A, A sagittal T2-weighted image demonstrates hyperintense T2 signal in the cord extending from the level of the third cervical vertebra to the sixth cervical vertebra (arrowheads). The hyperintense cord signal is consistent with cord gliosis secondary to myelopathic cord injury from severe motor tics. There has been surgical removal of the posterior elements from the fourth to sixth vertebrae. This posterior decompression provided more space to the spinal cord, which helps to minimize further cord injury. B, Axial T2-weighted image in the same patient demonstrates the bilateral nature of the cord signal, including involvement of the cord gray matter (arrowheads). Note the cord atrophy in panels A and B.
Cumulative incidence under the assumption of no competing risks estimated as 1 − the Kaplan-Meier estimate of survival function (with 95% CIs) of any cervical spine disorders in the subcohort of individuals who were followed up from age 3 years (3298 individuals with Tourette syndrome/chronic tic disorder and 32 980 matched controls). The shaded area represents the confidence intervals.
eTable 1. ICD-10 codes for the neuropsychiatric disorders included in the study and corresponding minimal age limits
eTable 2. ICD-10 codes and surgical codes for the outcomes included in the study
eTable 3. Characteristics of the study cohorts
eTable 4. Risk of cervical spine disorders in individuals with TS/CTD compared to matched unaffected subjects stratified by sex
eTable 5. Risk of cervical spine disorders in individuals with TS/CTD, excluding individuals with register diagnoses of rheumatic disorders, injuries due to traffic accidents, and fall- and sport-related injuries from both the TS/CTD exposed cohort and matched general population cohort
eFigure. Flowchart describing the inclusion and exclusion criteria of the final cohort
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Isung J, Isomura K, Larsson H, Sidorchuk A, Fernández de la Cruz L, Mataix-Cols D. Association of Tourette Syndrome and Chronic Tic Disorder With Cervical Spine Disorders and Related Neurological Complications. JAMA Neurol. 2021;78(10):1205–1211. doi:10.1001/jamaneurol.2021.2798
Are Tourette syndrome or chronic tic disorder associated with severe cervical spine disorders and related neurological complications?
In this matched cohort study, including 6791 individuals with Tourette syndrome or chronic tic disorder diagnosed in specialist settings and 67 910 unaffected individuals, patients had a 39% increased risk of having any cervical spine disorder (vascular or nonvascular).
Individuals with Tourette syndrome or chronic tic disorder may have an increased risk of cervical spine disorders, including those associated with potential long-term disability.
Severe forms of Tourette syndrome or chronic tic disorder (TS/CTD) may involve repeated head jerking. Isolated case reports have described a spectrum of severe neck disorders in individuals with TS/CTD. However, the nature and prevalence of cervical spine disorders in TS/CTD are unknown.
To establish if TS/CTD are associated with an increased risk of cervical spine disorders and related neurological complications compared with individuals from the general population.
Design, Setting, and Participants
All individuals born from 1973 to 2013 and living in Sweden between 1997 and 2013 were identified. Individuals with a record of TS/CTD diagnosed in specialist settings were matched on age, sex, and county of birth with 10 unexposed individuals randomly selected from the general population. Cox proportional hazards regression models were used to estimate the risk of vascular and nonvascular cervical spine disorders among exposed individuals, compared with unexposed individuals. Models were adjusted for other known causes of cervical spine injury. Data were analyzed from March 19 to May 16, 2021.
International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnoses of TS/CTD in the Swedish National Patient Register.
Main Outcomes and Measures
Records of cervical vascular disorders (ie, aneurysm, cerebral infarction, transitory cerebral ischemia) and cervical nonvascular disorders (ie, spondylosis, cervical disc disorders, fractures of the cervical spine, cervicalgia) and cervical surgeries. Covariates included rheumatic disorders, traffic injuries, fall- or sport-related injuries, and attention-deficit/hyperactivity disorder comorbidity.
A total of 6791 individuals with TS/CTD were identified (5238 [77.1%] were male; median [interquartile] age at first diagnosis, 15.6 [11.4-23.7] years) and matched to 67 910 unexposed individuals. Exposed individuals had a 39% increased risk of any cervical spine disorder (adjusted hazard ratio, 1.39; 95% CI, 1.22-1.59). Adjusted hazard ratios for cervical vascular and nonvascular disorders were 1.57 (95% CI, 1.16-2.13) and 1.38 (95% CI, 1.19-1.60), respectively. Risks were similar among men and women.
Conclusions and Relevance
Individuals with severe TS/CTD are at increased risk of cervical spine disorders. These outcomes are relatively rare but may lead to persistent disability in some individuals and thus require close monitoring to facilitate early interventions.
Tourette syndrome (TS) and chronic tic disorder (CTD) are neurodevelopmental disorders with a typical onset during early childhood.1 The cardinal symptoms are sudden, rapid, nonrhythmic movements, and/or sounds referred to as motor and vocal tics, respectively.1 Motor tics can be simple, involving a single muscle group, or complex, involving several muscle groups and more elaborate movements.1 Motor tics involving the neck region (eg, neck stretches/movements, head jerk movements, complex head/shoulder/neck movements) are common, occurring in up to 58% of children and adults with TS/CTD2 and also are reported to be among the most bothersome tics.3 The neck is particularly vulnerable to severe damage owing to repeated trauma.4 Single case reports5-11 and a case series4 of individuals with severe motor tics involving the head and neck region have described a range of severe injuries to the cervical spine with neurological complications. These include cervical disc herniation, cervical myelopathy with neurological complications, and even stroke secondary to traumatic vascular dissection of the carotid and vertebral arteries. Injury to the cervical cord following severe tics may require surgical interventions, such as removal of spinal elements for decompression (Figure 1), as well as highly specialized interventions such as deep brain stimulation to prevent further damage.12 Whether these anecdotal reports represent isolated cases or a more common phenomenon among individuals with TS/CTD is unknown. Given the potential seriousness of cervical spine disorders, it would be important to establish the prevalence and nature of cervical spine disorders in individuals with TS/CTD; this would help raise awareness and improve vigilance of individuals in need of individually tailored interventions to prevent long-term disability.
This Swedish population-based study aimed to provide estimates of the prevalence and types of cervical spine disorders in a large TS/CTD cohort. The analyses controlled for variables that are known to be associated with cervical spine disorders, such as rheumatic disorders,13 traffic injuries, fall- and sport-related injuries,14 and attention-deficit/hyperactivity disorder (ADHD) comorbidity.15
The study was approved by the Regional Ethical Review Board in Stockholm (reference 2013/862-31/5). Because the study was register based with no individuals being identifiable at any time, informed consent was waived.
We used the unique Swedish national identification number as key16 to link Swedish nationwide health and administrative registers. Data on demographic characteristics, migration, and deaths were extracted from the Swedish Total Population Register17 and the Cause of Death Register.18 Information on diagnoses given in inpatient (from 1969, with nationwide coverage for psychiatric disorders from 1973) and outpatient specialist services (since 2001), as well as surgical codes (according to the Nomesco Classification of Surgical Procedures), were retrieved from the National Patient Register (NPR). In the NPR, diagnoses are coded using the Swedish version of the International Classification of Diseases (ICD), with the ICD-10 classification being in use since 1997.19 Medication data were collected from the Prescribed Drug Register, which covers all prescribed medications dispensed in pharmacies in Sweden since July 2005.20
A population-based, matched-cohort design was used. The exposed cohort consisted of all Sweden-born individuals between 1973 and 2013 and living in the country at any time between 1997 and 2013 and who had a recorded diagnosis of TS/CTD in the NPR at 3 years or older. In line with prior research,21-26 exposure status was ascertained by using an algorithm that minimizes the inclusion of individuals with only transient tics and using a minimal age of diagnosis to avoid diagnostic misclassification (eTable 1 in the Supplement). The Swedish ICD-10 codes for TS/CTD diagnoses have been previously validated, with a reported excellent positive predictive value of 97%.27
The cohort entry date was the participants’ third birthday or January 1, 1997, whichever occurred last. The rationale for choosing the participants’ third birthday, instead of the actual date of first diagnosis, was that delays in help seeking and diagnosis are common in TS/CTD, and thus the date of the first diagnosis in the NPR does not generally reflect the actual onset of the disorder.28 Therefore, the assumption was that TS/CTD was present at the time of entering the cohort, regardless of the date of the first recorded diagnosis.
Individuals were excluded if they had a recorded diagnosis of one of the outcomes (see below) before the cohort entry date, if they had died or emigrated from Sweden before the cohort entry date, or if they had conflicting information (eg, individuals who appeared as having died or emigrated before being diagnosed with TS/CTD). For each of the exposed individuals, we identified from the Total Population Register 10 randomly selected individuals from the general population who were free from a diagnosis of TS/CTD during the study period and matched them by sex, birth year, and county of birth. To be considered for matching, unexposed individuals had to be living in Sweden and had to be free from any of the outcomes at the date when the corresponding index TS/CTD-exposed individual entered the cohort. The cohorts of TS/CTD-exposed individuals and matched controls were followed up from January 1, 1997, or their third birthday, whichever occurred last, to the date of the outcome, death, emigration from Sweden, or end of the study period (ie, December 31, 2013), whichever occurred first.
For the main analysis, records of cervical spine disorders and related neurological complications were collected from the NPR by means of their corresponding ICD-10 codes and operation codes. These included the following cervical vascular diagnoses: aneurysm and dissection of the internal carotid artery, aneurysm and dissection of the vertebral artery, aneurysm and dissection in other specified arteries, cerebral infarction, and transitory cerebral ischemia not otherwise specified. The included nonvascular disorders were spondylosis, cervical disc disorders (including with radiculopathy or myelopathy), fractures and distortions of the cervical spine, and cervicalgia. Additionally, to improve coverage of serious nonvascular disorders, we retrieved information on the following surgical codes: primary disc prosthesis of the cervical column, excisions and fusions of joints of the vertebral column, other surgical joint procedures of the vertebral column, fracture surgery of the vertebral column, and surgery on muscles and tendons of the cervical column (eTable 2 in the Supplement).
To control for other known causes of cervical spine disorders, we collected records of rheumatic disorders registered during the study period, including seropositive and seronegative rheumatoid arthritis (M05, M06), psoriatic arthritis (M07), juvenile idiopathic arthritis (M08), and ankylosing spondylitis (M45). These chronic conditions may be considered as potential confounders for the associations of interest given that autoimmune disorders have previously been suggested to be associated with TS/CTD.29 We also collected data on potential mediators, such as traffic injuries (V00-V99) and fall- and sport-related injuries (W00-W19). Information on injuries was collected if recorded from 1997 until the time each person was censored from follow-up. Additionally, to control for ADHD comorbidity during the study period, we retrieved the corresponding ICD-10 codes from the NPR, supplemented with dispensation data from the Prescribed Drug Register, following the methods described in previous Swedish register studies30 (eTable 1 in the Supplement).
Cox proportional hazards regression models were applied to estimate hazard ratios (HRs) with 95% CIs for the risk of cervical spine disorders in TS/CTD-exposed individuals, compared with matched (1:10) unaffected individuals, taking time in days after the start of follow-up as the underlying time scale. Analyses were performed separately for cervical vascular disorders and cervical nonvascular disorders (including surgical codes). Because the vascular and nonvascular groups are not mutually exclusive, the same individual may be included in the vascular and nonvascular disorder analyses. In addition, to increase statistical power for some of the analyses, we created a broader group, any cervical spine disorder, which included all the codes in the vascular and nonvascular groups.
The initial model automatically controlled for the matching variables (ie, birth year, sex, and county of birth; model 1, henceforth, unadjusted model). The second model adjusted for rheumatic disorders (model 2). The third model additionally controlled for traffic injuries (model 3). The final model further controlled for fall- and sport-related injuries (model 4). These 4 models were then repeated stratifying by sex. Finally, models 1 to 4 were rerun after excluding individuals with a comorbid diagnosis of ADHD from the exposed and unexposed cohorts. Data were analyzed from March 19 to May 16, 2021.
First, the main analysis was repeated limiting the TS/CTD cohort to those individuals who had at least 2 diagnoses of TS/CTD after age 18 years (as a proxy for the severity and long-term persistence of TS/CTD into adulthood).24 Those with a record of TS/CTD but who were no longer seen by a specialist after age 18 years and those who were seen only once after age 18 years were excluded, along with their matched unexposed controls.
Second, in the subgroup of individuals with complete data available from age 3 years, the expected cumulative incidence of any cervical spine disorder for exposed and matched unexposed individuals by the end of follow-up (ie, at age 20 years) was calculated using Kaplan-Meier survival estimates (under the assumption of no competing risks). These analyses ensure complete follow-up data, thus avoiding issues with left truncation (ie, exposure and/or outcome happening before the register started and thus missing).
Third, a post hoc sensitivity analysis excluded all individuals with records of rheumatic disorders, traffic injuries, and fall- and sport-related injuries from the exposed and unexposed cohorts.
A total of 12 917 387 individuals were identified as born in Sweden from 1973 to 2013. After applying all inclusion and exclusion criteria, 6791 individuals with a diagnosis of TS/CTD were identified and matched with 67 910 unexposed population controls (eFigure in the Supplement). The majority of identified individuals with TS/CTD were men (5238 [77.1%]). The median (interquartile range) age at first diagnosis was 15.6 (11.4-23.7) years. The median (interquartile range) length of follow-up was 17 (11.5-17.0) years for exposed individuals and 17 (11.0-17.0) years for unexposed individuals. Characteristics of the study cohorts are presented in eTable 3 in the Supplement.
A total of 237 individuals (3.5%) in the TS/CTD group and 1483 (2.2%) in the unexposed group had at least 1 diagnosis of a cervical spine disorder during follow-up, corresponding with a 39% increased risk in the fully adjusted model (adjusted HR [aHR], 1.39 [95% CI, 1.22-1.59]) (Table 1). Similarly, the risks of vascular and nonvascular cervical disorders were significantly higher in individuals with TS/CTD, compared with the matched controls (aHR, 1.57 [95% CI, 1.16-2.13] for vascular disorders and aHR, 1.38 [95% CI, 1.19-1.60] for nonvascular disorders) (Table 1). After excluding individuals with a comorbid diagnosis of ADHD from the TS/CTD-exposed and matched unexposed cohorts, the results from the fully adjusted model remained largely unchanged for all outcomes (Table 2).
In the sex-stratified cohorts, 3.2% (n = 167) of TS/CTD-exposed men and 4.5% (n = 70) of TS/CTD-exposed women had at least 1 record of any cervical spine disorder during follow-up (eTable 4 in the Supplement). The corresponding proportions among control men and women were 2.0% (n = 1032) and 2.9% (n = 451), respectively. The analyses revealed a similar risk of any cervical spine disorder for men (aHR, 1.39 [95% CI, 1.18-1.63]) and women (aHR, 1.41 [95% CI, 1.11-1.79]). Among men, the risk was higher for vascular (aHR, 1.90 [95% CI, 1.31-2.77]) than nonvascular disorders (aHR, 1.30 [95% CI, 1.09-1.55]). Among women, the risk of vascular outcomes was nonsignificant (aHR, 1.13 [95% CI, 0.66-1.92]), but the risk of nonvascular outcomes was significant (aHR, 1.66 [95% CI, 1.27-2.17]). However, these results need to be interpreted with caution owing to the limited statistical power in these analyses.
For the first additional analysis, we identified 1921 individuals with at least 2 diagnoses of TS/CTD after age 18 years (28.3% of the original TS/CTD cohort), thus corresponding with a group of more long-term disease. This subcohort was compared with their corresponding matched individuals without TS/CTD. Among individuals with persistent TS/CTD, 123 (6.4%) had at least 1 record of any cervical spine disorder, compared with 695 matched controls (3.6%). This translated into a 52% increased risk in the fully adjusted model (aHR, 1.52 [95% CI, 1.26-1.83]). The risk was also higher for vascular and nonvascular cervical spine disorders (Table 3).
In the second additional analysis, we identified a subcohort of individuals who were followed up from age 3 years (3289 individuals with TS/CTD and 32 890 matched controls) and had complete data. The Kaplan-Meier expected cumulative incidence of any cervical spine disorder at the end of follow-up was 2.1% (95% CI, 1.6%-2.9%) for the TS/CTD group and 1.5% (95% CI, 1.3%-1.8%) for the matched unexposed cohort (Figure 2). This corresponded with an aHR of 1.54 (95% CI, 1.16-2.05) in the fully adjusted model. The differences between the 2 cohorts for any cervical spine disorder were already significant (nonoverlapping 95% CIs) from midchildhood (Figure 2). There were too few cases to analyze vascular and nonvascular disorders separately.
In a final post hoc analysis, we reran the analyses excluding all individuals with records of rheumatic disorders, traffic injuries, and fall- and sport-related injuries from the exposed and unexposed cohorts. This resulted in slightly higher estimates but with confidence intervals that overlapped with those of the main analyses (eTable 5 in the Supplement).
This is the first systematic investigation of cervical spine disorders in individuals with TS/CTD, to our knowledge. The main finding was that individuals with TS/CTD diagnosed in specialist settings had a significantly increased risk (ranging from 39% to 54% in various analyses) of cervical spine disorders, compared with matched individuals from the general population. Analyses strictly controlled for a series of conditions that are known to be associated with neck injuries, including rheumatic disorders, traffic injuries, fall- or sport-related injuries, and ADHD comorbidity. The observed risks were similar among men and women with TS/CTD.
The outcomes were relatively rare, with only 237 of 6791 individuals (3.5%) with TS/CTD having a record of a cervical spine disorder during the 17-year follow-up period. However, these events are even rarer in the general population (2.2% of the matched controls). Considering that some of these disorders, such as cerebrovascular events and spine injuries with myelopathy, have the potential to cause severe long-term disabilities, our results suggest that cervical spine disorders are not just isolated events in TS/CTD but a phenomenon that requires awareness and vigilance. Our survival analyses (for the subcohort that was followed up from age 3 years) showed that the exposed and unexposed cohorts separated early in midchildhood (nonoverlapping confidence intervals). It can take many years for TS/CTD to be detected and diagnosed28; thus, delays in the implementation of evidence-based treatment for tics could have dire consequences for individuals with severe motor tics affecting the neck region.
Vascular cervical spine disorders in particular were very rare (less than 1% of the TS/CTD cohort) but are of particular concern. Cerebral arterial dissection with ischemic stroke in children is rare but potentially fatal.31 In a 2001 review, 21% of children died following dissection of the cerebral artery with a resulting ischemic stroke, and only 37% of the survivors experienced a full recovery.31 Physicians working in pediatric neurology settings should be aware that severe neck tics could be a risk factor for such medical complications.
Individuals with TS/CTD presenting with severe repetitive neck extension tics (so-called whiplash tics) and/or those who are in need of emergency care or even hospitalization (referred to as malignant TS)32 should be monitored in collaboration with movement disorder experts because they will likely require highly specialized interventions other than conventional behavioral therapy or medication, such as botulinum toxin injections, spinal surgery, or deep brain stimulation in selected cases.32,33
Some strengths of this study are the uniquely large cohort of individuals with TS/CTD and the 17 years of follow-up. The use of the Swedish registers, with national coverage and prospective data collection, minimized the risk of selection and recall bias.
The study has some limitations that are inherent to register-based epidemiological studies.34 The Swedish ICD-10 codes for TS/CTD are highly valid and reliable,27 but misdiagnosis of some cases is hard to avoid (for example, it is difficult to distinguish idiopathic from functional tics based on patient records alone). Thus, our results are only as valid as the original diagnoses given by the hundreds of clinicians involved in the care of patients. Importantly, our results may not generalize to milder or transient forms of tic disorders that do not require or seek contact with specialist services. We assumed that the observed cervical spine disorders are a consequence of repeated motor tics involving the head and neck region because the hazards overall remained unchanged when strictly controlling for a range of other known causes of neck disorders. While it is possible that there are other factors that were not accounted for, such as deliberate repetitive head banging,35 our results are consistent with the high prevalence of tics involving the neck region,2 previous case reports of serious cervical spine disorders in TS/CTD,4-11 and survey data reporting physical pain and even injuries directly related to motor tics.36 Surveillance bias is another potential threat to the validity of our results, although the outcomes are generally severe and unlikely to go unnoticed by the Swedish universal health care system, particularly vascular disorders. We did not have information on tic topography or severity, but analyses limiting the cohort of individuals with TS/CTD to those with more chronic presentations (a proxy of greater tic severity or persistence) resulted in stronger associations compared with the main analyses, which adds to the plausibility of the findings.
Individuals with TS/CTD diagnosed in specialist settings are at increased risk of cervical spine disorders, some of which can be potentially serious and be associated with long-term disability. These disorders are relatively rare but may appear early in midchildhood, highlighting the importance of increasing awareness of a potentially serious complication of severe motor tics. Early diagnosis and management of TS/CTD and close monitoring of at-risk individuals are warranted.
Corresponding Author: Josef Isung, MD, PhD, Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, Gävlegatan 22B, 113 30 Stockholm, Sweden (firstname.lastname@example.org).
Accepted for Publication: June 10, 2021.
Published Online: August 23, 2021. doi:10.1001/jamaneurol.2021.2798
Author Contributions: Dr Isomura had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Isung, Isomura, Sidorchuk, Fernández de la Cruz, Mataix-Cols.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Isung, Isomura.
Critical revision of the manuscript for important intellectual content: Isung, Larsson, Sidorchuk, Fernández de la Cruz, Mataix-Cols.
Statistical analysis: Isomura.
Administrative, technical, or material support: Mataix-Cols.
Supervision: Larsson, Mataix-Cols.
Conflict of Interest Disclosures: Dr Larsson reported grants from Shire/Takeda and personal fees from Evolan outside the submitted work. Dr Fernández de la Cruz reported grants from Swedish Research Council for Health, Working Life and Welfare (FORTE), Region Stockholm, Åke Wibergs stiftelse, and Karolinska Institutet and personal fees from UpToDate outside the submitted work. Dr Mataix-Cols reported personal fees from UpToDate and Elsevier outside the submitted work. No other disclosures were reported.
Additional Contributions: The authors wish to thank the anonymous patient in Figure 1 and James F. Leckman, MD, PhD, from the Child Study Center, Yale School of Medicine, and Robert K. Fulbright, MD, from the Department of Radiology and Biomedical Imaging, Yale School of Medicine for kindly providing the magnetic resonance imaging pictures to illustrate cervical lesions caused by severe motor tics. Compensation was not received.