BACKGROUND: The study was conducted to investigate whether hormonal contraceptives administered via the oral and vaginal route exert a similar effect on insulin sensitivity (SI). STUDY DESIGN: This is a prospective, randomized study performed in the University Hospital. Subjects were healthy lean young women, needing a hormonal contraceptive, randomly allocated to receive for 6 months (a) an oral contraceptive (OC) containing 30 mcg ethinylestradiol (EE)/150 mcg desogestrel (DSG) (high-estrogen group; n=12), (b) an OC containing 20 mcg EE/150 mcg DSG (low-estrogen group; n=12) and (c) a vaginal ring contraceptive releasing, per day, 15 mcg EE/120 mcg etonorgestrel, the active DSG metabolite (n=12). SI and glucose utilization independent of insulin (Sg) were evaluated by the minimal model method. Modifications of total, high-density lipoprotein (HDL) and low-density lipoprotein cholesterol and triglycerides were also evaluated. RESULTS: Sg did not vary with any treatment. SI decreased during OCs (5.74+/-0.49 vs. 3.86+/-0.44; p=.0005), independently of the high/low-estrogen dose. SI did not decrease during vaginal ring use (4.64+/-1.03 vs. 5.25+/-1.36; p=.57; p=.019 vs. oral). Total cholesterol and HDL cholesterol increased (p=.02) during OCs, independently of the dose. Triglycerides increased during both oral (p=.01) and vaginal (p=.032) contraceptive use. CONCLUSIONS: The present data indicate that in contrast to OC use, vaginal contraception with the ring does not deteriorate SI. The vaginal ring may represent an appropriate choice for long-term contraception in women at risk for developing diabetes mellitus or metabolic syndrome.
