Blood ketone bodies in patients with recent-onset type 1 diabetes (a multicenter study)

Abstract

Insulin deficiency with glucagon excess leads to the release of ketone bodies (KBs) by the liver and excretion in the urine. So far, only KB monitoring in urine has been used during assessment of children with diabetes. Currently used nitroprusside strips for urine KB detection react only with acetoacetate (AcAc) and not with the most prevalent KB moiety - 3beta-hydroxybutyrate (3HB) - that is in equilibrium with AcAc (up to 10:1 ratio). To verify the significance of 3HB in the blood compared to that of AcAc in the urine in recently diagnosed type 1 diabetic subjects independent of the presence of diabetic ketoacidosis (DKA). A total of 118 consecutive newly diagnosed subjects with type 1 diabetes from different centers in Italy underwent intensive insulin therapy and hydration when necessary (38/118 = 32% with DKA). Hourly urine and capillary blood samples were used to monitor KBs until metabolic control was achieved. Although DKA was present in 32% of patients, blood 3HB was significantly increased (3.56 +/- 1.7 mmol/L) in 83% of the patients and correlated with blood glucose (r = 0.39; p < 0.01) according to a bimodal model. The strongest association was found between 3HB and venous pH (r =-0.56; p < 0.0001). Time required for blood 3HB normalization depended strongly on the starting blood KB values (r = 0.44; p < 0.0001) and was significantly lower than that required for disappearance of KB from urine (17.4 +/- 13.6 h, range 1-69 h vs. 19.7 +/- 17.8 h, range 1-120 h; p = 0.004). However, urine KBs disappeared before blood 3HB normalization in 23% of the patients. Blood 3HB evaluation is a better indicator of metabolic control compared to urine KB detection and is useful to predict the time required for blood KB clearing. Further studies are needed to assess its use in the early detection and management of DKA.