(Dis)agreement and concordance of metabolic indices from the oral glucose tolerance test and mixed-meal tolerance test: Implications for application

Metabolic indices derived from the oral glucose tolerance test (OGTT) and mixed-meal tolerance test (MMTT) are often interpreted interchangeably; however, these tests represent distinct physiological stimuli. In order to examine potential differential metabolic responses to these tests, we characterized the (dis)agreement between OGTT- and MMTT-derived indices of insulin sensitivity, insulin secretion and β-cell function. We also described their convergent validity in individuals with normal glucose tolerance (NGT) or type 2 diabetes (T2D). We conducted a post hoc analysis of eight trials with participants who underwent either OGTT or MMTT, or both. Insulin sensitivity, insulin secretion and β-cell function were derived. Agreement was assessed using Bland-Altman regression approaches to estimate systematic and proportional bias and 95% limits of agreement. Linear mixed-effects models were used to estimate means by glycaemic status (NGT vs. T2D), and area under the receiver operating characteristic (AUROC) curve analyses were used to evaluate discriminative ability between glycaemic status for the respective test indices. No systematic bias was observed. Proportional bias was present for insulin secretion and β-cell function, with increasingly negative differences between methods (OGTT-MMTT) as outcome values increased. All indices exhibited relatively wide 95% limits of agreement, which widened further at higher magnitudes. The OGTT- and MMTT-derived means were significantly different between NGT and T2D. We did not observe differences in AUROC between tolerance tests. Agreement between OGTT- and MMTT-derived indices worsened as the respective value increased, suggesting differential stimulus-response characteristics in insulin secretion between tolerance tests. Despite this, descriptive analyses suggest that the test-specific indices demonstrate convergent validity.