Association between higher urinary normetanephrine and insulin resistance in a Japanese population.

Since activation of the sympathetic nervous system is associated with both impaired insulin secretion and insulin resistance, or namely with diabetes, evaluation of such activation in ordinary clinical settings may be important. Therefore, we evaluated the relationships between urinary concentration...

Full description

Bibliographic Details
Main Authors: Masaya Murabayashi, Makoto Daimon, Hiroshi Murakami, Tomoyuki Fujita, Eri Sato, Jutaro Tanabe, Yuki Matsuhashi, Shinobu Takayasu, Miyuki Yanagimachi, Ken Terui, Kazunori Kageyama, Itoyo Tokuda, Kaori Sawada, Kazushige Ihara
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2020-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0228787
Description
Summary:Since activation of the sympathetic nervous system is associated with both impaired insulin secretion and insulin resistance, or namely with diabetes, evaluation of such activation in ordinary clinical settings may be important. Therefore, we evaluated the relationships between urinary concentrations of the catecholamine metabolites, urinary normetanephrine (U-NM) and urinary metanephrine (U-M), and glucose metabolism in a general population. From 1,148 participants in the 2016 population-based Iwaki study of Japanese, enrolled were 733 individuals (gender (M/F): 320/413; age: 52.1±15.1), who were not on medication affecting serum catecholamines, not diabetic, and had complete data-set and blood glucose levels appropriate for the evaluation of insulin secretion and resistance, using homeostasis model assessment (HOMA-β and HOMA-R, respectively). Univariate linear regression analyses revealed significant correlations between both U-NM and U-M, and HOMA-β, but adjustment for multiple factors correlated with HOMA indices abolished these (β = -0.031, p = 0.499, and β = -0.055, p = 0.135, respectively). However, the correlation between U-NM and HOMA-R observed using univariate linear regression analysis (β = 0.132, p<0.001) remained significant even after these adjustments (β = 0.107, p = 0.007), whereas U-M did not correlate with HOMA-R. Furthermore, use of the optimal cut-off value of U-NM for the prediction of insulin resistance (HOMA-R >1.6) determined by ROC analysis (0.2577 mg/gCr) showed that individuals at risk had an odds ratio of 2.65 (confidence interval: 1.42-4.97) after adjustment for the same factors used above. Higher U-NM concentrations within the physiologic range are a significant risk factor for increased insulin resistance in a general Japanese population.
ISSN:1932-6203