Gait & Posture 40 (2014) 76–81
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Rectus femoris transfer in cerebral palsy patients with stiff knee gait§ Seung Yeol Lee b,1, Soon-Sun Kwon c,1, Chin Youb Chung a, Kyoung Min Lee a, Young Choi a, Tae Gyun Kim d, Woo Cheol Shin a, In Ho Choi e, Tae-Joon Cho e, Won Joon Yoo e, Moon Seok Park a,* a
Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, Kyungki, Republic of Korea Department of Orthopaedic Surgery, Mongji Hospital, Kyungki, Republic of Korea c Biomedical Research Institute, Seoul National University Bundang Hospital, Kyungki, Republic of Korea d Department of Orthopaedic Surgery, Konyang University Hospital, Daejon, Republic of Korea e Department of Orthopaedic Surgery, Seoul National University Children’s Hospital, Seoul, Republic of Korea b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 October 2013 Received in revised form 19 February 2014 Accepted 23 February 2014
Background: Although several studies have reported on the outcomes of rectus femoris transfer (RFT), few have investigated the multiple factors that could affect the results. Therefore, we evaluated the outcomes of RFT and analyzed factors that inﬂuence improvement and annual change in knee motion after surgery in patients with cerebral palsy (CP). Methods: We reviewed ambulatory patients with CP who were followed up after they had undergone RFT as part of a single-event multilevel surgery (SEMLS) and who had undergone preoperative and postoperative three-dimensional (3D) gait analysis between January 1995 and December 2012. Relevant kinematic values, including peak knee ﬂexion, knee range of motion, and timing of peak knee ﬂexion in the swing phase and gait deviation index (GDI) score, were the outcome measures. Improvements in rate of angle and GDI score were adjusted by multiple factors such as sex, Gross Motor Function Classiﬁcation System (GMFCS) level, anatomic type of CP, and concomitant surgeries as the ﬁxed effects, and follow-up duration, laterality, and each subject as the random effects, all of which was performed using a linear mixed model. Results: A total of 290 patients (487 limbs) and 612 3D gait analysis (2–4 per patient) results were ﬁnally included in this study. At 2 years after RFT, estimated mean peak knee ﬂexion (1.28, p = 0.005), estimated mean knee range of motion (10.78, p < 0.001), and estimated mean GDI score (7.3, p < 0.001) increased signiﬁcantly. Peak knee ﬂexion in the swing phase occurred 5.4% earlier after surgery compared with that at baseline (p < 0.001). In serial postoperative gait analyses, peak knee ﬂexion in the swing phase occurred 0.8% earlier per year in patients with GMFCS level I or II (p = 0.021). Conclusions: RFT as part of a SEMLS was effective in treating stiff knee gait. In serial postoperative gait analyses, patients with GMFCS level I or II showed better prognosis than those with level III with regard to timing of peak knee ﬂexion in the swing phase. Level of evidence: Prognostic level IV. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Rectus femoris transfer Cerebral palsy Single event multilevel surgery Stiff knee gait Outcome
1. Introduction Stiff knee gait, which results in limited ﬂexion/extension knee motion instance and restricted arc of motion during swing , is a
§ The authors thank Eun Young Lee, BS and Mi sun Ryu, BS for data collection. No beneﬁts in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. * Corresponding author at: Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, 300 Gumi-Dong, Bundang-Gu, Sungnam, Kyungki 463-707, Republic of Korea. Tel.: +82 31 787 7203; fax: +82 31 787 4056. E-mail address: [email protected]
(M.S. Park). 1 These authors contributed equally to the writing of this article.
http://dx.doi.org/10.1016/j.gaitpost.2014.02.013 0966-6362/ß 2014 Elsevier B.V. All rights reserved.
common gait disorder in ambulatory patients with cerebral palsy (CP), regardless of anatomic type . Such loss of knee function results in foot clearance problems and reduces gait velocity and step length. The main cause of stiff knee gait in patients with CP is spasticity of the rectus femoris [3,4]. To optimize knee function in these patients, proximal rectus femoris release initially was performed ; however, this procedure can result in stiff-legged gait because of its effect on the hip ﬂexor  Since rectus femoris transfer (RFT) has been suggested for restoring knee motion during the swing phase, it has become the standard surgical treatment for stiff knee gait in patients with CP and is frequently performed as part of a single-event multilevel surgery (SEMLS) [1,3,6–8].
S.Y. Lee et al. / Gait & Posture 40 (2014) 76–81
Several studies analyzing preoperative and postoperative gait have reported that RFT improves knee range of motion and peak knee ﬂexion in the swing phase [8–11]. However, most of these studies did not consider bilateral cases. Limb-based analysis could cause demographic data to be duplicated . The inclusion of both limbs in statistical analysis violates the underlying principle of statistical independence and could thereby bias the study results by exaggerating the levels of signiﬁcance and narrowing the conﬁdence intervals if the limbs in the same individuals are signiﬁcantly related to each other [12–14]. In addition, previous studies have not considered factors that could affect the results of the study, such as age, sex, follow-up interval and duration, and concomitant surgeries as part of an SEMLS, which should be considered in retrospective studies because of their heterogeneity. Gross Motor Function Classiﬁcation System (GMFCS) level, which may be related to the outcomes of surgery , should also be considered. A linear mixed model (LMM) is a parametric linear model for longitudinal data that quantiﬁes the relationships between a continuous dependent variable and various predictor variables, providing a simple and effective way to incorporate within-subject and between-subject variation and the correlation structure of longitudinal data . Analysis using a LMM is appropriate in longitudinal or repeated-measures studies, in which subjects are measured repeatedly over time or under different conditions . In a LMM, ﬁxed effects such as sex represent categorical levels that are measurable and not random. Random effects represent factors that can be speciﬁed to individuals within a population, which account for the variation within subjects . Therefore, the use of a mixed model application is appropriate to estimate the outcomes of RFT in terms of multiple inﬂuencing factors. Although several studies have studied the outcomes of RFT, they did not consider the multiple factors that could affect the results. Therefore, we performed this study to evaluate the outcomes of RFT and to analyze the factors that inﬂuence the improvement and annual change in knee motion after surgery in patients with CP by using an LMM application.
2. Materials and methods This retrospective study was approved by the institutional review board of our hospital, a tertiary referral center for CP. Informed consent was waived due to the retrospective nature of this study. We reviewed the medical records of patients with CP who were followed up after they had undergone RFT and who had undergone preoperative and postoperative three-dimensional (3D) gait analysis between January 1995 and December 2012. At our institution, postoperative gait is routinely analyzed approximately 1 year after SEMLS. In addition, we recommended that patients undergo regular gait analysis. From the third study, gait analysis was performed on the discretion of the patients or their parents. If a repeat surgery for the recurrence of contracture or deformity was performed during the follow-up period, only the data of the gait analyses performed before repeat surgery were included. Patients who had incomplete or missing data on 3D gait analysis and those who were nonambulatory (GMFCS level IV or V) were excluded. Age, sex, anatomic type (unilateral vs. bilateral involvement), GMFCS level, and details of concomitant surgeries were obtained by reviewing medical records. Intramuscular psoas lengthening (IMPL), femoral derotation osteotomy (FDO), distal hamstring lengthening (DHL), tendo-Achilles lengthening (TAL), and Strayer procedure (gastrocnemius recession) were considered as relevant concomitant surgeries, which could affect sagittal plane motion during gait.
2.1. Operative protocol RFT, as part of a SEMLS to improve gait pattern, was performed by 2 pediatric surgeons (CYC and MSP) with 25 and 9 years of experience in orthopedics, respectively. Both surgeons followed the same treatment approach. Preoperative 3D gait analysis was used to plan the procedures. Surgical procedures were performed based on considerations of both clinical and gait analysis ﬁndings. Indications for RFT were as follows: (1) positive Duncan-Ely test, (2) diminution of 158 or more of the total range of knee ﬂexion during the swing phase , and (3) delayed peak knee ﬂexion during the swing phase. In cases of concomitant DHL, which comprised gracilis lengthening, semitendinosus tendon transfer to the adductor magnus, and aponeurotic lengthening of the semimembranosus, the rectus femoris tendon was transferred to the gracilis tendon. When RFT was performed without DHL, the sartorius was used as the transfer site. As a SEMLS, FDO was performed at the intertrochanteric level of the proximal femur, and TAL was performed using coronal Z-plasty. For the Strayer procedure, the gastrocnemius tendon was resected at its most distal part. Then, the resected tendinous end of the gastrocnemius was sutured to the underlying soleus fascia . After the surgery, a removable knee immobilizer was applied to the patients for 3 weeks. The patients began physical therapy on the third to sixth postoperative week depending on the type of concomitant surgeries. 2.2. Acquisition of kinematic data and GDI score Three-dimensional gait analysis was performed a few days before surgery using a Vicon 370 (Oxford Metrix, Oxford, UK) equipped with 7 cameras and 2 force plates. Markers were placed according to the Helen Hayes marker set  by two assistant operators under the supervision of a senior operator, or by the senior operator himself. Kinematic data were archived as patients walked barefoot on a 9-m walkway 3 times with an interval of approximately 30 s. Three trials were averaged to determine the values of the index variables. Then, preoperative and postoperative kinematic variables were compared to assess the outcomes of RFT on knee kinematics. The gait deviation index (GDI) score  was calculated to determine functional improvement in gait after surgery. A GDI score of >100 denotes a nonpathological gait, and each 10-point decrement below 100 indicates 1 SD from normal kinematics. Relevant kinematic values, including peak knee ﬂexion, knee range of motion, and timing of peak knee ﬂexion in the swing phases and GDI score, were considered as the outcome measures. 2.3. Constructing a LMM For each of 4 measurements, changes in knee motion and GDI score were adjusted by multiple factors such as sex, GMFCS level, the anatomic type of CP, IMPL, FDO, TAL, and Strayer procedure as the ﬁxed effects, and follow-up duration, laterality, and each subject as the random effects, all of which were performed using an LMM. The covariance structure was assumed as the variance components (VCs). Restricted maximum likelihood estimation (REML) was used as a method for producing an unbiased estimator. By examining the individual pattern of the rate of change in angle along with the follow-up duration, a LMM with a random slope and random intercept was suggested. The linearity of the follow-up duration effect was interpreted to evaluate the estimation of the 4 measurements. The models were compared using the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC). A smaller AIC or BIC value is preferred in terms of model selection. All models had low AIC and BIC scores; therefore, the
S.Y. Lee et al. / Gait & Posture 40 (2014) 76–81
models were accepted as valid for the estimation of the 4 measurements. 2.4. Statistical analysis To summarize patient demographics, descriptive statistics such as mean and SD were used. The Kolmogorov–Smirnov test was used to verify the normality of the distribution of variables. The LMM was constructed to estimate the rate of angle based on the linearity of the follow-up duration effect, sex, GMFCS level, anatomic type of CP, IMPL, FDO, TAL, and Strayer as covariates. The slope indicated the annual change in the estimated value of 3D gait analysis. The LMM was applied to estimate the change in the value at (1) the total follow-up duration, (2) preoperative and ﬁrst postoperative evaluations, (3) subsequent postoperative evaluations, and (4) after stratiﬁcation of patients according to the GMFCS level (I and II vs. III). Statistical analyses were conducted with R version 2.15.2 (R Foundation for Statistical Computing, Vienna, Austria; ISBN 3900051-07-0; http://www.r-project.org) using the NLME package . All statistics were 2-tailed and p values < 0.05 were considered signiﬁcant.
3. Results Three hundred twenty-one patients met the inclusion criteria. After implementation of the exclusion criteria, a total of 290 patients (487 limbs) were ﬁnally included in this study. Most of the included patients had spastic type CP (269 patients) based on physiologic classiﬁcation, bilateral involvement (255 patients) based on anatomic classiﬁcation, and GMFCS level II (151 patients)
Table 1 Patient demographics. Parameter
Age at surgery (years) Follow-up duration (years) Follow-up interval (years) No. of follow-up
8.6 5.0 (3.4–33.5) 2.0 1.3 (1.0–10.0) 1.8 1.0 (1.0–7.0) 2 (2–4)
IMPL FDO DHL TAL Strayer procedure
39 270 470 338 108
(8.0%) (55.4%) (96.5%) (69.4%) (22.2%)
Age at surgery, follow-up duration, and follow-up interval; mean standard deviation (range), no. of follow-up; mode (range). IMPL, intramuscular psoas lengthening; FDO, femoral derotation osteotomy; DHL, distal hamstring lengthening; TAL, tendo-Achilles lengthening.
based on functional classiﬁcation (Fig. 1). The mean age of the patients at the time of surgery was 8.6 5.0 years (range, 3.4–33.5 years). The total number of surgical procedures performed was 1712, including RFTs (mean, 3.5 per limb): 39 IMPLs, 270 FDOs, 470 DHLs, 338 TALs, and 108 Strayer procedures. A total of 612 preoperative and postoperative 3D gait analyses were performed. The mean follow-up duration was 2.0 1.3 years (range, 1.0–10.0 years), with a mean follow-up interval of 1.8 1.0 years (range, 1.0–7.0 years). The ﬁrst postoperative 3D gait analysis was performed at a mean of 1.7 0.9 years (range, 1.0–6.9 years) after surgery. The number of 3D gait analyses per patient spanned 2–4 (Table 1). Preoperative and ﬁrst postoperative 3D gait analyses of 487 limbs were assessed to estimate
Fig. 1. Inclusion and exclusion criteria.
S.Y. Lee et al. / Gait & Posture 40 (2014) 76–81
Table 2 Changes in the estimated value of three-dimensional gait analysis at 2 years after rectus femoris transfer.a Peak knee ﬂexion in the swing phase (8)
Total GMFCS level I & II GMFCS level III
Knee range of motion in the swing phase (8)
Timing of peak knee ﬂexion in the swing phase (%)
55.8 56.3 59.7
57.0 57.6 60.5
0.005 0.003 0.692
17.9 22.7 17.8
28.6 33.9 24.4