Botulinum toxin type A in the management of equinus in children with cerebral palsy: an evidence-based economic evaluation.

INTRODUCTION: The aim of this study was to compare two methods of conservative management of calf spasticity and equinus gait--intramuscular injection of botulinum toxin type A (BTX-A) and serial casting. An economic evaluation framework was adopted to assess whether BTX-A offers value for money in the management of equinus gait due to calf spasticity in children with cerebral palsy. Short- to medium-term health care costs and outcomes were estimated for the comparison. This study was embarked upon to provide clinical and economic data as part of an application to the Pharmaceutical Benefits Advisory Committee (PBAC) for the reimbursement of BOTOX by the Australian Commonwealth Government. This is the primary mechanism for reimbursement of pharmaceuticals in Australia, as they are not routinely reimbursed through health insurance companies. The perspective of the analysis was that of the Australian health system. METHODOLOGY: Randomized controlled trials (RCTs) exist comparing one treatment cycle of BTX-A with serial casting (Corry et al., 1998; Flett et al., 1999). A long-term prospective study provided data on multiple cycles of BTX-A treatment in a more naturalistic setting (Boyd et al., 1999). A simple economic modelling approach was used to establish resource utilization by treatment arm. Only direct medical costs were considered (BTX-A, medical personnel time and medical consumables). MAIN MEASURES: Clinical efficacy was obtained from the randomized controlled trials (Corry et al., 1998; Flett et al., 1999). Patient/parent preference was obtained from long-term follow-up (Corry et al., 1998) and a preference questionnaire (Flett et al., 1999). Australian treatment patterns and patient demographics were obtained from the naturalistic study (Boyd et al., 1999). RESULTS: The RCTs demonstrated equivalent efficacy of BTX-A and serial casting; however, with BTX-A the effect lasted longer and was clearly the preferred treatment. For patients with hemiplegia the costs of an episode of treatment with BTX-A or serial casting are ($AUD) $595 and $435, respectively, and thus the additional costs associated with BTX-A are $160. The corresponding costs for patients with diplegia are $1045 for BTX-A treatment and $870 for serial casting and thus the additional cost associated with BTX-A is $175. With an overall treatment duration of 3.7 years and an average treatment interval of 10 months, patients would receive an average of 5.4 treatments. Thus, for patients with hemiplegia the total additional cost, discounted at 5% annually, for BTX-A is $793. For patients with diplegia the total additional cost for BTX-A is $867. CONCLUSIONS AND CLINICAL INTERPRETATIONS: BTX-A is an effective, safe and acceptable treatment modality and is associated with only a modest increase in direct medical costs per child per year. BTX-A can be considered a valuable and cost-effective treatment in the conservative management of equinus due to calf spasticity in children with cerebral palsy. This conclusion is supported by the acceptance of the Pharmaceutical Benefits Advisory Committee (PBAC) recommendation that BOTOX should attract a full Government subsidy in Australia.

Botulinum toxin A in hamstring spasticity.

Hamstring injection of Botulinum toxin A (BtA) may have a role in the conservative management of flexed knee gait in cerebral palsy or in simulating the effect of surgery. Ten children who were likely to require future hamstring lengthening were injected. Short term outcome was assessed by clinical examination and 3-D gait analysis. Mean popliteal angle decreased by 16 degrees and maximum knee extension in stance increased by 8 degrees, the latter relapsing by 12 weeks. Mean pelvic tilt tended to increase suggesting that isolated hamstring weakening be approached with caution. Energy cost of walking was not significantly changed in six of the ten patients. A small increase in knee extension in stance was often associated with patient satisfaction. There are theoretical grounds for expecting an associated increased longitudinal muscle growth after BtA injection.

Arthroscopic reconstruction of the anterior cruciate ligament. A comparison of patellar tendon autograft and four-strand hamstring tendon autograft.

We compared the outcome of anterior cruciate ligament reconstruction using hamstring tendon autograft with outcome using patellar tendon autograft at 2 years after surgery. Patients had an isolated anterior cruciate ligament injury and, apart from the grafts, the arthroscopic surgical technique was identical. Prospective assessment was performed on 90 patients with isolated anterior cruciate ligament injury undergoing reconstruction with a patellar tendon autograft; 82 were available for follow-up. The hamstring tendon autograft group consisted of the next 90 consecutive patients fulfilling the same criteria; 85 were available for follow-up. Clinical review included the Lysholm and International Knee Documentation Committee scores, instrumented testing, thigh atrophy, and kneeling pain. These methods revealed no difference between the groups in terms of ligament stability, range of motion, and general symptoms. Thigh atrophy was significantly less in the hamstring tendon group at 1 year after surgery, a difference that had disappeared by 2 years. The KT-1000 arthrometer testing showed a slightly increased mean laxity in the female patients in the hamstring tendon graft group. Kneeling pain after reconstruction with the hamstring tendon autograft was significantly less common than with the patellar tendon autograft, suggesting lower donor-site morbidity with hamstring tendon harvest.

Musculoskeletal modelling in determining the effect of botulinum toxin on the hamstrings of patients with crouch gait.

This study aimed to determine the effect of hamstring botulinum toxin A (Btx-A) injection in 10 children with crouch gait in terms of changes in muscle length and lower-limb kinematics. Before Btx-A injection limb kinematics were recorded. Maximum hamstring lengths and excursions were calculated by computer modelling of the lower limb. Data were compared with the averaged hamstring lengths of 10 control children. Hamstrings were defined as short if their length was shorter than the average maximum length minus one standard deviation. Gait analysis was repeated 2 weeks after isolated hamstring Btx-A injection. Pre- and postinjection kinematic data and muscle lengths were then compared. Four of 18 injected limbs in three subjects had short medial hamstring before injection, none of the subjects had short lateral hamstrings. Muscle excursion was significantly reduced in the short and adequate maximum muscle length groups. A significant increase in the semimembranosus and semitendinosus length in all of the injected limbs was noted. Only in the short muscle group was a significant increase in muscle excursion observed. Knee extension improved by 13 degrees in the adequate muscle length group and by 15.6 degrees in the short muscle length group. Pelvic tilt and hip flexion increased in both groups non-significantly. Average walking speed postinjection increased from 0.60 ms(-1) to 0.71 ms(-1). Short hamstrings are over-diagnosed in crouch gait. Hamstring Btx-A injection in patients with crouch gait produces significant, repeatable muscle lengthening and improved ambulatory function.

Botulinum toxin A compared with stretching casts in the treatment of spastic equinus: a randomised prospective trial.

Conservative therapies for equinus in cerebral palsy may help to postpone calf surgery in younger children. This study reports a prospective randomised trial of intramuscular botulinum toxin A (BtA) as an alternative to serial casting in 20 children with a dynamic component to calf equinus. Outcome was assessed in the short term to show the effect of one treatment cycle. Assessments were by clinical examination, video gait analysis, and three-dimensional gait analysis. BtA was of efficacy similar to that of serial casting. Tone reduction in the BtA group allowed a more prolonged improvement in passive dorsiflexion, which may allow more opportunity for increase in muscle length. Gait analysis showed an improved mean ankle kinematic pattern in a subsection of both groups, which was maintained at 12 weeks in the BtA group, whereas the cast group relapsed. There were fewer side effects in the BtA group. Median time to reintervention was similar.

Endoscopic reconstruction for isolated anterior cruciate ligament rupture.

We treated 90 patients with an isolated injury to the anterior cruciate ligament (ACL) by reconstruction using a patellar tendon autograft and interference screw fixation. Of these, 82 (91%) were available for review at 24 months. Two grafts and two contralateral ACLs had ruptured during sport and there was one case of atraumatic graft resorption. Using the assessment of the International Knee Documentation Committee (IKDC), 86% of the remaining patients were normal or nearly normal. The median Lysholm knee score was 95/100 and 84% of patients were participating in moderate to strenuous activity. All had grade-0 or grade-1 Lachman, pivot-shift and anterior-drawer tests. Measurement with the KT1000 arthrometer gave a side-to-side difference of <3 mm of anterior tibial displacement in 90%. Sixty-six radiographs were IKDC grade A and one was grade B. Pain on kneeling was present in 31% and graft site pain in 44%. At 24 months after operation all patients had excellent knee stability, a high rate of return to sport and minimal radiological evidence of degenerative change. Our series therefore represents a basis for comparison of results using other techniques and after more severe injuries.

Integration of hamstring tendon graft with bone in reconstruction of the anterior cruciate ligament.

Anterior cruciate ligament (ACL) reconstruction using four-strand hamstring graft with round-headed, cannulated, interference (RCI) screw fixation requires osteointegration of the tendon graft. This report describes the histology at the bone-tendon junction of two specimens retrieved from patients undergoing revision surgery after traumatic mid-substance ACL graft rupture at 6 and 10 weeks after initial reconstruction. Revision was performed at 12 and 15 weeks. Integration of the graft was evident by observation of collagen fiber continuity between bone and tendon. This histology plus the low incidence of early graft failure suggest that free tendon autograft attached to bone by RCI screw allows adequate osteointegration between 6 and 15 weeks after surgery.

Botulinum toxin A in the hemiplegic upper limb: a double-blind trial.

In a randomised, double-blind study, the effects of intramuscular injection of botulinum toxin type A (BtA) into the upper limb were compared with those of normal saline solution in 14 patients with cerebral palsy; their mean age was 9 years. Range of movement and function were assessed before injection and at 2 and 12 weeks after injection. BtA injection significantly increased maximum active elbow and thumb extension and significantly reduced tone at wrist and elbow. The hand grasp-and-release score improved, representing a modest functional change, but fine motor function, assessed by the ability to pick up coins, did not improve and in some cases deteriorated temporarily. The most notable subjective change was the cosmetic benefit of reduced involuntary elbow flexion. The tone-reducing effect of BtA was clinically detectable in comparison with the placebo and patients and parents perceived the change as beneficial. The median of changes in the treatment group was small but the range was large, suggesting that BtA can be useful in selected patients.

Three-dimensional gait analysis in spina bifida.

This study was designed to determine gait patterns in children with lumbar and sacral neurologic level spina bifida. We studied a group of 28 children: 10 had L4-level lesions and a mean age of 11 years; eight had L5-level lesions and a mean age of 8 years; and 10 had S1-level lesions with a mean age of 12 years. A group of 15 normal children, mean age 10 years, was used for comparison. Each child underwent three-dimensional gait analysis using the Vicon system. We found that there were recognisable gait patterns for each level of spina bifida and that the abnormalities accurately reflected the muscle deficiencies present. The gait patterns approximated more closely to those of the normal group as the neurological level descended. The most important findings were of increased pelvic obliquity and rotation with hip abduction in stance (reflecting the gross Trendelenburg-type gait seen in these children) and persistent knee flexion throughout stance as a result of the absence of the plantar flexion-knee extension couple. We found that gait was not improved by tendon transfers performed either at the hip (posterolateral psoas transfer) or at the ankle (tibialis anterior transfer).

Measurement of oxygen consumption in disabled children by the Cosmed K2 portable telemetry system.

Measurement of the energy cost of walking in children with cerebral palsy or spina bifida is difficult due to the cumbersome nature of equipment used to assess oxygen consumption. Such information collected with a lightweight telemetric system, the Cosmed K2, correlated well with that from a non-portable breath-by-breath system associated with a treadmill. The K2 did not significantly affect regular gait pattern as measured by gait analysis, and repeatability was satisfactory. Measurement of the energy cost of walking in the individual is unreliable in detecting differences of less than 10%; comparison between groups is more useful.

Energy consumption in children with spina bifida and cerebral palsy: a comparative study.

The authors looked for differences in the energy expenditure patterns of ambulant children with cerebral palsy and spina bifida. Oxygen consumption was measured according to type of cerebral palsy or level of spina bifida lesion, and in healthy children. The rate of oxygen consumption (mL/kg/min) was significantly higher in the children with diplegia than in those with hemiplegia or with spina bifida or the healthy children. Oxygen cost (mL/kg/m) was significantly higher and velocity was significantly slower in all the groups with disability than in the healthy children. The reason children with diplegia consumed more oxygen than other children when walking may be that their abnormal equilibrium reactions impaired their balance and their ability to control their walking speed.

Limb length after fracture of the femoral shaft in children.

Fifty children, aged < or = 10 years, with isolated fractures of the femoral shaft had scanograms performed at fracture union and a mean 3.9 years later (SD, 1.6 years). Overgrowth was a mean 6.9 mm (95% confidence limits, 5.2-8.6 mm). Age was an influential variable, overgrowth being less in children < 4 and those > 7 years. At fracture reduction, much less shortening should be accepted than is commonly recommended.

Botulinum toxin in the management of the lower limb in cerebral palsy.

The role of intramuscular botulinum toxin A in the treatment of 26 children with cerebral palsy was evaluated. The indication for injection was the presence of a dynamic contracture of lower-limb muscles interfering with positioning or walking. Spastic target muscles were identified by clinical examination and, in ambulant children, by gait analysis. Between 50 and 320 units of botulinum toxin were injected into each muscle group to a total dose of 100 to 400 units per child. The effects of injection were monitored by repeated clinical examination and gait analysis. There were no clinically detectable systemic side-effects, and all but one patient had a reduction in tone, which occurred within three days and persisted for two to four months. There were significant improvements in ambulatory status and in sagittal-plane kinematics. In some cases these gains persisted after the tone-reducing effects of the toxin had worn off.

Maximal aerobic power measurement in runners and swimmers.

Five cross-country runners and five competitive swimmers performed a pulling exercise with elastic shock cords and a treadmill run to exhaustion. The mean VO2 max related to lean body mass of the runners was significantly higher than the swimmers on the treadmill (p less than 0.05) while, on the pulling test, the mean VO2 max of the swimmers was significantly higher than the runners (p less 0.01). The maximum heart rates achieved pulling were 95% of the running maximum by runners and 96% by swimmers with no significant difference between them. Their mean oxygen pulse was almost the same for maximal running but the swimmers had a significantly higher oxygen pulse than the runners for maximal pulling (p less than 0.01). The swimmers could reach about 79% of their running VO2 max by pulling while the runners used 53% of their running VO2 max.