Dystonias of the Neck: Clinico-Radiologic Correlations

Since the clinical description of the four basic dystonic head/neck movements in 1953 by Hassler3 there have been no publications concerning the further differentiation of these forms of CD. Also, it has not been attempted to define muscles affected by dystonia with the help of imaging procedures. A new book called “Imaging of Movement Disorders” does not contain a single image of dystonic muscles in its eleven chapters4.


Introduction
Idiopathic cervical dystonias (CD), the most prevalent form of defined dystonias in adults, are characterized by involuntary, abnormal movements of the head, and/or by the involuntary adoption of various head postures. With the exception of a few patients, the course of the condition is not progressive 1,2 .
Since the clinical description of the four basic dystonic head/neck movements in 1953 by Hassler 3 there have been no publications concerning the further differentiation of these forms of CD. Also, it has not been attempted to define muscles affected by dystonia with the help of imaging procedures. A new book called "Imaging of Movement Disorders" does not contain a single image of dystonic muscles in its eleven chapters 4 .
Since botulinum toxin has become available for the symptomatic treatment of CD, an exact differentiation between dystonic and non-dystonic muscles is very important.
Botulinum toxin is an established treatment for CD 5 and has gained widespread acceptance since its first use in 1985 6 , based on a large body of evidence 7,8 . Response rates of approximately 80% have been achieved in patients with CD in open and double-blind studies [9][10][11] . Furthermore, treatment is generally well tolerated even following long-term administration 12,13 and the risk of adverse events may also be minimized using well-defined injection procedures.
Nevertheless, unsatisfactory treatment outcomes have been observed in patients with CD 14 . These may generally be attributed to incorrect muscles selected for treatment administration and suboptimal dosing and distribution to the affected muscles. Moreover, in some cases, poor treatment outcomes may reflect failure to treat all the affected muscles or the influence of toxins on healthy muscles. Although imaging 15 , electromyography 16 and modifications of injection technique 17 have all been used to improve treatment outcomes, the influence of these methods on treatment outcome requires further clarification 17 .
In patients with CD, dystonic nodding movements, rotatory movements and lateral flexion can originate from the head joints and from the cervical spine. The combination of nodding movements of the upper head joint with rotatory movements of the lower head joint can result in movements in all three spatial planes. Dystonic movement disorders occur in the majority of patients in two planes (50%), less commonly in one plane (approximately 35%) and, rarely, in three planes (11%) 18 .
The localization of dystonic movements is dependent on the origin and insertion of the muscles involved in dystonia. As not all potentially affected muscles are dystonic in each form of CD, clinical decisions must be based on findings from the clinical examination, including palpation, electromyographic analysis and, in cases of ambiguity, imaging (computed tomography [CT] and magnetic resonance imaging [MRI]). Although dystonic head postures are traditionally classified according to four different movement planes (rotation, lateral flexion, forward/backward flexion and sagittal shift), our clinical experience suggests that this system of classification is not sufficient for the accurate identification of muscles that should be targeted for botulinum toxin treatment. Thus, we conducted a large, non-intervention study using clinical examination, CT and MRI, with the overall aim of elucidating a more precise method of differentiating forms of head and neck postures in patients with CD. Initial results of the MRI analyses have been published previously 19 .

Setting and study population
Patients treated in our specialist movement disorders clinic with documented primary CDestablished by clinical examination, electromyography, MRI of the neurocranium, laboratory tests and data relating to the patient's medical history -were eligible for inclusion. Patients were enrolled from 2007 to 2009. This sample represents all the patients in our clinic in this time period, with the exception of pregnant patients (n=1) or those who refused to participate (n=1). Written informed consent was given by all included patients and the study was approved by the local ethics committee.

Study assessments
Characterization of the different forms of the abnormal head (-caput) and neck (-collis) postures in patients with CD was conducted by clinical evaluation and radiological examination. For complicated or unclear cases, CT and/or MRI was also used.
Radiological examinations involved: 1) CT scanning of the soft tissues of the neck as single layers at the section level of cervical vertebrae 3 and 7; and 2) MRI of the cervical spine and the soft parts of the neck (with T1-and T2-weighting in 2 mm slices), and the deep neck muscles (examined at an angle with T1 weighting). During the examination, patients were requested to assume a relaxed head or neck posture. For patients with lateral flexion, electronic 'straightening' of the tomograph was performed to enable muscles to be visualized at the same height on both sides of the image (Figure 1). For evaluation of MRI data, images from 50 patients who did not suffer from CD (prior condition documented as mild trauma or suspected radicular disorder) were used as retrospective controls.

Statistical analyses
Following CT and MRI, the relationships of the skull to the cervical spine and of the cervical spine to the thoracic spine were analyzed. CT images were also used to obtain measurements and shapes of muscles in the neck region, as clinically appropriate. Results of analyses for CT and MRI data for patients with CD are presented descriptively as percentages.

Results
Overall, 95 patients (55 female, 40 male) with established primary CD were included in the study. The patient population had a mean age of 48.5 years, with a mean age at disease onset of 41.6 years.
The four types of abnormal head posture were characterized further using clinical findings and MRI and CT imaging, and are summarized in Tables 2 and 3.   Table 3. Proposed subdivisions of cervical dystonia forms

Flexion
Clinical examination alone was sufficient to determine the variant of lateral flexion that was present in the majority of patients (Figure 2a and b). In a few cases, imaging was used to confirm the clinical decision (Figure 2c).
Clinical and imaging observations revealed that 19% of patients had lateral flexion that was located in the head joints, whereas flexion in the cervical spine or between the cervical spine and the thoracic spine was present in 22% of patients (Table 2). Most (59%) patients exhibited simultaneous lateral flexion of the head and of the cervical spine (Table 2).

Rotation
In analyses of head rotation, clinical differentiation between rotation in the head joints and the cervical spine region was frequently inaccurate (Figures 4 and 5). Guidance was provided by the position of the readily palpable incisura jugularis sterni with respect to the www.intechopen.com incisura thyreoidea superior (for example, in the case of rotation in the head joints, the notches are positioned directly above each other; Figure 4d). Results using CT-slices to compare the position of the vertebrae in both planes (C3 and C7) allowed more reliable differentiation between torticollis and torticaput. In the case of torticaput, rotation was observed only in the lower head joint (articulatio atlantoaxialis).

www.intechopen.com
Thus, in the C3 image, only the skull (easy to recognize by the lower jaw) is positioned in the rotated direction, whereas cervical vertebrae 3 and 7 are not rotated towards each other (Figure 4b and c). By contrast, the upper cervical vertebrae are rotated towards the lower cervical vertebrae in the presence of torticollis, and the third cervical vertebra points in the direction of the lower jaw (Figure 5b and c). Correspondingly, CT images demonstrated differences between the diameters of the muscles on each side, and therefore the muscles affected by dystonia.
a . b .
c. d.  Computer tomograph at C3 (the cervical vetebra is rotated towards C7 but not towards the skull).

Shift
Lateral shift was evident as a combination of lateral flexion of the cervical spine with flexion in the head joints to the opposite side (Figure 6a), whereas sagittal shift presented as a combination of forward flexion of the cervical spine with backward flexion in the head joints (Figure 6b). In contrast, backward shift developed from a combination of retrocollis and anterocaput (Figure 6c).

Muscle size
Measurement of maximum muscle diameters using MRI 15 revealed that, in addition to the large neck muscles, musculus obliquus capitis inferior was the only small muscle that was asymmetrical in the majority of patients (73%; Figure 7a). This muscle, when dystonic, induced rotation and backward flexion in the head joints. Therefore, separate analysis of this www.intechopen.com muscle is advised given the difficulty of localized treatment with botulinum toxin. For reasons of safety, treatment should only be performed if administration is monitored by CT (Figure 7b).
a. b.

Botulinum toxin targeting in CD based on clinic-radiologic correlates
This was a large study that was conducted to elucidate the characteristics of abnormal head and neck postures among patients with CD in order to maximize treatment outcomes with botulinum toxin. In contrast to the previously accepted classification of CD (four basic forms), our clinical and imaging findings support the differentiation of the disorder into 10 variations of posture and/or movement (Figure 8), which allows better delineation of the muscles involved in the particular form of CD ( Table 4).
Analyses of the prevalence of the characterized forms of CD in our study population revealed that, in the case of lateral flexion and rotation, the abnormal movement and/or posture involved only the head joints (latero-and/or torticaput) in approximately 20% of patients and only the region of the cervical spine (latero-and/or torticollis) in approximately a further 20% of patients. The remaining patients, approximately 60%, had both disorders, albeit with varying degrees of involvement of -caput and -collis. Thus, the incidence of these three forms represented a ratio of 1:1:3. A similar ratio of incidence was observed for forward and backward flexion forms involving the head joints (antero-and/or retrocaput) or the cervical spine (antero-and/or retrocollis) or both in our study population. Treatment strategies for the individual abnormal dystonic postures will differ according to the function of the muscles in the neck region. For example, in the event of lateral bending, dystonia of the muscles originating from, or inserted in, the head or the first cervical vertebra causes abnormal head posture only when normal alignment of the cervical spine is present (Figure 2). With the involvement of muscles that originate from, or are inserted in, the cervical spine, 'genuine' laterocollis occurs if the relationship of the head to the cervical spine is normal. If both muscle groups are dystonic in the same direction, both muscle groups must be treated. However, dosage must be evaluated for both groups, and final dose is dependent on which of the two groups is most affected. If both muscle groups are affected in opposite directions (i.e. laterocollis and laterocaput), this presents clinically as lateral shift ( Figure 6). In this study, sagittal backward shift, a rare presentation, was shown to be a complex cervical dystonia with involvement of the muscles for retrocollis and anterocaput. In contrast, forward sagittal shift, in most cases, was solely caused by bilateral dystonia of the musculus sternocleidomastoidei. Contrary to many earlier reports, bilateral tonus of the musculus sternocleidomastoidei does not cause anterocaput; the head is turned backwards in the sagittal plane and the upper cervical spine is pulled downwards, representing a forward shift ( Figure  6c). This was first described in 1895 20 , and has been confirmed more recently 21 .
The results of this study are intended to provide a more precise differentiation of the variants of CD and thus to maximize the effectiveness of treatment with botulinum toxin. For purposes of treatment, it is necessary to decide which of the muscle groups are primarily dystonic; it is often sufficient to treat only severely affected muscles 22 .
It is important to treat the most involved dystonic muscles, but it is just as important not to treat muscles that are not affected; a healthy muscle injected accidentally causes a significant change in the pattern of movement and posture and thus renders it difficult to select strategies for further treatment. If an injection plan is prepared on the basis of the clinical examination and consequently optimum treatment results are achieved (i.e. no complaints for several weeks), further diagnostic procedures are not required. If treatment results are inadequate, imaging procedures should also be used to modify the injection plan. www.intechopen.com

Recommendations for the differentiation and treatment of CD with botulinum toxin and conclusion
Based on our classification of CD, we propose the following recommendations for the treatment of CD with botulinum toxin: 1. To confirm lateral bending, it is usually possible to differentiate clinically between laterocollis and laterocaput. In addition, an analysis of the position of the incisura jugularis sterni towards the incisura thyreoidea superior (Figure 4d) is helpful. Should the diagnosis remain unclear, a simple anterior-posterior X-ray is normally sufficient to resolve the problem. In the event of a combination of laterocaput and laterocollis, the relationship of the angles between the thoracic spine and the cervical spine and/or between the cervical spine and the skull is adequate to determine the distribution of botulinum toxin dose between the two different muscle groups. 2. Lateral shift inevitably occurs when laterocollis occurs on one side and laterocaput on the other side. Thus, muscles attached to the cervical spine on the side of the shift, and the muscles attached to the skull on the opposing side, require treatment. 3. In the case of rotation, reliable clinical differentiation between torticollis and torticaput is difficult. The position of the larynx may be helpful. In the case of torticaput, it tends to remain in the middle, whereas in the presentation of torticollis it tends to be turned to the side. To confirm this form of CD, CT sections at planes C3 and C7 should always be obtained. By comparing the position of the vertebra in both planes, torticollis and torticaput can be reliably differentiated. Additionally, measurements of muscle diameters and shapes on both sides can be extracted from the CT images at C3 and C7, enabling capture of almost all of the muscles involved in CD. 4. An analysis of forward flexion (i.e. differentiation between anterocollis and anterocaput) can be performed by lateral examination of the angles between the cervical spine and the thoracic spine and/or between the cervical spine and the base of the skull. This also applies to the analysis of backward movement (the differentiation between retrocollis and retrocaput). 5. A forward sagittal shift (a combination of anterocollis and retrocaput) is generally caused by bilateral dystonic activity of the musculus sternocleidomastoidei. Conversely, a backward sagittal shift is a combination of anterocaput and retrocollis.
In considering our findings, a potential methodological limitation of note was that this study constituted neither a prospective nor a controlled design. However, by recruiting patients without using strict inclusion and exclusion criteria (with the exception of pregnancy and lack of consent), the observed results can be considered to be more representative of the real-life treatment population.
Botulinum toxin administration to the affected muscles is the treatment of choice for primary CD. In order to differentiate between the dystonia affecting muscles inserted in, or originating from, the cervical spine and/or muscles inserted in, or originating from, the skull, supplementation of the clinical examination with CT or MRI is advised. Given the involvement of different muscle groups, it is essential to differentiate between -collis andcaput forms of CD. We have proposed a classification system of 10 basic subtypes of abnormal dystonic positions and postures that we envisage will aid the selection of optimal treatment strategies with botulinum toxin.

Acknowledgments
Grateful thanks to Andrea Stenner, MD, Director of the Neurological Specialist Department of the MVZ 1 of the Paracelsus clinik, Zwickau, and Anke Jahn, MD, Member of staff of the Radiological group practice at the Paracelsus clinic, Zwickau, for their expert support and kind help. Thank you also to David Caird, PhD, Ipsen Pharma GmbH, Ettlingen, for help in preparation of the manuscript. Editorial assistance for the preparation of this manuscript was provided by Ogilvy Healthworld Medical Education; funding was provided by Ipsen Pharma GmbH . Dystonia has many facets, and among those, this book commences with the increasingly associated genes identified, including a construct on how biology interacts with the dystonia genesis. The clinical phenomenology of dystonia as approached in the book is interesting because, not only were the cervical, oromandibular/lingual/laryngeal, task-specific and secondary dystonias dealt with individually, but that the associated features such as parkinsonism, tremors and spasticity were also separately presented. Advances in dystonia management followed, and they ranged from dopaminergic therapy, chemodenervation, surgical approaches and rehabilitation, effectively complementing the approach in dystonia at the clinics. A timely critical pathophysiologic review, including the muscle spindle involvement in dystonia, is highlighted at the book's end.