Effect of Oral Contraceptive Phase on Mechanisms of Cutaneous Microvascular Function

Abstract

While the effect of oral contraceptive (OC) use on endothelial function has been previously studied, precise mechanisms of endothelial function in this group of women remains unclear. The purpose of this study was to investigate the contribution of cyclooxygenase (COX), endothelial‐derived hyperpolarizing factors (EDHF), and nitric oxide (NO) to cutaneous microvascular function in women during the low hormone (LH) and high hormone (HH) phases of OC. Five young, healthy women taking monophasic, combined OC of any generation visited the lab twice: once during the LH (d 1‐2 of placebo pills) and once during the HH phase (d 3‐6+ of active pills). Participants were instrumented with three microdialysis fibers, and each site was randomized as control (lactated Ringer's), COX inhibition (10 mM ketorolac), or EDHF inhibition (50 mM tetraethylammonium, TEA). Laser‐Doppler flowmetry and local heaters were used to measure skin blood flow and induce local thermal hyperemia, respectively. Each site was heated from 33°C to 39°C at a rate of 0.1°C/sec. Once a plateau to local heating was established, 20 mM L‐NAME, a NO synthase inhibitor, was infused at each site to quantify NO‐dependent vasodilation. Maximal vasodilation was induced by heating the skin to 43°C and infusing 28 mM sodium nitroprusside. Data are mean ± SD. For baseline, there was a main effect of OC phase (p = 0.03) and treatment (p < 0.001) but no interaction effect (p = 0.21) such that baseline was increased at COX‐inhibited sites (23 ± 3 %CVCmax) compared with control (14 ± 6 %CVCmax; p = 0.03) during the HH phase. For initial peak, there was a main effect of treatment (p < 0.001) but no effect of OC phase (p = 0.99) or interaction effect (p = 0.59) such that initial peak was decreased at EDHF‐inhibited sites during the HH phase (17 ± 6 %CVCmax) compared with control (51 ± 23 %CVCmax; p = 0.02). For local heating plateau, there was a main effect of treatment (p < 0.001) but no effect of OC phase (p = 0.28) or interaction effect (p = 0.21) such that plateau was decreased at EDHF‐inhibited sites during both phases (LH: 18 ± 12 %CVCmax; HH: 20 ± 9 %CVCmax) compared with respective control (LH: 63 ± 13 %CVCmax; HH: 67 ± 23 %CVCmax) and ketorolac sites (LH: 58 ± 21 %CVCmax; HH: 71 ± 12 %CVCmax; all comparisons: p < 0.01). For %NO, there was no observed effect of OC phase (p = 0.77), treatment (p = 0.21), or interaction (p = 0.09). While not statistically significant, there was a large effect size for %NO between LH and HH phases at control sites (LH: 72 ± 15 %NO; HH: 50 ± 23 %NO; Cohen’s d = 1.05). During the LH phase, large effect sizes were observed between control and TEA (48 ± 14 %NO; d = 1.65) sites and control and ketorolac (54 ± 22 %NO; d = 0.96) sites. During the HH phase, a large effect size was observed between control and ketorolac (69 ± 17 %NO; d = 0.94) sites. These preliminary data suggest mechanisms underlying endothelial function vary across OC phase in women. While overall endothelial function (i.e., magnitude of local heating plateau) is similar between OC phases, there may be basal constriction (i.e., increased baseline with COX inhibition) and reduced NO‐dependent dilation during the HH phase. This shift in mechanisms underlying endothelial function in women using OC may contribute to cardiovascular health outcomes.