Faculty Bibliography

This study sought to assess comorbidity profiles unique to early-onset-scoliosis (EOS) patients by employing cluster analytics and to determine the influence of isolated comorbidity clusters on perioperative complications, morbidity and mortality using a high powered administrative database. The KID database was queried for ICD-9 codes pertaining to congenital and idiopathic scoliosis from 2003, 2006, 2009, 2012. Patients <10 y/o (EOS group) were included. Demographics, incidence and comorbidity profiles were assessed. Comorbidity profiles were stratified by body systems (neurological, musculoskeletal, pulmonary, cardiovascular, renal). K-means cluster and descriptive analyses elucidated incidence and comorbidity relationships between frequently co-occurring comorbidities. Binary logistic regression models determined predictors of perioperative complication development, mortality, and extended length-of-stay (≥75th percentile). 25,747 patients were included (Age: 4.34, Female: 52.1%, CCI: 0.64). Incidence was 8.9 per 100,000 annual discharges. 55.2% presented with pulmonary comorbidities, 48.7% musculoskeletal, 43.8% neurological, 18.6% cardiovascular, and 11.9% renal; 38% had concurrent neurological and pulmonary. Top inter-bodysystem clusters: Pulmonary disease (17.2%) with epilepsy (17.8%), pulmonary failure (12.2%), restrictive lung disease (10.5%), or microcephaly and quadriplegia (2.1%). Musculoskeletal comorbidities (48.7%) with renal and cardiovascular comorbidities (8.2%, OR: 7.9 [6.6-9.4], p < 0.001). Top intra-bodysystem clusters: Epilepsy (11.7%) with quadriplegia (25.8%) or microcephaly (20.5%). Regression analysis determined neurological and pulmonary clusters to have a higher odds of perioperative complication development (OR: 1.28 [1.19-1.37], p < 0.001) and mortality (OR: 2.05 [1.65-2.54], p < 0.001). Musculoskeletal with cardiovascular and renal anomalies had higher odds of mortality (OR: 1.72 [1.28-2.29], p < 0.001) and extLOS (OR: 2.83 [2.48-3.22], p < 0.001). EOS patients with musculoskeletal conditions were 7.9x more likely to have concurrent cardiovascular and renal anomalies. Clustered neurologic and pulmonary anomalies increased mortality risk by as much as 105%. These relationships may benefit pre-operative risk assessment for concurrent anomalies and adverse outcomes. Level of Evidence: III - Retrospective Prognostic Study.

Background and Objective(s): Commercially available activity tracking devices present the opportunity to quantify functional changes in ambulatory status of patients with cerebral palsy (CP). The purpose of this study was to assess the accuracy of a waist-based and a wrist-based wearable device for tracking steps and walking distance of children with CP in a controlled setting. Study Design: Prospective cohort study. Study Participants & Setting: Participants were 27 children, mean age of 9.6 years (range, 4-15), with CP and a Gross Motor Function Classification System (GMFCS) score between I and III. Thirteen children were diagnosed with hemiplegia (48%), 13 with diplegia (48%), and 1 with quadriplegia (4%). Seventeen were classified as GMFCS I (63%), 3 as II (11%), and 7 as III (26%). Fourteen children wore ankle-foot orthoses (52%), one wore HKAFO (4%) and 12 wore no orthotics (44%). Four patients used Lofstrand crutches (15%), two used posterior rollators (7%), and 21 patients used no assistive devices (78%). Materials/Methods: Each participant was outfitted with both a waist-based activity tracker, FitBitTM One, and a wrist-based tracker, FitBitTM Flex. They ambulated at a self-selected speed for 670 feet in a hallway. Orthotics and assistive devices were utilized as per the child's normal routine. Number of steps and distance were collected from both trackers. A researcher simultaneously used a tally counter to manually count total number of steps. Pearson correlation coefficients were determined for number of steps from each tracker and the manual count. Mean absolute percent error (MAPE) was calculated for steps for each tracker and the manual count, and for the distance from each tracker and the hallway distance. Results: For number of steps, a strong positive correlation was found between the waist tracker and the manual count (r=0.997), whereas a weak positive correlation was found between wrist tracker and manual count (r=0.223). MAPE for steps was 1% for the waist tracker and 12% for the wrist tracker. For distance, MAPE was 56% for the waist tracker and 41% for the wrist tracker. Conclusions/Significance: The waist-based activity tracker provided an accurate step count. Neither waist-based nor wristbased tracker was accurate for distance measurement. Thus, the walking ability of ambulatory children with CP can be accurately quantified with a readily available inexpensive activity tracker. This has the potential to enable clinicians to assess the effects of various treatments on the real-world activity level of patients with CP

Foot and ankle deformities in cerebral palsy can be effectively treated with surgery. Surgery should be considered in patients with significant deformity and those who have pain or difficulty with orthotic and shoe wear. Equinus contracture of both gastrocnemius and soleus can be treated with open tendoachilles lengthening; ankle valgus with medial epiphysiodesis. Equinovarus is more commonly seen in hemiplegic patients and this deformity can usually be treated with tendon transfers. Triple arthrodesis is an option in children with severe degenerative changes. It is important to address all aspects of the child's pathology at the time of surgical correction.

BACKGROUND: The purpose of this study was to evaluate the necessity of early postoperative radiographs after pinning of supracondylar humerus fractures by determining both the percentage of patients who displayed change in fracture fixation and whether these changes affected their outcome. METHODS: A series of 643 consecutive patients who underwent operative management of Gartland type II and III fractures at our institution between January 2002 and December 2010 were reviewed. Demographic data were obtained through chart review, including age, sex, extremity, fracture type, and mechanism. Intraoperative fluoroscopic images were compared with postoperative radiographs to identify changes in fracture alignment and pin placement. RESULTS: A total of 643 patients (320 females, 323 males) with a mean age of 6.1 years (range, 1.1 to 16.0) were reviewed. Fifty-seven percent of fractures were classified as type II and 43% were type III. The overall complication rate was 8.8% (57/643). Pin backout or fracture translation was seen in 32 patients (4.9%) at the first postoperative visit. All of these patients sustained type III fractures. One of these patients required further operative management. Patients with changes in pin or fracture alignment did not demonstrate a statistically significant difference in time to first postoperative visit (P=0.23), days to pin removal (P=0.07), or average follow-up time (P=0.10). Fracture severity did not correlate with change in alignment (P=0.952). No postoperative neurological complications were observed in patients with alignment changes. CONCLUSIONS: Mild alignment changes and pin migration observed in postoperative radiographs after pinning of supracondylar humerus fractures have little effect on clinical management parameters or long-term sequelae. Radiographs can therefore be deferred until the time of pin removal provided adequate intraoperative stability was obtained. LEVEL OF EVIDENCE: Level IV.