Pathogenesis of metaphyseal radiolucent changes following ischemic necrosis of the capital femoral epiphysis in immature pigs. A preliminary report

BACKGROUND: Although metaphyseal radiolucent changes are often seen in patients with Legg-Calvé-Perthes disease, the pathogenesis of these changes remains controversial. The purpose of the present study was to determine the prevalence and histopathological characteristics of these metaphyseal radiolucent changes in a piglet model of ischemic necrosis of the capital femoral epiphysis. METHODS: Ischemic necrosis of the right femoral head was produced in fifty piglets by surgically placing a ligature tightly around the femoral neck. The contralateral, left hip of each animal was used as a control. Radiographs and histological sections of the femoral heads were examined at two, four, and eight weeks. The radiographs were used to measure the femoral neck length in order to assess growth disturbance. RESULTS: Thirteen of the fifty animals were found to have radiolucent changes in the proximal femoral metaphysis on the side of the infarcted femoral head. These changes were observed in none of the twelve animals that were evaluated at two weeks, in one of the fourteen animals that were evaluated at four weeks, and in twelve of the twenty-four animals that were evaluated at eight weeks. The radiolucent changes ranged from a focal cystic lesion to a diffuse area of radiolucency around the proximal femoral physis. Three distinct types of histological changes were observed in the metaphysis. Type-I changes were characterized by focal thickening of the physeal cartilage extending down into the metaphysis. Some of these lesions demonstrated cystic degeneration of the thickened cartilage. Type-II changes were characterized by central disruption of the physis and resorption and replacement of the metaphyseal bone in the region by fibrovascular tissue. Type-III changes were characterized by diffuse resorption of the physeal cartilage and resorption of the adjacent metaphyseal and epiphyseal bone. The mean femoral neck length on the infarcted side in animals with metaphyseal radiolucent changes was significantly shorter than that in animals without metaphyseal radiolucent changes (p = 0.02). CONCLUSIONS: Metaphyseal radiolucent changes frequently were observed in this piglet model at eight weeks after the induction of ischemia. The metaphyseal radiolucent changes were associated with histopathological lesions of the physis. The present study suggests that the presence of metaphyseal lesions can result in a greater growth disturbance of the proximal femoral physis than is seen in the absence of metaphyseal lesions. Clinical Relevance: The present study supports the clinical observation that the presence of diffuse metaphyseal radiolucent changes may be associated with substantial growth disturbance of the proximal part of the femur in patients with Legg-Calvé-Perthes disease. The study provides a histopathological basis for proximal femoral physeal growth disturbance that has not been clearly demonstrated in the past. These findings also may provide a histopathological basis for the fluidfilled metaphyseal cysts that sometimes are observed on the magnetic resonance imaging scans of these patients.     

The effect of osteogenic protein-1 in an in vivo physeal injury model

The physis has limited ability to undergo repair, and injury may result in growth arrest. Osteogenic protein-1 promotes bone formation in diaphyseal defects, chondrocyte proliferation, and matrix synthesis. The authors' goal was to determine if the presence of osteogenic protein-1 in a defect involving the physis would promote cartilage repair, and in doing so, to determine the effect of osteogenic protein-1 on physeal growth. An ovine model of growth plate damage was used, in which the proximal medial physis of the tibia was partially ablated. The defect was filled with a Type I collagen paste containing osteogenic protein-1 (350 microg) or collagen alone. Growth rate was measured at 4, 14, and 56 days, and the defects were analyzed histologically at 4, 14, and 56 days. Bone bridge formation occurred within the defect site. However, osteogenic protein-1 promoted outgrowth of the adjacent physeal cartilage. The physeal cartilage underwent expansion until the mineral forming within the defect site blocked its progress. The effect was localized because only that portion of the physis at the defect margin appeared to be affected.     

Free physeal transplantation in the rabbit. An experimental approach to focal lesions

We have developed a procedure to transplant free autogenous iliac-crest physeal grafts into defects created in the lateral aspect of the distal femoral physis of rabbits. Excision of a portion of the lateral part of the physis led consistently to formation of a bone bridge, growth arrest, and valgus deformity. There was no evidence of increased activity by the persisting physis to fill the defect or of differentiation of mesenchymal cells into a functioning physeal layer. We carried out preliminary studies to define the optimum conditions for obtaining and transplanting a free autogenous iliac-crest physeal graft into the femoral physeal defect. The iliac apophysis is composed of a fibrocartilaginous layer, cartilage similar to epiphyseal cartilage, and a physis. The graft included the physis and some overlying epiphyseal cartilage but the outermost fibrocartilaginous layer and the surrounding perichondrium were removed, as these two tissues were believed to inhibit diffusion. The graft was separated gently from the metaphysis. Histological studies showed that the separation occurred at the lower hypertrophic-chondrocyte zone. After optimum techniques had been developed for transplantation and the feasibility of the procedure had been established in twenty-one rabbits, the capacity of the physeal transplants to prevent bone-bridge formation, growth arrest, and valgus deformity was evaluated in thirty-nine rabbits, three to four months old, using the standardized transplantation procedure. The results were assessed in terms of the gross and radiographic appearance of the femur, measurements of length, and histological characteristics. The transplanted physis retained its normal morphology, united with the residual part of the original femoral physis, and participated in endochondral bone formation. Bone-bridge formation between the femoral epiphysis and metaphysis, growth arrest, and valgus deformity were prevented or minimized in most animals, and the physeal transplants yielded good or excellent results in 60 per cent of the procedures using stringent criteria. The results of this study show that free physeal transplants into focal defects can remain viable and function satisfactorily if they are fashioned properly and inserted gently. CLINICAL RELEVANCE: This study demonstrated that transplantation of an iliac-crest physis into a defect of the lateral aspect of the distal femoral physis can prevent bone-bridge formation, growth arrest, and the development of valgus deformity.(ABSTRACT TRUNCATED AT 400 WORDS)     

Juxtaphyseal aneurysmal bone cysts.

Aneurysmal bone cysts are benign primary or secondary lesions that commonly arise in long bones and often before skeletal maturity. Little has been written about aneurysmal bone cysts that abut the physeal plate. The records of 15 patients with juxtaphyseal aneurysmal bone cysts were reviewed. Fourteen of the patients were referred with abnormal radiographs after evaluation for pain in the affected limb. One patient presented with abnormal radiographs after fracture about the aneurysmal bone cyst. None of the patients had evidence of growth plate disruption. The children's ages ranged from 2 to 14 years, with a mean of 9.8 years. There were 10 boys and five girls. Lesion locations included: six in the proximal tibia, three in the distal fibula, two in the distal tibia, two in the proximal femur, one in the distal femur, and one in the distal radius. All of the lesions abutted the physeal plate and fell into one of the types in Campanacci's classification of juxtaphyseal aneurysmal bone cysts. Three lesions were classified as Type 1, eight were Type 2, and four were Type 3. This study included no cases of Type 4 or 5 lesions. Treatment of all lesions consisted of excision, curettage, and bone grafting with care taken to preserve the growth plate. Adjunctive cauterization was performed in two cases. There were no incidences of postoperative physeal plate arrest. Overgrowth of the fibula occurred in one patient. Three patients experienced recurrent lesions. One of the children underwent repeat curettage and bone grafting with no additional recurrence. In the other two children with recurrence, the lesion had grown away from the physeal plate while remaining static in size and asymptomatic. Based on this study, juxtaphyseal aneurysmal bone cysts may be treated satisfactorily with intralesional surgery and bone grafting with expectation of normal physeal growth.     

Physeal arrest about the knee associated with non-physeal fractures in the lower extremity

The cases of seven patients who had a physeal arrest about the knee in association with nonphyseal fractures in the lower extremity were reviewed. The patients were between ten and twelve and one-half years old at the time of injury, and the physeal arrest involved either the posterolateral part of the distal femoral physis or the anterior part of the proximal tibial physis. There was no evidence of iatrogenic trauma to the physis. Recognition of the physeal injury was delayed for an average of one year and ten months until a gross angular deformity appeared. Adolescents who have fractures of the lower extremities that do not appear to involve a physeal plate should nevertheless be evaluated and followed for possible physeal injury about the knee that can be detected only after additional growth has taken place.     

The effect of percutaneous pin fixation in the treatment of distal femoral physeal fractures

Distal femoral physeal fractures in children have a high incidence of physeal arrest, occurring in a mean of 40% of cases. The underlying nature of the distal femoral physis may be the primary cause, but other factors have been postulated to contribute to the formation of a physeal bar. The purpose of this study was to assess the significance of contributing factors to physeal bar formation, in particular the use of percutaneous pins across the physis. We reviewed 55 patients with a median age of ten years (3 to 13), who had sustained displaced distal femoral physeal fractures. Most (40 of 55) were treated with percutaneous pinning after reduction, four were treated with screws and 11 with plaster. A total of 40 patients were assessed clinically and radiologically after skeletal maturity or at the time of formation of a bar. The remaining 15 were followed up for a minimum of two years. Formation of a physeal bar occurred in 12 (21.8%) patients, with the rate rising to 30.6% in patients with high-energy injuries compared with 5.3% in those with low-energy injuries. There was a significant trend for physeal arrest according to increasing severity using the Salter-Harris classification. Percutaneous smooth pins across the physis were not statistically associated with growth arrest.     

Simple bone cysts of the proximal humerus complicated with growth arrest

A simple bone cyst of the proximal humeral metaphysis was found to cause growth disturbance with shortening and deformity in four patients. All had one pathological fracture. Three of them were treated with cortisone injections; the fourth patient, who presented an erosion of the physis, was treated with saline solution irrigation. Four hypotheses about the aetiology of growth arrest in the evolution of this benign lesion, are discussed: iatrogenic lesion of the physis, growth plate involvement by fracture, cortisone injection, and increased cyst pressure leading to erosion and even perforation of the growth plate.     

Acetabular development in developmental dysplasia of the hip complicated by lateral growth disturbance of the capital femoral epiphysis

BACKGROUND: Lateral growth disturbance of the capital femoral epiphysis is the most common type of physeal arrest complicating the treatment of developmental hip dysplasia. Although this type of physeal damage has been assumed to result in poor acetabular development, the natural history of dysplastic hips affected by this pattern of growth disturbance is still unclear. To investigate this issue, we evaluated acetabular development in a retrospective study of fifty-eight hips in forty-eight patients who had lateral physeal arrest after management of developmental hip dysplasia. METHODS: Of the fifty-eight hips, thirty-six were reduced closed and twenty-two were reduced open. The average age of the patients was twenty-two months (range, three to ninety-seven months) at the time of the reduction and twenty-one years (range, ten to fifty-five years) at the time of the latest follow-up evaluation. Hips rated as Severin class I (an excellent result) or II (a good result) were defined as having a satisfactory result, and those rated as Severin class III (a fair result) or IV (a poor result) were considered to have an unsatisfactory result. Specific femoral head changes were sought in the complete radiographic files on all hips. Various radiographic parameters of hip integrity, including the degree of lateral tilt of the capital femoral epiphysis, were measured over time, and comparisons were made between hips classified as satisfactory and those classified as unsatisfactory at four time-points: before the reduction, at two years after the reduction, at six to eight years of age, and at the time of the final follow-up. RESULTS: Lateral growth disturbance of the capital femoral epiphysis was first evident by an average of ten years of age (range, four to fourteen years of age). There was no consistent early pattern of changes in the epiphysis, physis, or metaphysis related to later development of valgus tilt of the epiphysis. Thirty-four hips (59 percent) were rated as satisfactory and twenty-four were rated as unsatisfactory at the latest follow-up evaluation. Hips classified as unsatisfactory exhibited poor acetabular development by an average age of seven years. The inclination of the epiphyseal plate became progressively more horizontal or even reversed over time; however, serial measurements of inclination were not significant predictors of Severin classification. CONCLUSIONS: Lateral growth disturbance of the capital femoral epiphysis is not necessarily associated with poor acetabular development, as when dysplasia does occur it is generally evident prior to the identification of the physeal arrest. It is important to monitor acetabular development after reduction rather than search for radiographic changes of physeal arrest, which are difficult to detect in young children.     

Growth arrest of the distal radius following a metaphyseal fracture: case report and review of the literature

We report a 12-year-old girl who developed growth arrest of the distal radius physis 9 months after sustaining a complete fracture of the distal radial and ulnar metaphysis with no involvement of the physis evident at time of injury. The girl sustained a fracture of the metaphysis of her right distal radius and ulna after a fall. Anterior-posterior, lateral and oblique radiographs at injury, and during subsequent healing show no evidence of the fracture involving the physis. She was treated with closed reduction and casting for 6 weeks and healed uneventfully. She returned 4 month later concerned about distal ulnar prominence. Radiographs revealed a loss of radial tilt and with suspicion of a physeal bar. Magnetic resonance imaging confirmed a physeal bar located in the dorsal radial region. A literature search of the Medline database was used to obtain prior case reports for review purpose. The patient underwent an epiphysiodesis of the distal radius and ulna along with an opening wedge osteotomy and bone grafting of the distal radius to restore radial height and inclination. She healed without complication and with restoration of the normal relationship of the distal radius and ulna. A review of the literature reveals five reported case of distal radial metaphyseal fractures not invloving the physis leading to growth arrest. By comparison, there are 31 reported cases of distal radius physeal arrest following fractures involving the physis. The physician should be aware that common distal radius metaphyseal fractures may rarely lead to growth arrest.     

Secondary tethers after physeal bar resection: a common source of failure?

Despite a standardized operative technique and appropriate patient selection for physeal bar resection, a bar size less than 50% of the physis, and a prospective growth period greater than 2 years, failure may result. Limited growth because of poor function of the remaining physis and secondary tethers (incomplete resection or recurrence of the bar) may prevent reestablishment of growth or lead to its premature cessation. The current study investigated patients with insufficient restoration of growth by means of magnetic resonance imaging or computed tomography or both. Twenty-two patients had 24 physeal bar resections with interposition of autologous fat as the index procedure (Langenski?ld technique). Fourteen patients had only fair or poor results. Five patients had premature arrest of the affected physis with a postoperative growth period less than 1 year. Radiologic and clinical findings revealed bridge recurrence in four patients but no obvious reasons in the remaining five patients. Postoperative magnetic resonance imaging or computed tomography or both detected a secondary tether because of incomplete bar resection (one patient) or recurrence of the bar (four patients). Graft dislocation out of the resection cavity with an associated recurrence of the bar proved to be the underlying problem in three of the eight patients with bar recurrence. Magnetic resonance imaging is not only useful in preoperative mapping of physeal bars but also may help to explain failures after growth plate surgery.     

Epiphysiodesis by electrocautery in the rabbit and dog

Numerous experiments have demonstrated arrest of longitudinal bone growth by traumatizing the epiphyseal growth plate. Recently, promising studies have been performed using less invasive means to produce epiphysiodesis than the conventional surgical technique. The distal femoral physeal plate was cauterized in 20 rabbits and 17 dogs through an inserted needle and a standard operating room electrocautery machine. The result showed a progressive distal femoral length reduction without angular deformities in the experimental limb averaging 3 mm at two weeks to 11.6 mm at 20 weeks after the surgical procedure in the rabbit and 11.5 mm at 12 weeks in the dog. Histologic sections demonstrated thermal injury to the physis followed by a progressive narrowing and eventual resorption of the growth plate.     

Periosteal entrapment in distal femoral physeal fractures: harbinger for premature physeal arrest ?

We report on two patients who sustained Salter-Harris II fractures of the distal femur with physeal widening after being tackled in football games. Preoperative MRI indicated entrapped periosteum at the physeal fracture site for both patients. Both patients underwent open reduction of the physeal fracture with removal of the entrapped periosteum and achieving an anatomic reduction. Follow-up MRI's revealed premature physeal arrest. Subsequent procedures were performed to address sequelae of premature physeal arrest. The presence of physeal widening and entrapped periosteum may reflect high-energy trauma to the physis. This can result in injury to both the epiphyseal blood supply and to the physeal cartilage (germinal zone) resulting in physeal arrest despite anatomic reduction after removal of the entrapped periosteum. Upon literature review, pre-operative MRI demonstrating entrapped periosteum has not been previously reported. We hypothesize that the presence of entrapped periosteum following distal femoral physeal fractures may be associated with an increased risk for premature physeal arrest.     

Traumatic growth arrest of the distal tibia: a clinical and radiographic review.

BACKGROUND: Injury to the physis of the distal tibia in children can lead to subsequent growth arrest. This can result in physeal bars, leg-length discrepancies and angular deformities. METHOD: The cases of 12 children with distal tibial growth arrest due to ankle trauma were reviewed. All were treated at a tertiary care institution between 1990 and 2002. RESULTS: The average age at initial injury was 9.7 years (range 5-13 yr). Salter-Harris classifications (SH) of their injuries were SH2 in 4 children, SH3 in 1, SH4 in 6 and SH5 in 1 child. Four involved open fractures; 5 were high-energy injuries. Six of the injuries resulted in simple physeal bars, 4 caused pure angular deformities and 2 resulted in leg-length discrepancies > 2 cm. Eight of the physeal arrests were treated either with bar excision, selective epiphysiodesis or osteotomy for angular correction. CONCLUSION: This series reinforces the importance of frequent follow-up of distal tibial physeal injuries in order to detect growth arrest early, thus facilitating corrective surgery.     

Effects of a tensioned tendon graft in a bone tunnel across the rabbit physis

Children who sustain anterior cruciate disruption often are denied the standard reconstructive procedures because of the concern that drilling across the physis of the tibia and femur and compression from a tensioned graft will result in growth plate arrest. To test this concept and to assess whether a tendon placed in the tunnel would function in a manner similar to a fat graft after the resection of a physeal bar, tunnels were made across the proximal tibial physis and distal femoral physis in a group of immature rabbits. Four tunnel diameters were used from 1.95 to 3.97 mm, in three rabbits at each diameter, with patellar tendon autografts being used as the reconstruction of the anterior cruciate ligament in two of the animals. The knees were radiographed every 4 weeks, and the animals were euthanized 4 months after surgery. The surgically treated and control knees were salvaged, and each knee was examined grossly, radiographically, and histologically. Eight of the 11 animals had growth arrest of one or both physes. The larger the drill hole diameter the more marked was the deformity. The proximal tibial physis seemed to be the most vulnerable for growth arrest, occurring in eight of the knees. The insertion of a tendon did not seem to offer any protection to physeal arrest. Because of these findings, it is not recommended that tunnels involving 1% or more of the area of the physis be placed across the tibial and femoral physis to reconstruct the anterior cruciate in very skeletally immature children.     

Physeal fractures of the distal radius and ulna: long-term prognosis

OBJECTIVES: The long-term prognosis of injuries to the distal physis of forearm bones, including complications such as radioulnar length discrepancy and styloid nonunion, has not been extensively studied. Reliable radiographic prognostic criteria to predict physeal disturbance at trauma are also lacking. The aim of this study is to investigate both issues. DESIGN: Retrospective study. SETTING: University hospital. PATIENTS/PARTICIPANTS: One hundred sixty-three lesions to the distal physis of the forearm bones in 157 patients were available for a long-term follow-up. Seventy-seven injuries of the distal radial physis were radiographically isolated, 54 were associated with a fracture of the ulnar styloid, and 26 with a fracture of the distal ulnar metaphysis. Of the six injuries of the distal ulnar physis, five were associated with a fracture of the distal radial metaphysis, and one was an isolated injury of the distal ulnar physis. INTERVENTION: Treatment consisted of wrist immobilization in a long-arm plaster cast for 6 weeks. Dorsal or volar displacement was reduced using general anesthesia. MAIN OUTCOME MEASUREMENTS: All patients had both clinical and radiographic evaluation, with an average follow-up of 25.5 years (range 14-46 years). The average age of the patients at injury was 11.6 years (range 5-17 years), whereas their average age at follow-up was 35.5 years (range 22-56 years). Both the Salter and Harris and the Ogden classifications were used to classify physeal injuries. RESULTS: According to Salter and Harris, of the 157 radial lesions, 18 were type 1 and 139 type 2. According to Ogden, 14 were type 1A, 4 type 1C, 84 type 2A, 13 type 2B, 17 type 2C, and 25 type 2D. Of the 6 ulnar lesions, 2 were Salter and Harris type 1 (Ogden type 1A), 3 type 2 (Ogden type 2A), and 1 type 4 (Ogden type 4A). Fifty-four radiographically evident fractures of the ulnar styloid associated with injuries of the distal radial physis were classified as Ogden type 7A. At follow-up, all of our patients were fully asymptomatic, except for those who had forearm bone growth failure of more than 1 cm. Shortening of the previously injured forearm bones ranging from 1 to 6.5 cm was observed in 2 open and subsequently infected lesions as well as in 5 uncomplicated lesions of the 157 distal radial physeal injuries (4.4%), and in 3 of the 6 distal ulnar physeal injuries (50%). Shortening of 1 cm or more was observed in the uncomplicated lesions of radial physeal injury with Ogden type 1C, 2B, and 2D lesions, and in ulnar physeal injuries Ogden type 1A, 2A, and 4A. Thirty-eight additional patients had radioulnar length discrepancy that ranged from 2 to 9 mm, and 53 patients had styloid nonunion, but all of them were asymptomatic. CONCLUSIONS: None of the patients reviewed at follow-up, including those with radioulnar length discrepancy of less than 1 cm and those with styloid nonunion, complained of any symptom related to their previous injury, not even those engaged in heavy manual labor. Of the 10 patients with either radial or ulnar shortening of more than 1 cm, only 2 with radial growth arrest and marked radioulnar length discrepancy had severe functional problems. Growth disturbances of more than 1 cm following distal radial physeal injury occurred only in Ogden type 1C, 2B, and 2D lesions, whereas in distal ulnar physeal injuries, growth disturbances occurred regardless of the Ogden classification type.     

Distal tibial physeal arrest after meningococcal septicaemia: management and outcome in seven ankles

Survivors of infantile meningococcal septicaemia often develop progressive skeletal deformity as a result of physeal damage at many sites, particularly in the lower limb. Distal tibial physeal arrest typically occurs with sparing of the distal fibular physis leading to a rapidly progressive varus deformity. There have been reports of isolated cases of this deformity, but to our knowledge there have been no papers which specifically describe the development of the deformity and the options for treatment. Surgery to correct this deformity is complex because of the patient's age, previous scarring and the multiplanar nature of the deformity. The surgical goal is to restore leg-length equality and the mechanical axis at the end of growth. Surgery should be planned and staged throughout growth in order to achieve the best functional results. We report our experience in six patients (seven ankles) with this deformity, who were managed by corrective osteotomy using a programmable circular fixator.     

The early identification and classification of growth disturbances of the proximal end of the femur

An analysis of the serial radiographs of sixty-eight patients who were treated for congenital dislocation of the hip revealed that a growth disturbance of the proximal end of the femur that was caused by partial or complete physeal closure developed in thirty-three of these patients. The disturbances were related to the character of the metaphyseal growth-disturbance lines and were subsequently classified according to the site and extent of physeal closure. In order to study this problem, we divided the physis of the proximal end of the femur into two contiguous sections: a medial and a lateral portion. Two typical patterns of premature closure were identified: one located at the junction of the medial and lateral portions of the physis and the other located in the area of the medial portion of the physis alone. The pattern of physeal closure, together with the age of the patient at the time of closure, determines the subsequent growth of the proximal end of the femur. The final outcome of growth of the proximal end of the femur can be predicted within six months after the initial treatment. Epiphyseal changes were found to be of no prognostic significance in the absence of physeal closure.     

Correction of severe wrist deformity following physeal arrest of the distal radius with the aid of a three-dimensional computer simulation

Growth arrest following physeal injury may result in severe limb deformity. We report a case of complex wrist deformity caused by injury to the distal radial physis resulting in radial shortening and abnormal inclination of the radial articular surface, which was successfully treated by gradual correction after computer simulation. The simulation enabled us to develop an appropriate operative plan by accurately calculating the axis of the three-dimensional (3D) deformity using computer bone models. In the simulative surgery with a full-size stereolithography bone model, an Ilizarov external fixator was applied to the radius such that its two hinges were located on the virtual axis of the deformity, which was reproduced in the actual surgery. This technique of 3D computer simulation is a useful alternative to plan accurate correction of complex limb deformities following growth arrest.     

Growth disturbance of the proximal part of the femur after treatment for congenital dislocation of the hip

The radiographs of ninety patients in whom treatment of unilateral congenital dislocation of the hip was complicated by disturbance of growth of the proximal part of the femur were studied retrospectively. All patients were followed until closure of the affected proximal femoral physis. We divided the patients into three groups, according to the degree of vascular insufficiency: patients who had mild vascular insufficiency of the hip, which had little effect on growth; those who had moderate vascular insufficiency, which produces partial arrest of growth; and those who had severe vascular insufficiency, which causes complete arrest of growth. Good correlation was found between the initial degree of vascular insufficiency and the radiographic results at the most recent follow-up. The radiographic signs that were used to predict the extent of physeal involvement were a crescent-shaped epiphysis, medial bowing of the femoral neck (a shorter and more concave curve between the lesser trochanter and the proximal femoral metaphysis [the lateral portion of the Shenton line]), lateral tilting of the capital epiphysis, and premature physeal closure. Signs that were diagnostic of existing physeal involvement were elevation of the greater trochanter and shortening of the affected extremity. The presence and severity of these signs correlated well with the degree of vascular insufficiency. Medial bowing was the most reliable prognostic factor for the determination of the fate of the hip joint at maturity.