[Accuracy of three methods of height prediction in a group of variant short stature children].
OBJECTIVE: The objective of this study was to determine how accurately three methods of height prediction estimate adult height in a group of 27 boys and 16 girls with untreated short stature who have reached adult height. Twenty of the patients were considered to have familial short stature, 8 constitutional delay of growth and puberty, 12 a combination of both conditions and 3 idiopathic short stature. PATIENTS AND METHODS: Height predictions at various chronological ages were retrospectively compared to final adult height according to the Tanner-Whitehouse Mark I, Bayley-Pinneau and Roche-Wainer-Thisen methods. Heights were expressed as centimeters and standard deviation scores (SDS) and related to Tanner standardized curves. RESULTS: The final heights reached were significantly below the target height [-2.0 (0.5) SDS versus -1.6 (0.6), p < 0.01]. The three methods over-estimated adult height. The Bayley-Pinneau method was especially accurate in girls and in familial short stature. The Tanner and Bayley-Pinneau predictions were similar in constitutional delay of growth and puberty. The Roche method gave the greatest errors in all groups. CONCLUSIONS: 1) Thee groups of normal variant short stature patients reached an adult height in the lower normal range. 2) The three method of height prediction over-estimated final height. 3) As in some pathological conditions, the Bayley-Pinneau was the most reliable method of height prediction in children with short stature. 4) These children may be considered as a historical group to evaluate the effect of growth-promoting therapies.