Childhood metabolic measurements predicted diabetes development years later

A child’s blood pressure, BMI, blood glucose level and other laboratory tests and simple office measures may predict the risk for developing type 2 diabetes nine and 26 years later, according to a new report.

Researchers analyzed longitudinal data from two studies. The National Growth and Health Study followed 1,067 black and white girls enrolled at ages 9 and 10 for nine years, and the Princeton Follow-up Study tracked 822 black and white school-aged children for 22 to 30 years beginning in 1973 to 1976.

In the Princeton Follow-up Study, individuals were more likely to have diabetes at age 39 years if they had high systolic BP, a high BMI, glucose levels of at least 100 mg/dL, low high-density lipoprotein levels and high triglyceride levels in childhood.

“When BMI, systolic BP and diastolic BP were all lower than the 75th percentile and there was no parental diabetes, the likelihood of children developing type 2 diabetes 22 to 30 years later was only 1%,” the researchers wrote.

In the National Growth and Health Study, childhood high systolic BP, insulin concentration and having a parent with diabetes increased the risk for having diabetes at age 19.

“If childhood BMI, systolic BP and diastolic BP were all lower than the 75th percentile, the likelihood of type 2 diabetes at age 19 years was 0.2%, 0.2% if the parents were also free of diabetes and 0.3% if childhood insulin was also less than the 75th percentile,” they wrote.

The data are important “since children with systolic BP, triglyceride, BMI and insulin in the top fifth percentile, a glucose concentration of at least 100 mg/dL and a parent with diabetes could be targeted for primary prevention of type 2 diabetes through diet, exercise and possibly insulin-sensitizing drug intervention, with special focus on overweight children with positive family history of diabetes,” they wrote. – by Katie Kalvaitis

Morrison JA. Arch Pediatr Adolesc Med. 2010;164:53-60.

This is an important and useful paper for those in pediatrics and family practice. First, it provides evidence that a family history of diabetes along with simple measurements of BMI and systolic BP can identify prospectively, and with a high degree of accuracy, which children are likely to develop type 2 diabetes as adults. The addition of fasting glucose, insulin and triglyceride values improves the prediction slightly, but are not absolutely essential if they are not readily affordable or available. Although the observation that children who are obese with a family history of diabetes are likely to ultimately develop diabetes is not new, this study confirms that those whose BMI is in the upper fifth percentile deserve the practitioner’s most targeted nutrition and lifestyle intervention efforts. Perhaps just as importantly, however, was the observation that children whose BMI and systolic BP is less than the 75th percentile are extremely unlikely to develop type 2 diabetes. Given the public concern about an ‘epidemic’ of type 2 diabetes, these data will allow the pediatrician and family practitioner to alleviate the concerns of parents of children whose BMI are less than the 75th percentile, even when there is a family history of diabetes. Additionally, these data will allow the pediatrician comfort and confidence in optimizing his or her counseling time directed toward those children who are most likely to develop type 2 diabetes as adults.

- Dennis M. Bier, MD
Professor of Pediatrics
Director USDA/ARS Children's Nutrition Research Center
Baylor College of Medicine Houston, Texas