Selective Dorsal Rhizotomy for Spasticity 

Updated: Dec 11, 2018
Author: Richard CE Anderson, MD; Chief Editor: Kim J Burchiel, MD, FACS 

Overview

Background

Selective dorsal rhizotomy (SDR) is used primarily to treat children with lower-extremity spasticity, also known as spastic diplegia or diparesis. The primary goal of SDR is to reduce spasticity and to improve lower-extremity function. Patients who ultimately benefit most from the procedure typically have pure spasticity involving the lower limbs, good cognitive function and strength, no fixed contractures, and postural stability.

Although the procedure has been successful in adolescents and even young adults, it is generally performed in a younger population (aged 3-8 years).[1] Patients with spastic quadriparesis (all 4 extremities involved) also benefit from SDR, but improvements in the upper extremities are typically less predictable and dramatic than those seen in the lower extremities. If a child with spastic quadriparesis has a significant component of dystonia, however, SDR may not be the best treatment option, and other modalities, including intrathecal baclofen, should be considered.[2]

Cerebral palsy is one of the most common congenital neurological conditions in children, affecting more than 10,000 newborns each year in the United States. Its incidence in 8-year-old children is approximately 3.6 per 1000.[3] Intractable spasticity can cause pain, sleep disorders, increased energy requirements, interference with positioning and difficulty transferring the child, and complications with dressing and body hygiene. Individuals with spasticity often develop contractures and pressure sores that compound treatment.[4] More than 90% of children survive into adulthood if motor and feeding skills are ensured, which rests on the abilities of the individual, as well as caregivers.

Ambulatory function and strength

In some cases, a child may use his or her spasticity for postural support in upright positions. If the child has poor underlying muscle strength, spastic tone often becomes the means by which the child is able to provide postural stability. Assessment of postural tone is important in the preoperative evaluation, as removal of this spastic tone may result in decreased ambulatory potential.[5]

Among patients with exceptionally poor strength and tone dependence, SDR may not lead to an optimal outcome.

History of the procedure

Dorsal rhizotomy, which literally means to cut or sever a dorsal nerve root, was initially presented as a surgical treatment by Dr. Otfrid Foerster in 1911 at a meeting for the Royal Society of Medicine.[6]

A subsequent review by Foerster, including 159 cases of spasticity, also recognized the importance of root conservation in order to preserve the balance between spasticity and flaccidity. By careful electrical stimulation of specific roots, he proposed selective root cutting in order to preserve a certain degree of function.[7]

Their work advanced the concept of a more selective dorsal rhizotomy, with rootlets chosen initially based on clinically identified spastic muscle groups segregated by their disabling or beneficial effects. Further modification of Foerster’s technique by Fasano et al, described as “functional posterior rhizotomy,” promoted the use of intraoperative electrostimulation for root selection instead of the Gros method of clinical identification. This allowed proprioceptive afferents to be saved and better functional outcomes to be achieved, further solidifying the procedure now known as SDR.[8]

The exact anatomical localization of the Ia sensory input to the spinal cord at the dorsal root entry zone by Sindou in 1974 made SDR a more functional possibility.[9] Severance of these specific inputs would allow for loss of tone without concomitant loss of other sensory input or motor control.

Since popularization and further revision by Peacock and Eastman in 1981, the procedure has remained an important treatment for spasticity control, specifically in those with cerebral palsy.[10]

Pathophysiology

Normal muscle tone depends on balanced excitatory and inhibitory influences on alpha motor neurons, which are nerves that begin in the spinal cord and travel via peripheral nerves to innervate skeletal muscles. Alpha motor neurons are the principle effectors of motor tone and control. Inhibitory interneurons influence alpha motor neurons within the spinal cord and are activated by cortical upper motor neurons. Excitatory influences come from both muscle spindles and Golgi tendon organs, which are located in the skeletal muscle/tendon, and send information to the spinal cord via afferent sensory fibers of peripheral nerves.

Brain injury in cerebral palsy prevents descending spinal cord tracts from providing the necessary inhibitory influence, creating an imbalance of overabundant excitatory input. The stretch reflex arc between the spinal cord and skeletal muscle continues to receive excitatory input from Ia afferents traveling from the muscle through the peripheral sensory nerve’s dorsal root, without adequate opposing inhibition from the brain’s descending tracts, ultimately causing what is termed spasticity. Spasticity is defined as a velocity-dependent increase in stretch reflexes, which translates to a clinical increase in resistance to passive movement. By decreasing the excitatory afferent input from the dorsal roots, the amount of excitation experienced by the alpha motor neurons can be reduced, therefore reducing spasticity.[11]

Schematic drawings representing the excitatory and Schematic drawings representing the excitatory and inhibitory influences on the spinal cord alpha motor neuron, which innervates the muscle fibers. (A) Normal physiology with a balance of inhibitory influence from descending neurons and excitatory influence from the sensory spinal reflex arc. (B) In children with spasticity, injury to the upper motor neuron results in a decrease in the descending inhibitory influence, leaving a hyperactive spinal cord reflex arc. By cutting some of the dorsal rootlets, selective dorsal rhizotomy can help restore balance to the alpha motor neuron by reducing the amount of excitatory influence on the alpha motor neuron. Courtesy of Tae S Park, MD.

Indications

SDR has been demonstrated to be beneficial in children with bilateral lower-extremity spasticity due to upper motor neuron injury in cerebral palsy. Patients who achieve the best outcomes are those who are personally motivated, younger, and ambulatory and who exhibit good proximal strength and motor control in the lower extremities.[12]

SDR is also likely to benefit patients with significant lower-extremity spasticity whose reduction in tone will be helpful in improving function and/or easing the burden of care.

The ability to participate in a rigorous postoperative physical therapy program is essential and must be discussed with families and caregivers in advance.[11]

Contraindications

In some cases, a child may use his or her spasticity for postural support in upright positions. If the child has poor underlying muscle strength, spastic tone often becomes the means by which the child is able to provide postural stability. Assessment of postural tone is important in the preoperative evaluation, as removal of this spastic tone may result in decreased ambulatory potential.[5]

Among patients with exceptionally poor strength and tone dependence, SDR may not lead to an optimal outcome.

Technical Considerations

Successful motor tone reduction in SDR and proper surgical technique are based on principles of nerve anatomy with regard to the sensorimotor reflex loop and the anatomic position of sensory and motor roots.

Sensory nerve roots contain efferent signals that contribute to the maintenance of motor tone, which is abnormally high in cerebral palsy because of a lack of supraspinal inhibition. Therefore, in order to reduce motor tone without causing motor weakness, only the sensory nerve rootlets (and not motor) must be cut in SDR. The sensory nerve rootlets can be identified by their anatomic position as they exit the canal dorsally and must also be confirmed as sensory nerve using intraoperative electrophysiologic monitoring.

Procedure Planning

Complication Prevention

Complications of SDR[13] include sensory changes such as dysesthesias or hyperesthesias, which may be avoided by limiting the number of dorsal rootlets cut.

Postoperative urinary retention or sphincter dysfunction may be avoided with intraoperative identification and avoidance of sectioning sacral nerve roots.

Postoperative cerebrospinal fluid leak is avoided with water-tight dural closure and overlying fascial closure, as well as dermal closure with a running nylon suture.

Immediately after surgery, patients typically have a significant reduction in spasticity, often unmasking underlying muscle weakness.[14] Pudendal monitoring, in an attempt to avoid such complications, is practiced by many institutions.[15]

Outcomes

Pediatric studies have demonstrated the importance of early intervention[16, 17] ; outcomes are often better in patients younger than 8 years.[1]

Since modernization of SDR, 3 randomized controlled trials have examined the clinical safety and improved outcomes of SDR compared with physical therapy alone; all 3 studies confirm that SDR reduces spasticity, although only 2 of the 3 studies demonstrated a statistically significant advantage in functional outcome.[10, 1, 18]

A subsequent meta-analysis demonstrated significant spasticity reduction and functional improvements after comprehensive treatment with SDR and follow-up physiotherapy.[19] Another more recent study comparing SDR to intrathecal baclofen pump determined that both result in significant improvements in tone, function, and passive range of motion, but SDR provided a larger magnitude of improvement.[20]

In a meta-analysis of the 3 randomized clinical trials of SDR,[21, 19] 82 patients younger than 8 years underwent either SDR with physiotherapy or physiotherapy alone. Outcome measures at 9-12 months included lower-extremity spasticity (Ashworth scale) and overall function (Gross Motor Function Measure). Both a statistically significant reduction in spasticity and a greater functional improvement were shown in children undergoing SDR with physiotherapy compared to those who underwent physiotherapy alone. Multivariate analysis also showed a direct correlation between the amount of dorsal root tissue transected and the amount of functional improvement, despite the variability of technique between institutions.

 

Periprocedural Care

Patient Education & Consent

It is essential for the family and interdisciplinary health care team to understand and agree upon the goals of surgical intervention in spasticity. With the correct patient selection and postoperative rehabilitation, children with spasticity can experience marked improvements in lower extremity function, range of motion, and global function following selective dorsal rhizotomy (SDR). With an irreversible procedure such as SDR, however, expectations should be clearly defined and realistic outcomes delineated.

The ability to participate in a rigorous postoperative physical therapy program is essential and must be discussed with families and caregivers in advance.[11]

Patient Instructions

All patients undergo general anesthesia so are required to be kept on nil per os (NPO; nothing by mouth) from midnight the night before surgery. Patients are sometimes instructed to use a special cleansing soap the night prior and morning of surgery in order to minimize the risk of surgical infection. Patients may expect to be kept flat in bed the first day or two postoperatively in order to minimize the risk of cerebrospinal fluid leakage.

Elements of Informed Consent

Patients must be aware the goal of surgery is reduction of motor tone, which has various functional implications based on the patient’s age, functional status, strength, and ability to engage in postoperative physical therapy. They must be aware that the consequences of rootlet transection are irreversible. The patient should also be informed of the risk of cerebrospinal fluid leak from an intradural operation.

Pre-Procedure Planning

Preoperative MRI and intraoperative ultrasonography are helpful to determine if additional bone removal is needed to achieve adequate exposure around the conus.

Equipment

Preoperative neurophysiologic monitoring electrodes are placed after intubation. Needle electrodes are typically placed in the adductor magnus (longus or brevis) (L2-4), vastus lateralis (L2-4), anterior tibialis (L4, L5), bicep femoris (L5, S1), gastrocnemius (S1, S2), and external anal sphincter muscles (S2-S4) in preparation for intraoperative electromyography (EMG).

Ultrasonography is used to ensure adequate bone removal

An operating microscope is used for the intradural portion of the case.

Rhizotomy stimulating electrodes and routine microinstruments should be made available.

Anesthesia

The anesthetic regimen is critical to intraoperative evaluation by the neurophysiologist and physiotherapy team. Premedication with a short-acting benzodiazepine and induction agents such as sevoflurane and remifentanil are frequently used. Subsequently, propofol, benzodiazepines, and paralytics are avoided because of the alteration of EMG activity.

Anesthesia is often maintained with fentanyl, nitrous oxide, and sevoflurane.

Positioning

The patient is placed in a prone position, with soft supports under the chest and hips, and appropriate padding of the knees, feet, and arms. The patient’s head is placed in a soft foam head-holder with the breathing tube to the side. The patient’s arms are typically placed at 90° at the shoulder and elbows, in a cephalad direction, in order to allow the surgeon access to the lumbosacral spine and anesthesiologist better access to lines. In smaller children, the arms may be left at the side of the hips in anatomical position, if desired.

If manual palpation of muscle groups is performed intraoperatively to assess response during nerve rootlet stimulation, the patient’s feet should be close to the bottom of the bed to allow adequate access for the therapists under the operative drapes.

If a traditional SDR is performed, a five- or six-level laminoplasty is typically done to expose the entire cauda equina. If a SDR centered at the conus is performed, fluoroscopy is used to localize L1. Preoperative MRI and intraoperative ultrasonography are helpful to determine if additional bone removal is needed to achieve adequate exposure around the conus.

Monitoring & Follow-up

Initial management includes transfer to the PICU and then to the neurosurgical ward with intravenous and then oral pain medications, as needed.

Antibiotics are continued for 24 hours, and steroids for 72 hours postoperatively.

The patient may be kept flat for 1-3 days postoperatively to promote healing and to reduce the risk of cerebrospinal fluid leak.

Recovery from surgery typically takes 2-3 days.

Immediately after surgery, patients typically have a significant reduction in spasticity, often unmasking underlying muscle weakness.[14] Pudendal monitoring, in an attempt to avoid such complications, is practiced by many institutions.[15] Although studies have proposed an increased incidence of scoliosis and hip subluxation following SDR, the common finding of these conditions in patients with spasticity who have not undergone SDR leave the role of SDR in the causation of these conditions unclear.[22, 23, 24]

Long-term monitoring

Several studies have documented positive outcomes from SDR. Muscle tone, flexibility, gait pattern, functional positioning, and the functional ability of the child have all been documented as improving following SDR.[20, 25, 26] Of note, SDR did not eliminate the need for further orthopedic surgery in approximately half of patients requiring further tendon-lengthening procedures. The need for subsequent orthopedic procedures has also been reported in other studies; however, it is likely that these procedures would have been necessary regardless of SDR.[27]

 

Technique

Approach Considerations

Although a traditional selective dorsal rhizotomy (SDR) involves a long skin incision with a 5- or 6-level laminectomy or laminoplasty, it can also be performed with a less-invasive approach through a smaller incision and a single level laminectomy.[28] Advantages of this approach include a smaller incision, fewer levels of bone removal or disruption, and less postoperative pain.[11]

Preoperative antibiotics and steroids are given prior to skin incision.

Selective Dorsal Rhizotomy

The steps of SDR are illustrated in the images below.

After the conus is clearly identified, a single la After the conus is clearly identified, a single laminectomyis done entirely with a Midas Rex craniotome. At least 5 mm of thecaudal conus should be exposed. The laminectomy extends laterallyclose to the facet joint. Courtesy of Tae S Park, MD.
After the dural incision, an operatingmicroscope i After the dural incision, an operatingmicroscope is brought into the field. The L-1 and L-2 spinal roots areidentified at the corresponding intervertebral foramina, and the filumterminale in the midline is found. Courtesy of Tae S Park, MD.
The L-2 dorsal root and the dorsal rootsmedial to The L-2 dorsal root and the dorsal rootsmedial to the L-2 root are retracted medially to separate the L2-S2 dorsalroots from the ventral roots. The thin S3-5 spinal roots exiting fromthe conus are identified. A cotton patty is placed over the ventralroots and lower sacral roots. Courtesy of Tae S Park, MD.
A 5 mm Silastic sheet is placed underthe L2-S2 dor A 5 mm Silastic sheet is placed underthe L2-S2 dorsal roots, after which the sugeon again inspects the L-2 dorsal root at the foraminal exit, the lateral surface of the conusbetween the dorsal and ventral roots, and the lower sacral roots near the filum terminale. The Inspection ensures placement of only the the L2-S2 dorsal roots on top of the Silastic sheet. Courtesy of Tae S Park, MD.
The L-2 dorsal root is easily identified. In an at The L-2 dorsal root is easily identified. In an attempt to identify the L3-S2 dorsal roots, all the dorsal roots are spread over the Silastic sheet and grouped into presumed individual dorsal roots. Then the innervation pattern of each dorsal root is examined with electromyographic (EMG) responses to electrical stimulation with a threshold voltage. Courtesy of Tae S Park, MD.
With a Scheer needle, each dorsal root is subdivid With a Scheer needle, each dorsal root is subdivided into three to five rootlet fascicles, which are subjected to EMG testing. Courtesy of Tae S Park, MD.
Stimulation of an L-2 rootlet fascicle elicits an Stimulation of an L-2 rootlet fascicle elicits an unsustained discharge to a train of titanic stimuli. Courtesy of Tae S Park, MD.
The rootlet is thus spared from sectioning and pla The rootlet is thus spared from sectioning and placed behind the Silastic sheet. Courtesy of Tae S Park, MD.
Stimulation of a rootlet is thus sectioned. Courte Stimulation of a rootlet is thus sectioned. Courtesy of Tae S Park, MD.
The rootlets spared from sectioning are under the The rootlets spared from sectioning are under the Silastic sheet, and the roots to be tested are on top of the Silastic sheet. Note the EMG testing and sectioning of the dorsal roots are performed caudal to the conus. Courtesy of Tae S Park, MD.