Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Selective peripheral denervation is an elegant surgical method for the treatment of cervical dystonia that can be tailored to the specific pattern of dystonia for an individual patient.
Selective peripheral denervation most frequently consists of a combination of sectioning the peripheral branch of the spinal accessory nerve to the sternocleidomastoid muscle and extraspinal posterior ramisectomy from C1 to C6.
Clinical improvement and reduced disability have been reported in 70% to 90% of patients in most series, with a wide variability in the degree of improvement.
Complications have been described rarely and were transient in most instances.
Patients may gain additional benefit from stepwise procedures over time.
Dystonia has become an established indication for deep brain stimulation (DBS) within the past few years. , In particular, cervical dystonia (CD) more recently has received great attention as an indication for DBS. When pallidal DBS was introduced in the late 1990s, it was initially considered for treating patients with medical refractory CD who were not candidates for peripheral surgery, including those with severe retrocollis or more mobile elements of dystonia. Meanwhile, pallidal DBS has gradually become the mainstay of surgical treatment in most countries. , Selective peripheral denervation is more rarely performed nowadays despite its proven efficacy, and it is no longer being recommended as the second-line treatment when the response to botulinum neurotoxin (BoNT) injection does not yield optimal benefit. New data on larger study populations are now available from Eastern countries such as China and South Korea. , Given the increased recognition of mild yet not uncommon bradykinetic side effects of pallidal DBS in patients with CD, selective peripheral denervation still holds great potential for future management of CD, and we may see a comeback of this elegant procedure in due time.
Selective peripheral denervation requires a special expertise not only in surgical technique but also in the phenomenologic differential diagnosis of CD. The term selective peripheral denervation was coined by Bertrand, who popularized the procedure in the 1970s. It consists of a combination of sectioning of the peripheral branch of the spinal accessory nerve to the sternocleidomastoid muscle and posterior ramisectomy from C1 to C6. There have been several modifications of the technique, and it has often been combined with targeted myotomies. In experienced hands, it is a safe procedure with minimal and infrequent side effects.
The term cervical dystonia is used consistently nowadays and has ousted the term spasmodic torticollis for several reasons: few patients present with simple turning of the head; the abnormal movement is not always spasmodic; and, most importantly, nondystonic neck postures are called torticollis as well. , Dystonia recently has been redefined as a movement disorder characterized by sustained or intermittent muscle contractions, causing abnormal and often repetitive movements, postures, or both. CD affects predominantly the neck, including the anterior and posterior neck muscles. In some patients, the shoulder is involved with protraction or elevation of the shoulder. In a subset of patients, CD is a feature of more widespread segmental dystonia also involving muscles of the face, larynx, or upper extremities. Frequently, there are also tremulous or jerking movements of the head. Dystonic tremor is more irregular than essential tremor and is most evident when the patient attempts to move the head in the direction contralateral to the force of the dystonia. Most frequently, CD is accompanied by neck pain that may be severe and lead to further incapacitation. In many instances, patients employ a “sensory trick” such as touching the chin, holding the neck, or other maneuvers to decrease the dystonic activity, also called geste antagoniste .
CD is the most frequent form of focal dystonia. , Its prevalence is estimated to range between 5 and 13 per 10,000 people in Western countries. The mean age of onset is 41 years, and, similar to other forms of focal dystonia, there is a slight female preponderance (about 1.2:1). In the vast majority of patients with CD, no underlying cause can be identified, while various genetic defects have been identified in a few instances. Until recently, both idiopathic and genetic manifestations would have been referred to as primary dystonia, but the new classification uses the terms idiopathic dystonia and inherited dystonia, respectively. The new nomenclature has been widely accepted in the past few years, but compliance with it has been comparatively lower in “surgical” studies. The term acquired has replaced the description of secondary dystonia in the rare case of CD caused by trauma or exposure to neuroleptic drugs.
Various basic patterns of CD have been defined according to head position. , Torticollis indicates rotation of the head about the head-body-axis (movement of the chin toward one shoulder in the horizontal plane); laterocollis is defined as a sideward tilt of the head (movement of the ear toward the ipsilateral shoulder); anterocollis is defined as flexion of the neck in the anterior-posterior-axis; and retrocollis is defined as extension in the anterior-posterior axis. Lateral shift depicts translation of the axis of the head in the horizontal plane, and sagittal shift in the sagittal plain. Commonly, a combination of these abnormalities is present, depending on the degree of involvement of different cervical muscles ( Fig. 118.1 ). It is important to understand which muscles are causing the dystonic posture or movements in an individual patient. This issue is even more important when deciding which muscles should be denervated because, in a given patient, dystonic activity may vary in different muscles yet produce the same dystonic posture.
The differential diagnosis of CD includes a variety of nondystonic disorders. , Pseudodystonia may be caused by atlantoaxial dislocation, degenerative disk disease, Klippel-Feil-syndrome, pediatric posterior fossa tumors, trochlear palsy, Sandifer syndrome, and various other disorders. More systemic basal ganglia disorders must be excluded in patients with onset of CD at a young age. In untreated or undertreated CD, persistent dystonic postures may result in accelerated degenerative cervical spine disease manifesting as cervical myelopathy or radiculopathy. Rarely, dystonic postures may even lead to permanent fixed deformities secondary to ossification of spinal ligaments or facet joints. ,
The first-line treatment of CD is chemodenervation with BoNT type A. , , The dose and the sites of injection must be individualized to obtain optimal results. The efficacy and safety of BoNT has been demonstrated in several randomized controlled and open trials. Most report improvement in about 90% of CD patients, with mild side effects in up to 28%. The effect of BoNT is usually noted about 1 week after injection, and the average duration of the benefit is 3 to 4 months. In patients who are resistant to BoNT type A, BoNT type B and other newer preparations have become an alternative. It must be noted, however, that the average duration of maximum improvement is much shorter, side effects are more frequent, and immunogenity is higher with BoNT B. Pharmacotherapy for CD includes mainly anticholinergic drugs or muscle relaxants. Overall, drug treatment plays a minor role in CD, although many patients also take analgesics for alleviation of chronic neck pain.
Surgical treatment for CD has a long history dating back to ancient Greece. , , Myotomies of the sternocleidomastoid muscles were performed as early as 1641, and denervation of cervical muscles was achieved by 1834. Intradural procedures initially targeted sectioning of the posterior cervical nerve roots, but soon attention shifted to the anterior roots. McKenzie developed intradural rhizotomy of the anterior upper cervical roots in combination with sectioning of the intradural spinal accessory nerve in the 1920s. Soon thereafter, Dandy, as well as Hamby and Schiffer, refined this procedure further. Until the early 1990s, upper cervical rhizotomies and spinal accessory nerve sectioning were the most common surgical procedures for CD. , This “standard” procedure, however, was rather nonselective and was accompanied by frequent side effects with sometimes devastating complications, therefore modifications were elaborated aiming to denervate dystonic muscles exclusively and to preserve normal activity. In different series, however, reported results and also complications were highly variable. Although most studies claimed useful postoperative improvement in 60% to 90% of patients, it was unclear to what extent symptomatic amelioration of abnormal postures or movements translated to improvement in functional disability, in particular regarding the high number of side effects. A weak or unstable neck was estimated to occur in up to 40% of patients after bilateral rhizotomy, and transient dysphagia was noted in up to 30% of patients. More favorable results with fewer side effects were seen with selective approaches targeting only those nerves supplying clearly dystonic muscles and with staging of the procedure.
Microvascular decompression of the spinal accessory nerve has been attempted in a limited number of patients based on the erroneous assumption that CD would be caused by vascular compression of the spinal accessory nerve, in analogy to other cranial neuropathies such as hemifacial spasm or trigeminal neuralgia. However, pathophysiologic concepts did not support microvascular decompression as a valid treatment option for CD, and the outcome data were ambiguous. Microvascular decompression was abandoned and is no longer considered a useful treatment option for CD. ,
When Bertrand introduced selective peripheral denervation, combining posterior ramisectomy with spinal accessory nerve sectioning, he demonstrated that this innovative concept was much more selective than earlier intradural procedures, produced more favorable outcomes, and resulted in a much lower frequency of side effects. , The most outstanding advantage of selective peripheral denervation as opposed to intradural rhizotomy certainly is that it is an extradural as well as an extraspinal procedure ( Fig. 118.2 ). Although rhizotomy can only approach the nerve roots from C1 to C3 bilaterally, respectively, from C1 to C4 on one side, ramisectomy can be performed down from C1 to C6, either on one side or bilaterally in the same session. When selective ramisectomy and peripheral denervation became increasingly popular, sometimes in combination with myotomy, rhizotomy was abandoned almost completely. With regard to its rather unfavorable risk-benefit profile, intradural rhizotomy without additional ramisectomies of lower segments now is considered obsolete.
Until recently, selective peripheral denervation was considered the second-line treatment in CD patients who could not achieve satisfactory benefit with medical treatment or repeated BoNT injections, while pallidal DBS was recommended as a third-line treatment only or was reserved for patients in whom selective peripheral denervation was not feasible. , This concept, however, has been challenged, and many patients now are referred directly for DBS. ,
It has been estimated that the frequency of primary nonresponders to BoNT injections is 6% to 14% of patients with CD. , In general, immunoresistance to BoNT may develop in about 3% to 10% of patients during long-term treatment. Although newer BoNT preparations tend to be less immunogenic, antibody-induced therapy failure with BoNT B has been described, even at a rate of 44%. Patients with predominantly tonic dystonia are better candidates for selective peripheral denervation than those with prominent phasic movements. Patients with marked myoclonus or with dystonic head tremor are not good candidates for the procedure. In some patients, selective peripheral denervation may also serve as an adjunct or as an alternative to BoNT injection. Likewise, it may be combined with pallidal DBS to achieve additional benefit. ,
It is pivotal to tailor the denervation procedure to each patient’s individual pattern of CD, which may involve several successive operative steps and also the use of additional myotomy or myectomy. , The primary goal is to selectively weaken the dystonic muscles while preserving normal muscle function to avoid side effects. The patient and the treating neurologist must be informed that a staged procedure may be necessary to achieve an optimal result. The muscles with the most prominent dystonic activity are usually identified by clinical examination observing the pattern of the patient’s dystonia and palpating the affected muscles. Electromyography guidance may be useful in some cases. In addition to the muscles most commonly involved in CD (sternocleidomastoid, trapezius, splenius, and semispinalis muscles), other muscles such as the levator scapulae, scalene, longus colli, and paraspinal erector trunci muscles should be investigated.
Because the operative approach is tailored to the pattern of dystonia, specific procedures may vary. For example, in a patient with torticollis with rotation of the head to the right side, the combined procedure would include right posterior ramisectomy and left selective peripheral sternocleidomastoid muscle denervation, possibly combined with myotomy or myectomy of the left sternocleidomastoid muscle. In a patient with retrocollis, bilateral posterior ramisectomy, eventually combined with muscle sectioning of the posterior neck muscles, would be most useful.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here