The influence of age at single-event multilevel surgery on outcome in children with cerebral palsy who walk with flexed knee gait

Transcription

1 DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE The influence of age at single-event multilevel surgery on outcome in children with cerebral palsy who walk with flexed knee gait MARTIN ƒvehlýk 1,2 GERHARD STEINWENDER 1 TANJA KRAUS 1 VINAY SARAPH 1 THOMAS LEHMANN 3 WOLFGANG E LINHART 1 ERNST B ZWICK 1 1 Paediatric Orthopaedic Unit, Department of Paediatric Surgery, Medical University of Graz, Graz, Austria. 2 Orthopaedic and Traumatology Department for Children and Adults, Charles University, Prague, Czech Republic. 3 Institute of Medical Statistics, Computer Sciences and Documentation, Jena University Hospital, Jena, Germany. Correspondence to Dr Martin ƒvehlík at Paediatric Orthopaedic Unit, Department of Paediatric Surgery, Medical University of Graz, Auenbruggerplatz 34, Graz A-8036, Austria. martin.spejlik@seznam.cz This article is commented on by Hoffinger on page 678 of this issue. PUBLICATION DATA Accepted for publication 17th March Published online 21st June ABBREVIATION GDI Gait Deviation Index AIM Information on the timing and long-term outcome of single-event multilevel surgery in children with bilateral spastic cerebral palsy (CP) walking with flexed knee gait is limited. Based on our clinical experience, we hypothesized that older children with bilateral spastic CP would benefit more from single-event multilevel surgery than younger children. Moreover, any improvement in older children could be maintained with fewer additional surgery events. METHOD We performed a retrospective analysis of the long-term outcomes of single-event multilevel surgery. Thirty-two children (17 males, 15 females) who had received single-event multilevel surgery between 1995 and 2000 with a mean age at the time of surgery of 10 years 6 months (range 5y 8mo 15y 6mo; SD 3y 1mo) and in Gross Motor Function Classification System level II (n=12) or III (n=20) were included in the study. The inclusion criteria required that all children were ambulatory with spastic bilateral CP, had a flexed knee gait, had a full set of data for single-event multilevel surgery preoperatively and at 1 year and 10 years postoperatively, had not had previous surgery on their lower limbs, had not had any treatment with botulinum toxin A before gait assessment, and had not received intrathecal baclofen medication. The follow-up time lasted for over 10 years until the participants reached adulthood (mean age at the last follow-up 21 years 4 months, SD 3y 4mo). Data were collected on six separate occasions: preoperatively, at 1 year, at 2 to 3 years, at 5 years, at 7 to 8 years, and at 10 or more years postoperatively. The primary outcome was the Gait Deviation Index, and the secondary outcomes were the number and type of initial and additional surgeries. A linear mixed model and Spearman s rank correlation coefficient were used to prove the hypothesis. RESULTS The older the child was at the time of the surgery, the better the long-term result ( b b Age,Time =0.15; p=0.03). We did not find any correlation between age at the time of surgery and the number of bony or soft-tissue procedures performed initially as well as during the 10 years of follow-up. INTERPRETATION Children with CP who require single-event multilevel surgery at an older age fare better in the long term than those who are younger at the time of surgery. The pubertal growth spurt is discussed as a contributing factor to gait deterioration. The treatment of ambulatory children with cerebral palsy (CP) requires an understanding of the interactions of existing motor system deficits and biomechanical function in the context of growth and development. Spasticity is the most common type of abnormal muscle tone in children with CP. Spastic muscles fail to grow in proportion with the bone and present dynamic tightness in young children. Most of these muscles develop fixed contractures over time. 1 Inappropriate muscle activity exerts abnormal forces on the growing skeleton, which leads to secondary bony deformities and joint instability. The interaction of joint contractures, muscle weakness, bony deformities, and joint instability occurring at multiple levels of the lower limbs affects the quality and efficiency of gait in children with CP. Therefore, the natural progression of gait in children with bilateral spastic CP is characterized by deterioration of gait kinematics and time distance parameters. 2 In the 1980s, state-of-the-art orthopaedic treatment of children with CP involved staged procedures on a yearly basis, which was often referred to as the birthday surgery syndrome. Rang 3 was probably the first to propose the method of single intervention at multiple levels. Single-event multi- 730 DOI: /j x ª The Authors. Developmental Medicine & Child Neurology ª 2011 Mac Keith Press

2 levl surgery entails correction of all the existing soft tissue and bony deformities at the same time, with the goal of having a single rehabilitation period. A favourable outcome of singleevent multilevel surgery has been described previously for ambulatory 4 as well as for non-ambulatory 5 children with bilateral spastic CP. However, reports on the long-term outcome of this type of orthopaedic intervention are lacking. Rodda et al. 6 published a comprehensive study on the operative treatment of severe crouch gait and reported on 5-year results in 10 children with spastic CP. Saraph et al. 7 demonstrated changes in gait parameters over 3 years after singleevent multilevel surgery in a group of children with spastic diplegia. They proposed that any evaluation of gait improvement surgery in CP should be performed at a minimum of 3 years after surgery to provide a sound evaluation of the treatment outcome. The follow-up times of other studies are even shorter. 8,9 Therefore, the evidence that guides the timing of multilevel surgery is very limited. There is a commonly held opinion that orthopaedic surgical interventions to improve gait should be postponed until motion patterns are well established. Interventions that are performed at an early age are believed to be associated with an increased risk of failure and relapse with less predictable results. 10,11 Molenaers et al. 12 performed a retrospective review of 424 children with CP and showed that the introduction of gait analysis increases the age at which the first orthopaedic surgical procedures are performed and that botulinum toxin type A treatment delays and reduces the frequency of surgical procedures. On the other hand, a recent report by Gough et al. 9 showed that surgical interventions to correct deformities in young ambulant children with bilateral spastic CP led to a favourable outcome when compared with a group of non-operated children. Furthermore, Sussman and Aiona, 3 in their review of treatment of spastic diplegia in children with CP, suggested that younger children may benefit from staged and initially minor operative procedures. As there is limited information on the timing and long-term outcome of single-event multilevel surgery in children with bilateral spastic CP who walk with a flexed knee gait, we decided to review our 10 years or more of postoperative results to evaluate the aspect of the timing (i.e. when to perform) of single-event multilevel surgery. Based on our clinical experience and the findings mentioned above, we hypothesized that older children with bilateral spastic CP would benefit more from single-event multilevel surgery than younger children. Moreover, we hypothesized that any achieved improvement in older children could be maintained with fewer consecutive surgery events than children who undergo surgery at a younger age. METHOD The paediatric orthopaedic unit of the Medical University of Graz, Austria, acts as a specialized assessment and treatment centre for children with CP, providing gait analysis as well as conservative and operative management. To guide postoperative conservative treatment, children are followed up with repeated gait analysis. We reviewed all children with bilateral What this paper adds This paper is the first to report on the long-term outcomes of single-event multilevel surgery in children with bilateral spastic CP. The study found that the older the child is at the time of the surgery the better the long-term result. The pubertal growth spurt is discussed as a contributing factor to gait deterioration. spastic CP who received single-event multilevel surgery between 1995 and Children were included if they (1) were ambulatory with spastic bilateral CP; (2) had a flexed knee gait, defined as 20 or more of minimum knee flexion in the single limb support; (3) had a full set of time distance and kinematic data preoperatively and 1 year and 10 or more years postoperatively; (4) had no previous surgery on the lower limbs before the single-event multilevel surgery; (5) had not received treatment with botulinum toxin in the 6 months before each of the gait assessments; and (6) had not received intrathecal baclofen medication. Based on these inclusion criteria, we performed a query to retrieve anonymized data sets from our gait analysis databank. Clinical data were retrieved from our hospital database system. Out of a total of 936 children with bilateral spastic CP, 32 (all diplegic) fulfilled the above-mentioned criteria. Three limbs did not fulfil the criteria for flexed knee gait preoperatively and, therefore, the data on these children were not analysed. The ages of the children ranged from 5 years 8 months to 15 years 6 months, while the mean age at the surgery was 10 years 6 months (SD 3y 1mo). All of the children were community ambulators and their level of function according to the Gross Motor Function Classification System 13 was either level II (n=12) or level III (n=20). The mean age at the last follow-up was 21 years 4 months (SD 3y 4mo). The number and type of surgical procedures as well as the number and type of consecutive procedures during the 10 or more years of follow-up are provided in TablesIandII. The details on the standardized protocol for the indications of orthopaedic surgery events are shown in Table III. The protocol for the postoperative rehabilitation at our unit has been published previously. 4 In brief, a plaster cast was applied to children who underwent foot surgery. The postoperative mobilization started 7 to 10 days after surgery and children were discharged once they could walk with reasonable confidence. The participants required dynamic ankle foot Table I: Summary of outcomes for primary and secondary outcome measures Mean (SD) time postsurgery,y Nr of participants Mean SD Age GDI PreOP PreOP GDI 1y PostOP 1.12 (0.25) GDI 2 3y PostOP 2.7 (0.57) GDI 5y PostOP 4.85 (0.63) GDI 7 8y PostOP 7.48 (0.31) GDI 10+y PostOP (1.22) GDI, Gait Deviation Index; PreOP, preoperatively; PostOP, postoperatively. Timing of Multilevel Surgery in CP Martin ƒvehlík et al. 731

3 Table II: Summary of outcomes for secondary outcome measures Bony procedures per limb Soft-tissue procedures per limb Total number of surgeries per limb Additional bony procedures per limb Additional soft-tissue procedures per limb Additional surgeries per limb Nr of participants Median Minimum Maximum orthoses for a period of 6 to 10 months, and they discontinued rigid knee foot orthosis night splints 1 year after the surgery. The ethics committee of the local medical university approved the study and informed consent was not required for a retrospective analysis of the biometric data. Computerized gait analysis was performed using a videobased motion capturing system (Vicon, Oxford Metrics, Oxford, UK). Marker placement followed a standard protocol. 14 Vicon Clinical Manager software (Oxford Metrics) was used for data processing. All children walked at self-selected speeds along a 10-metre walkway and, for each child, a minimum of five trials providing a clear foot force plate contact were captured and averaged. Data were collected on six separate occasions: preoperatively and at 1 year, 2 to 3 years, 5 years, 7 to 8 years, and 10 or more years postoperatively. The precise timing of postoperative follow-ups can be found in Table I. Similar to the study by Gough et al., 9 because of interdependency between the joints of the lower limbs during walking, individual joint kinematics were not analysed. Instead, the Gait Deviation Index (GDI) was calculated. 15 This index has been proposed as a summary measure of gait pathology. It compares nine kinematic variables of a child s gait against those of a typically developing comparison group. The kinematics from the pelvis and hip in all three planes, the knee and ankle in the sagittal plane, and foot progression are all used for the calculation. A GDI score of 100 and above denotes non-pathological gait, SD bands are scaled to 10- point intervals below 100. Statistical analysis AstheGDIisnotanindependentvariable,whenbothlimbs of an individual were included in the statistical evaluation, a linear mixed model was used to estimate the effect of age and follow-up time (1, 2.5, 5, 7.5 and 10y after single-event multilevel surgery) on the GDI. In particular, the interaction effect of age and follow-up time was of interest. To adjust for baseline value, preoperative GDI has been included in the model. Spearman s rank correlation coefficient was used to prove the hypothesis that older children need fewer consecutive surgeries than younger ones. The significance level was set at 5% for all statistical tests. Data analysis was performed using SPSS 18 software (SPSS Inc., Chicago, IL, USA). RESULTS The mean preoperative GDI values for the children under review were more than three SDs lower than those of typically developing children, and clearly demonstrated a pathological walking pattern. The preoperative GDI values of older children tended to be lower than those of younger ones, but this effect was not significant ( b b Age =1.00; p=0.076; Table III). However, our mixed linear model revealed a significant interaction between age at the time of surgery and follow-up time. It showed that the older the child was at the time of the surgery, the better the long-term result ( b b Age,Time =0.15; p=0.03; Fig. 1; Table IV). In other words, older children benefit more from single-event multilevel surgery on a long-term basis than younger children. A graph showing the relationship between theageatthetimeofsurgeryandthepredictedvaluesofgdi Table III: Indication criteria for the surgical procedures in management of children with cerebral palsy based on clinical and gait analysis examination Procedure Clinical criteria Gait analysis criteria Intrapelvic psoas lengthening Fixed flexion deformity of >15 Double bump pattern seen on sagittal plane pelvis kinematics Adductor lengthening Passive hip abduction after hamstring lengthening of <30 Medial hamstring lengthening Increased popliteal angle fixed flexion deformity under anaesthesia Decreased knee flexion at initial contact terminal swing Lateral hamstring lengthening Increased popliteal angle fixed flexion deformity under anaesthesia persisting Decreased knee flexion at initial contact terminal swing Distal rectus transfer Positive Duncan-Ely Test Inadequate knee flexion in swing Gastrosoleus lengthening Equinus deformity not correctable under anaesthesia Equinus at initial contact and, in stance, reversal of slope of ankle moments; energy generation in mid-stance Foot tendon lengthening Varus valgus deformity seen during observational transfers gait analysis Tibia derotation osteotomy Bony rotational deformity of more than 10 Persistent internal external rotation throughout the gait cycle Femur derotation osteotomy Bony rotational deformity of more than 10 Persistent internal external rotation throughout the gait cycle 732 Developmental Medicine & Child Neurology 2011, 53:

4 GDI change: after 10 years vs preoperative Age (y) Figure 1: Relationship of the Gait Deviation Index (GDI) change (preoperative vs 10y postoperative) to age of a child at initial surgery. The better long-term outcome of older children is illustrated. Table IV: Impact of age on preoperative Gait Deviation Index (GDI) and impact of age and follow-up time on postoperative GDI estimates of regression coefficients (linear mixed model, dependent variable: GDI) can be found in Figure 2. This is a major finding, considering that we did not find any correlation between age at the time of surgery and the number of bony or soft-tissue procedures performed initially as well as during the 10 years of follow-up (Table V). Out of the additional surgeries, a lengthening of the Achilles tendon had to be performed after an initial Baumann procedure in two children (three limbs). Indications for the other additional orthopaedic procedures were different from those at the initial single-event multilevel surgery and included Austin procedure or basal osteotomy of the first metatarsus for hallux valgus deformity; Green Grice subtalar arthrodesis or triple arthrodesis to treat an instable valgus deformity of the foot; supramalleolar tibia osteotomy, supracondylar femoral derotation osteotomy for excessive internal hip rotation; and triple pelvis osteotomy with psoas and rectus femoris lengthening to treat subluxation of the hip. DISCUSSION The presented data support our hypothesis that children with bilateral CP and flexed knee gait benefit more when a singleevent multilevel surgery is performed at an older age. However, contrary to our second hypothesis, we did not find any correlation between age at the time of surgery and the number of bony or soft-tissue procedures performed initially as well as during the follow-up time. To our knowledge, this is the first Estimate SE p value Impact of age on preoperative GDI Intercept ( b 0 ) <0.001 Age ( b Age ) ) Impact of age and follow-up time on postoperative GDI Intercept ( b b0) <0.001 Age ( b Age ) ) GDI PreOP ( b GDI PreOP) Time ( b Time ) ) Interaction of age and time ( b Age, b time ) PreOP, preoperatively; SE, standard error. Table V: Correlation of age at initial surgery and number of initial as well as additional procedures over the follow-up of 10 years Surgical procedure Spearman s rank correlation coefficient p-value Initial bony procedures Initial soft-tissue procedures Additional bony procedures Additional soft-tissue procedures ) Estimated GDI Time point: (y) 10 7½ 5 2½ Age (y) Figure 2: Predicted values of the Gait Deviation Index (GDI) calculated from age for each time-point after surgery (linear mixed model). Timing of Multilevel Surgery in CP Martin ƒvehlík et al. 733

5 report addressing the timing of single-event multilevel surgery based on a long-term follow-up period of over 10 years. There are a number of limitations to this study. As the design of the study is retrospective, some follow-up assessments between the first postoperative evaluation and the final assessment 10 or more years after surgery were missing for some children. This limits the interpretation of the time course of changes in gait function. Nevertheless, the preoperative, the 1-year, and the final postoperative data sets are complete, in accordance with the inclusion criteria. As the inclusion criteria were quite strict and the follow-up period was long, the size of our study group was small, which limits the power of the study. Therefore, extending the results to other types of CP or gait deviations must be done with caution. Another limitation that has to be addressed is the preoperative biological and functional state of the children in our study group. Although we did not find any difference in GDI before the surgery, it is possible that these children might have been widely dissimilar in the severity of CP, motor control, and pattern development. It is obvious that a child who needs a single-event multilevel surgery at the age of 6 years is not the same as another child undergoing the surgery in his or her teens. This important fact has to be kept in mind while interpreting this study. Comprehensive measures of gait pathology are useful in clinical practice. The computerized gait analysis renders objective data to describe the gait. However, owing to the complexity and volume of data generated by three-dimensional gait analysis, the interpretation of these data is not a simple task. A number of methods were developed to deal with this problem. A neural network, 16 cluster analysis, 17 or multivariate statistic 18 was employed to summarize the gait analysis data. Out of those, the Gillette Gait Index is probably the most widely used clinically. 18 However, even the authors of the Gillette Gait Index have highlighted a number of its limitations: the unbalanced and incomplete nature of the 16 univariate parameters that comprise the index; the nonnormality of the index; and the excessive sensitivity to laboratory-specific control data. 15 Therefore, a new summary measure the GDI was recently introduced. 15 The GDI scales correlate with the Functional Assessment Questionnaire Walking Scale and topographic classification within the diagnosis of CP, and is also strongly correlated with the Gillette Gait Index. 15 Furthermore, the GDI demonstrated its ability to distinguish between the different Gross Motor Function Classification System levels, which further confirms its clinical usability. 19 The strength of the GDI is in illustrating the overall change in pathological gait. However, the index fails to identify the extent of postintervention change at the single level. Nevertheless, it has been recently proved to be an appropriate outcome measure for the evaluation of the effects of surgical treatment in CP. 20 The substantial postoperative improvement in gait deviation documented in this study must be interpreted in the light of the fact that the natural progression of gait in children with CP shows deterioration. 2,21 An important aspect of the present study is that children were followed up for more than 10 years until late adolescence or adulthood. The onset of puberty has been described as a contributing factor to the deterioration of the natural progression of gait. 2 According to our clinical experience, the growth spurt at puberty contributes heavily to the development of muscle contractures and may therefore lead to secondary bony deformities. As the onset of puberty occurs earlier in children with CP, 22 older children could have already been over the period of the growth spurt, which may have produced more predictable outcomes. For the children studied, we do not have the data that could help document the onset of puberty and the pubertal growth spurt in order to support this hypothesis. However, we speculate that the timing of the pubertal growth spurt should be taken into consideration when planning surgical procedures in children with CP. For isolated calf lengthening, it has been shown that surgery at a younger age increases the risk of developing calcaneus gait. 10 In our study, lengthening of the Achilles tendon had to be performed in two children (three limbs) after an initial Baumann procedure. The fear of overcorrection and weakening of plantar flexors, which would be disabling for children with flexed knee gait, might have led us instead to slight under correction of equinus deformity. The children included in this study underwent surgery between 1995 and At that time, physiotherapy and surgical intervention were the main modalities to treat gait problems in CP. As already documented in the literature, the introduction of gait analysis increases the age at which the first orthopaedic surgical procedure is performed, and botulinum toxin type A treatment delays and reduces the frequency of surgical procedures. 12 Therefore, it is very likely that the older children in the study presented might be different from adolescents with CP nowadays. Repeated botulinum toxin type A treatment allow us to postpone the need for surgery even after the growth spurt. Today it is possible to treat all multilevel problems and lever-arm dysfunctions during the course of a single surgical procedure to achieve more predictable results. However, we do not suggest that all the surgical interventions should be postponed until the growth spurt is past. Surgery should always be considered once other conservative treatment methods have been deemed ineffective, and it should definitely not be delayed if walking function is deteriorating. Even if our study shows better long-term outcomes in older children with CP, in practice we often do not have the luxury of being able to sit back and wait for the adolescent growth spurt in order to achieve these better long-term results. In addition to the advantages of the single-event multilevel procedure and more predictable treatment planning, the adolescent with CP will have the opportunity to take part in the decision-making process, which will enhance his or her motivation to participate in the necessary postoperative rehabilitation programme. However, a demotivated teenager could fail to comply with the prolonged postoperative rehabilitation. Long-term postoperative physiotherapy is an integral and crucial part of the surgical treatment. Therefore, a more strongly motivated child or adolescent with CP might help us to not only improve the results of surgery but also slow down any early loss of mobility in young adults with CP Developmental Medicine & Child Neurology 2011, 53:

6 In conclusion, single-event multilevel surgery provides a favourable long-term outcome in children with spastic CP who walk with a flexed knee gait. We did not find any correlation between age at the time of surgery and the number of bony or soft-tissue procedures performed initially or during the follow-up time. The long-term functional improvement was more favourable for children who needed surgery at an older age than in those who underwent single-event multilevel surgery at a younger age. REFERENCES 1. Ziv I, Blackburn N, Rang M,Koreska J.Muscle growth in normal and spastic mice. Dev Med Child Neurol 1984; 26: Bell KJ, Ounpuu S, DeLuca PA, Romness MJ. Natural progression of gait in children with cerebral palsy. J Pediatr Orthop 2002; 22: Sussman MD, Aiona MD. Treatment of spastic diplegia in patients with cerebral palsy. J Pediatr Orthop B 2004; 13: S Saraph V, Zwick EB, Zwick G, Steinwender C, Steinwender G, Linhart W. Multilevel surgery in spastic diplegia: evaluation by physical examination and gait analysis in 25 children. J Pediatr Orthop 2002; 22: Khan MA. Outcome of single-event multilevel surgery in untreated cerebral palsy in a developing country. J Bone Joint Surg Br 2007; 89: Rodda JM, Graham HK, Nattrass GR, Galea MP, Baker R, Wolfe R. Correction of severe crouch gait in patients with spastic diplegia with use of multilevel orthopaedic surgery. J Bone Joint Surg Am 2006; 88: Saraph V, Zwick EB, Auner C, Schneider F, Steinwender G, Linhart W. Gait improvement surgery in diplegic children: how long do the improvements last? J Pediatr Orthop 2005; 25: Gough M, Eve LC, Robinson RO, Shortland AP. Shortterm outcome of multilevel surgical intervention in spastic diplegic cerebral palsy compared with the natural history. Dev Med Child Neurol 2004; 46: Gough M, Schneider P, Shortland AP. The outcome of surgical intervention for early deformity in young ambulant children with bilateral spastic cerebral palsy. J Bone Joint Surg Br 2008; 90: Borton DC, Walker K, Pirpiris M, Nattrass GR, Graham HK. Isolated calf lengthening in cerebral palsy. Outcome analysis of risk factors. J Bone Joint Surg Br 2001; 83: Fabry G, Liu XC, Molenaers G. Gait pattern in patients with spastic diplegic cerebral palsy who underwent staged operations. J Pediatr Orthop B 1999; 8: Molenaers G, Desloovere K, Fabry G, De Cock P. The effects of quantitative gait assessment and botulinum toxin a on musculoskeletal surgery in children with cerebral palsy. J Bone Joint Surg Am 2006; 88: Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997; 39: Ounpuu S, Gage JR, Davis RB. Three-dimensional lower extremity joint kinetics in normal paediatric gait. J Pediatr Orthop 1991; 11: Schwartz MH, Rozumalski A. The Gait Deviation Index: a new comprehensive index of gait pathology. Gait Posture 2008; 28: Zwick EB, Leistritz L, Milleit B, et al. Classification of equinus in ambulatory children with cerebral palsy discrimination between dynamic tightness and fixed contracture. Gait Posture 2004; 20: Toro B, Nester CJ, Farren PC. The development and validity of the Salford Gait Tool: an observation-based clinical gait assessment tool. Arch Phys Med Rehabil 2007; 88: Schutte LM, Narayanan U, Stout JL, Selber P, Gage JR, Schwartz MH. An index for quantifying deviations from normal gait. Gait Posture 2000; 11: Molloy M, McDowell BC, Kerr C, Cosgrove AP. Further evidence of validity of the Gait Deviation Index. Gait Posture 2010; 31: Cimolin V, Galli M, Vimercati SL, Albertini G. Use of the Gait Deviation Index for the assessment of gastrocnemius fascia lengthening in children with cerebral palsy. Res Dev Disabil 2011; 32: Johnson DC, Damiano DL, Abel MF. The evolution of gait in childhood and adolescent cerebral palsy. J Pediatr Orthop 1997; 17: Worley G, Houlihan CM, Herman-Giddens ME, et al. Secondary sexual characteristics in children with cerebral palsy and moderate to severe motor impairment: a cross-sectional survey. Pediatrics 2002; 110: Shortland AP. Muscle deficits in cerebral palsy and early loss of mobility: can we learn something from our elders? Dev Med Child Neurol 2009; 51: Timing of Multilevel Surgery in CP Martin ƒvehlík et al. 735