Short-term, high-dose testosterone treatment fails to reduce adult height in boys with constitutional tall stature

Height predictions based on three different methods (Bayley-Pinneau [BP], Tanner-Whitehouse Mark II [TW II], Roche-Wainer-Thissen [RWT]) were compared to adult heights in 19 males with constitutional tall stature previously treated with high-dose testosterone oenanthate for 6 months (group A) and 25 untreated tall males (group B). Their chronological ages (CA) at the initial evaluation of tall stature ranged from 12.1 to 16.6 years in group A and from 10.4 to 15.7 years in group B; at the time of assessment of adult height ages ranged from 18.0 to 26.5 years and from 18.4 to 25.1 years, respectively. Height measurements and predicted adult heights were expressed as height standard deviation scores (height SDS) for chronological age using the tables of Reinken and van Oost [14]. Height SDS in group A were 2.8 (range = 1.8–5.4) before testosterone treatment, 3.0 (range = 2.0–4.8) thereafter and finally 3.0 (range = 2.1–4.2) (P=NS) and in group B 2.7 (range = 0.5–4.3) and 2.4 (range = 1.3–3.5) (P=NS). A significant difference between adult height SDS and predicted height SDS according to BP was detected both in group A (3.0; range = 2.1–4.2 vs 3.6; range = 2.4–5.0; P≤0.004) and group B (2.4; range = 1.3–3.5 vs 3.0; range = 2.0–4.9; P≤0.0002), whereas no significant difference between adult height SDS and predicted height SDS according to TW II and RWT was found in either group. These data indicate that BP height predictions overestimated adult height in our patient group of treated and untreated males with constitutional tall stature. In contrast, the TW II and RWT methods were more accurate in predicting adult height in these patients, but also failed to demonstrate that testosterone therapy in boys with constitutional tall stature can be limited to a 6-month period in order to reduce adult height.