Effects of soft tissue surgery on transverse kinematics in patients with cerebral palsy

Gait disturbances, including flexed knee gait, stiff knee gait, and tip-toeing gait, are common in patients with cerebral palsy (CP) [1‐3]. Flexed knee gait is characterized by an abnormally high knee flexion, and hamstring spasticity with contracture has been known as one of the main factors [4‐7]. Stiff knee gait presents as delayed and/or reduced peak knee flexion during the swing phase [8, 9] and is considered to be caused by spasticity of the rectus femoris muscle. Tip-toeing gait, or equinus deformity of the ankle, is defined as a limitation of passive dorsiflexion beyond the neutral position, considered to be caused by spastic or contracted calf muscles [7, 10]. Several surgical procedures have attempted to resolve these abnormalities; distal hamstring lengthening (DHL), rectus femoris transfer (RFT), and tendo-Achilles lengthening (TAL) are commonly performed as part of a single event multilevel surgery (SEMLS) [11‐13].

Transverse plane deformities, such as, in-toeing gait due to excessive femoral anteversion or internal tibial torsion, are also common in patients with CP and can be corrected by derotational osteotomy of the femur or tibia [5, 14‐16]. Foot deformities including cavovarus or planovalgus deformity also contribute to problems in the transverse plane and can be corrected by corrective osteotomy and/or tendon surgery [14].

Several studies have reported improved motion in the sagittal plane after DHL, RFT, and TAL [3, 12, 13, 17‐21]. However, to our knowledge, there have been few studies on changes in transverse plane kinematics after soft tissue surgery. Furthermore, transverse plane kinematics have not been frequently reported because of the increased variability of joint motion angle in the transverse plane due to differences in marker placement [22, 23] and soft tissue artifact at the thigh [24, 25]. Chong et al. [26] reported that the medial hamstring plays an important role in in-toeing gait, and Steinwender et al. [27] reported that soft tissue surgery that included multilevel medial lengthening of this muscle improved the internal rotation of the hip. This finding was supported by the results obtained by Jung et al. [28], who reported that pelvic retraction and hip internal rotation improved following soft tissue surgery.

Considering this, we hypothesized that DHL, RFT, and TAL can affect kinematics in the transverse plane as well as in the sagittal plane. There are no reports available regarding kinematic changes in the transverse plane for parts of the lower extremities, except for the hip and pelvis, following soft tissue surgery such as DHL, RFT, or TAL. Therefore, we performed this study to evaluate the changes in transverse kinematics after soft tissue surgery—namely, DHL, RFT, or TAL—using three-dimensional (3D) gait analysis.

This retrospective study was approved by the institutional review board at our institute (a tertiary referral center for CP and the need for informed consent was waived.

Inclusion criteria were as follows: (1) consecutive ambulatory patients with CP who underwent SEMLS including DHL with semitendinosus transfer (STT), RFT, or TAL; and (2) patients for whom preoperative and 1-year postoperative 3D gait analysis were carried out. Exclusion criteria were as follows: (1) patients who underwent concomitant surgery which could affect transverse kinematics, such as femoral derotation osteotomy, tibial derotation osteotomy, or foot surgery (calcaneal lengthening osteotomy, tibialis posterior tendon split transfer, tibialis anterior tendon split transfer, Dwyer osteotomy, tibialis posterior tendon lengthening, triple arthrodesis, or triple osteotomy), 2) patients with a history of gait-correcting surgery or selective dorsal rhizotomy, and (3) patients who had incomplete or missing 3D gait analysis data.

From the medical record review, patients’ demographics including sex, age at surgery, gross motor function classification system (GMFCS) level, anatomical type of CP and type of concomitant surgeries, was obtained.

Between 2003 and 2016, preoperative and postoperative 3D gait analysis was carried out on 430 patients (706 limbs) who underwent SEMLS. After implementation of the inclusion and exclusion criteria, 156 patients (213 limbs) were finally enrolled in this study. The majority of patients were classified as GMFCS level I according to functional classification, and bilateral involvement according to type of involvement. The mean age at the time of surgery was 8.4 ± 3.2 years (range, 4.4 to 20.9). Average interval between preoperative and postoperative 3D gait analysis was 1.2 ± 0.8 years. The total number of surgical procedures was 507 (2.4 per limb) (Table 1).

Sagittal kinematics including knee flexion at initial contact, ankle dorsiflexion at initial contact, and mean ankle dorsiflexion in the stance phase were significantly improved after SEMLS (p < 0.001 in all cases). Transverse kinematics, including the mean tibial rotation and mean foot progression angle, were improved to a more external angle after soft tissue surgery (− 2.9°, p = 0.004 and − 9.5°, p < 0.001, respectively) (Table 2).

After adjusting the changes in transverse kinematics for multiple factors, the mean hip rotation was significantly improved to a more external angle by RFT (p = 0.019), and the foot progression angle was also significantly improved to a more external angle by TAL (p = 0.028) (Table 3). The change in mean hip rotation in patients who underwent RFT was to a more external rotation (4.7°) compared to those who did not undergo RFT (Fig. 1). Similarly, the change in foot progression angle in patients who underwent TAL was to a more external rotation (3.9°) compared to those who did not undergo TAL (Fig. 2).

Soft tissue surgeries including DHL with STT, RFT, and TAL are usually performed to correct abnormalities in the sagittal plane. Previous studies have reported good results following soft tissue surgery; however, kinematic changes in the transverse plane after soft tissue surgery remain unclear.

Initially, we hypothesized that kinematics in the transverse plane would be altered after DHL with STT, RFT, and TAL. Because all procedures target structures that affect the medial side of the lower leg, it is a reasonable assumption that the biomechanics will be different after these surgeries. The semitendinosus and gracilis are considered primary internal rotators of the tibia, and their points of insertion are located at the medial side of the tibia [30]. Therefore, reduction in ligament tightening following DHL alone or DHL with RFT would decrease the internal rotation torque. Furthermore, the pathway of the rectus femoris and its origin are located more laterally than those of the gracilis. If RFT is performed, the direction of force to the medial side of the knee will change from internal to external. We believe that this is why we observed altered hip and tibial rotation following DHL with STT and RFT.

In a study by Jung et al. [28], hip rotation after SEMLS became more external in CP patients with diplegia. The authors frequently performed DHL and RFT together, but the two operations were not separately analyzed. However, Lovejoy et al. [31] reported that hamstring lengthening caused decreased hip internal rotation, although not to the extent that an internal rotation gait pattern changed to an external rotation gait pattern. In addition, Arnold et al. [32] reported that the medial hamstring, adductor brevis, adductor longus, and gracilis were unlikely to substantially contribute to excessive internal rotation of the hip. For this reason, they suggested that surgical lengthening of the hamstrings or adductors is an inappropriate approach to reduce excessive internal rotation of the hip. Therefore, internal rotation of the hip cannot be changed to external rotation, even if DHL is performed. Overall, a rotational change in the transverse plane occurs after SEMLS, but the mechanisms behind this change are unknown. Our study showed that hip rotation in the transverse plane tends toward external rotation after RFT.

Lofterød et al. [33] reported that the foot progression angle became more external after SEMLS, although other factors did not have statistically significant relationships with SEMLS. Our study showed that not only the foot progression angle, but also tibial rotation became more external following SEMLS. However, while the above-mentioned study only considered patients with diplegia, we presented results for patients with both hemi- and diplegia obtained from a larger group (156 patients, 213 limbs). Hadley et al. [20] also reported that the foot progression angle changed toward a more external angle following multiple soft tissue release, although the reasons for this were unclear.

A previous study showed that the point of Achilles tendon insertion was approximately 2 cm from the midline to the medial side [34]. We believe that this may be the reason the foot progression angle became more external after TAL. In addition, the current study demonstrated that the change in foot progression angle was significantly associated with TAL. If the Achilles tendon is tightened, the calcaneus would rotate internally, causing other components of the foot to rotate internally along the calcaneus. Therefore, once the Achilles tendon was lengthened following TAL, the forces driving the internal rotation of the foot decreased.

There were some limitations in this study. First, we used the Helen Hayes marker set to perform 3D gait analysis. Knee rotation was calculated from the relative motion between the distal medial-lateral axis of the shank and medial-lateral axis of the thigh. However, we used the term “tibial rotation” instead of “knee rotation” because knee rotation is minimal and considered as tibial rotation in clinical practice. Therefore, the results of tibial rotation could be inaccurate if problems involving the knee were present. Second, several studies reported low repeatability in the transverse plane due to variabilities in the alignment of markers [22, 23]. However, markers were placed by a skillful operator for consistent identification of anatomical landmarks and positioning of markers in this study. Third, SEMLS is the treatment of choice for the correction of gait pattern in patients with CP. This poses significant challenges for the analysis of individual surgical methods such as DHL with STT, RFT, and TAL. Therefore, we have compensated for this restriction as much as possible through statistical methods. Fourth, a control group that consisted of patients with CP who had not undergone surgery was not included due to the retrospective design of this study. Bone remodeling occurs with age in children and can lead to changes in transverse kinematics. Two short-term longitudinal study showed remodeling of femoral anteversion in patients with CP who did not underwent femoral derotation osteotomy [15, 35]. However, several studies showed that remodeling of femoral anteversion did not occur during development in children with CP [36‐38]. Therefore, long-term longitudinal study regarding the natural history of femoral and tibial torsion in CP is required.

In conclusion, our study demonstrates that the transverse kinematics were significantly improved to a more external angle after soft tissue surgery in patients with CP. Therefore, clinicians should consider that soft tissue surgery can affect the transverse plane kinematics in patients with CP. However, appropriate osteotomy should be considered for the correction of transverse plane deformities because of the small degrees of change after soft tissue surgery.