Journal Logo

Review Article

Multiple Epiphyseal Dysplasia

Anthony, Steven DO; Munk, Richard MD; Skakun, William DO; Masini, Michael MD

Author Information
Journal of the American Academy of Orthopaedic Surgeons: March 2015 - Volume 23 - Issue 3 - p 164-172
doi: 10.5435/JAAOS-D-13-00173
  • Free


Multiple epiphyseal dysplasia (MED) is a genotypically and phenotypically heterogeneous skeletal dysplasia.1,2, MED was first described by Fairbank3 and has subsequently been further elucidated. Advances in genetics have uncovered six different mutations responsible for the condition, which may be inherited in either autosomal dominant or autosomal recessive forms.2,4-6 The end result is disorganized endochondral ossification of the epiphyses of long bones. Articular cartilage at these sites is initially normal, but rapidly degenerates because of the lack of underlying osseous support. This early articular degeneration leads to precocious end-stage osteoarthritis, with a predilection for large weight-bearing joints.6-9 Early accurate diagnosis of MED is vital for maximizing the functional outcomes of affected individuals and identifying those at risk for propagating the disease.

Epidemiology and Etiology

The reported prevalence of MED ranges from 1 in 10,000 to 1 in 20,000 people.10,11 A wide range of phenotypic expressions of the disease exists among the different genotypic forms.2,4,12,13 Most of the cases identified are autosomal dominant, with family studies demonstrating probands with multiple affected family members who require hip or knee arthroplasty at an early age.12

Currently, six genetic mutations have been identified as being responsible for MED. These mutations include the collagen oligomeric matrix protein (COMP), collagen type IX α-1 (COL9A1), collagen type IX α-2 (COL9A2), collagen type IX α-3 (COL9A3), matrilin-3 (MATN3) genes, and the sulfate transporter gene SLC26A2. All of these mutations are autosomal dominant except the SLC26A2 gene, which is autosomal recessive. The autosomal dominant varieties comprise approximately 75% of reported cases, of which 66% have been attributed to COMP mutations, 24% to MATN3, and 10% to COL9A mutations.14 Many cases of apparent MED without an identifiable genetic mutation still exist.14,15 This lends credence to the belief that additional genetic etiologies of MED are yet to be identified.2,14,15

Clinical Presentation

Autosomal Dominant Form

Autosomal dominant forms of MED often present in childhood. However, some patients do not present until after reaching skeletal maturity. In affected children, symptoms of these forms of MED include early fatigue during long walks or play; limited range of motion; limp; and periarticular pain in the knees, hips, and shoulders. Clinical findings include brachydactyly and mildly short stature with average trunk height. There may also be contractures about the hip, knee, and/or elbow. There is no facial or pelvic involvement. The spine is unaffected, but mild radiographic changes, such as irregular end plates and Schmorl nodes, may be observed.15 Often, there is a mild myopathic component to the condition.10,12 However, in general, motor function and intelligence are not compromised.

Autosomal Recessive Form

The autosomal recessive form of MED (rMED) typically presents later in life than does the autosomal dominant form. Most patients present in late adolescence or early adulthood. However, some of these patients can be identified at birth because approximately 44% have birth defects including clubfoot, cleft palate, or clinodactyly.4,16 The autosomal recessive type has more involvement of the hands and feet and often includes mild scoliosis. The developmental abnormalities of the epiphyses of long bones are similar to those of the autosomal dominant forms.4


History and Physical Examination

The accurate diagnosis of MED largely depends on a focused history and detailed physical examination. As described previously, early fatigue with walking or playing, limited range of motion, and periarticular pain in major weight-bearing joints are typical complaints.4,12 Family history can provide clues to the genetic etiology and should be evaluated in detail, especially if relatives have a similar history or older relatives have been treated with joint arthroplasty at a young age.

Radiographic Evaluation

As part of the initial evaluation, plain radiographs of the affected area should be obtained. Once MED is suspected, a skeletal survey is mandatory to evaluate other joints. Radiographic findings are symmetric. In children, the hip may show small, underdeveloped proximal femoral epiphyses and acetabular changes.7,17 The femoral neck is often shortened, widened, and in varus malalignment (Figures 1 and 2). The proximal tibial and distal femoral epiphyses will demonstrate loss of height, underdevelopment, and irregular contours. Most patients’ knees have some degree of valgus deformity, and genu varum has also been reported17-19 (Figure 3). There may also be ankle or shoulder involvement; changes are often seen at the distal tibia, talus, and humeral head3,17 (Figure 4). Brachydactyly and epiphyseal changes are common in the hands. End plate irregularities or Schmorl nodes may be noted in the spine15 (Figure 5).

Figure 1
Figure 1:
Frog-leg lateral radiograph of the pelvis demonstrating findings consistent with multiple epiphyseal dysplasia in an 8-year-old child. The proximal femoral epiphyses are small and round secondary to delayed development. There are significant and symmetric acetabular changes secondary to the lack of femoral epiphyseal development.
Figure 2
Figure 2:
AP radiograph of the pelvis of the patient shown in Figure 1. Note the underdeveloped, round femoral epiphyses and acetabular dysplasia. The metaphyses were spared.
Figure 3
Figure 3:
AP radiograph of the knees demonstrating underdeveloped distal femoral and proximal tibial physes with flaring of the metaphyses and significant valgus deformity of the left knee in a 9-year-old patient with multiple epiphyseal dysplasia.
Figure 4
Figure 4:
Mortise (A) and lateral (B) weight-bearing radiographs of the ankle in an adult with multiple epiphyseal dysplasia. Note the flattening of the talar dome and talar head on the lateral view. The medial talar dome appears flattened on the AP view, and there is valgus deformity.
Figure 5
Figure 5:
Lateral radiograph of the thoracic spine demonstrating end plate irregularities in a child with multiple epiphyseal dysplasia. The arrows point to anterosuperior and anteroinferior “cupping” of the end plates secondary to insufficient growth of the vertebral ring epiphyses.

In adults, additional findings can include flattening of the femoral condyles, with a shallow trochlear groove and depression of the lateral tibial plateau;20 decreased sphericity of the humeral and femoral heads; and flattening of the talar dome.3,17,21 The double-layered patella sign may be seen on the lateral radiograph of the knee4,22 (Figure 6). There are two separate ossification centers, anterior and posterior, that fail to fuse. This is characteristic of MED and is considered by some to be a pathognomonic finding, particularly for the recessive form.4,13,17,22

Figure 6
Figure 6:
Lateral radiograph of the knee demonstrating the classic double-layered patella sign. (Reproduced with permission from Unger S, Bonafé L, Superti-Furga A: Multiple epiphyseal dysplasia: Clinical and radiographic features, differential diagnosis and molecular basis. Best Pract Res Clin Rheumatol 2008;22[1]:19-22.)

The role of MRI in the evaluation of patients with confirmed MED has been assessed.23 Although MRI allows further detailed assessment of the affected joint, no superior benefits (compared with plain radiography) for the diagnosis and evaluation of MED have been demonstrated.

Genetic Evaluation

Genetic analysis can be a valuable tool for diagnosing and managing MED. Identifying the exact genetic etiology of the disease allows for patient (and family) education on the pattern of inheritance and the chance that subsequent offspring may be affected. Moreover, genetic analysis facilitates early and accurate diagnosis of MED. This is important for educating the patient and his or her family about the expected course of the disease, which may influence activity and occupation choices and allow for earlier initiation of joint-preserving treatment.18 Although genetic testing offers these significant benefits, it is expensive and imperfect. Complete MED panels range from $1,800 to $5,990,24 and 10% to 20% of patients with a clinical diagnosis of MED may fail to test positive for known genetic markers.13,15 To address these issues, targeted genetic testing protocols and genotype-phenotype correlations have been investigated.

Targeted genetic testing of suspected MED uses clinical and radiographic clues to generate a prioritized, cost-effective approach to testing.10,13,17 Most laboratories can test for individual gene mutations or perform a complete MED panel. Individual mutations can be tested at a fraction of the cost of the complete panel. Identifying clinical and radiographic differences between genotypes would increase the accuracy of individual gene testing, thus minimizing costs. Toward this goal, several studies have demonstrated some reliable genotype-phenotype correlations.

COMP mutations result in small, rounded capital femoral epiphyses and acetabular changes at the hip, widened metaphyses and shortened epiphyses at the knees, and brachydactyly.13,25,26MATN3 mutations also result in small, rounded femoral epiphyses, but acetabular changes and brachydactyly are uncommon.13,17COL9A mutations are associated with severe dysplastic changes at the knees but relative sparing of the hips.26 The double-layered patella sign may represent the recessive genotype.4 These findings and other clinical and radiographic markers10,13,17,25,26 are summarized in Table 1.

Table 1
Table 1:
Genotype-Phenotype Correlations

Any patient with a clinical diagnosis of MED should be offered genetic counseling, which provides patients and their families with information on the genetic nature of the disease and the costs and potential benefits of genetic testing. If they choose to go forward with genetic testing, the genotype-phenotype correlations should be used to prioritize testing. It should be noted that the genotype-phenotype correlations alone are not sufficient to make a genetic diagnosis. There is overlap among the correlations, and any phenotypic expression may be seen in any of the genetic forms of the disease.1,2,10,13,27 Therefore, although we recommend the use of these correlations to generate an efficient plan for genetic testing, we do not recommend attempting to make a genetic diagnosis based on clinical and radiographic findings alone.

Differential Diagnosis

The differential diagnosis of MED includes Legg-Calvé-Perthes disease (LCP), spondyloepiphyseal dysplasia (SED), congenital hypothyroidism, mucopolysaccharidoses, and several genetically related disorders, including pseudoachondroplasia (PSACH) and diastrophic dysplasia (Table 2). SED is caused by a mutation of the COL2A gene and presents with articular manifestations similar to MED.28 However, SED demonstrates vertebral body abnormalities and significant scoliosis, whereas MED typically presents with little or no spinal involvement.28 Mucopolysaccharidoses may be differentiated with urinalysis, and congenital hypothyroidism may be confirmed with evaluation of thyroid function tests.

Table 2
Table 2:
Differential Diagnoses for Multiple Epiphyseal Dysplasia

Genetically Related Disorders

A COMP gene mutation is associated with PSACH and MED. PSACH may result in short-limbed dwarfism, cervical instability, scoliosis, increased lumbar lordosis, bowing of the lower extremities and joint flexion contractures. These abnormalities are not typically seen with MED.29 Defects of the COL9A genes may result in Stickler syndrome types 4 and 5, which present with hearing and vision deficits, cleft palate, and flattened facies in addition to joint laxity and early arthritis. In addition to being the etiology of rMED, mutations of the SLC26A2 gene may cause atelosteogenesis type 2, achondrogenesis type 1B, or diastrophic dysplasia.

MED Versus LCP

MED can be confused with bilateral LCP disease, and distinguishing between the two is vital. Unlike the circumstances in patients with LCP, there is no potential for remodeling of the capital femoral epiphysis in those with MED. Thus, treatment with prolonged abduction and immobilization subjects pediatric patients to unwarranted therapeutic morbidity. However, if the child with LCP is deprived of such treatment, the outcome may be detrimental.18

MED is always bilateral with symmetric changes. This finding alone distinguishes MED from LCP in most cases because LCP typically presents with unilateral changes. In a study of 637 cases of LCP, Guille et al30 found that LCP was bilateral in 83 (13%). Of these 83 cases, 26 (31%) presented at the same Waldenstrom stage and appeared symmetrical, similar to MED. In these patients, a skeletal survey helps to differentiate between these two disease processes because LCP does not demonstrate the deformities about the knee and other joints typically seen with MED.18 Acetabular changes associated with MED are typically found at the initial workup; the dysplastic femoral head has not allowed for proper development of the acetabulum. Patients with LCP initially present with normal acetabula because the development of the femoral epiphysis was previously normal; however, compensatory acetabular changes often develop later (Figure 7). Cysts within the metaphyseal region of the proximal femur may also provide a clue to the proper diagnosis. These are often seen in patients with LCP, but not in those with MED.

Figure 7
Figure 7:
A, Inlet radiograph of the pelvis showing symmetric changes, dysplastic acetabula, and no metaphyseal changes in a patient with multiple epiphyseal dysplasia. B, Inlet radiograph of the pelvis in a patient with bilateral Legg-Calvé-Perthes disease showing asymmetric changes of the femoral epiphyses, metaphyseal changes, and sparing of the acetabula.



Treatment of children with MED begins with a comprehensive physical examination and radiographic evaluation to document baseline findings and determine appropriate therapies. The goals of treatment are to delay the onset of early osteoarthritis, improve function, and educate patients and their families on the natural history and genetic basis of the disease. Nonsurgical treatments, such as anti-inflammatory medications, pain control, physical therapy, and activity modification, are mainstays of initial care. Appropriate weight control and avoidance of high-impact activities are imperative to minimize stress on the affected joints. In addition, genetic counseling should be offered.12 If symptoms and/or deformities are significant, surgical intervention may be indicated.

Potential benefits of surgical intervention for children with MED include pain relief, improved motion, and realignment of the mechanical axis. To achieve this, arthroscopic or open débridement and various realignment procedures have been advocated. Patients with MED may be prone to painful articular loose bodies and meniscal or labral tears at a young age, which can be managed with open or arthroscopic débridement.31 Realignment procedures may provide improved function and stability and diminished joint stresses. Multiple surgical options have been described, including temporary and permanent hemiepiphysiodesis, various osteotomies, and external fixation for gradual correction of deformities.19,32,33 Although the long-term benefits of these treatments remain unproven, the goals of treatment are to delay the onset of early osteoarthritis and maintain joint anatomy and stability, making future total joint arthroplasties less complicated.19,32 Additionally, surgical correction may be undertaken for cosmesis because angular deformities can have a striking appearance.

Accurate measurement of angular deformities, especially those at the knee, should be obtained before initiating surgical correction to determine the extent of the deformity and create a baseline by which to measure correction. Akhmedov et al19 compared the reliability of several different angular measurements of the lower extremities in children with and without MED. The authors found that the most reliable measurements were the mechanical tibiofemoral angle, anatomic tibiofemoral angle, and mechanical axis deviation. Because of the abnormal development and small size of the epiphysis, measurements using the physis and epiphysis were found to be less reliable.

Growth modulation procedures through temporary (eg, staples, plates) or permanent hemiepiphysiodesis are used at the distal femoral and proximal tibial physes of skeletally immature patients. Hemiepiphysiodesis of the proximal femur performed when the greater trochanter and capital femoral epiphysis are still contiguous may be helpful in reversing developmental coxa vara. Although temporary hemiepiphysiodesis has historically yielded reliable outcomes, such treatment for MED should be undertaken cautiously because of the unpredictable activity of the physis after staple or bridge plate removal.32 Cho et al32 presented findings on nine patients (17 knees, 24 staple hemiepiphysiodeses) treated with hemiepiphyseal stapling to correct angular deformity about the knee. Although all the angular deformities were corrected appropriately, only 14 of 17 knees maintained the appropriate alignment. The authors found the surgery to be effective in initially correcting angular deformities but, because of the unpredictable nature of the involved physes, Cho et al32 recommended avoiding the standard overcorrection of the deformity and close monitoring.

Although osteotomies about the hip and knee may be beneficial, they should be used carefully so as not to make future total joint arthroplasty more difficult than it already is in these cases. The severity of deformities about the hip varies greatly and may or may not include significant acetabular changes.9 Osteotomies should be planned on an individual basis and should attempt to correct varus or valgus malalignment and maintain a more spherical and concentrically reduced femoral head. To this end, trochanteric advancement, varus- or valgus-producing osteotomies, and various acetabular osteotomies can be used, and good outcomes have been reported.34 Distal femoral and proximal tibial osteotomies can be performed to correct varus or valgus malalignment at the knees. These procedures are indicated for patients who are skeletally mature and have severe deformity or for skeletally immature patients in whom growth modulation has failed or is contraindicated. Whenever possible, hardware should be removed after the osteotomy has healed because of the high likelihood of future total joint arthroplasty (Figure 8).

Figure 8
Figure 8:
Full-length AP radiographs of the knees before (A) and after (B) distal femoral osteotomy and subsequent hardware removal.


Initially, adults with MED should be treated nonsurgically and genetic counseling should be offered. Most adults with MED present with advanced degenerative joint disease (DJD). Nonsurgical treatment may have little success because of significant damage to the affected joints, and total joint arthroplasty or arthrodesis should be considered.

Arthrodesis is often recommended for patients who are active laborers with advanced DJD of the hip, knee, shoulder, or other joints without MED. However, these patients typically suffer from isolated DJD secondary to trauma, osteonecrosis, or congenital deformity. Arthrodesis of the hip or knee places substantially more stress on the lumbar spine and surrounding joints.35 Patients with MED typically already have degeneration of these surrounding joints, and arthrodesis in this setting can lead to rapid deterioration at adjacent sites. In the setting of bilateral hip/knee DJD in a young patient with MED, total joint arthroplasty of the hip and/or knee is preferable to arthrodesis. However, fusion about the foot and ankle, elbow, and shoulder may be preferable to joint arthroplasty, particularly in the younger patient with joint instability.

Total joint arthroplasty in this population is less predictable than it is in the setting of osteoarthritis, but satisfactory outcomes can be attained.21,36,37 More sophisticated implants, such as custom prostheses, modular hips, or constrained knees, are often necessary to overcome the instability and dysplastic changes that often accompany joint disease in these patients.21,36,37

When considering total hip arthroplasty (THA) in the setting of MED, several anatomic variations must be understood.21,37 The true acetabulum is often very small, shallow, and poorly developed, with version abnormalities.21,36,37 This can make placement and secure fixation of an acetabular component challenging. The proximal femur frequently demonstrates a short varus neck, a widened metaphysis, and a narrow femoral canal.9,36,37 This can make adequate fixation of standard femoral prostheses challenging. Additionally, these patients often demonstrate proximal migration of the femur, which further complicates accurate placement of implants, postoperative stability, and achievement of equal leg length. Despite these anatomic changes, satisfactory outcomes can be achieved with THA in patients with MED.21,36,37

Ramaswamy et al37 reported on THA performed in patients with MED, with an average follow-up of 15 years. The authors demonstrated significant pain relief and improvement in quality of life, although their quantitative outcomes were lower than those reported after THA in patients without MED. No femoral revisions were required. The survival of the acetabular component was 93.7% (15 of 16 components) at 10 years and 76.7% (13 of 16 components) at 15 to 20 years. Two of the three failures were threaded acetabular components, which have poorer outcomes and are no longer in use in North America.37,38 Polyethylene exchange with retention of acetabular and femoral components was not performed in the first 10 years. However, by the 15-year follow-up, 11 (69%) had been performed.37 Considering the population, the polyethylene survival rate is acceptable and could be even better with modern cross-linking and sterilization procedures.

Pavone et al36 reported on seven patients with MED who presented in the third decade of life and underwent staged bilateral THA. At an average 6-year follow-up, the authors found that range of motion, clinical function, and quality of life were improved. There was only one case of aseptic loosening, with no cases of dislocation or periprosthetic fracture reported.

Lim et al21 reported on 23 modular cementless THAs performed in 13 patients with MED and end-stage arthritis of the hip. At a mean follow-up of 4.8 years, no revisions were required because of aseptic loosening. However, one patient underwent revision for polyethylene exchange and grafting 8 years after the index arthroplasty.

In the studies by Lim et al,21 Pavone et al,36 and Ramaswamy et al,37 16 of 53 hips (30%) developed heterotopic ossification. This is likely secondary to extensive soft-tissue stripping because larger exposures are necessary to address the dysplastic changes at the acetabulum and proximal femur. Surgeons should consider implementing standard prophylaxis to protect against heterotopic ossification when performing THA in this population. Other complications included osteolysis, which required polyethylene exchange and grafting, and aseptic loosening of the acetabular component. All studies demonstrated reliable pain relief and improvement in quality of life, with acceptable longevity in patients with MED treated with THA.21,36,37

Total knee arthroplasty and total shoulder arthroplasty may also be used in the setting of MED. Similar to hip arthroplasty, arthroplasties of the knee and shoulder require special consideration with regard to the anatomic deformities frequently seen with this condition. Specific considerations include proper sizing of the prosthesis because of metaphyseal-diaphyseal mismatch, correction of angular deformities, and joint stability. The metaphysis of the humerus, femur, and tibia are often enlarged relative to the diaphysis8,25 and require larger prostheses than expected. Angular deformity about the knee that has been present since childhood often leads to ligamentous laxity or flexion/extension contractures. At the shoulder, proximal diaphyseal bowing and significant glenoid dysplasia may be encountered.8 In these cases, augmentation of bony deformity and the use of custom or constrained prostheses may be necessary. Patients with MED also may present with a shallow trochlear groove and a multilayered patella.4,20 When present, these deformities predispose the knee to patellofemoral instability. Currently, no clinical studies have examined total knee or total shoulder arthroplasty in patients with MED.


MED is a genotypically and phenotypically heterogeneous disorder that may be inherited in an autosomal dominant or autosomal recessive form. It is responsible for the development of symmetric dysplasia of the epiphyses of long bones and results in an asymmetric, mildly short stature and precocious DJD. An accurate diagnosis requires a detailed physical examination, radiographic skeletal survey, and detailed family history. When a clinical diagnosis of MED is made, the patient should be offered genetic counseling. Treatment of MED in children is aimed at joint preservation and limb alignment. Treatment of adults with MED primarily consists of total joint arthroplasty or arthrodesis.


Evidence-based Medicine: Levels of evidence are described in the table of contents. In this article, references are 19, 21, 36, and 38 are level II studies. References 3, 17, 18, 20, 22, 23, 26, 27, 30-32, 34, 35, and 37 are level IV studies. References 5-16, 25, 29, and 33 are level V expert opinion.

References printed in bold type are those published within the past 5 years.

1. Chapman KL, Briggs MD, Mortier GR: Review: Clinical variability and genetic heterogeneity in multiple epiphyseal dysplasia. Pediatr Pathol Mol Med 2003;22(1):53–75.
2. Czarny-Ratajczak M, Lohiniva J, Rogala P, et al.: A mutation in COL9A1 causes multiple epiphyseal dysplasia: Further evidence for locus heterogeneity. Am J Hum Genet 2001;69(5):969–980.
3. Fairbank T: Dysplasia epiphysialis multiplex. Br J Surg 1947;34(135):225–232.
4. Ballhausen D, Bonafé L, Terhal P, et al.: Recessive multiple epiphyseal dysplasia (rMED): Phenotype delineation in eighteen homozygotes for DTDST mutation R279W. J Med Genet 2003;40(1):65–71.
5. Chapman KL, Mortier GR, Chapman K, Loughlin J, Grant ME, Briggs MD: Mutations in the region encoding the von Willebrand factor A domain of matrilin-3 are associated with multiple epiphyseal dysplasia. Nat Genet 2001;28(4):393–396.
6. Briggs MD, Hoffman SM, King LM, et al.: Pseudoachondroplasia and multiple epiphyseal dysplasia due to mutations in the cartilage oligomeric matrix protein gene. Nat Genet 1995;10(3):330–336.
7. Ingram RR: Early diagnosis of multiple epiphyseal dysplasia. J Pediatr Orthop 1992;12(2):241–244.
8. Ingram RR: The shoulder in multiple epiphyseal dysplasia. J Bone Joint Surg Br 1991;73(2):277–279.
9. Treble NJ, Jensen FO, Bankier A, Rogers JG, Cole WG: Development of the hip in multiple epiphyseal dysplasia: Natural history and susceptibility to premature osteoarthritis. J Bone Joint Surg Br 1990;72(6):1061–1064.
10. Unger S, Bonafé L, Superti-Furga A: Multiple epiphyseal dysplasia: Clinical and radiographic features, differential diagnosis and molecular basis. Best Pract Res Clin Rheumatol 2008;22(1):19–32.
11. Briggs MD, Chapman KL: Pseudoachondroplasia and multiple epiphyseal dysplasia: Mutation review, molecular interactions, and genotype to phenotype correlations. Hum Mutat 2002;19(5):465–478.
12. Dahlqvist J, Orlén H, Matsson H, Dahl N, Lönnerholm T, Gustavson KH: Multiple epiphyseal dysplasia. Acta Orthop 2009;80(6):711–715.
13. Kim OH, Park H, Seong MW, et al.: Revisit of multiple epiphyseal dysplasia: Ethnic difference in genotypes and comparison of radiographic features linked to the COMP and MATN3 genes. Am J Med Genet A 2011;155A(11):2669–2680.
14. Jackson GC, Mittaz-Crettol L, Taylor JA, et al.: Pseudoachondroplasia and multiple epiphyseal dysplasia: A 7-year comprehensive analysis of the known disease genes identify novel and recurrent mutations and provides an accurate assessment of their relative contribution. Hum Mutat 2012;33(1):144–157.
15. Briggs MD, Wright MJ, Mortier GR: Multiple epiphyseal dysplasia, dominant, in Pagon RA, Bird TD, Dolan CR, et al., eds: GeneReviews. Seattle, WA, University of Washington, Seattle.,2013. Accessed November 25, 2015.
16. Bonafé L, Mittaz-Crettol L, Ballhausen D, Superti-Furga A: Multiple epiphyseal dysplasia, recessive, in Pagon RA, Bird TD, Dolan CR, et al., eds: GeneReviews. Seattle, WA, University of Washington, Seattle.,2014. Accessed November 25, 2014.
    17. Mäkitie O, Mortier GR, Czarny-Ratajczak M, et al.: Clinical and radiographic findings in multiple epiphyseal dysplasia caused by MATN3 mutations: Description of 12 patients. Am J Med Genet A 2004;125A(3):278–284.
    18. Bajuifer S, Letts M: Multiple epiphyseal dysplasia in children: Beware of overtreatment! Can J Surg 2005;48(2):106–109.
    19. Akhmedov B, Sung KH, Chung CY, Lee KM, Park MS: Reliability of lower-limb alignment measurements in patients with multiple epiphyseal dysplasia. Clin Orthop Relat Res 2012;470(12):3566–3576.
    20. Miura H, Noguchi Y, Mitsuyasu H, et al.: Clinical features of multiple epiphyseal dysplasia expressed in the knee. Clin Orthop Relat Res 2000;380:184–190.
    21. Lim SJ, Park YS, Moon YW, Jung SM, Ha HC, Seo JG: Modular cementless total hip arthroplasty for multiple epiphyseal dysplasia. J Arthroplasty 2009;24(1):77–82.
    22. Hodkinson HM: Double patellae in multiple epiphysial dysplasia. J Bone Joint Surg Br 1962;44:569–572.
    23. van Mourik JB, Weerdenburg JP, Verhaar JA: Magnetic resonance imaging of multiple epiphyseal dysplasia (type 2) of the knee. Radiography 2001;7:61–64.
    24. GeneTests. Accessed December 15, 2014.
    25. Lachman RS, Krakow D, Cohn DH, Rimoin DL: MED, COMP, multilayered and NEIN: An overview of multiple epiphyseal dysplasia. Pediatr Radiol 2005;35(2):116–123.
    26. Unger SL, Briggs MD, Holden P, et al.: Multiple epiphyseal dysplasia: Radiographic abnormalities correlated with genotype. Pediatr Radiol 2001;31(1):10–18.
    27. Nakashima E, Ikegawa S, Ohashi H, Kimizuka M, Nishimura G: Double-layered patella in multiple epiphyseal dysplasia is not exclusive to DTDST mutation. Am J Med Genet A 2005;133A(1):106–107.
    28. Anderson IJ, Goldberg RB, Marion RW, Upholt WB, Tsipouras P: Spondyloepiphyseal dysplasia congenita: Genetic linkage to type II collagen (COL2AI). Am J Hum Genet 1990;46(5):896–901.
    29. Unger S, Hecht JT: Pseudoachondroplasia and multiple epiphyseal dysplasia: New etiologic developments. Am J Med Genet 2001;106(4):244–250.
    30. Guille JT, Lipton GE, Tsirikos AI, Bowen JR: Bilateral Legg-Calvé-Perthes disease: Presentation and outcome. J Pediatr Orthop 2002;22(4):458–463.
    31. Roy DR: Spectrum of intra-articular findings of the acute and subacute painful hip with multiple epiphyseal dysplasia/spondyloepiphyseal dysplasia. J Pediatr Orthop B 2011;20(5):284–286.
    32. Cho TJ, Choi IH, Chung CY, Yoo WJ, Park MS, Lee DY: Hemiepiphyseal stapling for angular deformity correction around the knee joint in children with multiple epiphyseal dysplasia. J Pediatr Orthop 2009;29(1):52–56.
    33. Sebik A, Sebik F, Kutluay E, Kuyurtar F, Ademoğlu Y: The orthopaedic aspects of multiple epiphyseal dysplasia. Int Orthop 1998;22(6):417–421.
    34. Li LY, Zhao Q, Ji SJ, Zhang LJ, Li QW: Clinical features and treatment of the hip in multiple epiphyseal dysplasia in childhood. Orthopedics 2011;34(5):352.
    35. Callaghan JJ, Brand RA, Pedersen DR: Hip arthrodesis: A long-term follow-up. J Bone Joint Surg Am 1985;67(9):1328–1335.
    36. Pavone V, Costarella L, Privitera V, Sessa G: Bilateral total hip arthroplasty in subjects with multiple epiphyseal dysplasia. J Arthroplasty 2009;24(6):868–872.
    37. Ramaswamy R, Kosashvili Y, Cameron H: Total hip replacement in patients with multiple epiphyseal dysplasia with a mean follow-up of 15 years and survival analysis. J Bone Joint Surg Br 2010;92(4):489–495.
    38. Clarius M, Jung AW, Streit MR, Merle C, Raiss P, Aldinger PR: Long-term results of the threaded Mecron cup in primary total hip arthroplasty: A 15-20-year follow-up study. Int Orthop 2010;34(8):1093–1098.
    © 2015 by American Academy of Orthopaedic Surgeons