Growth Hormone Stimulation Testing in Childhood: Diagnostic Pathways and Discordant Results in a Tertiary Centre Cohort

OBJECTIVE: Growth hormone stimulation tests (GHSTs) are central to the diagnosis of growth hormone deficiency (GHD) despite recognised limitations. Two failed GHSTs are recommended to confirm GHD; however, a single failed test may be considered sufficient in selected cases. This study aimed to describe GHST practice in a tertiary paediatric endocrine centre, to determine how often an initial subthreshold GHST was not followed by a final diagnosis of GHD and to assess the diagnostic performance of IGF-I, IGFBP-3 and the IGF-I/IGFBP-3 molar ratio. DESIGN: Retrospective observational study. PATIENTS: Seventy-three children who underwent GHST for suspected GHD at a tertiary Irish paediatric endocrinology centre. MEASUREMENTS: Clinical, biochemical and radiological data were reviewed, including height, BMI, height velocity, IGF-I, IGFBP-3, bone age and GHST results. RESULTS: GHD was diagnosed in 34 children (47%) and excluded in 39 (53%). Of 47 children with a first peak GH below 7 µg/L, GHD was subsequently excluded in 13, an apparent false-positive rate of 27.7% (95% CI 16.9-41.8). Six of the 13 had a severely subthreshold initial response (peak GH < 3 µg/L). Discordant results were clustered in peripubertal children. IGF-I had poor sensitivity and specificity (45.5% and 50.0%, respectively); IGFBP-3 had low sensitivity but high specificity (9.1% and 100%, respectively). The IGF-I/IGFBP-3 molar ratio did not differ between diagnostic groups. CONCLUSIONS: Approximately one in four children with a subthreshold first GHST was not ultimately diagnosed with GHD. These findings support the need for repeat testing before initiating treatment in children without independent evidence of pituitary disease.