Propensity scores as a novel method to guide sample allocation and minimize batch effects during the design of high throughput experiments.

BACKGROUND: We developed a novel approach to minimize batch effects when assigning samples to batches. Our algorithm selects a batch allocation, among all possible ways of assigning samples to batches, that minimizes differences in average propensity score between batches. This strategy was compared to randomization and stratified randomization in a case-control study (30 per group) with a covariate (case vs control, represented as ß1, set to be null) and two biologically relevant confounding variables (age, represented as ß2, and hemoglobin A1c (HbA1c), represented as ß3). Gene expression values were obtained from a publicly available dataset of expression data obtained from pancreas islet cells. Batch effects were simulated as twice the median biological variation across the gene expression dataset and were added to the publicly available dataset to simulate a batch effect condition. Bias was calculated as the absolute difference between observed betas under the batch allocation strategies and the true beta (no batch effects). Bias was also evaluated after adjustment for batch effects using ComBat as well as a linear regression model. In order to understand performance of our optimal allocation strategy under the alternative hypothesis, we also evaluated bias at a single gene associated with both age and HbA1c levels in the 'true' dataset (CAPN13 gene). RESULTS: Pre-batch correction, under the null hypothesis (ß1), maximum absolute bias and root mean square (RMS) of maximum absolute bias, were minimized using the optimal allocation strategy. Under the alternative hypothesis (ß2 and ß3 for the CAPN13 gene), maximum absolute bias and RMS of maximum absolute bias were also consistently lower using the optimal allocation strategy. ComBat and the regression batch adjustment methods performed well as the bias estimates moved towards the true values in all conditions under both the null and alternative hypotheses. Although the differences between methods were less pronounced following batch correction, estimates of bias (average and RMS) were consistently lower using the optimal allocation strategy under both the null and alternative hypotheses. CONCLUSIONS: Our algorithm provides an extremely flexible and effective method for assigning samples to batches by exploiting knowledge of covariates prior to sample allocation.

DNA Methylation Near DLGAP2 May Mediate the Relationship between Family History of Type 1 Diabetes and Type 1 Diabetes Risk.

Given the differential risk of type 1 diabetes (T1D) in offspring of affected fathers versus affected mothers and our observation that T1D cases have differential DNA methylation near the imprinted DLGAP2 gene compared to controls, we examined whether methylation near DLGAP2 mediates the association between T1D family history and T1D risk. In a nested case-control study of 87 T1D cases and 87 controls from the Diabetes Autoimmunity Study in the Young, we conducted causal mediation analyses at 12 DLGAP2 region CpGs to decompose the effect of family history on T1D risk into indirect and direct effects. These effects were estimated from two regression models adjusted for the human leukocyte antigen DR3/4 genotype: a linear regression of family history on methylation (mediator model) and a logistic regression of family history and methylation on T1D (outcome model). For 8 of the 12 CpGs, we identified a significant interaction between T1D family history and methylation on T1D risk. Accounting for this interaction, we found that the increased risk of T1D for children with affected mothers compared to those with no family history was mediated through differences in methylation at two CpGs (cg27351978, cg00565786) in the DLGAP2 region, as demonstrated by a significant pure natural indirect effect (odds ratio (OR) = 1.98, 95% confidence interval (CI): 1.06-3.71) and nonsignificant total natural direct effect (OR = 1.65, 95% CI: 0.16-16.62) (for cg00565786). In contrast, the increased risk of T1D for children with an affected father or sibling was not explained by DNA methylation changes at these CpGs. Results were similar for cg27351978 and robust in sensitivity analyses. Lastly, we found that DNA methylation in the DLGAP2 region was associated (P<0:05) with gene expression of nearby protein-coding genes DLGAP2, ARHGEF10, ZNF596, and ERICH1. Results indicate that the maternal protective effect conferred through exposure to T1D in utero may operate through changes to DNA methylation that have functional downstream consequences.

Prediction of Late Dysplasia Based on Ultrasound and Plain X-Ray at 6 Months.

BACKGROUND: Developmental dysplasia of the hip represents a spectrum of deformity. Residual dysplasia at 2 years of age is associated with an increased risk for osteoarthritis and functional limitations. We compared the prognostic value of 6-month imaging modalities and aimed to identify optimal diagnostic metrics for the prediction of residual dysplasia. METHODS: After IRB approval, patients who underwent Pavlik treatment between 2009 and 2018 with 2-year follow-up were identified. Sonographs [ultrasound (US)] and radiographs (x-ray) were obtained at 6-month and 2-year-old visits. Dysplasia at 2 years was defined as an acetabular index (AI) >24 degrees. Receiver operating characteristic curves were constructed to quantitatively compare the prognostic ability of US and x-ray-based measures at 6 months. Youden's index [(YI) (values range from 0 (poor test) to 1 (perfect test)] was used to evaluate existing cutoffs at 6 months of age (normal measurements: alpha angle (AA) ≥60 degrees, femoral head coverage (FHC) ≥50%, and AI <30 degrees) relative to newly proposed limits. RESULTS: Fifty-nine patients were included, of which 28.8% of patients (95% CI: 17.3 to 40.4%) had acetabular dysplasia at 2 years. After adjusting for sex, AA [Area under the Curve (AUC): 80] and AI (AUC: 79) at 6 months of age were better tests than FHC (AUC: 0.77). Current diagnostic cutoffs for AA (YI: 0.08), AI (YI: 0.0), and FHC (YI: 0.06) at 6 months had poor ability to predict dysplasia at 2 years. A composite test of all measures based on proposed cutoffs (AA ≥73 degrees, FHC > 62% and AI ≤24 degrees) was a better predictor of dysplasia at 2 years (Youden's index (YI): 0.63) than any single metric. CONCLUSIONS: The rate of residual dysplasia remains concerning. The 6-month x-ray and US both play a role in the ongoing management of the developmental dysplasia of the hip. The prediction of dysplasia is maximized when all metrics are considered collectively. Existing parameters were not accurate; We recommend the following cutoffs: AA ≥73 degrees, FHC > 62%, and AI ≤24 degrees. These cutoffs must be validated. LEVEL OF EVIDENCE: Prognostic Level II.

Application of Accepted Use Criteria for the Treatment of Developmental Dysplasia of the Hip Decreases the Number of Infants Treated With a Pavlik Harness.

BACKGROUND: This analysis examined how the application of the American Academy of Orthopedic Surgeons appropriate use criteria (AUC) for developmental dysplasia of the hip in infants would change treatment patterns and outcomes for Graf IIA hips at a single quaternary pediatric hospital. METHODS: After Institutional Review Board approval, patient medical records were reviewed and data were collected. Graf IIa hips were defined as alpha angle (AA) 50 to 59 degrees. AA and femoral head coverage (FHC) were measured from initial and 6-month ultrasounds and acetabular index (AI) was measured from radiographs at 6 months of age. Instability (positive Ortolani and Barlow tests) was noted. On the basis of the American Academy of Orthopedic Surgeons AUC for managing developmental dysplasia of the hip, hips were further categorized as normal (FHC ≥45%), borderline (FHC 35% to 44%), or dysplastic (FHC <35%). RESULTS: Overall, 13% (49/371) of Graf IIa hips (AA 50 to 59 degrees) were dysplastic (FHC <35%). Total 24% (89/371) were clinically unstable. Total 42% (37/89) of unstable Graf IIa hips were dysplastic. Only 4% of stable Graf IIa hips were dysplastic (12/282). Out of 371 Graf IIa hips, 256 were treated with Pavlik harness (n=250) or Rhino brace (n=6). Among stable, nondysplastic (SND) hips (those with normal and borderline FHC≥35%), 33% (52/158) were treated because of a more severe contralateral side. If the AUC had been applied, 67% (106/158) of SND Graf IIa hips would not have been treated. Among the n=162 hips that returned for a 6-month radiograph, there was no difference in AI in the 115 treated and 47 untreated SND hips (mean difference treatment vs. no treatment: -1.5, 95% CI, -3.1 to 0.2, P =0.0808). CONCLUSIONS: Using AUC recommendations, our center could reduce the number of SND Graf IIa hips we treat by 67%. Although 24% of Graf IIa hips were clinically unstable and 13% were dysplastic based on FHC, most Graf IIa hips had normal or borderline FHC per the AUC and may do well with observation and follow-up ultrasound at 12 weeks old. LEVEL OF EVIDENCE: Level III-diagnostic study.

Inverse probability weighting is an effective method to address selection bias during the analysis of high dimensional data.

Omics studies frequently use samples collected during cohort studies. Conditioning on sample availability can cause selection bias if sample availability is nonrandom. Inverse probability weighting (IPW) is purported to reduce this bias. We evaluated IPW in an epigenome-wide analysis testing the association between DNA methylation (261,435 probes) and age in healthy adolescent subjects (n = 114). We simulated age and sex to be correlated with sample selection and then evaluated four conditions: complete population/no selection bias (all subjects), naïve selection bias (no adjustment), and IPW selection bias (selection bias with IPW adjustment). Assuming the complete population condition represented the "truth," we compared each condition to the complete population condition. Bias or difference in associations between age and methylation was reduced in the IPW condition versus the naïve condition. However, genomic inflation and type 1 error were higher in the IPW condition relative to the naïve condition. Postadjustment using bacon, type 1 error and inflation were similar across all conditions. Power was higher under the IPW condition compared with the naïve condition before and after inflation adjustment. IPW methods can reduce bias in genome-wide analyses. Genomic inflation is a potential concern that can be minimized using methods that adjust for inflation.

How Do We Best Educate Our Patients' Caregivers? Comparing the Efficacy of Print Versus Media-based Education Materials in Peripheral Nerve Catheter and Pain Pump Education.

BACKGROUND: Media-based educational materials (EMs) are becoming prominent. The purpose of this study was to compare print versus media-based EMs given to caregivers of pediatric and young adult patients undergoing surgery. We aimed to see whether print or media-based EMs lead to greater caregiver satisfaction, comfort, and preparedness for outpatient peripheral nerve catheter and pain pump management. We also assessed caregiver preference for EM modality. HYPOTHESIS: We hypothesized that media-based EMs would demonstrate greater overall efficacy and thus generate higher caregiver preference. STUDY DESIGN: Randomized control trial. METHODS: After IRB approval, clinicaltrials.gov registration (17-0638), and informed consent, caregivers were randomized to either media or print-based EM groups. Caregivers reviewed their assigned EM and completed a standardized assessment of their comprehension. We assessed caregiver satisfaction, preparedness, and comfort level with the content on a 5-point Likert scale. On postoperative days 1 to 2, caregivers reported satisfaction, comfort, and preference for EM modality. An intent-to-treat analysis was used to compare the 2 groups. RESULTS: From our final cohort of 135 caregivers, we found no difference [P>0.05] in satisfaction, comfort level, level of preparedness, or discharge readiness scores between groups. After the caregivers were given both EMs, they were evenly split in their preference for print (49.6%) versus video (50.4%) based methods. CONCLUSIONS: We did not detect a significant difference in caregiver preference or feelings of preparedness between groups. Interestingly, a significant proportion of caregivers (25%) did not feel comfortable managing the peripheral nerve catheter and its pain pump at home. Future studies should work to improve caregiver comfort with educational content before patient discharge. CLINICAL RELEVANCE: Providers and institutions should feel comfortable providing both print and media-based patient and caregiver education. Caregiver education may be best suited based on caregiver preference of one EM modality versus the other. LEVEL OF EVIDENCE: Level I.

The oxylipin profile is associated with development of type 1 diabetes: the Diabetes Autoimmunity Study in the Young (DAISY).

AIMS/HYPOTHESIS: Oxylipins are lipid mediators derived from polyunsaturated fatty acids. Some oxylipins are proinflammatory (e.g. those derived from arachidonic acid [ARA]), others are pro-resolving of inflammation (e.g. those derived from α-linolenic acid [ALA], docosahexaenoic acid [DHA] and eicosapentaenoic acid [EPA]) and others may be both (e.g. those derived from linoleic acid [LA]). The goal of this study was to examine whether oxylipins are associated with incident type 1 diabetes. METHODS: We conducted a nested case-control analysis in the Diabetes Autoimmunity Study in the Young (DAISY), a prospective cohort study of children at risk of type 1 diabetes. Plasma levels of 14 ARA-derived oxylipins, ten LA-derived oxylipins, six ALA-derived oxylipins, four DHA-derived oxylipins and two EPA-related oxylipins were measured by ultra-HPLC-MS/MS at multiple timepoints related to autoantibody seroconversion in 72 type 1 diabetes cases and 71 control participants, which were frequency matched on age at autoantibody seroconversion (of the case), ethnicity and sample availability. Linear mixed models were used to obtain an age-adjusted mean of each oxylipin prior to type 1 diabetes. Age-adjusted mean oxylipins were tested for association with type 1 diabetes using logistic regression, adjusting for the high risk HLA genotype HLA-DR3/4,DQB1*0302. We also performed principal component analysis of the oxylipins and tested principal components (PCs) for association with type 1 diabetes. Finally, to investigate potential critical timepoints, we examined the association of oxylipins measured before and after autoantibody seroconversion (of the cases) using PCs of the oxylipins at those visits. RESULTS: The ARA-related oxylipin 5-HETE was associated with increased type 1 diabetes risk. Five LA-related oxylipins, two ALA-related oxylipins and one DHA-related oxylipin were associated with decreased type 1 diabetes risk. A profile of elevated LA- and ALA-related oxylipins (PC1) was associated with decreased type 1 diabetes risk (OR 0.61; 95% CI 0.40, 0.94). A profile of elevated ARA-related oxylipins (PC2) was associated with increased diabetes risk (OR 1.53; 95% CI 1.03, 2.29). A critical timepoint analysis showed type 1 diabetes was associated with a high ARA-related oxylipin profile at post-autoantibody-seroconversion but not pre-seroconversion. CONCLUSIONS/INTERPRETATION: The protective association of higher LA- and ALA-related oxylipins demonstrates the importance of both inflammation promotion and resolution in type 1 diabetes. Proinflammatory ARA-related oxylipins may play an important role once the autoimmune process has begun.

Predictors of oxylipins in a healthy pediatric population.

BACKGROUND: Oxylipins are formed from oxidation of omega-6 (n6) and omega-3 (n3) fatty acids (FAs). Evidence for inflammatory effects comes mostly from adults. METHODS: Oxylipins from n6 FA (27 n6-oxylipins) and n3 FA (12 n3-oxylipins) were measured through ultra-high-performance liquid chromatography-mass spectrometry (LC-MS/MS) in plasma from 111 children at risk of type 1 diabetes (age 1-17 years) studied longitudinally. Oxylipin precursor FAs (arachidonic acid, linoleic acid, alpha-linolenic acid, docosahexaenoic acid, eicosapentaenoic acid) were measured in red blood cell (RBC) membrane and plasma. Precursor FAs dietary intake was measured through food frequency questionnaire and environmental tobacco smoke (ETS) through questionnaires. Linear mixed models were used to test oxylipins with predictors. RESULTS: Age associated with 15 n6- and 6 n3-oxylipins; race/ethnicity associated with 3 n6- and 1 n3-oxylipins; sex associated with 2 n6-oxylipins. ETS associated with lipoxin-A4. Oxylipins associated with precursor FAs in plasma more often than RBC. RBC levels and dietary intake of precursor FAs more consistently associated with n3-oxylipins than with n6-oxylipins. CONCLUSIONS: In healthy children, oxylipin levels change with age. Oxylipins associated with precursor FAs more often in plasma than RBC or diet, suggesting that inflammatory regulation leading to FA release into plasma may also be a determinant of oxylipin generation. IMPACT: This is the first study to examine predictors of oxylipins in healthy children at risk of type 1 diabetes. In healthy children at risk of type 1 diabetes, many oxylipins change with age, and most oxylipins do not differ by sex or race/ethnicity. Environmental tobacco smoke exposure was associated with the presence of lipoxin A4. Omega-6- and omega-3-related oxylipin levels were consistently associated with their respective precursor fatty acid levels measured in the plasma. Proportionally more omega-3 compared to omega-6 oxylipins were associated with dietary intake and red blood cell membrane levels of the respective precursor fatty acid.

DNA methylation near the INS gene is associated with INS genetic variation (rs689) and type 1 diabetes in the Diabetes Autoimmunity Study in the Young.

OBJECTIVE: Mechanisms underlying the role of non-human leukocyte antigen (HLA) genetic risk variants in type 1 diabetes (T1D) are poorly understood. We aimed to test the association between methylation and non-HLA genetic risk. METHODS: We conducted a methylation quantitative trait loci (mQTL) analysis in a nested case-control study from the Dietary Autoimmunity Study in the Young. Controls (n = 83) were frequency-matched to T1D cases (n = 83) based on age, race/ethnicity, and sample availability. We evaluated 13 non-HLA genetic markers known be associated with T1D. Genome-wide methylation profiling was performed on peripheral blood samples collected prior to T1D using the Illumina 450 K (discovery set) and infinium methylation EPIC beadchip (EPIC validation) platforms. Linear regression models, adjusting for age and sex, were used to test to each single nucleotide polymorphism (SNP) -probe combination. Logistic regression models were used to test the association between T1D and methylation levels among probes with a significant mQTL. A meta-analysis was used to combine odds ratios from the two platforms. RESULTS: We identified 10 SNP-methylation probe pairs (false discovery rate (FDR) adjusted P < .05 and validation P < .05). Probes were associated with the GSDMB, C1QTNF6, IL27, and INS genes. The cg03366382 (OR: 1.9, meta-P = .0495), cg21574853 (OR: 2.5, meta-P = .0232), and cg25336198 (odds ratio: 6.6, meta-P = .0081) probes were significantly associated with T1D. The three probes were located upstream from the INS transcription start site. CONCLUSIONS: We confirmed an association between DNA methylation and rs689 that has been identified in related studies. Measurements in our study preceded the onset of T1D suggesting methylation may have a role in the relationship between INS variation and T1D development.

Dimeglio Score Predicts Treatment Difficulty During Ponseti Casting for Isolated Clubfoot.

BACKGROUND: The Dimeglio score (DS) is widely used to assess clubfoot severity, but its ability to predict long-term outcomes following Ponseti treated isolated clubfoot (IC) is controversial. This study tested the association between the initial DS and its individual parameters with the number of Ponseti clubfoot casts required to achieve correction and the rate of early recurrence following treatment. METHODS: Data were retrospectively collected from patients who underwent treatment of IC between March 2012 and March 2015 and were followed for ≥2 years. DSs were collected at the initial casting visit. The number of Ponseti casts required to achieve clubfoot correction before tenotomy and recurrence of deformity were collected as the primary outcome variables. Recurrence was defined as any loss of correction leading to repeat casting or tenotomy during the bracing phase. Negative binomial and logistic regression analyses were used to test the association between the 8 Dimeglio parameters and number of casts and incidence of recurrence, respectively. RESULTS: A total of 53 patients (37 male and 16 female) were included in the study. The median number of casts required to achieve an acceptable correction was 5 (range, 2 to 16). The incidence of recurrence was 24.53% (13/53). An increase in derotation, varus, equinus, muscle condition, and total DSs at the initial cast visit were associated with a significant (P<0.05) increase in the number of casts required to achieve an acceptable correction. The derotation parameter [rate ratio: 1.30, 95% confidence interval (CI): 1.13-1.50, P=0.0003] was most strongly associated with number of casts. Total DSs at initial visit was the only variable significantly associated with the incidence of deformity recurrence (odds ratio: 1.36, 95% confidence interval: 1.01-1.84, P=0.0482). CONCLUSION: Initial DS is correlated with the number of casts required for correction in Ponseti treated IC. DS may help physicians establish realistic expectations for families with regard to the length of treatment and the possibility of recurrence following Ponseti treatment. LEVEL OF EVIDENCE: Level II-retrospective prognostic study.

Fassier-Duval Rods are Associated With Superior Probability of Survival Compared With Static Implants in a Cohort of Children With Osteogenesis Imperfecta Deformities.

BACKGROUND: The survival of Fassier-Duval (FD) telescoping rods as compared with static implants in children affected by osteogenesis imperfecta is not well characterized. The purpose of this study was to compare risk of lower extremity implant failure in FD rods versus static implants. METHODS: Data were retrospectively collected from patients with osteogenesis imperfecta who underwent surgical treatment using either FD rods or static implants (Rush rods, flexible nails, or Steinmann pins) between 1995 and 2015. The timing of implant failure was the primary outcome variable of interest. Comparisons were limited to limbs with no previous history of implants. Cox-proportional hazards regression analyses were used to compare the hazard of implant failure across implants. Negative binomial regression analyses were used to compare the incidence of surgical procedures in the 2 implant groups. RESULTS: The final cohort consisted of 64 limbs (n=21 patients). The static implant group (n=38) consisted of 24 Rush rods (63%), 14 flexible nails (37%), and 2 Steinmann pins (5%). The hazard of implant failure in the static implant group was 13.2 times [95% confidence interval (CI), 2.5-69.6; P=0.0024] the hazard of implant failure in the FD rod group. The hazard of implant failure among females was 4.8 (95% CI, 1.4-16.7; P=0.0125) times the hazard of implant failure among males. The total surgery rate in the static implant group was 7.8 (95% CI, 1.8-33.0; P=0.0056) times the total surgery rate in the FD group. CONCLUSIONS: Among surgically naive limbs, FD rods were associated with significantly improved probability of survival compared with static implants. LEVEL OF EVIDENCE: Level II-retrospective study.

The modified Dunn procedure provides superior short-term outcomes in the treatment of the unstable slipped capital femoral epiphysis as compared to the inadvertent closed reduction and percutaneous pinning: a comparative clinical study.

PURPOSE: The aim of this study was to compare clinical outcomes and radiographic correction after modified Dunn procedure versus inadvertent closed reduction and percutaneous pinning for the treatment of unstable slipped capital femoral epiphysis (SCFE). METHODS: We evaluated 45 patients with unstable SCFE treated using the modified Dunn procedure (n = 27) or percutaneous pinning (n = 18) during a minimum follow-up of one year. Clinical outcomes were assessed using the Heyman and Herndon scores. The Southwick angle, alpha angle, and femoral head-neck offset were used to assess radiographic correction. The occurrence of complications and unplanned re-operations were recorded. RESULTS: At latest follow-up, 67% (18/27) in the modified Dunn procedure group and 28% (5/18) in the in situ pinning group had good or excellent Heyman and Herndon outcomes (p = 0.016). The morphology of the femoral head and neck was improved in the modified Dunn procedure group compared to percutaneous pinning (Southwick angle, alpha angles; femoral head-neck offset; p < 0.001). The proportion of osteonecrosis (26 vs. 28%; p > 0.999) and unplanned re-operations (26 vs. 33%; p = 0.894) was similar in both groups. CONCLUSION: Compared to inadvertent reduction and percutaneous pinning, the modified Dunn procedure provided better clinical and radiographic outcomes with similar proportion of osteonecrosis and unplanned re-operations following an unstable SCFE.

Prevalence of Persistent Median Arteries in the Pediatric Population on Ultrasonography.

OBJECTIVES: To use real-time ultrasonography to estimate the prevalence of persistent median arteries in a cohort of pediatric orthopedic patients. METHODS: With Institutional Review Board approval, patients between the ages of 3 months and 19 years were recruited for this cross-sectional study. Variables of interest included demographics, underlying diagnoses, and the presence of a Doppler-confirmed median artery. Ultrasonographic examinations were performed on both upper extremities by a single investigator. Patient- and limb-level analyses were performed. A multivariable generalized logistic regression analysis was used to test the association between the prevalence of limb-specific persistent median arteries and age. A generalized estimating equation was used to adjust for the inclusion of multiple limbs from the same patient. RESULTS: A total of 135 patients (270 limbs) were evaluated. The patient-specific prevalence rate (persistent median arteries present in 1 or both limbs) was 26.7%. Among these patients (n = 36), a persistent median artery was present bilaterally in 55.6% (n = 20). The limb-specific prevalence rate (proportion of limbs with a persistent median artery) was 20.7%. After adjusting for race, for every 1-year increase in age, the odds of a persistent median artery decreased by 4.4%. After adjusting for age, African Americans were significantly more likely to present with a persistent median artery (odds ratio, 3.78; 95% confidence interval, 1.25-11.48). CONCLUSIONS: Ultrasonography can effectively visualize anatomic variants, such as persistent median arteries, in the pediatric population. The prevalence of persistent median arteries was higher than anticipated, especially among African American patients.

The Prevalence and Treatment of Hip Dysplasia in Prader-Willi Syndrome (PWS).

BACKGROUND: Prader-Willi syndrome (PWS) is a genetic disorder with multisystem involvement. There are a number of associated orthopaedic manifestations, the most recognized of which is scoliosis. The aim of this study was to assess the prevalence of hip dysplasia and to investigate its treatment in patients with PWS. METHODS: Following IRB approval, all patients seen at our institution's Prader-Willi multidisciplinary clinic were retrospectively reviewed. Only patients with an ultrasound, anteroposterior (AP) spine, AP abdomen, AP hip radiograph, and/or skeletal survey were included in the study. The presence of hip dysplasia was determined based on ultrasonographic and/or radiographic measurements performed by a single fellowship trained pediatric orthopaedic surgeon. A multivariable logistic regression analysis was used to test the association between patient demographics and the prevalence of hip dysplasia. Age at diagnosis, treatment type, and outcomes were recorded for patients that underwent treatment for hip dysplasia. RESULTS: Hip dysplasia was identified in 30% (27/90) of the patient population. Two of the 27 patients (7.4%) had normal films but had a history of resolved hip dysplasia. Prevalence was not associated with sex (P=0.7072), genetic subtype (P=0.5504), race (P=0.8537), ethnicity (P=0.2191), or duration of follow-up (P=0.4421). Eight of the 27 patients (30%) underwent hip treatment by Pavlik harness (2/8), Pavlik harness and closed reduction (1/8), closed reduction (3/8), open reduction (1/8), and unspecified hip surgery (1/8). The mean age at diagnosis was 2 months for the patients that were successfully treated for hip dysplasia (3/8) and 12 months for those who had residual dysplasia following the treatment (5/8). CONCLUSIONS: Our study demonstrates a higher prevalence of hip dysplasia in patients with PWS than previously documented. The age at which hip dysplasia develops remains unknown; therefore, we recommend an ultrasound screening for all infants with PWS at 6 weeks of age and subsequent radiographic studies at 1, 2, 5, 10, and 15 years of age to allow for early diagnosis and intervention. LEVEL OF EVIDENCE: Level III-retrospective comparative study.

Graf Type-IV Hips Have a Higher Risk of Residual Acetabular Dysplasia at 1 Year of Age Following Successful Pavlik Harness Treatment for Developmental Hip Dysplasia.

BACKGROUND: We asked whether patient-specific factors and ultrasound (US) measurements of hip dysplasia severity at initial examination influence short-term residual acetabular dysplasia in patients successfully treated with Pavlik harness for developmental hip dysplasia. METHODS: After IRB approval, 134 hips (84 patients) successfully treated by the Pavlik method between August 2011 and October 2014 with follow-up at 12 months of age were identified. Early successful treatment was defined as normal examination and US after approximately 12 weeks of Pavlik treatment. Multivariate linear and logistic regression models were used to identify factors associated with acetabular index (AI) measurements at 12 months as well as factors associated with an increase in AI between the 6- and 12-month timepoints (dysplastic progression). RESULTS: The study consisted of 134 hips (84 patients). The distribution of dysplastic, Barlow, and Ortolani hips was 44.8% (N=60), 30.6% (N=41), and 24.6% (N=33), respectively. The crude incidence of residual dysplasia at the 6-month visit (AI>30 degrees) was 11.7% (12/102). The incidence of residual dysplasia at the 12-month visit (AI>28 degrees) was 11.8% (13/111). Graf type was the only variable associated with AI values at the 12-month visit (mean difference: Graf type-IV-Other, 2.6; 95% confidence interval, 0.3-4.9; P=0.026). CONCLUSIONS: The risk of residual acetabular dysplasia after normal hip US following Pavlik treatment is not negligible. Radiographic surveillance is warranted to monitor and screen for dysplasia progression. Patients with dislocated Graf type-IV hips at diagnosis are at increased risk of residual acetabular dysplasia at 1 year after successful treatment with the Pavlik method. LEVEL OF EVIDENCE: Level III-therapeutic study.

Local Infiltration Analgesia Compared With Epidural and Intravenous PCA After Surgical Hip Dislocation for the Treatment of Femoroacetabular Impingement in Adolescents.

BACKGROUND: Open treatment of femoroacetabular impingement (FAI) through a surgical hip dislocation (SHD) approach has been reported to allow for improvement in pain and function. However, the approach require a trochanteric osteotomy and may be associated with high level of pain after surgery. Currently, there is no systematic approach for pain management after SHD for treatment of FAI. METHODS: A retrospective chart review was used to collect data from 121 subjects (12 to 21 y and below) who received periarticular local infiltration analgesia (LIA, n=20), epidural analgesia (n=72), or intravenous patient-controlled analgesia (PCA, n=29) after SHD from January 2003 to June 2014. Verbal pain scores, opioid consumption, incidence of side effects/complications, and length of hospital stay (LOS) were recorded. All nonopioid medications with analgesic potential were included in the statistical models as potential confounding variables RESULTS:: Twelve hours after surgery, the odds of moderate/severe pain were higher in the PCA group (odds ratio, 20.5; 95% confidence interval (CI), 1.7-243.8; P=0.0166] and epidural group (odds ratio, 5.2; 95% CI, 0.7-92.0; P=0.3218) compared with the LIA group. There was no difference in pain scores across all groups 1 hour (P=0.0675) or 24 hours (P=0.3473) postoperatively. Total opioid consumption in the LIA group was 59.8% (95% CI, 15.0%-81.0%; P=0.0175) lower than the total opioid consumption in the epidural group and 60.7% (95% CI, 17.3-81.3; P=0.0144) lower than the total opioid consumption in the PCA group. LOS was increased in the epidural (mean difference, 22.1; 95% CI, 6.8-37.4 h; P=0.0051) and PCA (mean difference, 16 h; 95% CI, 1-31.5 h; P=0.0367) groups relative to the LIA group. There was 0 (0%) complication in the LIA group compared with 11 (15.3%) in the epidural group. CONCLUSIONS: LIA was more effective at controlling pain 12 hours after surgery in comparison with PCA with similar pain control to epidural. LIA was associated with significantly lower need for opioids and shorter LOS compared with the PCA and epidural protocols. Periarticular infiltration should be considered for pain management after SHD for treatment of FAI in adolescents. LEVEL OF EVIDENCE: Level III-retrospective comparative study.

Clinical and Radiographic Outcomes After Submuscular Plating (SMP) of Pediatric Femoral Shaft Fractures.

BACKGROUND: To evaluate lower extremity strength and alignment among children who underwent submuscular plating (SMP). METHODS: Subjects who underwent SMP for a length unstable femoral fracture returned for isokinetic strength testing ≥2 years after surgery. Extensor and flexor strength deficits (percent difference) between the operative and nonoperative limbs were evaluated. Radiographic measurements of mechanical alignment [anatomic lateral distal femoral angle (aLDFA)] and qualitative measurements (The Pediatric Outcomes Data Collection Instrument, PODCI) were obtained from all subjects. The upper 97.5% confidence interval approach to establishing clinical equivalency was utilized to compare differences in strength and alignment between the operative and nonoperative limbs. An extensor strength deficit of >19% and an aLDFA discrepancy of >5 degrees were considered to be clinically significant. RESULTS: The average age at surgery of the 10 subjects included in the study was 8.7 years. The hardware was placed an average of 27.9 mm from the distal femoral physis and was removed 6.4 months postsurgery. Among all subjects, the median PODCI scores were ≥97 according to all subscales. There was no significant difference in extension torque between the operative versus nonoperative limbs at 60 degrees/s (P=0.5400), 120 degrees/s (P=0.4214), or 180 degrees/s (P=0.8166). More importantly, extension strength deficits between the operative and nonoperative limbs were not clinically significant at 60 degrees/s [upper 97.5% confidence interval (CI), 10.9%], 120 degrees/s (upper 97.5% CI, 11.0%), or 180 degrees/s (upper 97.5% CI, 10.7%). The difference in aLDFA between the operative and nonoperative limb was less than the predefined clinically significant threshold of 5 degrees for all subjects. CONCLUSIONS: SMP achieves satisfactory clinical and functional results. In this series, extensor strength deficits and/or lower extremity malalignment were not clinically meaningful. High patient satisfaction can be expected after implant removal. LEVEL OF EVIDENCE: Level IV-case series.

Does Surgeon Experience Impact the Risk of Complications After Bernese Periacetabular Osteotomy?

BACKGROUND: Bernese periacetabular osteotomy (PAO) is a technically challenging procedure with potential risk for major complications and a previously reported steep learning curve. However, the impact of contemporary hip preservation fellowships on the learning curve of PAO has not been established. QUESTIONS/PURPOSES: (1) What was the frequency of major complications during the PAO learning curve of two surgeons who recently graduated from hip preservation fellowships? (2) Is increasing level of experience associated with the risk of a complication and with operative time, a possible surrogate measure of surgical performance? METHODS: We retrospectively studied 81 PAOs performed by one of two surgeons who recently graduated from a hip preservation fellowship during their first 4 years of practice in two institutions. One of the surgeons participated as a fellow in 78 PAOs with an increasing level of responsibility during the course of 1 full year. The other surgeon performed 41 PAOs as a fellow during 6 months, also with an increasing level of responsibility during that time. There were 68 (84%) female and 13 (16%) male patients (mean age, 18 years; range, 10-36 years). The frequency of complications was recorded early and at 1 year after surgery and graded according to a validated classification system describing five grades of complications. Complications that required surgical intervention (Grade III) and life-threatening complications (Grade IV) were considered major complications. Persistent pain after surgery, although considered a failure of PAO, was not considered a surgical complication as a result of the multifactorial etiology of pain after hip-preserving surgery. However, early reoperation and revision surgery were counted as complications. To evaluate the association between increasing level of experience and the occurrence of complications, we divided each surgeon's experience into his first 20 procedures (initial interval) and his second 20 (experienced interval) to test whether the incidence of complications or operative time was different between the two intervals. Because the association between experience and the likelihood of a complication was estimated to be consistent between the two surgeons, the analysis was performed with data pooled from the two surgeons. To test whether there was a difference in the likelihood of a complication in the initial and the experienced intervals, a multivariate logistic regression analysis was performed and the adjusted risk of a complication between the two intervals was calculated. Linear regression analyses were used to test the association between surgeon level of experience and operative time. RESULTS: The overall incidence of major (Grade III or Grade IV) complications was 6% (95% confidence interval [CI], 2%-14%). These included deep infection (3% [three of 81]), intraoperative posterior column fracture (1% [one of 81]), and pulmonary embolism (1% [one of 81]). With the numbers available, the risk of a complication did not decrease with increasing surgeon experience. After controlling for body mass index and surgeon, the frequency of a complication did not decrease in the experienced interval relative to the initial interval (odds ratio, 0.78; 95% CI, 0.25-2.4; p = 0.6623). The adjusted risk difference between the experienced interval relative and the initial interval was 6% (95% CI, -11% to 23%). When experience was modeled as a continuous variable (number of PAOs performed), increasing experience was not associated with a lower likelihood of a complication (odds ratio per one PAO increase in experience, 0.99; 95% CI, 0.94-1.04; p = 0.5478). However, after adjusting for body mass index and surgeon, increased experience was associated with a reduction in operative time (slope [change in log operative time per one procedure increase in experience], -0.005; 95% CI, -0.009 to -0.0005; p = 0.0292). For every one additional PAO increase in experience, there was a 0.45% decrease in operative time (95% CI, 0.05%-0.86% decrease]. CONCLUSIONS: With a case exposure greater than 40 PAOs and progressive surgical responsibility during contemporary structured training, two young surgeons were able to perform PAO with a low risk of complications. However, even with that surgical experience before independent practice, surgical time decreased over the first 40 PAOs they performed independently. Our data may help guide orthopaedic residency and hip preservation fellowship programs in establishing training requirements and assessing competency in PAO. LEVEL OF EVIDENCE: Level III, therapeutic study.

Measurement of Femoral Version by MRI is as Reliable and Reproducible as CT in Children and Adolescents With Hip Disorders.

BACKGROUND: Femoral version measurement techniques based on magnetic resonance imaging (MRI) studies have been developed as an alternative to the high levels of ionizing radiation associated with computed tomography (CT)-based methods. Previous studies have not evaluated the reliability, repeatability, and accuracy of MRI-based femoral version measurements in an adolescent population. METHODS: Subjects who underwent MRI and CT studies for clinical suspicion of hip pain secondary to hip dysplasia or femoroacetabular impingement between 2011 and 2013 were identified. Rapid sequence femoral version images were obtained from MRI Hip dGEMRIC and/or postarthrogram studies. Femoral version images were also obtained from bilateral CT lower extremity, without contrast, studies. Measurements were made by 1 fellowship-trained, pediatric hip preservation attending surgeon, 2 pediatric orthopaedic surgical fellows, and 1 fellowship-trained musculoskeletal radiologist on 2 separate occasions. Linear mixed models were used to estimate the reliability and repeatability associated with CT-based and MRI-based measurements (intraclass correlation coefficients) and to estimate the agreement (CT-MRI) between the 2 techniques. RESULTS: The mean age of 36 subjects was 15.4 years (±4.1 y). Interrater reliability was 0.91 (95% CI, 0.86-0.95) for the CT technique compared with 0.90 (95% CI, 0.86-0.94) for the rapid sequence MRI technique. Intrarater reliability for the CT technique was 0.96 (95% CI, 0.91-0.98) compared with 0.95 (95% CI, 0.90-0.97) for the MRI technique. The agreement between the MRI-based and CT-based techniques (bias: 1.9 degrees, limits of agreement: -11.3 to 14.9 degrees) was similar to the agreement between consecutive MRI measurements (bias: 0.4 degrees, limits of agreement: -7.8 to 8.6 degrees) as well as consecutive CT measurements (bias: 0.5 degrees, limits of agreement: -8.8 to 9.9 degrees). CONCLUSIONS: The interrater and intrarater reliability and repeatability estimates (intraclass correlation coefficient values) associated with both techniques was excellent (>0.90). Acquirement of axial images at the pelvis and knee during MRI for investigation of adolescents with hip pain allows for reliable measurement of femoral version. LEVEL OF EVIDENCE: Level II-diagnostic study.

Higher Pavlik Harness Treatment Failure Is Seen in Graf Type IV Ortolani-positive Hips in Males.

BACKGROUND: Patients with developmental dysplasia of the hip (DDH) whose hips are dislocated but reducible (Ortolani positive) are more likely to experience Pavlik harness treatment failure than are patients with dysplastic and reduced but dislocatable (Barlow positive) hips. However, data regarding factors associated with failure are limited and conflicting. QUESTIONS/PURPOSES: We asked: (1) What is the frequency of Pavlik harness treatment failure among Ortolani-positive hips, Barlow-positive hips, and dysplastic hips? (2) What are the factors predictive of failure of Pavlik harness treatment for Ortolani-positive hips? METHODS: In this retrospective study we identified 150 patients who underwent the Pavlik harness method for treatment of DDH between August 2011 and July 2015. Six patients initially treated at an outside facility, four patients with associated conditions, and three who pursued treatment elsewhere were excluded. A total of 137 patients (215 hips) with a median age at the time of Pavlik placement of 30 days (range, 4-155 days) were included. Of the 215 hips, 78 (36.3%) were Ortolani positive, 60 (27.9%) were Barlow positive, and 77 (35.8%) were stable, with the diagnosis of dysplasia made on ultrasound. All patients were treated with the Pavlik harness method. The primary outcome was failure of the Pavlik harness to achieve and maintain concentric hip reduction assessed by examination and ultrasound. All patients were followed after completion of Pavlik treatment for a minimum of 2 months (mean, 3 months; range, 2-4 months). In addition, 90% (122 of 137) of the patients were followed for a minimum of 6 months. Patient-specific data including family history, breech versus cephalic presentation at birth, age, sex, laterality, and hip abduction were recorded. Ultrasound data at the time of diagnosis included Graf classification, alpha angle, and percentage of femoral head coverage. RESULTS: The Pavlik harness method failed in 27% (21 of 78) of hips that were Ortolani positive, 8% (six of 77) with dysplasia, and 5% (three of 60) that were Barlow positive. After controlling for potential confounding variables, such as range of hip abduction, male sex (adjusted odds ratio [OR], 6.9; 95% CI, 2.0-24.2; p = 0.002) and Graf Type IV ultrasound classification (dislocated hip with alpha angle less than 43° and labrum displaced downward) (OR, 4.4; 95% CI, 1.3-15.4; p = 0.019) were identified as independent predictors of failure of Pavlik treatment among Ortolani-positive hips. CONCLUSIONS: Ultrasound imaging of the hip should be part of the initial assessment for Ortolani-positive hips, as the ultrasound classification was found to have prognostic implications. Parents of male infants with Graf Type IV hips should be counseled regarding the higher risk of Pavlik failure. Future well-designed prospective controlled studies are necessary to establish whether alternative strategies to the Pavlik harness might improve the early outcomes of DDH in males with Graf Type IV hips. LEVEL OF EVIDENCE: Level III, therapeutic study.