Aim: The creation of a database with somatometric parameters (body weight and eight) from school screening children and the comparison of nonscoliotic children with their counterparts suffering of scoliosis to Cobb angle >10° curves.

Material and Method: 3631 screened children where divided in 3 groups. The 1st group comprised normal children with 0° angle of trunk inclination, (ATI). The 2nd group comprised children with ATI >1° and < 6°. The third group comprised children with ATI >7° and Cobb angle >10°. The mean/median and standard deviation (± 1 SD) of body weight and height, the body mass index (BMI = weight/height²) and the corrected for the scoliotic curve height were calculated by age. Statistical analysis included descriptives (mean, ± 1 SD, median) and Mann - Whitney non-parametric test.

Results: In boys of the 1st and 2nd group 4,25% had obesity (BMI = 30-35), 2,9% severe obesity (BMI = 30-35) and 1,7% morbid obesity (BMI = 40-45)- while 6,4%, 1,9% and 1,3% in girls respectively. In the 3rd group girls, 27,2% were underweight (BMI = 16-20) and 11,3% severely underweight (BMI < 16), while among boys 42% were severely underweight. In the 3rd group there were no obese girls and only 5% obese boys. The comparison of body weight between scoliotic (3rd group) and nonscoliotic children (1st and 2nd group) fails to show any statistical difference.

In the 1st and 2nd group, the girls’ mean height is greater than that of boys aged 9–12 years but less when boys are 13-18 years old. In the 3rd group a mean of 1,15 cm increase is observed after height correction for the scoliotic curve, in boys and 1,3 cm in girls respectively. The comparison of body height (both uncorrected and corrected) between scoliotic and nonscoliotic children fails to show any statistical difference.

Discussion-Conclusions: A variety of findings regarding the stature and weight of AIS children has been published. In this studied Mediterranean sample of the population, the somatometric parameters of height and weight in children with scoliosis, regardless of curve type and site, are not statistically different from their nonscoliotic counterparts.