- Email: [email protected]
Superior functional outcome after femoral derotation osteotomy according to gait analysis in cerebral palsy
Gait & Posture 41 (2015) 52–56
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Superior functional outcome after femoral derotation osteotomy according to gait analysis in cerebral palsy M. Niklasch a, T. Dreher a,*, L. Do¨derlein b, S.I. Wolf a, K. Ziegler a, R. Brunner c, E. Rutz c a Pediatric Orthopaedics and Foot Surgery, Clinic for Orthopaedic and Trauma Surgery, Heidelberg University Hospital, Schlierbacher Landstrasse 200a, 69118 Heidelberg, Germany b Orthopaedic Hospital for Children, Behandlungszentrum Aschau GmbH, Bernauerstrasse 18, 83229 Aschau i. Chiemgau, Germany c Paediatric Orthopaedic Department, University Children’s Hospital Basle (UKBB), Spitalstrasse 33, 4056 Basel, Switzerland
A R T I C L E I N F O
A B S T R A C T
Article history: Received 13 February 2014 Received in revised form 4 August 2014 Accepted 18 August 2014
The femoral derotation osteotomy (FDO) is seen as the golden standard treatment in children with cerebral palsy and internal rotation gait. Variable outcomes with cases of over- and undercorrection mainly in the less involved patients have been reported. The determination of the amount of derotation is still inconsistent. 138 patients (age: 11 (3.3) years) with cerebral palsy and internal rotation gait were examined pre- and 1 year postoperatively after distal or proximal FDO, using standardized clinical examination and 3D gait analysis. Three groups were defined retrospectively depending on the amount of derotation in relation to the mean hip rotation in stance (MHR) during gait analysis: Group A (derotation angle > MHR þ 108), Group B (derotation angle = MHR 108), Group C (derotation angle < MHR 108), and compared according to their postoperative mean hip rotation. ANOVA with Bonferroni post hoc test was used for statistics (p < 0.05). Group B had the greatest benefit with the highest rate (86%) of good results (postoperative MHR = 158). In contrast there were 14% cases of overcorrection and 5% cases of deterioration in Group A with only 81% good results and only 79% good results in Group C. It can be concluded, that it is less likely to have unsatisfactory outcomes if the amount of FDO is defined according to the findings of gait analysis compared with clinical examination. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Cerebral palsy Internal rotation gait Femoral derotation osteotomy Indication Outcome
1. Introduction Internal rotation gait is one of the most common gait abnormalities in children with cerebral palsy [1]. The pathogenesis of internal rotation gait is still not fully understood and potentially includes both static and dynamic components. The increased femoral anteversion should be taken into consideration as a static component [2], although not all children with cerebral palsy and internal rotation gait show an increase of femoral anteversion [3,4]. Dynamic components might arise by muscular imbalance, increased muscle tone, spasticity and altered moment arms [5]. Precisely there exist increased moment arms of the internal rotator muscles combined with hip flexion deformity [6] and decreased moment arms of the abductors. The anterior parts of the glutei and the musculus tensor fasciae latae were considered as well as internal rotators [7].
* Corresponding author. Tel.: +49 6221 5635438; fax: +49 6221 5626725. E-mail addresses: [email protected], [email protected] (T. Dreher). http://dx.doi.org/10.1016/j.gaitpost.2014.08.011 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
Internal rotation gait relevantly influences gait since it may lead to lever-arm dysfunction [8,9], clearance problems (‘‘knocking knees’’), decrease of foot lever arm [6] and in consequence tripping. Increased pelvic rotation and excessive tibial torsion may be present concomitantly to compensate the hip malposition or due to lever-arm-dysfunction [8,10,11]. Despite the underlying cause is still not fully understood, addressing the static component by femoral derotation osteotomy (FDO) is seen as the golden standard treatment in patients with cerebral palsy [4,12–14]. The osteotomy can be performed either proximally as intertrochanteric osteotomy or distally as supracondylar osteotomy with comparable static and functional results [12,13,15,16]. Various studies reported the effectiveness of this procedure to correct internal rotation gait [4,13–15]. Nevertheless recent studies describe a variable outcome with cases of over- and undercorrection mainly in less involved patients [4,17] and recurrence [18,19]. One unsolved problem in the treatment of internal rotation gait is how to determine the amount of intraoperative derotation. Previous studies showed that static examination methods such as measuring the femoral anteversion by CT scan [3,20] or
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
measuring clinical passive range of motion do not well correlate with mean hip rotation observed in three-dimensional gait analysis [4,21]. Schwartz et al. [22] outlined that in patients where the amount of internal rotation in gait analysis matches the amount of anteversion angle, the outcome after FDO is superior compared to cases where anteversion angle is not reflected in dynamic parameters. Similar results were reported by Dreher et al. [4], who showed that outcome is good in cases where static (clinical examination) and dynamic (gait analysis) amount of internal rotation are comparably high. However, in most of the past studies, intra-operative derotation amount primarily aimed to correct increased femoral anteversion. Hence a major question is if outcome becomes more predictable when the intra-operative amount of derotation is determined according to data from gait analysis. Wren [23] recently showed clinically and statistically significant improvement of in-toeing after FDO when the gait analysis advice was followed, but poor outcome, when no gait analysis report was available or the recommendations were not implemented. Unfortunately the conclusion is limited as only 7 of 39 recommended FDOs were performed. The purpose of this study was to evaluate if for patients in which derotation amount was close to mean hip rotation in stance during gait analysis FDO leads to a superior outcome compared to those, in which derotation amount differs from dynamic data. 2. Materials and methods Standardized three-dimensional gait analysis and clinical examination were routinely performed for all ambulatory patients with cerebral palsy before and 1 year after surgery (14 months 5.0) at both participating institutions. For the present cohort study, all ambulatory children with bilateral cerebral palsy (GMFCS levels I–III) and internal rotation gait that were treated with femoral derotation osteotomy under supervision of four surgeons (authors of the present paper) in the context of single-event multilevel surgery between 2000 and 2011 were selected retrospectively from the gait laboratory databases. 235 affected limbs from 138 children with a mean age of 11 (3.3) years [range: 3.8–17.8] matched these criteria. In children with bilateral FDO one limb was selected randomly. Femoral derotation osteotomy was performed either distally at the supracondylar level or proximally at the intertrochanteric level. In the past, the primary indication was, a clinically observed, disturbing internal rotation gait accompanied by an increased passive internal rotation in clinical examination and an increased femoral anteversion. The aim of FDO was mainly to correct increased femoral anteversion and to achieve a neutral clinical midpoint [24] between passive internal and external rotation. According to the findings of previous investigations [4] with cases of overcorrection, the indication for FDO changed over the time and the results of three-dimensional gait analysis became more important for the planning of derotation amount. The amount of derotation was measured intraoperatively by using K-wires placed proximally and distally to the osteotomy. The intraoperative amount of derotation averaged 258 78 (range: 58–408). Inclusion criteria were the availability of gait data both pre- and 1 year postoperatively and a documentation of the amount of intraoperative derotation. Patients who underwent additional tibial (de)rotational osteotomy were excluded from this study. The evaluation included physical examination, videotaping and instrumented gait analysis according to a standardized protocol. For three-dimensional gait analysis a six camera Vicon system (Oxford Metrics, Oxford, UK) (Vicon 370 until 2002 in Heidelberg, since then Vicon 612; Vicon 460 in Basel) was used. Equivalency of both systems in Heidelberg was meticulously checked. Skin
53
mounted markers were applied to bony landmarks of the patients according to the protocol of Kadaba et al. [25] and a knee alignment device was used to reduce mistakes in the transversal plane. The patients were asked to walk down a 7 m walkway barefoot at selfdetermined speed. The examinations were all performed by a physiotherapist and a study nurse specially trained in pediatric neuro-developmental therapy and with more than 8 years of experience with gait analysis. Postoperative management consisted of early mobilization with weight-bearing transfers after the first 2–6 weeks and subsequent ambulation, depending on the patient’s body weight and concomitant procedures. Thigh orthoses fixing both legs in about 108 of external rotation were worn during the night for at least 6 months postoperatively to maintain correction. The patients were retrospectively classified into three groups by the amount of derotation in relation to the mean hip rotation in stance during gait analysis. Group A was defined by a derotation amount of more than 108 (twice the estimated measurement error [4]) larger than indicated by mean hip rotation in stance. Group B was defined by a derotation amount within 108 of gait analysis advice. Group C was defined by a derotation amount more than 108 less than mean hip rotation in stance. Improvement of mean hip rotation in stance was calculated by subtracting postoperative from preoperative mean hip rotation in stance. Limbs with a postoperative mean hip rotation in stance phase in a range of 158 were defined as good results. Limbs with a postoperative mean hip rotation in stance phase more than 158 external are considered to be overcorrected. A postoperative mean hip rotation with an absolute value more than 108 worse than preoperative is described as ‘‘worsening’’. Moderate results present a postoperative mean hip rotation more than 158 internal, without fitting the criteria for worsening. 2.1. Statistical methods The outcome variables examined in this study were mean pelvic and mean hip rotation in the stance phase of gait, mean foot progression in the stance phase of gait and the improvement in mean hip rotation in stance. Statistical analysis was performed using IMB SPSS Statistics 19. First the normal distribution of the outcome variables was confirmed by Shapiro Wilk test. All tests were two-tailed, and the significance level was set at p < 0.05. First, outcomes were compared between the pre- and postoperative gait analysis parameters for each of the three defined groups using Student’s t-test. Then an analysis of variance (ANOVA) with Bonferroni post hoc tests was performed to compare the parameters between the defined groups. The amount of derotation osteotomy and the preoperative anteversion measured by clinical examination were not normally distributed. For both variables non-parametric tests were used (Mann–Whitney U-test and Kruskal–Wallis test).
3. Results 57 limbs matched the criteria for Group A (excessive FDO), 67 limbs were allocated Group B (moderate FDO) and 14 limbs made up Group C (conservative FDO). Between these groups there were no significant differences concerning the age of the patients (p = 0.347), the functional level (GMFCS; p = 0.399; Table 1), the side of derotation (p = 0.653) and the proportion of distal to proximal femoral osteotomies (p = 0.485). Concomitant procedures are listed in Table 1. Among the three defined groups, there is no significant difference in preoperative mid-point of passive hip rotation (hip extended [24]) and clinical anteversion (p > 0.379; Table 2). In contrast the three defined groups show statistically significant (p < 0.001) differences in preoperative mean hip rotation in stance (Group A: 88 108, Group B: 228 98; Group C: 418 88).
54
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
Table 1 Demographic data: distribution of the GMFCS level in the different groups and number of surgical procedures performed in single event multilevel surgery. All (138)
A (57)
B (67)
C (14)
GMFCS I GMFCS II GMFCS III
16 (12%) 87 (63%) 35 (25%)
6 (11%) 35 (61%) 16 (28%)
10 (15%) 42 (63%) 15 (22%)
0 (0%) 10 (71%) 4 (29%)
Procedures Pelvic osteotomy FDO (proximal) FDO (distal) Intramuscular psoas lengthening Proximal rectus femoris release Adductor lengthening Hamstring lengthening Distal rectus femoris transfer Patella tendon shortening Calf muscle lengthening Bony foot stabiliatzion Soft tissue procedures, foot Pelvic osteotomy
12 45 93 18 30 7 71 78 28 85 43 29 12
6 20 37 8 14 4 29 32 7 14 18 10 6
4 19 48 8 12 1 35 40 17 12 23 15 4
(9%) (33%) (67%) (13%) (22%) (5%) (51%) (57%) (20%) (62%) (31%) (21%) (9%)
(11%) (35%) (65%) (14%) (25%) (7%) (51%) (56%) (12%) (25%) (32%) (23%) (18%)
(6%) (28%) (72%) (12%) (18%) (2%) (52%) (60%) (25%) (18%) (34%) (22%) (6%)
2 6 8 2 4 2 7 6 4 4 2 4 2
(14%) (43%) (57%) (14%) (29%) (14%) (50%) (43%) (29%) (29%) (14%) (29%) (14%)
Table 2 Preoperative clinical examination and intraoperative measured amount of derotation; positive values indicate internal rotation.
Passive internal rotation (measured in hip extension) Passive external rotation (measured in hip extension) Mid-point [24] (measured in hip extension) Clinical anteversion Amount of derotation osteotomy
Group A
Group B
Group C
678 128
708 138
588 158
308 148
258 158
208 148
19 128
22 128
19 108
308 98 278 68
288 108 248 78
278 98 248 78
All groups present a significant decrease in both mean hip rotation in stance (p < 0.001) and mean foot progression angle (p < 0.001) in stance from preoperative to postoperative. The change in mean pelvic rotation during stance is not significant (p > 0.374) in any of those groups. There is no significant difference between the mid-point in clinical examination [24] and the mean hip rotation in stance during gait analysis in Group B (p = 0.920), whereas these two parameters in Groups A (p < 0.001) and C (p < 0.001) are both significantly different (Tables 2 and 3). Nevertheless the correlation between mean hip rotation in stance and mid-point in clinical examination is weak (r = 0.351, p = 0.006). For results of gait analysis parameters pre- and postoperative please refer to Table 3. Positive values indicate internal rotation. The average improvement in mean hip rotation in stance differed significantly (p < 0.001) among the groups and averaged 108 (148) in Group A, 198 (138) in Group B and 348 (158) in Group C. The improvement in relation to the preoperative mean hip rotation according to the three groups is presented in Fig. 1. Group B had the greatest benefit with the highest ratio (86%) of good results (mean postoperative hip rotation in stance 158). An overcorrection (mean postoperative hip rotation more than 158 external) was found in two cases (3%). There was no case of worsening (mean postoperative hip rotation more than 108 worse than mean preoperative hip rotation). Group C had only 79% good results, but no case of overcorrection or worsening. Group A had the poorest outcome with 81% good results, but 14% overcorrection and 3% worsening (Fig. 2).
4. Discussion Internal rotation gait is a complex problem which arises from the interaction of static and dynamic components. However, FDO addresses only the static component and variable results are reported after surgery. The purpose of this study was to evaluate if patients, in which derotation amount is close to mean hip rotation in stance during gait analysis, have a better predictable outcome than those in which derotation amount differs from dynamic data. A high rate of over- and undercorrection and a high variance of results in mean hip rotation in stance after FDO have been reported with poorer outcome in the less involved limbs [4]. Previous studies suggested that decision-making, guided by physical evaluation, does not result in satisfying outcome; dynamic evaluation is necessary to select those who will benefit from a FDO and to quantify the amount of derotation needed [4,23]. The results of this study indicate that the functional results are superior if the derotation amount is determined according to mean hip rotation in stance during gait analysis compared to determination by clinical examination. This study includes data of two clinical centers and thus has the potential to evaluate a large number of patients over a long period of time. In both gait laboratories a 6 camera Vicon system was used in combination with a marker set according to the protocol of Kadaba et al. [25]. Gait data of each patient is collected preoperative and 1 year postoperative and compared. In a systematic review [26] errors in transverse plane kinematics are reported between 28 and 78 in knee and hip rotation and between 28 and 58 for the measured foot progression angle even within a 3 month period [27]. Gorton et al. [27] describe a standard deviation in hip rotation of 0.78 within sessions, of 0.58 between 12 different motion analysis systems, of 0.78 between different sessions and overall (including a 2 year period) of 7.38. According to these references, gait data from both laboratories can be compared with only marginal limitation. In consideration of the variability between the measurements at intervals of 1 year and published recommendations [4], the three defined groups are separated at 108 difference between intraoperative measured derotation angle and mean hip rotation during preoperative gait analysis. Good results have been defined as postoperative mean hip rotation between 158. Three groups were defined retrospectively depending on the amount of derotation in relation to the mean hip rotation in stance during gait analysis. There have not been any significant differences in the preoperative mid-point of passive hip rotation [24] and preoperative clinical anteversion between these three groups. In contrast, the affected limbs related to the respective groups showed significant differences in the mean hip rotation in stance during preoperative three dimensional gait analysis. In Group B (limbs, that have been derotated according to mean hip rotation during gait analysis (108)), the mid-point of passive hip rotation [24] does not significantly differ from the dynamically measured amount of internal rotation during gait analysis. The amount of derotation did not exceed the measured mean hip rotation for more than 108: this group presented the best functional outcome with good results in 87%. This study suggests – underlined
Table 3 Results of instrumented three-dimensional gait analysis. Preop: preoperative measurement, Postop: postoperative measurement, positive values indicate internal rotation. Parameters
Mean pelvic rotation in stance Mean hip rotation in stance Mean foot progression angle in stance
Group A
Group B
Group C
Preop
Postop
Preop
Postop
Preop
Postop
18 98 88 108 108 198
18 78 58 118 38 118
18 78 228 98 138 168
08 78 48 108 18 118
28 108 418 88 218 158
1 108 118 138 18 168
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
55
Fig. 1. Improvement in relation to preoperative mean hip rotation. Positive values indicate internal rotation; improvement = preoperative mean hip rotation postoperative mean hip rotation; the area between the broken lines marks the good results (postoperative mean hip rotation is in between 158), results below the line demonstrate a remaining internal rotation gait (more than 158 internal); results above the line demonstrate overcorrected limbs (more than 158 external).
Fig. 2. Distribution of good, moderate and bad results within the different groups. Good results: mean hip rotation in stance 158, moderate results: improvement, but mean postoperative hip rotation in stance more than 158 internal; worsening: absolute value of mean postoperative hip rotation more than 108 worse than mean preoperative hip rotation; overcorrection: mean postoperative hip rotation more than 158 external.
by the findings of previous investigations [4,22,23] – that derotation osteotomy should include gait analysis for indication. Furthermore, the data of the present study indicate that the determination of the derotation amount by gait analysis leads to superior outcome compared to clinical examination. This is underlined by the weak correlation (r = 0.351) between the mean hip rotation during gait and the passive hip rotation even in this group. Corroborating previous findings [4,17] the worst outcome was seen in the less involved patients in this study. All patients treated with an excessive FDO were related to Group A. In this group the mean hip rotation in stance during gait analysis was significantly smaller than the mid-point of passive hip rotation [24]. The required amount of derotation was predominantly overestimated and this group reached the poorest outcome with 14% of overcorrection and two cases of worsening of internal rotation gait. The poor outcome in Group A is a major concern, because overcorrection is a severe impairment of lever-arm function with reduction of the foot lever arm, altered moment arms at the knee
and weakness of hip muscles [6,28] in most of the cases and therefore, should be avoided. The large standard deviation in the preoperative mean hip rotation in this group has to be mentioned as a limitation. However, this inhomogeneity underlines once more the difficulty of addressing internal rotation gait in these patients. Further prospective investigations should focus on optimized treatment strategies for these patients. In Group C the static component of internal rotation gait was relatively small compared to the amount of internal rotation gait. Defining a good treatment strategy is challenging for these patients, as mild femoral anteversion restricts the derotation angle and retroversion should be avoided at all costs [4,20]. For these patients it should be critically considered, if additional soft tissue procedures, such as the steel procedure (transfer of the gluteus medius and gluteus minimus) [7,29] are needed to address the dynamic component. The results of the recent study go along with the findings of Schwartz et al. [22]. They reported the best outcome after FDO for
56
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
patients with excessive femoral anteversion and internal rotation gait (three dimensional gait analysis). Conversely patients with excessive femoral anteversion without internal rotation in gait analysis are compromised and show an excessive external foot progression angle postoperatively. This study demonstrates that the determination of the derotation angle relevantly influences the results after FDO. Additionally, multivariate analyses will be needed to explore further influences by concomitant procedures performed during single event multilevel surgery and to analyze factors that influence treatment decisions [30]. It remains a major question, if dynamic problems can really be solved by an isolated static treatment strategy like the FDO or if additional procedures are needed at least for those patients with a high discrepancy between static and dynamic tests. It can be concluded, that according to the results of this study, it is less likely to have unsatisfactory outcomes if the amount of FDO is defined according to the findings of gait analysis compared with clinical examination. The indication for FDO and the amount of derotation angle should be critically considered especially in patients with an increased femoral anteversion and mild internal rotation gait. Static examinations (clinical and radiographic) are however needed to avoid retroversion [4,20].
References [1] Wren TA, Rethlefsen S, Kay RM. Prevalence of specific gait abnormalities in children with cerebral palsy: influence of cerebral palsy subtype, age, and previous surgery. J Pediatr Orthop 2005;25(1):79–83. [2] O’Sullivan R, Walsh M, Hewart P, Jenkinson A, Ross LA, O’Brien T. Factors associated with internal hip rotation gait in patients with cerebral palsy. J Pediatr Orthop 2006;26(4):537–41. [3] Aktas S, Aiona MD, Orendurff M. Evaluation of rotational gait abnormality in the patients cerebral palsy. J Pediatr Orthop 2000;20(2):217–20. [4] Dreher T, Wolf S, Braatz F, Patikas D, Do¨derlein L. Internal rotation gait in spastic diplegia – critical considerations for the femoral derotation osteotomy. Gait Posture 2007;26(1):25–31. [5] Arnold AS, Delp SL. Rotational moment arms of the medial hamstrings and adductors vary with femoral geometry and limb position: implications for the treatment of internally rotated gait. J Biomech 2001;34(4):437–47. [6] Delp SL, Hess WE, Hungerford DS, Jones LC. Variation of rotation moment arms with hip flexion. J Biomech 1999;32(5):493–501. [7] Steel HH. Gluteus medius and minimus insertion advancement for correction of internal rotation gait in spastic cerebral palsy. J Bone Joint Surg Am 1980;62(6):919–27. [8] King HA, Staheli LT. Torsional problems in cerebral palsy. Foot Ankle 1984;4(4):180–4. [9] Gage JR, Novacheck TF. An update on the treatment of gait problems in cerebral palsy. J Pediatr Orthop B 2001;10(4):265–74. [10] Staheli LT. Medial femoral torsion. Orthop Clin North Am 1980;11(1):39–50.
[11] Inan M, Ferri-de Baros F, Chan G, Dabney K, Miller F. Correction of rotational deformity of the tibia in cerebral palsy by percutaneous supramalleolar osteotomy. J Bone Joint Surg Br 2005;87(10):1411–5. [12] Ounpuu S, DeLuca P, Davis R, Romness M. Long-term effects of femoral derotation osteotomies: an evaluation using three-dimensional gait analysis. J Pediatr Orthop 2002;22(2):139–45. [13] Pirpiris M, Trivett A, Baker R, Rodda J, Nattrass GR, Graham HK. Femoral derotation osteotomy in spastic diplegia. Proximal or distal? J Bone Joint Surg Br 2003;85(2):265–72. [14] Dreher T, Wolf SI, Heitzmann D, Swartman B, Schuster W, Gantz S, et al. Longterm outcome of femoral derotation osteotomy in children with spastic diplegia. Gait Posture 2012;36(3):467–70. [15] Saraph V, Zwick EB, Zwick G, Dreier M, Steinwender G, Linhart W. Effect of derotation osteotomy of the femur on hip and pelvis rotations in hemiplegic and diplegic children. J Pediatr Orthop B 2002;11(2):159–66. [16] Kay RM, Rethlefsen SA, Hale JM, Skaggs DL, Tolo VT. Comparison of proximal and distal rotational femoral osteotomy in children with cerebral palsy. J Pediatr Orthop 2003;23(2):150–4. [17] Rutz E, Donath S, Tirosh O, Graham HK, Baker R. Explaining the variability improvements in gait quality as a result of single event multi-level surgery in cerebral palsy. Gait Posture 2013;38(3):455–60. [18] Dreher T, Vegvari D, Wolf SI, Geisbusch A, Gantz S, Wenz W, et al. Development of knee function after hamstring lengthening as a part of multilevel surgery in children with spastic diplegia: a long-term outcome study. J Bone Joint Surg Am 2012;94(2):121–30. [19] de Morais Filho MC, Kawamura CM, dos Santos CA, Mattar Jr R. Outcomes of correction of internal hip rotation in patients with spastic cerebral palsy using proximal femoral osteotomy. Gait Posture 2012;36(2):201–4. [20] Braatz F, Wolf SI, Gerber A, Klotz MC, Dreher T. Do changes in torsional magnetic resonance imaging reflect improvement in gait after femoral derotation osteotomy in patients with cerebral palsy? Int Orthop 2013. [21] Wren TA, Elihu KJ, Mansour S, Rethlefsen SA, Ryan DD, Smith ML, et al. Differences in implementation of gait analysis recommendations based on affiliation with a gait laboratory. Gait Posture 2013;37(2):206–9. [22] Schwartz MH, Rozumalski A, Novacheck TF. Femoral derotational osteotomy: surgical indications and outcomes in children with cerebral palsy. Gait Posture 2014;39(2):778–83. [23] Wren TA, Lening C, Rethlefsen SA, Kay RM. Impact of gait analysis on correction of excessive hip internal rotation in ambulatory children with cerebral palsy: a randomized controlled trial. Dev Med Child Neurol 2013;55(10):919–25. [24] Kerr AM, Kirtley SJ, Hillman SJ, van der Linden ML, Hazlewood ME, Robb JE. The mid-point of passive hip rotation range is an indicator of hip rotation in gait in cerebral palsy. Gait Posture 2003;17(1):88–91. [25] Kadaba MP, Ramakrishnan HK, Wootten ME. Measurement of lower extremity kinematics during level walking. J Orthop Res 1990;8(3):383–92. [26] McGinley JL, Baker R, Wolfe R, Morris ME. The reliability of three-dimensional kinematic gait measurements: a systematic review. Gait Posture 2009;29(3): 360–9. [27] Gorton 3rd GE, Hebert DA, Gannotti ME. Assessment of the kinematic variability among 12 motion analysis laboratories. Gait Posture 2009;29(3):398– 402. [28] Arnold AS, Asakawa DJ, Delp SL. Do the hamstrings and adductors contribute to excessive internal rotation of the hip in persons with cerebral palsy? Gait Posture 2000;11(3):181–90. [29] Do¨derlein L. Infantile Zerebralparese Diagnostik, konservative und operative Therapie. 1st ed. Darmstadt: Steinkopff Verlag; 2007. [30] Schwartz MH, Rozumalski A, Novacheck TF. Femoral derotational osteotomy: surgical indications and outcomes in children with cerebral palsy. Gait Posture; 2013.
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Superior functional outcome after femoral derotation osteotomy according to gait analysis in cerebral palsy M. Niklasch a, T. Dreher a,*, L. Do¨derlein b, S.I. Wolf a, K. Ziegler a, R. Brunner c, E. Rutz c a Pediatric Orthopaedics and Foot Surgery, Clinic for Orthopaedic and Trauma Surgery, Heidelberg University Hospital, Schlierbacher Landstrasse 200a, 69118 Heidelberg, Germany b Orthopaedic Hospital for Children, Behandlungszentrum Aschau GmbH, Bernauerstrasse 18, 83229 Aschau i. Chiemgau, Germany c Paediatric Orthopaedic Department, University Children’s Hospital Basle (UKBB), Spitalstrasse 33, 4056 Basel, Switzerland
A R T I C L E I N F O
A B S T R A C T
Article history: Received 13 February 2014 Received in revised form 4 August 2014 Accepted 18 August 2014
The femoral derotation osteotomy (FDO) is seen as the golden standard treatment in children with cerebral palsy and internal rotation gait. Variable outcomes with cases of over- and undercorrection mainly in the less involved patients have been reported. The determination of the amount of derotation is still inconsistent. 138 patients (age: 11 (3.3) years) with cerebral palsy and internal rotation gait were examined pre- and 1 year postoperatively after distal or proximal FDO, using standardized clinical examination and 3D gait analysis. Three groups were defined retrospectively depending on the amount of derotation in relation to the mean hip rotation in stance (MHR) during gait analysis: Group A (derotation angle > MHR þ 108), Group B (derotation angle = MHR 108), Group C (derotation angle < MHR 108), and compared according to their postoperative mean hip rotation. ANOVA with Bonferroni post hoc test was used for statistics (p < 0.05). Group B had the greatest benefit with the highest rate (86%) of good results (postoperative MHR = 158). In contrast there were 14% cases of overcorrection and 5% cases of deterioration in Group A with only 81% good results and only 79% good results in Group C. It can be concluded, that it is less likely to have unsatisfactory outcomes if the amount of FDO is defined according to the findings of gait analysis compared with clinical examination. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Cerebral palsy Internal rotation gait Femoral derotation osteotomy Indication Outcome
1. Introduction Internal rotation gait is one of the most common gait abnormalities in children with cerebral palsy [1]. The pathogenesis of internal rotation gait is still not fully understood and potentially includes both static and dynamic components. The increased femoral anteversion should be taken into consideration as a static component [2], although not all children with cerebral palsy and internal rotation gait show an increase of femoral anteversion [3,4]. Dynamic components might arise by muscular imbalance, increased muscle tone, spasticity and altered moment arms [5]. Precisely there exist increased moment arms of the internal rotator muscles combined with hip flexion deformity [6] and decreased moment arms of the abductors. The anterior parts of the glutei and the musculus tensor fasciae latae were considered as well as internal rotators [7].
* Corresponding author. Tel.: +49 6221 5635438; fax: +49 6221 5626725. E-mail addresses: [email protected], [email protected] (T. Dreher). http://dx.doi.org/10.1016/j.gaitpost.2014.08.011 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
Internal rotation gait relevantly influences gait since it may lead to lever-arm dysfunction [8,9], clearance problems (‘‘knocking knees’’), decrease of foot lever arm [6] and in consequence tripping. Increased pelvic rotation and excessive tibial torsion may be present concomitantly to compensate the hip malposition or due to lever-arm-dysfunction [8,10,11]. Despite the underlying cause is still not fully understood, addressing the static component by femoral derotation osteotomy (FDO) is seen as the golden standard treatment in patients with cerebral palsy [4,12–14]. The osteotomy can be performed either proximally as intertrochanteric osteotomy or distally as supracondylar osteotomy with comparable static and functional results [12,13,15,16]. Various studies reported the effectiveness of this procedure to correct internal rotation gait [4,13–15]. Nevertheless recent studies describe a variable outcome with cases of over- and undercorrection mainly in less involved patients [4,17] and recurrence [18,19]. One unsolved problem in the treatment of internal rotation gait is how to determine the amount of intraoperative derotation. Previous studies showed that static examination methods such as measuring the femoral anteversion by CT scan [3,20] or
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
measuring clinical passive range of motion do not well correlate with mean hip rotation observed in three-dimensional gait analysis [4,21]. Schwartz et al. [22] outlined that in patients where the amount of internal rotation in gait analysis matches the amount of anteversion angle, the outcome after FDO is superior compared to cases where anteversion angle is not reflected in dynamic parameters. Similar results were reported by Dreher et al. [4], who showed that outcome is good in cases where static (clinical examination) and dynamic (gait analysis) amount of internal rotation are comparably high. However, in most of the past studies, intra-operative derotation amount primarily aimed to correct increased femoral anteversion. Hence a major question is if outcome becomes more predictable when the intra-operative amount of derotation is determined according to data from gait analysis. Wren [23] recently showed clinically and statistically significant improvement of in-toeing after FDO when the gait analysis advice was followed, but poor outcome, when no gait analysis report was available or the recommendations were not implemented. Unfortunately the conclusion is limited as only 7 of 39 recommended FDOs were performed. The purpose of this study was to evaluate if for patients in which derotation amount was close to mean hip rotation in stance during gait analysis FDO leads to a superior outcome compared to those, in which derotation amount differs from dynamic data. 2. Materials and methods Standardized three-dimensional gait analysis and clinical examination were routinely performed for all ambulatory patients with cerebral palsy before and 1 year after surgery (14 months 5.0) at both participating institutions. For the present cohort study, all ambulatory children with bilateral cerebral palsy (GMFCS levels I–III) and internal rotation gait that were treated with femoral derotation osteotomy under supervision of four surgeons (authors of the present paper) in the context of single-event multilevel surgery between 2000 and 2011 were selected retrospectively from the gait laboratory databases. 235 affected limbs from 138 children with a mean age of 11 (3.3) years [range: 3.8–17.8] matched these criteria. In children with bilateral FDO one limb was selected randomly. Femoral derotation osteotomy was performed either distally at the supracondylar level or proximally at the intertrochanteric level. In the past, the primary indication was, a clinically observed, disturbing internal rotation gait accompanied by an increased passive internal rotation in clinical examination and an increased femoral anteversion. The aim of FDO was mainly to correct increased femoral anteversion and to achieve a neutral clinical midpoint [24] between passive internal and external rotation. According to the findings of previous investigations [4] with cases of overcorrection, the indication for FDO changed over the time and the results of three-dimensional gait analysis became more important for the planning of derotation amount. The amount of derotation was measured intraoperatively by using K-wires placed proximally and distally to the osteotomy. The intraoperative amount of derotation averaged 258 78 (range: 58–408). Inclusion criteria were the availability of gait data both pre- and 1 year postoperatively and a documentation of the amount of intraoperative derotation. Patients who underwent additional tibial (de)rotational osteotomy were excluded from this study. The evaluation included physical examination, videotaping and instrumented gait analysis according to a standardized protocol. For three-dimensional gait analysis a six camera Vicon system (Oxford Metrics, Oxford, UK) (Vicon 370 until 2002 in Heidelberg, since then Vicon 612; Vicon 460 in Basel) was used. Equivalency of both systems in Heidelberg was meticulously checked. Skin
53
mounted markers were applied to bony landmarks of the patients according to the protocol of Kadaba et al. [25] and a knee alignment device was used to reduce mistakes in the transversal plane. The patients were asked to walk down a 7 m walkway barefoot at selfdetermined speed. The examinations were all performed by a physiotherapist and a study nurse specially trained in pediatric neuro-developmental therapy and with more than 8 years of experience with gait analysis. Postoperative management consisted of early mobilization with weight-bearing transfers after the first 2–6 weeks and subsequent ambulation, depending on the patient’s body weight and concomitant procedures. Thigh orthoses fixing both legs in about 108 of external rotation were worn during the night for at least 6 months postoperatively to maintain correction. The patients were retrospectively classified into three groups by the amount of derotation in relation to the mean hip rotation in stance during gait analysis. Group A was defined by a derotation amount of more than 108 (twice the estimated measurement error [4]) larger than indicated by mean hip rotation in stance. Group B was defined by a derotation amount within 108 of gait analysis advice. Group C was defined by a derotation amount more than 108 less than mean hip rotation in stance. Improvement of mean hip rotation in stance was calculated by subtracting postoperative from preoperative mean hip rotation in stance. Limbs with a postoperative mean hip rotation in stance phase in a range of 158 were defined as good results. Limbs with a postoperative mean hip rotation in stance phase more than 158 external are considered to be overcorrected. A postoperative mean hip rotation with an absolute value more than 108 worse than preoperative is described as ‘‘worsening’’. Moderate results present a postoperative mean hip rotation more than 158 internal, without fitting the criteria for worsening. 2.1. Statistical methods The outcome variables examined in this study were mean pelvic and mean hip rotation in the stance phase of gait, mean foot progression in the stance phase of gait and the improvement in mean hip rotation in stance. Statistical analysis was performed using IMB SPSS Statistics 19. First the normal distribution of the outcome variables was confirmed by Shapiro Wilk test. All tests were two-tailed, and the significance level was set at p < 0.05. First, outcomes were compared between the pre- and postoperative gait analysis parameters for each of the three defined groups using Student’s t-test. Then an analysis of variance (ANOVA) with Bonferroni post hoc tests was performed to compare the parameters between the defined groups. The amount of derotation osteotomy and the preoperative anteversion measured by clinical examination were not normally distributed. For both variables non-parametric tests were used (Mann–Whitney U-test and Kruskal–Wallis test).
3. Results 57 limbs matched the criteria for Group A (excessive FDO), 67 limbs were allocated Group B (moderate FDO) and 14 limbs made up Group C (conservative FDO). Between these groups there were no significant differences concerning the age of the patients (p = 0.347), the functional level (GMFCS; p = 0.399; Table 1), the side of derotation (p = 0.653) and the proportion of distal to proximal femoral osteotomies (p = 0.485). Concomitant procedures are listed in Table 1. Among the three defined groups, there is no significant difference in preoperative mid-point of passive hip rotation (hip extended [24]) and clinical anteversion (p > 0.379; Table 2). In contrast the three defined groups show statistically significant (p < 0.001) differences in preoperative mean hip rotation in stance (Group A: 88 108, Group B: 228 98; Group C: 418 88).
54
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
Table 1 Demographic data: distribution of the GMFCS level in the different groups and number of surgical procedures performed in single event multilevel surgery. All (138)
A (57)
B (67)
C (14)
GMFCS I GMFCS II GMFCS III
16 (12%) 87 (63%) 35 (25%)
6 (11%) 35 (61%) 16 (28%)
10 (15%) 42 (63%) 15 (22%)
0 (0%) 10 (71%) 4 (29%)
Procedures Pelvic osteotomy FDO (proximal) FDO (distal) Intramuscular psoas lengthening Proximal rectus femoris release Adductor lengthening Hamstring lengthening Distal rectus femoris transfer Patella tendon shortening Calf muscle lengthening Bony foot stabiliatzion Soft tissue procedures, foot Pelvic osteotomy
12 45 93 18 30 7 71 78 28 85 43 29 12
6 20 37 8 14 4 29 32 7 14 18 10 6
4 19 48 8 12 1 35 40 17 12 23 15 4
(9%) (33%) (67%) (13%) (22%) (5%) (51%) (57%) (20%) (62%) (31%) (21%) (9%)
(11%) (35%) (65%) (14%) (25%) (7%) (51%) (56%) (12%) (25%) (32%) (23%) (18%)
(6%) (28%) (72%) (12%) (18%) (2%) (52%) (60%) (25%) (18%) (34%) (22%) (6%)
2 6 8 2 4 2 7 6 4 4 2 4 2
(14%) (43%) (57%) (14%) (29%) (14%) (50%) (43%) (29%) (29%) (14%) (29%) (14%)
Table 2 Preoperative clinical examination and intraoperative measured amount of derotation; positive values indicate internal rotation.
Passive internal rotation (measured in hip extension) Passive external rotation (measured in hip extension) Mid-point [24] (measured in hip extension) Clinical anteversion Amount of derotation osteotomy
Group A
Group B
Group C
678 128
708 138
588 158
308 148
258 158
208 148
19 128
22 128
19 108
308 98 278 68
288 108 248 78
278 98 248 78
All groups present a significant decrease in both mean hip rotation in stance (p < 0.001) and mean foot progression angle (p < 0.001) in stance from preoperative to postoperative. The change in mean pelvic rotation during stance is not significant (p > 0.374) in any of those groups. There is no significant difference between the mid-point in clinical examination [24] and the mean hip rotation in stance during gait analysis in Group B (p = 0.920), whereas these two parameters in Groups A (p < 0.001) and C (p < 0.001) are both significantly different (Tables 2 and 3). Nevertheless the correlation between mean hip rotation in stance and mid-point in clinical examination is weak (r = 0.351, p = 0.006). For results of gait analysis parameters pre- and postoperative please refer to Table 3. Positive values indicate internal rotation. The average improvement in mean hip rotation in stance differed significantly (p < 0.001) among the groups and averaged 108 (148) in Group A, 198 (138) in Group B and 348 (158) in Group C. The improvement in relation to the preoperative mean hip rotation according to the three groups is presented in Fig. 1. Group B had the greatest benefit with the highest ratio (86%) of good results (mean postoperative hip rotation in stance 158). An overcorrection (mean postoperative hip rotation more than 158 external) was found in two cases (3%). There was no case of worsening (mean postoperative hip rotation more than 108 worse than mean preoperative hip rotation). Group C had only 79% good results, but no case of overcorrection or worsening. Group A had the poorest outcome with 81% good results, but 14% overcorrection and 3% worsening (Fig. 2).
4. Discussion Internal rotation gait is a complex problem which arises from the interaction of static and dynamic components. However, FDO addresses only the static component and variable results are reported after surgery. The purpose of this study was to evaluate if patients, in which derotation amount is close to mean hip rotation in stance during gait analysis, have a better predictable outcome than those in which derotation amount differs from dynamic data. A high rate of over- and undercorrection and a high variance of results in mean hip rotation in stance after FDO have been reported with poorer outcome in the less involved limbs [4]. Previous studies suggested that decision-making, guided by physical evaluation, does not result in satisfying outcome; dynamic evaluation is necessary to select those who will benefit from a FDO and to quantify the amount of derotation needed [4,23]. The results of this study indicate that the functional results are superior if the derotation amount is determined according to mean hip rotation in stance during gait analysis compared to determination by clinical examination. This study includes data of two clinical centers and thus has the potential to evaluate a large number of patients over a long period of time. In both gait laboratories a 6 camera Vicon system was used in combination with a marker set according to the protocol of Kadaba et al. [25]. Gait data of each patient is collected preoperative and 1 year postoperative and compared. In a systematic review [26] errors in transverse plane kinematics are reported between 28 and 78 in knee and hip rotation and between 28 and 58 for the measured foot progression angle even within a 3 month period [27]. Gorton et al. [27] describe a standard deviation in hip rotation of 0.78 within sessions, of 0.58 between 12 different motion analysis systems, of 0.78 between different sessions and overall (including a 2 year period) of 7.38. According to these references, gait data from both laboratories can be compared with only marginal limitation. In consideration of the variability between the measurements at intervals of 1 year and published recommendations [4], the three defined groups are separated at 108 difference between intraoperative measured derotation angle and mean hip rotation during preoperative gait analysis. Good results have been defined as postoperative mean hip rotation between 158. Three groups were defined retrospectively depending on the amount of derotation in relation to the mean hip rotation in stance during gait analysis. There have not been any significant differences in the preoperative mid-point of passive hip rotation [24] and preoperative clinical anteversion between these three groups. In contrast, the affected limbs related to the respective groups showed significant differences in the mean hip rotation in stance during preoperative three dimensional gait analysis. In Group B (limbs, that have been derotated according to mean hip rotation during gait analysis (108)), the mid-point of passive hip rotation [24] does not significantly differ from the dynamically measured amount of internal rotation during gait analysis. The amount of derotation did not exceed the measured mean hip rotation for more than 108: this group presented the best functional outcome with good results in 87%. This study suggests – underlined
Table 3 Results of instrumented three-dimensional gait analysis. Preop: preoperative measurement, Postop: postoperative measurement, positive values indicate internal rotation. Parameters
Mean pelvic rotation in stance Mean hip rotation in stance Mean foot progression angle in stance
Group A
Group B
Group C
Preop
Postop
Preop
Postop
Preop
Postop
18 98 88 108 108 198
18 78 58 118 38 118
18 78 228 98 138 168
08 78 48 108 18 118
28 108 418 88 218 158
1 108 118 138 18 168
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
55
Fig. 1. Improvement in relation to preoperative mean hip rotation. Positive values indicate internal rotation; improvement = preoperative mean hip rotation postoperative mean hip rotation; the area between the broken lines marks the good results (postoperative mean hip rotation is in between 158), results below the line demonstrate a remaining internal rotation gait (more than 158 internal); results above the line demonstrate overcorrected limbs (more than 158 external).
Fig. 2. Distribution of good, moderate and bad results within the different groups. Good results: mean hip rotation in stance 158, moderate results: improvement, but mean postoperative hip rotation in stance more than 158 internal; worsening: absolute value of mean postoperative hip rotation more than 108 worse than mean preoperative hip rotation; overcorrection: mean postoperative hip rotation more than 158 external.
by the findings of previous investigations [4,22,23] – that derotation osteotomy should include gait analysis for indication. Furthermore, the data of the present study indicate that the determination of the derotation amount by gait analysis leads to superior outcome compared to clinical examination. This is underlined by the weak correlation (r = 0.351) between the mean hip rotation during gait and the passive hip rotation even in this group. Corroborating previous findings [4,17] the worst outcome was seen in the less involved patients in this study. All patients treated with an excessive FDO were related to Group A. In this group the mean hip rotation in stance during gait analysis was significantly smaller than the mid-point of passive hip rotation [24]. The required amount of derotation was predominantly overestimated and this group reached the poorest outcome with 14% of overcorrection and two cases of worsening of internal rotation gait. The poor outcome in Group A is a major concern, because overcorrection is a severe impairment of lever-arm function with reduction of the foot lever arm, altered moment arms at the knee
and weakness of hip muscles [6,28] in most of the cases and therefore, should be avoided. The large standard deviation in the preoperative mean hip rotation in this group has to be mentioned as a limitation. However, this inhomogeneity underlines once more the difficulty of addressing internal rotation gait in these patients. Further prospective investigations should focus on optimized treatment strategies for these patients. In Group C the static component of internal rotation gait was relatively small compared to the amount of internal rotation gait. Defining a good treatment strategy is challenging for these patients, as mild femoral anteversion restricts the derotation angle and retroversion should be avoided at all costs [4,20]. For these patients it should be critically considered, if additional soft tissue procedures, such as the steel procedure (transfer of the gluteus medius and gluteus minimus) [7,29] are needed to address the dynamic component. The results of the recent study go along with the findings of Schwartz et al. [22]. They reported the best outcome after FDO for
56
M. Niklasch et al. / Gait & Posture 41 (2015) 52–56
patients with excessive femoral anteversion and internal rotation gait (three dimensional gait analysis). Conversely patients with excessive femoral anteversion without internal rotation in gait analysis are compromised and show an excessive external foot progression angle postoperatively. This study demonstrates that the determination of the derotation angle relevantly influences the results after FDO. Additionally, multivariate analyses will be needed to explore further influences by concomitant procedures performed during single event multilevel surgery and to analyze factors that influence treatment decisions [30]. It remains a major question, if dynamic problems can really be solved by an isolated static treatment strategy like the FDO or if additional procedures are needed at least for those patients with a high discrepancy between static and dynamic tests. It can be concluded, that according to the results of this study, it is less likely to have unsatisfactory outcomes if the amount of FDO is defined according to the findings of gait analysis compared with clinical examination. The indication for FDO and the amount of derotation angle should be critically considered especially in patients with an increased femoral anteversion and mild internal rotation gait. Static examinations (clinical and radiographic) are however needed to avoid retroversion [4,20].
References [1] Wren TA, Rethlefsen S, Kay RM. Prevalence of specific gait abnormalities in children with cerebral palsy: influence of cerebral palsy subtype, age, and previous surgery. J Pediatr Orthop 2005;25(1):79–83. [2] O’Sullivan R, Walsh M, Hewart P, Jenkinson A, Ross LA, O’Brien T. Factors associated with internal hip rotation gait in patients with cerebral palsy. J Pediatr Orthop 2006;26(4):537–41. [3] Aktas S, Aiona MD, Orendurff M. Evaluation of rotational gait abnormality in the patients cerebral palsy. J Pediatr Orthop 2000;20(2):217–20. [4] Dreher T, Wolf S, Braatz F, Patikas D, Do¨derlein L. Internal rotation gait in spastic diplegia – critical considerations for the femoral derotation osteotomy. Gait Posture 2007;26(1):25–31. [5] Arnold AS, Delp SL. Rotational moment arms of the medial hamstrings and adductors vary with femoral geometry and limb position: implications for the treatment of internally rotated gait. J Biomech 2001;34(4):437–47. [6] Delp SL, Hess WE, Hungerford DS, Jones LC. Variation of rotation moment arms with hip flexion. J Biomech 1999;32(5):493–501. [7] Steel HH. Gluteus medius and minimus insertion advancement for correction of internal rotation gait in spastic cerebral palsy. J Bone Joint Surg Am 1980;62(6):919–27. [8] King HA, Staheli LT. Torsional problems in cerebral palsy. Foot Ankle 1984;4(4):180–4. [9] Gage JR, Novacheck TF. An update on the treatment of gait problems in cerebral palsy. J Pediatr Orthop B 2001;10(4):265–74. [10] Staheli LT. Medial femoral torsion. Orthop Clin North Am 1980;11(1):39–50.
[11] Inan M, Ferri-de Baros F, Chan G, Dabney K, Miller F. Correction of rotational deformity of the tibia in cerebral palsy by percutaneous supramalleolar osteotomy. J Bone Joint Surg Br 2005;87(10):1411–5. [12] Ounpuu S, DeLuca P, Davis R, Romness M. Long-term effects of femoral derotation osteotomies: an evaluation using three-dimensional gait analysis. J Pediatr Orthop 2002;22(2):139–45. [13] Pirpiris M, Trivett A, Baker R, Rodda J, Nattrass GR, Graham HK. Femoral derotation osteotomy in spastic diplegia. Proximal or distal? J Bone Joint Surg Br 2003;85(2):265–72. [14] Dreher T, Wolf SI, Heitzmann D, Swartman B, Schuster W, Gantz S, et al. Longterm outcome of femoral derotation osteotomy in children with spastic diplegia. Gait Posture 2012;36(3):467–70. [15] Saraph V, Zwick EB, Zwick G, Dreier M, Steinwender G, Linhart W. Effect of derotation osteotomy of the femur on hip and pelvis rotations in hemiplegic and diplegic children. J Pediatr Orthop B 2002;11(2):159–66. [16] Kay RM, Rethlefsen SA, Hale JM, Skaggs DL, Tolo VT. Comparison of proximal and distal rotational femoral osteotomy in children with cerebral palsy. J Pediatr Orthop 2003;23(2):150–4. [17] Rutz E, Donath S, Tirosh O, Graham HK, Baker R. Explaining the variability improvements in gait quality as a result of single event multi-level surgery in cerebral palsy. Gait Posture 2013;38(3):455–60. [18] Dreher T, Vegvari D, Wolf SI, Geisbusch A, Gantz S, Wenz W, et al. Development of knee function after hamstring lengthening as a part of multilevel surgery in children with spastic diplegia: a long-term outcome study. J Bone Joint Surg Am 2012;94(2):121–30. [19] de Morais Filho MC, Kawamura CM, dos Santos CA, Mattar Jr R. Outcomes of correction of internal hip rotation in patients with spastic cerebral palsy using proximal femoral osteotomy. Gait Posture 2012;36(2):201–4. [20] Braatz F, Wolf SI, Gerber A, Klotz MC, Dreher T. Do changes in torsional magnetic resonance imaging reflect improvement in gait after femoral derotation osteotomy in patients with cerebral palsy? Int Orthop 2013. [21] Wren TA, Elihu KJ, Mansour S, Rethlefsen SA, Ryan DD, Smith ML, et al. Differences in implementation of gait analysis recommendations based on affiliation with a gait laboratory. Gait Posture 2013;37(2):206–9. [22] Schwartz MH, Rozumalski A, Novacheck TF. Femoral derotational osteotomy: surgical indications and outcomes in children with cerebral palsy. Gait Posture 2014;39(2):778–83. [23] Wren TA, Lening C, Rethlefsen SA, Kay RM. Impact of gait analysis on correction of excessive hip internal rotation in ambulatory children with cerebral palsy: a randomized controlled trial. Dev Med Child Neurol 2013;55(10):919–25. [24] Kerr AM, Kirtley SJ, Hillman SJ, van der Linden ML, Hazlewood ME, Robb JE. The mid-point of passive hip rotation range is an indicator of hip rotation in gait in cerebral palsy. Gait Posture 2003;17(1):88–91. [25] Kadaba MP, Ramakrishnan HK, Wootten ME. Measurement of lower extremity kinematics during level walking. J Orthop Res 1990;8(3):383–92. [26] McGinley JL, Baker R, Wolfe R, Morris ME. The reliability of three-dimensional kinematic gait measurements: a systematic review. Gait Posture 2009;29(3): 360–9. [27] Gorton 3rd GE, Hebert DA, Gannotti ME. Assessment of the kinematic variability among 12 motion analysis laboratories. Gait Posture 2009;29(3):398– 402. [28] Arnold AS, Asakawa DJ, Delp SL. Do the hamstrings and adductors contribute to excessive internal rotation of the hip in persons with cerebral palsy? Gait Posture 2000;11(3):181–90. [29] Do¨derlein L. Infantile Zerebralparese Diagnostik, konservative und operative Therapie. 1st ed. Darmstadt: Steinkopff Verlag; 2007. [30] Schwartz MH, Rozumalski A, Novacheck TF. Femoral derotational osteotomy: surgical indications and outcomes in children with cerebral palsy. Gait Posture; 2013.