Key Points
Question
Are patients with cervical spondylosis more likely than patients with lumbar spondylosis to have cervical dizziness?
Findings
In this cohort study of 3638 patients with cervical spondylosis and 3638 matched controls with lumbar spondylosis, the patients with cervical spondylosis had a slightly higher risk of dizziness than matched controls. Although the 1-year incidence of dizziness after diagnosis of cervical spondylosis was 10.2%, it was only 1.6% higher than that in the control group.
Meaning
This study supports the association between dizziness and cervical spondylosis but reveals that cervical dizziness is uncommon even in patients with cervical spondylosis.
This cohort study assesses the risk of dizziness among patients with cervical spondylosis compared with those with lumbar spondylosis.
Abstract
Importance
The dizziness associated with cervical spondylosis is a controversial topic given that many experts believe that cervical spondylosis is a common cause of dizziness, whereas others do not believe it exists.
Objective
To compare the risk of dizziness between patients with cervical spondylosis and matched controls (ie, patients with lumbar spondylosis after propensity score matching [PSM]).
Design, Setting, and Participants
This cohort study used medical claims data from the National Health Insurance Research Database of Taiwan for patients 60 years or older with cervical or lumbar spondylosis newly diagnosed in any outpatient department between January 1, 2010, and December 31, 2015. Patients diagnosed with cervical spondylosis were included as the study cohort, and those diagnosed with lumbar spondylosis who were matched to the study cohort via PSM were selected as the control cohort. Both cohorts were followed up for 1 year unless they were diagnosed with dizziness, censored by death, or withdrew from the health insurance program. Data analysis was performed from August 9 to September 20, 2022.
Main Outcomes and Measures
The main outcome was the date of outpatient diagnosis of dizziness. The risks of dizziness were compared between groups. The relative risk and incidence rate difference were calculated.
Results
A total of 3638 patients with cervical spondylosis (mean [SD] age, 67.9 [7.1] years; 2024 [55.6%] male) and 3638 patients with lumbar spondylosis (mean [SD] age, 68.0 [7.1] years; 2024 [55.6%] male) after PSM were selected as the study and control cohorts, respectively. The patients with cervical spondylosis had higher risk of dizziness than matched controls, with a 1-year relative risk of 1.20 (95% CI, 1.03-1.39). The 1-year incidence of dizziness was 10.2% (95% CI, 9.2%-11.2%) in patients with cervical spondylosis and 8.6% (95% CI, 7.7%-9.5%) in the matched group of lumbar spondylosis. The incidence rate difference between the groups was 1.6% (95% CI, 0.3%-3.0%).
Conclusions and Relevance
These data support the association between dizziness and cervical spondylosis, but the small difference between groups reveals that dizziness associated with cervical spondylosis is uncommon. Clinicians should be wary of diagnosing a cervical cause for dizziness based on an actual history of cervical spondylosis.
Introduction
Dizziness is one of the most common symptoms in general practice but also a diagnostic challenge because it can be caused by vestibular, neurologic, cardiovascular, metabolic, or even psychiatric disorders. Among these varied causes, a specific cause that includes the experience of dizziness as part of a cervical disorder is termed cervical dizziness, cervicogenic dizziness, or cervical vertigo.1 Cervical dizziness was first reported by Barré2 in 1926 and then by Ryan and Cope3 in 1955. The report of dizziness attributed to neck pathology is accompanied by neck pain and typically triggered by head and neck movement. Thus, although cervical dizziness has been described in the literature for nearly 100 years, it remains an ambiguous diagnosis because there are neither accepted consensus diagnostic criteria nor agreed-on diagnostic tests.1,4 Regardless, dizziness associated with cervical spondylosis is particularly controversial given that some experts consider it one of the most common causes of dizziness, whereas other experts doubt its existence.1,2,3,4,5
A community-based study of older adults with dizziness found that the most common causes of dizziness were related to central vascular disease (70.5%) and cervical spondylosis (65.8%).6 Another study of 1000 patients with dizziness seen in multiple outpatient departments found that 89% did in fact have cervical spondylosis, of whom 90% reported significant improvement in their dizziness after using prescribed muscle relaxants.7 However, opponents to the theory that dizziness can arise from cervical spondylosis suggest the association is coincidental and not causal because both of them have a high prevalence in the general population.8,9 The common report of neck movement–induced dizziness (ie, head-motion dizziness)10 cannot be attributed to cervical spondylosis because the vestibular system is also engaged, making it difficult to distinguish any cervical contribution. Furthermore, neck pain is not exclusive to pathology of the cervical spine given that people with chronic vestibular disorders often fix their head to prevent head-motion dizziness, which results in neck stiffness and pain.1 Recently, a select team of clinician scientist members of the Barany Society concluded that the current evidence supporting the correlation between dizziness and cervical spondylosis is lacking and further controlled studies are warranted.1
The purpose of this retrospective cohort study was to examine the association between dizziness and cervical spondylosis. We compared the risk of dizziness between older patients with cervical and lumbar spondylosis. The patients with lumbar spondylosis serve as a patient control cohort that is relevant to cervical spondylosis given both diseases are caused by degenerative spinal conditions common in older adults. We hypothesized that if dizziness can be associated with cervical spondylosis, the incidence of dizziness should be higher in the cohort with cervical spondylosis.
Methods
We conducted a population-based, retrospective cohort study from the Longitudinal Generation Tracking Database (LGTD) using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnoses from outpatient departments in Taiwan. This study design culled data from the National Health Insurance Research Database of Taiwan; personal information is deidentified. We defined each cohort for the primary analysis as patients with cervical spondylosis (study cohort) and patients with lumbar spondylosis (control cohort). This study was performed in accordance with the ethical standards of the Declaration of Helsinki11 and its later amendments and was approved by the institutional review board of the Research Ethics Committee of Taichung Tzu Chi Hospital. The committee waived the need for informed consent given the use of deidentified data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Data Source
The National Health Insurance Research Database of Taiwan contains records of approximately 23 million enrollees dating back to inception in March 1995 and represents 98% of the total population in Taiwan. We reviewed records from the LGTD, which includes claims data for 2 million enrollees randomly sampled from all beneficiaries of the National Health Insurance program in 2005. The medical records in the LGTD include clinic visits, emergency department visits, and inpatient hospitalizations ranging from 2000 to 2017. To ensure confidentiality, the enrollees’ personal information is protected using anonymous unique identification numbers. We extracted data based on the International Classification of Diseases. Before 2016, the ICD-9-CM was used for recording diagnosis in the National Health Insurance Research Database, and the International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) has been used since 2016.
Study Population
Figure 1 illustrates the selection process of the individuals in this study. Patients 60 years or older with cervical or lumbar spondylosis newly diagnosed in any outpatient department between January 1, 2010, and December 31, 2015, were identified from the LGTD. The study group of patients with cervical spondylosis included those with cervical spondylosis with myelopathy (ICD-9-CM code 721.1) and without myelopathy (ICD-9-CM code 721.0). The control group of lumbar spondylosis included lumbar spondylosis with myelopathy (ICD-9-CM code 721.42) and without myelopathy (ICD-9-CM code 721.3). To ensure recruitment purity, patients with both cervical and lumbar spondylosis were excluded. All patients who had dizziness (ICD-9-CM codes 780.4, 386.XX) for any other reason (eg, vestibular hypofunction) that existed before the diagnosis of cervical or lumbar spondyloses were excluded.
Figure 1. Selection Process of the Study Participants.
LGTD indicates Longitudinal Generation Tracking Database; PSM, propensity score matching.
Outcome Measures
Our primary outcome was the date of initial outpatient diagnosis of dizziness (ICD-9-CM codes 780.4, 386.XX; ICD-10-CM codes R42, H81-83). All individuals (both cohorts) were followed up for 1 year unless they were diagnosed with dizziness, censored by death, or withdrew from the health insurance program.
Propensity Score Matching
Propensity score matching is a statistical method used to construct an artificial subgroup from the original control group (ie, lumbar spondylosis group) that is comparable to the study group by matching covariates between the study and control groups. In our study, several covariates, such as having cerebrovascular disorders, may confound the probability of cervical spondylosis diagnosis in these 2 groups. Propensity score matching allows for the construction of an artificial subgroup with a balanced covariate distribution to ensure a fair estimate. The propensity score represents each patient’s probability of being diagnosed with cervical spondylosis based on 25 covariates that were potential confounders or comorbidities with dizziness. We used a 1:1 PSM with the nearest neighbor matching method. We set a caliper value of 0.2 as the tolerance for maximum allowable differences in propensity scores between the matched pairs. Additionally, we enforced an exact match on sex.
Statistical Analysis
We evaluated the balance of individual covariates after PSM by the standardized differences, which is a measure of distance between 2 group means. After PSM, we compared the risk of dizziness between both groups using the Kaplan-Meier method and Cox proportional hazards regression model to estimate the hazard ratios (HRs) of dizziness. We calculated the 1-year relative risk between the groups. The 1-year incidence of dizziness after diagnosis of cervical spondylosis and lumbar spondylosis was calculated. The incidence rate difference between groups was analyzed. Data analysis was performed from August 9 to September 20, 2022.
Results
In the database, 3645 patients diagnosed with cervical spondylosis and 8365 diagnosed with lumbar spondylosis met the study’s inclusion and exclusion criteria. Through the process of PSM, a subset of 3638 patients in the lumbar spondylosis group (mean [SD] age, 67.9 [7.1] years; 2024 [55.6%] male and 1614 [44.4%] female) were matched to 3638 patients with cervical spondylosis (mean [SD] age, 68.0 [7.1] years; 2024 [55.6%] male and 1614 [44.4%] female). After PSM, the differences in demographics and comorbidities between the 2 groups were not significant and thus considered minimal and an indicator of good balance (Table).
Table. Demographic Data of the Cervical Spondylosis and Lumbar Spondylosis Cohorts After Propensity Score Matchinga.
| Variable | Spondylosis | Standardized differenceb | |
|---|---|---|---|
| Cervical (n = 3638) | Lumbar (n = 3638) | ||
| Age, mean (SD), y | 67.9 (7.1) | 68.0 (7.1) | 0.014 |
| Sex | |||
| Male | 2024 (55.6) | 2024 (55.6) | 0 |
| Female | 1614 (44.4) | 1614 (44.4) | |
| Comorbidities | |||
| Cardiovascular disorders | 475 (13.1) | 459 (12.6) | 0.032 |
| Cerebrovascular disorders | 218 (6.0) | 229 (6.3) | 0.037 |
| Visual disorders | 724 (19.9) | 762 (21.0) | 0.048 |
| Hearing loss | 37 (1.0) | 28 (0.8) | 0.159 |
| Migraine | 19 (0.5) | 22 (0.6) | 0.129 |
| Anxiety | 217 (6.0) | 232 (6.4) | 0.049 |
| Depression | 102 (2.8) | 104 (2.9) | 0.026 |
| Traumatic brain injury | 63 (1.7) | 62 (1.7) | 0 |
| Hypertension | 1595 (43.8) | 1590 (43.7) | 0.003 |
| Diabetes | 684 (18.8) | 657 (18.1) | 0.033 |
| Dyslipidemia | 44 (1.2) | 45 (1.2) | 0 |
| COPD | 357 (9.8) | 376 (10.3) | 0.039 |
| Cancer | 231 (6.4) | 220 (6.1) | 0.036 |
| Dementia | 57 (1.6) | 52 (1.4) | 0.096 |
| Parkinson disease | 37 (1.0) | 35 (1.0) | 0 |
| Hypotension | 6 (0.2) | 9 (0.3) | 0.284 |
| Chronic kidney disease | 146 (4.0) | 163 (4.5) | 0.087 |
| Medications | |||
| Sedatives | 437 (12.0) | 443 (12.2) | 0.013 |
| Muscle relaxants | 1144 (31.5) | 1141 (31.4) | 0.003 |
| ≥5 Medications | 2746 (75.5) | 2776 (76.3) | 0.031 |
| Urbanization | |||
| Urban | 2266 (62.3) | 2273 (62.5) | 0.006 |
| Suburban | 947 (26.0) | 940 (25.8) | 0.007 |
| Rural | 425 (11.7) | 425 (11.7) | 0 |
| No. of outpatient visits in 1 y before index date, median (IQR) | 32 (20-48) | 31 (19-48) | NA |
| CCI score, median (IQR) | 1 (0-1) | 0 (0-1) | NA |
Abbreviations: CCI, Charlson Comorbidity Index; COPD, chronic obstructive pulmonary disease; NA, not applicable.
Data are presented as number (percentage) of patients unless otherwise indicated.
The standardized difference is the measure of distance between 2 group means, with 0.2, 0.5, and 0.8 representing small, medium, and large effect sizes, respectively.
Figure 2 shows the cumulative incidences of dizziness in each patient cohort. The patients with cervical spondylosis had a higher risk of dizziness than those with lumbar spondylosis (HR, 1.20; 95% CI, 1.03-1.39). The incidence of dizziness within 1 year after diagnosis of cervical spondylosis was 10.2% (95% CI, 9.2%-11.2%). The incidence of dizziness after diagnosis of lumbar spondylosis (ie, matched control) was 8.6% (95% CI, 7.7%-9.5%). The incidence rate difference between the cervical spondylosis group and control group was 1.6% (95% CI, 0.3%-3.0%). The 1-year relative risk of developing dizziness after spondylosis was 1.20-fold higher in the cervical spondylosis group (95% CI, 1.03-1.39).
Figure 2. Cumulative Incidences of Dizziness After Diagnosis of Cervical and Lumbar Spondylosis.
The patients with cervical spondylosis had a higher risk of dizziness than those with lumbar spondylosis (hazard ratio, 1.20; 95% CI, 1.03-1.39).
Discussion
In this study, we found that the risk of developing dizziness in patients with cervical spondylosis was slightly greater than the risk of dizziness in matched controls with lumbar spondylosis, which supports the possibility of cervical causes of dizziness in some older patients with cervical spondylosis. However, although the 1-year incidence of dizziness after diagnosis of cervical spondylosis was 10.2%, the difference in the incidence rate between cervical and lumbar spondylosis was only 1.6%. This small effect size suggests that even for those who already have a degenerative cervical spine disorder, cervical spondylosis is an uncommon cause of their dizziness. Dizziness has been linked with cervical spondylosis for almost 100 years,2,3 and many theories, including vascular,12,13 proprioceptive,3,14,15 and autonomic mechanisms,2,16 have been proposed to explain their association; however, correctly diagnosing cervical dizziness in patients with cervical spondylosis is still difficult due to the absence of both accepted diagnostic criteria and validated diagnostic tests.1 Clinicians should not make hasty diagnoses of cervical dizziness just because the patient has a history of cervical spondylosis or a cervical spine radiograph that reveals degenerative features. Instead, detailed history taking and clinical examinations are crucial.
Specific clinical examinations, such as the cervical torsion test and cervical relocation test, may further help diagnosis, yet their validity is variable.17 An excellent review and suggestive clinical algorithm can be read in the article by Reiley et al.18 Accordingly, cervical dizziness is still a diagnostic challenge that can only be made after comprehensive exclusion of other origins. For example, dizziness during neck movement or a time-locked association between neck pain and dizziness implies the possibility of cervical dizziness; however, neck motion would also stimulate the vestibulo-ocular reflex or trigger a head motion dizziness common in migraine.19,20,21 Our data support that the dizziness related to cervical spondylosis exists but is rare. Most causes of dizziness in patients with cervical spondylosis are not cervicogenic.
To our knowledge, there is no pathophysiologic association between dizziness and lumbar spondylosis. However, in our study, up to 8.6% of the patients with lumbar spondylosis developed dizziness within 1 year of diagnosis. Therefore, it appears that dizziness and cervical or lumbar spondylosis develop through complex medical interactions. We offer 3 possibilities: (1) the dizziness is not associated with the cervical or lumbar spondylosis but instead is associated with a comorbid, different medical diagnosis (eg, Meniere disease or anemia); (2) the dizziness is indirectly associated with the spondylosis via covariates but not cervical pathology (eg, dizziness is from prescribed pain medications for cervical or lumbar spondylosis); or (3) the dizziness is truly associated with having cervical pathology (ie, cervical dizziness) that we do not understand. In our study, the dizziness in the patients with cervical spondylosis included each of the 3 conditions we describe above, whereas dizziness in the patients with lumbar spondylosis was only possible because of the first 2 conditions (Figure 3). After PSM, the incidences of the first and second explanations for dizziness would be similar in the 2 groups because the demographics, comorbidities, and other covariates were mostly balanced via PSM. Therefore, the relative and excessive risks (eg, incidence rate difference)22 of dizziness between cervical spondylosis and control groups represent, to some extent, the commonness of the cervical cause of dizziness. As the results in our study show, the low relative and excessive risks between the study groups suggest that dizziness associated with cervical spondylosis is uncommon.
Figure 3. Model of Complex Interactions Between Dizziness and Cervical and Lumbar Spondylosis.
Limitations
Our study has several limitations. First, in this database study, we used the date of diagnosis and not the symptom onset date because time of onset was unavailable. There might be a time bias between symptom onset and date of diagnosis. Second, we could not verify the coded diagnosis. Third, even though the 25 covariates we selected were balanced between groups via PSM, there may be other covariates that we are unaware of and thus did not capture. Fourth, we were unable to access the clinical details to identify critical correlates, such as the nature of the dizziness (vertigo, presyncope, disequilibrium, or lightheadedness), the patterns of dizziness (acute, episodic, or chronic), any triggers of dizziness, or the findings of a physical examinations.
Conclusions
This cohort study showed that patients with cervical spondylosis have a higher risk of dizziness than patients with lumbar spondylosis (matched controls). This finding suggests that dizziness may be associated with cervical spondylosis. However, our study also shows that cervical dizziness is not common in patients with cervical spondylosis. Therefore, clinicians should not diagnose a patient with cervical dizziness just because their personal history or radiographic record shows cervical spondylosis. Clinicians should thoroughly exclude other possible causes before making the diagnosis of cervical dizziness, particularly given an absent consensus on the diagnostic criteria for cervical dizziness.
Data Sharing Statement
References
- 1.Seemungal BM, Agrawal Y, Bisdorff A, et al. The Bárány Society position on ‘cervical dizziness’. J Vestib Res. 2022;32(6):487-499. doi: 10.3233/VES-220202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barré JA. Sur un syndrome sympathique cervical poste ´rieur et sa cause frequente, la’rthrite cervical. Rev Neurol (Paris). 1926;1:1246-1248. [Google Scholar]
- 3.Ryan GM, Cope S. Cervical vertigo. Lancet. 1955;269(6905):1355-1358. doi: 10.1016/S0140-6736(55)93159-7 [DOI] [PubMed] [Google Scholar]
- 4.Brandt T, Bronstein AM. Cervical vertigo. J Neurol Neurosurg Psychiatry. 2001;71(1):8-12. doi: 10.1136/jnnp.71.1.8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vural M, Karan A, Albayrak Gezer İ, et al. Prevalence, etiology, and biopsychosocial risk factors of cervicogenic dizziness in patients with neck pain: a multi-center, cross-sectional study. Turk J Phys Med Rehabil. 2021;67(4):399-408. doi: 10.5606/tftrd.2021.7983 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Colledge NR, Barr-Hamilton RM, Lewis SJ, Sellar RJ, Wilson JA. Evaluation of investigations to diagnose the cause of dizziness in elderly people: a community based controlled study. BMJ. 1996;313(7060):788-792. doi: 10.1136/bmj.313.7060.788 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Takahashi S. Importance of cervicogenic general dizziness. J Rural Med. 2018;13(1):48-56. doi: 10.2185/jrm.2958 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lv Y, Tian W, Chen D, Liu Y, Wang L, Duan F. The prevalence and associated factors of symptomatic cervical spondylosis in Chinese adults: a community-based cross-sectional study. BMC Musculoskelet Disord. 2018;19(1):325. doi: 10.1186/s12891-018-2234-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bisdorff A, Bosser G, Gueguen R, Perrin P. The epidemiology of vertigo, dizziness, and unsteadiness and its links to co-morbidities. Front Neurol. 2013;4:29. doi: 10.3389/fneur.2013.00029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bisdorff A, Von Brevern M, Lempert T, Newman-Toker DE. Classification of vestibular symptoms: towards an international classification of vestibular disorders. J Vestib Res. 2009;19(1-2):1-13. doi: 10.3233/VES-2009-0343 [DOI] [PubMed] [Google Scholar]
- 11.World Medical Association . World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
- 12.Bayrak IK, Durmus D, Bayrak AO, Diren B, Canturk F. Effect of cervical spondylosis on vertebral arterial flow and its association with vertigo. Clin Rheumatol. 2009;28(1):59-64. doi: 10.1007/s10067-008-0983-0 [DOI] [PubMed] [Google Scholar]
- 13.Olszewski J, Majak J, Pietkiewicz P, Luszcz C, Repetowski M. The association between positional vertebral and basilar artery flow lesion and prevalence of vertigo in patients with cervical spondylosis. Otolaryngol Head Neck Surg. 2006;134(4):680-684. doi: 10.1016/j.otohns.2005.11.023 [DOI] [PubMed] [Google Scholar]
- 14.Brandt T, Huppert D. A new type of cervical vertigo: head motion-induced spells in acute neck pain. Neurology. 2016;86(10):974-975. doi: 10.1212/WNL.0000000000002451 [DOI] [PubMed] [Google Scholar]
- 15.Freppel S, Bisdorff A, Colnat-Coulbois S, et al. Visuo-proprioceptive interactions in degenerative cervical spine diseases requiring surgery. Neuroscience. 2013;255:226-232. doi: 10.1016/j.neuroscience.2013.09.060 [DOI] [PubMed] [Google Scholar]
- 16.Morinaka S. Musculoskeletal diseases as a causal factor of cervical vertigo. Auris Nasus Larynx. 2009;36(6):649-654. doi: 10.1016/j.anl.2009.04.009 [DOI] [PubMed] [Google Scholar]
- 17.L’Heureux-Lebeau B, Godbout A, Berbiche D, Saliba I. Evaluation of paraclinical tests in the diagnosis of cervicogenic dizziness. Otol Neurotol. 2014;35(10):1858-1865. doi: 10.1097/MAO.0000000000000506 [DOI] [PubMed] [Google Scholar]
- 18.Reiley AS, Vickory FM, Funderburg SE, Cesario RA, Clendaniel RA. How to diagnose cervicogenic dizziness. Arch Physiother. 2017;7:12. doi: 10.1186/s40945-017-0040-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Thompson-Harvey A, Hain TC. Symptoms in cervical vertigo. Laryngoscope Investig Otolaryngol. 2018;4(1):109-115. doi: 10.1002/lio2.227 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Reid SA, Callister R, Katekar MG, Rivett DA. Effects of cervical spine manual therapy on range of motion, head repositioning, and balance in participants with cervicogenic dizziness: a randomized controlled trial. Arch Phys Med Rehabil. 2014;95(9):1603-1612. doi: 10.1016/j.apmr.2014.04.009 [DOI] [PubMed] [Google Scholar]
- 21.Ashina S, Bendtsen L, Lyngberg AC, Lipton RB, Hajiyeva N, Jensen R. Prevalence of neck pain in migraine and tension-type headache: a population study. Cephalalgia. 2015;35(3):211-219. doi: 10.1177/0333102414535110 [DOI] [PubMed] [Google Scholar]
- 22.Kim HY. Statistical notes for clinical researchers: risk difference, risk ratio, and odds ratio. Restor Dent Endod. 2017;42(1):72-76. doi: 10.5395/rde.2017.42.1.72 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
