LONG-TERM METABOLIC EFFECTS OF GH THERAPY IN CHILDREN WITH IDIOPATHIC GH DEFICIENCY

Background Increased insulin levels during GH replacement therapy are often reported both in adults and in children with GH deficiency (GHD), although discordant data fully describe the behavior of metabolic parameters during GH therapy in children and overall the majority of published data are limited to the first 12 months of follow-up and in small populations. Aim To evaluate the long-term metabolic influences of GH replacement in a large selected cohort of prepubertal children with idiopathic GHD during GH therapy. Subjects and Methods Data of 194 prepubertal GHD children (150 M, 44 F; age 10.85 ± 2.77 yrs) were analyzed. Before GH therapy and yearly up to 36 months we measured body mass index (BMI), waist circumference (WC), IGF-1, total-, HDL- and LDL-cholesterol, triglycerides, glucose and insulin levels, Homa-IR and HbA1c. In a subgroup of 34 patients (28 M, 8 F) we performed yearly OGTT and calculated ISI-Matsuda (ISI), Insulinogenic Index (InsIn) and Oral Disposition Index (DIo). Results No subject showed overt dysglycemia or dyslipidemia at baseline and during GH therapy. Compared to baseline (17.62 ± 3.14 Kg/m2), BMI significantly increased after 12 (17.81 ± 3.36 Kg/m2; p<0.001), 24 (18.24 ± 3.22 Kg/m2; p<0.001) and 36 months (18.43 ± 3.58 Kg/m2; p=0.014), without significantly changes in WC. Glucose (88.24 ± 9.09 vs 82.52 ± 9.00 mg/dl; p<0.001), insulin (7.41 ± 4.40 vs 4.96 ± 4.20 mg/dl; p<0.047), Hba1c (5.07 ± 0.43 vs 4.78 ± 0.52 %; p<0.001) and triglycerides (73.62 ± 34.88 vs 66.59 ± 34.30 mg/dl; p=0.029) were significantly higher after 1 year of therapy when compared to baseline, although within the normal values, without additional increase at 24 and 36 months. Homa- IR significantly increased from baseline to 12 months (5.07 ± 0.43 vs 1.03 ± 1.01; p<0.001) with a subsequent significant reduction after 24 months (2.36 ± 1.64; p=0.020), without further modifications at 36 months (2.06 ± 1.45; p=0.242). In the subgroup of 34 children, ISI (7.22 ± 0.38), InsI (0.75 ± 0.04) and DIo (5.52 ± 1.34) did not significantly change during the entire follow-up. Conclusions A slight deterioration of glucose and lipid metabolism seems to occur during the first year of GH treatment, with a re-establishment in the subsequent years of follow-up, although no overt worsening in insulin sensitivity occurs. A longer follow-u