervical decompression and fusion. Having more levels involved does not preclude the surgical option, but it does raise the risk of the patient reporting nonsuccessful outcome and pseudoarthrosis. Therefore, the patient needs to clearly understand this preoperatively. On the other hand, in an older population with advanced degeneration, these joints have less motion, and my experience has been favorable.⁹ The physiologically younger patient with expectation of high physical demand will be less satisfied with attempts at multilevel treatment.

To answer a patient’s question about what they may expect from surgical management, the reader needs to assess their level of satisfaction with the current surgical literature support for that treatment. A clever paper by Smith and Pell, titled ‘Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials,’ points out that advocates of evidence-based medicine ought not to neglect the information we obtain from observational studies.⁶⁹ When we see similar reports of good to excellent outcomes spanning six decades, from various centers, utilizing similar diagnostic evaluations and surgical interventions, we can place greater confidence in the predictability of those data (Table 68.1).^9,^17,^20,^45,^50–58,⁶⁰ The pioneers of ACDF surgery, Robinson and Smith, reported on 56 patients in their 1962 paper.^16,¹⁷ Forty-five of 56 had neck pain, 38 of 56 had traumatic etiology, 25 of 56 had occipital pain, 26 of 58 had interscapular pain, and discography was done in 47 of 56 patients. They reported 73% good to excellent results, 22% fair, and 5.5% poor outcome, with 9 of 56 patients having pseudoarthrosis.

Fig. 68.2 This case represents a common presenting scenario. The patient is a 39-year-old female first seen in November 2000 with a 9-month history of 90% neck pain and headaches and 10% arm pain. She was seen again in early 2003 with continued symptoms which are progressively worse. (A, B) These show flexion extension radiographs which reveal moderate cervical spondylosis at C5–C6 with no obvious instability. (C) This shows a sagittal MRI with small posterior protrusion at C5–C6 with disc desiccation. There appears to be small annular thickening posteriorly at C3–C4. There is no gross cord compression. (D) Transverse image at the same level shows no gross cord compression. (E) The patient underwent discography with 10/10 concordant reproduction of pain at C5–C6, with small pressure sensations at other levels. (F, G) These show flexion extension radiographs 1 year postoperatively with solid arthrodesis at C5–C6 and no obvious degenerative change or stability at the adjacent segments. Her pain was substantially improved. She was working full-time and taking no pain medication.

Highlighting the data from a recent series from our center can give an appreciation of what to expect from this type of surgical approach.⁹ Out of a prospective series of 112 patients with axial mechanical neck pain, 87 were available at average 4.4-year follow-up which included an extensive self-reported outcome data questionnaire. This series documented that 82% (71/87) self-rated their outcome to be good, very good, or excellent. Pain improvement was evidenced by an average diminution of the VAS from 8.4 preoperative to 3.8 at follow-up, and was seen in 93% of individuals. The average 3.8 follow-up score includes those with recurrent pain, new injury, and who did not respond, thus reflecting a higher average VAS. Self-rated function improved 50% as documented on both a cervical-modified Oswestry and Roland and Morris disability index. We specifically looked at our use of discography in the selection process (Table 68.2). If a clean anatomic pattern was present, that is greater than or equal to 6/10 concordant reproduction of pain, with little or no pain at an anatomically near-normal control level, and all of the affected levels were treated, 91% (32/35) reported good to excellent outcome. In those with ‘nonclean anatomic pattern,’ i.e. most commonly greater than or equal to 6/10 nonconcordant pain at what will become a proposed adjacent segment with mild to moderate morphologic abnormality, only 68% (21/31), reported good to excellent outcome. In the group where we did not utilize discography, i.e. those with single- or two-level pathology on MRI with a lesser component of clinically correlating radicular pain, 86% (18/21) self-reported good to excellent outcome. The take-home message is that the better one can objectify the pathology, the more likely is it that surgery can remedy the patient’s symptomatic complaints. The reader can review similar reports of surgical outcomes that span six decades from multiple centers (Table 68.1).^9,^17,^20,^45,^50–58,⁶⁰ These authors report remarkably similar outcomes and what appears to be a similar diagnostic grouping. This gives greater confidence that these studies do cumulatively support this type of surgical approach for discogenic neck pain.