Original research articleEffect of four oral contraceptives on hemostatic parameters
Introduction
Several case-control and cohort studies have reported on an association between use of oral contraceptives (OCs) and the risk of venous thromboembolic events (VTEs). The decrease in the incidence of VTEs after introduction of low-dose OCs containing ≤35 μg ethinylestradiol (EE) suggested a causal connection with the dose of the estrogen component [1], [2], [3], [4], [5]. This appeared to be plausible because changes in coagulation and fibrinolysis variables were significantly lower in OCs with 30 μg EE than with 50 μg EE [6]. Therefore, development of new OC formulations was characterized by the intention to further reduce the EE dose to ≤20 μg, which had to be compensated for by more potent progestins in order to maintain the contraceptive efficacy.
Reports on a higher VTE risk during use of low-dose OCs containing the so-called third-generation progestins as compared to OCs containing levonorgestrel (LNG) or norethisterone, led to controversial discussions [5], [7], [8], [9], [10], [11]. However, subsequent findings suggested that progestins might play a role in the development of VTEs during the use of OCs, as there were differences in their effects on the intrinsic and extrinsic procoagulant activity [6], [12], [13], [14], [15], [16]. An unfavorable effect of the progestin component might prevent further reduction of risk of VTEs when using OCs with 20 μg EE. Although when taken alone progestins have no or only minor effects on hemostasis, they can modify the action of EE on certain coagulation and fibrinolysis parameters, depending on their androgenic activity [6], [12], [14]. The changes in hemostasis parameters observed during clinical studies cannot explain the association between OC use and increased risk of VTEs. In patients with risk factors, e.g., deficiencies of coagulation inhibitors or resistance to activated protein C, OCs may enhance a preexisting imbalance in the control of coagulation [17]. Such a predisposition might explain the observation that the risk is highest in the first year of OC use [5].
Natural estrogens like estradiol (E2) are believed to be associated with a lower risk of VTE than EE, because their effect on hemostasis parameters is weaker. In a small study with 21 healthy oophorectomized women, 4 developed a venous thrombosis within 80 days of treatment with 20 μg mestranol, which is rapidly converted to EE after intake [18].
Attempts to use E2 or estradiol valerate (EV) in combined OCs were not very successful, because rate of irregular bleeding was high. This has been found to be due to the progestin-induced stimulation of endometrial 17β-hydroxysteroid dehydrogenase, which causes a rapid local inactivation of E2. Contrary to this, EE cannot be metabolized by this enzyme, as the ethinyl group blocks oxidation of the 17β-hydroxy group. Therefore, it has been proposed to stabilize the cycle control of combined OCs containing E2, by the addition of 10 μg EE [19]. Indeed, treatment of fertile women with a combination of 10 μg EE + 2 mg EV and 2 mg dienogest (DNG) caused a reliable suppression of ovulation and good cycle control [20].
In the present study, the effect on various hemostasis parameters of this new formulation was compared with that of combinations of 2 mg DNG with either 20 μg or 30 μg EE and a preparation with 20 μg EE and 100 μg LNG. This trial was carried out to examine whether partial replacement of EE by a natural estrogen would result in less impact on hemostasis. Moreover, the effect of LNG, a progestin with androgenic properties, was compared with that of DNG, the only nortestosterone derivative with an antiandrogenic effect.
Section snippets
Study design
The double-blind, controlled, randomized study was carried out at two centers (Essen and Frankfurt). Included in this study were 100 healthy volunteers between 18 and 35 years of age with regular menstrual cycles and without contraindications for use of OCs. The women had not used any hormonal medication for at least 4 weeks prior to the study and did not use drugs that were known to influence the effects of OCs.
A general and gynecological examination including a Papanicolaou (Pap) smear and a
Results
Screening was carried out with 110 subjects, and 100 were randomized to receive one of four medications. Eight subjects discontinued the study prematurely, and no data from the treatment phase were available from one subject. Ninety-one subjects completed the study. Reasons for discontinuation were: bleeding abnomalities (n = 1); withdrawal of consent (n = 1); inclusion criteria not met (n = 1); did not appear for final examination (n = 1); adverse events (n = 2); pregnancy (n = 1); other (n =
Discussion
Most studies investigating the effects of OCs on hemostasis parameters showed an increase in both the procoagulant and the fibrinolytic activity, suggesting a shift of the equilibrium between coagulation and fibrinolysis to a higher level of fibrin turnover as indicated by a rise in the levels of prothrombin fragment 1+2 and D-dimer-fibrin split products. Most of the changes in hemostasis parameters were caused by the estrogen component, but only some of them, e.g., factor VII activity,
Acknowledgements
The study was financially supported by Jenapharm GmbH & Co KG, Jena, Germany). The support of Mrs. Sonja Hägele (Jenapharm GmbH & Co KG) who did the language reviewing, of Dr. Claudia Moore (Jenapharm GmbH & Co KG) who was responsible for the planning and performance of the study, and of Dr. Uwe Mellinger (Jenapharm GmbH & Co KG) who did the statistical evaluation of the results, is gratefully acknowledged.
References (68)
Oral contraceptives, clotting factors and thrombosis
Am J Obstet Gynecol
(1982)- et al.
First-time use of newer oral contraceptives and the risk of venous thromboembolism
Contraception
(1997) Newer oral contraceptives and the risk of venous thromboembolism
Contraception
(1998)Oral contraceptives and arterial and venous thrombosisa clinician's formulation
Am J Obstet Gynecol
(1998)Risk of venous thromboembolism with third-generation oral contraceptivesa review
Am J Obstet Gynecol
(1999)Effects of progestins on haemostasis
Maturitas
(1996)- et al.
Low-dose oral contraceptives and acquired resistance to activated protein Ca randomised cross-over study
Lancet
(1999) - et al.
Importance of levonorgestrel dose in oral contraceptives for effects on coagulation
Lancet
(1999) - et al.
Increased risk of venous thrombosis in oral contraceptive users who are carriers of factor V Leiden mutation
Lancet
(1994) - et al.
A randomised controlled double-blind study of the effects on hemostasis of two progestin-only pills containing 75 μg desogestrel or 30 μg levonorgestrel
Contraception
(1998)