BACKGROUND: GH has an insulin antagonist effect, and GH treatment has therefore been suggested to impair glucose metabolism and increase risk of diabetes mellitus. SETTING: Data from 11,686 GH-treated patients in the Genetics and Neuroendocrinology of Short Stature International Study (GeNeSIS), a multinational observational study of children with growth disorders, were analyzed for diabetes incidence. Baseline diabetes prevalence was determined from a GH-naive subgroup. METHODS: Prevalence and incidence (by standardized incidence ratio) were compared with results from patients aged less than 20 yr in the U.S. SEARCH for Diabetes in Youth study. RESULTS: Baseline type 1 diabetes prevalence per 1000 persons was 4.92 (95% confidence interval = 1.91-12.58) in GeNeSIS and 1.03 (0.97-1.10) in SEARCH for 0- to 9-yr-olds, and 7.33 (4.20-12.77) and 2.99 (2.78-2.98), respectively, for 10- to 19-yr-olds; there were no GeNeSIS cases of type 2 diabetes before GH initiation. During a median 1.8 yr of GH treatment, diabetes standardized incidence ratios for U.S. patients were 1.4 (0.5-3.1) for type 1 and 8.5 (2.8-19.5) for type 2, and for all patients was 1.4 (0.7-2.4) for type 1 and 6.5 (3.3-11.7) for type 2. Among the 11 patients with incident type 2 diabetes, risk factors for diabetes were identified in 10 patients. Glucose concentrations normalized for seven of nine patients for whom glycemic status could be determined (three of whom continued GH therapy and four who discontinued). CONCLUSION: The incidence of type 2 diabetes was higher in GH-treated children than the general population. Monitoring of glucose, before and periodically during GH treatment, is recommended for those with preexisting type 2 diabetes risk factors.
BACKGROUND: GH has an insulin antagonist effect, and GH treatment has therefore been suggested to impair glucose metabolism and increase risk of diabetes mellitus. SETTING: Data from 11,686 GH-treated patients in the Genetics and Neuroendocrinology of Short Stature International Study (GeNeSIS), a multinational observational study of children with growth disorders, were analyzed for diabetes incidence. Baseline diabetes prevalence was determined from a GH-naive subgroup. METHODS: Prevalence and incidence (by standardized incidence ratio) were compared with results from patients aged less than 20 yr in the U.S. SEARCH for Diabetes in Youth study. RESULTS: Baseline type 1 diabetes prevalence per 1000 persons was 4.92 (95% confidence interval = 1.91-12.58) in GeNeSIS and 1.03 (0.97-1.10) in SEARCH for 0- to 9-yr-olds, and 7.33 (4.20-12.77) and 2.99 (2.78-2.98), respectively, for 10- to 19-yr-olds; there were no GeNeSIS cases of type 2 diabetes before GH initiation. During a median 1.8 yr of GH treatment, diabetes standardized incidence ratios for U.S. patients were 1.4 (0.5-3.1) for type 1 and 8.5 (2.8-19.5) for type 2, and for all patients was 1.4 (0.7-2.4) for type 1 and 6.5 (3.3-11.7) for type 2. Among the 11 patients with incident type 2 diabetes, risk factors for diabetes were identified in 10 patients. Glucose concentrations normalized for seven of nine patients for whom glycemic status could be determined (three of whom continued GH therapy and four who discontinued). CONCLUSION: The incidence of type 2 diabetes was higher in GH-treated children than the general population. Monitoring of glucose, before and periodically during GH treatment, is recommended for those with preexisting type 2 diabetes risk factors.