Evaluation of the growth response of children with growth hormone deficiency according to the peak growth hormone levels in provocation tests

Abstract

Background

We aimed to evaluate the relation between the peak growth hormone (GH) levels in provocation tests and response to recombinant human GH (rhGH) therapy in patients with GH deficiency (GHD).

Methods

This was a cross-sectional, single-center, and retrospective study. A total of 135 patients under the age of 16 years who were diagnosed with GHD through insulin tolerance tests and L-DOPA stimulation tests and who received rhGH therapy for at least 2 years in the Pediatric Endocrinology Clinic of Akdeniz University Hospital between 1997 and 2021 were included in the study.

Results

The patients were divided into two groups: idiopathic GHD (group I, n = 119) and multiple pituitary hormone deficiencies or organic pathology on magnetic resonance imaging (group II, n = 16). The patients in group I were classified into three subgroups according to the peak GH values in the provocation tests (group Ia: peak GH <3 μg/L, n = 34; group Ib: peak GH between 3 and 7 μg/L, n = 71; group Ic: peak GH between 7 and 10 μg/L, n = 34). The median age was 11.5 years in group I (8.8 in group Ia, 12.1 in group Ib, 12.3 in group Ib) and 8.8 years in group II. The height standard deviation score (SDS) was −2.93 in group I (−2.85 in group Ia, −2.99 in group Ib, −2.94 in group Ic) and −3.79 in group II. The median Δheight SDS was 0.61 in group I and 1.05 in group II at the end of the first year of treatment and 0.31 in group I and 0.45 in group II at the end of the second year (p = 0.005 and p = 0.074, respectively). When the subgroups of group I were compared, height SDS, Δheight SDS, and height velocity (HV) SDS were all higher in group Ia at the end of the first year of rhGH therapy (p = 0.040, p = 0.029, and p = 0.005, respectively). The height SDS was still significantly higher in group Ia (p = 0.033) while the HV SDS and Δheight SDS were similar between the groups at the end of the second year of therapy (p = 0.164 and p = 0.522, respectively). There was a statistically significant association between the first-year HV SDS and the peak GH value in provocation tests in multiple regression analyses (p<0.001). In addition, the final model revealed that height SDS and weight SDS at the start of the treatment and the first-year HV SDS are the factors with a statistically significant effect on the second-year HV SDS (p = 0.022, p = 0.001, and p<0.001, respectively).

Conclusion

Our findings show that the lower the GH peak in provocation tests, the better the response to treatment. The best HV was observed in the first year of rhGH therapy, and the diagnosis should be checked in those patients who had a low first-year HV and did not have a severely low GH peak in provocation tests.