The very large sizes can be very disabling and at the origin of a difficult psychological experience. Some families are in demand for a growth-curbing treatment. Currently, only hormonal treatments are offered in France but they are at the origin of many side effects. Northern European countries applied a surgical technique, deviated from the one used for a long time in leg asymmetries, to sterilize the proximal tibial and distal femoral growth cartilage, with most of the growth taking place at this level. The goal of this project is to be able to offer this technique to families in demand.

By "very large size" we mean a size greater than +3 SD (standard deviation) on the updated French reference curves. Most of the very large sizes are constitutional.

The first step in the management of the paediatrician endocrinologist with very large body height is to collect the auxological parameters (family height, not just parental height, birth height, weight, height, head circumference, body mass index, height, sitting height) and plot the corresponding adolescent growth curves. The clinical stage (interview, clinical examination) will attempt to rule out a diagnosis other than a very large constitutional size: a recent statural acceleration related to a hormonal cause with specific medical treatment (hyperthyroidism or hyper androgeny or early puberty or acromegaly or more rarely estrogen deficiency by mutation of the CYP19A1 aromatase gene or by mutation of the ERα receptor) and will seek a syndrome in which the very large size is a sign among others.

Among the main syndromes are Klinefelter's syndrome (prevalence 1 person/1,200 or 1/600 boy births), Marfan's syndrome (about 1/6,000 births, prevalence 1 to 5 persons/ 10000), triple X syndrome (1 to 5 persons/10,000).

Additional examinations will be guided by diagnostic hypotheses; the very large constitutional diagnosis remains a diagnosis of elimination.

The very large constitutional size is defined as family, i.e. related to the size of at least 1 of the 2 parents or close family members of the parents.

The target size can be calculated as an indication:

• Target size =[father size (cm) + mother size (cm) +13] /2 for boys
• Target size =[father size (cm) + mother size (cm) -13] /2 for girls However, experience shows that the target size is a poor prognostic tool in very large constitutional sizes: growth is always above it.

Very large constitutional size remains the most frequent diagnosis among etiological diagnoses of very large sizes.

The therapeutic question follows the diagnostic stage. Only very large sizes secondary to a hormonal cause can benefit from specific medical treatment.

Outside these situations, a therapeutic discussion will be proposed based on the size prognosis estimated by the pediatrician endocrinologist. In the event of a predicted very large size and if this represents an inconvenience for the adolescent, the paediatrician endocrinologist may propose a growth-blocking treatment.

Very large sizes can be disabling and cause a difficult psychological experience, so some families are in need of growth-restricting treatment, especially girls who consult more often and earlier than boys.

The estimation of adult prognosis by the pediatrician endocrinologist, formerly based on Bayley and Pinneau's tables, has been replaced by more modern software integrating Greulich and Pyle's tables, Tanner Whitehouse 2 and 3, and taking into account parental sizes and ethnic origin. This automated reading of bone maturation makes it possible to be more accurate in predicting final size, an essential step before any therapeutic discussion.

In the international literature (mainly the United States, Northern Europe and Australia) it is relatively consensual to consider a growth-blocking treatment for a prognosis ≥185 cm for girls and ≥ 200 cm for boys. For our study, we will use a prognosis ≥ 184 cm for girls and ≥ 198 cm for boys, which corresponds to a size of +3 SD on the updated French curves.

Currently in France, the therapeutic possibilities to slow down growth and thus reduce the final adult size are:

• Either medical treatments: excluding MA and causing many side effects;
• Or surgical treatments: performed anecdotally in France and based on publications from Northern Europe.

Inclusion Criteria:

• Adolescent (e) consultant for very large size with a demand for growth inhibition
• Very large idiopathic or secondary height without any other etiological treatment possible
• Tanner score ≥ 3 for boys and ≥ 2 for girls
• A0 ≤ 14 years for boys and ≤ 12.5 years for girls
• A chronological age of 10 to 16 for boys and 8 to 14 for girls
• Actual size from minimum 167.5 cm to 10 years old to at least 180 cm to 16 years old for boys and from minimum 160 cm to 8 years old to at least 174 cm to 14 years old for girls
• Predicted height ≥ 198 cm for boys and ≥ ≥ 184 cm for girls (refer to chapter 6.6)
• Radiological persistence of tibial and femoral tibial cartilage estimated residual growth ≥ 8 cm
• Absence of psychological contraindication to a curbing treatment of growth
• Agreement of the holders of the parental authority and the teenager for a surgery of epiphysiodesis type
• Agreement of the holders of the parental authority and the teenager to take part in the study
• Affiliated to a social security scheme

Exclusion Criteria:

• Very large size related to a medically curable etiology (hypersecretion of growth hormone called acromegaly)
• Refusal of the holders of the parental authority or the teenager for a surgery of the epiphysiodesis type
• Refusal of the holders of the parental authority or the teenager to take part in the study
• Medical or psychological contraindication to surgery.