BMI, BUN Can Predict Death from Diabetic Ketoacidosis in Children

Higher BMI and blood urea nitrogen levels are associated with an increased risk for death from diabetic…. 

Higher BMI and blood urea nitrogen levels are associated with an increased risk for death from diabetic ketoacidosis among children, according to data from a retrospective study. 

Suresh Havalad, MD, associate professor of pediatrics at Chicago Medical School and division director of Pediatric Critical Care at Advocate, analyzed data for all children aged younger than 19 years who presented to the hospital with diabetic ketoacidosis who survived (controls, n=122) and died (cases, n=10) during a 6-year period.

They analyzed medical charts and autopsy reports and collected data on sample and treatment characteristics, serum chemistries, and histological or radiographic evidence for cerebral edema. Case-control comparisons were conducted using multivariate logistic-regression models and selected clinical parameters, including serum glucose, creatinine, blood urea nitrogen (BUN), calculated osmolality and BMI.

Death from diabetic ketoacidosis was significantly associated with BMI (OR=1.18; 95% CI, 1.07-1.31), BUN (OR=1.11; 95% CI, 1.04-1.18), glucose (OR=1; 95% CI, 1-1.01), creatinine (OR=3.94; 95% CI, 1.51-10.33) and calculated osmolality (OR=1.06; 95% CI, 1.03-1.08) in univariate analysis. In multivariate model, however, significant predictors of mortality were only BMI (OR=1.14; 95% CI, 1.03-1.27) and BUN (OR=1.09; 95% CI, 1.01-1.18).

Using established criteria for obesity and azotemia, children with diabetic ketoacidosis were more likely to die if BMI was .30 (OR=30.83; 95% CI, 5.35-177.36) and BUN was .40 mg/dL (OR=98.4; 95% CI, 9.23-1049.17).

Despite the small sample size, the results suggest that children with higher BMI and BUN are at increased risk of death from diabetic ketoacidosis. Further study is needed to confirm and extend these findings,” the researchers wrote in their abstract.

Presented at: 2012 Pediatric Academic Societies Annual Meeting; April 28-May 1; Boston. Vaikom House AK. Abstract #1515.272