CASE:This is a case of a female patient born with thrombocytopenia-absent radius syndrome, with bilateral upper extremity phocomelia, bilateral hip dislocations, and congenital fusion of the right knee with progressively worsening flexion contracture. At age 3 years and 5 months, the patient was treated with excision of the knee ankylosis and Van Nes rotationplasty. This proved durable at age 20 years (final follow-up) without any need for further surgery and without complication.CONCLUSION:This is the first known report of Van Nes rotationplasty as a durable treatment option in the management of congenital knee ankylosis, which may avoid reoperation and eliminate risk of recurrence.

To understand if anatomic physeal-sparing ACL reconstruction in the immature host preserves range of motion, permits a return to sports, and avoids limb length discrepancy and accelerated intra-articular degeneration with a cross-sectional radiographic, physical examination and patient-reported outcomes analysis. A cross-sectional recall study included 38 patients aged 7–15 who underwent all-epiphyseal ACL reconstruction with hamstring allograft performed by a single surgeon at a large academic medical center. All-epiphyseal reconstructions were performed using a modified Anderson physeal-sparing technique, with the femoral tunnel placed using an “inside-out” technique. Assessments consisted of a physical exam, long leg cassette radiographs, KT-1000 measurements, subjective patient metrics, and magnetic resonance imaging. Thirty-eight (56.7%) of 66 eligible patients returned for in-person clinical and radiographic exams. Patients were 11.4 ± 1.8 years at the time of surgery. Five patients were females (13.2%). Mean follow-up was 5.5 ± 2.4 years. ACL re-injuries occurred in four patients (10.5%), all of whom underwent revision reconstructions. Thirty-three of the remaining 34 (97.1%) patients returned to sports following their reconstruction, and 24 (70.6%) returned to their baseline level of competition. Mean limb length discrepancy (LLD) was 0.2 ± 1.4 cm. Nine patients had an LLD of > 1 cm (26.5%), which occurred at an equivalent age as those with < 1 cm LLD (10.8 ± 2.0 vs. 11.7 ± 1.7, n.s.). Pre-operative Marx scores (13.1 ± 3.5) were not significantly different from post-operative values (12.3 ± 5.1, n.s.). Patients who required ACL revisions had significantly lower Marx scores than those with intact primary grafts (8.3 ± 7.1 vs. 13.4 ± 4.5, p = 0.047). Cohort mean International Knee Documentation Committee (IKDC) score was 89.7 ± 12.7. Anatomic all-epiphyseal anatomic ACL reconstruction appears to be useful in patients with significant projected remaining growth, with good return-to-sport outcomes and minimal risk of clinically significant physeal complications. However, given the limited patient recall possible in the present study, further large sample size, high-quality works are necessary to validate our findings. Level IV.

Prospective cohort study. The objective of this study was to examine intermediate-term progression for a large series of patients with adolescent idiopathic scoliosis (AIS) with curves 40° or greater. Curve progression in AIS has been well documented for smaller curves in adolescence up to skeletal maturity; however, the data on curve progression past 40° or into adulthood are limited. With many surgeons recommending surgical correction when patients reach this threshold, it is important to understand the radiographic progression of curves into adulthood. A database of all patients seen by a single surgeon from 1984 through 2018 with AIS curves progressing to at least 40° entered prospectively was utilized for this study. This included a total of 738 patients. Curve progression was analyzed overall and stratified by length of follow-up, curve location, and Risser stage at the time of presentation among other variables. Curve magnitude and Risser stage designations in this study were validated by performing a separate inter- and intrarater agreement study using four independent reviewers reading 50 patients’ Cobb angle and Risser stage blinded in triplicate to examine the reliability of the study measurements. Annualized curve progression (ACP) averaged 6.3 ± 10.4°. ACP varied with length of follow-up: patients with up to 1 year of follow-up had an average ACP of 11.5 ± 17.0°, while those with 1–2 years had 8.2 ± 8.8°, and 2–5 years had 3.7 ± 4.1°, tapering off further from there. Risser stage 0 or 1 was associated with the highest ACP as compared to Risser stage 2–3 or 4–5. Intraclass correlation (ICC) values for Cobb angle measurement and Risser stage designations from four raters measuring 50 patients’ measures, blinded and in triplicate, were all > 0.80, signifying a high degree of reliability within and between readers. Annualized curve progression for 40° and greater curves was not linear over time; it was greatest immediately after a curve reaches 40° and tapered off over the next decade. Immature Risser stage at presentation was strongly associated with increasing ACP at all time frames. Prognostic Level I

Background:Closed reduction of pediatric fractures is commonly performed by orthopaedic residents using conscious sedation in the emergency department (ED). The purpose of this study was to determine the rate of satisfactory reductions as performed by residents, and to determine the outcomes of these procedures.Methods:A retrospective review was performed of all fractures that underwent closed reduction under conscious sedation in the ED of a level 1 pediatric trauma center between January 1, 2010 and November 30, 2014. Initial and subsequent radiographs were reviewed and a determination was made as to whether the initial reduction was satisfactory, based on predetermined criteria for angulation and displacement. If a second reduction attempt in the operating room was necessary, this was noted. Chart notes were reviewed until a documented endpoint was reached, such as uneventful healing, malunion, nonunion, or growth arrest.Results:A total of 838 subjects were identified. The upper extremity was involved in 85% of the fractures. Of the initial 838 fracture reductions performed, 39 (4.7%) were unsatisfactory. Residents on their first pediatric orthopaedic rotation had a higher unsatisfactory reduction rate compared with more experienced residents (7.0% vs. 3.4%, P=0.01). A second reduction was performed for 94 of 749 (12.6%) fractures. Of these, 35 (37.2%) required an open procedure to accomplish a satisfactory reduction. Fractures with initially satisfactory reductions were significantly less likely to require a second reduction attempt than those with initially unsatisfactory reductions (9.2% vs. 80.0%, P<0.01). The likelihood of a satisfactory reduction was significantly higher in the upper extremity than in the lower extremity. Overall, the vast majority (99.2%) of fractures had a satisfactory final outcome.Conclusions:Most attempts at closed reduction of pediatric fractures in the ED by orthopaedic residents are successful, and the likelihood of a satisfactory reduction was associated with increased levels of resident experience. Fractures with an initially successful reduction were far less likely to require remanipulation.Level of Evidence:Level IVthis is a therapeutic case series.

Background:The Joint Commission on Accreditation of Healthcare Organizations specifically mandates the dual interpretation of musculoskeletal radiographs by a radiologist in addition to the orthopaedist in all hospital-based orthopaedic clinics. Previous studies have questioned the utility of this practice. The purpose of this study was to further investigate the clinical significance of having the radiologist provide a second interpretation in a hospital-based pediatric orthopaedic clinic.Methods:A retrospective review was performed of all patients who had plain radiographs obtained in the pediatric orthopaedic clinic at an academic children's hospital over a 4-month period. For each radiographic series, the orthopaedist's note and the radiology interpretation were reviewed and a determination was made of whether the radiology read provided new clinically useful information and/or a new diagnosis, whether it recommended further imaging, or if it missed a diagnosis that was reflected in the orthopaedist's note. The hospital charges associated with the radiology read for each study were also quantified.Results:The charts of 1570 consecutive clinic patients who were seen in the pediatric orthopaedic clinic from January to April, 2012 were reviewed. There were 2509 radiographic studies performed, of which 2264 had both a documented orthopaedist's note and radiologist's read. The radiologist's interpretation added new, clinically important information in 1.0% (23/2264) of these studies. In 1.7% (38/2264) of the studies, it was determined that the radiologist missed the diagnosis or clinically important information that could affect treatment. The total amount of the professional fees charged for the radiologists' interpretations was $87,362. On average, the hospital charges for each occurrence in which the radiologist's read provided an additional diagnosis or clinically important information beyond the orthopaedist's note were $3798.Conclusions:The results of this study suggest that eliminating the requirement to have the radiologist interpret radiographs in the pediatric orthopaedic clinic would have few clinical consequences.Level of Evidence:Level IIIThis is a diagnostic retrospective cohort study.

Background: This study sought to determine the hip pathology of family members of patients with developmental dysplasia of the hip (DDH). The authors evaluated 120 people from 19 families known to have at least 1 member with surgically treated DDH. Each individual’s functional outcome scores and pelvic radiographs were assessed for hip symptoms or pathology. Methods: Using a genetic population database and a pediatric hospital patient population, 19 families with high rates of DDH were identified. All family members (n=120) underwent physical examination, radiographic assessment, and completion of outcome instruments [American Academy of Orthopedics (AAOS) Hip and Knee; Harris Hip Score (HHS); and Western Ontario and McMaster Universities Arthritis Index (WOMAC)]. Results: The 120 subjects ranged from 1 to 84 years, 34 had orthopaedically treated DDH. Of the remaining 86 supposedly normal subjects, 23 (27%) had occult acetabular dysplasia (OAD) as defined by center edge angle (CEA) <20 and/or a Severin score of III or greater. Sixty percent of the 86 individuals were less than 30 years old, 74% of the OAD group were less than 30. Outcome scores of the treated DDH patients (AAOS, HHS, and WOMAC) were worse on the involved side regardless of age. Over age 30 individuals with OAD had statistically significant decreases in their AAOS Hip and Knee and WOMAC scores on the dysplastic side, but their HHS scores were not significantly different. Conclusions: Twenty-seven percent of first-degree and second-degree relatives of patients with DDH had unsuspected radiographic acetabular dysplasia in our study. Most of the subjects with OAD were younger than 30. After age 30, many of these patients developed symptoms. Clinical Relevance: In families with a significant history of DDH, radiographic screening of siblings of patients with DDH to define OAD may be prudent. Level of Evidence: Level I—diagnostic study.

BACKGROUND:Performance on the Orthopaedic In-training Examination (OITE) has been correlated with performance on the written portion of the American Board of Orthopaedic Surgery examination. Herein we sought to discover whether adding a regular pediatric didactic lecture improved residents' performance on the OITE's pediatric domain.METHODS:In 2012, a didactic lecture series was started in the University of Pittsburgh Medical Center (UPMC) Hamot Orthopaedic Residency Program (Hamot). This includes all topics in pediatric orthopaedic surgery and has teaching faculty present, and occurs weekly with all residents attending. A neighboring program [UMPC Pittsburgh (Pitt)] shares in these conferences, but only during their pediatric rotation. We sought to determine the effectiveness of the conference by comparing the historic scores from each program on the pediatric domain of the OITE examination to scores after the institution of the conference, and by comparing the 2 programs' scores.RESULTS:Both programs demonstrated improvement in OITE scores. In 2008, the mean examination score was 19.6±4.3 (11.0 to 30.0), and the mean percentile was 57.7±12.6 (32.0 to 88.0); in 2014, the mean examination score was 23.5±4.2 (14.0 to 33.0) and the mean percentile was 67.1±12.1 (40.0 to 94.0). OITE scores and percentiles improved with post graduate year (P<0.0001). Compared with the preconference years, Hamot residents answered 3.99 more questions correctly (P<0.0001) and Pitt residents answered 2.93 more questions correctly (P<0.0001). Before the conference, site was not a predictor of OITE score (P=0.06) or percentile (P=0.08); there was no significant difference found between the mean scores per program. However, in the postconference years, site did predict OITE scores. Controlling for year in training, Hamot residents scored higher on the OITE (2.3 points higher, P=0.003) and had higher percentiles (0.07 higher, P=0.004) than Pitt residents during the postconference years.CONCLUSIONS:This study suggests that adding a didactic pediatric lecture improved residents' scores on the OITE and indirectly suggests that more frequent attendance is associated with better scores.LEVEL OF EVIDENCE:Level III-retrospective case-control study.

STUDY DESIGN:Case control comparative series.OBJECTIVE:Describe surgical range adolescent idiopathic scoliosis (AIS) patients electing to forgo surgery and compare health-related quality-of-life outcomes to a similar cohort of operated AIS patients by the same single surgeon.SUMMARY OF BACKGROUND DATA:No data have been published either documenting SRS-22r scores of nonoperated patients with curves ≥40° or comparing these scores to a demographically similar operated cohort.METHODS:Individuals with curves ≥40°, age ≥18 years, and electing to forgo surgery were identified. All patients completed an SRS-22r questionnaire. This nonoperated cohort's SRS-22r scores were compared to those of a large demographically similar cohort operated by the same surgeon. Group differences between the SRS-22r scores were evaluated by comparing these to published Minimal Clinically Important Differences (MCID) for the SRS-22r.RESULTS:One hundred ninety subjects with nonoperated curves were compared to 166 individuals who underwent surgery. The nonoperated cohort averaged 23.5 years of age, averaged 7.7 years since curve reached 40°, and had an average 50° Cobb angle at last follow-up. No statistical significant differences were found between the groups on the Pain, Function, or Mental Health domains of the SRS-22r. Statistically significant differences in favor of the operative cohort were found for self-image, satisfaction, and total score. The observed group differences did not meet the established thresholds for minimal clinically important differences in any of the domain scores, the average total score, or raw scores.CONCLUSION:There are no meaningful clinically significant differences in SRS-22r scores at average 8-year follow-up between AIS patients with curves ≥40° treated with or without surgery. These data in conjunction with an absence of long-term evidence of serious medical consequences with nonsurgical management of curves ≥40° should encourage surgeons to reevaluate the benefits of routine surgical care.LEVEL OF EVIDENCE:3.

The growth plate, also known as the physis or epiphyseal plate, is essential for longitudinal growth of bones in the immature skeleton. A variety of insults to the growth plate from trauma to infection to idiopathic causes can lead to physeal bar formation, an interruption in normal growth plate cartilage, where a bony or fibrous bridge develops between the metaphysis and epiphysis. This bridge restricts subsequent bone growth, leading to limb shortening and/or angular deformities. Early recognition of the presence of a physeal bar can help direct appropriate surgical management to restore linear growth of the bone.

Study Design. Biomechanical cadaver study. Objective. To evaluate the relative effectiveness of Ponte osteotomies for spinal release in deformity correction. Summary of Background Data. Controversy exists as to the role of Ponte osteotomy in deformity correction surgery. Very little has been written about the biomechanical effects of Ponte osteotomy. Past biomechanical studies have been limited to application of forces through endplates, single functional units, or lack of comparison with anterior release. Methods. Twelve fresh-frozen human full thoracic spinal units were tested for motion in axial rotation, flexion/extension, and lateral bending in a custom-designed robotic environment. Testing was repeated after sequential facetectomy and Ponte osteotomy (6 specimens) and compared with partial and full discectomy (6 specimens). Results. Motion in axial rotation is increased 21% by Ponte osteotomy compared with 35% for full discectomy. Anterior displacement of the spinal column, creating lordosis, was increased 15% by Ponte osteotomy and 40% by full discectomy. Posterior displacement of the spinal column, creating kyphosis, was increased 23% by Ponte osteotomy and 89% by full discectomy. Finally, in coronal force application the Ponte osteotomy had virtually no effect (2%) compared with 40% increased motion by full discectomy. Conclusion. Posterior Ponte osteotomy releases produced more motion than facetectomy alone in axial rotation and sagittal correction maneuvers, but had no effect on coronal correction. Anterior discectomy release destabilized spinal column significantly more than posterior releases in all force applications. Despite ample clinical experience demonstrating the effectiveness of posterior-only surgery, the biomechanical effect of Ponte osteotomies is modest. Level of Evidence: N/A

Retrospective. To identify the clinical factors suggestive of infected and non-infected drainage to help clinical decision making. Summary of Background Data: Differentiating between drainage caused by a benign seroma and deep spinal infection may be difficult in the early postoperative period. Institutional spine surgery database was searched to identify the cases that were taken back to the operating room for drainage from the surgical wound in the early postoperative period between 2000 and 2012. A total of 38 cases of early wound drainage (within 6 weeks postoperatively) were identified that were treated with opening all layers, irrigation, and debridement. Intraoperative cultures were sent in all cases. Twenty-five patients proved to have non-infected drainage and did not require further treatment. In 13 patients, infection was confirmed with intra-operative findings and cultures; these patients were treated with serial debridements. In 4 cases, implants had to be removed after multiple debridements (after a quiescent period). The group with non-infected drainage differed from the infection group in that most patients (21 of 25) had non-neuromuscular deformities, whereas 77% of the infected group had neuromuscular etiology (10 of 13) (p =.0004). Average number of days to revision was 8.5 (range, 5–14 days) for the non-infected group. Of the 25 patients, 23 presented in the first 10 days. In the infected group, average number of days to revision was 19. Ten of the 13 patients presented on postoperative day 14 or later. Logistic regression analysis showed a significant association between increased likelihood of infection and increased time from the index procedure (p =.0085). The findings suggest that early presenting drainage in pediatric idiopathic spine deformity is often not infected. Drainage, especially presenting after the second postoperative week in neuromuscular patients, proved to be mostly deep spinal wound infections.

STUDY DESIGN:Retrospective case series.OBJECTIVE:This study evaluated the incidence of postoperative neurological symptoms after a freehand pedicle screw insertion technique in idiopathic posterior scoliosis surgery.SUMMARY OF BACKGROUND DATA:It is generally accepted that pedicle screws can be inserted by a freehand technique in the thoracic and lumbar spine in patients with adolescent idiopathic scoliosis (AIS) with a very low frequency of major complications. The prevalence of clinically significant screw misplacement, with or without the need for revision surgery is less well defined.METHODS:Between January 1, 2000, and October 2, 2012, five hundred fifty-nine patients with AIS had thoracolumbar posterior instrumented spine surgery at the Children's Hospital of Pittsburgh. Each patient's chart and radiographs were reviewed and only those with AIS were included. Patients with neuromuscular and syndromic diagnoses were excluded as well as those with congenital or traumatic etiologies, incomplete charts, less than 3 months of follow-up and those without pedicle screws. The records were studied for complaints of radicular pain, neurological deficit, or severe headache that could be indicative of potential screw misplacement.RESULTS:Four hundred eighty-one patients with 5923 pedicle screws met the inclusion criteria. Nine patients (1.9%) developed symptoms and underwent computed tomographic scanning. Six patients were found to have pedicle screw malposition (8 screws) and 3 of these patients underwent revision surgery. Of the 3 revision patients, 2 presented with radicular symptoms (leg pain) and 1 with an orthostatic headache due to cerebrospinal fluid leakage. At the final follow-up, all revision patients had complete symptom resolution. In total, there were 8 symptomatic, misplaced pedicle screws (0.14%) in 6 patients (1.25%).CONCLUSION:During a 12-year period in a dedicated pediatric orthopedic hospital using the freehand placement technique, the incidence of symptomatic misplaced pedicle screws was exceedingly low.LEVEL OF EVIDENCE:4.

Severe varus and flexion deformity of the proximal femur may be secondary to various etiologies but, regardless of the cause, the patient walks with a dramatic Trendelenburg limp. Surgical correction is challenging and requires a complete understanding of all components of the deformity. A utilitarian approach (SuperHip procedure) has been described by Dror Paley and addresses every element of this complex deformity and provides excellent correction in young children. This article illustrates the application of this technique in a case of congenital femoral deficiency.(C) 2013 Elsevier Inc. All rights reserved.

Developmental dysplasia of the hip describes an anatomical abnormality of the hip joint in which femoral head and acetabulum have an incongruent relationship, ranging from acetabular dysplasia with insufficient acetabular coverage of the femoral head to complete dislocation. Treatment of the ambulatory child with developmental dysplasia of the hip usually requires a surgical reduction, often combined with a shortening femoral osteotomy and a pelvic osteotomy. This paper describes the techniques employed in the surgical management of a dislocated hip in an ambulatory child. (C) 2013 Elsevier Inc. All rights reserved.

The management of young patients with severe hip arthritis is very difficult, as treatment options, such as total hip arthroplasty, which is excellent for relieving pain while preserving function, are often avoided because of a high requirement for subsequent revision surgeries. Hip arthrodesis, although perceived as having inferior functional outcomes in these young individuals, offers a treatment option that relieves pain, maintains function, and allows for conversion to a total hip arthroplasty in future, if desired. This article reviews the indications and technique of hip fusion.

The neuromuscular disorder-affected child with severe hip subluxation presents a difficult challenge for the pediatric orthopaedic surgeon. Surgical correction is often necessary and can range from soft tissue releases to bony procedures including proximal femoral osteotomy, proximal femoral resection, arthrodesis, arthroplasty, and pelvic osteotomy. In this article, the indications and surgical techniques for these procedures are discussed. In addition, some general considerations for the care for the child with neuromuscular disorder are considered. (C) 2013 Elsevier Inc. All rights reserved.

BACKGROUND:Developmental dysplasia of the hip (DDH) is a familial condition with a wide phenotypic expression. Families with high rates of DDH may have individuals with subtle phenotypic expression that can progress to osteoarthritis and require total hip arthroplasty (THA). This study compares the rates of THA in relatives of individuals with DDH with individuals in control families.METHODS:Probands with a diagnosis of DDH were identified using medical records linked to the Utah Population Database. Ten age-matched and sex-matched controls were randomly selected from a pool of unaffected individuals within the Utah Population Database. Diagnostic and procedural codes were used to determine the incidence of hip and knee osteoarthritis (HOA and KOA) and of THA and total knee arthroplasty (TKA) among the cases and controls and their relatives. Relative risks (RR) for HOA and KOA and for THA and TKA were calculated for the probands/controls and their family members.RESULTS:The RR of HOA was significantly increased in probands (RR=82.4; P<2e-16), their parents (R=2.22; P=0.0003), and in their grandparents (RR=1.33; P=0.011). The RR of THA was also significantly increased in probands (RR=1168; P <3e-08) and in their grandparents (RR=2.06; P=0.01). The RR of KOA was significantly increased in probands with a diagnosis of DDH (RR=20.96; P=2.2e-8) but not in their parents or grandparents. The RR of TKA was also increased in probands alone (RR=57.47; P=1.7e-05).CONCLUSIONS:Parents and grandparents of individuals with diagnosed DDH are significantly more likely to be diagnosed with HOA and undergo THA than members of the general population. These first-degree and second-degree relatives were not at higher risk for KOA or TKA. Given the known familial association of DDH, this association with osteoarthritis of the hip suggests a risk of undiagnosed hip dysplasia in individuals whose families have a high rate of DDH.LEVEL OF EVIDENCE:Level III.

Background: We posed 2 questions: what is the long-term result of open reduction surgery in developmental dysplasia of the hip, and is there an age at surgery above which the outcome was too poor to recommend the operation?Methods: Between 1955 and 1995, 148 patients with 179 dislocated hips had open reduction surgery for developmental dysplasia of the hip (141 anterior and 38 Ludloff medial approaches). We attempted to locate all 148 patients for the follow-up evaluation.Results: Fifty-three patients (36%) with 66 hips (37%) were located and participated in the study. These 66 hips represented 34% of the anterior open reductions and 47% of the Ludloff medial reductions. Twenty-two of the 66 hips had Severin IV or worse outcomes and included 7 with total hip arthroplasties and 2 with hip fusions. Age at surgery was significantly lower for Severin I, II, and III, compared with Severin IV and above (P = 0.003, 0.001, 0.003) with outcomes deteriorating substantially after age 3. Approximately half of the hips required further surgery for dysplasia. All hips that sustained osseous necrosis had Severin IV or worse outcomes, and hips that redislocated and required revision surgery only achieved Severin I or II ratings 18% of the time. Nine "normal" hips became dysplastic and 3 had pelvic osteotomies as teenagers. Two other normal hips developed osseous necrosis during treatment of the contralateral hip.Conclusions: Results deteriorate as the age at surgery increases. Osseous necrosis and redislocation predict a poor functional and radiographic result. The "normal" hip may develop insidious dysplasia and also may be injured during treatment of the involved hip. Above age 3, some patients may not have sufficient acetabular growth to remodel a surgically reduced hip.Level of Evidence: Level IV-case series.

Objectives: Isolated dislocations of the radiocapitellar joint or those associated with fractures of the ulna (Monteggia fracture-dislocations) are often missed in the emergency room. When left untreated radial head dislocations can cause pain, loss of motion, and progressive cubitus valgus. This report describes open reduction and annular ligament reconstruction to treat chronic radial head dislocations. Design: Retrospective chart review of 11 patients with chronic radial head dislocations Setting: Texas Scottish Rite Hospital and Cook Fort Worth Children's Hospital,USA Subjects: Eleven patients had chronic radial head dislocations; six Monteggias, two intra-articular fracture-dislocations, and three isolated radial head dislocations Intervention: Four patients were left untreated. Seven patients had surgical reconstruction with open reduction of the radiocapitellar joint and repair (1 patient) or reconstruction (6 patients) of the annular ligament. Three osteotomies were performed to straighten any ulnar malunion. Main Outcome Measures: Two of the four untreated patients demonstrated progressive cubitus valgus. Five of the seven operated patients had normal carrying angles and recovered full flexion and extension but lost 30% of their pronosupination. The two intra-articular fracture-dislocations had very poor results due to extremely stiff joints. Conclusions: We recommend surgical relocation and annular ligament reconstruction to treat chronic radial head dislocations. This reconstruction prevented cubitus valgus in patients with chronic radial head dislocations and restored elbow flexion. Patients with fracture-dislocations do not respond as well.