Height and weight in children with Prader-Willi syndrome, diagnosed by standard clinical criteria, follow a specific developmental pattern resulting in early childhood obesity, absent pubertal growth spurt and adolescent short stature. New molecular techniques (methylation analysis, fluorescence in situ hybridization) now allow the unequivocal diagnosis of Prader-Willi syndrome (PWS). We investigated the possibility of a bias in syndrome-specific growth standards based on clinically diagnosed patients by comparing these standards with new standards derived from 100 German patients with molecularly confirmed PWS, none of whom had received a growth-promoting therapy. Height centile curves of the German patients fall in the tall range of standards derived from American patients. This is mainly due to an elevation of the lower centile ranges in both sexes. When the height standards derived from German patients are compared to those of a large multinational cohort of patients, 78% of whom were not confirmed by genetic testing, only minor differences in the height centiles become apparent. The population background therefore does not appear to play a major role for the observed differences. In a marked proportion of patients a decreased sitting height/height ratio is found. This was usually associated with scoliosis. Weight standards from our study group show that after 14 y of age German girls with PWS are heavier than their American counterparts. Standards for the body mass index of German patients of both sexes are increased over normal reference standards (p < 0.0001) and do increase with age (boys: p = 0.0038; girls: p = 0.0004). PWS genotypes or sex had no apparent influence on SDS for height, weight and body mass index.
Because of the observed differences to other growth standards, use of the newly constructed centile curves is advocated in German patients with molecularly confirmed PWS to avoid delay in the diagnosis of additional growth-compromising conditions.