Is prolactin the missing link in adipose tissue dysfunction of polycystic ovary syndrome patients?

The aims of the study were to evaluate whether adiposity was among the determinants of circulating prolactin levels and to determine whether serum prolactin independently predicted metabolic abnormalities in patients with polycystic ovary syndrome (PCOS). A total of 322 PCOS patients with normal serum prolactin levels were recruited between January 2007 and January 2014. Anthropometric, metabolic, and hormonal parameters were measured in all of the patients. HOMA-IR was calculated as an index of insulin resistance. Serum prolactin was negatively correlated with age (p < 0.0001), all the adiposity indices [body mass index p < 0.0001; waist circumference p < 0.0001; waist-hip ratio (WHR) p < 0.0001], visceral adiposity index (VAI, p = 0.043), fasting insulinemia (p = 0.002), and HOMA-IR (p = 0.002), and was positively correlated with serum adiponectin (p = 0.034), but not with circulating androgens or serum leptin levels. Serum adiponectin, but not HOMA-IR or fasting insulinemia, was independently associated with serum prolactin after adjustment for age, leptin, and anthropometrical adiposity parameters. Of the adiposity parameters, only WHR and VAI were independent predictors of serum prolactin after adjustment for adiponectin. Circulating prolactin was also negatively correlated with fasting glycemia (only in patients with normal glucose metabolism, p = 0.037) and was inversely correlated with the presence of metabolic syndrome (p < 0.001), but this association was not maintained after adjustment for possible confounders. In PCOS patients, serum prolactin level was related to adipose tissue quantity and function, and adiponectin was a possible mediator of this relationship. Low serum prolactin levels were associated with an unfavorable metabolic profile, but this association seemed to be due to the complex interplay among prolactin, adiposity, and insulin resistance rather than to a direct metabolic effect of prolactin.