Interobserver Agreement of Contrast-Enhanced Ultrasonography for Assessing Insulin- Mediated Microvascular Recruitment.
Mary, Nydia   (Vanderbilt University)
Mentor: Biaggioni, A. Gamboa (Vanderbilt University)
It has been vastly demonstrated the correlation that exists between glucose-metabolism, oxygen and microvascular blood volume (MBV). At present, it has been a challenge to measure MBV in vivo, in the clinical scenario; also, there is limited data on the reproducibility of measurement systems for insulinmediated microvasculature blood volume changes. Hence, we propose to test the hypothesis that Contrast-Enhanced Ultrasonography (CEUs) has a low interobserver variability, in measurements of three variables of interest: Peak intensity (Microvascular blood flow), Slope (Microvascular blood flow velocity) and AUC (Area under the curve). We obtained images at baseline levels and after 90 min. of insulin infusion during a hyperinsulinemic-euglycemic clamp in a crossover randomized design, once with saline control and another with an intraarterial infusion of phentolamine; an alpha-adrenergic antagonist. Data were analyzed using QLAB software, as a real-time recording of vascular replenishment; obtaining pulsing intervals versus video intensity traces from which values of the variables of interest were derived. Ninety-two pairs of interobserver measurements from peak intensities, slopes and AUC; depicted in Bland- Altman plots, showed less than 10% interobserver variability, and only 7.52%, 5.37%, and 7.60 %; respectively; were outside the 95% limits of agreement (LoA) estimated by the mean difference/bias ) and the standard deviation of the differences (s), calculated as) ) ± 2s. Moreover, insulin tended to increase microvascular recruitment when combined with established doses of Phentolamine. These data are suggestive for CEUs as a highly reproducible method for measurement of microvascular recruitment and independent of the person performing of-line analysis.