Resistin as a local factor in polycystic ovary syndrome: a novel view of "adipo(cyto)kines"?

We read with great interest the report by Seow et al. (2004) on circulating and adipocyte resistin in women with polycystic ovary syndrome (PCOS). In our opinion, this study addresses aspects of signi®cant methodological and, possibly, clinical importance. Seow et al. reported no difference in serum resistin levels between obese women with PCOS and normal-weight controls. No association with body mass index (BMI) was observed. In our opinion, this ®nding should be attributed to the small number of the women studied (17 women with PCOS and 10 controls) since, in a recently published study including 70 women with PCOS (35 with BMI >25 kg/m2 and 35 with BMI <25 kg/m2) and 20 healthy, normal weight controls (Panidis et al., 2004), our team has found signi®cantly higher levels of this hormone in the group of PCOS women with BMI >25 kg/ m2, compared with both groups of normal weight women. A signi®cant correlation between circulating resistin and BMI was also demonstrated. However, in our study, no difference was found in circulating resistin between lean women with PCOS and BMI-matched controls. Moreover, no BMI-independent correlation between serum resistin levels and indices of insulin resistance was observed. Therefore, we seem to agree with Seow et al. on the key conclusion that systemically acting resistin does not appear to be actively involved in the pathogenesis of PCOS. Nevertheless, Seow et al. also reported a signi®cantly higher expression of the resistin gene in omental adipocytes isolated from obese women with PCOS, compared with controls, despite the similar serum levels of this hormone in the two groups. It is well known that circulating levels do not always re ̄ect the state of tissue hormonal circuits which, nonetheless, can be of substantial physiological and clinical signi®cance. Angiotensin II is probably the best studied example (Frohlich, 2003). Therefore, the ®ndings of Seow et al. are of signi®cant medical importance, since their study implies that, in PCOS, resistin might have a local paracrine, rather than a systemic mode of action. Intriguingly, in the seminal proposal of resistin as the link between central obesity and insulin resistance (Steppan et al., 2001), the highest mRNA levels in female mice were measured in the perigonadal fat. This detail is seldom mentioned in the literature. Conclusively, the ®ndings reported by Seow et al., combined with our own recent results and previously published data, support a novel thesis: although circulating resistin does not seem to be actively involved in PCOS, it is probable that this hormone, and, possibly, other `adipo(cyto)kines' might act as local determining factors in this syndrome.