Skeletal dysplasias and connective tissue disorders represent a heterogenous group of orthopaedic pathologies resulting from an array of genetic mutations and chromosomal anomalies affecting the growth and development of connective tissues, including bone. Minute changes to the structure of the physis, composition of connective tissues, or the process of ossification can have far-reaching effects on skeletal development. Orthopaedic surgical management of these children requires insight into their potential for growth and development, as well as consideration of their medical comorbidities.

Marfan syndrome results from a mutation in the gene encoding for fibrillin, an extracellular glycoprotein involved in the formation of elastic fibers in connective tissues. A consequent increase in growth factor availability is responsible for changes in the mechanical properties of soft tissues in patients with Marfan syndrome, most importantly affecting the aortic root and the ocular lens. Classically, patients with Marfan syndrome have tall stature, with long thin limbs and spiderlike digits (ie, arachnodactyly). Inheritance is commonly autosomal dominant, although spontaneous mutations occur in 15% to 30% of cases.¹

The revised Ghent nosology emphasizes family history, aortic root aneurysm, and ectopia lentis as cardinal features of Marfan syndrome² (Table 1). Diagnosis also involves a systemic score, including the following musculoskeletal manifestations: reduced elbow extension, wrist and thumb signs, pectus deformity, pes planovalgus, protrusio acetabuli, and scoliosis. Spine deformity is a common presenting feature of Marfan syndrome and, consequently, orthopaedic surgeons should be prepared to screen for this underlying condition. Patients with findings of Marfan syndrome should be referred for genetic workup, as well as ophthalmologic and cardiac evaluations to rule out associated abnormalities (eg, lens dislocation, aortic dilatation) that can lead to significant morbidity and mortality.

Although treatment algorithms for scoliosis in Marfan syndrome are comparable with those of idiopathic scoliosis, several distinguishing features should be noted. From
an anatomic standpoint, patients with Marfan syndrome can have dural ectasia, associated small pedicles, and osteopenia, all of which add challenge to spinal instrumentation. Preoperative axial imaging (CT and MRI) is therefore recommended for this population (Figure 1). According to a 2021 study, females and patients with positive wrist signs have been noted to be at particular risk for progression to severe scoliosis.³ Hypokyphosis, increased Cobb angle, and chest wall deformity are associated with reduced preoperative pulmonary function.⁴ Brace treatment appears to be less effective than for idiopathic scoliosis, with reported success of only 17% for mild to moderate curves.⁵

From a surgical standpoint, a 2019 study showed that patients with Marfan syndrome have a 2.4% risk of neurologic complication from spinal fusion.⁶ Surgeons should be cautious if using traction because vertebral subluxation can occur or worsen, especially in the presence of kyphosis.⁴ Other surgical risks include dural tear, infection, pseudarthrosis, curve decompensation, and cardiovascular complication. In combination, these factors lead to a 10% rate of readmission within 90 days of discharge after spinal fusion.⁷ Although growth-friendly spinal instrumentation effectively increases thoracic height in patients with Marfan syndrome with early-onset scoliosis, a 2021 study has shown that there is a particularly high rate of implant failure during their treatment.⁸

Osteogenesis imperfecta is a genetic connective tissue disorder with a suspected incidence of 1:20,000 characterized by low bone density, fractures, spine and extremity deformity, and several extraskeletal manifestations.⁹,¹⁰ It is genetically and phenotypically heterogeneous with a wide range of clinical severity.

The widely used Sillence classification was initially described in 1979 based on clinical features before the characterization of the underlying genetic etiology.¹¹ Four types were described: type I: mild, nondeforming; type II: lethal perinatal; type III: severe, progressively deforming; and type IV: phenotypically variable, white sclera.

In the early 1980s, type I collagen mutations were first associated with autosomal dominant osteogenesis imperfecta. COL1A1 and COL1A2 code for type I collagen, a heterotrimer containing two alpha1 chains and one alpha2 chain that form a triple helix and provide strength to the extracellular matrix of bone. Osteogenesis imperfecta may be caused by an issue with the quality or quantity of type I collagen. A clinically distinct form of osteogenesis imperfecta, type V, was identified in patients demonstrating calcification of the interosseous membrane, hyperplastic callus, and an autosomal dominant inheritance not associated with type I collagen mutations, subsequently found to be related to an IFITM5 mutation.¹² Over the past decade, an expanding number of recessive forms of osteogenesis imperfecta responsible for approximately 15% of cases have been identified, mostly caused by mutations in genes encoding proteins involved in the synthesis or processing of type I collagen.¹³

There is currently no cure for osteogenesis imperfecta. The mainstays of medical management include
nutritional optimization of calcium intake and vitamin D levels and activity. There is a known positive correlation between muscle strength and bone strength in the setting of osteogenesis imperfecta. Diphosphonates are a class of antiresorptive drugs shown to increase bone mass, improve vertebral size and shape, potentially reduce fracture frequency, and anecdotally improve bone pain.¹⁴ Additional agents are currently being investigated in the setting of osteogenesis imperfecta, including sclerostin antibody as an anabolic treatment and denosumab, a human monoclonal antibody that blocks RANKL, an essential cytokine in the osteoclastogenesis pathway.

Most fractures in children with osteogenesis imperfecta may be managed nonsurgically with a period of immobilization. To avoid a cycle of muscle weakness, disuse osteopenia, and a fracture cluster, immobilization periods should be as brief as possible permitted by early healing. Surgical intervention in the form of realignment and intramedullary rodding is indicated for progressive long bone deformity interfering with motor development or function or associated with recurrent fractures, otherwise operatively indicated fractures and symptomatic nonunions¹⁵,¹⁶ (Figure 2). The concept of multiple osteotomies and intramedullary fixation has long been used, and this surgical concept continues to be used with less invasive osteotomies. Both fixed-length rods and telescopic rods, currently most commonly the Fassier-Duval implant, have been described.¹⁷,¹⁸,¹⁹ A 2020 multicenter review demonstrated that most individuals with moderate and severe forms of osteogenesis imperfecta undergo rodding procedures and that individuals with severe osteogenesis imperfecta who underwent rodding have improved mobility outcomes and lower fracture rates.²⁰ A meta-analysis of 359 primary nonelongating rodding procedures of femurs and
tibias in children with osteogenesis imperfecta with a mean follow-up of 63 months showed a revision surgery rate of 39.4%.²¹ The aim of telescopic rods such as the Fassier-Duval rod is to reduce the number of revisions required because of growth; however, telescopic rods are notable for a similar revision and complication rate. The authors of a 2019 study²² reported the use of Fassier-Duval rods over static implants in pediatric patients with osteogenesis imperfecta to demonstrate a higher implant survival during the first 48 months after index surgery as well as a decrease in the incidence of total number of surgeries (planned and unplanned) among Fassier-Duval rods compared with static rods. Additionally, transfixion pin backout and prominence are common in the setting of Fassier-Duval rods.²³ Center-center position of the male component of the Fassier-Duval rod in the epiphysis of the distal femur and tibia contributes to rod longevity.²⁴ There is a growing body of data that tourniquets and noninvasive blood pressure cuffs may be used for children with osteogenesis imperfecta.²⁵,²⁶

Acetabular protrusio, more common in those with more clinically severe osteogenesis imperfecta, is progressive over time and remains a challenging problem.²⁷,²⁸ It is also associated with femoral neck fractures and there is to be a high index of suspicion in diagnosing nondisplaced femoral neck fractures, particularly when nondisplaced.²⁹ Elbow deformities are also common in the setting of osteogenesis imperfecta including radial head dislocation, most common in patients with osteogenesis imperfecta type V. Another issue about the elbow includes olecranon fractures. A 2019 review of 358 patients found an incidence of olecranon fractures of 8.1%, predominantly in type I patients with a 41% chance of sustaining a contralateral olecranon fracture within the subsequent 5 months.³⁰

Spinal manifestations in osteogenesis imperfecta include scoliosis, kyphosis, craniocervical junction abnormalities (ie, basilar invagination), spondylolysis, and spondylolisthesis. The prevalence of scoliosis in the population of patients with osteogenesis imperfecta
ranges from 39% to 80%, more prevalent in the more severe forms of osteogenesis imperfecta. It is known to progress if untreated well into adulthood. Spinal fusion to halt curve progression is considered when curves reach 45°, but the patient’s age and truncal height need to be considered to avoid thoracic insufficiency syndrome. Contemporary techniques and a multimodal strategy to address the poor bone quality have shown promising results for spinal fusion in the setting of osteogenesis imperfecta. There have also been recent reports of growth-friendly surgical treatment of scoliosis in the setting of osteogenesis imperfecta.³¹

Craniocervical junction abnormalities are seen in 37% of patients with osteogenesis imperfecta, including basilar invagination (13%), basilar impression (15%), and platybasia (29%).³² Skull base abnormalities are correlated with the severity of disease and older age. Advanced imaging modalities including CT and MRI are recommended to best understand the complex anatomy that can be quite difficult to discern on plain radiography. Basilar invagination may incur debilitating neurologic consequences, and ongoing screening of the osteogenesis imperfecta population, particularly those more severely affected, is an important aspect of care.³³ Management of basilar invagination may be by odontoid resection, foramen magnum decompression, or both and may require occipitocervical fusion with adjunctive preoperative halo traction.

Neurofibromatosis type 1 is a common, autosomal dominant single-gene disorder affecting production of neurofibromin, a protein implicated in skeletal growth and development. It is defined using criteria established in 1987 by the National Institutes of Health (Table 2), with at least two such findings required for diagnosis. Notably, these features may develop at different ages, meaning that orthopaedic surgeons should maintain suspicion for this underlying condition as they follow very young patients meeting only one criterion (eg, tibial dysplasia). Common orthopaedic manifestations of neurofibromatosis type 1 include scoliosis, dysplasia of long bones, limb overgrowth, and malignant transformation of tumors.

Scoliosis in neurofibromatosis type 1 can be categorized as dystrophic or nondystrophic. Dystrophic curves are typically short, sharp thoracic curves that progress rapidly and are associated with characteristic radiographic findings (Table 3). When these features are present, preoperative CT and MRI provide valuable information regarding the presence of dural ectasia, rib head dislocations, and paraspinal masses (Figure 3). Early surgical management is generally undertaken for dystrophic scoliosis, although there remains some debate surrounding the relative merits of definitive fusion versus growth-friendly surgery in the early-onset neurofibromatosis type 1 population.³⁴,³⁵,³⁶ When growing rods are used, dual-rod constructs are preferred because of a high rate of implant-related complications.³⁷ In contrast, nondystrophic curves can be managed with similar principles to idiopathic scoliosis, with careful observation for potential modulation (ie, development of dystrophic features) over time.

Tibial dysplasia is commonly associated with neurofibromatosis type 1 and can present in infancy as an anterolateral tibial bow, with or without pseudarthrosis (ie, congenital pseudarthrosis of the tibia). In this condition, the dysplastic site is surrounded by hamartoma that predisposes to fracture and complicates healing. When tibial dysplasia presents in a prefracture state, treatment has traditionally involved protective clamshell bracing; however, distal tibial guided growth has recently shown encouraging results in preventing fracture while promoting remodeling.³⁸ Once fracture occurs, surgical treatment generally aims to achieve and maintain union, while correcting the underlying deformity. Therefore, aggressive débridement of the pathologic tissue is advocated, followed by autografting and stabilization with intramedullary rods or external fixation.³⁹,⁴⁰ Recent literature has emphasized the importance of fibular procedures (including intentional cross union) in achieving union, as well as the potential adjunctive value of diphosphonate infusions and off-label use of bone morphogenetic protein.⁴¹,⁴² Patients undergoing multiple surgeries have been treated with vascularized fibular autografts, Ilizarov frames, and induced-membrane techniques, although amputation remains a possibility in these challenging scenarios.³⁷,⁴³,⁴⁴

Ehlers-Danlos syndrome (EDS) is a heterogeneous group of heritable connective tissue disorders affecting collagen production and metabolism. The most recent classification of EDS (2017) includes 13 subtypes, of which the classic (cEDS) and hypermobile (hEDS) subtypes are most commonly encountered in clinical practice.⁴⁵ Although each subtype is unique in its genetic basis, presenting symptomatology and diagnostic criteria, prominent features generally include joint hypermobility, skin hyperelasticity, and tissue fragility. Both cEDS and hEDS have autosomal dominant inheritance patterns, although the specific molecular basis of hEDS remains unknown.⁴⁵ Patients with EDS often present to orthopaedic surgeons with recurrent joint dislocations (eg, glenohumeral, patellofemoral) and chronic musculoskeletal pain.⁴¹,⁴⁶,⁴⁷ Less commonly, EDS can manifest with spinal deformity or instability.⁴²

The Beighton score is used in the clinical evaluation of generalized joint hypermobility and is also included in diagnostic criteria for EDS⁴⁵ (Table 4). The following Beighton score cutoffs are used (among other criteria) for hEDS: more than 6 points for prepubertal children, more than 5 for adults up to age 50 years, and more than 4
for patients older than 50 years.⁴⁵ Certainly, there exists a wide spectrum of joint laxity among orthopaedic patients, and principles relevant to EDS can likely be applied when treating other patients with high Beighton scores.⁴⁷

Management of recurrent joint instability in EDS often begins with physical therapy, focusing on improving active range of motion while strengthening dynamic stabilizers. If surgical management is considered, patients should be counseled regarding the potential for poor results with standard soft-tissue reconstructions.⁴⁶,⁴⁷ Although various adjunctive procedures (eg, allografts, bony augmentations, osteotomies) have been developed to enhance joint stability in patients with EDS, there are few high-quality studies reporting on their outcomes.⁴⁶ Similar principles should be used when planning arthroplasty for patients with EDS, including consideration of constrained components (eg, dual-mobility cups) to account for lax soft tissues.⁴⁸,⁴⁹ Careful, multilayered closure and appropriate postoperative immobilization can help to optimize wound healing.

Chronic joint pain can be a challenging problem in EDS, limiting function and affecting health-related quality of life.⁴⁵ A 2019 study has also shown higher rates and durations of opioid use among patients with EDS compared with matched control patients.⁴¹ These issues highlight the importance of maintaining a broad, multidisciplinary approach when treating patients with connective tissue disorders.

Achondroplasia is the most common form of skeletal dysplasia, affecting approximately 250,000 people worldwide. This form of disproportionate dwarfism is the result of a gain-of-function mutation in the fibroblast growth factor receptor 3 (FGFR3) gene,⁵⁰ resulting in increased tyrosine kinase activity and impaired cartilage differentiation and endochondral ossification. The resulting phenotype is characterized by short stature, rhizomelia, macrocephaly, and frontal bossing.

Other mutations of the FGFR3 gene can result in various associated skeletal dysplasias, including thanatophoric dysplasia, severe achondroplasia with developmental delay and acanthosis nigricans, hypochondroplasia, and Crouzon syndrome. Thanatophoric dysplasia is usually fatal by age of 2 years because of cardiopulmonary failure and is associated with severe rhizomelia, protuberant abdomen, and a small, restrictive thoracic cavity leading to cardiopulmonary failure.

Infants with achondroplasia historically had an increased risk of mortality, likely because of foramen magnum stenosis resulting in cervico medullary compression and central sleep apnea, although mortality rates now approach those of unaffected infants because of improved surveillance. Comprehensive care of these children requires regular clinical assessment for symptoms of compression, including detailed history of motor development, neurologic examination, and polysomnography. MRI is not necessary for every child with achondroplasia and should only be obtained to confirm suspected cases. Foramen magnum stenosis and treatment with surgical decompression is required in 20.5% of children with achondroplasia, with 10% requiring a second decompression.⁵¹

Thoracolumbar kyphosis develops in nearly all infants with achondroplasia and usually progresses until children are able to ambulate independently.⁵² This deformity is flexible in infants, and progression can be limited by educating the family on sitting modifications. Although treatment with bracing has been considered in the past, the evidence is hard to interpret in light of the natural history of thoracolumbar kyphosis in achondroplasia, with 73% spontaneous resolution within 1 year of walking⁵³ and 89% resolution by 10 years of age.⁵⁴ As the thoracolumbar kyphosis decreases in these children, there is a compensatory increase in lumbar lordosis and sacral slope, maintaining overall sagittal balance.⁵⁵ In those patients with persistent kyphosis, there is a risk of progression and exacerbation of existing spinal stenosis leading to myelopathy, progressing rarely to paraplegia. Although correction of thoracolumbar kyphosis can be performed by a combined anterior-posterior approach, most modern treatment is performed using an all-posterior approach with osteotomies, achieving correction without lengthening the spinal cord. Kyphosis that involves wedged and posteriorly translated apical vertebrae can be managed with vertebral
column resection, but according to a 2021 study, rates of neurologic injury and failure of instrumentation remain high (57%)⁵⁶ (Figure 4). Historically, indication for surgical correction of thoracolumbar kyphosis was deformity greater than 45°, and although modern indications remain undefined, surgery should be considered for progressive deformity greater than 60° and symptomatic spinal stenosis.