The impact of hamstring lengthening on stance knee flexion at skeletal maturity in ambulatory cerebral palsy.

This study reports the long-term outcomes of hamstring lengthening to treat flexed knee gait in children with ambulatory cerebral palsy (CP) after skeletal maturity. This retrospective longitudinal observational study used instrumented gait analysis (GA) <8 and >15 years old in children with bilateral CP. The primary variable was knee flexion in stance phase. Eighty children (160 limbs) were included; 49% were male, 51% female. Mean age at first GA was 6.0 (SD: 1.2) years and 19.6 (SD: 4.5) years at final GA. Mean follow-up was 13.7 (SD: 4.7) years. Children were classified as Gross Motor Function Classification System I-8, II-46 and III-26. Average Gross Motor Function Measure Dimension D was 72% (SD: 20%). Hamstring lengthenings occurred once in 82, twice in 54 and three times in 10 limbs. From initial to final GA, average knee flexion in stance was unchanged, 27.8° (SD: 14.8°) to final 27.0° (SD: 11.2°; P = 0.54). Knee flexion at foot contact was 39.6° (SD: 13.0°), improving to final GA of 30.7° (SD: 10.6°; P < 0.001). Initial gait deviation index was 65.8 (SD: 31.9), improving to final 78.9 (SD: 28.2; P < 0.001). Older age, males and concomitant plantar flexor lengthening predicted change toward more flexed knee gait. Hamstring lengthening did not lead to back-kneeing gait at maturity while maintaining childhood stance phase knee flexion. A subgroup still developed significant flexed knee gait posture and may have benefited from more aggressive treatment options. This outcome may also be impacted by diverse functional levels, etiologies and treatments of flexed knee gait.

The change in sagittal plane gait patterns from childhood to maturity in bilateral cerebral palsy.

BACKGROUND: The longitudinal stability of sagittal gait patterns in diplegic cerebral palsy (CP), stratified using the Rodda classification, is currently unknown. RESEARCH QUESTION: What is the trajectory of sagittal plane gait deformities as defined by the Rodda classification in a large cohort treated with orthopedic surgery guided by gait analysis? METHODS: A retrospective study utilized gait analysis to evaluate sagittal gait parameters before age 8 and after age 15 years. Individual limbs were categorized at each time point according to the Rodda classification based on mean sagittal plane knee and ankle angle during stance. Welch's t-tests compared gait variables from early childhood with maturity and examined changes associated with plantarflexor lengthening surgery. RESULTS: 100 youth with CP were evaluated twice: at a mean age of 5.49 ± 1.18 and 19.09 ± 4.32 years, respectively. Gross Motor Function Classification System distribution at maturity was I (10.5 %), II (55.2 %), III (28.6 %), and IV (5.7 %). At the initial visit, most limbs were in either true equinus (30 %) or jump-knee gait (26.5 %). At maturity, crouch gait (52.5 %) was the most common classification, of which 47.6 % were mild (1-3 standard deviations from age-matched norm; 21°-30°) and 52.4 % moderate or severe. For the entire cohort, at initial and final visits, respectively, mean knee flexion in stance was 26.8°±14.8° and 25.9°±11.4° (p = 0.320), ankle dorsiflexion in stance increased from -0.3°±11.5° to 9.0°±6.0° (p < 0.001), and passive knee flexion contracture was -2.3°±7.0° and -3.9°±8.0° (p = 0.043). In children who started in true equinus, apparent equinus, and crouch, there was no difference in stance phase knee flexion at maturity between those who underwent plantarflexor lengthenings versus those who did not (p > 0.18). SIGNIFICANCE: The trend in this cohort was toward crouch with increased stance phase ankle dorsiflexion from early childhood to maturity. Plantarflexor lengthenings were not a significant factor in the progression of stance phase knee flexion.

Growth in individuals with Saul-Wilson syndrome.

Saul-Wilson syndrome (SWS) is a rare autosomal recessive disorder characterized by microcephalic primordial dwarfism, spondyloepimetaphyseal dysplasia, characteristic facial findings, clubfoot, brachydactyly, bilateral cataracts, and hearing loss. Recently, recurrent mutations in COG4, encoding a component of the Conserved Oligomeric Golgi (COG) complex, were identified. We created detailed growth curves for stature, weight, and head circumference, as well as weight-for-length and weight velocity charts for younger children, derived from hundreds of data points obtained by retrospective chart review from 14 individuals with molecularly-confirmed SWS. In addition, we performed statistical comparisons of height-for-age model fits before and after initiation of growth hormone supplementation, and found that this therapy does not appear to influence height in individuals with SWS. We hope that these charts will represent valuable tools for clinicians, both in assessing whether SWS seems an appropriate diagnosis, as well as to monitor growth of affected individuals. In particular, we hope that our detailed growth characterization will reduce morbidity resulting from unnecessarily aggressive nutritional interventions by well-intentioned physicians trying to promote weight gain, an unrealistic goal in this genetically-determined cause of primordial dwarfism.

Comparing groups of time dependent data using locally weighted scatterplot smoothing alpha-adjusted serial T-tests.

BACKGROUND: In statistical analysis of time series researchers often pick key points from curves and run the venerable analysis of variance (ANOVA) to determine if a difference exists between groups. However, this approach fails to compare most of the data across time and thereby may throw out potentially valuable inferences. RESEARCH QUESTION: This study illustrates a novel method termed LOESS alpha-adjusted serial t-testing (LAAST). LAAST employs locally weighted scatterplot smoothing (LOESS) on the data, serial correlation to make alpha adjustments, and point-wise Welch's t-tests to determine regional significance when comparing groups of time dependent data. It was expected that LAAST gives similar results to random field theory (RFT) based inferences while overcoming its shortcomings with respect to longitudinal data analysis. METHODS: Two data sets were analyzed with LAAST and RFT. The first contained two groups of five simulated random sinusoidal waveforms such that both inline time-series and equivalent time-offset longitudinal conditions were represented. The second data set was comprised of publicly available medial gastrocnemius forces from individuals with (N = 27) and without (N = 16) pain. RESULTS: Results for both data sets indicated similar corrected alpha levels regardless of analysis type, but the applied alpha level corrections were less conservative for LAAST than RFT or Holm-Bonferroni corrections, but often more conservative than Hochberg corrections. SIGNIFICANCE: Analysis methods employing functional ANOVA and RFT have enabled researchers to effectively run comparisons between groups at all points within the time series and are gaining popularity. However, in some correction methods for multiple comparisons the alpha level correction can in turn lead to inflation of type II error. These results suggest that LAAST is comparable to RFT while also being appropriate for longitudinal type time series data analysis. Additionally, its use of Welch's t-tests improves its validity on non-normally distributed data.

Factors influencing post-surgical variability in StepWatch data in youth with cerebral palsy.

BACKGROUND: Over the past several years, activity monitors have become very popular in the general population, and due to their low cost and ease of use, are starting to be seen as clinical tools for the assessment of interventions. This presents researchers with the opportunity to better understand how activity, or lack thereof, is related to the recovery of patients. However, even in individuals without disabilities, there is a high degree of variability in activity monitor data which must be better understood in order to produce clinically meaningful interpretation of such data. RESEARCH QUESTION: What sources of variability contribute the most to the daily scatter in activity data as measured by StepWatches in youth with Cerebral Palsy (CP)? In particular, do non-clinical factors such as weather and location contribute to this variability significantly? METHODS: This was a retrospective study making use of data from our activity monitoring protocol of youths with CP who obtain single event multi-level surgeries. Before and after these surgeries, 57 such youths aged 4.2-21.3 years were issued StepWatches to monitor daily activity for 8 day periods over 24 months duration. Weather data and walk scores for the patients' home locations were collected from online databases. Steps per hour were predicted from clinical and environmental data using bootstrapped regression to determine the stability of regression coefficients and the percent variability explained by each variable. RESULTS: Time since surgery, age, season, GMFCS level, and surgical burden were significant variables in the model. Of them, GMFCS level was most important and explained nearly 16% of the variability in the data. Temperature, precipitation, and walk score had small effects on step count variance. SIGNIFICANCE: Understanding sources of variability in step-counts is important if such a measure is to be used as a clinical measure of recovery, and may be important in the consideration of future surgical planning.

Reliability and minimal detectable change in foot pressure measurements in typically developing children.

BACKGROUND: In pedobarography, clinically meaningful comparison of measurements within or between subjects is limited by data variability and measurement error. This study aims to determine the components of the minimal detectable change (MDC) in impulse across all foot regions and the reliability of these measures. METHODS: A convenience sample of foot pressures from 108 visits by normal, healthy subjects aged 2-17 years was studied. Each subject had three pedobarograph measurements taken per foot, with six subjects returning for a second visit for assessment of day-to-day variability. Using a five-region mask, segmental impulses were determined, and from these we obtained the coronal plane pressure index (CPPI). Inter-rater, intra-rater, and day-to-day data were analyzed using intraclass correlation coefficients (ICC) to quantify reliability. Variability of the data was analyzed to quantify the MDC. RESULTS: Inter- and intra-rater reliability was high for all measurements while variability was low, indicating small direct measurement error. Generally, the largest contributing factor to the MDC was day-to-day variability. Step-to-step variability was more dependent on foot segment than age although minor age-related changes were noted. Finally, the high relative variability in the CPPI and the medial mid foot impulse resulted in very high MDCs for these measures.

Concurrent surgeries are a factor in predicting success of rectus transfer outcomes.

Cerebral palsy (CP) patients exhibit a wide range of functional responses to specific surgical interventions. The variability associated with surgical outcomes within this patient population precipitates the need to determine the functional effects in advance of the surgery. Pilot studies have indicated that the application of statistical models towards predicting functional outcomes of surgery produces clinically meaningful results. The purpose of this study was to evaluate the influence of pre-operative kinematic parameters in conjunction with concurrent surgeries on rectus transfer (RT) outcomes. Gait analysis data were collected from 68 patients (94 legs) pre- and post-operatively within a year of surgery. Patients were divided into four groups. Group 1 (N=14) contained only legs experiencing a rectus transfer. Group 2 (N=30) contained legs experiencing RT and hamstring lengthening (HL). Group 3 (N=7) contained legs experiencing RT and lower leg surgeries: either Achilles lengthening (AL), or gastrocnemius lengthening (GL). Finally, Group 4 (N=43) contained legs which experienced all four surgeries (RT+HL+AL+GL) concurrently. Multivariate jackknifed linear regression was used to predict the change in knee range of motion, DeltaROM, due to surgery on a group-by-group basis. In each case, the sole significant pre-surgical variable was the range of motion. Group 1 legs had the best correlation (R=0.846) between pre-surgical ROM and DeltaROM whereas Group 4 legs had the worst correlation (R=0.661). Group 3 results were not significant. Group 1 DeltaROM were also significantly lower as a whole, than all other groups, due to a larger percentage of patients needing the surgery to change timing of flexion in swing phase. When compared to modeling without consideration of surgery, modeling by surgical group produced results with fewer outliers and better correlation coefficients. Regardless of surgical group, patients with a lower pre-surgical ROM (<40 degrees ) were more likely to experience an increase in ROM as a result of surgical intervention compared to their peers with an elevated pre-surgical ROM who were likely to see a decrease in ROM. Pre-surgical ROM therefore serves as a useful predictor of outcome in rectus transfer surgery.