OBJECTIVES: To evaluate the prevalence and risk factors of nocturnal hypoglycemia (NH) in children and adolescents with insulin-dependent diabetes mellitus. STUDY DESIGN: A total of 150 patients, 87% of whom were receiving conventional therapy, were admitted to the hospital for one night. Blood glucose (BG) levels were measured hourly from 10 PM to 8 AM. RESULTS: The prevalence of NH was 47%; NH was asymptomatic in 49% of the cases. Risk factors were as follows: at least two episodes of severe hypoglycemia from onset of insulin-dependent diabetes mellitus (p = 0.0004), insulin dosage > 0.85 IU/kg per day (p = 0.02), more than 5% of BG measurements < or = 3.3 mmol/L during the last month of monitoring (p = 0.04). The risk decreased significantly with age (p = 0.0001). Both high predictive values and significant relative risk were found for BG thresholds < or = 5.2 mmol/L at dinner time (p < 0.0001) and < or = 6.7 mmol/L at 7 AM (p < 0.0001). When BG values at 10 PM were used, prediction of NH was weak. CONCLUSIONS: Nocturnal hypoglycemia occurred frequently in children and adolescents with insulin-dependent diabetes mellitus. Our study found risk factors that will help pediatricians to identify those children with a high risk of NH. Especially in these patients, counseling based on the BG values before dinner and early in the morning is indicated to reduce the prevalence of NH.
OBJECTIVES: To evaluate the prevalence and risk factors of nocturnal hypoglycemia (NH) in children and adolescents with insulin-dependent diabetes mellitus. STUDY DESIGN: A total of 150 patients, 87% of whom were receiving conventional therapy, were admitted to the hospital for one night. Blood glucose (BG) levels were measured hourly from 10 PM to 8 AM. RESULTS: The prevalence of NH was 47%; NH was asymptomatic in 49% of the cases. Risk factors were as follows: at least two episodes of severe hypoglycemia from onset of insulin-dependent diabetes mellitus (p = 0.0004), insulin dosage > 0.85 IU/kg per day (p = 0.02), more than 5% of BG measurements < or = 3.3 mmol/L during the last month of monitoring (p = 0.04). The risk decreased significantly with age (p = 0.0001). Both high predictive values and significant relative risk were found for BG thresholds < or = 5.2 mmol/L at dinner time (p < 0.0001) and < or = 6.7 mmol/L at 7 AM (p < 0.0001). When BG values at 10 PM were used, prediction of NH was weak. CONCLUSIONS:Nocturnal hypoglycemia occurred frequently in children and adolescents with insulin-dependent diabetes mellitus. Our study found risk factors that will help pediatricians to identify those children with a high risk of NH. Especially in these patients, counseling based on the BG values before dinner and early in the morning is indicated to reduce the prevalence of NH.
Authors: Victor W Zhong; Jamie L Crandell; Christina M Shay; Penny Gordon-Larsen; Stephen R Cole; Juhaeri Juhaeri; Anna R Kahkoska; David M Maahs; Michael Seid; Gregory P Forlenza; Elizabeth J Mayer-Davis Journal: J Diabetes Complications Date: 2017-04-20 Impact factor: 2.852