Long-term results of multilevel surgery in adults with cerebral palsy.

BACKGROUND: Deterioration of gait in adolescent and adult patients with cerebral palsy can be associated with multiple factors. Multilevel surgery (MLS) is one option in adults with cerebral palsy to improve gait function with encouraging short-term results. It is a question whether these improvements are maintained over time. METHODS: In a retrospective consecutive cohort study, adults with bilateral spastic cerebral palsy (BSCP) treated with MLS between 1995 and 2011 were scanned for potential inclusion. Patients needed to fulfill the following inclusion criteria: age at MLS > 17, standardized three-dimensional gait analysis (3D-GA) including clinical examination at pre-operative (E0), a short-term follow-up (E1) and at least seven years (E2) after the index MLS. Twenty adults (10 women, 10 men) with a Gross Motor Function Classification Level (GMFCS) I-III and a mean age at MLS of 24.8 years were included in this study. The average long-term follow-up was 10.9 years. The Gait Profile Score (GPS) was used as primary outcome measure. RESULTS: The GPS improved significantly from 13.8° before surgery to 11.2° at short-term (p = 0.007) and to 11.3° at long-term follow-up (p = 0.002). Mean GPS showed a slight deterioration between E1 and E2 due to a minority of six patients (30%) who showed a significant loss of correction. CONCLUSION: Surgical treatment in adults with BSCP was feasible and effective in the long-term. Significant improvement of gait and function was maintained in the majority of patients, while some patients were prone to develop crouch gait, hip flexion contractures, or pain.

Proximal versus distal femoral derotation osteotomy in bilateral cerebral palsy.

AIM: Femoral derotation osteotomy (FDO) can be conducted either proximally or distally to correct internal rotation gait (IRG) and increased anteversion in children with cerebral palsy (CP). Previous studies with limited numbers of participants have presented comparable short-term static and kinematic outcomes for both techniques. The objective of this retrospective multicentre study was to verify this thesis with a larger number of patients. METHOD: In total, 119 children with CP and IRG were included after matching the groups on preoperative mean stance hip rotation: 67 intertrochanteric (proximal group) FDO (average age at surgery 9y [SD 3y]); 52 supracondylar (distal group) FDO (average age at surgery 12y [SD 3y]). One random limb of each child was analyzed. Both transverse plane kinematic gait data and torsional parameters of clinical examination were assessed preoperatively and postoperatively. RESULTS: On average, both groups' mean hip rotation and midpoint of hip rotation improved postoperatively, and groups did not differ preoperatively or postoperatively for these variables (p>0.05). INTERPRETATION: The osteotomy location does not influence short-term gait kinematics nor static measurements. Therefore, the choice of performing proximal or distal FDO in children with CP for treatment of IRG and increased anteversion should be motivated by considerations other than these outcome parameters. WHAT THIS PAPER ADDS: Femoral derotation osteotomy (FDO) location does not influence mean stance hip rotation in cerebral palsy (CP). FDO location does not influence mid-point hip rotation in CP. FDO location should be motivated by concomitant procedures performed.

Mid-term development of hamstring tendon length and velocity after distal femoral extension osteotomy in children with bilateral cerebral palsy: a retrospective cohort study.

AIM: Flexed knee gait can be treated with distal femoral extension osteotomy (DFEO) and additional patellar tendon advancement (PTA) in children with cerebral palsy (CP). This study assesses changes in hamstring muscle tendon length (MTL) and velocity after DFEO (+PTA). METHOD: Nineteen children (mean age 13y [standard deviation 3y] at surgery) with CP and flexed knee gait who were treated with DFEO (15 limbs) or DFEO+PTA (10 limbs) were retrospectively included in this study. Gait analyses were performed preoperatively (E0), 1 year postoperatively (E1), and for 10 limbs additionally 2 to 5 years postoperatively (E2). Hamstring MTL and velocities were assessed at all examination dates using OpenSim. RESULTS: Hamstring MTL and velocity did not change significantly over time. From E0 to E1, knee flexion in stance improved for both DFEO and DFEO+PTA (p<0.05), knee flexion in swing only improved after DFEO+PTA (p<0.05). The improved knee flexion in stance and swing was maintained at E2. INTERPRETATION: DFEO led to a significant improvement in knee kinematics at E1 which was maintained at E2. DFEO seems to prevent recurrent hamstring tightness but does not lead to lengthened or fastened hamstrings. WHAT THIS PAPER ADDS: Distal femoral extension osteotomy (DFEO) does not change hamstring muscle tendon length. DFEO does not change hamstring lengthening velocity. DFEO leads to a significant improvement in knee kinematics. Changes in knee kinematics after DFEO can be maintained at mid-term. DFEO seems to prevent recurrent hamstring tightness.

Three-dimensional motion analysis for validation of shoulder internal rotation.

BACKGROUND: 10% of the points for the Constant-Murley score (CMS) are allocated for the capacity for internal rotation (IR), measured as unassisted active movement of the dorsum of the hand or the thumb to reach different anatomical landmarks. However, there is little information about the validity of this method and no three-dimensional measurement of the degree of IR that is necessary to reach these landmarks. METHODS: Sixteen volunteers with healthy shoulders were recruited. The degree of IR was defined using the following landmarks as described in the CMS: (1) lateral aspect of thigh, (2) buttock, (3) sacroiliac joint, (4) level of waist, (5) vertebra T12, (6) interscapular. The validity of IR measurement was assessed by simultaneous 3D motion analysis. RESULTS: Using the thumb as pointer, there were significant increases in IR from 39.3° at position 1 to 80.4° at position 2, followed by 105.1°, 108.6°, 110.1°, and 125.3° at position 3-6. Taking the dorsum of the hand as pointer, there were significant increases in IR between all positions, starting from 71.2° (position 1) and followed by 99.3°, 104.1°, 110.3°, 115.2°, and 119.7° at positions 2 to 6. Comparing the two measurement methods, a significant difference was found for the amount of IR between positions 1 and 2. CONCLUSION: Measurement of IR as described in the CMS is a suitable method. However, there was an increase of only 10° in IR between positions 3 and 5, which may be hard to measure with a standard goniometer in clinical practice.

Motion patterns in activities of daily living: 3- year longitudinal follow-up after total shoulder arthroplasty using an optical 3D motion analysis system.

BACKGROUND: Total shoulder arthroplasty (TSA) can improve function in osteoarthritic shoulders, but the ability to perform activities of daily living (ADLs) can still remain impaired. Routinely, shoulder surgeons measure range of motion (ROM) using a goniometer. Objective data are limited, however, concerning functional three-dimensional changes in ROM in ADLs after TSA in patients with degenerative glenohumeral osteoarthritis. METHODS: This study included ten consecutive patients, who received TSA for primary glenohumeral osteoarthritis. The patients were examined the day before, 6 months, and 3 years after shoulder replacement as well. We compared them with a control group (n = 10) without any shoulder pathology and measured shoulder movement by 3D motion analysis using a novel 3 D model. The measurement included static maximum values, the ability to perform and the ROM of the ADLs "combing the hair", "washing the opposite armpit", "tying an apron", and "taking a book from a shelf". RESULTS: Six months after surgery, almost all TSA patients were able to perform the four ADLs (3 out of 40 tasks could not be performed by the 10 patients); 3 years postoperatively all patients were able to carry out all ADLs (40 out of 40 tasks possible). In performing the ADLs, comparison of the pre- with the 6-month and 3-year postoperative status of the TSA group showed that the subjects did not fully use the available maximum flexion/extension ROM in performing the four ADLs. The ROM used for flexion/extension did not change significantly (preoperatively 135°-0° -34° vs. 3 years postoperatively 131° -0° -53°). For abduction/adduction, ROM improved significantly from 33°-0° -27° preoperatively to 76° -0° -35° postoperatively. Compared to the controls (118°) the TSA group used less ROM for abduction to perform the four ADLs 3 years postoperatively. CONCLUSION: TSA improves the ability to perform ADL and the individual ROM in ADLs in patients with degenerative glenohumeral osteoarthritis over the course of 3 years. However, TSA patients do not use their maximum available abduction ROM in performing ADLs. This is not related to limitations in active ROM, but rather may be caused by pathologic motion patterns, impaired proprioception or both.

Proprioception 3 years after shoulder arthroplasty in 3D motion analysis: a prospective study.

BACKGROUND: This prospective study evaluated the development of proprioception over the course of 3 years after shoulder arthroplasty. METHODS: Twenty-one patients were enrolled who underwent total shoulder arthroplasty (n = 10) or hemiarthroplasty (HEMI) (n = 11) for shoulder osteoarthritis. All patients were examined 1 day before the operation, 6 months and 3 years after surgery in a motion analysis study with an active angle-reproduction (AAR) test. RESULTS: Overall proprioception measured by the AAR deteriorated significantly 3 years after surgery [from 6.6° (SD 3.1) to 10.3° (SD 5.7); p = 0.017] and was significantly worse than in the control group [10.3° (SD 5.7) vs. 7.8° (SD 2.3); p = 0.030). In the HEMI subgroup, 3 years after shoulder replacement, there is a significant deterioration of proprioception at 30° of external rotation [from 3.1° (SD 3.5) to 12.8° (SD 10.7); p = 0.031]. On average, in the TSA subgroup proprioception deteriorated from 7.1° (SD 3.1) to 8.6° (SD 1.4) and in the HEMI subgroup from 6.1° (SD 2.1) to 12.4° (SD 8.3). The comparison of postoperative impairment of proprioception between the TSA and HEMI subgroup showed significantly worse proprioception for the HEMI subgroup at 30° of external rotation [9.8° (SD 10.1) vs. 1.6° (SD 6.3) in the TSA group; p = 0.046]. CONCLUSION: In conclusion, proprioception that was measured by an AAR test remained unchanged or deteriorated 3 years after shoulder arthroplasty. The postoperative deterioration of proprioception was more distinctive in HEMI than in TSA group.

Preoperative hip rotation moments do not predict long-term development after femoral derotation osteotomy in children with cerebral palsy.

BACKGROUND: Femoral derotation osteotomy (FDO) is the standard treatment for internal rotation gait (IRG) in children with cerebral palsy (CP) although high rates of recurrence have been reported recently. Various factors associated with recurrence could be identified, but no predictor named. RESEARCH QUESTIONS: Does FDO lead to a change of internal transversal hip moments? Are preoperative internal transversal hip moments a predictor for recurrence of IRG? METHODS: 41 children with spastic bilateral CP and 72 limbs that received a FDO (10.4 ±â€¯2.7 years at surgery) were included retrospectively. Kinematic data were analyzed pre- (2 ±â€¯3 months), postoperatively (12 ±â€¯3 months) and at long-term follow-up (at least five years postoperatively; 84 ±â€¯13 months), internal transversal hip moments were analyzed pre- and postoperatively. RESULTS: The maximum peaks of the internal hip rotation moment during loading response decreased significantly (p = 0.003). The minimum during the second half of the stance phase increased significantly (p = 0.004) and the initially internal externally rotating moment changed to an internal internally rotating moment. No correlation between changes in hip rotation from postoperatively to the long-term follow-up and the preoperative internal hip rotation moment could be identified. SIGNIFICANCE: FDO leads to changes in internal hip rotation moments. Preoperative internal hip rotation moments can't be used as predicting factor for recurrence of IRG. The data suggest, that recurrence of IRG depends less on patient specific motion patterns, but more on the time point of surgery and the therapy of all concomitant deformities during SEMLS.

Long-term development of overcorrection after femoral derotation osteotomy in children with cerebral palsy.

BACKGROUND: Recent studies showed rates of recurrence of internal rotation gait (IRG) after femoral derotation osteotomy (FDO) up to 40%. Some surgeons even advice overcorrection during FDO to avoid a later recurrence. RESEARCH QUESTION: Evaluation of the long-term development of limbs with initial overcorrection after FDO. METHODS: 29 limbs of 20 children (9.9 ±â€¯3.2 years at surgery) with IRG, cerebral palsy (CP) and more than 5° external hip rotation postoperatively were included retrospectively. A gait analysis and clinical examination were performed preoperatively (less than one year, E0), postoperatively (9-23 months, E1) and at the long-term follow-up (at least five years postoperatively, E2). Differences between those children that remained overcorrected at E2 and those with a hip rotation within normal range at E2 were evaluated. RESULTS: At E2 41% of these limbs remained overcorrected, 52% showed a hip rotation within normal range and 7% showed recurrence of IRG. A comparison of those limbs that remained overcorrected and those ending within normal range revealed neither a difference in age at surgery nor in static and dynamic torsional parameters at E0 and E1 except for pelvic rotation. A significantly larger pelvic internal rotation at E1 for those with remaining overcorrection could be identified. SIGNIFICANCE: A general overcorrection during FDO in children with CP to avoid recurrence of IRG cannot be recommended, as 41% remain overcorrected. Preoperative predictors for long-term development couldn't be identified. If pelvic mal-rotation is corrected, hip rotation may change into normal range over the time in combination with the development of a flexed knee gait.

Equinus Correction During Multilevel Surgery in Adults With Cerebral Palsy.

BACKGROUND: Equinus foot deformity constitutes a common gait disorder in ambulatory adults with bilateral spastic cerebral palsy (BSCP). The outcome after intramuscular aponeurotic lengthening in the context of single-event multilevel surgery (SEMLS) in adulthood has not been investigated. METHODS: We followed a group of 31 ambulatory adults with BSCP and equinus who underwent SEMLS including gastrocnemius-soleus intramuscular aponeurotic recession or Achilles tendon lengthening. All patients were analyzed preoperatively and at least 1 year (mean follow-up period: 1.6 years) postoperatively by clinical examination and 3-dimensional instrumented gait analysis including the Gait Profile Score (GPS). RESULTS: Clinical examination showed no significant improvement of ankle dorsiflexion ( P = .5) and an unchanged plantarflexion ( P = .7) with knee extended but a significant postoperative reduction of spasticity in the calf muscle ( P = .0001) as measured by clinical examination following the modified Ashworth scale. Significant improvement of mean ankle dorsiflexion in stance and swing ( P = .0001) was found. The GPS decreased and improved significantly (15.9 ± 4.6 to 11.4 ± 3.1; P = .0001). Persistence of equinus and calcaneal gait indicating under- and overcorrection at follow-up was found in 1 patient (3%), respectively. CONCLUSION: Intramuscular gastrocnemius-soleus aponeurotic recession is part of multilevel surgery corrected equinus deformity in adults. The increase in muscle length led to significant improvement of kinetic and kinematic parameters during walking without a loss of muscle strength and push-off capacity. The risk of overcorrection after equinus correction in adults with BSCP was found to be relatively low. LEVEL OF EVIDENCE: Level IV, retrospective case series.

Electromagnetic bone segment tracking to control femoral derotation osteotomy-A saw bone study.

Correction of rotational gait abnormalities is common practice in pediatric orthopaedics such as in children with cerebral palsy. Femoral derotation osteotomy is established as a standard treatment, however, different authors reported substantial variability in outcomes following surgery with patients showing over- or under-correction. Only 60% of the applied correction is observed postoperatively, which strongly suggests intraoperative measurement error or loss of correction during surgery. This study was conducted to verify the impact of error sources in the derotation procedure and assess the utility of a newly developed, instrumented measurement system based on electromagnetic tracking aiming to improve the accuracy of rotational correction. A supracondylar derotation osteotomy was performed in 21 artificial femur sawbones and the amount of derotation was quantified during the procedure by the tracking system and by nine raters using a conventional goniometer. Accuracy of both measurement devices was determined by repeated computer tomography scans. Average derotation measured by the tracking system differed by 0.1° ± 1.6° from the defined reference measurement . In contrast, a high inter-rater variability was found in goniometric measurements (range: 10.8° ± 6.9°, mean interquartile distance: 6.6°). During fixation of the osteosynthesis, the tracking system reliably detected unintentional manipulation of the correction angle with a mean absolute change of 4.0° ± 3.2°. Our findings show that conventional control of femoral derotation is subject to relevant observer bias whereas instrumental tracking yields accuracy better than ±2°. The tracking system is a step towards more reliable and safe implementation of femoral correction, promising substantial improvements of patient safety in the future. © 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 35:1106-1112, 2017.

What is the price for the Duchenne gait pattern in patients with cerebral palsy?

Duchenne gait is characterized by trunk lean towards the affected stance limb with the pelvis stable or elevated on the swinging limb side during single limb stance phase. We assessed the relationship between hip abduction moments and trunk kinetics in patients with cerebral palsy showing excessive lateral trunk motion. Data of 18 subjects with bilateral spastic cerebral palsy (CP) and 20 aged matched typically developing subjects (TD) were collected retrospectively. Criteria for patient selection were barefoot walking without aid presenting with excessive lateral trunk motion. Subjects had been monitored by conventional 3D gait analysis of the lower extremity including four markers for monitoring trunk motion. Post-hoc, a generic musculoskeletal full body model (OpenSim 3.3) assuming a rigid trunk articulated to the pelvis by a single ball joint was applied for analyzing joint kinematics and kinetics of the lower limb joints including this spine joint. Joint angle ranges of motion, maximum joint moments and powers in the frontal plane as well as mechanical work were calculated and averaged within groups showing prominent differences between groups in all parameters. To the best of our knowledge, this is the first work explicitly looking into the kinetics of Duchenne gait in patients with CP, clinically known as compensation for unloading hip abductor muscles. The results show that excessive lateral trunk motion may indeed be an extremely effective compensation mechanism to unload the hip abductors in single limb stance but for the price of a drastic increase in demand on trunk muscle effort and work.

The influence of timing of knee recurvatum on surgical outcome in cerebral palsy.

Recent reports have shown that timing of genu recurvatum (GR) might be caused by different underlying factors and that equinus leads to GR especially during early stance. The purpose of this study was to investigate the reduction of GR after surgical correction of equinus in children with bilateral spastic cerebral palsy and whether the children with early and late type GR show differences in reduction of knee hyperextension after a surgery. In 24 limbs (mean age 10.3 years, GMFCS I-III) showing equinus and GR the kinematics of the knee and ankle as well as the kinetics of the knee were evaluated before and one year (mean follow up period: 12.8 months) after surgical correction of equinus. The study was approved by the local ethical committee. Limbs with early type GR showed a reduction by 11.1° (p<0.001) and those with late type GR by 6.0° (p<0.049) in GR after surgery. Before surgery limbs with early type GR showed increased external extending moments, which decreased significantly after surgery. In contrast limbs with late GR did not show a significant reduction of those moments. The findings of this study underline the influence of equinus on early GR as an underlying factor. As equinus is attributed to early knee hyperextension and proximal factors are more important as underlying factors in late type GR, a classification into early and late onset GR is useful to identify underlying factors and to choose adequate treatment.

3D motion capture using the HUX model for monitoring functional changes with arthroplasty in patients with degenerative osteoarthritis.

BACKGROUND: Although shoulder hemiarthroplasty (SHA) can improve function in osteoarthritic shoulders, the ability to perform activities of daily living (ADL) may remain impaired. Shoulder surgeons routinely measure parameters such as range of motion, pain, satisfaction and strength. A common subjective assessment of ADL is part of the Constant Score (CS). However, there is limited objective evidence on whether or not shoulder hemiarthroplasty can restore normal range of motion (ROM) in ADL. METHODS: The study included eight consecutive patients (n=8; seven women, one man), who underwent SHA for glenohumeral osteoarthritis. The patients were examined the day before, as well as 6 months and 3 years after shoulder replacement. They were compared with a control group with no shoulder pathology, and shoulder movement was measured with 3D motion analysis using the "Heidelberg Upper Extremity" (HUX) model. Measurements included static maximum values and four ADL. RESULTS: Comparing the preoperative to the 3-year postoperative static maximum values, there were significant improvements for abduction from 50.5° (SD ± 3 2.4°) to 72.4° (SD ± 38.2°; p=0.031), for adduction from 6.2° (SD ± 7.7°) to 66.7° (SD ± 18.0°; p=0.008), for external rotation from 15.1° (SD ± 27.9°) to 50.9° (SD ± 27.3°; p=0.031), and for internal rotation from -0.6° (SD ± 3.9°) to 35.8° (SD ± 28.2°; p=0.031). There was a trend of improvement for flexion from 105.8° (SD ± 45.7°) to 161.9° (SD ± 78.2°; p=0.094) and for extension from 20.6° (SD ± 17.0°) to 28.0° (SD ± 12.5°; p=0.313). The comparison of the 3-year postoperative ROM between the SHA group and controls showed significant differences in abduction; 3-year postoperative SHA ROM 72.4° (SD ± 38.2°) vs. 113.5° (SD ± 29.7°) among controls (p=0.029). There were no significant differences compared to the control group in adduction, flexion/extension and rotation 3 years after SHA surgery. In performing the ADL, the pre- to the 6-month and 3-year postoperative status of the SHA group resulted in a significant increase in ROM in all planes (p<0.05). Comparing the preoperative to the 3-year postoperative ROM used in ADL, there was an improvement in the flexion/extension plane, showing an improvement trend from preoperative 85°-0°-25° to postoperative 127°-0°-38° (p=0.063). In comparison, controls used a significantly greater ROM during ADL with mean flexion/extension of 139°-0°-63° (p=0.028). For the abduction/adduction plane, ROM improved significantly from preoperative 25°-0°-19° to postoperative 78°-0°-60° (p=0.031). In comparison to controls with abduction/adduction of 118°-0°-37° 3 years postoperative, the SHA group also used significantly less ROM in the abduction/adduction plane (p=0.028). CONCLUSION: While SHA improves ROM in ADL in patients with degenerative glenohumeral osteoarthritis, it does not restore the full ROM available for performing ADL compared to controls. 3D motion analysis with the HUX model is an appropriate measurement system to detect surgery-related changes in shoulder arthroplasty.