SHOX-Related Skeletal Features
- 45,X (and mosaic/structural variants) in females; many skeletal features, and the short stature, are attributable to SHOX haploinsufficiency - the gene whose mutations cause Leri-Weill dyschondrosteosis
- Classic signs: disproportionate short stature, cubitus valgus, genu valgum, short 4th/5th metacarpals (metacarpal sign), Madelung deformity, high-arched palate, short webbed neck, shield chest
- Spinal deformity in 12% (37 of 306), 73% of it mild; severe curves in about 2% of the cohort, and 4 patients came to fusion
- Growth hormone was not a risk factor for a curve, its severity or its progression (P greater than 0.05 for all) - do not withhold it on that basis
- Reduced bone mineral density, partly from the oestrogen deficiency of ovarian failure, with increased fracture risk
- Management is largely surveillance within a multidisciplinary team, treating significant deformity on its merits alongside cardiac, renal and endocrine care
- “No validated predictor of curve progression in Turner syndrome exists
- “A short fourth metacarpal supports the diagnosis but does not make it: also pseudohypoparathyroidism / Albright hereditary osteodystrophy and other SHOX disorders
- “Cardiac assessment before elective surgery: bicuspid aortic valve, coarctation, aortic root dilatation and dissection
Overview and the SHOX Mechanism
Turner syndrome is a chromosomal disorder of females caused by complete or partial absence of one X chromosome, classically 45,X, with mosaic and structural variants. It causes short stature, gonadal dysgenesis with ovarian failure, and characteristic somatic features.
The SHOX mechanism. Many of the skeletal features, and the short stature itself, are attributable to haploinsufficiency of the SHOX (short-stature homeobox) gene. It is the same gene whose mutations cause Leri-Weill dyschondrosteosis and contribute to idiopathic short stature.
A multisystem disorder. Turner syndrome involves the heart, including the aorta, the kidneys and the endocrine system. Any orthopaedic intervention is undertaken with awareness of these comorbidities and in coordination with the wider team.
The Skeletal Signature
Recognising the classic SHOX-related skeletal signature can even prompt the diagnosis:
- Disproportionate short stature
- Cubitus valgus, an increased carrying angle at the elbow
- Genu valgum
- Short fourth, and sometimes fifth, metacarpals, giving a positive metacarpal sign
- Madelung deformity of the wrist
- High-arched palate
- Short webbed neck
- Shield (broad) chest
Madelung deformity. The radiographs and photographs below come from a family with a pathogenic SHOX-locus rearrangement and illustrate the distal radial deformity shared with Turner syndrome. Significant deformity is treated on its merits (see Management).
The Metacarpal Sign (Short Fourth Metacarpal)
Brachymetacarpia. The fourth metacarpal, and sometimes the fifth, is abnormally short, a dysplastic feature of the SHOX skeletal phenotype. When the patient makes a fist the affected knuckle is depressed: a dimple replaces the normal knuckle prominence over the short metacarpal.
The Archibald sign. On a PA hand radiograph, draw a tangent touching the distal heads of the fourth and fifth metacarpals and extend it proximally. Normally the line passes just distal to the head of the third metacarpal and misses it. With a short fourth metacarpal the line intersects or passes through the third metacarpal head, a positive metacarpal (Archibald) sign.
Significance. The sign is a useful pointer to Turner syndrome but is not specific, so it supports the diagnosis rather than making it. The same short fourth metacarpal is seen in:
- Pseudohypoparathyroidism and pseudopseudohypoparathyroidism (Albright hereditary osteodystrophy)
- Other SHOX disorders
- An isolated familial or idiopathic trait
Spinal Deformity and Growth Hormone
Scoliosis and kyphosis are more common in Turner syndrome. "Increased risk" is right but unquantified, and the numbers change how you counsel a family. They come from the largest single-institution review, 306 patients followed from 2007 to 2021.
How common. 37 of 306 (12%) had a radiographically relevant spinal deformity. That is several times the roughly 2 to 3 per cent quoted for adolescent idiopathic scoliosis in the general population, so the increased risk is real, but it also means almost nine in ten girls with Turner syndrome do not develop a curve.
How severe. Most curves are mild. Of the 37, 27 (73%) were mild, 4 (11%) moderate and 6 (16%) severe, so severe deformity affected 6 of the whole cohort of 306, about 2%.
Surgery is uncommon. Of the 6 with severe curves, 4 underwent spinal fusion and 1 was braced. The sixth was braced before dying of a cardiovascular cause, a reminder in the middle of a spinal discussion that the cardiac disease is what kills in this syndrome.
Growth hormone is not the culprit. It is often taught that growth hormone, by accelerating growth velocity, unmasks or worsens a curve. In the same review 190 of 306 (62%) received GH, and GH therapy was not a risk factor for the diagnosis of a major spinal curve, for a more severe curve at diagnosis, or for curve progression (P greater than 0.05 for all). Among the 37 with a curve, 24 (65%) had received GH and 13 (35%) had not, proportions close to the cohort as a whole.
How far to trust it. This is the best available evidence rather than proof of safety. It is a retrospective, univariate chart review from a single tertiary paediatric centre with only 37 curve events, so a modest GH effect could have been missed: it argues against a large effect rather than proving none. Ascertainment depends on who was imaged, and referral bias may inflate the rate relative to an unselected Turner population.
What predicts progression. The authors call for further study, and no validated risk factor for curve progression in Turner syndrome has been identified.
What to tell the family. Screen the spine through growth, expect that a curve found will most likely be mild, and reserve alarm for the small minority.
Monitor the spine through growth because scoliosis occurs in this population, not because GH is presumed to drive it, and do not withhold or interrupt GH on that reasoning alone.
Bone Health and the Endocrine Interface
Reduced bone density. Reduced bone mineral density (osteoporosis) is common, related in part to the oestrogen deficiency of ovarian failure, and fracture risk is increased. Bone health and timely oestrogen replacement therefore matter.
Growth hormone and oestrogen. Growth-hormone therapy is used to improve final adult height. Oestrogen replacement for the gonadal failure is important for pubertal development and for bone density, and the orthopaedic surgeon manages the skeletal features in this context.
Management
Management is largely surveillance and supportive care within a multidisciplinary team, alongside the systemic cardiac, renal and endocrine care of the syndrome:
- Monitor stature and growth, the spine and limb alignment
- Optimise bone health with oestrogen, vitamin D and calcium, and weight-bearing activity
- Treat significant deformities on their merits, for example Madelung deformity or symptomatic genu valgum
Turner syndrome carries a high rate of bicuspid aortic valve and coarctation and a predisposition to aortic root dilatation and dissection. Before any elective orthopaedic surgery and general anaesthetic, cardiac assessment (echocardiography, cardiology input) is essential, and blood-pressure and aortic surveillance continue perioperatively.
Renal anomalies. A horseshoe kidney or collecting-system duplication, for example, affects fluid and drug handling and positioning.
Fixation in low-density bone. The reduced bone mineral density means poorer implant purchase and a higher fracture risk, so fixation is planned accordingly and oestrogen, vitamin D and calcium are optimised. There is no evidence base for implant selection or fixation technique in this bone.
Mnemonics & Memory Aids
TURNER
Hook:TURNER: Turner (SHOX), Ulnar/elbow cubitus valgus, Reduced bone density, Neck short/webbed, Extremity (short metacarpals/Madelung), Rule the spine in.
Clinical Decision Scenarios
Practise clinical reasoning and management decisions out loud
“A short-statured girl has cubitus valgus, short fourth metacarpals and a Madelung deformity. What syndrome and mechanism do you consider, and what are the orthopaedic priorities?”
What it is
- 45,X (and mosaic/structural variants) in females
- Short stature, ovarian failure, characteristic somatic features
- Many skeletal features from SHOX haploinsufficiency
Classic skeletal signs
- Cubitus valgus, genu valgum
- Short fourth/fifth metacarpals (metacarpal sign); Madelung deformity
- High-arched palate, short webbed neck, shield chest
Spine & bone
- Spinal deformity in 12% (37 of 306): 73% mild, 11% moderate, 16% severe; 4 came to fusion
- Growth hormone NOT a risk factor for curve presence, severity or progression (P greater than 0.05 for all)
- No validated predictor of curve progression exists - screen through growth regardless
- Reduced bone mineral density (oestrogen-related fracture risk)
- Optimise oestrogen, vitamin D/calcium, weight-bearing activity
Management
- Growth-hormone therapy (stature); oestrogen replacement (puberty/bone)
- Surveillance of stature/spine/alignment; treat significant deformity (Madelung, genu valgum)
- Multidisciplinary care (cardiac/aortic, renal, endocrine comorbidities)
Evidence & Key Studies
SHOX gene deletions and the skeletal features shared with Turner syndrome
- SCOPE - READ THIS FIRST: the 900 children studied all had a NORMAL KARYOTYPE and so did NOT have Turner syndrome. The association between SHOX and the Turner phenotype is the paper's premise, not its finding. Its own result concerns SHOX mutations in idiopathic short stature.
- 900 patients with height below the third percentile (or minus 2 SD) and no obvious Leri-Weill skeletal features were analysed - 750 for intragenic mutations by SSCP and sequencing, 150 for whole-gene deletions by FISH. This was the largest such series at the time.
- SHOX mutations were found in 2.4% of children with short stature: 9 of 750 intragenic mutations (of which at least 3 were judged functional) and 3 of 150 complete gene deletions. The mutation spectrum is biased, the vast majority being COMPLETE GENE DELETIONS. The authors note this prevalence is similar to that of growth-hormone deficiency or of Turner syndrome itself.
- The link to the Turner skeletal signature comes from FAMILY STUDIES: older relatives carrying the same mutation showed mild Turner-like features - high-arched palate, short neck, ABNORMAL AURICULAR DEVELOPMENT, cubitus valgus, genu valgum, short fourth metacarpals and Madelung deformity. This is the observation that supports attributing those features to SHOX haploinsufficiency.
- Limitations: a genetic prevalence study, not a skeletal one - no radiographic scoring, no measure of how often each feature occurs, and the family-member phenotypes are descriptive.
Scoliosis and kyphosis prevalence in Turner syndrome: a retrospective review at a paediatric tertiary care medical centre
- The largest single-institution review of spinal deformity in Turner syndrome known to its authors: 306 patients, retrospective chart review 2007-2021.
- 37 of 306 (12%) had a radiographically relevant spinal deformity - 27 of 37 (73%) mild, 4 (11%) moderate and 6 (16%) severe. Of the 6 severe curves, 4 had spinal fusion and 1 was braced; the sixth was braced before a cardiovascular-related death.
- KEY NEGATIVE FINDING: growth hormone was used by 190 of 306 (62%). On univariate analysis GH therapy was NOT a risk factor for the diagnosis of a major spinal curve, for a more severe curve at diagnosis, or for curve progression (P greater than 0.05 for all) - contradicting the common teaching that GH accelerates curves in Turner syndrome.
- Among those with a curve, 24 of 37 (65%) had received GH versus 13 (35%) who had not, closely mirroring the 62%/38% split of the whole cohort.
- Limitations: retrospective, single tertiary paediatric centre, univariate analysis only, and just 37 curve events - so this is evidence against a large GH effect rather than proof of none, and referral bias may inflate the 12% relative to an unselected Turner population. The authors state that further study is needed to determine risk factors for curve progression, so none is currently validated.
The spinal figures - 37 of 306 (12%) with a curve, the 73/11/16 per cent mild-moderate-severe split, the 4 fusions, and the finding that growth hormone was not a risk factor for curve presence, severity or progression - come from Marx (DOI), a single-centre retrospective review. The Turner-like skeletal features in SHOX-mutation carriers (high-arched palate, short neck, abnormal auricular development, cubitus valgus, genu valgum, short fourth metacarpals, Madelung deformity) and the 2.4% SHOX-mutation rate in short stature come from Rappold (DOI) - noting that its 900 patients had NORMAL karyotypes and did not have Turner syndrome, so it supports the SHOX mechanism rather than measuring anything in Turner patients. The 45,X karyotype and ovarian failure, the reduced bone mineral density and oestrogen-related fracture risk, the metacarpal (Archibald) sign, and the cardiac, aortic and renal comorbidities are standard, well-established teaching.
What is not established: no validated risk factor for spinal-curve progression in Turner syndrome has been identified - the authors of the only sizeable series call for further study. There is no Turner-specific fracture-rate figure quoted here, no threshold bone-density value that triggers intervention, no evidence base for implant selection or fixation technique in the reduced bone density, and no trial of any orthopaedic intervention in this population. The 12% curve prevalence comes from one tertiary centre and may not transfer to an unselected Turner population.