SELECTIVE DORSAL RHIZOTOMY IN CEREBRAL PALSY: SELECTION CRITERIA AND POSTOPERATIVE PHYSICAL THERAPY PROTOCOLS

ABSTRACT Objective: To identify selection criteria for selective dorsal rhizotomy (SDR) in cerebral palsy, to analyze the instruments used for evaluation, and to describe the characteristics of physical therapy in postoperative protocols. Data sources: Integrative review performed in the following databases: SciELO, PEDro, Cochrane Library, and PubMed. The terms in both Portuguese and English for “cerebral palsy”, “selective dorsal rhizotomy”, and “physical therapy” were used in the search. Studies whose samples enrolled individuals with cerebral palsy who had attended physical therapy sessions for selective dorsal rhizotomy according to protocols and describing such protocols’ characteristics were included. Literature reviews were excluded and there was no restriction as to period of publication. Data synthesis: Eighteen papers were selected, most of them being prospective cohort studies with eight-month to ten-year follow-ups. In most studies, the instruments of assessment encompassed the domains of functions, body structure, and activity. The percentage of posterior root sections was close to 50%. Primary indications for SDR included ambulatory spastic diplegia, presence of spasticity that interfered with mobility, good strength of lower limbs and trunk muscles, no musculoskeletal deformities, dystonia, ataxia or athetosis, and good cognitive function. Postoperative physical therapy is part of SDR treatment protocols and should be intensive and specific, being given special emphasis in the first year. Conclusions: The studies underline the importance of appropriate patient selection to obatin success in the SDR. Postoperative physical therapy should be intensive and long-term, and must necessarily include strategies to modify the patient’s former motor pattern.


INTRODUCTION
Spasticity is the main clinical feature of patients with spastic cerebral palsy (CP) and is considered the most important cause of discomfort, gait abnormalities, and functional limitations. 1 It also generates muscle shortenings that influence bone growth and lead to deformities. Controlling it, therefore, is crucial to the treatment of CP. 2 Selective dorsal rhizotomy (SDR) is a neurosurgical procedure performed in children with bilateral spastic CP to reduce lower limb spasticity. 3 It is mostly performed at the lumbosacral level and consists of the interruption of the afferent stimulus by the monosynaptic stretch reflex. 3 In order to preserve the sensory and sphincter functions, the dorsal root is divided into radicles and only a portion of these is sectioned. 3 SDR was first described by Foerster in 1908, after he observed that the dorsal (sensory) radicles section could decrease spasticity; significant muscle weakness with sensory and proprioceptive losses was also observed after the procedure. 2 Thus, in 1978, Fasano presented the intraoperative electrophysiological stimulation and the section of a portion of dorsal radicles, and both techniques are currently used. 2 The method was then adopted and popularized by Peacock and Arens in 1980. 2 SDR results indicate spasticity reduction, muscle strength gain, gait speed and kinematics increase, and gross motor function improvement. [4][5][6] Patients submitted to SDR and physical therapy are compared to with those who only received physical therapy, significant reduction in spasticity and functional improvement are seen in the first group. 7,8 Specific physical therapy plays a central role in the postoperative phase, as spinal bone procedures such as laminectomy or laminotomy require special care in the first weeks of this period, in addition to formal conduct. 9,10 The centers that offer SDR follow special protocols for the postoperative period. In Brazil, this technique is starting to be disseminated and, due to peculiarities related to postoperative treatment, this review of protocols described in the literature aims to help professionals to better understand the role of physical therapy in rehabilitation. The objectives of this study were, therefore, to identify SDR selection criteria and to describe the characteristics of physical therapy postoperative protocols.

DATA SOURCE
This is an integrative literature review. The electronic search for references was carried out in August 2016 in the databases SciELO, PEDro, Cochrane Library, and PubMed. The terms used in the search, both in Portuguese and in English, were: "paralisia cerebral"/"cerebral palsy", "rizotomia dorsal seletiva"/"dorsal selective rhizotomy", and "fisioterapia"/"physical therapy". Headings, abstracts and, when necessary, the full study were reviewed to determine whether they would match inclusion criteria: studies conducted with individuals with CP who had attended physical therapy sessions according to SDR protocols and depicting such protocols' characteristics. No filters were applied to search, as well as there were no restrictions as to age group of sample subjects or period of publication. Literature reviews were excluded. The lists of references of selected papers were also searched for other relevant manuscripts.
After selection, the authors made a critical reading to grasp the main information, which was then presented in the following categories: • characteristics of studies; • characteristics of study samples.
• SDR selection criteria; • characteristics of physical therapy protocols.

DATA SYNTHESIS
According to the pre-established criteria, 18 articles were selected for this review. Figure 1 is the flowchart of papers' search and selection.
Evaluation tools according to domains of the International Classification of Functioning, Disability and Health The evaluation of SDR candidates should be as comprehensive as possible and encompass elements described by the International Classification of Functioning, Disability and Health (ICF). One of the studies 7 used the quantitative classification by ICF, and another one made evaluations considering ICF domains, as described below. The literature brings a variety of information with respect to items to be evaluated, with domain, structure, and body function as per ICF considered in all studies but one. 23 The instruments used in studies to evaluate domain, structure, and body function were: spasticity evaluation (Ashworth scale, [5][6][7][8][11][12][13][14]16,22 clinical signs of spasticity, 11,15 quantitative spasticity assessment (QSA) 6,11,19 by isokinetic dynamometer 4,17,18 ), motion range, [5][6][7][8][11][12][13][14][15]19,22 , reflex range, 12,14 14 by hips and spine radiography, 7 selectivity assessment 7 , and isometric contraction assessment by electromyography. 21 Only two studies 17,21 did not measure the activity domain, and in those addressing it, the instruments used were: Gross Motor Function Measure (GMFM), [4][5][6][7][8]11,12,14,16,18,22 walking status 6,8,11,14 , Gross Motor Classification System (GMFCS), 5,7,22,23 three-dimensional gait analysis, 4,6,7,12,18 observational gait analysis 7,16 (Observational Gait Score 7 , Edinburgh Visual Gait Score 16 ), urodynamics, 7 Peabody Fine Motors Scale, 8 self-care evaluation, 8,20 walking distance in 60 seconds 12 , and Physiological Cost Index. 8 Six studies 5,7,8,16,20,23 addressed the domain participation and its evaluation instruments: Pediatric Evaluation of Disability Inventory (PEDI), 5,7,16,20,23 Canadian Occupational Performance Measure (COMP) 7 , and self-care evaluation. 8,20 SDR evaluation should follow more comprehensive protocols with postoperative analysis of the same instruments, thus allowing a more accurate evaluation of results and better conclusion-drawing. Table 2 shows the characteristics of samples of the included studies. All of them enrolled individuals with spastic PC. The study by Chan et al mentioned a participant with hereditary spastic paraparesis (HSP), in addition to 20 individuals with CP. 7 Although CP diagnosis was one of the inclusion criteria of this review, SDR can also be indicated for patients with spasticity resulting from other diagnoses such as multiple sclerosis, Leigh syndrome, 25   Only one study did not include individuals with CP due to spastic diplegia 19 and four studies enrolled individuals with spastic quadriplegia. 7,14,19,20 Regarding GMFCS levels, only half of the studies 4,5,7,8,16-18,22.23 referred to this classification, and individuals presenting all levels are mentioned. Overall, SDR is the procedure of choice for spasticity of lower limbs in children with diplegia, 9 since they have more involvement of the lower limbs and dystonia is not always present. 9 Patients with spastic quadriplegia are more likely to present dystonia and involvement of both upper and lower limbs, and the treatment with continuous intrathecal baclofen infusion is more indicated, 9 although some studies support SDR. 7,14,19,20 Another aspect to be considered when indicating SDR is ambulation potential, 9 which includes GMFCS levels I, II, and III. However, investigations have performed SDR for levels IV and V with specific goals and suggested that this is an alternative to the use of continuous intrathecal baclofen infusion, given the management and monitoring complexity of this method. 28 Most studies had section of 50% of the selected posterior rootlets from L1 or L2 to S1 or S2. A meta-analysis showed direct relationship between the percentage of cut and function gain, that is, the decrease in spasticity helps in the acquisition of functional abilities. 1

CNS malformation 4,18
Spastic diplegia [4][5][6][7][8][11][12][13][14][15][16][17][18]20,21 Spastic quadriplegia with remarks 7,14,17,[20][21][22][23] Visual impairment limiting mobility 11 Good head-trunk control 6,7,22,23 LL reasonable muscle strength 12,22,23 Depends on spasticity to stand up or walk 22,23 Ability or potential to wander with and without supportive device, 4,6,12,18 for three meters 12 Able to walk barefoot for eight minutes, with or without support, 4.18 to sit, kneel and crawl independently for short periods, 16 to crouch seven times, 16 sit on a bench with free arms and to stand up with support 8,13 Fixed LL contractures 5,7 Severe fixed contractures: 11,12,16,17,22,23 >30º on hips and knee; 12 >15º on hips and knee and >30º on ankle; 11 >20º on hips, knee, and ankle and >80º popliteal angle GMFCS I -III, 4,18 I-V, 5 II-III 16 Progressive subluxation of the hips 8,11 36-month or more intellectual function 6,11 Minimum cognitive skills to actively participate 4,5,18 Children with intellectual disabilities 23 Spinal deformities, uncontrolled epilepsy, contraindication for prolonged anesthesia 11 Spasticity of LL interfering with functional tasks such as sitting, standing, and walking 4,5,7,8,[12][13][14][16][17][18][19][20]22,23 Spasticity in at least six muscle groups of both LL 16 Orthopedic surgery, 4,12,18 in the previous year 4,5,18 or near-term planning 8 Botulinum toxin or plaster in six months 4,18 Availability for intensive physical therapy 5,8,11 Good family and rehabilitation support 11,16 Severe cognitive disability 4,5,7,11,18 LL: lower limbs; GMFCS: Gross Motor Function Classification System; CNS: central nervous system. family support are cited, 16,29 as well as good rehabilitation 11,16 and the capacity to collaborate in rehabilitation (cognitively and emotionally). 18 Even though this is not the population to whom SDR is ideally indicated, some studies indicate it for patients with spastic quadriplegia 7,14,17,20 with the following criteria: 3.7 significant spasticity interfering with positioning, care, and passive stretching; absence of other motor disorders; and absence of fixed contractures in multiple joints. In both topographies, abnormalities of movement (dystonia, ataxia, choreoathetosis, hypotonia, stiffness), 4,6,7,11,13,17,18,22 hips instability, 11 significant scoliosis, 11 presence of significant fixed contractures, 7,11,12,16,17,22 absence of antigravity muscle strength, 11 and visual impairments limiting mobility 11 are contraindications for the procedure. The correct indication of SDR is fundamental for the success of treatment. 3,30 Criteria have been described and the literature supports that it is important that this decision is made by a multidisciplinary team trained and specialized in the care of spasticity in CP patients and with access to all treatment options. [1][2][3]10,31 This team should consist of a physical therapist, a pediatrician, an orthopedist, and a neurosurgeon, all of them trained and specialized. 1,31 The whole staff, including patients' family members, should agree with the SDR decision and with the individual treatment goals for each child. 2,9 A recent systematic review stated that these selection criteria vary across studies and are based more on clinical reasoning than on scientific evidence, and it is important that specialists come to a consensus on the subject. 3 Characteristics of physical therapy protocols Table 4 lists the characteristics of post-SDR physical therapy protocols, including start of sessions, length of hospital stay and frequency. Studies typically show that, after SDR, patients undergo intensive physical therapy rehabilitation lasting approximately one year, starting on the first days after surgery and staying hospitalized from six days to six weeks. Two studies 13,15 reported preoperative physical therapy and three 12,20,21 mentioned postoperative occupational therapy as well.
Half of the studies report that after the in-hospital physical therapy period, specific treatment guidelines are sent to local therapists, with whom the responsible therapist had made prior contact, in order to maintain consistency of the treatment plan.
As for the physical therapy program itself, early mobilization of the lower limbs is made during the first week after SDR to maintain a range of motion and positioning, including prone, supine and siting positions with extended lower limbs. 5,7,12,16 The first five days are specific for muscle strength exercises with hip abductors and extensors, knee extensors, dorsiflexors, and practice of normal orthostasis and gait patterns are initiated. 16 The onset of orthostasis is described as initiated by the use of parapodium in the 8 th day, 12 or with the use of fixed or ground-reaction Ankle Foot Orthoses (AFO) to stimulate knee extension on the sixth day, 16 and adaptation equipment. 14 Muscle strengthening is described as rehabilitation practice in most studies, 4,7,8,[11][12][13][14]16,17,20,21 with emphasis on the lower limb extensor and hip abductors muscles, knee extensors and dorsiflexors, 8,12,13,16 in addition to upper limbs 12 and trunk muscle., 4,12 The exercises are performed using isolated training, 20 progressive resistance training, 11 and selective or functional control. 21 Gait training starts on the second 7 or third week 12 and focuses on normal motor pattern with the use of supportive devices 17 if necessary. In addition, the use of normal movement pattern facilitation (neurodevelopmental theory) is also described, 8,[11][12][13]21 as well as fine motor skills training, 12 functional activities, 4,5,7,12,14,17,[20][21][22] daily-living activities, 5.7 posture control and alignment, 8,13,14,22 and postural transfer training with emphasis to balance when siting, kneeling, crawling, standing from floor and chair, standing, and on gait. 5,7,12,17,21,22 Hydrotherapy, 5.16 equotherapy, 5,16 and physical activities 5,22,23 are also mentioned.
According to the most recent recommendations by the National Institute for Health and Clinical Excellence (NICE), when it comes to treatment of spasticity in children and adolescents with non-progressive brain disorders, an intensive physical therapy program is essential after clinical approach to spasticity by SDR 31 and also determinant for successful outcomes. 30

FINAL REMARKS
Several studies have reported the treatment of spasticity by SDR associated with physical therapy. At large, they emphasize the importance of adequate indication of the procedure to be made by a multidisciplinary team that includes a physical therapist. The most important indication is for outpatients presenting spastic diplegia, as a means to improve gait and motor function patterns. A less frequent indication is for patients with spastic quadriplegia, with specific goals of positioning, spasticity control, sitting, hygiene, and daily care for both patient and relatives. Intensive and long-term postoperative physical therapy (especially in the first postoperative year) is emphasized and should cover strategies to modify the patient's former motor patterns.