Clinical Description
Hereditary multiple osteochondromas (HMO) (also known as multiple hereditary exostoses [MHE]) is characterized by growths of multiple osteochondromas. Shortened stature compared to unaffected family members and angular deformities of the forearms and legs are common. The risk for malignant degeneration to osteochondrosarcoma increases with age, although the lifetime risk for malignant degeneration is low (~2%-10%). To date, more than 1,000 individuals with a pathogenic variant in EXT1 or EXT2 have been identified. The following description of the phenotypic features associated with this condition is based on these reports.
Table 2.
Hereditary Multiple Osteochondromas: Frequency of Select Features
View in own window
| Feature | % of Persons w/Feature | Comment |
|---|
| Osteochondromas | 100% | More lesions in persons w/EXT1-related HMO than in those w/EXT2-related HMO 1 |
| Shortened stature |
| More pronounced in persons w/EXT1-related HMO than EXT2-related HMO 2 |
| Angular deformities of forearms or legs | 40%-74% 3 | |
| Leg length discrepancy | 10%-50% 4 | |
| Chondrosarcoma | 2%-10% 5 | Predominantly localized to pelvis, scapula, proximal femur, & humerus; typically solitary, low-grade lesions |
HMO = hereditary multiple osteochondromas
- 1.
- 2.
- 3.
- 4.
- 5.
Onset. The proportion of individuals with HMO who have clinical findings increases from approximately 5% at birth to 96% at age 12 years [Legeai-Mallet et al 1997]. The median age at diagnosis is three years.
Osteochondromas. The number of osteochondromas varies widely even within families. Involvement is usually symmetric. The most commonly involved bones are the femur (30%), radius and ulna (13%), tibia (20%), and fibula (13%). Anatomic distribution and number of osteochondromas depends on genotype and sex of the affected person [Clement & Porter 2014b]. Osteochondromas grow in size and gradually ossify during skeletal development and stop growing with skeletal maturity, after which no new osteochondromas develop. One case report also suggests the possibility of regression of these lesions and symptomatic to asymptomatic transition with maturity. This supports the utilization of conservative management prior to skeletal maturity even for transiently symptomatic lesions [Ikeda et al 2025]. However, there is an overall lack of evidence that regression should be the expected outcome and families should be counseled appropriately.
Osteochondromas typically arise in the juxtaphyseal region of long bones and from the surface of flat bones (pelvis, scapula). An osteochondroma may be sessile or pedunculated. Sessile osteochondromas have a broad-based attachment to the cortex. The pedunculated variants have a pedicle arising from the cortex that is usually directed away from the adjacent growth plate. The pedunculated form is more likely to irritate overlying soft tissue, such as tendons, and compress peripheral nerves or blood vessels. The marrow and cancellous bone of the host bone are continuous with the osteochondroma.
Shortened stature. It has been stated that at least 40% of individuals with HMO have shortened stature (stature shorter than predicted based on the heights of unaffected parents and sibs). Although interference with the linear growth of the long bones of the leg often results in reduction of predicted adult height, the height of most adults with EXT2-related HMO and many with EXT1-related HMO falls within the normal range [Porter et al 2004]. Shortened stature is more pronounced in persons with EXT1-related HMO [Pedrini et al 2011, Clement et al 2012, Li et al 2017]. Height has been found to be directly proportional to leg length, and in many individuals with EXT1- and EXT2-related HMO, height is below the 10th centile [Li et al 2017]. Multivariate analysis determined that the presence of a distal femoral osteochondroma was an independent predictor of knee deformity, diminished knee joint range of motion, and short stature [Clement & Porter 2014a].
Bone deformity. Abnormal bone remodeling may result in shortening and bowing with widened metaphyses [Porter et al 2004]. Hand deformity resulting from shortened metacarpals is common. Abnormal growth and development of the forearm and leg in untreated individuals with HMO is common, including both proportionate and disproportionate shortening of the two bones of the forearm or leg, producing shortened and angulated limbs, respectively. In a study of 46 kindreds in Washington State, United States, 39% of individuals had a deformity of the forearm, 10% had an inequality in limb length, 8% had an angular deformity of the knee, and 2% had a deformity of the ankle [Schmale et al 1994]. Angular deformities (bowing) of the forearm and/or ankle are the most clinically significant orthopedic issues [Shin et al 2006]. Forearm deformities [Masada et al 1989, Jo et al 2017] and ankle deformities [Ahn et al 2019] have previously been classified. The classification reported by Ahn et al [2019] is the most widespread and accepted for HMO lower leg deformities. A combination of the Masada et al [1989] and Jo et al [2017] classifications are typically used for forearms with a preference for the Jo et al [2017] classification due to better reliability [Farr et al 2021]. Additionally, individuals with HMO and associated ankle deformities were reported to have up to 19% risk of early secondary arthritis by age 42 years [Noonan et al 2002]. A novel classification system has since been developed for forearm abnormalities caused by HMO that stratifies these deformities into three groups: high, moderate, and low risk of radial head dislocation. Individuals with distal ulnar lesions were classified as high or moderate risk based on radiographic parameter of proportional length (ulnar-to-radial length ratio), and those without distal ulnar lesions were considered low risk [Chan et al 2025].
Hip dysplasia frequently results from osteochondromas of the proximal femur and coxa valga. Decreased center-edge angles and increased uncovering of the femoral heads may lead to early thigh pain and abductor weakness and subsequent early arthritis [Makhdom et al 2014, Wang et al 2015]. Femoral-acetabular impingement may also arise from proximal femoral osteochondromas, limiting hip motion [Viala et al 2012, Higuchi et al 2016, Duque Orozco et al 2018].
Symptoms secondary to mass effect. Compression or stretching of peripheral nerves usually causes pain but may also cause sensory or motor deficits [Göçmen et al 2014, Onan et al 2014, Payne et al 2016]. Spinal cord compression and myelopathy from cervical osteochondromas have been reported [Aldea et al 2006, Giudicissi-Filho et al 2006, Pandya et al 2006, Ashraf et al 2013, Veeravagu et al 2017, Akhaddar et al 2018, Gigi et al 2019, Montgomery et al 2019], as has dysphagia from a cervical osteochondroma [Gulati et al 2013]. Bilateral inferior cervical osteochondromas have been found to produce neurogenic and vascular thoracic outlet syndrome [Abdolrazaghi et al 2018]. Syringomyelia and tethered cord / fibrolipoma in individuals undergoing screening imaging without evidence of spinal osteochondromas have also been described [Legare et al 2016]. Mechanical obstruction of joint motion may result from large osteochondromas impinging on the adjacent bone of a joint. Overlying muscles and tendons may be irritated or entrapped, resulting in pain and loss of motion [Andrews et al 2019]. Nerves and vessels may be displaced from their normal anatomic course, complicating attempts at surgical removal of osteochondromas. Rarely, urinary or intestinal obstruction results from large pelvic osteochondromas. Thoracic osteochondromas have been reported to lead to diaphragmatic rupture [Abdullah et al 2006], pneumothorax [Chawla et al 2013, Imai et al 2014, Dumazet et al 2018], hemothorax [Yoon et al 2015, Lin et al 2017], coronary artery compression [Rodrigues et al 2015], and severe chest pain [Kanthasamy et al 2020]. Osteochondromas have led to pseudoaneurysms [Oljaca et al 2019, Iqbal et al 2020] that can mimic sarcoma. Biopsy of a misdiagnosed pseudoaneurysm can have life-threatening consequences [Iqbal et al 2020].
Scoliosis. Some authors have documented scoliosis caused by or in the setting of osteochondromas with a prevalence of 72% [Matsumoto et al 2015, Veeravagu et al 2017, Gigi et al 2019]. The mean main curve was 15.3 degrees and minor curve 10.6 degrees. Despite the documented prevalence, a curve necessitating surgical intervention is uncommon.
Chondrosarcoma. The most serious sequela of HMO is malignant transformation of an osteochondroma. Axial sites such as the pelvis, scapula, ribs, and spine are more commonly the location of transformation of osteochondromas to chondrosarcoma [Porter et al 2004]. Rapid growth and increasing pain, especially in a physically mature person, are signs of sarcomatous transformation, a potentially life-threatening condition.
A bulky cartilage cap (best visualized with MRI or CT) thicker than 2-3 cm is highly suggestive of chondrosarcoma [Shah et al 2007, Bernard et al 2010]. After skeletal maturity, increased radionucleotide uptake on serial technetium bone scans may also be evidence of malignancy. High metabolic activity in the cartilage as evidenced by uptake of gadolinium on T2-weighted MRI may also be indicative of malignancy [De Beuckeleer et al 1996]. FDG-PET imaging may be useful in the workup for malignant transformation in HMO. A standardized uptake value maximum (SUVmax) of 2 has been reported as the cutoff above which chondrosarcomatous transformation of an osteochondroma has likely occurred, although lesions with an SUVmax as low as 1.3 have been found in grade I chondrosarcoma [Aoki et al 1999, Feldman et al 2005, Purandare et al 2019].
The incidence of malignant transformation to chondrosarcoma, or less commonly to other sarcomas, is estimated at 2%-10%. In a large cohort of 529 affected individuals, the rate of malignant transformation was calculated to be 5% [Pedrini et al 2011]. A survey of an international heterogeneous cohort of 757 individuals with HMO revealed 21 (2.7%) with malignant transformation, with pelvis and scapula the most common sites of malignant change from benign osteochondromas [Czajka & DiCaprio 2015]. However, rates as high as 10% have been reported in individuals with HMO [Tepelenis et al 2021].
Malignant transformation can occur during childhood or adolescence, but the risk increases with age [Schmale et al 2010]. Based on a study of HMO in Washington State, United States, it was estimated that HMO may increase the risk of developing a chondrosarcoma by a factor of 1,000-2,500 over the risk for individuals without HMO.
Other
Vascular abnormalities. In addition to vascular abnormalities from osteochondromas close to vascular structures, an association between osteochondromas and venous malformations has been reported [
Albokhari et al 2023].
Leukemia. There have been reports of a possible role of
EXT1 and
EXT2 in leukemogenesis through several mechanisms. To date, three individuals have been reported with HMO and leukemia [
Comisi et al 2025].
Bursal formation between osteochondromas and surrounding tissues occurs in at least 1.5% of individuals with HMO and can result in painful mass formation that tends to increase with size. This is most frequent in lesions of the scapula and shoulder joint [
Murphey et al 2000]. It is important to differentiate symptomatic bursal formation from malignant transformation when an individual presents with new worsening pain.
Prognosis. Males tend to be more severely affected than females [Pedrini et al 2011]. Although pain is commonly reported, most individuals with HMO lead active, healthy lives.