Oral Versus Patch Hormonal Contraceptive Effects on Metabolism, Clotting, Inflammatory Factors and Vascular Reactivity

Overview

The purpose of this study is to compare the effects of oral versus patch administration of hormonal contraception on hormone sensitive proteins such as lipoproteins, clotting factors and inflammatory proteins as well as blood sugar and insulin levels, antioxidant status and flow-mediated dilation of arm and forearm vessels. The hypothesis is that oral administration of contraceptive hormones will result in higher plasma levels of estrogen sensitive proteins originating from the liver while patch administration of contraceptive hormones will result in greater systemic effects of estrogen on vascular reactivity and antioxidant status.

Study Rationale: The metabolic effects of hormonal contraceptives differ when administered by oral or systemic routes. The oral route magnifies the hepatic production of hormone sensitive proteins including lipoproteins, clotting factors and inflammatory proteins such as CRP. These effects are due to the first pass of hormone to the liver from the portal circulation. The first pass hepatic effect also reduces the systemic exposure to orally administered estrogen and progestin, due to glucuronidation and sulfation of sex hormones and excretion in the bile. Systemic administration by the patch is more analogous to physiologic hormone release from the ovary, as it avoids the first pass effect and the peripheral tissues are exposed to higher hormone concentrations than with oral administration. Conversely, the liver may have a lower exposure to hormone by patch administration than by oral administration. With respect to the extended cycle (2-month) patch contraception formulation under study, similar questions may be asked about the metabolic, inflammatory and vascular effects of this regimen compared to the one-month cycles. The purpose of this research is: 1. to measure the plasma estrogen and progestin levels observed with each formulation and the physiological parameters that affect vascular health including lipids, glucose and insulin, redox state, clotting and inflammatory factors, and vascular reactivity; and 2. to study the relationships of hormones to physiological parameters among the three regimens. The underlying rationale is that the net effect of estrogen-progestin combinations is a function of the dose and biological effect of each hormone and that these effects differ by route of administration. Study Design: Three contraception regimens will be compared in an open label, randomized, crossover design. The three regimens are Ortho-Cyclen® as the standard reference oral contraceptive, Ortho Evra® patch formulation, and an investigational two month patch formulation. The two month patch formulation will consist of applying Ortho Evra® for 7 weeks in a row with the 8th week off compared to 3 weeks of patch application with the 4th week off in the standard patch cycle. Subjects will take each hormonal regimen for two months. Eligible subjects will have an initial one-month run-in period on Ortho-Cyclen®. In addition to washing out any previous hormone exposure, this washout will serve as a compliance hurdle that will help in selecting subjects that are willing to undergo the discipline of the study. Washouts between treatments will not be performed in order to provide continuous contraception for the study subjects and facilitate compliance once subjects are randomized into the study. As each treatment period will last two months, the first month of each two-month treatment period will be considered as the washout from the previous treatment protocol. The second month of each two-month treatment period will be the investigative month. Plasma hormones monitored in the second month of each two month period are ethinyl estradiol and norelgestromin. Norelgestromin is the primary active metabolite of norgestimate metabolism. Clotting parameters to be measured are those monitored in our previous study of a desogestrel (D) vs. levonorgestrel (L) contraceptive comparison at nearly equal estrogen exposures over 9 months. In this study, PT and PTT increased equally from a pre-treatment baseline, factor V decreased on (D) and increased on (L), and free protein S decreased on (D) and did not change with (L). The impression from this study was that the changes were internally compensating. PAI-1, an inhibitor of plasminogen activation, will also be measured. These will be measured at 1, 7, 21 and 28 days of the investigative month. Inflammatory parameters will include highly sensitive C-reactive protein (hsCRP), which is of hepatic origin, and serum amyloid A (SAA), which varies in parallel to CRP in the metabolic syndrome but is transported almost entirely in HDL. SAA transport in HDL is considered to be a barometer of impaired HDL function, which occurs in inflammatory states. These measurements will be obtained at days 1 and 21. Antioxidant status is related to inflammatory and metabolic stress and is of additional interest in light of a current NIH NICHD study to examine the changes in oxidant stress during the menstrual cycle. The primary parameter we will use is total antioxidant capacity (TOAC). Several other parameters of plasma redox status can be measured if interesting trends are detected. These parameters will also be measured at days 1 and 21 of the respective cycles. We have previously described the antioxidant effects of estrogen and prooxidant effects of progestins and differences among the progestins in in vitro systems. The lipoprotein measures selected are the standard lipid profile consisting of triglyceride, cholesterol, HDL cholesterol measured by precipitation and estimation of LDL by the Friedwald algorithm. The HDL2 and HDL3 sub-fractions and apoproteins AI and AII are sensitive measures of estrogen-progestin balance, as is sex-hormone binding globulin (SHBG). Lipoprotein lipids, HDL sub-fractions and SHBG will be measured at 1, 7, 21 and 28 days. The apoproteins will be measured at days 1 and 21. We will measure vascular reactivity after arterial ischemia in two ways, by brachial artery reactivity and by venous plethysmography. Both techniques are sensitive to nitric oxide mediated vascular dilation. Both procedures will be done without the nitroglycerin step. Differences in flow mediated dilation and arterial compliance have been described in the normal menstrual cycle, and should be sensitive to differing biologic effects of estrogen and progestin in the respective contraceptive cycles. We expect that the vasodilatory response to ischemia will represent the net opposing biologic effects of estrogen and progestin, as do the hormone-sensitive plasma proteins. These measurements will be made at day 21 of the second month of each of the three treatment arms, as day 21 is the time of maximum hormone exposure in all three regimens. The subject's overall satisfaction with each hormonal regimen will be quantified in a daily diary along with a record of menstrual history.