Advanced glycation end-products and its soluble receptor are not independent predictors of incident dysglycaemia or metabolic syndrome in women with polycystic ovary syndrome: a prospective observational study

BACKGROUND: To evaluate the association of serum advanced glycation end-products (AGEs) and its soluble receptor of AGE (sRAGE) levels with dysglycaemia and metabolic syndrome in women with polycystic ovary syndrome (PCOS). METHODS: This was an analysis of a cohort of women with PCOS who were prospectively recruited for a longitudinal observational study on their endocrine and metabolic profile between January 2010 and December 2013. The association of serum AGEs and sRAGE levels with dysglycaemia and metabolic syndrome at the second-year visit (the index visit) and the sixth-year visit (the outcome visit) were determined. Comparisons of continuous variables between groups were made using the Mann–Whitney U-test. Spearman test was used for correlation analysis. Multivariate binary logistic regression analysis was employed to identify the factors independently associated with the outcome events. RESULTS: A total of 329 women were analysed at the index visit. Significantly lower serum levels of sRAGE (both p < 0.001), but no significant difference in AGEs, were observed in those with dysglycaemia or metabolic syndrome. At the outcome visit, those with incident metabolic syndrome had a significantly lower initial serum sRAGE levels (p = 0.008). The association of serum sRAGE with dysglycaemia and metabolic syndrome at the index visit was no longer significant in multivariate logistic regression after controlling for body mass index, free androgen index and homeostatic model assessment for insulin resistance (HOMA-IR). sRAGE was also not significantly associated with incident metabolic syndrome at the outcome visit on multivariate logistic regression. CONCLUSIONS: Serum sRAGE levels are significantly lower in women with PCOS who have dysglycaemia or metabolic syndrome, and in those developing incident metabolic syndrome in four years. However, it does not have a significant independent association with these outcome measures after adjusting for body mass index, free androgen index and HOMA-IR.