Secondary Logo

Journal Logo


A Brief History of Legg-Calvé-Perthes Disease

Wenger, Dennis R. MD; Pandya, Nirav K. MD

Author Information
Journal of Pediatric Orthopaedics: September 2011 - Volume 31 - Issue - p S130-S136
doi: 10.1097/BPO.0b013e318226028c
  • Free


Legg-Calvé-Perthes (LCP) disease has provided a continual intellectual challenge for the orthopaedic community, with ongoing debate about etiology, pathogenesis, and treatment. From the time of its initial description by Legg, Calvé, and Perthes (1909 to 1910), the condition has puzzled clinicians across the globe (Fig. 1). To better understand LCP disease, one can benefit from a review of the development of our knowledge of the disorder.

The “founding fathers” who first described the condition. A, A. Legg—United States, B, J. Calvé–France, C, G. Perthes—Germany.


The 19th century brought the rapid development of orthopaedic knowledge, particularly in Germany, France, and Austria despite radiographs not having been discovered. With the increasing use of general anesthesia, direct surgical exploration of the hip led to an expanded understanding of hip pathology. Despite these efforts, clarification of different childhood hip diseases remained rather limited beyond hip sepsis and tuberculosis (TB); both of which were easily recognized owing to the significant destruction of the hip joint. It is from the study of hip infections in children that much of the early work centering on what we now know as LCP disease was unknowingly performed.

While performing a detailed histological analysis of hip TB in the late 1800s, Rokitansky, the renowned Viennese bone pathologist, described a milder form of childhood hip disease that closely mirrored what we now call LCP disease.1 Thomas, Baker, and Wright also reported on what they believed to be infectious processes affecting the pediatric hip whose presentation was retrospectively noted to be very similar to LCP.2–4

As a result, in lieu of radiographs, the proposed etiologies of the irritated pediatric hip in the 19th century (which were difficult to differentiate) were generally grouped into infection, coxitis/synovitis, slipped capital femoral epiphysis,5 osteochondritis,6 or pseudocoxalgia.7

Further analysis in the study of childhood hip disease awaited the radiographic era.


Roentgen's late 19th century discovery of the x-ray revolutionized the orthopaedic world, particularly the understanding of childhood hip disease.8 Within months after his report (around Christmas time, 1895 in Wurzburg, Germany), x-ray machines were installed in hospitals in most major European cities, confirming that truly important new scientific developments do not require advertisement and self-promotion. A revolution of eponymic attachment to orthopaedic disorders (Kohler, Sever, Osgood, Schlatter, Scheurmann) soon followed in an era described by Mercer Rang as “osteochondritismanship.” It was during this era in 1905 that Kohler made an initial description of the x-ray changes we now know as LCP disease,9 however his report was not widely recognized.

The first formal description of the disease occurred nearly simultaneously in 1909 and 1910 by Legg in Boston, Calvé in France, and Perthes in Germany; all of whom postulated different disease etiologies. Legg presented a series of limping patients with flattened femoral heads, which he believed were due to trauma.10 In contrast, Calvé reported on 10 patients with noninflammatory hip pain and a flattened femoral head that he felt was due to abnormal osteogenesis.11 Finally, Perthes reported on 6 patients with hip pain that he felt was due to an inflammatory condition.12 Concurrently, Waldenstrom from Sweden described the radiographic features of this condition although he mistakenly thought it was due to TB.13

The different theories regarding the etiology of the limping children in these early papers not only demonstrated the rapid impact of Roentgen's discovery in helping to provide a reproducible visual picture of a disease, but also that how little was known as to the cause of these flattened femoral heads as seen on x-ray.


Initially, the disease described by Legg, Calvé, and Perthes was lumped into the general category of childhood hip osteochondritis (what all abnormalities of bone growth and shape in childhood came to be referred to at the time). Yet, the true etiology remained unknown. Was it because of an embolic disorder, hyperemia, trauma, endocrine dysfunction, infection, or a blood supply deficiency?

Histological studies provided a better understanding of the pathophysiology of LCP disease, and a subsequent understanding of the radiographic stages of the disease. In 1921, Phemister in Chicago reported on his histologic findings after curettage of a proximal femur of a 10-year-old boy with several months of hip symptoms.14 He found a combination of necrotic and healing bone intermingled with granulation tissue and osteoclasts. This confirmed that the radiographic changes seen in LCP disease were likely owing to aseptic necrosis followed by a healing phase in which revascularization occurred. This histological sequence was described as schleichend ersetzung by the Germans (subsequently interpreted by Phemister as “creeping substitution”) in which bone was first resorbed by osteoclasts and then replaced with new bone by osteoblasts. The concept of osteoblasts and osteoclasts working simultaneously in the proximal femur was further supported by the work of Riedel in 1922.15 Yet, Phemister felt that his findings were due to both an inflammatory and infectious process, and the true etiology remained unanswered. Multiple other theories were entertained, including bacillary embolism,16 arterial hyperemia,17 transient synovitis,18 endocrine dysfunction,19 and external toxins.20

A disruption of the blood supply to the femoral head was also entertained, and has now come to the forefront as the most important etiologic factor in Perthes disease (Fig. 2). Konjetzny was the first to show in 1926 that an interruption to the vascular supply to the femoral head was present in LCP disease.21 This vascular work was furthered by Theron (obstruction of the superior retinacular artery),22 Atsumi et al (lateral epiphyseal artery interruption),23 Sanchis et al and Inoue et al (double infarction theory),24,25 and Kleinman and Bleck (increased blood viscosity).26 Currently, thrombophilia (owing to dysfunction in the levels of protein C and S) is being postulated as a possible factor leading to vascular thrombosis and subsequent osteonecrosis.27

Diagrammatic depiction of femoral head circulation. All etiologic theories focus on how this vascularity is somehow temporarily disturbed.

The above histological studies provided the framework for a better understanding of LCP pathophysiology, and the reasons for the changes in femoral head and neck shape that occurred during the disease process. Early, acute pain in LCP was thought to be the result of a shear fracture between the necrotic bone segment and the more elastic living bone under the cartilage surface (Caffey's sign).28 Growth in the femoral neck width (owing to appositional bone growth) during a period in which longitudinal growth had slowed or stopped created a short/wide femoral neck. The femoral head widened as well, leading to decreased coverage by the acetabulum with the acetabular rim impinging on the femoral head. Weight-bearing on this “biologically” plastic femoral head led to further deformity, hinge abduction, and finally an incongruent joint, which leads to premature arthritis.


Armed with a better understanding of disease etiology as gained from histological studies, new classification systems were developed after the initial era of radiographic discovery. In 1922, Waldenstrom (Fig. 3A, B) described 4 chronologic radiographic stages of LCP: initial, fragmentation, re-ossification, and healed.29 He noted that femoral head deformity developed and worsened during the initial and fragmentation stage. During the re-ossification stage, femoral head deformity could worsen, improve, or remain unchanged.

A, H. Waldenstrom (Sweden) also described the radiographic features now known as LCP disease but thought the changes were due to tuberculosis. B, The chronologic radiographic stages of LCP as described by Waldenstrom: initial, fragmentation, re-ossification, and healed. The chronologic sequence occurs over an approximate 2-year time period (Illustration courtesy of Stuart Weinstein, MD). LCP indicates Legg-Calvé-Perthes.

Nearly 50 years later (1971), Catterall30 developed a prognostic radiographic classification based on the extent of the femoral head infarct (I: anterior epiphysis involvement, II: central segment fragmentation and collapse, III: anterior, central, and lateral involvement; IV: entire head involvement). In addition, he defined 4 head “at-risk” signs, including lateral subluxation, calcification lateral to the epiphysis, Gage's sign (V shaped defect laterally), and a horizontal growth plate. A fifth at risk sign was later contributed by Smith in 1982 (presence of a metaphyseal cyst).31

In 1980, Salter and Thompson32 recognizing that Catterall groups I and II were distinct from groups III and IV, developed a simpler classification dividing patients into 2 groups (A: less than ½ of the head involved, B: greater than ½ of the head involved). The extent of femoral head involvement was determined by the extent of the subchondral fracture (Caffey's sign).

Seventy years after Waldenstrom (1992), Herring et al33 developed another prognostic classification based on the height of the lateral epiphyseal pillar during the fragmentation stage of disease (group A: no collapse, group B: lateral pillar with >50% of original height, group C: lateral pillar with <50% of original height). They later added a “B-C” border classification, making the system as complex as that of Catterall.34

These classification systems with follow-up analysis clarified that patients with “at-risk” signs and/or Catterall stages III and IV (Salter and Thompson B), and Herring group C patients were destined to have poor outcomes owing to progressive head deformity and joint incongruity. Stulberg et al (1981) developed an important 5-part classification system, which could be made at the time of femoral head healing, clarifying which patients would have premature arthritis.35 The use of alternative imaging modalities in addition to plain radiograph have become increasingly important in defining disease severity (Fig. 4A–C).

Alternative imaging methods to analyze LCP disease. A, Bone scan, B, MRI, C, 3-D CT Scan.


Early treatment approaches for LCP focused on long periods of hospitalization and complete nonweight-bearing, either by bed rest or a wheelchair. Subsequent methods focused on childhood mobility and included a wide variety of special types of braces, including the “patten-ended caliper” designed by Hugh Owen Thomas, and other braces and belts designed to prevent weight-bearing and unload the hip for up to 2 years to allow for femoral head healing.36

This period of bracing and belts was followed by the abduction plaster cast era brought on by A. O. Parker and Brian McKibben in Cardiff, Wales (Fig. 5A).37 They found that patients could be placed in wide abduction plaster casts for an extensive period of time (often 1 year or more) to provide femoral head containment within the acetabulum during revascularization using the acetabulum as a mold for the “soft” femoral head. The abduction casting concept was popularized in North America by Gordon Petrie 1971 at the Montreal Shriner's Hospital.38 Although demanding and rigorous to apply, the Petrie cast method showed that weight-bearing with the femoral head contained was not harmful (Fig. 5B).

A, Cardiff, Wales—where A.O. Parker and Brian McKibben developed the concept of abduction cast containment. B, Modern fiberglass cast used for the nonoperative treatment of LCP disease. LCP indicates Legg-Calvé-Perthes.

Problems with Petrie casting were common, including multiple clinic visits and even hospitalizations for repeating casting, regaining hip and knee range of motion, and dealing with femoral condylar flattening that sometimes was seen with prolonged treatment. As a result, abduction casting was replaced by more civilized, removable braces that mimicked Petrie casts (Fig. 6A–C).39,40 The Toronto brace (developed by Bobechko), which allowed full knee flexion while maintain hip internal rotation proved too cumbersome to be practical. The Atlanta brace, which initially became popular owing to its ease of wear, soon fell out of favor because it failed to internally rotate the femur (and thus could not predictably contain the femoral head). The ineffectiveness of this brace style was shown by the work of both Meehan et al and Martinez et al.41,42

Traditional brace methods to provide femoral head containment. A, Traditional abduction brace, B, Toronto brace (Bobechko), C, Atlanta brace.


By the early 1960s, it became clear that keeping a child in a cast or brace for a prolonged period was not only difficult and psychologically stressful to a child, but also often ineffective because the child would not dutifully wear the brace. The demise of government/charity hospitals where a child could be hospitalized for weeks or even months was also drawing to a close. This caused surgeons to consider operative methods to contain the femoral head.

The concept of surgical containment of the femoral head via femoral varus osteotomy was first introduced by Soeure in Belgium 1952,43 followed by Axer in Israel in 1965,44 and Lloyd-Roberts in London in 1976.45 The femoral varus osteotomy centered the head in the acetabulum, allowing the acetabulum to serve as a mold (Fig. 7A).

Surgical containment methods for LCP (A) Femoral varus osteotomy, (B) Salter osteotomy. LCP indicates Legg-Calvé-Perthes.

In 1962, Salter added the innominate osteotomy as a method for containment in Perthes disease (a method that he initially used for developmental dysplasia of the hip, Fig. 7B).46 He felt that acetabular rotation provided a better method for femoral head containment while avoiding the potential limb shortening and limp often associated with femoral osteotomy. Subsequently it has been noted that the Salter osteotomy alone cannot provide adequate acetabular rotation to fully contain the femoral head in older patients with extensive necrosis. These older children with severe head involvement sometimes benefit from advanced containment methods such as the combination of femoral osteotomy plus Salter osteotomy or triple pelvic osteotomy.47,48


Although LCP disease has been recognized for a century, full agreement on its etiology and treatment remains controversial. Despite a lack of full understanding of the disease, modern treatment methods allow most children to be treated without lengthy hospitalization, casting, or bracing. Advanced imaging modalities such as 3-dimensional CT scan and dynamic MRI have given us greater insight into the anatomy and growth patterns of this disorder.49 Surgical containment is generally successful, but there are cases, which fail and require further surgery. Older patients who may develop femoral acetabular impingement after advanced coverage procedures can later be treated by the methods developed by Ganz, including femoral head neck recontouring or femoral head reduction.50 Future treatment of LCP is also likely to include biological and pharmacologic methods.51


1. Rokitansky F, von C A Manual of Pathological Anatomy, vol 3 (translated by Moore CH. 1850 London Sydenham Society
2. Thomas HO Contributions to Surgery and Medicine. Part II. Principles of Treatment of Diseased Joints.. 1883 London HK Lewis
3. Baker WM. Epiphyseal necrosis and its consequences Br Med J.. 1883;2:416
4. Wright GA. The value of determining the primary lesion in joint disease as an indication for treatment Br Med J. 1883;2:419
5. Muller E. Uber die verbiegung des schenkelhalses in wachstumalts Beitr Klin Chir.. 1888;4:137
6. Konig F. Uber freie korper in den gelenken Dtsch Z Chir.. 1888;27:90
7. Duplay S. On pseudocoxalgia Medical Press and Circular.. 1896;62:567
8. Roentgen WC. Ueber eine neue art von strahlen (vorlaufige mitteilung) Sber Physik-med.. 1895;9:132–141
9. Kohler A Atlas der anatomie und pathologie des huftgelenkes. 1905 Hamburg Graefe and Sillem
10. Legg AT. An obscure infection of the hip joint Boston Med Surg J.. 1910;162:202
11. Calvé J. Sur une forme particuliere de coxalgie greffe sur des deformations caracteristiques de l'extremite superieure de femur Rev Chir.. 1910;42:54
12. Perthes G. Uber arthritis deformans juvenilis Dtsch Z Chir.. 1910;10:111
13. Waldenstrom H. Der obere tuberkulose collumnerd Z Orthop Chir.. 1909;24:487
14. Phemister DB. Perthes disease Surg Gynecol Obstet.. 1921;33:87
15. Riedel G. Pathologic anatomy of osteochondritis deformans coxae juvenilis Zentralbl Chir.. 1922;49:1447
16. Axhausen G. Kohler's disease and Perthes' disease Zentralbl Chir.. 1923;50:553
17. Bentzon PGK. Experimental studies on the pathogenesis of coxa plana Acta Radiol.. 1926;6:155
18. Jacobs BW. Early recognition of osteochondrosis of capital epiphysis of femur JAMA.. 1960;172:527
19. Gill AB. Relationship of Legg-Perthes' disease to the function of the thyroid gland J Bone Joint Surg Am.. 1943;25:892
20. Glimcher MJ. Legg-Calvé-Perthes' syndrome: biological and mechanical considerations in the genesis of clinical abnormalities Orthop Rev.. 1979;8:33–40
21. Konjetzny GE. Zur patholgie and pathologischen anatomie der Perthes-Calvé schen krankheit Act Chir Scand.. 1943;74:361
22. Theron J. Angiography in Legg-Calvé-Perthes disease Radiology.. 1980;135:81
23. Atsumi T, Yoshihara S, Hiranuma Y. Revascularization of the artery of the ligamentum teres in Perthes disease Clin Orthop Relat Res.. 2001;386:210–217
24. Sanchis M, Freeman MAR, Zahir A. Experimental stimulation of the blood supply to the capital epiphysis in the puppy J Bone Joint Surg Am.. 1973;55:335
25. Inoue A, Freeman MAR, Vernon-Roberts B, et al. The pathogenesis of Perthes disease J Bone Joint Surg Br.. 1976;58:483
26. Kleinman RG, Bleck EE. Increased blood viscosity in patients with Legg-Perthes' disease: a preliminary report J Pediatr Orthop.. 1981;1:131
27. Glueck CJ, Crawford A, Roy D, et al. Association of antithrombotic factor deficiencies and hypofibrinolysis with Legg-Perthes disease J Bone Joint Surg Am. 1996;78:3
28. Caffey J. The early roentgenographic changes in essential coxa plana: the significance in pathogenesis Am J Roentgenol.. 1968;103:620–634
29. Waldenstrom H. The definitive forms of coxa plana Acta Radiol.. 1922;1:384
30. Catterall A. The natural history of Perthes' disease J Bone Joint Surg Br. 1971;53B:37
31. Smith SR, Ions GK, Gregg PJ. The radiological features of the metaphysis in Perthes' disease J Pediatr Orthop. 1982;2:401–404
32. Salter RB, Thompson GH. Legg-Calvé-Perthes disease: the prognostic significance of the subchondral fracture and two group classification of the femoral head involvement J Bone Joint Surg Am.. 1984;66:479–89
33. Herring JA, Neustadt J B, Williams JJ, et al. The lateral pillar classification of in Legg-Calvé-Perthes disease J Pediatr Orthop. 1992;12:143–150
34. Herring JA, Kim HT, Browne R. Part I: Classification of radiographs with use of the modified lateral pillar and Stulberg classifications J Bone Joint Surg Am.. 2004;86:2103–20
35. Stulberg SD, Cooperman DR, Wallenstein R. The natural history of Legg-Calvé-Perthes disease J Bone Joint Surg Am.. 1981;63:1095
36. Friberg S. The roentgenological results after caliper treatment of coxa plana with varying degrees of epiphyseal involvement Acta Orthop Scand.. 1975;46:234–45
37. Brotherton BJ, McKibbin B. Perthes disease treated by prolonged recumbency and femoral head containment: a long term appraisal J Bone Joint Surg.. 1977;59:8–14
38. Petrie JG, Bitenc I. The abduction weightbearing treatment in Legg-Perthes' disease J Bone Joint Surg Br.. 1971;53:54
39. Purvis JM, Dimon JH III, Meehan PL, et al. Preliminary experience with the Scottish Rite Hospital abduction orthosis for Legg-Perthes disease Clin Orthop Relat Res.. 1980;150:49–53
40. Bobechko WP. The Toronto brace for Legg-Perthes disease Clin Orthop Relat Res.. 1974;102:115–117
41. Meehan PL, Angel D, Nelson JM. The Scottish Rite abduction brace orthosis for the treatment of Legg-Perthes' disease: a radiographic analysis J Bone Joint Surg Am. 1992;74:2
42. Martinez AG, Weinstein SL, Dietz FR. The weight bearing abduction brace for the treatment of Legg-Perthes' disease J Bone Joint Surg Am.. 1992;74:12–21
43. Soeur R, DeRacker CH. L'aspect anatomopathologique de l'osteochondrite et les theories pathogeniques quis'y rapportent Acta Orthop Belg.. 1952;18:57–102
44. Axer A. Subtrochanteric osteotomy in the treatment of Perthes' disease: a preliminary report J Bone Joint Surg Br.. 1965;47:489–499
45. Lloyd-Roberts GC, Catterall A, Salamon PB. A controlled study of the indications for and the results of femoral osteotomy for the treatment Perthes' disease J Bone Joint Surg Br.. 1976;58:31–36
46. Salter RB. Perthes' disease: treatment by innominate osteotomy Inst Course Lect.. 1973;22:309
47. Wenger DR. Selective surgical containment for Legg-Perthes' disease: recognition and management of complications J Pediatr Orthop. 1981;1:153
48. Crutcher JP, Staheli LT. Combined osteotomy as a salvage procedure for severe Legg-Calvé-Perthes disease J Pediatr Orthop. 1992;12:151
49. Kim H, Wenger DR. “Functional retroversion” of the femoral head in Legg-Calvé-Perthes disease and epiphyseal dysplasia: analysis of head-neck deformity and its effect on limb position using three-dimensional computed tomography J Pediatr Orthop. 1997;17:240
50. Eijer H, Podeszwa DA, Ganz R, et al. Evaluation and treatment of young adults with femoro-acetabular impingement secondary to Perthes' disease Hip Int. 2006;16:273–280
51. Aya-ay J, Athavale S, Morgan-Bagley S, et al. Retention, distribution, and effects of intraosseously administered ibandronate in the infracted femoral head J Bone Miner Res.. 2007;22:93–100

Legg-Calvé-Perthes disease; pediatric hip; history; treatment

© 2011 Lippincott Williams & Wilkins, Inc.