Timothy C. Hain, MD • Page last modified: March 20, 2021
The process is generally uncertain and frustrating. Cervical vertigo is reported to be most common in the 30-50 year old age group, as well as being more common in the female population (Heikkila, 2004). As is common in poorly defined syndromes, criteria for diagnosis are inconsistent and these criteria must be taken as being tentative at best.
According to Heikkila (2004), neck pain must be present. The author of this review does not agree with this statement -- as it would obviously exclude several mechanisms above (e.g. vascular, cord compression). Heikkila also suggests that tinnitus and low-frequency hearing loss are possible. We would disagree that these symptoms are diagnostic. Heikkala suggested that there may be photophobia and blurred vision. We would generally attribute these sorts of symptoms to migraine, ocular disturbances (e.g. post cataract removal), concussion, or much more remotely, vertebral insufficiency.
There is no consensus on how to diagnose cervical vertigo (Brandt, 1996). The author of this page uses a combination of criteria (see above). First, one excludes other causes of vertigo such as vestibular neuritis (with an ENG and/or rotatory chair test), and BPPV (with a positional test). Other entities that need to be ruled out including inner ear disease such as Meniere's syndrome, central vertigo, psychogenic vertigo (often including malingering when there are legal issues), and medical causes of vertigo. As cervical vertigo often is associated with a head injury, in this situation, the various causes of post-traumatic vertigo should be considered. There should be a sufficient cause of neck injury (whiplash injury or severe arthritis). Symptoms elicited by massage of the neck or vibration to the neck add to the clinical suspicion.
There should be little or no hearing symptoms or findings, other than an occasional low-tone sensorineural hearing reduction (an audiogram and OAE is recommended). There may be ear pain (otalgia), as part of the ear is supplied by sensory afferents from the high cervical nerve roots.
On physical examination, there should be no spontaneous nystagmus, but there may be positional nystagmus. Many patients who have vertigo in the context of neck disease have a BPPV type nystagmus on positional testing. More precisely, there is an upbeating nystagmus supine. (Hozl et al, 2009). This suggests that the neck afferents may interact strongly with vestibular inputs derived from the posterior canal. It is known that neck input interacts with caloric responses (Kobayshi et al, 1991). In our experience however, we also find that persons with Migraine may have BPPV type nystagmus, and the literature suggests that Migraine commonly is associated with positional nystagmus.
Use of VNG to diagnose cervical vertigo: Although the idea is logical, the author has not generally found it helpful clinically to compare positional results with the head kept constant on body to positional tests where there is head on trunk movement. While persons with herniated disks often do develop nystagmus when their head is turned while upright (see next section), current ENG technology is usually insufficient to document it. It seems likely to us that a methodology where there was a method of keeping the eye in the center of the orbit (perhaps with a "winking" light, and a method of quantifying head position on the body (perhaps with a still picture grabbed at the right point), would do the job. Nobody has this technology implemented yet in their ENG systems. In the author's practice, this test is done entirely at the bedside.
Often it is helpful to compare nystagmus elicited with the head prone to with the head supine, as if the nystagmus does not reverse, cervical vertigo seems fairly certain.
Head-turning upright test (neck torsion nystagmus). Another potentially useful maneuver is to turn the head to one side to the limit of range, while the examinee is upright and simply wait for 30 seconds (Cherchi and Hain, 2010). The figure below shows a weak positive and the movie below in the case section shows a strong positive. Clinically, nystagmus that changes direction according to the direction of the head on neck, rather than with gravity, makes cervical vertigo likely. It is the author's personal observations that persons who are positive on this test nearly always have a disk abutting their cervical cord, generally at C5-6. Similar to Devaraja (2018), we feel that this test is not especially useful. Our observations is that it is neither sensitive nor specific for cervical vertigo.
The craniocervical flexion test (CCFT) are difficult ones implement as well as accept as rational. The both involve using EMG. The related "COR" test is discussed separately below.
Posturography with the head held in different angles on the neck has been used in an attempt to diagnose cervical vertigo (Dejong and Bles 1986; Kogler et al, 2000). This process, seems to have too many free variables -- in other words, people can simply sway more due to anxiety or voluntarily sway more, in situations where there is benefit to be obtained from being diagnosed as having cervical vertigo. This situation, of course, naturally arises in people litigating after an auto accident.
Static posturography does not appear to be useful. Dynamic posturography, incorporating sway referencing, may be more sensitive (Alund et al, 1991).
Joint position error. The physical therapy literature has numerous enthusiastic reports concerning this test for cervical vertigo. Like posturography, this test is vulnerable to intent, bias, and malingering. Furthermore, this test presumes a mechanism (proprioception), which may be true in some patients, but certainly not in all. It may be helpful though in a subset of patients with cervical vertigo (i.e. the subset that has proprioceptive loss rather than vertebral artery compromise or cervical stenosis).
Smooth pursuit neck torsion (SPNT) test is another procedure favored by papers in the literature produced by physical therapists. This is a strange procedure where a smooth pursuit is measured with the head turned on 45 degrees to either side. Logically, the connection between neck torsion and smooth pursuit is very difficult to follow, and we think that it is likely overly enthusiastic investigators and overly cooperative subjects. .
|Cervical nystagmus recorded with head turned to left.|
Cervico-ocular reflex (COR) testing
This is a difficult endeavor involving turning the body while recording the eyes. The most practical method is to have the patient sit on a swivel stool, wearing video Frenzel goggles, with vision denied (i.e. in total darkness). Then the patient turns themselves about 30 deg to either side, while the examiner holds the head still in space and watches on the video monitor. This process is nearly always unproductive of any visible eye deviation in persons with complete bilateral vestibular loss. So the straightforward approach doesn't work.
Kelders et al (2005) reported that the COR is increased in whiplash patients. Schubert et al (2004) reported that it is weak, unreliable, and is not increased in patients with unilateral vestibular loss. Bronstein and Hood (1986) reported that it is increased in patients with absent vestibular function. In our clinic, we have not found the COR to be very helpful or reliable. All together, the COR does not seem to be a good way to diagnose cervical vertigo.
Laboratory studies: If cervical vertigo still seems likely after excluding reasonable alternatives, one next needs to look for positive confirmation. Routine studies in working up cervical vertigo include:
CT angiogram with 3-dimensional reconstruction. Left vertebral (left lower) is large and dominant. Right vertebral (right lower) is small and hypoplastic.
Note that reconstructions are not nearly as reliable as simple images, as reconstructions are done by computers that may have their own ideas about what is a blood vessel and what is noise.
Angiography: CT-angiography has been rapidly improving in recent years and it is excellent for detection of vertebral hypoplasia -- which is as much as you may be able to determine anyway. Three-dimensional reconstructions can be very helpful.
The "gold standard test" for the cervical vertigo due to compression of the vertebral arteries is selective vertebral angiography with the head turned to either side. Vertebral angiography is preferred for head-turning tests because there is less dye put into the body than for CT-angiography. However, because selective vertebral angiography is a risky procedure by itself, often it is decided not to proceed to this step. Our position is that one should not attempt vertebral angiography, but simply do CT-angiography as long as kidney function is adequate.
There is also another problem --
Catch 22: A basic flaw with any "head turned" radiographic procedures is the "Catch 22" problem -- if there is a risk of head turning -- it may not be detectable. The reason is that if there is even a tiny risk of a stroke during a radiographic procedure, radiologists may simply choose not to turn the head. Or to put this another way - - a risk averse radiologist will be unable to diagnose a vertebral occlusion associated with head turning because they will refuse to turn the head sufficiently to diagnose it. Practically, it is generally impossible to monitor one's radiology department sufficiently to be sure that they do turn the head to end rotation. Thus, in some settings, it may be simply impossible to diagnose vertebral artery occlusion because of radiologist risk aversion.
Ordinary MRA and vertebral doppler procedures are rarely abnormal, and sometimes are used as a screening procedure to decide whether vertebral angiography is necessary. We are unenthusiastic about this as it seems unreasonable to us to use methods that are unreliable as screening procedures.
An MRI scan of the neck and flexion-extension X-ray films of the neck are suggested in all. We strongly advise against "open MRI", or "stand up MRI", as the image quality from these procedures is not as good as higher field methods.
Fluoroscopy of the neck may be used in persons with abnormal flexion-extension views. ENG testing is recommended, largely to exclude alternative causes. Vertebral artery doppler may be helpful in some. (Sakaguchi et al. 2003), but we presently prefer CT-angiography.