Context: Whether baseline preserved ratio impaired spirometry (PRISm) is associated with the risk of developing type 2 diabetes (T2D) and if this association could be mediated by circulating metabolites remains to be elucidated.
Objective: To measure the prospective association of PRISm with T2D and potential metabolic mediators thereof.
Methods: This study used data from the UK Biobank and included 72 683 individuals without diabetes at baseline. PRISm was defined as the predicted forced expiratory volume in 1 second (FEV1) <80% and the FEV1/forced vital capacity ratio ≥0.70. Cox proportional hazards modeling was performed to assess the longitudinal relation between baseline PRISm and incident T2D. Mediation analysis was used to explore the mediation effects of circulating metabolites in the path from PRISm to T2D.
Results: During a median follow-up of 12.06 years, 2513 participants developed T2D. Individuals who had PRISm (N = 8394) were 47% (95% CI, 33%-63%) more likely to develop T2D compared with those who had normal spirometry (N = 64 289). A total of 121 metabolites showed statistically significant mediation effects in the path from PRISm to T2D (false discovery rate <0.05). Glycoprotein acetyls, cholesteryl esters in large high-density lipoprotein (HDL), degree of unsaturation, cholesterol in large HDL, and cholesteryl esters in very large HDL were the top 5 metabolic markers, with mediation proportions (95% CI) being 11.91% (8.76%-16.58%), 11.04% (7.34%-15.55%), 10.36% (7.34%-14.71%), 9.87% (6.78%-14.09%), and 9.51% (6.33%-14.05%), respectively. A total of 11 principal components that explained 95% variance of the metabolic signatures accounted for 25.47% (20.83%-32.19%) of the relation between PRISm and T2D.
Conclusions: Our study revealed the association of PRISm with T2D risk and the potential roles of circulating metabolites in mediating this association.
Keywords: mediation analysis; metabolomics; preserved ratio impaired spirometry; type 2 diabetes.
© The Author(s) 2023. Published by Oxford University Press on behalf of the Endocrine Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.