Cervical vertigo in severe cervical spondylosis: frequent or over- diagnosed?


 Background
Cervical vertigo (CV) is a grossly over-diagnosed entity although, in all probability, it does not exist. The aim of this study was to test following hypothesis: that even in patients with one of the most severe forms of cervical spondylosis, degenerative cervical myelopathy (DCM), stimulation of the cervical proprioceptors does not provoke or increase vertigo
Methods
The study was performed in a cohort of 38 patients with DCM confirmed by clinical manifestation and MRI-detected degenerative cervical cord compression. The incidental presence of vertigo in these patients was investigated by means of a questionnaire and a clinical neurological examination. The cervical torsion test (to stimulate cervical proprioceptors), and ultrasound examinations of the extracerebral carotid and vertebral arteries (to exclude impaired blood circulation in the cervical region) were performed. All patients with vertigo underwent a diagnostic work-up designed to reveal its cause.

Results: Subjective symptoms of vertigo occurring in the six months previous to examination were reported by 18 patients (47%). None of these patients responded positively to the cervical torsion test, while highly probable explanations involving an etiology for vertigo other than from the cervical region were found in all patients suffering from it. No patient exhibited significant stenosis of the vertebral arteries.
Conclusions: There were no clear signs of CV in the cohort of patients with DCM. If patients with cervical spondylosis suffer from vertigo, its cause may, more easily be explained by common and treatable etiologies that do not lie in the cervical region.

3 physician. The lifetime prevalence in adults is around 20%, reaching 40% in the elderly (1,2). A report reviewing United States emergency admissions indicated that vertigo and dizziness accounted for 2.5% of all presentations (3). Many studies suggest the existence of what has become known as "cervical (or cervicogenic) vertigo" (CV), but physicians lack sufficient data to form definite opinions and to give clinical guidelines for its diagnosis and treatment. The overall prevalence of CV is not known, but it is a term very frequently used in clinical practice. Colledge et al. reported 65% of dizziness cases attributable to cervical spondylosis (4), and Takahashi estimated a prevalence of CV as high as 90% in out-patients suffering from vertigo and presenting at a neurosurgical department (5).
This study adopts a new approach to the problem, based on testing the movement of the neck while keeping the head stationary, in a group of patients suffering from degenerative cervical myelopathy (DCM), the most severe symptomatic form of spondylosis. The aim of the study is to test the hypothesis that vertigo is not provoked or exacerbated by stimulation of the cervical proprioceptors and that the etiology of coexisting vertigo, even in these patients, may be explained by mechanisms other than lesion(s) of the nervous system in the cervical region.

Materials And Methods
The study sample consisted of a cohort of consecutive subjects referred to the All subjects in the study complied with the following inclusion criteria: 4 MR signs of spondylogenic or discogenic compression of the cervical spinal cord with or without concomitant change in signal intensity from the cervical cord on T2/T1 images (see ''Imaging'' below) Presence of at least one of the clinical signs and one of the symptoms that could be attributed to cervical cord involvement.

Exclusion criteria:
Previous surgery on the cervical spine (possibly limiting rotation of the spine) Non-spondylogenic compression or causes of cervical myelopathy other than spondylogenic The entire cohort was a completely new sample. All subjects gave their written, informed consent to participate in the study. The following clinical and demographic data were also noted: Age Sex Degree of disability as assessed by mJOA score, the generally-accepted disability scale for DCM patients (6) Vertigo questionnaire A questionnaire adapted from Filippopulos et al. (7) was given to all patients. The prevalence of vertigo was assessed by the question "Have you suffered from dizziness or vertigo during the last six months?" which could be answered only by "yes" or "no". If "yes", the respondents were asked to specify further the type of vertigo: "spinning vertigo, as if on a carrousel" (spinning vertigo), "swaying vertigo. as if on a small boat" (swaying vertigo), "feeling of impending black-out when standing up rapidly" (orthostatic dizziness), or "none of these three types" (unspecified dizziness). The body positions and movements related to the different vertigo types were assessed by the following questions, which could be answered by only "yes" or "no": Is the vertigo a) triggered or aggravated by head movements? b) triggered by a change of position (e.g. standing up from lying down? c) also present when sitting or lying down? d) only present when standing or walking?

Cervical torsion test
A cervical torsion test was performed in all patients. The procedure was adapted after the work of L´Hereux-Lebeau (8). Subjects were seated in a rigid but fully rotatable chair that provided support to the entire body. Their legs were flexed with a slight bend at the knees. They were securely held in the chair with shoulder-and lap-belts. It was requested that their eyes should be kept open during the procedure. The examination took place in 6 natural light (there was a large window in the room and the lights were turned off). First, the subject´s trunk was passively turned 70 degrees to the right, with the head still, then returned to centre, followed by turning the trunk 70 degrees to the left, and returning to centre. Each position was held for 30 seconds with the head stabilized by the observer for all positions. Nystagmus was evaluated with a video-Frenzel apparatus. The test was

Results
Subjective feelings of vertigo in the previous six months were reported by 18 patients (47%). Detailed neurological examination of these patients disclosed the following causes of vertigo: orthostatic dizziness in 8 patients (44% of patients with vertigo, 22% of all patients), uncompensated hypertension in 5 (28% and 14% respectively), benign paroxysmal vertigo in 4 (22% and 11%, respectively) and psychogenic dizziness in 1 (6% and 3%, respectively). These results are summarized in Tables 1, 2, and 3.
None of 38 patients studied provided positive results on the cervical torsion test and none had subjective feelings of vertigo during this test, whether or not they had subjectively described vertigo in previous six months or not.
Three patients (0.8%) exhibited haemodynamically significant stenosis of the carotid arteries (two of them suffered from recently-diagnosed, uncompensated hypertension, while one had orthostatic dizziness). None of the patients studied had significant stenosis of the vertebral arteries.

Discussion 8
In this cohort study of a group of DCM patients, it was impossible to confirm the high prevalence of CV previously reported elsewhere and attributed either to advanced symptomatic spondylosis of the cervical spine and/or stenotic changes of the extracerebral carotid or vertebral arteries (4,5).
In recent decades, cervical vertigo has emerged as a special category of dizziness, generating considerable controversy. The diagnosis remains debatable; there remains a lack of validated tests to confirm this entity, and exclusion clinical diagnosis appears to be the default standard (8). A diagnosis of CV, however, is made too often by many physicians, largely because the simultaneous occurrence of vertigo and cervical spondylosis is very common. The overall incidence of so-called CV remains unknown.
Despite the fact that some authors suggest a prevalence of up to 90% (in out-patients suffering from vertigo and presenting at a neurosurgical department) (5), other authors doubt the diagnosis entirely (11).
Several explanations of the etiology of cervical vertigo have been published. Disturbed cervical proprioception is suggested by what is probably the most-cited study (12).
Vertigo, ataxia, and nystagmus have been induced in animals by injecting local anesthetics into the neck (13). This presumably interrupted the flow of afferent information from neck-muscle and joint receptors. Ataxia in human beings has been associated with a broad-based, staggering gait and hypotonia of the ipsilateral arm and leg (13). Posturography, however, proved normal and nystagmus was absent (13). In patients with cervicocranial syndrome, unilateral blockade with a local anesthetic at the level of the C2 vertebral body has resulted in an ipsilateral gait deviation without ocular changes, also without ataxia or hypermetria (14). These findings accord with previous data derived from animal studies (14). Some authors have suggested that spinal cord compression is the most frequent cause of cervical vertigo (15,16). This may well be true if the posterior column pathways are compressed. However, the objection may be raised that such vertigo could originate in any part of the sensory pathway from the receptors all the way to the cortex (e.g. in polyneuropathy, radiculopathy, myelopathy and encephalopathy). If cervical vertigo is the subject under discussion, the cause should, ipso facto, be located in the neck and should be provoked and treated in the neck.  (19). These reflexes are associated with the vestibular nuclei and thus with ocular reactions (nystagmus). Quantitative data on COR demonstrate that this is strongly suppressed in healthy people, and the vestibulo-ocular reflex (VOR) is the main gazestabilizing system during rapid head movements (20). Suppression, however, ceases in complete bilateral vestibular deficiency, in severe brain injuries or in very special circumstances, such as fixed position of the head in a dark room, etc (21). COR gain rises with increasing age and there is a significant covariation between gains in VOR and COR, meaning that when VOR increases, COR decreases and vice versa (22). However, it can hardly be expected that, in routine clinical practice, the origin of the vertigo is located in the neck in such circumstances (stationary head, rotation of the trunk and cervical spine without visual control). This led to the authors' decision not to use video-nystagmography in a darkened room, since this would probably elicit merely COR, which is largely subclinical and may be elicited even in healthy people. If the head is kept strictly stationary and the experiment is performed in daylight, rotation of the trunk does not evoke vertigo -as is evident from the cohort herein.
A further hypothesis is that CV may arise out of impaired blood circulation in the vertebrobasilar arteries. In 1933, DeKleyn and Versteegh first described a syndrome of vertigo and nystagmus produced by head movement. In post-mortem studies, they noted compromised circulation in the vertebral arteries (VA) with head rotation. Later, stroke accompanying maximum rotation of the head was described in archery (23). Nevertheless, because of the collateral blood flow through the contralateral VA and the circle of Willis, VA occlusion does not lead to symptoms in most individual cases.Thus, cases of symptomatic rotational vertebral artery occlusion are very rare (24)(25) The "goldstandard test" for this diagnosis is DSA (digital subtraction angiography) of the VA with the head turned to either side (26). This is, however, an intrinsically risky procedure, often therefore ruled out of routine practice. Investigation of the effect of the position of the head on flow rate in the vertebral arteries, as measured by Doppler ultrasound at rotations of 30degrees up to 60 degrees to either side, revealed no changes in blood flow in healthy subjects, which means that common rotation of the cervical spine cannot elicit vertigo (27). In this study, three patients (0.8%) had haemodynamically significant carotid artery stenosis, all of them from the symptomatic (patients with vertigo) group. This prevalence is similar to that reported in a large international population study in Europe (28). All three of these patients exhibited cardiovascular risk factors. None of the patients in this study appeared to have significant stenosis of the vertebral arteries, either symptomatic or asymptomatic. The prevalence of asymptomatic vertebral artery stenosis in the general population is not well known, but it constitutes approximately 13% of patients with symptomatic internal carotid artery stenosis (29), so the percentage is very low. Thus, in conclusion, the available literature indicates that hypoperfusion in the vertebrobasilar territory has no close correlation with clinical symptoms of vertigo, and should not be raised as the sole reason in explaining CV (30).
Vertigo, in general, is a common condition, yet definitions vary and management guidelines are often contradictory (31). A survey in Germany, for example, reported that vertigo had a prevalence of 22.9% in an adult population recruited over 12 months (1); a cross-sectional study in Scotland reported 21% prevalence per life (2). In the elderly, the prevalence rises to 30%-40% of persons over the age of 70, resp. to 45%-51% at ages 88-90 (32). The most commonly-reported presenting symptom is dizziness (33), but symptoms may also include nausea (34), vomiting (35), light-headedness, and difficulty standing or walking (33). Vertigo is not a single disease entity but a symptom of a wide range of diseases of varying etiology. These may arise from the inner ear, the brainstem, and/or the cerebellum, or they may be of psychosomatic origin. More than half the patients suffering from dizziness have non-vestibular diagnoses (36).
Patients with intrinsic problems (cardiovascular, pulmonary etc.), are unlikely to suffer from pure rotational vertigo and the severity of this condition is often overrated by their clinicians (11). Orthostatic dizziness in the adult population has accounted for 42% of all participants with vertigo and for 55% of non-vestibular dizziness diagnoses (37). These findings correlate with the results of this study -in 44% of symptomatic (vertigosuffering) patients, orthostatic etiology was confirmed by detailed internal examination.
Five patients were diagnosed with uncompensated hypertension, making up 28% of the symptomatic group. In general, hypertension and dizziness are both highly prevalent and significantly associated, highlighting a pressing need for investments in preventive measures (38).
Benign paroxysmal positional vertigo (BPPV) is the most common of the peripheral types of vertigo. Tan noted that 9% of elderly patients undergoing general geriatric assessment exhibited unrecognized BPPV (39). This percentage proved even higher in a larger series of patients -approximately 34% (40). Our study disclosed four patients with BBPV (22% of symptomatic subjects), but the group was too small to draw any definite conclusions. A particular explanation also suggests itself here: that some patients with vertigo in the cohort herein had undergone series of special vestibular tests before they were admitted to our centre.

Conclusion
In conclusion, despite a comparatively high prevalence of vertigo in this cohort of DCM patients, it proved impossible to demonstrate that it could be provoked by the cervical torsion test, or related to stenotic changes in the vertebral arteries. No indications emerged for the existence of CV in DCM patients. If patients with cervical spondylosis suffer from vertigo, it is therefore necessary to doubt so-called "cervical vertigo" and to   BPPV: benign paroxysmal positional vertigo