Effect of discontinuation of long-term growth hormone treatment on carbohydrate metabolism and risk factors for cardiovascular disease in girls with Turner syndrome.

GH treatment increases insulin levels in girls with Turner syndrome (TS), who are already predisposed to develop diabetes mellitus and other risk factors for developing cardiovascular disease. Therefore, in the present study, we investigated carbohydrate metabolism and several other risk factors that may predict development of cardiovascular disease in girls with TS after discontinuation of long-term GH treatment. Fifty-six girls, participating in a randomized dose-response study, were examined before, during, and 6 months after discontinuing long-term GH treatment with doses of 4 IU/m(2).d ( approximately 0.045 mg/kg.d), 6 IU/m(2).d, or 8 IU/m(2).d. After a minimum of 4 yr of GH treatment, low-dose micronized 17beta-estradiol was given orally. Mean (SD) age at 6 months after discontinuation of GH treatment was 15.8 (0.9) yr. Mean duration of GH treatment was 8.8 (1.7) yr. Six months after discontinuation of GH treatment, fasting glucose levels decreased and returned to pretreatment levels. The area under the curve for glucose decreased to levels even lower than pretreatment level (P < 0.001). Fasting insulin levels and the area under the curve for insulin decreased to levels just above pretreatment level (P < 0.001 for both), although being not significantly different from the control group. No dose-dependent differences among GH dosage groups were found. At 6 months after discontinuation, impaired glucose tolerance was present in 1 of 53 girls (2%), and none of the girls developed diabetes mellitus type 1 or 2. Compared with pretreatment, the body mass index SD-score had increased (P < 0.001), and the systolic and diastolic blood pressure SD-score had decreased significantly at 6 months after discontinuation of GH treatment (P < 0.001 for both) although remaining above zero (P < 0.001, P < 0.05, and P < 0.005, respectively). Compared with pretreatment, total cholesterol (TC) did not change after discontinuation of GH treatment, whereas the atherogenic index [AI = TC/high-density lipoprotein cholesterol (TC/HDL-c)] and low-density lipoprotein cholesterol (LDL-c) had decreased; and both HDL-c and triglyceride levels increased (P < 0.001 for AI, LDL-c, and HDL-c; P < 0.05 for triglyceride). Compared with the control group, AI, serum TC, and LDL-c levels were significantly lower (P < 0.001 for all), whereas HDL-c levels were significantly higher (P < 0.05). In conclusion, after discontinuation of long-term GH treatment in girls with TS, the GH-induced insulin resistance disappeared, blood pressure decreased but remained higher than in the normal population, and lipid levels and the AI changed to more cardio-protective values.