Dystonia : emerging concepts in pathophysiology.
The essential pathophysiological feature of dystonia is co-contraction of antagonistic muscles. This may be due to derangement of the spinal cord or cortical mechanism. In the cord, there is disruption of the normal reciprocal inhibition of antagonists during agonist contraction. This decreased reciprocal inhibition is due to reduced presynaptic inhibition of muscle afferent input to the inhibitory interneuron. The reduced presynaptic inhibition may in turn be either due to defective suprasegmental control or to changes in the tonic afferent input to the interneuron from cutaneous and muscle afferents. Alternatively, genesis of dystonia may entirely be a cortical mechanism. Overactivity of the premotor cortices, which receive projections from basal ganglia via ventral thalamus, could result in dystonia by abnormal activation of cortical motor neurons. This may again be due to a dopaminergic dysfunction of basal ganglia.
Dystonia is defined as a syndrome of sustained muscle contractions frequently causing twisting and repetitive movements or abnormal postures. Dystonic movements are almost always aggravated during voluntary movement (action dystonia). Idiopathic dystonia commonly begins with a specific action i.e. action dystonia. The abnormal movements appear with a specific action and are not present at rest. As the dystonic condition progresses, less specific action of the affected leg may activate dystonia (e.g. when tapping the floor). With further evolution, action in other parts of the body can induce dystonic movements of involved leg (overflow). With further worsening, dystonia is present even at rest. Eventually the limb goes into sustained posturing. In addition to the specific body part showing progression, dystonia often spreads to involve other parts of the body. Most often the spread is to contiguous body parts. As a general rule, the younger the patient at onset, the more likely it is that dystonia will spread. Dystonic movements tend to increase with fatigue, stress and emotion; they tend to be suppressed with relaxation, hypnosis and sleep. One characteristic and unique feature is that they can be decreased by tactile or proprioceptive sensory tricks. Thus, touching the involved or adjacent body part can often reduce muscle contraction. The pathophysiology of dystonia is largely unknown. The concept that basal ganglia disease is exclusively responsible for genesis of dystonia is no longer tenable, since lesions in various parts of the nervous system including that of peripheral nerves can cause dystonia.
In this article, an attempt has been made to postulate a reasonable explanation for the pathophysiology of dystonia based on various studies. Even though it is well known that dystonia results from co-contraction of antagonistic muscles, it will be interesting to find out what causes this co-contraction, why it is precipitated by certain task specific action and why it is relieved by stimulation of certain trigger zones.
PATHOPHYSIOLOGY OF DYSTONIA
Pathophysiologically, the essential feature is an abnormal co-contraction of agonists and antagonists, precipitated by certain task specific action and relieved by certain sensory tricks. There is also contraction of adjoining muscles (overflow) particularly during voluntary movement. The EMG in primary torsion dystonia shows co-contraction of antagonistic muscles with prolonged bursts and overflow to extraneous muscles. This abnormal co-contraction, the essential feature of dystonia, could theoretically result from a fault at the spinal cord level or at the cortical level.
At the cord level, there is abnormality of the normal reciprocal inhibition between agonist and antagonists in cases of dystonia. Studies have shown that the second longer phase of reciprocal inhibition is absent or very much prolonged in patients with dystonia. This decreased reciprocal inhibition is due to the reduced presynaptic inhibition of the muscle afferent input to the inhibitory interneuron. This reduced presynaptic inhibition could in turn be theoretically due to (a) defective descending control or (b) changes in the tonic afferent input to the interneuron from cutaneous and muscle afferents.
Intramuscular injection of botulinum toxin in dystonic patients has been found to improve the involuntary muscle activity and improve the reciprocal inhibition by increasing the presynaptic inhibition. This finding suggests that the toxin alters the spinal cord segmental motor system, by altering the tonic sensory inflow from the injected muscles. The fact that muscle afferents have a role in producing dystonic movements, is also exemplified by the work of Kayi et al, who studied the tonic vibration reflex in writer's cramp. A vibrator applied on the hand induced a dystonic posture and local injection of lidocaine into muscles markedly decreased dystonic movements. Grunewald et al studied involvement of muscle spindles in the pathophysiology of dystonia, as muscle spindles are involved in the sensation of position and movement of the body. Their experiments showed abnormal perception of motion, but not position, in dystonic subjects.
On the other hand, it is theoretically possible that co-contraction produced by alteration of reciprocal inhibition may be a cortical mechanism, this is by altering the descending control on the presynaptic inhibition. Inzelberg et al studied the kinematic properties of upper limb trajectories in torsion dystonia and found that movements in dystonic patients were slow and more variable; and movement trajectories had a prolonged decelerative component, which became even longer without visual feedback from the limb. These findings suggest that dystonic patients have defects in central motor mechanism. Van der Kamp et al observed that peak amplitude of movement related electroencephalogram potential is reduced in patients with arm dystonia. Similarly, abnormalities in cortical preparatory process for voluntary muscle relaxation or motor inhibition was found in focal hand dystonia by Yazawa et al, who studied the Bereitschafts potentials preceding voluntary muscle contraction and relaxation.
Inhibitory control of basal ganglia output to thalamocortical projection plays an important role in normal cortical activity in the current model of the basal ganglia motor circuit. Excess or collapse of the basal ganglia output can explain hypokinetic and hyperkinetic movement disorders of basal ganglia origin. An abundance of evidence indicates that parkinsonian akinesia results from hyperactivity of the basal ganglia output. Reversal of akinesia by lesions of the internal division of the globus pallidus (GPi) or its excitatory source, the subthalamic nucleus, is consistent with this pathological schema. Ballism associated with subthalamic lesions, and dopa-induced dyskinesia are regarded as hyperkinetic disorders resulting from suppressed subthalamopallidal projection. Decreased firing rate in GPi was reported in both disorders. However, pallidotomy has recently been postulated to abolish both ballism and dopa-induced dyskinesia. A possible mechanism for the effect of GPi destruction in these hyperkinetic disorders may be blockade of the generation or conduction of phasic neuronal activities driving choreic movements. Symptomatologically, dystonia has aspects of both hypokinetic and hyperkinetic disorders. Overactivity of the premotor cortices, which receive projections from the basal ganglia via the ventral thalamus, was found both at rest and on movement in idiopathic dystonia. This abnormal cortical activity may arise from underactivity of basal ganglia output. Vitek et al did microelectrode recording in basal ganglia of 3 patients with generalised dystonia and found that the mean discharge rates of neurons in both segments of the pallidum were considerably lower. In addition the pattern of spontaneous neuronal activity was highly irregular, occurring as intermittent grouped discharges separated by periods of pauses. This finding of decreased neuronal discharge rates of the pallidal neurons suggests that the dystonia is physiologically a hyperkinetic movement disorder. However, the amelioration of dystonia with pallidotomy suggests a complex pathomechanism of the pallidothalamic system in dystonia.
The technique of transcranial stimulation provided evidence of changes in the excitability of motor areas. Mavroudakis et al showed that the percentage increase in the area underneath the muscle action potential trace evoked by brain stimulation is greater in patients with dystonia than in normal individuals. This suggests increased cortical motor excitability in dystonia. Similarly, increased excitability of the corticospinal motor system was found in writer's cramp, by noting an increase in the corticomotor output to the affected hand during repetitive transcranial magnetic stimulation.
In idiopathic dystonia, an inappropriate overactivity of striato-frontal projection and impaired activity of motor executive areas has been found. Using PET regional blood flow studies, it has been shown that there is overactivity in the contralateral premotor cortex, rostral sensory motor area (SMA), area 8, anterior cingulate area 32, ipsilateral dorsolateral prefrontal cortex and bilateral lentiform nucleus. Under activity has been found in the caudal SMA, bilateral sensory motor cortex, posterior cingulate and mesial parietal cortex., These findings may explain the coexistence of dystonic postures and bradykinesia in these patients. The over-activity of premotor cortex and lentiform nucleus suggest that hyperkinetic movements observed in dystonia is due to overactivity of the premotor cortex, which in turn is linked to lentiform nucleus hyperfunction. The lentiform nucleus hyperfunction leads to structural disruption of basal ganglia inhibitory control resulting in overactivity of the premotor cortex. Thus primary dystonia results from a functional disturbances of basal ganglia, particularly in the striatal control of globus pallidus. This causes altered thalamic control of cortical planning and executive areas and abnormal regulation of brain stem and spinal cord inhibitory interneuronal mechanism. However, decreased activation of premotor cortex was seen in some studies in certain task specific dystonias like writer's cramp. Ibanez et al studied regional blood flow in patients using PET to identify regions of malfunction. They noted decreased activation of premotor cortex during writing and decreased correlation between premotor cortical regions and putamen. This suggests a dysfunction of premotor cortical net work in patients with writer's cramp, possibly arising from basal ganglia. Thus the dysfunction is compatible with loss of inhibition during generation of motor commands, which in turn could be responsible for the dystonic movements.
Reduced activity of the cortical inhibitory circuits in patients with dystonia has been found by studying the cortical excitability by magnetic stimulation. This reduced function of the cortical inhibitory circuits may be one factor contributing to excessive and inappropriate muscle contraction, which occurs during tasks in dystonic patients. Rona et al studied the cortical inhibitory mechanisms with the technique of paired transcranial magnetic stimulation in patients with different types of dystonias. They propose that the alterations observed in patients with dystonia are the result of impaired feedback from the basal ganglia to motor cortical areas, with the ultimate effect of a flattening of the excitability curve of the cortical motorneuron pool during voluntary muscle contraction.
What are the evidences that basal ganglia is responsible for the overactivity of the premotor cortex observed in dystonia? Innumerable clinical case studies have shown that the most common site of affection in symptomatic dystonia is basal ganglia, that too mainly the putamen. Evidence from PET studies suggests that the loss of cortical inhibition in primary torsion dystonia (PTD) might be caused by abnormal basal ganglia circuitry with an imbalance between the direct and indirect pathways. Using network analyses of PET, Eidelberg et al found relative bilateral increases in lateral frontal and paracentral cortices, associated with relative covariate hypermetabolism of the contralateral lentiform, pons, and midbrain. In dystonia, in contrast with Parkinson's disease, lentiform and thalamic metabolism were dissociated, suggesting excess activity of the direct putamino-pallidal pathway. Magyar-Lehmann et al confirmed the pathophysiologic role of the putamino-pallidum pathway in their study of patients with torticollis. They found increased bilateral lentiform glucose metabolism that did not correlate with disease severity, duration or side of chin turning.
An alternative view of brain networks in primary dystonia based on both PET studies using spiperone binding and animal models using MPTP has been presented., It was found that patients with primary focal dystonia have decreased putaminal binding of spiperone, a radioligand that binds predominantly to D2-like receptors. MPTP treated baboons developed transient hemidystonia prior to hemiparkinsonism. Dystonia corresponded temporally with decreased striatal dopamine content and transient decrease in D2-like receptor number. It was suggested that dystonia occurs after preferential decrease in D2 mediated inhibition of striatopallidal inhibitory neurons of the indirect pathway. Under this scheme, medial pallidal inhibitory output is increased to yield a loss of surround inhibition and overflow of the motor command. Unlike parkinsonism, however, the direct pathway is not affected so that selection of the motor action is not impeded. There is evidence to suggest a disturbance of dopaminergic function in the pathophysiology of early onset torsion dystonia. Recent work has revealed that the causative mutation in most cases is deletion of a glutamate residue from the carboxy terminal of Torsin A, a 332 aminoacid protein encoded by the DYT 1 gene. This protein has similarity to the family of heat shock protein and Clp proteases. However, its function and putative role in the nervous system remains uncertain. Angood et al have mapped the expression of the DYT1 gene in normal human postmortem brain. DYT1 mRNA is highly enriched in the dopaminergic neurons of substantia nigra of pars compacta. The prominent expression of the DYT1 gene within the pars compacta of substantia nigra which provides dopaminergic innervation to the basal ganglia implicates a disturbance of dopaminergic function in the pathophysiology of early onset dystonia. This shows a disturbance of dopaminergic function in the pathophysiology of early onset torsion dystonia.
With the above concepts in mind, let us try to answer the important pathophysiogical mechanisms in dystonia.
1. Why there is abnormal co- contraction? : This is due to the down regulation of the reciprocal inhibition due to changes in the presynaptic inhibition, which again may be either centrally mediated or due to affection of the peripheral afferent input ( either cutaneous or muscle).
2. Why dystonia is task-specific? : The abnormal co-contraction, the major determinant of dystonia, is not the sole abnormality in dystonia. There is an element of hypokinesia in patients with dystonia due to impaired activity of the motor executive areas. Underactivity of caudal SMA, bilateral sensory motor cortex, posterior cingulate cortex and medial parietal cortex has been shown by PET studies. Thus there may be a motor programming deficit, which specifically affects certain action than others. This may explain the task specificity of dystonia.
3. Why dystonia is relieved by certain trigger point stimulation? : As already discussed, the cutaneous and muscle afferents have a key role in modulating the reciprocal inhibition mechanism of the cord, in addition to the descending motor control. Botulinum toxin injection is found to improve the reciprocal inhibition by altering tonic sensory inflow through muscle afferents.4 Similarly, it is highly likely that cutaneous stimulation in certain points may modulate the presynaptic inhibition so as to improve the reciprocal inhibition and thereby alleviating the co-contraction between agonists and antagonists.
4. What is the mechanism of occupational dystonia? : It is possible that repetitive strain may result in alteration or degradation of the so-called `motor engrams' in the sensory motor cortical area for a particular task. This is supported by a recent study demonstrating the development of dystonia after repetitive strain in primates. Physiological studies in patients with such occupational dystonia show deficiencies in spinal reciprocal inhibition and abnormalities of central sensory processing and motor output that may be related to reduced cortical inhibition. Recent studies in primates support the notion that repetitive motions can induce plasticity changes in sensory cortex leading to degradation of topographic representation of the hand and raise the possibility that sensory training may be beneficial. Subjects with writer's cramp were found to have impaired capacity to integrate sensory information in the motor programming during precision tasks, despite normal sensibility.
Dystonia can also develop following a temporary disability to a limb, e.g. in wrist drop. It has been suggested that a dystonic movement disorder may be caused by persistent motor strategies adopted to compensate for a temporary disability. It is well known that lesions other than those of basal ganglia can also cause dystonia. Maximum number of cases of secondary dystonia are due to affection of basal ganglia, especially lentiform nucleus, caudate nucleus and thalamus.,, It is noteworthy that basal ganglia lesion can also produce focal dystonia e.g. basal ganglia lesions can produce cervical dystonia,,,, and blepharospasm. Diffuse or focal dystonia has been reported due to lesions of mesencephalon, cerebellar lesion pontine lesion due to central pontine myelinolysis without extra pontine myelinolysis,, medullary lesion and spinal cord lesion., Focal dystonia with or without associated causalgia can also occur due to peripheral nerve lesion. Generalized dystonia has been reported in homocystinuria, HIV infection, after methanol ingestion and as a remote effect of carcinoma.
To sum up, one should envisage dystonia as a phenomenon which is not specific to particular site of lesion. One may postulate two types of dystonias pathophysiologically, accepting that the major determinant of dystonia is abnormal co-contracion of muscles: (1) a central one, demonstrating task specificity having variation with stress, fatigue etc. (2) a peripheral one producing only co-contraction resulting in abnormal postures without task specificity or overflow.