In the past decade, several population-based studies reported the relationship between circulating Fetuin-A concentrations and PCOS. However, the conclusions of these studies were contradictory. In previous studies, circulating Fetuin-A levels were increased, decreased, or unchanged in PCOS patients compared with healthy women [22, 31-35]. In the current investigation, we found that circulating Fetuin-A concentrations were markedly elevated in women with PCOS compared with healthy women. Our results were consistent with those of Enli et al. [33 -35] but contrary to those of Díaz et al. [22]. The cross-sectional nature of the study and population heterogeneity might be related to differences in outcomes. Furthermore, previous studies had small sample sizes, and some had no healthy controls or did not include obese patients with PCOS. In this study, we avoid these shortcomings. In addition, we employed newly diagnosed PCOS women to avoid the effects of medication, lifestyle interventions, and the duration of the disease that were associated with those patients who were under treatment. The reason for the rise in circulating Fetuin-A was unknown. We speculate that the metabolic disorders and hyperandrogenism caused by IR (hyperinsulinemia) may promote the synthesis and release of Fetuin-A in vivo. In addition, increased Fetuin-A might be derived, at least in part, by the status of low-grade inflammation, since inflammatory cytokines, such as CRP, were increased in women with PCOS.
Two previous studies found that Fetuin-A inhibited the insulin receptor tyrosine kinase and Toll-like receptor 4 in liver and muscle cells to suppress insulin signaling and stimulate inflammatory signaling pathways [13, 14]. However, population-based studies showed that there was no correlation between Fetuin-A and IR in diabetic patients, and there was no correlation between Fetuin-A and the risk of diabetes [36-37]. In the current work, we found that serum Fetuin-A was positively correlated with BMI, WHR, TG, TC, LDL-C, HOMA-IR, LH, T, and DHEA-S, suggesting that Fetuin-A was associated with hyperinsulinemia and hyperandrogenism. These data supported the results of Pal et al. and Srivas et al. [13]. Because circulating Fetuin-A levels were associated with both T and IR, and hyperandrogenism and IR were essential characteristics of PCOS, it was of clinical significance to consider Fetuin-A a biomarker for PCOS.
Surprisingly, serum Fetuin-A concentration increased significantly in obese/overweight PCOS women but did not change in normal women with obese/overweight. This result was consistent with a previous study [38]. The reason for this was unknown. We speculate that the main factors affecting circulating Fetuin-A might be hyperinsulinemia and hyperandrogenemia, not adipose mass. In our study cohort, a small number of overweight women might affect the results. In addition, under hyperinsulinemia and hyperandrogenism, elevated circulating Fetuin-A raised the question of whether lowering Fetuin-A concentrations were the key to improving IR and hyperandrogenism. Secondly, because serum Fetuin-A levels were related to hyperandrogenemia, it was important to observe whether circulating Fetuin-A concentration would change with the menstrual cycle due to the change of hormone levels. To address these questions, further study is necessary.
We analyzed the ROC curve to explore the best cut-off point for predicting PCOS with circulating Fetuin-A. Our results showed that cyclic Fetuin-A was a good predictor for PCOS patients. Thus, we consider that the association of Fetuin-A with PCOS may be due to the high incidence of IR in PCOS population.
Our research also has some limitations. Firstly, the study population is young women, so our results may not apply to an elderly population. Secondly, our results are based on a single measurement of Fetuin-A. Without repeated measures at different time points, the introduction of random measurement errors in determining biochemical variables is possible. Finally, the nature of the cross-sectional study makes it impossible for our results to explain the causal relationship between increased Fetuin-A levels and the occurrence of IR and PCOS.