PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women's and Children's Health (UK). Spasticity in Children and Young People with Non-Progressive Brain Disorders: Management of Spasticity and Co-Existing Motor Disorders and Their Early Musculoskeletal Complications. London: RCOG Press; 2012 Jul. (NICE Clinical Guidelines, No. 145.)

10Selective dorsal rhizotomy

Introduction

Selective dorsal rhizotomy (SDR) is a neurosurgical operation on nerves entering the spinal cord. The aim of SDR is to improve gross motor function, particularly the ability to walk, by reducing muscle spasticity. The operation was first performed in 1908 and developed further in the 1980s by Peacock who was responsible for introducing SDR into the USA. SDR is currently available in a number of centres in the USA and Canada, but only one centre in England and Wales has performed the operation on a regular basis and published results.

SDR involves identifying nerve roots coming into the spinal cord from leg muscles and severing some of them. One of two approaches may be used to access the nerve roots: the first involves removing six to eight lamina (multilevel approach); the second – less invasive – approach is to remove and replace just one or two lamina (single level approach). Resection of the nerve roots interrupts the abnormal circuit of nerve impulses that keeps muscle tone high. The nerve roots must be identified correctly during the operation using electrical stimulation. If nerve roots coming into the spinal cord from the skin, bladder or bowel are cut then the patient may develop numbness or bladder or bowel incontinence.

SDR is irreversible and selecting appropriate children and young people to undergo the procedure is very important. The surgical technique requires good exposure of nerve roots and meticulous attention to identification of roots that will be cut. In the literature, the percentage of nerve roots cut varies from 14% to 50%. Nerve roots to be cut are from lumbar 1 (L1) level to sacral 2 (S2) level, although some surgeons avoid cutting S2 roots to reduce the risk of incontinence.

Potential complications of SDR may be temporary or permanent, and kyphoscoliosis (curvature of the spine) or spondylolisthesis (slipped vertebrae) may occur afterwards. As with any other irreversible operation, SDR should only proceed if the benefits outweigh the potential complications.

Most children and young people who have undergone SDR have had spastic diplegic cerebral palsy and, since the aim of the operation is to improve the child or young person's ability to walk, most were in Gross Motor Function Classification System (GMFCS) level II or III.

After SDR, most children and young people are weak, and they may initially lose motor ability. An intensive period of rehabilitation is required after the surgery and the setting (inpatient or outpatient care during the rehabilitation period) will be a consideration. The full benefits of SDR might not be realised for up to 1 year after the surgery, and the continuing need for physical therapy is a major commitment for the child or young person and their family.

Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010) contains the following recommendations:

  • Evidence relating to SDR for spasticity in cerebral palsy highlights a risk of serious but well-recognised complications. The evidence on efficacy (that is, how well the procedure works in the studies in which it has been evaluated) is adequate and the procedure may be used provided that normal arrangements for clinical governance and audit are in place.
  • As part of the consent process parents and carers should be informed that the procedure is irreversible, and that SDR sometimes leads to deterioration in walking ability or bladder function, or later complications including spinal deformity. Parents and carers should understand that prolonged physical therapy (specifically physiotherapy) and aftercare will be required and that additional surgery may be required.
  • Selection of patients and their treatment should be carried out by a multidisciplinary team with specialist training and expertise in the care of spasticity in patients with cerebral palsy, and with access to the full range of treatment options. The team would normally include a physiotherapist, a paediatrician and surgeons, all with specific training and expertise.
  • NICE encourages further research into SDR, especially in relation to long-term outcomes. Outcome measures should include the incidence of neurological impairment and spinal deformity, the need for additional operations, and assessment of disability, social inclusion, and quality of life.

Although Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010) makes recommendations on the safety and efficacy of SDR, it does not address whether or not the NHS in England and Wales should fund SDR. The remit of this clinical guideline includes evaluation of the clinical and cost effectiveness of SDR. The GDG prioritised consideration of SDR combined with physical therapy compared with physical therapy and no SDR (with or without other interventions) in children and young people who have spasticity, with or without other motor disorders (dystonia, muscle weakness or choreoathetosis) as a result of a non-progressive brain disorder.

The search strategy used for this question was the same as the search strategy used during development of Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010). Thus, the GDG considered all the evidence identified for inclusion in Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010), plus evidence published more recently. In accordance with the NICE guideline development process, a specific review protocol was developed for the guideline. The guideline review protocol identified specific populations, interventions (combinations of SDR with other interventions such as physical therapy), comparators and outcomes on which to base decisions regarding clinical and cost effectiveness of SDR. The guideline review process differed further from the process used in Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010) in that the GRADE approach was used to grade the quality of the evidence included in the guideline review, and the GDG's interpretation of the evidence and formulation of recommendations was explicitly linked to the graded evidence. In particular, the guideline review focused on the best quality evidence, and so it included only prospective comparative studies and case series involving more than 200 children or young people. In contrast, Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010) included evidence from small non-comparative studies and retrospective comparative studies. Compared with Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010), the GDG prioritised additional outcomes for consideration, including active range of movement (AROM). The GDG also considered outcomes measured at different follow-up points (for example 6 months, 9 months, 12 months and 24 months) separately, rather than pooled outcomes over all time points. This approach has the potential to distinguish between temporary and sustained (or immediate and delayed) outcomes.

Review question

What is the clinical effectiveness of SDR in children and young people with spasticity caused by a non-progressive brain disorder?

Description of included studies

Seven studies were identified for inclusion for this review question (Abbott 1992; Buckon 2004b; Engsberg 2006; Kim 2001; McLaughlin 1998; Steinbok 1997; Wright 1998). The studies addressed two comparisons (SDR plus physical therapy versus physical therapy alone, and SDR plus physical therapy versus orthopaedic [soft tissue] surgery plus physical therapy), although two of the studies were non-comparative (see below).

Three parallel randomised controlled trials (RCTs; McLaughlin 1998; Steinbok 1997; Wright 1998) and one non-randomised prospective study (Engsberg 2006) compared SDR plus physical therapy to physical therapy alone. A total of 90 children and young people, all of whom had diplegia, were included in the three RCTs. One RCT included children aged 3–7 years (Steinbok 1997), another included children and young people aged 3–18 years (McLaughlin 1998) and the remaining study did not specify the age range of the participants (the mean age was 4 years 10 months; Wright 1998). The non-randomised prospective study (Engsberg 2006) presented outcomes for 84% (65/77) of the children and young people with spastic diplegic cerebral palsy (GMFCS level I, II or III) and 40 children and young people with no disability who were included in the study. The mean ages of the children and young people were 9.0 (standard deviation [SD] 5.3) years in the SDR plus physical therapy group and 9.7 (SD 4.5) years in the physical therapy alone group.

Two of the RCTs reported that all SDR operations were performed by the same surgeon (McLaughlin 1998; Wright 1998). Two trials conducted rhizotomies from L2 to S2 (Steinbok 1997; Wright 1998) and the other trial conducted rhizotomies from L1 to S2 (McLaughlin 1998). The percentages of dorsal roots transected were: 58% for L2 to S1 and 40% for S2 (Steinbok 1997); 50% on average of each dorsal root (Wright 1998); and 26% (range 14% to 50%) from L1 to S2 (McLaughlin 1998). The non-randomized prospective study conducted rhizotomies from L1 to S2 transecting approximately 65% of rootlets (Engsberg 2006).

Similar quantities and types of physical therapy (specifically physiotherapy) were received by both groups in one RCT (Steinbok 1997). The techniques used included passive movements, strengthening and neurodevelopmental treatment (NDT). Weight-bearing exercises were emphasised in both treatment groups. Measures were taken to maintain blinding of physiotherapists. In another RCT (Wright 1998) all participants received similar types of physical therapy, but those who underwent SDR plus physical therapy had higher treatment intensity during their 6-week post-operative stay to improve strength in the trunk and lower extremities. The physical therapy techniques used in both treatment groups in this RCT included: range of movement; strengthening through functional activities; facilitation of normal movement patterns and postural control; standing and gait-related activities; and work on fine motor skills and functional abilities. In the third RCT (McLaughlin 1998), the techniques used were described in less detail, but they were reported to be tailored to the individual child or young person's needs. The emphasis and techniques used were reported to be appropriate for children and young people undergoing SDR, and 20 different categories of treatment were documented by the treating community physical therapists. In the non-randomised prospective study, the SDR plus physical therapy group received physical therapy sessions in their home towns four times per week for 8 months after discharge. Treatments were then reduced to three times per week for an additional 12 months. The physical therapy alone group received the same number of physical therapy sessions. Treatment in both treatment groups concentrated on the trunk and lower extremities, on strengthening and on functional activities. Billing data were used to confirm that both groups received similar amounts of physical therapy (Engsberg 2006).

Caregivers were masked to treatment allocation in two RCTs (Steinbok 1997; Wright 1998) but not in the other (McLaughlin 1998). Outcome assessors were masked to treatment allocation in all three studies. One RCT (Wright 1998) reported that assessors were able to distinguish between treatment groups, but they were not involved in providing care for the participants. Children and young people in both groups in the non-randomised prospective study were similar at baseline for age, sex, weight, GMFCS level and gait status, and all were judged to be suitable candidates for SDR. Details of the recruitment process, inclusion and exclusion criteria and baseline clinical assessments were reported in the article (Engsberg 2006).

Outcomes were reported at 6 months in one RCT, 9 months in one RCT, 12 months in two RCTs and 24 months in one RCT. All three RCTs used Modified Ashworth Scale (MAS) scores to assess tone and reported the Gross Motor Function Measure (GMFM) scores. One RCT reported range of movement and one reported walking. No evidence was identified for Goal Attainment Scaling (GAS), Pediatric Evaluation of Disability Inventory (PEDI), acceptability and tolerability (as reported by the child or young person or their parent or carer) or the Child Health Questionnaire (CHQ) for quality of life. None of the RCTs reported mortality rates. In one RCT (McLaughlin 1998) back and lower extremity pain and urinary problems were reported via an adverse effects questionnaire administered by the investigators every 3 months over the 24-month follow-up period. Outcomes were reported at 8 months and 20 months in the non-randomised prospective study (Engsberg 2006).

One non-randomised prospective study (Buckon 2004b) compared SDR plus physical therapy to orthopaedic (soft tissue) surgery plus physical therapy. Twenty-five children with spastic diplegia (age range 4–10 years; mean age 71.3 months) and their parents were invited to choose between SDR and soft tissue surgery after receiving information about both procedures. The orthopaedic surgeon and neurosurgeon who performed the procedures were reported to be in clinical equipoise in relation to their judgements about the effectiveness of the treatments. The selection criteria for SDR were:

  • age 4–10 years
  • predominantly spastic disorder
  • good trunk control
  • lower extremity contractures less than 10 degrees
  • able to isolate lower-extremity movements
  • follow-up physical therapy available (three or four times per week)
  • history of prematurity
  • no significant ataxia, athetosis or scoliosis
  • good lower-extremity antigravity strength
  • ambulatory with or without assistive devices
  • co-operative.

The inclusion criteria for soft tissue surgery were:

  • kinematic dysfunction
  • evidence of dynamic limitation of movement and spasticity on static examination that would benefit from muscle and tendon lengthening, release or transfer.

Parents were given a booklet, counselling from both surgeons and the opportunity to talk to physical therapists and other physicians, and were assisted in finding published articles to inform their decisions. Parents returned 1 month after the initial assessment to have any remaining questions answered, and to inform the clinical staff of the family's decision.

Eighteen families chose SDR and the other seven chose soft tissue surgery. The children in the SDR group had a mean age of 71.3 months; 17 were community ambulators (11 without and six with assistive devices) and one was a household ambulator (GMFCS level I, n = 3; level II, n = 8; level III, n = 7). The children in the soft tissue surgery group had a mean age of 78.6 months; six were community ambulators (three without and three with assistive devices) and one was a household ambulator (GMFCS level I, n = 2; level II, n = 2; level III, n = 4). The majority of orthopaedic procedures performed were releases and lengthenings, although two children also had osteotomies. The participants received post-surgical physical therapy that was standard for the intervention that they received. Functional outcomes were assessed using the Gross Motor Performance Measure (GMPM), GMFM and PEDI at baseline and at 6 months, 12 months and 24 months after surgery.

Two case series (Abbott 1992; Kim 2001) reported non-comparative evidence on post-operative and long-term urinary problems, post-operative ileus, scoliosis and hip subluxation in children and young people who underwent SDR. One case series included children and young people aged 2–13 years (Kim 2001) and the other did report not the age range of the participants (Abbott 1992).

Evidence profiles

Selective dorsal rhizotomy plus physical therapy versus physical therapy alone

Reduction of spasticity and optimisation of movement

All three RCTs identified for inclusion used MAS scores to assess tone at the elbow, hip, knee and ankle, and overall tone. Outcomes were assessed at 6 and 12 months (Wright 1998), at 9 months (Steinbok 1997) and at 12 and 24 months (McLaughlin 1998). Range of movement was measured at 9 months (Steinbok 1997), while AROM and passive range of movement (PROM) were measured at 6 months and 12 months (Wright 1998) and AROM was measured at 8 months and 20 months in the non-randomised prospective study (Engsberg 2006).

Table 10.1. Evidence profile for selective dorsal rhizotomy and physical therapy compared with physical therapy only in children with diplegia; tone and joint movement assessment

Optimisation of function

All three RCTs (McLaughlin 1998; Steinbok 1997; Wright 1998) reported GMFM outcomes for each dimension and total scores. Outcomes were assessed at 6 months, 9 months, 12 months or 24 months, depending on the study. The non-randomised prospective study reported GMFM percentage scores at 8 months and 20 months. A timed walk and gait analysis was conducted at 12 months in one RCT (Wright 1998) and at 8 months and 20 months in the non-randomised prospective study (Engsberg 2006).

Table 10.2. Evidence profile for selective dorsal rhizotomy and physical therapy compared with physical therapy only in children with diplegia; functioning assessment

Quality of life

None of the studies identified for inclusion reported quality of life.

Adverse effects

Two of the RCTs (McLaughlin 1998; Steinbok 1997) and both case series (Abbott 1992; Kim 2001) reported adverse effects. One RCT (McLaughlin 1998) used a structured adverse event questionnaire administered to the parents by the investigators in person or by telephone at 3-month intervals. The case series comprised retrospective reviews of children and young people who had undergone SDR in hospitals in New York from 1986 to 1992 (Abbott 1992) or in Korea for the 10 years leading up to 2000 (Kim 2001).

None of the studies identified for inclusion reported mortality rates.

Outcomes assessing pain were reported in one RCT (McLaughlin 1998) and in one case series (Kim 2001). The RCT reported that six of the 21 children and young people in the SDR plus physical therapy group experienced a total of 14 incidents of back pain during the 24-month follow-up period, compared with no incidents at all among the 17 children and young people in the physical therapy group. (MODERATE) Lower extremity pain was reported by 10 of the 21 children and young people (a total of 11 incidents) in the SDR plus physical therapy group during the same follow-up period, compared with 16 out of the 17 children and young people (19 incidents) in the physical therapy group. (MODERATE) The case series (Kim 2001) reported that all 208 patients experienced post-operative back pain, which was controlled well using an intravenous fentanyl drip for 3 days post-operatively. The incidence of long-term back pain among children and young people who underwent SDR plus physical therapy was 3.4% (7/208). (VERY LOW)

Both case series reported outcomes related to urinary problems (bladder dysfunction), although the precise outcomes evaluated varied from study to study. Across both case series (Abbott 1992; Kim 2001), 7.2% (33/458) children who underwent SDR plus physical therapy experienced post-operative urinary retention. (VERY LOW) One RCT (Steinbok 1997) reported transient urinary retention in one of 14 children who underwent SDR plus physical therapy, and this resolved by the fourth post-operative day; no cases were reported in the physical therapy group. (MODERATE) One case series (Abbott 1992) reported that 0.4% of children (1/250) who underwent SDR plus physical therapy required catheterisation 18 months after surgery. (VERY LOW) The other case series (Kim 2001) reported that 1% (2/208) of children who underwent SDR plus physical therapy experienced long-term urinary incontinence (no further details reported). (VERY LOW) One RCT (McLaughlin 1998) recorded urinary adverse effects as part of the questionnaire administered to parents. Three of the 21 children and young people in the SDR plus physical therapy group reported one urinary adverse event each during the 24-month follow-up period, compared with no events among the 17 children and young people in the physical therapy group. (MODERATE)

One case series (Abbott 1992) reported an incidence rate of 1.2% (3/250) for post-operative ileus following SDR plus physical therapy. (VERY LOW)

One case series (Kim 2001) reported scoliosis rates in children following SDR surgery using laminectomy or laminoplasty; 8.6% (5/58) of children and young people developed scoliosis after laminectomy and 1.3% (2/150) developed scoliosis after laminoplasty. (VERY LOW)

Both case series examined outcomes relating to hip dislocation. In one study (Abbott 1992), 2.4% (6/250) of children and young people developed hip dislocation requiring a varus derotation osteotomy. In the other study (Kim 2001), 1% (2/208) of children and young people developed progressive hip migration requiring orthopaedic surgery.

Acceptability and tolerability

None of the studies identified for inclusion reported outcomes related to acceptability and tolerability.

Reduction of pain

The evidence relating to pain is presented under adverse effects (see above).

Selective dorsal rhizotomy plus physical therapy versus orthopaedic (soft tissue) surgery

Reduction of spasticity and optimisation of movement

The only study identified for inclusion (Buckon 2004b) did not report reduction of spasticity and optimisation of movement.

Optimisation of function

Quality of life

The study did not report quality of life.

Acceptability and tolerability

The study did not report outcomes related to acceptability and tolerability.

Reduction of pain

The study did not report reduction of pain.

Adverse effects

The study did not report adverse effects.

Evidence statement

Selective dorsal rhizotomy plus physical therapy versus physical therapy alone

Reduction of spasticity and optimisation of movement

With regard to trunk rotation, one non-randomised prospective study provided evidence that, compared with baseline, there was a reduction in AROM at 8 months and 20 months in children who received SDR plus physical therapy compared with no change in those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) With regard to pelvic rotation, the same non-randomised prospective study provided evidence that, compared with baseline, there was a reduction in AROM at 8 months and 20 months in children who received SDR plus physical therapy compared with improvements in those who received physical therapy alone. Although the statistical significance of these findings could not be determined, the authors reported that the reduction in AROM in pelvic rotation at 20 months from baseline was statistically significant in children who received SDR plus physical therapy. (VERY LOW)

With regard to pelvic tilt, one non-randomised prospective study provided evidence that, compared with baseline, there was a reduction in AROM at 8 months and 20 months in children who received SDR plus physical therapy compared with improvement and no change (respectively) in those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) The authors reported that the reduction in AROM in children who received SDR plus physical therapy compared with those who received physical therapy alone was statistically significant at 20 months, but not at 8 months (mean final score comparison across groups).

Concerning hip joints, one RCT reported that, compared with baseline, there was a statistically significant reduction in tone (evaluated using MAS scores) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone. (MODERATE) One RCT provided evidence that, compared with baseline, there was a reduction in AROM in hip extension at 6 months in children who received SDR plus physical therapy compared with an increase in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The same RCT provided evidence that, compared with baseline, there was less improvement in AROM in hip extension at 12 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

One non-randomised prospective study provided evidence that, compared with baseline, there was a greater improvement in AROM in hip flexion or extension at 8 months and 20 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) The authors reported that the increase in AROM in children who received SDR plus physical therapy compared with those who received physical therapy alone was statistically significant at 20 months, but not at 8 months (mean final score comparison across groups). One RCT reported that, compared with baseline, there was a statistically significant improvement in range of movement at the hip joint at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone. (HIGH) One RCT provided evidence that, compared with baseline, there was less improvement in PROM in hip extension at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The same RCT provided evidence that, compared with baseline, there was greater improvement in PROM in hip extension at 12 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE)

With regard to knee joints, one RCT provided evidence that, compared with baseline, there was a reduction in tone at the knee joint (evaluated using MAS scores) at 6 months and 12 months in children who received SDR plus physical therapy compared with no change in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the reduction in tone in children who received SDR plus physical therapy compared with children who received physical therapy alone was statistically significant at 6 months and 12 months (mean final score comparison across groups). One RCT reported that, compared with baseline, there was a statistically significant reduction in tone (evaluated using MAS scores) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone. (MODERATE)

One RCT provided evidence that, compared with baseline, there was an improvement in AROM in knee extension at 6 months in children who received SDR plus physical therapy compared with a deterioration in those who received physical therapy alone, and a greater improvement at 12 months in children who received SDR plus physical therapy compared with those who receive physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were not statistically significant. One non-randomised prospective study provided evidence that, compared with baseline, there was a greater improvement in AROM in knee flexion or extension at 8 months and 20 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) The authors reported that the differences in mean final scores across treatment groups were not statistically significant. One RCT reported that, compared with baseline, there was a statistically significant improvement in range of movement at the knee joint at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone. (HIGH) One non-randomised prospective study provided evidence that, compared with baseline, there was a greater reduction in AROM in knee flexion at initial contact at 8 months n children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) The same non-randomised prospective study provided evidence that, compared with baseline, there was a reduction in AROM in knee flexion at initial contact at 20 months in children who received SDR plus physical therapy compared with an improvement in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (VERY LOW) The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

One RCT provided evidence that, compared with baseline, there was less of an improvement in PROM in knee extension at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The same RCT provided evidence that, compared with baseline, there was greater improvement in PROM in knee extension at 12 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were not statistically significant. One RCT provided evidence that, compared with baseline, there were reductions in PROM of the popliteal angle at 6 months and 12 months in children who received SDR plus physical therapy compared with improvements in those who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) No further details relating to the popliteal angle were reported. The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

With regard to the ankle joint, one RCT provided evidence that, compared with baseline, there was a reduction in tone at the ankle joint (evaluated using MAS scores) at 6 months and 12 months in children who received SDR plus physical therapy compared with no change in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the reduction in tone in children who received SDR plus physical therapy compared with those who received physical therapy alone was statistically significant at 6 months and 12 months (mean final score comparison across groups). One RCT reported that, compared with baseline, there was a statistically significant reduction in tone (evaluated using MAS scores) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone. (HIGH) One RCT provided evidence that, compared with baseline, there were improvements in AROM in ankle dorsiflexion at 6 months and 12 months in children who received SDR plus physical therapy compared with reductions in function in those who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were statistically significant.

One non-randomised prospective study provided evidence that, compared with baseline, there was an improvement in AROM in ankle dorsiflexion or plantarflexion at 8 months in children who received SDR plus physical therapy compared with no change in those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) The same study provided evidence that, compared with baseline, there was less improvement in AROM in ankle dorsiflexion or plantarflexion at 20 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) The authors reported that the differences in mean final scores across treatment groups were not statistically significant. One RCT reported that, compared with baseline, the improvement in range of movement at the knee joint was greater at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, althoughy the statistical significance of these findings could not be determined. (MODERATE) One non-randomised prospective study provided evidence that, compared with baseline, there was an improvement in AROM in ankle dorsiflexion or plantarflexion at initial contact at 8 months in children who received SDR plus physical therapy compared with no change in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (VERY LOW) The same non-randomised prospective study provided evidence that, compared with baseline, improvements in AROM in ankle dorsiflexion or plantarflexion at initial contact at 20 months in children who received SDR plus physical therapy were the same as those who received physical therapy alone, although the statistical significance of this finding could not be determined. (VERY LOW) The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

The same study provided evidence that, compared with baseline, there was a greater reduction in extension foot progression angle at 8 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (VERY LOW) At 20 months, there was a reduction in extension foot progression angle in children who received SDR plus physical therapy compared with an improvement in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (VERY LOW) The authors reported that the differences in mean final scores across treatment groups were not statistically significant at 8 months, but they were statistically significantly different at 20 months.

One RCT provided evidence that, compared with baseline, there was a greater improvement in PROM in ankle dorsiflexion with knee extended at 6 months in children who received SDR plus physical therapy compared with children who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The same RCT provided evidence that, compared with baseline, there was improvement in PROM in ankle dorsiflexion with knee extended at 12 months in children who received SDR plus physical therapy compared with a reduction in children who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were statistically significant at 6 and 12 months. With regard to total MAS scores, one RCT provided evidence that, compared with baseline, there were greater improvements at 6 months (MODERATE) and 12 months (LOW) in children who received SDR plus physical therapy compared with those who received physical therapy alone. At 24 months children who received SDR plus physical therapy showed improvement in total MAS scores compared with no change in those who received physical therapy alone (MODERATE). The statistical significance could not be determined for any of these three findings. The authors reported that the differences in mean final scores across treatment groups were not statistically significant at 6 months but were statistically significant at 12 and 24 months.

Optimisation of function

With regard to GMFM-A (lying and rolling), one RCT provided evidence that, compared with baseline, there was less improvement in function (evaluated using GMFM-88) at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was not statistically significant. With regard to GMFM-A (lying and rolling), one RCT provided evidence that, compared with baseline, there was less improvement in function (evaluated using GMFM-88) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors did not report the statistical significance of the difference in mean final scores across treatment groups or of the difference in mean change from baseline scores across treatment groups.

Two RCTs reported results for GMFM-A (evaluated using GMFM-88) at 12 months. One RCT reported that, compared with baseline, there was a reduction in function in children who received SDR plus physical therapy compared with an improvement in those who received physical therapy alone that was not statistically significant. (LOW) The second RCT provided evidence that, compared with baseline, there was an improvement in function in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (LOW) The authors of the second RCT reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT reported that, compared with baseline, GMFM-88 scores for lying and rolling at 24 months were lower in children who received SDR plus physical therapy compared with those who received physical therapy alone that was not statistically significant. (MODERATE)

With regard to GMFM-B (sitting), one RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM-88) at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM, version not reported) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors did not report the statistical significance of the difference in mean final scores across treatment groups or of the difference in mean change from baseline scores across treatment groups.

Two RCTs reported results for function (evaluated using GMFM-88) at 12 months. One RCT reported that, compared with baseline, there was a greater improvement in children who received SDR plus physical therapy compared with those who received physical therapy alone that was not statistically significant. (LOW) The second RCT provided evidence that, compared with baseline, there was a greater improvement in function in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (LOW) The authors of the second RCT reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT reported that, compared with baseline, GMFM-88 scores for sitting at 24 months were lower in children who received SDR plus physical therapy compared with those who received physical therapy alone that was not statistically significant. (MODERATE)

Concerning GMFM-C (crawling and kneeling), one RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM-88) at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM, version not reported) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors did not report the statistical significance of the difference in mean final scores across treatment groups or of the difference in mean change from baseline scores across treatment groups. Two RCTs reported results for function (evaluated using GMFM-88) at 12 months. One RCT reported that, compared with baseline, there was less improvement in function in children who received SDR plus physical therapy compared with those who received physical therapy alone but that this finding was not statistically significant. (LOW) The second RCT provided evidence that, compared with baseline, there was a greater improvement in function in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (LOW) The authors of the second RCT reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT reported that, compared with baseline, GMFM-88 scores for crawling and kneeling at 24 months were lower in children who received SDR plus physical therapy compared with children who received physical therapy alone that was not statistically significant. (MODERATE)

With regard to GMFM-D, one RCT provided evidence that, compared with baseline, GMFM-88 scores for standing at 6 months were higher in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was statistically significant. One RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM, version not reported) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors did not report the statistical significance of the difference in mean final scores across treatment groups or of the difference in mean change from baseline scores across treatment groups.

Two RCTs reported results for function (evaluated using GMFM-88) at 12 months. Both RCTs reported that, compared with baseline, there were greater improvements in standing in children who received SDR plus physical therapy compared with those who received physical therapy alone. The findings for the first RCT were not statistically significant and the statistical significance of the findings for the second RCT could not be determined. (LOW) The authors of the second RCT reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT reported that, compared with baseline, GMFM-88 scores for standing at 24 months were lower in children who received SDR plus physical therapy compared with those who received physical therapy alone although this difference was not statistically significant. (MODERATE)

With regard to GMFM-E, one RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM-88) at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM, version not reported) at 9 months in children who received SDR plus physical therapy compared with children who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors did not report the statistical significance of the difference in mean final scores across treatment groups or of the difference in mean change from baseline scores across treatment groups.

Two RCTs reported results for function (evaluated using GMFM-88) at 12 months. Both RCTs reported that, compared with baseline, there were greater improvements in walking, running and jumping in children who received SDR plus physical therapy compared with those who received physical therapy alone. The findings for the first RCT were not statistically significant and the statistical significance of the findings for the second RCT could not be determined. (LOW) The authors of the second RCT reported that the improvement in function in children who received SDR plus physical therapy compared with those who received physical therapy alone was statistically significant (mean final score comparison across groups). One RCT reported that, compared with baseline, GMFM-88 scores for walking, running and jumping at 24 months were higher in children who received SDR plus physical therapy compared with those who received physical therapy alone although this difference was not statistically significant. (MODERATE)

With regard to total GMFM scores, one RCT provided evidence that, compared with baseline, there was a greater improvement in function (evaluated using GMFM-88) at 6 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was not statistically significant. One RCT provided evidence that, compared with baseline, there was a statistically significant improvement in function (evaluated using GMFM, version not reported) at 9 months in children who received SDR plus physical therapy compared with those who received physical therapy alone. (MODERATE) Two RCTs reported results for function (evaluated using GMFM-88) at 12 months. Both RCTs reported that, compared with baseline, there were greater improvements in function in children who received SDR plus physical therapy compared with those who received physical therapy alone. The findings for the first RCT were not statistically significant but they were for the second RCT. Pooled results for the two RCTs were not statistically significant (VERY LOW) The authors of the second RCT reported that the improvement in function in children who received SDR plus physical therapy compared with those who received physical therapy alone was statistically significant (mean final score comparison across groups). One RCT reported that, compared with baseline, GMFM-88 total scores at 24 months were lower in children who received SDR plus physical therapy compared with those who received physical therapy alone although this difference was not statistically significant. (MODERATE)

One non-randomised prospective study provided evidence that, compared with baseline, there was no difference between the improvements seen in both treatment groups in GMFM percent scores at 8 months, but at 20 months greater improvement was seen in children who received SDR plus physical therapy compared with those who received physical therapy alone. The statistical significance of these findings could not be determined. (VERY LOW) The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

With regard to walking, one RCT provided evidence, that compared with baseline, there was a less improvement at 6 months in the distance children were able to walk in 60 seconds when they had received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The same RCT provided evidence that, compared to baseline, there was an improvement in timed walking at 12 months in children who received SDR plus physical therapy compared with a reduction in children who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

One non-randomised prospective study provided evidence that, compared with baseline, there was an increase in gait speed at 8 months in children who received SDR plus physical therapy compared with a decrease in those who received physical therapy alone, although the statistical significance of this finding could not be determined. (VERY LOW) The same non-randomised prospective study provided evidence that, compared with baseline, there was a greater improvement in gait speed at 10 months in children who received SDR plus physical therapy compared with a reduction in children who received physical therapy alone, although the statistical significance of this finding could not be determined. (MODERATE) The authors reported that the differences in mean final scores across treatment groups were not statistically significant.

One RCT reported that, compared with baseline, there was less improvement in walking velocity at 12 months in children who received SDR plus physical therapy compared with those who received physical therapy alone, although the statistical significance of these findings could not be determined. (MODERATE) The authors reported that the difference in mean final scores across treatment groups was not statistically significant.

Quality of life

No studies reported quality of life.

Adverse effects

No studies reported mortality rates.

Although one RCT and one case series evaluated back pain as an outcome, the clinical importance of the results was unclear because the studies did not report whether the results excluded back pain experienced routinely in the first few days or weeks after any type of back surgery. (MODERATE) Lower extremity pain was reported in fewer children and young people in the SDR plus physical therapy group compared with the physical therapy-only group during a 24-month follow-up period. (MODERATE) A case series reported that all 208 patients experienced short-term post-operative back pain that was controlled well using intravenous fentanyl for 3 days. The incidence of long-term back pain was 3.4% (7/208) among children and young people who underwent SDR plus physical therapy. (VERY LOW)

Two case series reported outcomes related to urinary problems (bladder dysfunction). Across both case series 7.2% (33/458) of children who underwent SDR plus physical therapy experienced post-operative urinary retention. (VERY LOW) An RCT reported transient urinary retention in one of 14 children who underwent SDR plus physical therapy, and this resolved by the fourth post-operative day; no cases were reported in the physical therapy-only group. (MODERATE) One case series reported that 0.4% of children (1/250) who underwent SDR plus physical therapy required catheterisation 18 months after surgery. (VERY LOW) Another case series reported that 1% (2/208) of children who underwent SDR plus physical therapy experienced long-term urinary incontinence. (VERY LOW) One RCT reported that three of the 21 children and young people in the SDR plus physical therapy group experienced one urinary adverse event each during the 24-month follow-up period, compared with no events among the 17 children and young people in the physical therapy-only group. (MODERATE)

One case series reported an incidence rate of 1.2% (3/250) for post-operative transient ileus following SDR plus physical therapy. (VERY LOW)

One case series reported scoliosis rates in children following SDR surgery using laminectomy (L1 to S1) or laminoplasty (L1 to L5) and subsequent upper sacral laminectomy; 8.6% (5/58) of children and young people developed scoliosis after laminectomy and 1.3% (2/150) developed scoliosis after laminoplasty. (VERY LOW)

Both case series examined outcomes relating to hip dislocation. One study reported that 2.4% (6/250) of children and young people developed hip dislocation requiring a varus derotation osteotomy. In the other study, 1% (2/208) of children and young people developed progressive hip migration requiring orthopaedic surgery.

Acceptability and tolerability

No studies reported acceptability and tolerability.

Reduction of pain

The evidence relating to pain is presented above (under Adverse effects).

Selective dorsal rhizotomy plus physical therapy versus orthopaedic (soft tissue) surgery

Reduction of spasticity and optimisation of movement

No studies reported reduction of spasticity and optimisation of movement.

Optimisation of function

One non-randomised comparative study compared the effects of SDR and orthopaedic (soft tissue) surgery at 6, 12 and 24 months using PEDI self-care, mobility and social function domains within the functional skills and caregiver assistance scales. Compared with baseline, there was evidence that there were greater improvements that were not statistically significant in children who received SDR and physical therapy compared with those who received orthopaedic surgery for most assessments of function at 6, 12 and 28 months. However, there were greater improvements seen in children who received orthopaedic surgery compared with those who received SDR and phyiscal therapy for the self-care domain score, caregiver assistance scale, at 12 months and the social function domain score, caregiver assistance scale, at 6 months. These findings were not statistically significant. (VERY LOW)

The same study also compared the effects of SDR and orthopaedic (soft tissue) surgery at 6 months, 12 months and 24 months using the GMFM-88 assessment tool. With regard to GMFM-A scores, there were no changes from baseline for children who received SDR and physical therapy or for those who received orthopaedic surgery because all children could perform the lying and rolling task at baseline. (VERY LOW) With regard to GMFM-B and GMFM-C scores there was evidence that, compared with baseline, there were greater improvements in sitting and crawling and kneeling that were not statistically significant at 6, 12 and 28 months in children who received SDR and physical therapy compared with those who received orthopaedic surgery. (VERY LOW) With regard to GMFM-D scores there was evidence that, compared with baseline, there were reductions in standing function that were not statistically significant at 6, 12 and 28 months in children who received SDR and physical therapy compared with those who received orthopaedic surgery. (VERY LOW)

With regard to GMFM-D scores there was evidence that, compared with baseline, there were less improvements in standing function at 6, 12 and 28 months in children who received SDR and physical therapy compared with those who received orthopaedic surgery. These findings were not statistically significant. (VERY LOW) With regard to GMFM-E scores there was evidence that, compared with baseline, there were greater improvements in walking, running and jumping at 6 and 28 months in children who received SDR and physical therapy compared with those who received orthopaedic surgery, but those who received orthopaedic surgery alone showed greater improvement in walking, running and jumping at 12 months than children who received SDR and physical therapy. These findings were not statistically significant. (VERY LOW)

With regard to total GMFM-88 scores there was evidence that, compared with baseline, there was a greater improvement in overall function at 6 months in children who received SDR and physical therapy compared with those who received orthopaedic surgery. However, at 12 and 28 months, those who received orthopaedic surgery alone showed greater improvement in total GMFM-88 scores at 12 months than children who received SDR and physical therapy. None of these findings was statistically significant. (VERY LOW)

Quality of life

No studies reported quality of life.

Acceptability and tolerability

No studies reported acceptability and tolerability.

Reduction of pain

No studies reported reduction of pain.

Adverse effects

No studies reported adverse effects.

Other comparisons of interest

The GDG also prioritised evaluation of the following interventions and comparators, but no studies were identified for inclusion:

Health economics

No economic evaluations for SDR were identified in the literature search conducted for the guideline for this review question. The NICE Interventional Procedures Advisory Committee (IPAC) which developed Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010) had access to an unpublished dissertation presenting a pilot economic analysis of SDR in the UK (Edwards 2010). The cost analysis from the dissertation is outlined here (further details are presented in Chapter 11).

The clinical evidence identified relating to SDR included short- and medium-term outcomes (that is, outcomes measured at up to 24 months) for two treatment comparisons: SDR plus physical therapy versus physical therapy alone; and SDR plus physical therapy versus orthopaedic surgery (soft tissue surgery). In the comparison of SDR plus physical therapy with physical therapy alone, a statistically significant reduction in tone in lower extremity joints was reported, whereas no statistically significant difference was reported for timed walking, gait analysis, optimisation of function, individual dimensions of the GMFM or total GMFM scores. In the comparison of SDR plus physical therapy with soft tissue surgery, no evidence was identified relating to reduction of spasticity or optimisation of movement (this is not the same as identifying evidence of no difference). For optimisation of function, however, the evidence identified reported no statistically significant differences in individual domains of PEDI, individual dimensions of the GMFM or total GMFM scores.

The cost of SDR is approximately £25,362 and this includes the cost associated with 7 weeks of hospital inpatient rehabilitation (Edwards 2010). Further details of the resource use related to SDR are presented in Chapter 11. Since no good-quality long-term outcome data (that is, outcomes measured after 24 months and, preferably, into adult life) are available, it is not possible to determine whether the initial reduction in tone reported in the clinical evidence would lead to clinically important long-term benefits. Conducting a cost effectiveness analysis requires estimates of long-term outcomes, such as improvements in quality of life. The only statistically significant benefit reported in the clinical evidence reviewed for the guideline was a reduction in tone in lower extremity joints. However, the GDG was unable to extrapolate this to a clinically important long-term improvement in function that would represent an increase in quality of life. Based only on the available short- and medium-term clinical outcomes, SDR cannot be said to be cost effective in terms of the NICE £20,000 threshold for willingness to pay for a quality adjusted life year (QALY) gain.

Evidence to recommendations

Relative value placed on the outcomes considered

SDR is a procedure intended to reduce muscle spasticity and so outcome measures should focus on changes in tone in relevant muscles. In particular, the GDG wished to know if reduced tone resulted in improvements in function, including the child or young person's abilities in terms of self-care and walking (such as speed of walking). Independence in the tasks of daily living that required walking and standing were considered important. Measures of stamina (distance walked in a given time) were not reported in the evidence identified for inclusion.

Much of the evidence reported findings in terms of scores intended to measure changes in muscle tone (for example Ashworth scores) or range of movement for a particular joint. The GDG considered these findings far less valuable than those relating to function, independence or quality of life as they found it difficult to interpret the reported scores in a clinically or socially meaningful way.

Pain is a symptom of spasticity and presence of pain affects quality of life. The GDG considered reduction in pain to be an important outcome measure.

As SDR is an irreversible procedure, the risks of complications of the surgery, including non-specific risks (such as infection) associated with other types of surgery and the specific complications of cutting dorsal nerve roots and performing laminectomy, are critical in decision making where the benefits for the child or young person may be marginal over time.

The GDG considered that the ideal long-term outcome would be the ability to maintain independent walking into adult life, but the evidence identified for inclusion did not report that length of follow-up.

Consideration of clinical benefits and harms

Some short- and medium-term improvements in motor function, as measured by individual dimensions of the GMFM or GMFM total scores, were statistically significant. However, even for those dimensions where such effects were demonstrated (for example GMFM-D (standing) or GMFM total score) the effects were not consistent or sustained across all durations of follow-up considered in the evidence (6–24 months). The GDG considered that if the observed improvements could be maintained through to adult life then the outcomes of SDR would be clinically important. The improvements take time to appear, however, and the GDG believed that in the first 6–12 months after the procedure, quality of life for the child or young person and their family may decrease temporarily because of post-operative adverse effects of the surgery itself, the need for a period of inpatient physical therapy and the prolonged rehabilitation period that follows.

The short- and medium-term reductions in spasticity and optimisation of movement demonstrated by improvements in muscle tone or range of movement in hip, knee and ankle joints were not consistent or sustained across all durations of follow-up considered in the evidence (6–24 months).

Although the risks of permanent morbidity following surgery are low, the potential consequences are serious. Children and young people, and their parents and carers, should be informed about the risks so that they can make informed decisions. The GDG noted differences in techniques for exposing dorsal nerve roots (laminectomy) and considered whether better exposure reduced the risks of damage to roots from the skin, bladder or bowel. The GDG noted that in one published study laminectomy of L1 to S1 was associated with a greater incidence of post-operative scoliosis than laminoplasty of L1 to L5 followed by upper sacral laminectomy.

Pre-existing muscle shortening and bony deformity may interfere with post-operative rehabilitation and limit improvement in motor function. If surgery is postponed the child or young person will need to undergo a further period of post-operative recovery. It may take a child or young person up to 2 years to recover fully from major orthopaedic surgery, and so it may be appropriate to consider performing orthopaedic surgery before or at the same time as SDR.

The non-randomised study that compared SDR and orthopaedic (soft tissue) surgery showed no significant differences between the two treatment groups in relation to any of the outcomes reported. The GDG noted, however, that the evidence from this study was of very low quality and concluded that it did not support a recommendation to offer soft tissue surgery instead of SDR, although the GDG recognised that SDR and orthopaedic surgery might be performed sequentially for some children and young people.

SDR does not avoid the need for orthopaedic surgery in the longer term. The onset of muscle shortening, bone or joint deformity or scoliosis may cause pain or impair function and it is important, therefore, that the child or young person is offered regular reviews until they are fully grown (when the risk of new orthopaedic complications is much lower). Once a child or young person has undergone SDR, the epidural space is obliterated and epidural anaesthesia during subsequent orthopaedic surgery, or during childbirth, will not be possible.

The GDG considered that rehabilitation after SDR is a process that would continue until the child or young person was fully grown and it requires, therefore, a long-term commitment from the child or young person and their family or carers. There might be need for further periods of intensive inpatient rehabilitation involving physiotherapy and use of additional or different orthoses compared with before surgery. Post-operative weakness in leg muscles is common, and targeted strength training will be an important component of post-surgery physical therapy. Orthoses and other supportive devices (such as walking frames) may be required to allow the child or young person to practice new skills and gain strength and balance. The GDG recognised that children and young people may gain weight after SDR and this may affect rehabilitation and motor function adversely.

The GDG considered that it would be important to ensure that the commitment required to follow a rehabilitation programme after SDR did not adversely affect other aspects of the child or young person's life (such as education).

The GDG concluded that while long-term reduction in spasticity might be expected after SDR, evidence for a long-term improvement in gross motor function was lacking.

Consideration of net health benefits and resource use

The GDG considered that the high initial cost of SDR would be justified only if improvements in motor function were maintained into adult life (for example if the child or young person were to progress through one or more levels of the GMFCS). Alternatively, if a clinically important improvement in quality of life following SDR could be demonstrated then the procedure might be shown to be cost effective even in the absence of progression in terms of GMFCS levels.

The GDG also considered that a sustained reduction in spasticity might reduce the long-term requirement for targeted resources, such as physical therapy, orthoses and mobility equipment, and this could result in significant cost savings to the NHS.

Quality of evidence

The quality of the evidence for reductions in spasticity and optimisation of movement ranged from very low to high. The quality of the evidence for improvement in function also ranged from very low to high. None of the evidence addressed long-term outcomes (that is, more than 24 months after surgery and preferably through to adult life). The interventions and comparators evaluated in the included studies varied in relation to:

  • the numbers of nerve roots divided and spinal segment levels involved (for example, in one RCT a mean of only 26% of sensory nerve rootlets were divided, which does not reflect the procedure as it is undertaken currently)
  • the content of the physical therapy components of the interventions and comparators (in one study the children and young people who underwent SDR received a more intensive initial physical therapy programme than did the physical therapy-only group).

The numbers of children and young people involved in the studies were small and no subgroup analyses were undertaken to try to identify clinical characteristics that might be associated with better outcomes after SDR.

Other considerations

The GDG considered that children and young people undergoing SDR should be followed up according to a standardised framework until they reached adulthood. Given the lack of good quality outcome data, the GDG further considered that anonymised data should be collected through a national audit of outcomes of SDR procedures that have already been undertaken, including long-term outcomes and adverse effects. Since any centre offering SDR is likely to perform the procedure on only a small number of children and/or young people each year, a national audit would allow more rapid collection of robust data, with the potential for comparing different centres in the long term, provided the same validated outcome measures are recorded in each centre. Collating and publishing data on adverse effects would provide information about the benefits and risks associated with SDR, and this would be of importance to children and young people considering SDR and their parents and carers. Such data might also allow comparisons to be made between outcomes of different practices or techniques used during SDR, such as the extent of bone removal and the number of rootlets cut.

In formulating their recommendations, the GDG members considered existing guidance contained in Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010). In particular, the GDG noted the importance of care being delivered by a multidisciplinary team with specialist training and expertise in the care of spasticity and with access to the full range of treatment options. Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010) emphasises that the SDR team would normally include a physiotherapist, a paediatrician and surgeons, all with specific training and expertise. The GDG recognised that current practice is to coordinate all aspects of clinical care for children and young people who have spasticity and co-existing motor disorders and their early musculoskeletal complications as a result of a non-progressive brain disorder through multidisciplinary teams comprising similar groups of healthcare professionals to those recommended in Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010). The group recommended that, as a general principle, care for children and young people with spasticity should be delivered in the context of a network of care that incorporates integrated team working, agreed care pathways and effective communication (see Chapter 4).

Key conclusions

In the experience of the GDG, many children and young people have serious difficulties with walking because of the degree of spasticity that is present, as well as other related problems that affect walking, such as weakness and poor selective motor control. The GDG recognised the longstanding knowledge of neurophysiological processes that result in spasticity, including the theoretical basis for expecting SDR to reduce muscle tone. The limited evidence available demonstrated that SDR does indeed reduce tone, and the GDG recognised that there was no reason to suspect that tone would increase subsequently (over a period of years) because the procedure is irreversible. There was, however, a lack of evidence supporting any clinical benefit of SDR in relation to optimisation of function. The GDG highlighted the evidence suggestive of benefit in this area, particularly the improvements in the GMFM-D (standing) at 6 months and the GMFM total score at 9 months (although these effects were not consistently observed across all studies nor sustained across all periods of follow-up, and most of the evidence was of low or moderate quality). No evidence at all was identified relating to quality of life.

The GDG concluded that a strong recommendation to offer SDR could not be supported in the absence of high quality evidence of a consistent and sustained (long-term) improvement of more than 2 years (for example sustained into adult life) in motor function or pain control. GDG members were aware that anecdotal evidence from an unpublished dissertation (Edwards 2010) suggests that in appropriately selected children and young people SDR may achieve such outcomes, but were of the view that further research, preferably conducted using RCTs or comparative observational studies, is needed to evaluate the clinical and cost effectiveness of the procedure in terms of long-term outcomes.

Nevertheless, the GDG considered that the available evidence, current clinical practice and existing NICE guidance supported a weak recommendation for the further clinical use of SDR as a treatment to improve walking ability in appropriately selected children and young people. Its decision to include this recommendation hinged on the caveat that centres offering SDR participated in national information-gathering activities that would increase the overall knowledge base for this treatment through pooling of outcome data. The GDG discussed and agreed six clinical criteria for identifying children and young people to whom SDR could be offered as part of research:

  • abnormal tone (pure spasticity)
  • good leg muscle strength
  • straight legs and minimal muscle shortening
  • good selective motor control in the legs
  • good cognitive skills
  • not overweight.

The GDG considered that the clinical pattern represented by the combination of the six criteria was most likely to be present in children and young people with cerebral palsy who have bilateral spasticity affecting the leg and who are in GMFCS level I, II or III. The GDG considered that the possible functional gain in children and young people in GMFCS level I was not sufficient to outweigh the risks of complications, and so would not recommend considering SDR for children and young people in this group. Children and young people in GMFCS level II or III were, however, thought likely to be able to derive the clinical benefit of improved walking ability through undergoing SDR. Thus the GDG recommended that SDR be considered for children and young people in GMFCS level II or III and it prioritised further research into the effectiveness and safety of SDR in this group of children and young people. The GDG also highlighted in its research recommendations the importance of physical therapy (particularly physiotherapy) as an adjunctive treatment to improve the chances of a successful outcome after SDR, since this reflected the evidence currently available.

In the GDG's view, SDR is more likely to be effective if spasticity is judged to be the major factor impairing movement. If weakness, dystonia, poor motor control or musculoskeletal deformities are the main causes of motor impairment, then SDR is much less likely to be effective. Poor selective motor control and dystonia will not be improved by SDR and will significantly affect the child or young person's ability to benefit from physical therapy during rehabilitation. Muscle weakness may become apparent immediately after SDR and the child or young person will require intensive strengthening physiotherapy.

No evidence was identified to support the use of SDR in more severely affected children, in children with unilateral spasticity affecting the leg, or in children and young people who have spasticity as the result of a head injury. The GDG acknowledged that in more severely affected children and young people, pain from spasticity affects quality of life and using SDR to reduce spasticity even when there is no likelihood of improved function might be justified once other treatments have been considered or used. The available evidence was, however, considered to be insufficient to recommend SDR in this context without further research.

The GDG noted that severe scoliosis might make SDR more difficult to perform. The GDG also noted that hip dislocation, unless surgically corrected before SDR, would reduce the effectiveness of the procedure and make post-operative rehabilitation difficult (because the child or young person might be in pain, and sitting and standing might be difficult).

The post-operative rehabilitation period places significant demands on the child or young person and their family or carers. Providing physical therapy regularly for up to 2 years after performing SDR may present difficulties when children and young people live in geographically remote areas. Physical therapists may need to rely heavily on the child or young person's parents and other family members or carers to supervise exercises, and this could have an impact on family life, including quality of life for parents and siblings. Children and young people with spasticity and co-existing learning difficulties or sensory impairments might have difficulty coping with rehabilitation programmes, and this would need to be considered carefully by parents or carers before consenting to treatment. Further research should, therefore, consider the practicalities of life for children and young people who have undergone SDR and their parents or carers, and how healthcare services can be developed to support families in a variety of circumstances.

The GDG recognised that SDR is one of a number of treatment options for children and young people and stressed that healthcare professionals might prefer to consider treatments with lower risks of adverse effects. Alternative treatments might include BoNT-A injections or ITB. However, ITB has mainly been considered for children and young people with more severe disability than those undergoing SDR and no evidence was identified to allow comparison of the clinical benefits and harms of SDR with those of ITB. Nevertheless, SDR is irreversible, and so everyone involved in making decisions about whether to choose SDR should first ensure that the procedure is appropriate for the individual child or young person.

The GDG recognised that children and young people and their parents or carers may wish to explore all available treatment options. Despite SDR being used in USA since the 1980s, and more than 1000 children and young people having undergone the procedure in one centre, there is no good quality evidence that the procedure results in clinically important improvements in motor function that are sustained over several years. The GDG was aware that SDR has been performed in several other countries during the past 30 years (including Australia, Canada, Japan and South Africa). Children who can walk with walking aids before the age of 10 years may lose that ability in their teenage years because of weight gain or further muscle shortening or weakness. The available evidence does not identify whether the loss of walking ability can be prevented by SDR. It is important that children and young people considering SDR, and their parents or carers, are aware of the shortcomings of the evidence. In formulating aspects of both their clinical and research recommendations relating to information for children and young people and their parents or carers, the GDG members mirrored existing guidance in Selective dorsal rhizotomy for spasticity in cerebral palsy (NICE IPG 373, 2010).

Recommendations

Table

Table

Copyright © 2012, National Collaborating Centre for Women's and Children's Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Cover of Spasticity in Children and Young People with Non-Progressive Brain Disorders
Spasticity in Children and Young People with Non-Progressive Brain Disorders: Management of Spasticity and Co-Existing Motor Disorders and Their Early Musculoskeletal Complications.
NICE Clinical Guidelines, No. 145.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2012 Jul.

NICE (National Institute for Health and Care Excellence)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...