Utility of casual postprandial glucose levels in type 2 diabetes management.

Abstract

Because readily available glycemic indicators are needed to guide clinical decision-making for intensification of diabetes therapy, our goals were to define the relationship between casual postprandial plasma glucose (cPPG) levels and HbA(1c) in patients with type 2 diabetes and to determine the predictive characteristics of a convenient glucose cutoff. We examined the relationship between cPPG levels (1-4 h post meal) and HbA(1c) levels in 1,827 unique patients who had both determinations during a single office visit. The population studied was predominantly African American and middle-aged, with average cPPG of 201 mg/dl and HbA(1c) of 8.4%. The prevalence of HbA(1c) > or = 7.0% was 67% and HbA(1c) >6.5% was 77%. Overall, cPPG and HbA(1c) were linearly correlated (r = 0.63, P < 0.001). The correlation between cPPG and HbA(1c) was strongest in patients treated with diet alone (n = 348, r = 0.75, P < 0.001) and weaker but still highly significant for patients treated with oral agents (n = 610, r = 0.64, P < 0.001) or insulin (n = 869, r = 0.56, P < 0.001). A cutoff cPPG >150 mg/dl predicted an HbA(1c) level > or = 7.0% in the whole group, with a sensitivity of 78%, a specificity of 62%, and an 80% positive predictive value. The same cPPG cutoff of 150 mg/dl predicted an HbA(1c) level >6.5%, with a sensitivity of 74%, a specificity of 66%, and an 88% positive predictive value. When rapid-turnaround HbA(1c) determinations are not available, a single cPPG level >150 mg/dl may be used during a clinic visit to identify most inadequately controlled patients and allow timely intensification of therapy.