Blood Glucose Area Under the Curve: Methodological Aspects

Abstract

To specify the influence of methods used in estimating area under the curve (AUC) and the meaning of total and incremental AUC, 75 glycemic responses to a mixed meal were studied in 75 diabetic patients, 39 with insulin-dependent diabetes mellitus and 36 with non-insulin-dependent diabetes mellitus. AUC was integrated with five computerized methods: polynomial interpolation of third and fourth degree, trapezoidal rule, Simpson's integration, and cubic interpolatory splines. Although these methods gave significantly different results (P < 0.001), a strong correlation was found between estimations of AUC with different methods (r > 0.99, P < 0.001). In addition, variation between methods was ≤2%, whereas the coefficient of variation between subjects was 38%. Total AUC was strongly correlated with basal blood glucose value (r = 0.90, P < 0.001), whereas incremental and positive AUC were not (r = 0.12 and 0.07, respectively, NS). Incremental and positive AUC were strongly correlated with glycemic rise (r = 0.89 and 0.93, respectively, P < 0.001), whereas total AUC was only slightly so (r = 0.31, P < 0.01). Incremental and positive AUC gave slightly but significantly different information on glucose response. These results suggest that variations related to the method used in estimating AUC are not clinically relevant and that a simple method such as trapezoidal rule can be used. Total AUC is a descriptive factor related to basal blood glucose value, whereas incremental and positive AUC more accurately describe glycemic response to foods.