Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Selective Posterior Rhizotomies for Spasticity in Children

A. Leland Albright, M.D.

Introduction

Spasticity, from the Greek word spastikos - to pull or draw - is perhaps the most common neurological pathophysiologic condition of children, occurring in thousands with cerebral palsy, cerebral injuries and cerebral malformations. It is thought to result from an imbalance of inhibitory and excitatory impulses converging on the anterior horn cells. Muscle tone is normal if inhibitory impulses descending through tectospinal, reticulospinal and vestihulospinal pathways are balanced with excitatory impulses entering the spinal cord from muscle spindles, through la fibers in posterior nerve roots. Injuries to the cerebrum and basal ganglia decrease the descending inhibitory impulses and result in a relative excess of excitatory impulses. Those excess impulses can now be reduced by selective posterior rhizotomies (SPR).

History of SPR

Posterior (sensory) rhizotomies were first performed by Alfred Foerster, a German neurosurgeon who was knowledgeable of the now-classic experiments of Charles Sherrington. In 1898, Sherrington had caused extensor rigidity in cats by sectioning their mid-brains. He found that the rigidity could be abolished if the posterior nerve roots to a limb were divided.² In 1908, Foerster reported that he had divided the L2-SI posterior nerve roots bilaterally, excluding the knee extensors (usually L4), in 153 children with spastic cerebral palsy, and that spasticity in their legs was profoundly reduced.³ Because the posterior nerve root sectioning was extensive, sensation and joint position sense were seriously impaired and the operation fell into disuse.

From 1920-1970, neurosurgeons treated spasticity either by anterior (motor) rhizotomics, which were effective but which abolished any motor function and caused muscle atrophy, or by destructive procedures, i.e., cordotomy and myleotomy. Most of the operations' side effects were greater than their benefits. In most children with cerebral palsy (CP), spasticity itself was untreated and the contractures it caused were treated by orthopedic surgeons.

In 1978, Fasano, the Italian neurosurgeon, reported the technique of selective posterior rhizotomy (SPR).⁴ He reported that if normal lumbosacral posterior nerves were stimulated at 1,5 or 10hz, the innervated muscle would contract one, five or 10 times per second. However, if the stimulus were increased to 50 hz, the response differed between normal nerves and the nerves of children with cerebral palsy. In normal children, a 50 hz stimulus was followed by a brief muscle contraction then immediate relaxation. In children with spastic cerebral palsy, the 50 hz stimulus caused a tetanic muscle contraction throughout the one second stimulus and sometimes longer. He divided each posterior nerve root into several rootlets, stimulated each rootlet, and found that stimulation of some rootlets caused a "normal" muscle contraction whereas others caused the "abnormal" tetanic response. Fasano postulated that rootlets causing the tetanic response were bringing excess excitatory impulses into the cord, and he divided (cut) those rootlets. In reality, the "abnormal" rootlets are probably normal, butterminate in the spinal cord on alpha motor neurons or interneurons that have lost inhibitory afferent impulses, so that stimula- tion induces excessive contraction. Postoperatively, spasticity in those muscles was relieved and sensation and movement were unaffected. At the time of Fasano's 1978 report, spasticity had not recurred in children treated by SPR with follow-up periods of up to eight years.

Fasano's article was not appreciated in the U.S., but was noted by Warwick Peacock, a pediatric neurosurgeon beginning practice in South Africa. Peacock reported excellent relief of spasticity with the technique and modified it by stimulating the posterior nerve roots near the individual lumbar foramina rather than at the site of attachment to the conus medullaris, where nerve root identification was more difficult.⁵ Peacock's article also went unnoticed in the U.S. until 1985 when he immigrated to the U.S. and presented his results to the Pediatric Section of the American Association of Neurological Surgeons and to the American Academy for Cerebral Palsy. Descriptions of his work were reported in the New York Times and the Reader's Digest. Peacock was inundated with consults and trained several other pediatric neurosurgeons to perform the operation.

SPR has been performed on approximately 1,500 children in the U.S. by approximately 30 pediatric neurosurgeons, and patients have been evaluated by independent physical therapists and gait analysts.

Whenever any new operation is developed, there is-and should be-serious skepticism as to its effectiveness. Ideally, the procedure should be performed with equal results by several surgeons, with appropriate controls and with outcomes evaluated by independent examiners. Appropriate controls are difficult in children with cerebral palsy, partly because neurosurgeons have no oper ation of known effectiveness against which to evaluate SPR. Studies are underway, however, comparing children treated by SPR with those treated by only intensive physical therapy and with those treated by orthopedic operations.

Patient Selection

The Children's Hospital of Pittsburgh and other centers performing SPR often evaluate children in a multidisciplinary spasticity cliiiic attended by physical therapists occupational therapists, physiatrists, pediatric orthopedists and pediatric neurosurgeons. SPR at present is considered only for spasticity of the lower extremities. Peacock performed SPR for spasticity of the upper extremities in eight children and found that their spasticity was relieved, but their function was not improved, presumably because upper extremity function is far more complex than that in the lower extremities.

The two primary goals of SPR are to improve function and to facilitate care. Ideal patients to improve function are ambulatory children with spastic diplegia secondary to prematurity, with near-normal trunk and upper extremity function, and normal mentation Operations arc ideally done at three to five years of age, but can be done to improve function at two to 12 years of age in girls and two to 16 years of age in boys, presumably because the greater proximal leg strength in boys permits more functional improvement postoperatively. In children with spastic diplegia, the goal is to improve gait and to make their spontaneous movements easier, the goal is not to make gait normal.

The second indication for SPR is in children at the other end of the severity scale of CP, those with severe spastic quadriparesis. Their paravertebral spasticity impedes sitting in a chair, their adductor spasticity impedes spreading the legs for perineal care, and their hamstring spasticity impedes straightening their legs to put on pants. The goal of SPR in these children is simply to relieve spasticity to permit such simple funetions.

Decisions about SPR are more difficult in the large group of children whose involvement is intermediate-older children with spastic quadriparesis or triparesis who are ambulatory, usually with iliopsoas and ham- string contractures, who appear to be using their spasticity to maintain an erect posture. SPR in such children will alleviate their spasticity, but without spasticity they may lose the ability to ambulate, and their contractures are unaffected. In general, SPR is not indicated for such children, and is not indicated for children with athetoid CP.

SPR is ideally performed before contractures have developed and before orthopedic operations have been performed. If tendons have been lengthened by orthopedic operations and spasticity is subsequently treated by SPR the lengthened muscles may not be able to contract to normal lengths. Rhizotomies can be beneficial for unilateral leg spasticity as well as bilateral. Although SPR has been used primarily to treat spasticity of cerebral palsy, it also appears to be effective for spasticity secondary to head trauma.

Operation

SPR is performed under general anesthesia with children in the prone position, and requires four to six hours. LMG electrodes are inserted to monitor muscle contractions in individual lower extremity muscles, ideally monitoring the iliopsoas quadriceps, adductors, hamstrings, dorsiflexors and plantar flexors bilaterally. Muscle relaxants may be used during induction, but must be reversed by the time of nerve stimulation. A midline lumbar incision is made and paravertebral tissues are dissected from L2 to the sacrum. An L2-L5 laminectomy or osteoplastic Iaminotomy (to replace L2-L5 at the end of the operation) is performed. The dura is opened and the operating microscope is brought into use.

The S1 nerve root is identified. At Sl and at each subsequent level, the anterior motor root is

separated from posterior sensory root just proximal to the root foramen. The motor root is always anteriorly positioned, has fewer cross-striations than the sensory root, and has a lower stimulus threshold than the sensory root. A cottonoid pad is placed over the motor root to protect it. The sensory root is divided into three to eight rootlets and is stimulated with electrodes 5-10 mm apart. Stimuli are 0.1-0.3 ms, one stimulus approximately every three to four seconds, and the voltage is increased until a distinct muscle contraction occurs. A train of stimuli, 50 hz for one second, is then applied at that voltage and the legs are observed for abnormal" responses: (1) clonus; (2) contraction of muscles not normally innervated by that nerve (i.e., contraction of the adductors after S1 stimulation); or (3) contraction of contralateral leg muscles. The FMG is observed for abnormal" responses: (1) a crescendo pattern during the stimulus; (2) continuation of contraction after the stimulus ceases; and (3) a sustained, tetanic contraction throughout the stimulus. The words normal and abnormal are in quotations because they describe the responses we believe to be normal or abnormal, but there is a dearth of test data from normal children.

If abnormal contractions are seen, the individual rootlets are stimulated and the clinical and LMG responses are observed. If there are no abnormal clinical responses and the only abnormal EMG response is a tetanic contraction, the rootlets causing the strongest clinical and EMG tetanic contraction are divided. All nerves innervating the spastic muscles are tested, usually L2-S2 bilaterally. Approximately 30-60% of the posterior rootlets give abnormal responses and are divided. Occasionally, none of the rootlets of one nerve are abnormal and none are divided, and occasionally all are abnormal and all are divided.

After nerve testing and division is complete, the dura is closed, the lamina are replaced (if an osteoplastic laminotomy was performed), and the tissues are closed. An epidural catheter is often left in the epidural space and brought out through a separate puncture wound to instill epidural morphine for one or two days postoperatively.

Postoperative Treatment

Two to three days of postoperative bed rest is expected for children until their discomfort abates, then they begin physical therapy, primarily passive range of motion. They are hospitalized approximately seven days and are then either transferred to an inpatient rehabilitation facility for one to three months or are discharged to their homes to receive outpatient physical therapy. If the goal of operation is to improve function, therapy is given either once a day (outpatient) or twice a day (inpatient) tive days a week for six months, then outpatient therapy is given two or three days a week for another six months. Therapy stretches tendons, strengthens individual lower extremity muscles, and retrains the pattern of gait. Therapists have used functional electrical stimulation during therapy in nine of the 51 children I have performed SPR on, primarily to strengthen foot dorsiflexors. We reevaluate children with examinations, videotaping and gait analysis three months, six months, and one, two, and five years after SPR.

Risks and Morbidity

The operative risks of SPR are small, primarily a 1-2% risk of wound infection and a 1% risk of CSF leak. The risk of true weakness is less than 1%. Motor nerves are carefully protected and are never knowingly divided. Most morbidity of the operation occurs in the first two to three days afterward and consists of lumbar pain, which is substantial unless epidural morphine is given, plus irritability, low-grade fever, anorexia and headache. The lower extremities are of ten held flexed at the hips and knees for two to three days. Approximately 5% of children require intermittent bladder catheterization for a few days and 5% have leg muscle spasms that require diazepam or clonazepam. Older children may notice leg numbness for two to three weeks before the numbness resolves, but rarely complain of dysesthesias.

Results

The reduction in tone is evident immediately after operation.⁶ In 40 mildly or moderately handicapped children, Peacock found that SPR reduced tone and improved function in all.⁷ Side-sitting was improved in 39 children and standing was affected variably, improved in 11 out of 15, unchanged in two, and worse in two. Gait was improved by SPR in 12 out of 14 children with poor posture who were able to walk independently prior to SPR. Of eight pre-SPR children able to walk with a walking aid, five after SPR could walk independently. Two continued to require an aid and one needed above the knee braces to walk with an aid. In general, patients who are not ambulatory before SPR cannot walk postoperatively either.

In my experience with SPR in 51 children, only five who were non-ambulatory preoperatively are ambulatory postoperatively, all with assistive devices. Also in my experience, 25 pre-SPR children used orthoses, 14 used walkers and 10 used wheelchairs. Eollowing SPR, 19 children used orthoses, 20 used walkers and 11 used wheelchairs. Seventeen children have needed the same assistive devices before and after SPR, six needed more devices and 12 needed fewer. Cahan, et al, analyzed gait before and after SPR in 15 patients ranging in age from four to 20 years. From six to nine months after the operation, stride characteristics, motion about the leg joints, foot placement, and ambulatory LMG monitoring improved.⁸

Fasano monitored patients up to IS years after SPR and reported that relief of spasticity persisted in 95% of his patients.⁹ Upper extremity function of pre- and post-SPR by occupational therapists has been tested and shows improvement, but never normalizes in 50-60% of children. Presumably this occurs because the rhizotomy abolishes impulses which enter the spinal cord via lumbar roots and ascend via intersegmental interneurons that synapse with anterior horn cells exciting to the upper extremities. ¹⁰

SPR effectively alleviates spasticity in the lower extremities. Whether it does so more effectively than non-selective rhizotomy is unknown. Because of the neurophysiologic rationale supporting SPR, it is unlikely that the two will be compared. For children who use their spasticity to maintain an erect posture, the drawback of SPR is that its effects cannot be modulated until the desired reduction in spasticity is achieved. Our current studies indicate that intrathecal baclofen will probably be effective for such children in the near future.

A. Leland Aibright, M.D., is Associate Professor of Neurosurgery, University of Pittsburgh School of Medicine.

References:

1. Young, R.R., "The Physiology of Spasticity and Its Response to Therapy," Annals N.Y. Acad. Sci., 531, 1988, pp.146-149.
2. Sherrington, C.S., "Decerebrate Rigidity and Reflex Coordination of Movements," J. Physiol. Lond, 22, 1898, pp.319-337.
3. Foerster, 0., "On the Indications and Results of the Excision of Posterior Spinal Nerve Roots in Men," Surg. Gyn. and Obsir., 5, 1913, pp.463- 474.
4. Fasano, V.A., "Surgical Treatment of Spasticity in Cerebral Palsy," Child's Brain, 4, 1978, pp. 289-3()5.
5. Peacock, W.J. and L.J. Arens, "Selective Posterior Rhizotomy for the Relief of Spasticity," S. Afr. Med. J., 62, 1982, pp.119-124.
6. Peacock, W.J., Personal Communication, 1988.
7. Peacock, W.J., L.J. Arens and B. Berman, "Cerebral Palsy Spasticity: Selective Posterior Rhizotomy," Pediat. Neuroscience, 13, 1987, pp. 61-66.
8. Cahan, L.D., L. Beeler, J. Adams et al, "Gait Analysis After Selective Posterior Rhizotomy," Abstract Paper, Pediatric Section, Amer. Assoc. Neurol. Surg., Scottsdale, Arizona, Dec. 1988.
9. Fasano, V.A., Personal Communication, 1988.
10. Peacock, Arens, Berman, "Cerebral Palsy Spasticity, "pp.61-66.