Hyperbolic relationship between insulin secretion and sensitivityThe relationship between two variables x and y is said to be hyperbolic if 𝑥 × 𝑦 = 𝑐, where c is a constant. The x and y variables can be log-transformed (ln(𝑥 × 𝑦) = ln(𝑥) + ln(𝑦) = ln(𝑐)), and the hyperbolic relationship between these two variables can be re-expressed as linear model such that: ln(𝑦) = ln(𝑐) − 1ln(𝑥), where the regression coefficient in the linear model of ln(𝑦) as a function of ln(𝑥) is 𝛽 = −1. 1,2 Note that β = -1 regardless of the units in which y and x are expressed and regardless of the base of logarithms used. The hypothesis that the relationship between insulin secretion and insulin sensitivity is hyperbolic is usually tested by determining if the slope of a linear model of ln(insulin secretion) as a function of ln(insulin sensitivity) is not significantly different from -1 (i.e., the 95% CI for the slope includes -1). [1][2][3][4] Reasons for using SMA regression to model relationships between insulin secretion and sensitivityWe studied the relationship between insulin secretion and sensitivity by estimating ln(insulin secretion) as a function of ln(insulin sensitivity). The models provided by the SMA regression were chosen after comparing their fit with models from linear and orthogonal regression using standard error of the residuals and confidence intervals for intercepts and slopes, Supplementary Tables S1 and S2. One reason for not considering linear regression models in this case is that this method only accounts for variability in y, while orthogonal and SMA regression models account for variability in both x and y. 5 Also, in our study, linear regression models had the highest standard error of the residuals in all cases, Supplementary Tables S1 and S2. Standard errors of the residuals were the same or slightly lower for orthogonal regression models when compared to those from SMA regression. However, SMA models were chosen because orthogonal regression is most useful when one is studying the relationship between two variables that are estimates of the same entity. 5 Another reason for choosing SMA models was that, their confidence intervals are estimated more efficiently than those for orthogonal regression models, and the confidence intervals for SMA slopes tend to be exact or close to exact in most practical instances. 6 Generally, SMA and orthogonal regression models led to the same conclusions (same qualitative results). In the OGTT dataset, both SMA and orthogonal regression indicated that the slope for ln(CIR120) as a function of ln(ISI0) was essentially -1 in the whole dataset (SMA: (95% CI -0•999 to -0•908); orthogonal: (95% CI -0•996 to -0•686)), significantly different from -1 in NGR subjects only (SMA: (95% CI -0•948 to -0•854); orthogonal: (95% CI -0•869 to -0•656)), and that it was not significantly different from -1 in the IGR subjects only (SMA: (95% CI -1•093 to -0•918); orthogonal: (95% CI -1•180 to -0•853)). In the IV/CLAMP dataset, both approaches suggested that the slope for ln(AIR) as a funct...
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Context Insulin secretion and sensitivity regulate glycemia, with inadequately compensated deficiencies leading to diabetes. Objective We investigated effects of weight loss, an intensive lifestyle intervention (ILS), and metformin on the relationship between insulin secretion and sensitivity using repository data from 2931 participants in the Diabetes Prevention Program clinical trial in adults at high risk of developing type 2 diabetes. Methods Insulin secretion and sensitivity were estimated from insulin and glucose concentrations in fasting and 30-minute post-load serum samples at baseline and 1, 2, and 3 years after randomization, during the active intervention phase. The non-linear relationship of secretion and sensitivity was evaluated by standardized major axis regression to account for variability in both variables. Insulin secretory demand and compensatory insulin secretion were characterized by distances along and away from the regression line, respectively. Results ILS and metformin decreased secretory demand while increasing compensatory insulin secretion, with greater effects of ILS. Improvements were directly related to weight loss; decreased weight significantly reduced secretory demand [b=-0.144 SD; 95% CI (-0.162, -0.125)/ 5 kg loss] and increased compensatory insulin secretion [b=0.287 SD, 95% CI (0.261, 0.314)/ 5 kg loss]. In time-dependent hazard models, increasing compensatory insulin secretion [HR=0.166 per baseline SD, 95% CI (0.133, 0.206)] and weight loss [HR=0.710 per 5 kg loss, 95% CI (0.613, 0.819)] predicted lower diabetes risk. Conclusions Diabetes risk reduction was directly related to the amount of weight loss, an effect mediated by lowered insulin secretory demand (due to increased insulin sensitivity) coupled with improved compensatory insulin secretion.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.