Primary disorders of connective tissue encompass a diverse range of single gene disorders that alter the function of connective tissue cells and their extracellular matrices. The mutations have major gene effect sizes with autosomal dominant, autosomal recessive, or X-linked patterns of inheritance. The resultant connective tissue phenotypes can manifest during fetal life but usually do so for the first time during childhood, adolescence, and early adulthood. Many of the phenotypes involve the musculoskeletal system and thus patients are often referred to a pediatric rheumatologist. The reasons for referral are usually joint pain and/or swelling or restricted movement. Some are associated with significant impairment during growth and increased susceptibility to premature osteoarthritis in early adulthood.
Many studies have shown that different mutations of a given gene expressed by connective tissue cells may have different effect sizes with severe, moderate, and mild phenotypes. In general, the severity of the phenotypes correlates with the ages of clinical onset of the disorders with severe phenotypes manifesting earlier and milder phenotypes later. Severe phenotypes associated with particular mutations are usually of similar severity in other individuals with the same mutations. However, phenotypes of moderate and mild severity are often more variable in other individuals bearing the same mutations. It is likely that the latter variability is due to interactions with other gene variants and the environment that modify the phenotype to a greater extent than is usually observed in individuals bearing mutations with larger effect sizes. This makes the clinical diagnosis especially challenging in the absence of genetic testing.
Recent advances in mutational analyses have resulted in the identification of causative genes and their variants in an increasing number of primary connective tissue disorders. These and previous studies over the past few decades have identified genes that are important for normal musculoskeletal health—a musculoskeletal gene set. These findings are of great value in the diagnosis and care of children with primary connective tissue disorders. In addition, these findings have contributed to the identification of gene variants associated with common adult musculoskeletal phenotypes, such as primary osteoarthritis and postmenopausal osteoporosis. In the latter disorders, genetic variants contribute to about 60% to 80% of the phenotypes, but each of the numerous identified genetic variants accounts for only a small percentage of the total genetic contribution.
For many decades, the rare primary disorders of connective tissue and the common adult disorders of primary osteoarthritis and postmenopausal osteoporosis have been considered separately. However, it is likely that these common disorders of adulthood have their origins at conception and during the subsequent years leading up to their clinical manifestations in adulthood. As a result, an alternative approach is to consider the rare and common primary connective tissue disorders as part of a spectrum. For example, a genetic arthritis spectrum may comprise rare fetal and childhood disorders at one end and common forms of primary osteoarthritis at the other end. Such an integrated approach to these disorders is likely to enhance the identification of potential new therapies for children and adults with genetic disorders of the musculoskeletal (MSK) system. This chapter focuses on the primary disorders of connective tissue that manifest during childhood–the childhood part of such spectra.
Classifications of primary connective tissue disorders were started in the 1950s and have undergone regular revisions since then. In 1956, McKusick reported a structured approach to the classification of heritable disorders of connective tissue. A large number of these disorders alter the radiographic appearance of the skeleton and are referred to as skeletal dysplasias. The nosology and classification of genetic skeletal disorders was revised in 2010 by the International Skeletal Dysplasia Society. A smaller number of disorders produce connective tissue laxity syndromes, such as the Ehlers–Danlos syndrome, in which the major impact is on the structure and function of soft tissues such as ligaments, tendons, vessels, and skin. A revised nosology was produced by the Ehlers–Danlos National Foundation (United States) and the Ehlers–Danlos Support Group (United Kingdom) in 1997. A consortium, which arose out of the first international meeting on Ehlers–Danlos syndrome convened in Ghent in 2012, plans to update the current classification.
This chapter will describe some of the more important disorders of connective tissue that may present initially to a pediatric rheumatologist, with specific focus on skeletal dysplasias and Ehlers–Danlos syndrome.
The classification of skeletal dysplasias has undergone revisions as new clinical and molecular information has become available. The International Skeletal Dysplasia Society is responsible for the current revision, which was prepared in 2010 by an expert group with clinical, radiological, and molecular expertise. Four hundred fifty-six different conditions were included and placed into 40 groups defined by molecular, biochemical, and radiographic criteria—a hybrid system ( Table 54-1 ).
|GROUP NUMBER AND NAME OF DISORDER|
|1.||Fibroblast growth factor receptor 3 (FGFR3) chondrodysplasia group||2.||Type 2 collagen group and similar disorders|
|3.||Type 11 collagen group||4.||Sulfation disorders group|
|5.||Perlecan group||6.||Aggrecan group|
|7.||Filamin group and related disorders||8.||Transient receptor potential cation channel, subfamily V, member 4 (TRPV4) group|
|9.||Short-ribs dysplasias (with or without polydactyly) group||10.||Multiple epiphyseal dysplasia (MED) and pseudoachondroplasia (PSACH) group|
|11.||Metaphyseal dysplasias||12.||Spondylometaphyseal dysplasias (SMD)|
|13.||Spondyloepi(meta)physeal dysplasias (SE[M]D)||14.||Severe spondylodysplastic dysplasias|
|15.||Acromelic dysplasias||16.||Acromesomelic dysplasias|
|17.||Mesomelic and rhizomesomelic dysplasias||18.||Bent bones dysplasias|
|19.||Slender bone dysplasia group||20.||Dysplasias with multiple joint dislocations|
|21.||Chondrodysplasia punctata (CDP) group||22.||Neonatal osteosclerotic dysplasias|
|23.||Increased bone density group (without modification of bone shape)||24.||Increased bone density group with metaphyseal and/or diaphyseal involvement|
|25.||Osteogenesis imperfecta and decreased bone density group||26.||Abnormal mineralization group|
|27.||Lysosomal storage diseases with skeletal involvement (dysostosis multiplex group)||28.||Osteolysis group|
|29.||Disorganized development of skeletal components group||30.||Overgrowth syndromes with skeletal development|
|31.||Genetic inflammatory/rheumatoid-like osteoarthropathies||32.||Cleidocranial dysplasia and isolated cranial ossification defects group|
|33.||Craniosynostosis syndromes||34.||Dysostoses with predominant craniofacial involvement|
|35.||Dysostoses with predominant vertebral with or without costal involvement||36.||Patellar dysostoses|
|37.||Brachydactylies (with or without extraskeletal manifestations)||38.||Limb hypoplasia—reduction defects group|
|39.||Polydactyly-syndactyly-triphalangism group||40.||Defects in joint formation and synostoses|
The large number of skeletal dysplasias was grouped in accordance with their phenotypic or molecular characteristics. Groups 1 to 8 are defined by their molecular characteristics. They include disorders with heterogeneous, although often overlapping, clinical phenotypes due to molecular anomalies of fibroblast growth factor receptor 3 ( FGFR3 ), type 2 collagen, type 11 collagen, sulfation, perlecan, aggrecan, filamin, and transient receptor potential cation channel, subfamily V, and member 4 ( TRPV4 ). Each of the remaining groups is defined by its clinical and radiological phenotypes. Ambiguities in the current hybrid classification are likely to be resolved as disease-genes are identified in the dysplasias without currently assigned genotypes. Three classifications of skeletal dysplasias may emerge: one using the clinical phenotypes, a second using the molecular characteristics, and a third using both. Each of the classifications may have their strengths and weaknesses depending on whether they are being used for clinical or research purposes.
There are many comprehensive resources available that can help with the approach to the diagnosis of skeletal dysplasias. For example, many relevant databases are available online from the National Center for Biotechnology Information (NCBI), USA. Included within its many options is the “Online Mendelian Inheritance in Man” (OMIM) database (available at www.ncbi.nlm.nih.gov/omim ), which provides up-to-date summaries of each skeletal dysplasia and their associated gene variants. Links are embedded within the summaries to many related databases. Other useful diagnostic resources include: Pictures of Standard Syndromes and Undiagnosed Malformations (POSSUM), available from the Murdoch Research Institute, Melbourne ( www.possum.net.au/ ), and the Winter-Baraitser Dysmorphology Database from London Medical Databases Ltd (available at www.lmdatabases.com/ ). Key phenotypic features are entered and a list of possible diagnoses is presented in sorted order. Both systems can display clinical photographs and radiographs for each of the listed potential diagnoses thus allowing the user to rapidly include and exclude potential diagnoses.
Thorough assessment of a patient with a suspected skeletal dysplasia involves detailed prenatal, perinatal, postnatal, and family histories as well as detailed general and musculoskeletal examinations. A review of growth charts and head circumference is extremely helpful and important in considering a diagnosis.
Many skeletal dysplasias produce typical dysmorphic clinical features that are instantly recognizable to expert clinicians. Achondroplasia with its rhizomelic short stature and characteristic craniofacial and hand features is an example of a relatively common skeletal dysplasia that is easily diagnosed clinically. However, dysplasias that produce mild short stature without dysmorphic features cannot be diagnosed clinically. Genetic testing is extremely important to confirm the clinical impression and to provide genetic counseling and management.
Diagnostic imaging is an essential requirement in the evaluation of children with a suspected skeletal dysplasia. A complete skeletal survey is recommended and should include plain radiographs of the lateral skull; lateral thoracic and lumbar spines; thorax; pelvis, including the hips; long bones; and hands. Detailed evaluations of the size, shape, and structure of each bone and joint are undertaken. The findings are often summarized in accordance with the predominant pattern of changes in the skeleton, such as spondylo, epiphyseal, metaphyseal, and diaphyseal abnormalities. These terms may be grouped together to reflect the diversity of skeletal phenotypes. Examples include multiple epiphyseal dysplasia, spondyloepiphyseal dysplasia, and spondyloepimetaphyseal dysplasia. Other patterns may involve changes in the density and shape of bones of the appendicular and axial skeletons.
The clinical plus radiographic features are usually sufficient to enable a diagnosis or differential diagnosis to be made. It is important to note that the radiographic features often change with growth. Many of the typical radiographic features may not be manifested early, while many other features may be lost following skeletal maturity. Consequently, it is advisable to assess all radiographs because they may provide valuable diagnostic information concerning epiphyseal, physeal, and metaphyseal growth abnormalities. Similarly, serial radiographs of selected bones are often useful in establishing a diagnosis in a child when their initial clinical and radiographic evaluations are inconclusive.
Magnetic resonance imaging (MRI) and computerized tomography (CT) are rarely used to establish the diagnosis of a skeletal dysplasia. However, they are often used to evaluate specific complications that can usually be predicted when the underlying diagnosis is known. For example, MRI is often used to evaluate cervical spinal cord compression in patients with achondroplasia or various forms of spondyloepiphyseal dysplasia as a result of stenosis of the foramen magnum or instability of the atlantoaxial joints.
Biochemical investigations should be performed in children with suspected rickets, mucopolysaccharidoses, mucolipidoses, or chondrodysplasia punctata. These investigations, coupled with molecular diagnosis, should be undertaken as soon as these dysplasias are suspected so that appropriate treatments can also be commenced early, when they are most effective.
Most of the diagnoses can be made without pathological confirmation. However, histology of autopsy material, usually skeletal tissues from lethal skeletal dysplasias, can confirm clinical and radiographic diagnoses. Postnatal iliac crest cartilage biopsy has limited value in the diagnosis of chondrodysplasias, but quantitative histomorphometry of iliac crest bone is routinely undertaken in many centers to monitor bisphosphonate therapy of patients with osteopenia due to osteogenesis imperfecta and related conditions. Histological studies of excised tissue from dysplasias that predispose to malignancy, for example multiple hereditary exostosis, is routine. Bone marrow histology is also undertaken in skeletal dysplasias that are associated with bone marrow anomalies including predisposition to malignancies such as the leukemias and lymphomas. Examples include patients with cartilage-hair hypoplasia or Schwachman–Bodian–Diamond syndrome.
Molecular diagnosis is an important tool for the confirmation of the clinical and radiological diagnosis of a skeletal dysplasia. It is valuable in confirming inheritance patterns and risks of recurrence as well as for genetic counseling. Molecular diagnosis has also rapidly expanded the amount of genetic information concerning skeletal dysplasias and the wide spectrum of genes involved in normal skeletal development.
Few of the molecular advances in the skeletal dysplasias have been translated into specific therapies directed to the disease genes, the disease gene products, or to physiological or biochemical pathways that are impaired by the mutations. In mucopolysaccharidoses, enzyme replacement therapy by administration of exogenous enzyme or by endogenous production from transplanted allogeneic bone marrow hematopoietic stem cells, may normalize hepatic and splenic abnormalities but is less effective in normalizing the skeletal phenotypes. In contrast, the bone phenotype and bone marrow function improves in some children with severe forms of osteopetrosis treated by allogeneic bone marrow hematopoietic stem cell transplantation.
Therapies for some of the other skeletal dysplasias involve pharmacological modulation of physiological processes that are impaired because of the mutations, and include the administration of phosphate and vitamin D to patients with various forms of genetic rickets and the administration of bisphosphonates to decrease the abnormally high levels of bone turnover in many forms of osteogenesis imperfecta. Recent studies in a mouse model of osteogenesis imperfecta provided evidence that a neutralizing antibody to the osteocyte secreted protein, sclerostin, increased bone formation and, as a result, it may be suitable for clinical trials as a much needed bone anabolic agent for this disorder.
For the majority of patients with skeletal dysplasias there are currently no specific treatments. Rehabilitation services are frequently required because of the musculoskeletal impairments. Surgical treatments may be needed to correct progressive skeletal deformity, stenosis, or instability. Surgery often has high complication and recurrence rates. Total joint replacements may be needed for premature osteoarthritis in patients with epiphyseal dysplasias. Other specialists, such as ophthalmologists, otolaryngologists, and endocrinologists, may need to be involved depending on the extraskeletal manifestations. Some dysplasias are associated with bone marrow dysfunction and the risk of malignancies. Careful surveillance of such patients enables preventative care and early treatments to be provided.
Some of the skeletal dysplasias that may present to the pediatric rheumatologist because of musculoskeletal pain or restricted joint range of motion are discussed below. Some features that can lead to suspicion of a suspected skeletal disorder versus juvenile idiopathic arthritis (JIA) are shown in Box 54-1 .
More than one family member affected
Family history of joint replacement at an early age (<40 years old)
Absence of evidence of systemic or synovial inflammation
Absence of ANA and rheumatoid factor
Absence of joint erosions despite chronic disease
Presence of two or more dysmorphic features
Multiple Epiphyseal Dysplasia
The multiple epiphyseal dysplasias (MED) have been selected for review as a typical example of a group of skeletal dysplasias. The anomalies of epiphyseal and physeal development are shared with large numbers of other skeletal dysplasias such as the heterogeneous spondyloepiphyseal and spondyloepimetaphyseal dysplasias.
The MEDs (see Tables 54-1 and 54-2 ) are a relatively common group of conditions that are included within groups 4, 6, and 10 of the current classification of genetic skeletal disorders. They are characterized by abnormal development of the epiphyses of the appendicular skeleton ( Figs. 54-1 & 54-2 ) with mild or no visible changes in the axial skeleton.
|NAME OF DISORDER||INHERITANCE||GENE||PROTEIN|
|Multiple epiphyseal dysplasia type 1 [EDM 1]||AD||COMP||Cartilage oligomeric matrix protein [COMP]|
|Multiple epiphyseal dysplasia type 2 [EDM 2]||AD||COL9A2||Collagen 9, α2 chain|
|Multiple epiphyseal dysplasia type 3 [EDM 3]||AD||COL9A3||Collagen 9, α3 chain|
|Multiple epiphyseal dysplasia type 4 [EDM 4]||AR||DTDST||SLC26A2 sulfate transporter|
|Multiple epiphyseal dysplasia type 5 [EDM 5]||AD||MATN3||Matrilin 3|
|Multiple epiphyseal dysplasia type 6 [EDM 6]||AD||COL9A1||Collagen 9, α1 chain|
|Multiple epiphyseal dysplasia, other types||AD|
|Familial osteochondritis dissecans||AD||AGC1||Aggrecan|
|Stickler syndrome, recessive type||AR||COL9A1||Collagen 9, α1 chain|
|Familial hip dysplasia (Beukes)||AD|
|Multiple epiphyseal dysplasia with microcephaly and nystagmus (Lowry-Wood)||AR|
The autosomal dominant forms are more common than the autosomal recessive forms. Children with the autosomal dominant forms appear to be normal at birth, although families with affected members are often able to identify affected babies because of their short and broad hands and feet. In early childhood, parents often note slowing of longitudinal growth, painful hips and knees, altered gait, and genu valgum (see Fig. 54-2 ). At initial presentation, these changes may be quite mild. Nonetheless, a skeletal survey may reveal delayed and abnormal development of multiple small and misshapen upper and lower limb epiphyses. The femoral heads may show similar changes but may also show multiple small centers of ossification that coalesce in late childhood and adolescence. Many such children are referred with a diagnosis of Perthes disease, a form of avascular necrosis of the hip. A skeletal survey is essential because of the difficulty in distinguishing Perthes disease from multiple epiphyseal dysplasia if radiological studies are limited to the pelvis and hips. This distinction is also important because some patients with multiple epiphyseal dysplasia may develop typical changes of Perthes disease in one or both hips during childhood. The vertebrae are often normal but may be ovoid with mild irregularity of the vertebral endplates.
Table 54-2 lists the known genes associated with autosomal dominant forms of multiple epiphyseal dysplasia. They include the genes encoding cartilage oligomeric matrix protein (COMP), matrilin 3, and the three α-chains of collagen 9. Although their clinical and radiological features are similar, there are also some differences. For example, joint laxity and a mild myopathy are found in those with COMP mutations, MED1, because of the expression of COMP protein in ligaments, tendons and muscles as well as in hyaline cartilage and bone. Patients with MED1 also have ovoid vertebral bodies and mild irregularity of the vertebral end plates. Their clinical and radiological features indicate that MED1 is a mild form of pseudoachondroplasia (PSACH) which is a form of spondyloepimetaphyseal dysplasia also caused by mutations of COMP ( Fig. 54-3 ). Muscular weakness, usually without the joint laxity of MED1, is also observed in some patients with mutations of COL9A2 and COL9A3. In the latter patients, the muscle weakness involves the proximal muscles of the limbs.
The radiographic features of familial osteochondritis dissecans caused by an autosomal dominant mutation of AGC1, that encodes aggrecan, overlap with those of MED2 due to mutations of COL9A2 in that some epiphyses show features typical of multiple epiphyseal dysplasia while others show features typical of osteochondritis dissecans.
All forms of autosomal dominant multiple epiphyseal dysplasia are associated with progressive skeletal impairments. Some patients develop progressive genu valgum, which may require surgical correction, usually lateral hemiepiphyseodesis of the distal femoral physes (see Fig. 54-2 ). Progressive osteoarthritis, particularly of the hips and knees, is common. Total joint replacements of hips or knees or both are often needed in the fourth decade but may be needed earlier, particularly in those with severe deformities of the femoral heads at skeletal maturity.
It is often difficult to diagnose the particular type of autosomal dominant multiple epiphyseal dysplasia based on clinical and radiological findings alone. Specific clinical features such as ligament laxity, muscle weakness, and the pattern of skeletal findings may be helpful in suggesting the specific type of MED that the patient has. However, clinical and radiographic typing is often inaccurate because of overlap in phenotypes between the types and because of phenotypic variability within each type. Consequently, each of the genes associated with autosomal dominant multiple epiphyseal dysplasia and familial osteochondritis dissecans need to be included in the genetic analyses. Despite this, molecular studies of a large cohort of patients with autosomal dominant multiple epiphyseal dysplasia did not identify mutations in the known genes in over one half of a series of 29 patients with MED. The latter finding suggests that many autosomal dominant mutations may reside beyond the sequenced regions of the known MED genes or reside in other genes. The latter patients are classified as multiple epiphyseal dysplasia—other types ( Table 54-2 ).
Multiple epiphyseal dysplasia type 4 (MED4) is the most common autosomal recessive variant. Patients with MED4 are born with club feet and later develop clinical and radiological features of multiple epiphyseal dysplasia. Some patients also have recurrent subluxation and dislocation of the patellae. Homozygous or double heterozygous mutations of the DTDST ( SLC26A2 ) gene, which encodes a sulfate transporter, have been identified in patients with MED4. The femoral heads are often severely affected and progressive osteoarthritis requiring total hip joint replacements in the second decade may develop. A double-layered patella is common in patients with MED. It is noted on lateral radiographs of the knee as a result of separate anterior and posterior centers of ossification. However, the patella findings are not confined to patients with MED because similar radiographic findings have been observed in patients with pseudoachondroplasia due to autosomal dominant mutations of COMP and in MED2 due to autosomal dominant mutations of COL9A2.
Examples of Syndromes Associated with Joint Hypermobility
Joint mobility changes with increasing age. When the joints are excessively mobile children may develop joint pain with or without swelling. The Beighton criteria (see Chapter 51 ) are used to define patients with the hypermobile joint syndrome. A number of well-defined syndromes are associated with hypermobile joints and must be recognized by the pediatric rheumatologist.
The Ehlers–Danlos syndrome is heterogeneous with severe, moderate, and mild soft tissue laxity phenotypes, particularly affecting the dermis and joints. The current numerical classification is shown in Table 54-3 and is based on clinical phenotypic patterns and gene mutations that resulted in abnormal fibrillogenesis of the collagen fibrils found in affected soft tissues. However, the classification needs to be upgraded because some of the previously included types were poorly characterized or were reclassified. In addition, many newly recognized types of Ehlers–Danlos syndrome, often associated with a more diverse range of molecular anomalies, need to be incorporated into the classification or reclassified elsewhere. The new types are included within the group labeled “other” in Table 54-3 .
|NUMBER||NAME||GENES||INHERITANCE||MAIN CLINICAL FEATURES|
|I||Classical (gravis)||COL5A1||AD||Severe joint and skin laxity; bruising; poor skin healing|
|II||Classical (mitis)||COL5A2||AD||Milder form of classical (gravis) EDS|
|AD||Marked joint laxity with minor skin anomalies|
|IV||Vascular||COL3A1||AD||Easy bruising; vascular and bowel ruptures; thin skin|
|VIA||Kyphoscoliosis||PLOD1||AR||Joint hypermobility, severe kyphoscoliosis with fragility of skin and eyes|
|VIB||Musculocontractural||CHST14||AR||Digit contractures; hypermobility; scoliosis; thin, lax skin; and ocular anomalies|
|VIIA||Arthrochalasia multiplex congenita||COL1A1||AD||Severe joint hypermobility and hip dislocations|
|VIIB||Arthrochalasia multiplex congenita||COL1A2||AD||Severe joint hypermobility and hip dislocations|
|VIIC||Dermatosparaxis||ADAMTS2||AR||Severe skin fragility, joint hypermobility, and blue sclerae|
|VIII||Periodontitis||Unknown but one locus at 12p13||AD||Periodontal loss, soft skin, and joint hypermobility|
|Progeroid||B4GALT7||AR||Dysmorphic with thin elastic skin|
|B3GALT6 related||B3GALT6||AR||Skin fragility, joint laxity, contractures, and spondyloepimetaphyseal dysplasia|
|Cardiac valvular||COL1A2||AR||Cardiac valve incompetence with skin and joint laxity|
|Classical EDS with vascular ruptures||COL1A1||AD||Features of type I/II EDS and vascular ruptures|
|FKBP14 related||FKBP14||AR||Scoliosis, joint hypermobility, hearing loss, and myopathy|
|Spondylocheiro dysplasia||SLC39A13||AR||Lax skin, easy bruising, spondyloepiphyseal dysplasia|
|Tenascin-X deficient||TNXB||AR||Joint laxity, skin laxity, easy bruising with normal skin healing|
|Periventricular heterotopia||FLNA||XL||Joint laxity with periventricular heterotopia|