Introduction
Legg-Calvé-Perthes disease (LCPD) is an idiopathic avascular necrosis of the capital femoral epiphysis in children.[1] This condition most commonly affects children between 4 and 10 years of age and shows a marked male predominance. Patients usually present with a limp, hip or groin pain, or referred knee pain, accompanied by restriction of hip motion, particularly abduction and internal rotation.[1][2]
The condition was independently described in 1910 by Legg, Calvé, and Perthes.[2] The pathologic process begins with interruption of the blood supply to the femoral head, followed by avascular necrosis, resorption, and reossification. The femoral head receives its primary perfusion from the lateral epiphyseal vessels, branches of the medial femoral circumflex artery. Disruption of these vessels leads to ischemia of the capital femoral epiphysis and potential deformation of the femoral head if containment in the acetabulum is not maintained.[3][4] The overarching goal of management is to allow the femoral head to heal in a spherical, congruent shape within the acetabulum, thereby minimizing the risk of femoroacetabular impingement and early osteoarthritis.[4][5]
Etiology
Register For Free And Read The Full Article
Search engine and full access to all medical articles
10 free questions in your specialty
Free CME/CE Activities
Free daily question in your email
Save favorite articles to your dashboard
Emails offering discounts
Learn more about a Subscription to StatPearls Point-of-Care
Etiology
The etiology of LCPD remains incompletely understood and is likely multifactorial.[6] The unifying pathologic event is interruption of blood supply to the capital femoral epiphysis, but the initiating mechanisms vary. Current models propose that genetic susceptibility, adverse environmental exposures, and local mechanical loading interact to produce an ischemic event in the capital femoral epiphysis of a vulnerable child.[7]
Factors
Vascular and coagulation Abnormalities in the vascular anatomy or function of the lateral epiphyseal vessels and venous outflow from the femoral head are implicated. Venous congestion, intraosseous hypertension, and arterial occlusion may predispose to ischemia. Several studies report a higher prevalence of thrombophilic abnormalities (eg, factor V Leiden, protein C or S deficiency, elevated lipoprotein[a]) in children with LCPD compared with controls, suggesting that a hypercoagulable state may contribute to epiphyseal infarction in susceptible patients.[7]
Mechanical and traumatic Repetitive microtrauma or an acute injury may precipitate vascular compromise in a biomechanically vulnerable hip. Children with LCPD often have increased activity levels or participate in impact sports, but trauma alone does not explain the disease and likely acts as a secondary trigger rather than a primary cause.[7]
Environmental exposures and systemic Exposure to tobacco smoke (maternal smoking during pregnancy or secondhand exposure in childhood) correlates with increased risk, possibly through effects on vascular development, endothelial function, or coagulation.[8] Lower socioeconomic status, urban residence, and nutritional deprivation are also associated but may serve as proxies for other environmental and health factors. Many affected children have short stature or delayed skeletal maturation. Lower circulating levels of insulin-like growth factor 1 and other growth factors have been reported, which may result in a smaller, more vulnerable epiphysis and altered endochondral ossification.[7]
Genetic and molecular Most cases appear sporadic, but familial clustering and an 8% to 12% prevalence of positive family history suggest a heritable component.[1][8] Mutations in COL2A1 and other genes involved in cartilage matrix integrity, angiogenesis, and coagulation have been reported in small series. Current molecular and cellular research implicates endothelial cell dysfunction, aberrant osteoclast activation, altered inflammatory signaling, and disturbed bone–cartilage cross-talk in the pathogenesis of femoral head necrosis.
Epidemiology
LCPD is rare but shows geographic, racial, and temporal variation. Reported incidence ranges from 0.4 to 29 per 100,000 children under 15 years of age.[1] Most series report onset between 3 and 12 years, with a peak around 5 to 7 years and a mean age at diagnosis of 6 to 7 years.[1][8]
Boys are affected approximately 3 to 5 times more often than girls.[9] Bilateral involvement occurs in 10% to 24% of patients, though the hips may be affected asynchronously.[8][10] Recent registry and national database studies confirm these patterns and suggest that incidence has declined in some regions over recent decades, possibly related to changes in smoking prevalence, environmental exposures, and socioeconomic conditions.[1]
LCPD is more common in children of European descent and in populations at higher latitudes and in urban environments, and it is relatively uncommon in children of African or East Asian ancestry.[1] Lower socioeconomic status, parental smoking, and crowding correlate with increased risk.[8] Obesity has also been reported as a risk factor in some contemporary cohorts. A positive family history is present in a minority (approximately 8% to 12%), indicating a heritable predisposition in some cases.[1]
Pathophysiology
LCPD progresses through a predictable sequence of femoral head epiphyseal ischemia, necrosis, resorption, and repair. Waldenström described four radiographic stages that remain clinically useful and correlate with underlying pathology.[3]
Initial (necrotic) stage
The loss of blood supply to part or all of the capital femoral epiphysis results in death of bone and marrow cells. The articular cartilage remains viable due to synovial fluid nutrition. Radiographs may show increased epiphyseal density (sclerosis) and subtle loss of epiphyseal height. A subchondral lucent line (the “crescent sign”) indicates a subchondral fracture through necrotic bone and heralds impending collapse.[4]
Fragmentation stage
Revascularization begins at the periphery of the epiphysis. Osteoclasts resorb necrotic bone, and the epiphysis appears fragmented, mottled, and flattened on radiographs. The lateral pillar may lose height. During this phase, the femoral head is biomechanically weakest and most susceptible to deformation under axial load and muscle forces. Containment in the acetabulum is critical because lateral extrusion promotes aspheric collapse.[3]
Reossification (healing) stage
New bone progressively replaces necrotic bone from the periphery inward. The femoral head gradually reossifies, and radiographic density normalizes. The final shape of the femoral head and congruency with the acetabulum are largely determined during this phase. If containment is maintained and the head remains centered, the likelihood of a spherical outcome increases.[3]
Healed (residual) stage
The disease process is complete, and the femoral head is fully reossified, typically 2 to 4 years after onset. The residual morphology may be normal or show deformities such as coxa magna, coxa plana, or aspherical enlargement. The acetabulum may remodel to accommodate the altered head shape. Residual deformity is graded at skeletal maturity using the Stulberg classification, which correlates with increased risk of osteoarthritis.[11] Biomechanically, an aspherical or laterally extruded femoral head results in abnormal joint loading, focal cartilage stress, femoroacetabular impingement (including hinge abduction), and early degenerative changes.[12]
Histopathology
Histologic findings vary with disease stage and reflect the underlying vascular insult and subsequent repair. However, histopathologic analysis is not required for diagnosis and is mainly described in experimental and historical studies.[4][13]
- Early ischemic phase: Empty lacunae, loss of osteocytes, and absence of hematopoietic cells in the marrow indicate bone and marrow necrosis. Trabeculae retain gross architecture but lack viable cells.
- Resorptive (fragmentation) phase: Osteoclast-mediated resorption of necrotic trabeculae occurs with ingrowth of fibrovascular granulation tissue. The epiphysis becomes structurally weak due to trabecular loss.
- Repair (reossification) phase: Woven bone is deposited on remaining trabeculae and within the fibrovascular tissue. Over time, this woven bone remodels into lamellar bone with restoration of normal marrow elements.
- Residual stage: The healed femoral head shows mature lamellar bone with variable deformity. Areas of cartilage and bone overgrowth, abnormal physeal architecture, and subchondral sclerosis reflect the earlier insult.
History and Physical
LCPD often presents insidiously. Parents typically report a painless or minimally painful limp in a child aged 4 to 10 years. The limp may worsen with activity and improve with rest. Pain usually localizes to the hip or groin and often refers to the knee, leading to misdirected evaluation.[1][8] Systemic signs (fever, erythema, elevated inflammatory markers) are absent and should prompt evaluation for septic arthritis, osteomyelitis, or inflammatory arthropathies.
History
- Gradual onset of limp development over weeks to months
- Activity-related hip, groin, thigh, or knee pain; sometimes only knee pain
- Morning stiffness or end-of-day fatigue in the affected limb
- Occasional night pain in more advanced disease
- No systemic symptoms; children are afebrile and otherwise well
Physical examination
- Antalgic or Trendelenburg gait
- Limited hip range of motion, especially abduction and internal rotation; flexion contracture and loss of full extension may develop
- Pain at extremes of motion, particularly abduction and internal rotation
- Relative leg length discrepancy due to femoral head collapse and growth disturbance
- Thigh and gluteal muscle atrophy in chronic cases
- In bilateral disease (15%–20%), findings may be symmetric or sequential; can be subtle early [14]
Evaluation
Diagnosis is primarily clinico-radiographic. The goals of evaluation are to confirm LCPD, exclude other causes of hip pain, stage the disease, and assess severity and containment.
Plain Radiographs
An anteroposterior pelvis radiograph and frog-leg lateral view are standard. Radiographic findings depend on the stage:
- Early/initial stage: subtle increased density of the capital femoral epiphysis, small epiphysis, or joint effusion; films may be normal.
- Fragmentation: patchy sclerosis and lucency, fragmentation, reduced epiphyseal height, lateral pillar collapse, and widening of the joint space due to cartilage hypertrophy and effusion. The crescent sign indicates a subchondral fracture.
- Reossification: gradual filling in of lucent areas, with reconstitution of trabecular patterns.
- Residual stage: final femoral head and acetabular morphology, including coxa magna, coxa plana, or asphericity.[4][15]
Radiographs also allow application of Catterall, Salter–Thompson, and Herring lateral pillar classifications, as well as evaluation of femoral head coverage and extrusion.
Magnetic Resonance Imaging
MRI is more sensitive than radiography in early disease and can detect marrow changes and loss of perfusion before radiographic abnormalities appear.[15] This imaging is particularly useful when:
- Radiographs are normal or equivocal, but clinical suspicion remains high
- Early assessment of the extent of necrosis is needed to guide treatment decisions
- Differentiation from transient synovitis, early slipped capital femoral epiphysis, or infection is required
Gadolinium-enhanced and perfusion MRI can quantify the proportion of the epiphysis that remains perfused. A smaller perfused volume correlates with more severe subsequent lateral pillar collapse and poorer prognosis.[8][15]
Other Imaging
Ultrasound can identify joint effusion in acute presentations, but does not diagnose LCPD. Bone scintigraphy is largely historical, having been supplanted by MRI. Computed tomography has a limited role in routine imaging but may aid in assessing complex residual deformity and preoperative planning in late disease.[8][15]
Laboratory Studies
Routine blood tests are normal in LCPD and are primarily obtained to exclude infection or inflammatory disease. Coagulation and metabolic workups may be considered in atypical cases (very young or older than usual age, bilateral disease, recurrent episodes, or family clustering), but no specific laboratory test confirms LCPD.[7]
Treatment / Management
Principles of Management
LCPD is self-limited, as the femoral head eventually revascularizes and heals. The critical determinant of long-term outcome is the sphericity of the femoral head and congruency with the acetabulum at skeletal maturity. Management aims to:
- Maintain or restore containment of the femoral head within the acetabulum
- Preserve hip range of motion, especially abduction and internal rotation
- Minimize femoral head deformation during the fragmentation and reossification stages
- Relieve pain and maintain functional mobility [3][11][16]
Treatment decisions are guided by age at onset, disease stage, extent of femoral head involvement (eg, Herring lateral pillar group), degree of extrusion, and hip range of motion.[3][11][16]
Broadly:
- Children younger than 6 years, especially with mild disease (Herring A, some B), usually do well with nonoperative management.
- Children older than 8 years and those with more extensive involvement (Herring B/B-C or C, Catterall III–IV) are at higher risk for poor outcomes and are more likely to benefit from surgical containment if adequate motion is preserved.[11][16][17] (A1)
Nonoperative Management
Activity modification and analgesia
Initial treatment in most patients includes:
- Avoidance of high-impact activities and prolonged running or jumping
- Relative weight-bearing restriction as needed, using crutches or a walker during painful phases
- Short courses of nonsteroidal anti-inflammatory drugs (eg, ibuprofen) to reduce pain and synovitis
- Short-term bed rest or in-hospital skin traction may be used for severe pain and muscle spasm, particularly when hip range of motion is markedly restricted [1] (B3)
Physical therapy
Physiotherapy is central to nonoperative care. Daily, structured exercises focus on:
- Maintaining or restoring abduction and internal rotation
- Stretching hip adductors, flexors, and internal rotators
- Strengthening abductors and core musculature
Hydrotherapy allows gentle motion with minimal joint loading. Home exercise programs, supported by periodic supervised sessions, help preserve motion and reduce the risk of fixed contractures. Examination under anesthesia and arthrography may be used when motion is limited to determine whether the hip remains containable.[1][18](B3)
Containment by casting or bracing
Nonoperative containment strategies aim to keep the femoral head well seated in the acetabulum in abduction:
- Petrie casts: Bilateral long-leg casts in abduction with a connecting bar; often used after adductor tenotomy to regain motion and containment
- Abduction orthoses (eg, Atlanta Scottish Rite orthosis): Maintain hips in abduction while permitting ambulation
These methods are most commonly used in younger children with moderate disease severity who have adequate motion but high-risk features (eg, Herring B). Duration often spans months to portions of the fragmentation and early reossification stages.[18] Evidence for bracing is mixed; some series report high rates of congruent hips with diligent use of wide-abduction orthoses, whereas systematic reviews and meta-analyses show no consistent advantage of bracing over active physiotherapy and activity modification, especially in children younger than 6 to 8 years with mild disease.[11][18] Compliance is often challenging and must be weighed against potential benefits.
Operative Management:
Surgery is typically considered in:
- Children older than 8 years at onset
- Patients with Herring B/B-C or C, Catterall III–IV, or extensive necrosis on MRI
- Hips with progressive lateral extrusion or loss of containment
- Patients with persistent loss of abduction and internal rotation despite intensive therapy
The main goal is containment: increasing acetabular coverage of the femoral head and centralizing it in the socket.[11][16]
Femoral varus osteotomy
Proximal femoral varus osteotomy (FVO) reorients the femoral head deeper into the acetabulum. This procedure is usually performed at the intertrochanteric or subtrochanteric level and fixed with a plate. Benefits include improved containment, reduced hinge abduction, and improved abduction range. Downsides include limb shortening, increased limp, and potential over-varus.[16]
Short- to mid-term studies suggest that FVO improves radiographic head sphericity and functional outcomes in appropriately selected patients, particularly those older than 6 to 8 years with Herring B/B-C disease and preserved motion.[16] Age and lateral pillar classification strongly influence outcomes, with younger age and less severe lateral pillar collapse correlating with better outcomes.
Pelvic osteotomy
Pelvic procedures reorient or augment the acetabulum to improve coverage:
- Salter innominate osteotomy: Single-cut redirectional osteotomy of the pelvis to increase anterolateral coverage.
- Pemberton or Dega acetabuloplasty: Reshaping osteotomies that hinge the acetabular roof to increase coverage.
- Shelf acetabuloplasty: Bone graft added to the lateral acetabular rim without cutting the pelvis, augmenting coverage.
Pelvic osteotomies are favored in children around 6 to 11 years with good hip motion and moderate disease, often when the femoral neck is already relatively short or valgus, and further varus is undesirable.[19][20] Comparative studies suggest that femoral and innominate osteotomies yield broadly similar radiographic and functional outcomes, so surgeon experience and patient anatomy often guide the choice.[11][19]
Combined femoral and pelvic osteotomies
In very severe cases (eg, Herring C with significant extrusion), some centers perform combined femoral and pelvic osteotomies to maximize containment. These procedures increase complexity and morbidity and are reserved for select patients.[16][19]
Valgus femoral osteotomy and Chiari osteotomy (salvage procedures)
In late disease with healed or near-healed deformity and symptomatic hinge abduction, containment of the entire head is no longer feasible. Salvage options aim to improve congruency between a remaining viable portion of the head and the acetabulum:
- Valgus extension femoral osteotomy: reorients the more intact portion of the femoral head into the weight-bearing zone, relieving lateral impingement and improving range of motion.
- Chiari medial displacement osteotomy: shifts the acetabulum medially over a deformed femoral head to increase coverage and improve load distribution.
These procedures are considered in older children and adolescents with painful, stiff hips and significant deformity.[19]
Hip arthrodiastasis (joint distraction)
Arthrodiastasis uses a hinged external fixator to distract the hip joint slightly while allowing controlled motion. The concept is to unload the femoral head, promote cartilage recovery, and maintain centralization. Systematic reviews suggest that, in older children and those with severe disease or early hinge abduction, articulated distraction can improve range of motion and pain, with many hips achieving Stulberg II–III outcomes.[19] However, the evidence base remains limited, the technique is resource-intensive, and pin-tract infections are common; it is best regarded as a niche strategy in specialized centers.
Experimental biological approaches
Experimental studies have explored core decompression and biologic augmentation, including bone marrow–derived mesenchymal stem cell transplantation, to enhance revascularization and remodeling. Early preclinical and small clinical studies suggest potential benefit but remain investigational, without robust evidence to support routine use.
Management of Residual and Adult Sequelae
Residual deformity after healed LCPD can produce femoroacetabular impingement, labral tears, chondral damage, and acetabular dysplasia.[12][21]
Options include:
- Hip arthroscopy or surgical hip dislocation with osteochondroplasty to address femoral head–neck offset abnormalities and labral pathology in skeletally mature individuals
- Trochanteric epiphysiodesis or distal trochanteric transfer to address abductor insufficiency from trochanteric overgrowth (often after varus osteotomy)
- Periacetabular osteotomy for symptomatic acetabular dysplasia.
- Total hip arthroplasty (THA) for advanced degenerative osteoarthritis
Differential Diagnosis
The differential diagnosis of Legg-Calvé-Perthes disease includes conditions that present with hip pain or a limp in children, such as slipped capital femoral epiphysis, transient synovitis, septic arthritis, or juvenile idiopathic arthritis (see Table. Differential Diagnosis of Legg-Calvé-Perthes Disease). Imaging and clinical features, including age at onset, pattern of femoral head involvement, and presence of systemic symptoms, help distinguish Perthes disease from these alternative diagnoses.
Table. Differential Diagnosis of Legg-Calvé-Perthes Disease
|
Septic arthritis of the hip |
Any age, often <5–6 years; acute onset | Acute, severe hip or groin pain; child often refuses to bear weight; fever, systemic illness; marked irritability with any motion | Elevated ESR/CRP, leukocytosis; ultrasound: joint effusion; MRI: synovitis, possible adjacent osteomyelitis; joint space widening; rapidly progressive destructive changes if untreated | Think sepsis if pain is sudden, severe, with fever and inability to bear weight; labs and ultrasound/MRI are essential LCPD is more insidious, afebrile, and chronic |
| Osteomyelitis (proximal femur or pelvis) | Childhood; can be any age | Focal bone pain, fever, systemic symptoms; may mimic hip pain/limp but often more localized tenderness over metaphysis or pelvis; frequently more acute than LCPD | MRI: marrow edema, abscess; plain films may show periosteal reaction or lytic lesions later; labs: elevated ESR/CRP, WBC | Consider osteomyelitis when local bony tenderness and systemic signs are prominent; MRI and labs clarify; LCPD shows epiphyseal AVN without systemic inflammatory picture |
| Transient synovitis of the hip | 3–8 years; often younger LCPD age range | Acute or subacute limp, mild–moderate hip pain; may follow viral illness; afebrile or low-grade fever; symptoms improve over days–1–2 weeks | Ultrasound: joint effusion; plain x-ray typically normal; MRI: no epiphyseal AVN, may show synovitis only; labs usually normal or minimally elevated | Time course is key: transient synovitis resolves with rest/NSAIDs; persistent or recurrent limp or pain beyond 2–3 weeks should prompt reevaluation for LCPD |
| Slipped capital femoral epiphysis (SCFE) | Typically 10–16 years; overweight adolescents; rare under 9 | Hip, thigh, or knee pain; limp; decreased internal rotation; obligate external rotation with hip flexion; often gradual, but can be acute slip | AP + frog-lateral x-rays: posterior–inferior displacement of epiphysis (Klein line, Trethowan sign); no AVN initially (unless unstable and complicated) | Age and body habitus are key: SCFE is an adolescent disorder; LCPD usually 4–10 years Radiographs show epiphyseal slip rather than fragmentation and collapse |
| Developmental dysplasia of the hip | Infancy; sometimes detected later in toddlers | Painless limp, leg length discrepancy, limited abduction; positive Ortolani/Barlow in infants; in older child, Trendelenburg gait; often no acute pain | US in infants: abnormal alpha angle, dislocated/subluxed hip; x-ray in older child: shallow acetabulum, lateralized femoral head, increased acetabular index; no AVN pattern of LCPD | DDH typically presents earlier, often painless; imaging shows dysplastic acetabulum and subluxation, not epiphyseal necrosis and fragmentation |
| Multiple epiphyseal dysplasia | Late childhood to adolescence | Bilateral, often symmetric joint pain; involvement of multiple joints (hips, knees, ankles); short stature may be mild; family history often positive | X-rays: irregular, small, flattened epiphyses in multiple joints; often bilateral hips; no typical unilateral, staged AVN pattern of LCPD | Think MED when multiple joints and both hips are involved and there is a family history; LCPD is usually unilateral and isolated to hip. |
| Spondyloepiphyseal dysplasia | Infancy/childhood | Disproportionate short-trunk dwarfism, spinal deformity (kyphosis, scoliosis), early hip pain; often bilateral hip involvement | Skeletal survey: flattened vertebral bodies (platyspondyly), abnormal epiphyses in multiple joints; bilateral hip changes; no classic Waldenström stages | Systemic skeletal findings and spinal changes distinguish SED; LCPD is localized hip AVN without generalized dysplasia |
| Other skeletal dysplasias | Variable, often early childhood | Short stature, limb deformities, multiple joint involvement; may have characteristic facies or other systemic findings | Skeletal survey shows generalized pattern (eg, metaphyseal or epiphyseal changes in many bones); hips not affected in isolation | Broad category; the presence of generalizedskeletal abnormalities, not an isolated hip process, differentiates these from LCPD |
| Sickle cell disease with femoral head osteonecrosis | School age to adolescence; patients with known sickle cell disease | Hip pain, limp; often bilateral; may have history of vaso-occlusive crises, anemia, jaundice | X-rays/MRI: femoral head osteonecrosis but usually bilateral; other sites (humeral head) may also be involved; lab evidence of hemolytic anemia; positive hemoglobin electrophoresis | Consider when osteonecrosis is bilateral in a child with known sickle cell disease; LCPD is usually unilateral and in otherwise healthy children |
| Gaucher disease | Childhood to adulthood | Bone pain, pathologic fractures, hepatosplenomegaly, cytopenias; hip pain may be presenting complaint | Imaging: medullary expansion, Erlenmeyer flask deformity of femur, areas of osteonecrosis; labs: low glucocerebrosidase activity, genetic testing positive | Systemic signs (organomegaly, cytopenias) and characteristic long-bone changes make Gaucher distinct from isolated LCPD |
| Endocrinopathies (eg, hypothyroidism) | Variable; often school age or adolescence | Short stature, delayed growth and puberty, fatigue, weight gain; hip pain may be part of broader musculoskeletal complaints; may coexist with SCFE | Delayed bone age; labs: abnormal TSH/T4 or other endocrine markers; imaging may show generalized delayed ossification, possible SCFE rather than LCPD | Think endocrine when growth failure and multiple systemic symptoms are present; LCPD patients often small for age but otherwise systemically well |
| Meyer dysplasia (benign developmental epiphyseal variant) | 2 to 6 years | Often asymptomatic or mild limp; usually incidental finding; minimal or no pain; normal general health | X-rays: irregular, fragmented appearing ossification center of femoral head, but normal contour, no collapse; changes are symmetric or bilateral and resolve spontaneously over time | Key mimic in younger children: time and follow-up are critical. Meyer dysplasia shows benign, self-limited ossification irregularity without the staged collapse and remodeling typical of LCPD |
AVN, avascular necrosis; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; LCPD, Legg-Calvé-Perthes disease; MED, medial; NSAID, nonsteroidal anti-inflammatory drugs; SCFE, slipped capital femoral epiphysis; SR/CRP, sedimentation rate/C-reactive protein; TSH, thyroid-stimulating hormone; WBC, white blood cell count
Staging
Several classification systems help stage disease, quantify severity, guide management, and predict prognosis.[5][11][22] In clinical practice, age at onset, combined with the Herring lateral pillar group, forms the basis of most contemporary treatment algorithms.[5][22]
Herring Lateral Pillar Classification
The lateral pillar (outer third) of the capital femoral epiphysis is critical for maintaining head shape. Herring classified hips during fragmentation based on lateral pillar height:
- Group A: No loss of lateral pillar height
- Group B: <50% loss of lateral pillar height
- Group C: >50% loss of lateral pillar height [5]
A border B/C (“B50”) category has been added with some modifications. The Herring classification is widely used due to relatively good reliability and strong correlation with outcome: Group A hips almost always have good outcomes, Group B have intermediate outcomes, and Group C have a high risk of severe deformity.[5][22]
Waldenström Radiographic Stages
Chronologic staging into initial (necrotic), fragmentation, reossification, and healed stages, as described under Pathophysiology. This system is useful for timing interventions; containment strategies are most effective when initiated before or early in the fragmentation process.
Catterall Classification
A 4-group system based on the extent of epiphyseal involvement on radiographs:
- Group I: <25% of the head involved
- Group II: 25%–50%
- Group III: >50%
- Group IV: almost the entire head
Catterall also described “head-at-risk” signs (eg, lateral calcification, Gage sign, horizontal physis, lateral subluxation) that indicate a poorer prognosis. Limitations include interobserver variability and reduced reliability when used alone.
Salter–Thompson Classification
A 2-group system based on the extent of the subchondral fracture line (crescent sign) on early AP radiographs:
- Group A: Fracture involves <50% of the head
- Group B: Fracture involves >50%
This classification provides an early estimate of the extent of necrosis but can only be applied when a clear subchondral fracture is present.
Stulberg Classification
Applied at skeletal maturity, this system describes residual morphology and congruency:
- Class I: normal spherical head and acetabulum
- Class II: spherical head with minor abnormalities (eg, mild coxa magna)
- Class III: aspherical but congruent head and acetabulum (ovoid joint)
- Class IV: flat, aspherical head with shallow, deformed acetabulum; congruent but markedly abnormal
- Class V: flat, aspherical head with normal acetabulum; incongruent joint
Stulberg I–II corresponds to good, III to fair, and IV–V to poor outcomes, and correlates strongly with the risk of early osteoarthritis and the need for total hip arthroplasty.[11]
Prognosis
LCPD is self-healing but not benign. The long-term outcome depends on age at onset, extent of epiphyseal involvement, containment during fragmentation, and residual morphology.[1][5][11]
Key Prognostic Factors
- Age at onset: Children younger than 6 years, especially with limited involvement, have a high likelihood of near-normal head shape and favorable long-term outcomes. Onset after age 8 to 9 years carries a significantly higher risk of residual deformity and early osteoarthritis. Very late onset (>11–12 years) often behaves more aggressively with limited remodeling capacity.[5][11]
- Extent of necrosis and lateral pillar collapse: Greater epiphyseal involvement (Catterall III–IV, large necrotic volume on MRI) and Herring C lateral pillar collapse predict poor sphericity at healing. Herring C hips frequently progress to Stulberg IV–V even with treatment.[5][11]
- Femoral head extrusion: Lateral subluxation and loss of acetabular coverage correlate with worse outcomes. Hips that maintain good containment (>80% coverage) tend to remodel more favorably.[5][11]
- Hip range of motion: Maintaining abduction and internal rotation is associated with better radiographic and functional outcomes. Fixed contractures and hinge abduction portend poorer outcomes.[3]
- Sex and bilaterality: Girls, particularly older girls, often have worse outcomes than boys, though they represent a minority of cases. Bilateral disease may reflect greater systemic vulnerability and can complicate management.[1]
Natural History and Long-Term Outcomes
Results from long-term cohort studies of largely nonoperatively treated patients show that 60% to 80% of hips with Stulberg I–II morphology remain asymptomatic or mildly symptomatic into mid-adulthood, whereas hips with Stulberg III–V have a substantially increased risk of pain, reduced function, and radiographic osteoarthritis by the fourth or fifth decade of life.[11] Appropriate age- and severity-based treatment can improve femoral head sphericity in high-risk patients and may reduce the likelihood or delay the timing of major reconstructive procedures.[5][11] Many patients with severe residual deformity ultimately require total hip replacement in early to mid-adulthood. While modern arthroplasty offers good pain relief and function, the technical demands and complication profile are less favorable than in primary osteoarthritis without deformity.[12]
Complications
Complications arise both from the disease process and from its treatment.[3][12] As a result, long-term surveillance into adolescence and early adulthood is often warranted for patients with significant deformity or prior reconstructive surgery.
Disease-Related Complications
- Coxa magna (enlarged femoral head)
- Coxa plana (flattened head)
- Aspherical femoral head and shortened femoral neck with reduced offset
- Lateral subluxation or extrusion of the femoral head
- Premature physeal closure causing limb-length discrepancy
- Acetabular dysplasia or remodeling with a shallow, widened socket
- Femoroacetabular impingement (cam-type), labral tears, and cartilage damage
- Early degenerative osteoarthritis and need for hip preservation surgery or total hip arthroplasty
Treatment-Related Complications
- Nonunion or malunion of osteotomies
- Over- or under-correction (excess varus or valgus) with residual deformity or limb-length discrepancy
- Abductor weakness, Trendelenburg gait, and trochanteric overgrowth after varus osteotomy
- Pin-tract infections and stiffness after arthrodiastasis
- Neurovascular injury or intraoperative fractures with complex reconstructive or arthroplasty procedures
- Complications related to prolonged casting or bracing, including skin breakdown and deconditioning
Deterrence and Patient Education
Primary prevention of LCPD is not currently possible, given its multifactorial and incompletely understood etiology. However, several patient- and family-directed measures can influence disease recognition, risk modification, and outcomes.[1][8]
- Early recognition: Parents and primary care clinicians should be educated that a persistent limp or activity-related hip, groin, thigh, or knee pain in a child warrants prompt evaluation, including hip radiographs. Early diagnosis allows timely containment strategies before advanced deformation.
- Risk factor counseling: Families should receive strong counseling regarding the avoidance of tobacco smoke exposure during pregnancy and childhood. Addressing obesity, poor nutrition, and limited physical activity may also have broader health benefits, though direct causal links to LCPD remain under investigation.
- Expectation setting: Clinicians should explain the protracted natural history, typical duration (2–4 years), and need for serial radiographs and examinations. Families should understand that the goal is a durable, well-shaped hip rather than rapid symptom resolution.
- Activity modification and adherence: Clear guidance on allowed and restricted activities, weight-bearing status, brace or cast wear, and home exercises is essential. Children and caregivers should understand the rationale for restrictions and the importance of maintaining hip motion for the long-term outcome.
- Transition to adult care: Adolescents and young adults with residual deformity should be counseled about symptoms of impingement or early arthritis and the potential need for hip-preserving or reconstructive procedures.
Pearls and Other Issues
Pearls and other issues concerning Legg-Calvé-Perthes disease include the following:
- Age at onset and lateral pillar (Herring) classification are the most practical and powerful prognostic tools and drive modern treatment algorithms.
- Most children younger than 6 years with limited involvement (Herring A or many B hips) do well with nonoperative management focusing on motion and symptom control.
- Children older than 8 years with Herring B/B-C disease benefit most from early surgical containment if the hip is still containable and motion is adequate.
- Herring C hips have a guarded prognosis; early aggressive containment may help, but often cannot prevent residual deformity.
- Preservation ofthe hip range of motion is as important as containment. Loss of abduction and internal rotation, and the presence of hinge abduction are red flags that should trigger re-evaluation of the treatment plan.
- Bilateral disease and atypical age at presentation warrant careful assessment for underlying dysplasias, metabolic disease, or hematologic conditions.
- Many adults with a history of LCPD present with femoroacetabular impingement rather than end-stage arthritis. Hip-preserving surgery may delay or avoid total hip arthroplasty in selected patients.
Enhancing Healthcare Team Outcomes
Perthes disease, or Legg-Calvé-Perthes disease, is a childhood orthopedic condition caused by avascular necrosis of the femoral head due to disrupted blood supply. Most often affecting boys ages 4 to 10, this condition presents with hip or groin pain, limping, and restricted hip mobility. The disease progresses through necrosis, fragmentation, reossification, and healing, with outcomes depending on timely diagnosis and appropriate management. Imaging, particularly x-ray and MRI, plays a crucial role in staging and prognosis, while treatment ranges from activity modification and physical therapy to surgical containment procedures, depending on severity and age at presentation.
Effective care requires clinicians to integrate diagnostic accuracy, evidence-based treatment strategies, and long-term monitoring. Physicians, general practitioners, and advanced practitioners must collaborate to determine individualized management plans, while nurses provide patient education and support for therapy adherence. Pharmacists contribute to safe pain management, and physical therapists maintain hip mobility and function. Interprofessional communication and coordinated strategies ensure patient-centered care, optimize outcomes, and safeguard long-term hip function, reducing complications such as deformity and early arthritis.
References
Rodríguez-Olivas AO, Hernández-Zamora E, Reyes-Maldonado E. Legg-Calvé-Perthes disease overview. Orphanet journal of rare diseases. 2022 Mar 15:17(1):125. doi: 10.1186/s13023-022-02275-z. Epub 2022 Mar 15 [PubMed PMID: 35292045]
Level 3 (low-level) evidencePerry DC. Unravelling the enigma of Perthes disease. Annals of the Royal College of Surgeons of England. 2013 Jul:95(5):311-6. doi: 10.1308/003588413X13629960046192. Epub [PubMed PMID: 23838491]
Leroux J, Abu Amara S, Lechevallier J. Legg-Calvé-Perthes disease. Orthopaedics & traumatology, surgery & research : OTSR. 2018 Feb:104(1S):S107-S112. doi: 10.1016/j.otsr.2017.04.012. Epub 2017 Nov 16 [PubMed PMID: 29155310]
Schwert A. [Legg-Calvé-Perthes disease]. Radiologie (Heidelberg, Germany). 2023 Oct:63(10):736-744. doi: 10.1007/s00117-023-01182-z. Epub 2023 Jul 8 [PubMed PMID: 37422572]
Herring JA, Kim HT, Browne R. Legg-Calve-Perthes disease. Part II: Prospective multicenter study of the effect of treatment on outcome. The Journal of bone and joint surgery. American volume. 2004 Oct:86(10):2121-34 [PubMed PMID: 15466720]
Level 2 (mid-level) evidenceSleem B, Abdul Khalek J, Kanbar K, Bitar E, Castaneda P, Masrouha K. Genetics and Epigenetics of Legg-Calvé-Perthes Disease. JBJS reviews. 2025 Mar 1:13(3):. doi: 10.2106/JBJS.RVW.24.00209. Epub 2025 Mar 25 [PubMed PMID: 40130954]
Pavone V, Chisari E, Vescio A, Lizzio C, Sessa G, Testa G. Aetiology of Legg-Calvé-Perthes disease: A systematic review. World journal of orthopedics. 2019 Mar 18:10(3):145-165. doi: 10.5312/wjo.v10.i3.145. Epub 2019 Mar 18 [PubMed PMID: 30918798]
Level 1 (high-level) evidenceDillman JR, Hernandez RJ. MRI of Legg-Calve-Perthes disease. AJR. American journal of roentgenology. 2009 Nov:193(5):1394-407. doi: 10.2214/AJR.09.2444. Epub [PubMed PMID: 19843759]
Ibrahim YH, Kersh MAAL, Fahmy H. Arthrodiastasis in the management of Perthes disease: a systematic review. Journal of pediatric orthopedics. Part B. 2020 Nov:29(6):550-555. doi: 10.1097/BPB.0000000000000690. Epub [PubMed PMID: 31856037]
Level 1 (high-level) evidenceBoghosian T, Momtaz D, Lawand J, Jahn J, Peterson B, Ghali A, Hosseinzadeh P. Risk Factors for Developing Perthes Disease: A Comprehensive National Analysis Spanning 2 Decades. Journal of pediatric orthopedics. 2025 May-Jun 01:45(5):e443-e448. doi: 10.1097/BPO.0000000000002914. Epub 2025 Mar 5 [PubMed PMID: 40048383]
Terjesen T, Wiig O, Svenningsen S. The natural history of Perthes' disease. Acta orthopaedica. 2010 Dec:81(6):708-14. doi: 10.3109/17453674.2010.533935. Epub 2010 Nov 11 [PubMed PMID: 21067434]
Accadbled F, Pailhé R, Launay F, Nectoux E, Bonin N, Gicquel P, SOFCOT. "Femoroacetabular impingement". Legg-Calve-Perthes disease: from childhood to adulthood. Orthopaedics & traumatology, surgery & research : OTSR. 2014 Oct:100(6):647-9. doi: 10.1016/j.otsr.2014.03.029. Epub 2014 Sep 10 [PubMed PMID: 25217031]
Alpaslan AM, Aksoy MC, Yazici M. Interruption of the blood supply of femoral head: an experimental study on the pathogenesis of Legg-Calve-Perthes Disease. Archives of orthopaedic and trauma surgery. 2007 Aug:127(6):485-91 [PubMed PMID: 17598118]
Dobbe AM, Gibbons PJ. Common paediatric conditions of the lower limb. Journal of paediatrics and child health. 2017 Nov:53(11):1077-1085. doi: 10.1111/jpc.13756. Epub [PubMed PMID: 29148202]
Anderson JA, Laucis N, Symanski J, Blankenbaker D. Imaging of Disease and Normal Variant Patterns in Pediatric Hips. Seminars in musculoskeletal radiology. 2024 Aug:28(4):447-461. doi: 10.1055/s-0044-1786153. Epub 2024 Jul 29 [PubMed PMID: 39074727]
Braun S, Adolf S, Brenneis M, Boettner F, Meurer A. Legg-Calvé-Perthes disease- surgical treatment options. Archives of orthopaedic and trauma surgery. 2025 Mar 12:145(1):186. doi: 10.1007/s00402-025-05801-3. Epub 2025 Mar 12 [PubMed PMID: 40072635]
Caldaci A, Testa G, Dell'Agli E, Sapienza M, Vescio A, Lucenti L, Pavone V. Mid-Long-Term Outcomes of Surgical Treatment of Legg-Calvè-Perthes Disease: A Systematic Review. Children (Basel, Switzerland). 2022 Jul 27:9(8):. doi: 10.3390/children9081121. Epub 2022 Jul 27 [PubMed PMID: 36010012]
Level 1 (high-level) evidenceMazloumi SM, Ebrahimzadeh MH, Kachooei AR. Evolution in diagnosis and treatment of Legg-Calve-Perthes disease. The archives of bone and joint surgery. 2014 Jun:2(2):86-92 [PubMed PMID: 25207324]
Maleki A, Qoreishy SM, Bahrami MN. Surgical Treatments for Legg-Calvé-Perthes Disease: Comprehensive Review. Interactive journal of medical research. 2021 May 3:10(2):e27075. doi: 10.2196/27075. Epub 2021 May 3 [PubMed PMID: 33938444]
Caruso G, Haoudou R, Schrooyen J, Malherbe C, Docquier PL. Natural history of shelf gap and paleo-acetabular depth after shelf acetabuloplasty for Perthes disease. Acta orthopaedica Belgica. 2020 Dec:86(4):599-605 [PubMed PMID: 33861905]
Ramachandran M, Azegami S, Hosalkar HS. Current concepts in the treatment of adolescent femoroacetabular impingement. Journal of children's orthopaedics. 2013 Mar:7(2):79-90. doi: 10.1007/s11832-012-0473-4. Epub 2013 Jan 11 [PubMed PMID: 24432063]
Stančák A, Kautzner J, Chládek P, Adamec O, Havlas V, Trč T. Predictors of radiographic outcomes of conservative and surgical treatment of Legg-Calvé-Perthes disease. International orthopaedics. 2022 Dec:46(12):2869-2875. doi: 10.1007/s00264-022-05584-x. Epub 2022 Sep 29 [PubMed PMID: 36173477]