We designed a retrospective case-control study that recruited children who visited Severance Children's Hospital from March 2011 to July 2020. These children's heights were below a standard deviation score (SDS) of -2, and these children underwent two GH stimulation tests and sella MRI. Initial evaluation included 287 children. Patients with underlying diseases that could affect their growth such as hypothyroidism, panhypopituitarism, tubulointerstitial nephropathy, congenital heart disease, Russell-Silver syndrome, Turner syndrome, Noonan syndrome, congenital megacolon, rickets, and achondroplasia (n=47) were excluded. Also excluded were those with anatomical abnormalities on sella MRI such as pituitary adenoma, craniopharyngioma, medulloblastoma, empty sella syndrome, Rathke cleft cyst, and supratentorial brain tumor (n=58). Additionally, patients who were lost to follow-up and did not have electronic medical records through 1 year after treatment (n=74) were excluded. After exclusions, the total number of participants in the study was 152. GH deficiency was defined as: (1) short stature of a height below the third percentile for age and sex; (2) delayed bone age (BA); and (3) peak GH level below 10 ng/mL on two stimulation tests. For analysis, we also divided the GH deficiency into severe deficiency (peak GH < 5 ng/mL), moderate deficiency (peak GH 5–6.9 ng/mL), and mild deficiency (peak GH 7–9.9 ng/mL) groups. ISS was defined as a height below the third percentile for age and sex without any other identifiable hormonal, skeletal, or systemic disorders. Of the 152 participants in the study, 105 had GH deficiency and 47 had ISS (Fig. 1).

All participants underwent a physical examination, biochemical study, and imaging study at the start of treatment; height was measured every six months thereafter. Anthropometric measures were obtained using a Harpenden stadiometer and digital scale. Growth parameters, including height and midparental height (MPH), were standardized as a standard deviation score (SDS) according to the 2017 Korean growth standard [9]. MPH was defined as the mean of the parental heights minus 6.5 cm in girls and plus 6.5 cm in boys. The GH, insulin-like growth factor I (IGF-I), and IGF binding protein-3 (IGFBP-3) levels were measured by chemiluminescence immunoassay using the Immulite 2000 Immunoassay System (Siemens Healthcare Diagnostics Inc., Erlangen, Germany). The IGF-I SDS and IGFBP-3 SDS were calculated according to the Korean standard reference [10]. For GH stimulation tests, we assessed serum GH concentration levels at 0, 30, 60, 90, and 120 minutes after provocation by insulin (0.1 unit/kg, intravenously) and L-dopa (125 mg, body weight under 12.5 kg; 250 mg, body weight 12.5 kg or more, orally) or arginine (0.5 g/kg to a maximum of 40 g, intravenously). An x-ray of the left hand, including the wrist, was obtained for BA determination by two pediatric endocrinologists using the Greulich-Pyle method [11]. All participants underwent brain MRI to rule out intracranial lesions or to determine pituitary gland morphology. The patients were scanned on various 3.0 T MRI units (Discovery MR750/750, GE Healthcare [Chicago, IL, USA]; or Ingenia CX, Philips Medical Systems [Amsterdam, The Netherlands]). The imaging protocols included pre-contrast sagittal T1-weighted, sagittal T2-weighted, high-resolution coronal T2-weighted, and contrast-enhanced T1-weighted imaging (T1C). The sequence parameters of the high-resolution coronal T2-weighted images were: TR/TE=2447/80 msec, slice thickness=1.5 mm, intersection gap=0 mm, field of view=20 × 20 cm, and pixel spacing=0.391 × 0.391 mm. A radiology specialist measured the outermost boundaries of anteroposterior (AP) diameter and height of pituitary glands on midsagittal images, and the maximal horizontal diameter was measured on coronal images (Fig. 2). The PVs were calculated using the formula of an ellipsoid (V=[height × depth × width]/2). All measured values are calculated in SDS according to the normative data for pituitary diameters in a Turkish pediatric population [12].

All patients were administered the GH dose based on their body weight. The first treatment dose was 0.72 units/kg/wk for patients with GH deficiency and 0.75–0.9 units/kg/wk for those with ISS [13]. We adjusted the GH dose to maintain the IGF-1 level in the normal range for age (0 to +2 SD) [13]. Height SDS gains after the first year of GH treatment were also assessed. We defined a good response to GH treatment as a gain of 0.6 or more in height SDS in the first year of treatment [14].

All data are presented as number (%) or mean±standard deviation. Chi-square test was used for categorical variables, and the t-test was used to compare the means of the single variant analysis. Statistical significance was set at P-value < 0.05. Data analyses were performed using IBM SPSS Statistics ver. 20.0 (IBM Co., Armonk, NY, USA).

This study was approved by the Institutional Review Board (IRB) of Severance Hospital, Yonsei University College of Medicine, Seoul, Korea (No. 4-2021-0012). The requirement for informed consent was waived by the IRB.

In this study, 152 children participated, of which 105 were diagnosed with GH deficiency and 47 were diagnosed with ISS. Table 1 shows the clinical characteristics and PVs of patients with GH deficiency or ISS. There was no significant differences in sex and age at the start of treatment and no significant difference in MPH between the GH deficiency and ISS groups. Initial height SDS was substantially smaller in patients with ISS than in those with GH deficiency (P=0.001). IGFBP-3 level SDS in GH deficiency was lower than that in ISS (P<0.05), but IGF-I level SDS did not show a significant difference between groups (P=0.248). AP and horizontal diameter, the height of the pituitary gland, and PV SDS did not show differences between the GH deficiency and ISS groups (P=0.249, P=0.851, P=0.314, P=0.536, respectively).

The PV SDS seemed to be smaller in the severe GH deficiency group than in the mild deficiency and ISS groups, although it was not significantly different (P=0.082). The height of the pituitary gland was associated with severity of GH deficiency (P<0.005) (Fig. 3).

In patients with GH deficiency, PV was the only factor related to the treatment response. Initial height SDS, IGF1 level, and PV showed significant differences between the good and poor groups in univariable analysis (P<0.001, P=0.012, P=0.003, respectively). However, only PV was significant when multiple variable analysis was performed simultaneously (P=0.001) (Table 2). The association between PV and response to GH treatment was not consistent between GH deficiency patients and ISS patients. The PV was not significantly different between good and poor response groups in patients with ISS. The IGFBP3 level SDS was the only statistically correlated factor for responsiveness in univariable analysis (P<0.05). In multiple linear regression analysis, there was no correlation for responsiveness of initial height, IGF-1 level, or PV SDS in patients with ISS (P=0.879, P=0.252, P=0.086, respectively) (Table 3).

Pearson correlation analysis determined that a height SDS gain during the first year of GH treatment was statistically correlated with the PV SDS in patients with GH deficiency but not in ISS patients (Pearson correlation coefficient [r]=-0.2, P=0.028; Pearson correlation coefficient [r]=-0.2, P=0.169, respectively) (Fig. 4).