Short digits…What's up?



Short digits, metatarsal bones, and metacarpal bones result from a premature closure of the physeal plate and can be a manifestation of several genetic and acquired disorders. We present a case of hereditary multiple exostosis (HME) with coexistent juvenile idiopathic arthritis (JIA) and short metacarpal and metatarsal bones. This case raised the question of whether the skeletal abnormalities were genetic or acquired and led us to review the mechanisms that regulate physeal closure.

Case Report

A 15-year-old girl known to have HME since age 9 was referred to our department with a 9-month history of increasing discomfort and swelling of her left elbow, knee joint, and Achilles tendon. Her family history revealed that her mother and 2 older brothers were also known to have HME, due to a defined mutation in the exostosin 2 (EXT2) gene, exon 7 (IVS6-2 A>G). There was no family history of disorders related to HLA–B27. Physical examination disclosed a monarthritis of the left elbow, a tendinitis of the left Achilles tendon, and no signs of psoriasis or uveitis. There were small, palpable exostoses at the basal phalanx of the third and fourth fingers of the right hand and large, palpable exostoses at both knee joints. The third metatarsal of her right foot and the fourth metacarpal of her left hand were short (Figure 1A) and there were no other signs of skeletal dysplasia. Laboratory examination showed no acute-phase reaction, rheumatoid factor and antinuclear antibodies were negative, and the patient was HLA–B27 positive. Radiologic examination showed erosive changes at the left ankle and no abnormalities at the left elbow and sacroiliac joints, and confirmed the existence of multiple exostoses at both knees (Figure 1B). The patient was diagnosed with enthesis-related JIA (1), coexisting with her HME. Alternative diagnoses such as pseudohypoparathyroidism and pseudopseudohypoparathyroidism were rejected as additional laboratory evaluation revealed normal levels of calcium, phosphate, alkaline phosphatase, and parathyroid hormone (PTH), and neither the patient nor her family had other symptoms of Albright's hereditary osteodystrophy.

Figure 1.

A, Shortening of fourth metacarpal of the left hand. B, Radiologic examination showing multiple exostosis at the distal end of the femur and proximal end of the tibia on both knee joints.

Treatment with diclofenac and sulfasalazine was initiated with satisfactory clinical results within months. By consulting the literature and the department of human genetics at our center, we learned that short metacarpal and metatarsal bones can be a rare manifestation of HME (2).


HME is an autosomal dominant disorder of the endochondral bone formation characterized by the development of multiple osteochondromas around the epiphysis of long bones. The disease has an estimated occurrence of 1 in 50,000–100,000 in white persons and is often diagnosed during the first decade of life when patients present with pain, cosmetic concerns, and short stature. The size and number of lesions varies widely among affected individuals and lesions may become symptomatic as they grow proportionately to the overall growth, sometimes resulting in compression of surrounding nerves, vessels, and nails (2).

There are 3 known inactivating mutations in tumor suppressor genes involved in the pathogenesis of HME: EXT1, EXT2, and rarely EXT3 located on chromosomes 8q24, 11p11–p12, and 19p, respectively. Genotype-phenotype analysis indicates that mutations in the EXT1 gene are associated with more severe phenotypes (e.g., large numbers of exostosis, vertebral location, short stature, and increased risk of malignant transformation of the osteochondromas) whereas EXT2 mutations are more frequently associated with moderate phenotypes, as was the case in our patient (2, 3). HME can vary widely in presentation, due to variable levels of gene expression, but most commonly the distal end of the femur and proximal ends of the tibia, fibula, femur, and humerus are affected. Metatarsal and metacarpal bones are involved only in 9% and 7% of the patients, respectively.

Brachydactyly, short metacarpals, and several other skeletal deformities can be inherited or acquired. Interestingly, the pathogenesis of brachydactyly, observed in several human skeletal dysplasias, shares a similar final common pathway that involves the Indian hedgehog(IHH)/PTH/PTH-related peptide (PTHrP) (3). This is a tightly controlled feedback loop within the growth plate that can be disrupted at several levels resulting in accelerated progression of the chondrocytes from proliferation to early apoptosis (Figure 2). HME and other inherited skeletal dysplasias that result from alterations in the IHH/PTH/PTHrP pathway are listed in Table 1 and are not discussed in detail in this article.

Figure 2.

During endochondral bone formation, prehypertrophic chondrocytes secrete the Indian hedgehog (IHH) protein, which diffuses to its receptor in cells of the perichondral region via heparan sulfate proteoglycans (HSPGs) that are glycosylated by exostosin 1 (EXT1) and EXT2. Binding of IHH to its receptors, patched (Ptc), smoothened (Smo), and hedgehog-interacting-protein (Hip), enhances the secretion of parathyroid hormone (PTH)–related peptide receptor (PTHrP). The latter diffuses and binds to the PTH/PTHrP receptor on a subpopulation of proliferating (pro) and prehypertrophic (pre) chondrocytes. Terminal differentiation and apoptosis of these cells is postponed by direct or indirect up-regulation of Bcl-2, an antiapoptotic protein, via Gs-α-1 protein of the adenylyl cyclase complex (GNAS1). Several fibroblast growth factors (FGF) only bind to their receptors in the presence of HSPGs. Proinflammatory mediators such as tumor necrosis factor α (TNFα), interleukin-1 (IL-1), oncostatin M (OSM), and macrophage migration inhibitory factor (MIF) also inhibit proliferation and promote chondrocyte apoptosis, presumably by inducing matrix metalloproteinases (MMPs). Insulin-like growth factor (IGF) and IL-4 partially reverse the effects of TNF and IL-1 on chondrocytes of the physeal plate. OPG = osteoprotegerin; hyp = hypertrophic chondrocytes; apop = apoptotic chondrocytes. Adapted, with permission, from ref. 3.

Table 1. Hereditary disorders that influence the IHH/PTH/PTHrP pathway*
MutationOMIM entryGene map locusDiseaseOMIM entryInheritance
  • *

    Genetic skeletal dysplasias, which can affect the growth plate but do not interfere with the IHH/PTH/PTHrP pathway, such as those due to mutations in the genes encoding for collagen, arylsulfatase, and lysosomal enzymes. IHH = Indian hedgehog; PTH = parathyroid hormone; PTHrP = PTH-related peptide receptor; HSPGs = heparan sulfate proteoglycans; EXT = exostosin; HME = hereditary multiple exostosis; GNAS = Gs-α-1 protein of the adenylyl cyclase complex; AHO = Albright's hereditary osteodystrophy; FGFR = fibroblast growth factor receptor; DTDST = diastrophic dysplasia sulfate transporter.

  • See text.

  • Many dominant and some recessive cases.

 *6007262q33–q35Brachydactyly type A1#112500Autosomal dominant
   Acrocapitofemoral dysplasia#607778Autosomal recessive
 EXT1*6081778q24.11–q24.13HME type I#133700Autosomal dominant
 EXT2*60821011p11–p12HME type II#133701Autosomal dominant
 EXT3%60020919pHME type III%600209Autosomal dominant
GNAS complex     
 +13932020q13.2AHO in pseudohypoparathyroidism type Ia; pseudopseudohypoparathyroidism#103580Autosomal dominant
   Prolonged bleeding time, brachydactyly, and mental retardation+139320Unknown
 PTHR1*16846813p22–p21.1Metaphyseal chondrodysplasia, Murk Jansen type#156400Autosomal dominant
   Blomstrand chondroplasia (lethal)#215045Autosomal recessive
 FGFR1*1363508p11.2–p11.1Pfeiffer's syndrome#101600Genetic heterogeneity
   Osteoglophonic dysplasia#166250 
   Kallmann syndrome#147950 
 FGFR2*17694310q26Apert's syndrome#101200Mostly sporadic
   Crouzon's syndrome#123500Autosomal dominant
   Jackson-Weiss syndrome#123150Autosomal dominant
   Beare-Stevenson cutis gyrata#123790 
 FGFR3*1349344p16.3Achondroplasia#100800Autosomal dominant
   Hypochondroplasia#146000Autosomal dominant
   Thanatophoric dysplasia (lethal)#187600Genetic heterogeneity
   Muenke syndrome#602849 
 *6067185q32–q33.1Diastrophic dysplasia#222600Autosomal recessive
   Achondrogenesis type Ib (lethal)#600972Autosomal recessive
  12q13.11–q13.2Atelosteogenesis type 2 (lethal)#200610Autosomal recessive
   Multiple epiphyseal dysplasia#226900Autosomal recessive

The growth plate development and the endochondral ossification are not only controlled by hormonal and nutritional factors but are also influenced by acquired inflammatory conditions, e.g., cytokines, growth factors, and metalloproteinases.

In the case of JIA, early physeal closure is often due to inflammation in the adjacent joints (which our patient had never presented), circulating proinflammatory factors, or steroid therapy (which our patient did not receive previously). Clinical data on patients with JIA indicate that interleukin-1β (IL-1β) and tumor necrosis factor α (TNFα) are the main mediators of the chronic inflammatory process (4, 5). Together with oncostatin M (OSM), a cytokine of the IL-6 family, IL-1β and TNFα are present in the synovial fluid of patients with JIA (6, 7). Experimental models have shown that TNFα and IL-1β act in synergy to inhibit longitudinal growth in fetal rat metatarsal bones, decrease chondrocyte proliferation, and induce chondrocyte apoptosis and metalloproteinase expression in the joint (7). These effects are partially reversible by insulin-like growth factor 1 or IL-4 and are not mediated by IL-6 (8, 9). Furthermore, the expression of OSM and macrophage migration inhibitory factor (MIF) in the joint enhances the production of IL-1 and TNFα and induces proteoglycan depletion and disorganization of the growth plate, which is fully prevented in the absence of IL-1 (7, 10).

Chondrocytes also have receptors for TRAIL, which, upon binding to their ligand, induce chondrocyte apoptosis (11). The binding of TRAIL to its endogenous antagonist osteoprotegerin (OPG) prevents chondrocyte apoptosis. This mechanism may also be involved in the early physeal closure in JIA, as indicated by the compensatory high levels of circulating OPG found in these patients (12).

The question arises whether these inflammation-induced signals also interact with the IHH/PTH/PTHrP pathway that regulates physeal closure. We postulate that this is indeed the case because several matrix metalloproteinases (MMPs), including MMP-13, MMP-14, and MMP-9 as most abundant, are expressed in the growth plate, where they coordinate the chondrocyte proliferation, migration, and apoptosis (13). The expression of MMP-13 and other MMPs is induced by TNFα and other proinflammatory cytokines including MIF (14), and interestingly, MMP-13 is also a downstream target of the IHH/PTH/PTHrP pathway (15). MMP-13 knockout mice and other mutation or knockout models that target MMPs show profound defects in the growth plate and endochondral ossification that mirror defects observed in human hereditary chondrodysplasias (13). In patients with JIA, several MMPs, including MMP-13, are overexpressed in synovial tissue and the levels of circulating MMPs are elevated, correlating with local and systemic inflammation (16).

We therefore propose that MMPs are the common pathway to genetic and acquired conditions leading to premature physeal closure. Proinflammatory cytokines such as IL-1, TNFα, OSM, and MIF are responsible for the early physeal closure in patients with JIA and other inflammatory conditions either directly or by inducing MMP expression at the growth plate.


Dr. Barrera had full access to all of the data in the study and takes responsibility for the integrity of the data.

Study design. Aarntzen, Barrera.

Acquisition of data. Aarntzen, Barrera.

Analysis and interpretation of data. Aarntzen, Barrera.

Manuscript preparation. Aarntzen, Barrera.