Retinol-binding protein 4 and insulin resistance in polycystic ovary syndrome

Abstract

Objective: Polycystic ovary syndrome (PCOS) is an insulin-resistant state with insulin resistance being an established therapeutic target; however, measurement of insulin resistance remains challenging. We aimed to 1) determine serum retinol-binding protein 4 (RBP4) levels (purported to reflect insulin resistance) in women with PCOS and control subjects, 2) examine the relationship of RBP4 to conventional markers of insulin resistance, and 3) examine RBP4 changes with interventions modulating insulin resistance in overweight women with PCOS.

Research design and methods: At baseline, 38 overweight women (BMI >27 kg/m(2)) with PCOS and 17 weight-matched control subjects were compared. Women with PCOS were then randomly assigned to 6 months of a higher-dose oral contraceptive pill (OCP) (35 microg ethinyl estradiol/2 mg cyproterone acetate) or metformin (1 g b.i.d.). Outcome measures were insulin resistance (total insulin area under the curve) on an oral glucose tolerance test, RBP4, and metabolic/inflammatory markers.

Results: Overweight women with PCOS were more insulin resistant than control subjects, yet RBP4 levels were not different in women with PCOS versus those in control subjects (35.4 +/- 4.3 vs. 28.9 +/- 3.1 microg/ml, P = 0.36). RBP4 correlated with cholesterol and triglycerides but not with insulin resistance. Metformin improved insulin resistance by 35%, whereas the OCP worsened insulin resistance by 33%. However, RBP4 increased nonsignificantly in both groups (43.7 +/- 6.3 vs. 42.6 +/- 5.5 microg/ml, P = 0.92).

Conclusions: Overweight women with PCOS were more insulin resistant than control subjects, but this finding was not reflected by RBP4 levels. RBP4 correlated with lipid levels but not with insulin resistance markers. RBP4 levels did not change when insulin resistance was reduced by metformin or increased by the OCP. These data suggest that RBP4 is not a useful marker of insulin resistance in PCOS but may reflect other metabolic features of this condition.

Figure 1—

Changes in RBP4 (A) and AUC insulin on an OGTT (B) after metformin (♦) or the OCP (□) for 6 months. Data are means ± SEM. Data were assessed using repeated-measures ANOVA with time as the within-subject factor and intervention as the between-subject factor. *Time-by-intervention effect (P < 0.01) with a decrease in AUC insulin for metformin (P = 0.03) and increase for OCP (P = < 0.01)