Growth hormone therapy in children: predictive factors and short-term and long-term response criteria.

PURPOSE: The definition of growth response in growth hormone (GH)-treated children is controversial. This study aims at: (1) evaluating short-term and long-term efficacy of GH treatment in a cohort of short children with GH deficiency (GHD); (2) assessing and compare various poor response criteria; (3) identifying predictive factors of growth response.

METHODS

Our study included 94 children, affected by isolated GHD and treated with GH until they reached final height. Criteria used for calculating the proportion of poor responders to GH for the first year were gain in height (ΔHt) SDS < 0.5 ("Bang criterion"), <0.3 or <0.4 SDS for less-severe and severe GHD, respectively ("Ranke criterion"), height velocity (HV) < mean -1 SDS ("Bakker criterion"); for adult height "Cianfarani criterion" was total ΔHt < 1 SDS.

RESULTS

After 1 year of treatment we defined "poor responders" 55.3% of patients according to Bang criterion, 40.9% according to Bakker criterion and 23.4% according to Ranke criterion. At the end of the treatment, poor responders according to Cianfarani criterion were 22.34%; almost everyone in our population (97.9%) achieved mMid-parental height (MPH). Median final Ht was -1.11 SDS. Our analysis revealed a significant negative association between ΔHt and age at diagnosis.

CONCLUSIONS

Bang criterion generated the highest number of poor responders, but had a low negative predictive value (67.5%); Ranke and Cianfarani criteria displayed similar rate of poor response. There is no reliable predictive factor of growth hormone response. However, almost all children treated reached MPH, suggesting good treatment efficacy.