WhatsAppinfo@ri-hub.it
Ri-Hub Junior

Child with One Leg Shorter: When It Really Matters

Team Ri-Hub15 min lettura

Revisione clinica: Daniele Carretto, Fisioterapista (Albo 2019) · Pubblicato il 28 luglio 2026

Due bambini che si godono una giornata di sole all'aperto con un'auto giocattolo, mostrando innocenza giocosa.
Condividi

You watch him from behind as he runs around the garden and something does not look right: one hip seems higher, his trousers fall differently on the two legs. Or someone else has told you, “his pelvis is crooked”, “he has one leg shorter”, and since that day you cannot watch him walk without looking for the difference.

Here is the answer straight away, because it turns the starting point on its head: almost everyone has one leg shorter than the other. In the review that gathered data obtained with reliable radiographic methods, the prevalence of an anatomical difference was 90%, with an average of 5.2 mm (standard deviation 4.1). Five millimetres, not five centimetres. The useful question is not “does he have one?” but how much, what it depends on, and whether it is changing.

The number that changes the conversation: 90%

That review, from 2005, gathered data published from 1970 onwards on the difference in length between the lower limbs, keeping only data obtained with accurate radiographic methods. Its conclusion, in its own words: an anatomical difference is almost universal, but the average size is small and it is unlikely to be clinically significant.

It should be read with its labels: it is a narrative review, not a meta-analysis, published in a chiropractic journal and built largely on adult case series. It is not the kind of evidence on which a guideline is written. It is, however, the figure that explains why so many children “with one leg shorter” are within normal limits.

A 2021 review in Clinics in Orthopedic Surgery reaches us from another direction: differences are typically under 10 mm and asymptomatic, while it is above 20 mm that they can change biomechanics and the distribution of loads.

And teenagers? An Italian cross-sectional study measured 44 adolescents aged 10 to 18 followed for idiopathic scoliosis or postural disorders: a difference in 79.5%, average 0.7 ± 0.6 cm, and all of them mild. The authors' comment is the part you need: those values reflect the norms of the general population. Even in a group selected because they have “something wrong with the back”, half a centimetre is not a finding: it is the background.

The two thresholds: below what nothing is done

Size of the difference What the sources say
Up to about 5 mm The average in the general population
Under 10 mm Typically asymptomatic, compensated spontaneously
Over 10 mm Measurable changes in gait may appear
Over 20 mm Threshold above which clinical relevance starts to be discussed
Over 50 mm predicted at the end of growth Surgical lengthening during growth can be considered

The 10 mm line needs a caveat, because it is the least solid. It comes from a 2017 systematic review of 12 studies: only 4 had measured gait in people with a real anatomical difference; the other 8 had simulated it with lifts of 1 to 5 cm under the shoe of healthy subjects. But a lift put on this morning is not a body that has grown that way: the simulation removes precisely the variable that weighs most in a child, adaptation.

And here comes the most useful and least discussed point on the subject. The 2020 review in the Deutsches Ärzteblatt International opens by declaring that there are no evidence-based studies on the indication for treatment, and later repeats that the evidence base is poor and that what we have are informal consensus recommendations. It adds that the link with back pain is debatable, while a modest increase in knee osteoarthritis appears likely, and it closes with a sentence that works as a rule: the decision to treat is always elective.

Those lines in the table are not laws of nature: they are lines drawn on a continuum by competent people who still had to decide where to put them. A 2002 review in Gait & Posture already summed up the state of the art this way: the authors agree neither on how much a difference actually causes problems, nor on what size is needed to produce them.

Silhouette di un padre e di un bambino sulla riva, circondati da uccelli.

Why the eye is deceived (and so is the tape measure)

There is a distinction that, once understood, spares you half the anxiety: a leg can be genuinely shorter, or can just look it.

An apparent difference comes from alignment, not from the bone: a rotated or tilted pelvis, a knee held bent, a foot that gives way more on one side and lowers that half of the body, a stiffness that pulls one hip up. Things that shorten a leg in the eyes of the observer, without any bone being shorter.

The second part of Knutson's review addresses exactly this: the asymmetry found unloaded, lying down, is a different phenomenon from an anatomical difference and may depend on hypertonia of the muscles above the pelvis. Hence an order of priorities: the functional asymmetry should be resolved before treating any anatomical difference. The author himself warns that more data are needed on the reliability of that measurement.

The back enters the picture too, though not in the direction you might expect. Functional scoliosis is by definition an apparent curve on a structurally normal spine: it is not progressive and regresses completely or partly when the length difference is corrected. In the study of the 44 adolescents, in postural lumbar curves the shorter limb was always on the convexity side; in true idiopathic scoliosis, no consistent association. Half a centimetre of leg does not explain a scoliosis: the topic is covered in scoliosis in adolescents: exercises and brace.

The same goes for the foot: a foot that looks flatter than the other is among the possible sources of an apparent difference, which according to the 2021 review can arise at any point between the iliac crest and the foot, and the topic is covered in flat feet in children: when treatment is really needed.

How it is really measured

Here the sources agree, which is rare.

A review in Clinical Orthopaedics and Related Research analysed 42 articles: tape measures and standing lifts are useful screening tools, but not as accurate as imaging. Among radiological tests, the authors describe the full-length standing radiograph as the most complete technique, and recommend a CT scanogram especially when the knee is held bent.

The systematic review in PLoS One from 2021, covering 37 studies on clinical methods and 15 on imaging, reaches the same conclusion: among clinical tests the block test (calibrated lifts under the short foot until the pelvis is level) is the most useful, and the full-length standing radiograph is the most valid reference. With a limitation that says a lot: many of the studies on clinical methods lacked a reference standard.

A study of 71 subjects compared three clinical methods with radiographic telemetry: the Weber-Barstow manoeuvre reached a kappa of 0.52 (moderate agreement), the pelvic gauge was not found to be valid, and the block method gave values in line with the radiograph (0.48 cm against 0.51 cm).

What you can do at home: no numbers. What you see when you look at him standing is a hypothesis, not a measurement, and this includes the photo taken from behind. It is more useful to note since when you have seen it, whether it is stable or increasing, whether there is pain, whether he limps, and whether there have been fractures or hip problems.

The real causes to rule out

The cause matters more than the millimetre, because it decides the trajectory. The 1982 review of 803 patients showed it: not all differences grow at a steady rate, and the five patterns described over time are distributed differently depending on the starting condition and on when it appeared.

These are the sources a doctor looks for, with what we know about them:

  • Thigh fractures. After a femur fracture the broken bone can grow more: it is a problem well enough known to have generated a practical question, namely whether it is worth leaving the fragments overlapping to compensate for it. In a series of 75 children with a closed femoral shaft fracture, measured twice on radiographs over four years, the answer was no: shortening the fragments showed no clear advantage in preventing overgrowth, and in older children it risked leaving a persistent shortening.
  • Treated congenital clubfoot. In a retrospective series of 278 children with idiopathic clubfoot treated with the Ponseti method, a leg-length difference documented on radiograph was present in 14.4%, average 1.1 ± 0.8 cm; much more frequent in unilateral forms (27% against 4.5% of bilateral ones) and in relapsed cases (21.5% against 6.2%). The pathway is covered in congenital clubfoot.
  • Persistent knee arthritis. An inflammation lasting months can make the affected leg grow more. In a retrospective comparison between two centres of 30 children with pauciarticular juvenile arthritis (now called oligoarticular juvenile idiopathic arthritis) diagnosed before the age of 7, those who received corticosteroid injections within two months of diagnosis, repeated when the synovitis returned, showed a significantly smaller difference (p = 0.005) and fewer prescriptions for lifts (p = 0.002). The lesson is not “have the injections”: it is that a knee swollen for weeks with no trauma is not a minor detail.
  • The hip. Among acquired causes the 2021 review lists the necrosis that follows hip dysplasia and Perthes' disease: here the entry signal is not length, it is limping or pain. Everything on this is in the child who limps and, for the first months, in hip dysplasia in newborns.
  • A whole side of the body larger. This is the item that needs separate handling, and it is in the section below.

If your question is about the shape of the legs rather than their length, bowed legs or knees that touch, that is a different chapter with its own sequence by age: bowed legs and knocking knees.

Red flags: when not to wait

Today, the same day, even in A&E: a fever together with refusal to put the foot down, or a joint that is hot, swollen and painful.

This week, without waiting for the next health check:

  • night pain that wakes him, or pain at rest that does not go away within a few days;
  • a limp lasting more than a few days, or getting worse instead of better;
  • a difference that has appeared or increased within a few weeks or months: the speed of change is more informative than the millimetres;
  • a knee or ankle swollen for weeks with no trauma at all;
  • loss of motor skills that your child already had, or clear weakness on one side.

And one sign that deserves a line of its own. If it is not just the leg that is longer, but a whole side of the body is larger (face, trunk, limb), the right place is not a physiotherapist's clinic: it is a paediatric genetic assessment. In the multicentre prospective study of 168 children with isolated hemihyperplasia, 9 children developed 10 tumours, an overall incidence of 5.9%, all of embryonal type (Wilms tumour, hepatoblastoma and others), and the authors conclude that those data justify a surveillance protocol. Along the same lines is the 2018 international consensus document on Beckwith-Wiedemann syndrome, which includes lateralised overgrowth among the signs of the condition and provides for cancer surveillance calibrated to the molecular subtype. This suspicion is not postponed until the next check.

And shoe lifts? What the evidence really says

Very little, and it is worth knowing before buying one.

The systematic review from 2018 in Archives of Physical Medicine and Rehabilitation looked for all studies on lifts for leg-length difference: it found 10, of which only one randomised trial, with methodological quality very low or poor. In the non-randomised studies, 88% ± 3% of the 349 participants treated reported partial or complete pain relief, a number that sounds excellent until you notice that a comparison group is almost always missing. In the one randomised trial, all 22 in the treated arm had relief, with an average reduction of 27 ± 9 mm on a 150 mm scale. And of the 9 guidelines examined, only 2 recommended lifts, by consensus. The authors close by calling for research into which threshold is worth correcting.

Two caveats: that review concerns adults aged 18 and over with musculoskeletal pain, not children who are well; and none of those studies shows that a lift given to an asymptomatic child with a 6 mm difference prevents anything.

On "preventing", it is worth being precise, because it is the argument that sells the most. The cohort study of 3,026 people aged 50 to 79, already with knee osteoarthritis or at high risk, measured the difference with a full-length radiograph: those with 1 cm or more showed radiographic osteoarthritis in the shorter leg in 53% of cases against 36% (odds ratio 1.9, 95% CI: 1.5-2.4). This is an association observed in middle-aged adults and older: it does not say that the difference caused the osteoarthritis, nor that a lift fitted at eight years old changes anything.

The practical sense: in a child, below the thresholds of clinical relevance and without symptoms, the lift is not a treatment, it is a response to adult anxiety. If instead the difference is significant or growing, the game is played elsewhere.

How it is decided, when the difference is real

In congenital forms the short limb grows in proportion to the long one: this is the basis of the multiplier method described in 2000, which estimates the final difference even from a single measurement. The 2020 review does, however, temper confidence in these predictions: predictive algorithms come with an approximation of about 2 cm. And if the predicted difference exceeds 5 cm, lengthening can already be considered during the years of growth.

For a parent the consequence is reassuring: today almost nothing has to be decided. It is the trajectory that dictates the choice, not this morning's snapshot.

What a physiotherapist can do (and what not)

Let us say it plainly: no exercise lengthens a bone, and no manipulation “levels the legs”. Anyone who promises you that is selling you something.

What we can do is smaller and more useful than it sounds: distinguish an apparent difference from a real one, which is the step where people most often go wrong. Look at how he stands, how he loads, how he walks, whether a hip or knee has lost range of movement, whether there is pain and where. And tell you which of three boxes you are in: "this is normal", "re-measure in a few months", or "needs a paediatric orthopaedic surgeon".

With Ri-Hub Junior you start with a free twenty-minute introductory consultation by video call with a physiotherapist on the Italian professional register who works with children and teenagers: you show us your child standing and walking, we go through the history together (fractures, infections, how he walked at two) and we tell you whether it makes sense to observe or to have him seen. For children up to 10 years old, sessions last 50 minutes and cost 59,90 €; from 11 years up you move to the standard session at 29,90 €.

Two limits, stated before rather than after. If one of the signs in the list above appears, the right place is the doctor, straight away. And if behind it there is a complex neurological or neuromuscular condition, those pathways require in-person assessment: we do not follow them online.

In three lines

A few millimetres of difference between the two legs is normal: prevalence around 90%, average 5.2 mm. Below a centimetre there is almost never anything to do, and much of what you see by eye is apparent (pelvis, foot, stiffness) rather than anatomical. The trajectory matters more than the number: if the difference appears quickly, hurts, makes him limp or affects a whole side of the body, you go to a doctor.

If the concern is about the way he walks more than the length of his legs, start from child who walks on tiptoe; if the topic is evening pain in the legs, it is growing pains.

Sources

Last source check: September 2026.

Domande frequenti

How big does the difference between the two legs have to be to matter?

Much bigger than you might think. In the review that gathered data obtained with reliable radiographic methods, the prevalence of an anatomical difference between the two legs was 90% and the average difference was 5.2 mm (standard deviation 4.1): a few millimetres is the norm, not the exception. The same author places the threshold above which the difference starts to be clinically significant at around 20 mm, and the narrative review published in Clinics in Orthopedic Surgery in 2021 says the same thing from another angle: differences are typically under 10 mm, asymptomatic or compensated without effort, while it is above 20 mm that they can change biomechanics and the distribution of loads. An honest caveat: these are lines drawn on a continuum, and the German review from 2020 in the Deutsches Ärzteblatt International states plainly that there are no evidence-based studies on the indication for treatment, only informal consensus recommendations.

My son has half a centimetre of difference: does he need a shoe lift?

Most probably not, and the reason is that there is nothing to correct. Half a centimetre falls within normal variation, and the evidence on shoe lifts is much weaker than the pages selling them suggest. The systematic review published in Archives of Physical Medicine and Rehabilitation in 2018 found ten studies, of which only one was randomised, with methodological quality judged very low or poor; in the non-randomised studies, 88% of the 349 participants treated reported partial or complete pain relief, but without an adequate comparison group, and only 2 of the 9 guidelines examined recommended lifts, on the basis of consensus. Above all, that review concerns adults over 18 with musculoskeletal pain, not children who are well. The authors themselves close by calling for research into which threshold is worth correcting, because today we do not know it.

How do I measure it at home?

You cannot, and that is not your problem: it is a problem of method. The systematic review in PLoS One from 2021, which assessed 37 studies on clinical methods and 15 on imaging methods, concludes that among clinical tests the block test (progressive lifts under the foot while standing) is the most useful, while the true reference remains the full-length standing radiograph of the lower limbs. The review in Clinical Orthopaedics and Related Research, which analysed 42 articles, is equally explicit: tape measures and lifts are screening tools, not as accurate as imaging. And a study of 71 subjects compared three clinical methods with radiographic telemetry: the Weber-Barstow manoeuvre reached a kappa of 0.52, the pelvic gauge was not found to be valid, while the block method gave values in line with the radiograph (0.48 cm against 0.51 cm). Translated: what you see when you look at your child standing is a hypothesis, not a measurement.

If he has one leg shorter, will he develop scoliosis?

The link is much weaker and works in both directions, more than it is usually portrayed. An Italian cross-sectional study of 44 adolescents aged 10 to 18 with idiopathic scoliosis or postural disorders found a leg-length difference in 79.5% of cases, with an average of 0.7 ± 0.6 cm and all differences classed as mild: the authors comment that these values simply reflect the norms of the general population. In postural lumbar curves the shorter limb was consistently on the same side as the convexity, while in true idiopathic scoliosis no consistent association emerged between leg-length difference and the characteristics of the curve. This is consistent with what the reviews say: functional scoliosis is an apparent curve on a structurally normal spine, which regresses completely or partly when the length difference is corrected. The study is small and the difference was measured with a tape measure, with ultrasound only when needed: take it for what it is.

The physiotherapist said his pelvis is rotated: is that the same as a shorter leg?

No, and it is the most important distinction in the whole subject. One thing is an anatomical difference in bone length; another is an alignment asymmetry that makes one leg look shorter: a rotated or tilted pelvis, a knee held slightly bent, a foot that collapses more on one side, muscle stiffness pulling one side up. The second part of Knutson's review addresses exactly this: the length asymmetry found when lying down, unloaded, is a different phenomenon from an anatomical difference and may depend on hypertonia of the muscles above the pelvis; the two can interact when standing but not when lying down, and the operational conclusion is that a functional asymmetry should be resolved before treating any anatomical difference. The author himself notes that more data on the reliability of the measurement are needed, so this is not dogma: it is an order of priorities.

Will the difference increase as he grows?

It depends on the cause, which is why the cause matters more than today's millimetres. The review of data from 803 patients published in the Journal of Bone and Joint Surgery in 1982 showed that not all differences grow at a steady rate: five patterns over time have been described, including increasing, increasing then slowing, increasing then plateauing, and two more complex ones, and the pattern depends on the condition that caused the difference and on when it appeared. In congenital forms the short limb grows in proportion to the other, which is the principle behind the multiplier method described in 2000, which allows the final difference to be estimated even from a single measurement. The 2020 review quantifies how reliable these estimates are: predictive algorithms reach an approximation of around 2 cm. This is why the check is repeated over time rather than deciding everything today.

When should I really worry and not wait?

Straight away, the same day, if there is a fever together with refusal to put the foot down, or a hot, swollen joint. This week, without waiting for the next health check, if there is night pain that wakes him or pain at rest that does not go away within a few days, if he has been limping for more than a few days, or if the difference has appeared or grown within a few weeks. And there is one sign that deserves a separate line: if it is not just the leg that is longer, but a whole side of the body is larger (face, trunk, limb), what is needed is a genetic assessment, not a physiotherapist. In the multicentre prospective study of 168 children with isolated hemihyperplasia, 9 children developed 10 tumours, an overall incidence of 5.9%, all of embryonal type: Wilms tumour, hepatoblastoma and others. This is exactly why surveillance protocols exist, and why this suspicion is not postponed.

Dal singolo esercizio al percorso

Questi esercizi sono generici. Il tuo no.

Quelli che hai appena letto valgono per tutti, e per questo non valgono per nessuno in particolare. Un fisioterapista iscritto all'Albo guarda il tuo caso nel primo colloquio — gratuito, venti minuti — e da lì costruisce il programma su di te: da casa quando basta, in uno studio vicino a te quando servono le mani. Il dolore che torna non toglie solo il movimento: toglie il sonno, la pazienza, la voglia di uscire di casa. Rimetterti in movimento è il nostro mestiere.

Trova il tuo fisioterapista →

Hai solo una domanda? Scrivila su WhatsApp →

Fino a 10 anni 50 minuti a 59,90€ · dagli 11 ai 18 sedute da 29,90€ · nessun abbonamento

Ri-Hub ti mette in contatto con un fisioterapista online: primo colloquio gratuito in videochiamata con un professionista iscritto all'Albo, che ti segue per tutto il percorso. E se al tuo caso servono le mani, è chi ti segue a dirtelo e a mandarti da uno dei nostri, vicino a te. Se preferisci partire dalla tua zona, guarda il fisioterapista vicino a te, città per città. Se ti stai chiedendo quanto costa una seduta di fisioterapia, abbiamo confrontato studio privato, domicilio, ticket SSN e online.

#bambini#età evolutiva#dismetria#gambe#bacino#sviluppo motorio

Articoli correlati