Posttraumatic movement disorders: A clinical and imaging profile of 30 patients
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.263212
Source of Support: None, Conflict of Interest: None
Keywords: Dystonia, head injury, neuroimaging, posttraumatic movement disorders, trauma, tremor
Traumatic brain injury is a major cause of death and disability, resulting in major health and socioeconomic problems due to the increase in motor-vehicle use.,,, Traumatic brain injury predisposes long-term survivors to various neurodegenerative disorders such as Alzheimer's disease, Parkinson's disease, amyotrophic lateral sclerosis,,,, and possibly, frontotemporal dementia. Long-term sequelae following head injury (HI) results in significant impairment of various modalities affecting: (i) cognition—executive dysfunction, memory impairment, dementia; (ii) behavioral changes—mood disturbances, suicidal tendencies, depression, apathy, impulsivity, aggression; (iii) neurological dysfunction—oculomotor disturbances, gait disturbances, ataxia, speech disturbances, motor neuron disease, and movement disorders (MDs). These have been extensively studied in people indulging in contact sports, and is especially seen among football players, boxers, and soccer players where it is termed as chronic traumatic encephalopathy, dementia pugilistica, or punch drunk syndrome (in association with boxing). The pathophysiological basis of brain injury in combative sports, military, assaults, falls, and motor vehicle accidents remains the same.,
Majority of the researches have focused on the cognitive and behavioral changes of the long-term sequelae following HI. Studies on the neurological motor impairment, in particular MDs, are very few.,, Various MDs have been reported following cerebral trauma  due to disruption of the extrapyramidal pathway. The prevalence of posttraumatic movement disorders (PTMDs) has been variably reported, ranging from 13 to 66%., The objectives of our study were to characterize the pattern (phenomenology) of MDs in patients with PTMD and to correlate these with the sites of lesion on brain imaging.
Subjects were recruited from the outpatient and inpatient services from the Department of Neurology, National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore, Karnataka, India, which is a tertiary neurology referral center. Patients presenting with MD, with no positive family history and having an antecedent traumatic HI were classified as PTMD. This was a prospective study carried out on PTMD patients, examined by a MD specialist to characterize the phenomenology and subsequently followed up. Of these patients, 14 formed a subgroup of a large (103) population of secondary MDs which have been published previously. All the patients underwent a detailed neurological evaluation, including history and examination, to rule out other causes of MDs. The characterization of the MD was performed by a MD specialist. The demographic profile, duration of illness, latency of development of the MD after trauma, nature of trauma and its severity, and medications received were noted. MDs were classified based on their semiology as Parkinsonism More Details (PAR), dystonia (DYS), dystonia with choreoathetosis (DYS+), hemiballismus (HMB), tremor (TRM), tremor with dystonia (TrD), myoclonus (MYO), and mixed MD (when a patient had a combination of more than one MD).
The severity of the trauma was graded as follows: (i) Mild—if the patient did not develop significant neurological disturbances and improved without intensive care unit (ICU) admission, and the hospital stay was <3 days, with the Glasgow coma score (GCS) of 14–15; (ii) Moderate—when the patient required hospitalization with mild to moderate neurological disturbances for >3 days without ICU or ventilator support and with the GCS of 9–13; (iii) Severe—when the patient had a significant decrease in the GCS score <8, requiring ICU care and ventilator support with or without a tracheostomy.
Imaging of the brain either with computed tomography (CT) or magnetic resonance imaging (MRI) was performed depending on the affordability and suitability of use of general anesthesia/sedation, particularly in children. The imaging findings were classified as: (i) isolated discrete lesions confined to a single anatomical site of the basal ganglia, thalamus, subthalamus, or the brainstem, (ii) diffuse lesions involving the cortical gray and/or, white matter, and (iii) a combination of (i) and (ii). The anatomical sites identified on MRI were caudate (C), putamen (P), globus pallidus (GP), striatum (STR), lentiform (L), striatopallidum (SP), thalamus (T), subthalamus (ST), pons (PO), midbrain (MB), gray matter (GM), and white matter (WM). Correlations were done between the type of PTMD and the sites of lesion. Depending on the clinical suspicion, type of MDs, and imaging characteristics, some of the patients were evaluated for the presence of Kayser–Fleischer (KF) ring, copper studies, and biochemical parameters to rule out Wilson's disease or any other MD.
The study was approved by the Ethics Committee of the institute, and all patients gave written informed consent before their participation in the study.
Thirty patients (28 men, 2 women) with PTMD were recruited for the study. The mean age, age at onset, and duration of illness [Table 1] were 32.6 ± 16.4, 29.1 ± 16.0, and 3.6 ± 5.0 years, respectively. The right upper limb was the initial body part affected in 36.7% of the patients followed by generalized involvement (20%), head (16.7%), left upper limb (10%), face (6.7%), head and left upper limb (3.3%), left lower limb (3.3%), and left upper and lower limb (3.3%). Severe and moderate trauma were observed in an equal number (46.7%) of the patients respectively, whereas mild trauma was observed in 6.7% of the patients. MDs developed predominantly after road traffic accidents in 83.3% patients, while the rest sustained an accidental fall from a height (13.3%) or the piercing of glass onto the face (3.3%). The latency (in years) was short for TrD (0.6 ± 0.8), DYS+ (0.7 ± 0.5), myoclonus (0.1), and mixed MD (0.4 ± 0.4), while it was longer in pure dystonia (1.6 ± 1.8) and pure tremor (2.1 ± 2.0). The longest latency of 2.4 ± 4.3 years was observed in patients with parkinsonism [Table 2].
Type of MDs[Figure 1]: Tremor (TRM or TrD) was the most common MD seen in 44.7% of the patients, followed by dystonia (DYS or DYS+) in 24.1%. The other MDs were parkinsonism (17.2%), mixed MD (10.3%), and myoclonus (3.4%). Almost all types of MDs were observed in the cohort except hemiballismus, pure chorea, and tics. There was a difference in the semiology of the MDs among the different age groups. Tremor with or without dystonia (TRM, TrD), pure dystonia (DYS), and mixed MD were found in all age groups, whereas dystonia with choreoathetosis (DYS+) was observed only in children <10 years. Adults above 40 years had either parkinsonism or myoclonus as the predominant MD.
CT scan of the brain: CT brain was the sole imaging modality performed in seven patients. MRI could not be done in view of either financial constraints or the inability to undergo general anesthesia for MRI purpose. The imaging findings with the type of MD were: (i) diffuse WM involvement without affecting the basal ganglia seen in five patients. Of them, two patients each had tremor and mixed MD, and one had parkinsonism; (ii) discrete lesion in the putamen in a patient with dystonia with choreoathetosis; and, (iii) normal imaging in one patient with parkinsonism.
MRI of the brain [Figure 2]: MRI brain was performed in 23 patients. The imaging findings were:
All patients received medications, which were titrated and gradually increased up to the maximum dosage or when side effects prevented dose escalation.
Imaging correlation of the various MD
Tremor (TRM): Eight patients had pure tremor, which was the most common MD. Their characteristics are described in [Table 2]. Right upper limb was the most common body part affected (50%), followed by head (37.5%) and generalized (12.5%) involvement. Rest and action tremor was the predominant type of tremor found in 75%, followed by pure action tremor in 25%. MRI brain revealed WM involvement as the most common site of abnormality (50%), followed by thalamus and brainstem combination (25%), putamen (12.5%), and subthalamus (12.5%).
Tremor with dystonia (TrD): Five patients had tremor with dystonia in combination. All of them had rest and action tremor in combination with dystonia. WM involvement was the most common site of abnormality seen in 60%, followed by thalamus and brainstem combination in 40% patients.
Patients with tremor (either pure tremor—TRM, or in combination with dystonia—TrD) had imaging abnormalities involving the WM in nearly two-third (69.2%) of the patients, followed by thalamus in one-third (30.8%) of the patients.
Dystonia (DYS): Seven patients had dystonia in the form of pure dystonia (4 patients) or in combination with choreoathetosis (3 patients) [Table 2]. Focal dystonia was the most common type of dystonia (71.4%), followed by generalized involvement (28.6%). Imaging was normal in one patient followed by involvement of the putamen (2), WM (2), pons (1), and thalamus (1). The imaging correlation was not possible due to the lesser number of patients.
Parkinsonism (PAR): Five patients had parkinsonism and manifested secondary MDs. The age and age at onset were predominantly in the fifth decade, compared to other MDs, where the age and age at onset varied from the first to second decade. Three (60%) patients had onset of symptoms in the right upper limb followed by left upper limb, and there was generalized involvement in one (20%) patient each. Latency of onset of parkinsonism after trauma was the longest (2.4 ± 4.3 years) compared with other MDs. MRI was abnormal in three patients, who showed diffuse WM involvement. CT scan in a patient with mild HI was normal. As the CT scan is not sensitive enough to pick up abnormalities, it is possible that subtle abnormalities were not picked up. Another patient with parkinsonism, who also had a history of mild HI, had a normal MRI. In this patient, MRI had the usual sequences of T2, T1, and FLAIR images, without gradient echo sequence, which could have detected diffuse axonal injuries.
Myoclonus (MYO): One patient with facial myoclonus following road traffic accident had lesions of bilateral globus pallidus on MRI.
Mixed MD (MIX): Four patients had mixed (a combination of any of the above categories of MD) MDs, as described later. One patient had basal ganglia, thalamus, and brainstem involvement, while rest of the three patients had diffuse WM involvement without involvement of the deep GM. Patients with lesions in the putamen, thalamus, and midbrain had a combination of parkinsonism, dystonia, and myoclonus. Three patients had diffuse WM involvement. The mixed MD was: (i) a combination of dystonia with parkinsonism, (ii) head tremor with dystonia, and (iii) tremor and action myoclonus in the upper limbs.
There were no patients with isolated chorea, tics, or hemiballismus. The symptoms in patients with parkinsonism, myoclonus, and mixed MDs could not be correlated with the imaging findings in view of the lesser number of patients.
Age-wise correlation of PTMD: Some MDs had a predilection for a certain age.
The most common MD in our cohort was tremor (44.7%) followed by dystonia (24.1%). One-third of the patients had parkinsonism, myoclonus, and mixed MD. In a consecutive analysis of 398 patients  with head trauma, 22.6% patients developed a MD. Tremor (84%) and dystonia (18%) were the common MDs reported by them. Many other studies have also reported tremor as the predominant MD  secondary to HI. Iwadate et al., reported eight patients with posttraumatic tremor. All of them had sustained severe HI with the Glasgow Coma Scale (GCS) score <8. CT brain showed diffuse changes in the cerebral hemispheres and the midbrain. Carvallo et al., described an 18-year old boy with left upper limb dystonic tremor secondary to HI. The MRI of the brain showed a lesion in the right thalamus. The patient responded to deep brain stimulation (DBS) of the globus pallidus after failure of conservative treatment.
These variations in the latency of development of MDs and the type of MDs seen have been also reported in other types of secondary MDs, such as postvascular MD., Pure tremor, dystonia, and parkinsonism have the longest delay in onset after a vascular insult , compared with other MDs. This type of delayed onset of parkinsonism and tremor has been similar to our study. The reasons for the delayed onset of MD could be (i) the time required for recovery of motor function and the consequent development of the pathological neuronal circuit; or, (ii) the time required for changes in neuronal synaptic activities.
In our cohort, patients with tremor had imaging abnormalities involving the WM in nearly two-third patients, followed by involvement of the thalamus in the rest. Raina et al., described 29 patients with Holmes tremor, among whom 17.24% had tremor secondary to traumatic brain injury. Of them, 60% had lesions in the thalamus and midbrain, while 40% had subcortical WM involvement. Apart from lesions in the thalamus and the Guillain-Mollaret's triangle, lesions in the subcortical WM may result in tremor due to disruption of the cerebello-thalamo-cortical pathways.
The pathophysiology of PTMD is poorly understood. Diffuse axonal injury has been implicated to be the primary mechanism in traumatic brain injury, which results in alteration of the neurophysiological processes and accumulation of abnormal protein aggregates., There are both primary and secondary injury mechanisms implicated after traumatic brain injury.,,, The primary damage involves focal contusions, particularly of the basal ganglia and its pathways (direct, indirect striatopallidal and lenticulothalamic, cerebellothalamic), or due to the mechanical effects of shearing, stretching, and tearing of the neurons, glia, axons, and blood vessels. Secondary injury results from delayed metabolic, neurochemical, and cellular changes. Other contributory factors include hypoxia, hypotension, and increased intracranial pressure that contribute to the extent of the lesion and the subsequent development of MDs. A positron emission tomography study demonstrated an increased microglial activation in traumatic brain injury that triggers a chronic inflammatory response in subcortical regions., There is disruption of the blood–brain barrier, resulting in an exposure of the brain to neurotoxins, cytokines, tumor necrosis factor-α, and other neuroinflammatory agents that induce neuroinflammation. Increased excitotoxicity due to increased glutamate levels results in neuronal cell death. α-Synuclein levels have been found to be elevated in the cerebrospinal fluid following HI, which results in neurological dysfunction. The other pathophysiological mechanisms implicated are mitochondrial dysfunction, calcium overload, increase in free-radicals that results in neuronal cell death, inflammation, and neurodegeneration. Thus, traumatic brain injury induces the inflammatory cascade with the resultant accumulation of tau and α-synuclein, that ultimately result in oxidative stress, neuroinflammation, and neurodegeneration.
Our study had a few limitations. The details of the insult, the GCS at the time of trauma, the type of HI, the additional effects of hypoxia, and the effects of seizures, sepsis, as well as operative procedures were not available in most of the cases; and, the precise information regarding the onset and evolution of MDs were sometimes not available. This is because such information is often given by the patients or their relatives based on their memory; recall bias may have not yielded adequate information and many of them did not have documents detailing the episode of HI. The correlation of PTMD with the anatomical site of lesions is difficult, as most of the MDs are mixed MDs and majority of the patients have widespread lesions in the brain following trauma. Moreover, most of our patients were from lower socioeconomic status, had only a CT scan and an MRI could not been done in them. It is difficult to discern which lesion was responsible for the MD, except in patients with tremor; they had predominant lesions diffusely in the WM and thalamus. The other MDs could not be anatomically correlated in view of lesser number of cases.
In conclusion, traumatic injuries have long-term serious adverse effects on the central nervous system. The severity of HI, age at onset of trauma, anatomical sites of the lesion, and the inherent genetic susceptibility have synergistic actions in the development of PTMDs. Future studies are required to study the different mechanisms including the genetic risk factors, the severity of HI, and the role of diffuse axonal injury visualised on the gradient sequences of MRI in the genesis of PTMDs. Neurosurgical procedures, particularly lesioning surgery, may play a role in the treatment, along with medications, in resource-poor countries where deep brain stimulation may not be affordable.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2]