- Email: [email protected]
Effect of two oral contraceptives with different ethinyl estradiol and levonorgestrel concentrations on the urinary excretion of biochemical vasoactive markers
Contraception 64 (2001) 357–362
Original research article
Effect of two oral contraceptives with different ethinyl estradiol and levonorgestrel concentrations on the urinary excretion of biochemical vasoactive markers Alfred O. Muecka,*, Harald Seegera, Gudula Petersenb, Ernst Schulte-Wintropb, Diethelm Wallwienera a
Section of Gynecological Endocrinology and Menopause, Department of Obstetrics and Gynecology, University of Tuebingen, Tuebingen, Germany b Wyeth-Pharma GmbH, Muenster, Germany Received 6 December 2000; received in revised form 13 September 2001; accepted 10 October 2001
Abstract In the present study the effect on the urinary excretion of vasoactive markers of two oral contraceptives (OCs), i.e., Leios, containing 0.02 mg ethinyl estradiol and 0.1 mg levonorgestrel, and Stediril 30, containing 0.03 mg ethinyl estradiol and 0.15 mg levonorgestrel, was investigated. cGMP, prostacyclin and its antagonist thromboxane, serotonin, and urodilatin, a natriuretic and diuretic peptide formed in the kidney, were measured as markers. In a comparative, double-blind, randomized, parallel group study, 34 women received Leios and 33 women Stediril 30. Nocturnal urine was collected before treatment and during cyclic treatment after 3 and 12 cycles. Both contraceptives significantly enhanced cGMP excretion after 12 cycles. The prostacyclin metabolite remained unchanged for both formulations, but the excretion of the thromboxane metabolite was significantly decreased after 12 cycles. Thus, the ratio of prostacyclin to thromboxane, crucial for the resulting effect on vascular tone, increased significantly. For the serotonin metabolite, no changes were observed for both contraceptives. The excretion of urodilatin significantly increased for both preparations after 12 cycles compared to the pretreatment values. These results indicate that the low-dose OCs Leios and Stediril 30 may stimulate the production of some vasoactive markers, at least after 12 cycles of treatment. The positive influence of these contraceptives on the various markers investigated may improve vascular tone, impede development of atherosclerosis and arterial thrombosis, and improve water and electrolyte homeostasis. These effects most likely can be attributed to the estrogenic component. Levonorgestrel may elicit no impact on these estrogen-induced changes that, however, seem only to be manifested after a longer treatment period. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Contraceptive pills; Vasoactive markers; Urinary excretion
1. Introduction The natural estrogen, 17-estradiol, elicits vasodilation in different human vessels as demonstrated by studies using noninvasive techniques such as Doppler sonography and plethysmography [1]. This effect may contribute to the reduction of morbidity and mortality of cardiovascular diseases [2]. The mechanism, however, by which estradiol exerts its vasoactive actions is not fully understood, but up-regulation of the synthesis of potent vasodilators, such as nitric oxide (NO) or prostacyclin, and concomitantly down* Corresponding author. Tel.: ⫹49-7071-29-80770; fax: ⫹49-7071-294801. E-mail address: [email protected] (A.O. Mueck).
regulation of vasoconstrictors, such as endothelin, may play a crucial role [3]. In previous studies we have demonstrated that estradiol replacement therapy in postmenopausal women can influence the urinary excretion of vasoactive markers such as cGMP and the ratio of prostacyclin to thromboxane, indicating a vasodilative effect of estradiol [4,5]. Progestin addition to estrogen replacement can lead to vasoconstriction and/or antagonize beneficial vasodilating estrogen effects, as was shown by blood flow measurements with hormone replacement therapy as well as with contraceptives [6,7]. By measuring biochemical vasoactive markers with estrogen replacement in postmenopausal women combined with sequential norethisterone acetate, we were able to show a change of the excretion of prostacyclin and
0010-7824/01/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 1 0 - 7 8 2 4 ( 0 1 ) 0 0 2 8 7 - 6
358
A.O. Mueck et al. / Contraception 64 (2001) 357–362
thromboxane metabolites, indicating a vasoconstrictory progestin action [8]. We also were able to observe effects of progestin addition on other vasoactive markers [9 –11]; however, the progestin effect on the prostacyclin to thromboxane ratio seems to be the most sensitive marker reflecting vasoactive action. As yet, little is known about the effect of the synthetic estrogen, 17␣-ethinyl estradiol, on the synthesis of these markers, especially when combined with progestins as usually used in oral contraceptives (OCs). Therefore, in the present study, the effect on the renal excretion of vasoactive markers of 2 levonorgestrel-containing OCs, i.e., Leios, containing 0.02 mg ethinyl estradiol and 0.1 mg levonorgestrel, and Stediril 30, consisting of 0.03 mg ethinyl estradiol and 0.15 mg levonorgestrel, were investigated. The following biochemical markers, which can modulate vascular tone by direct vascular action, were assessed: cGMP, which can reflect the systemic production of nitric oxide (NO) and/or atrial natriuretic peptide (ANP); prostacyclin and thromboxane; serotonin; as well as urodilatin, a diuretic and natriuretic peptide mainly produced in the kidney.
2. Patients and methods In a comparative, double-blind, randomized, parallel group study design, 34 women were treated for 12 cycles with Leios, i.e., 0.02 mg ethinyl estradiol and 0.1 mg levonorgestrel, and 33 women with Stediril 30, i.e., 0.03 mg ethinyl estradiol and 0.15 mg levonorgestrel. As usual with OC administration, one cycle is defined as 21 days with treatment, followed by a pill-free interval of 7 days. Women who had taken medication that could influence the production of the various markers under investigation were excluded from the study. Other medication was allowed to be continued during the study. The intake of drugs such as caffeine and tobacco and of foods containing serotonin, such as chocolate and bananas, which possibly can interfere with the excretion of the markers investigated, was reduced as much as possible. Also excluded were patients with neoplasias, thromboembolia, and diseases of the liver or kidneys. The urine excreted between 10 PM and 6 AM was collected before treatment (collected once during Days 17– 21) and during cyclic treatment after 3 and 12 cycles (on Days 17–21). The urine was stabilized by the addition of hydrochloric acid. Aliquots were frozen at ⫺20°C until measured. The advantages of nocturnal collection of urine are that a convenient and reliable collection of sample is possible and that definite, comparable periods of time free of stress are involved. Furthermore, urine collection is a noninvasive method that is important in this study because invasive procedures such as blood collection can influence the production of vasoactive markers.
2.1. Measurement of the markers in the urine 2.1.1. Prostacyclin/thromboxane metabolites Because prostacyclin and thromboxane are unstable compounds, the stable metabolites 2,3-dinor-6-keto-prostaglandin-F1␣ (DNPGF) and 11-dehydro-thromboxane-B2 (DHTxB), respectively, were measured. Measurement of these metabolites took place following solid phase extraction by using enzyme immunoassays (EIAs), as recommended by the assay manufacturers. The extraction was conducted with Bond-Elut-C18 columns. After pre-conditioning of the columns with water and ethanol, the samples were applied and eluted with methylformiate after washing with hexane. Recovery rate of the separation was about 65–70% as evaluated by an internal standard (3H-6-ketoprostaglandin-F1␣). The prostacyclin metabolite was measured by EIA developed by Hiramatsu et al. [12]. The interand intra-assay coefficients of variation were 10.1% and 8.9%, respectively, and the sensitivity of the assay was 10 ng/L. The accuracy of the assay was 101–112%, and crossreactivity with other important urinary prostanoids such as 6-keto-prostaglandin F1␣ was ⬍0.01%. The thromboxane metabolite was measured by EIA with the use of a commercial kit from Cayman, USA. The inter- and intra-assay coefficients of variation were 9.3% and 8.2%, respectively, and the sensitivity of the assay was 10 ng/L. The accuracy of the assay was 96 –108%; the cross-reactivity for thromboxane B2 was 0.05% and for 2,3-dinor thromboxane B2 ⬍0.01%. 2.1.2. cGMP cGMP was measured directly in the urine by using a commercial EIA kit (Cayman). The inter- and intra-assay coefficients of variation were 8.1% and 9.9%, respectively, and the sensitivity of the assay was 5 nmol/L. The accuracy of the assay was 98 –107%, the cross-reactivity for GMP, AMP, and cAMP was ⬍0.001%. 2.1.3. Serotonin Because serotonin is also unstable, its stable metabolite 5-hydroxyindole acetic acid (5-HIAA) was measured. It was measured directly in the urine by using a commercial EIA (IBL, Hamburg, Germany). The inter- and intra-assay coefficients of variation were 9.2% and 8.1%, respectively, and the sensitivity of the assay was 0.6 mmol/L. The accuracy of the assay was 99 –117%, cross-reactivity for HIAA was ⬍0.01%. 2.1.4. Urodilatin Urodilatin was measured directly in the urine by radioimmunoassay (Immundiagnostik, Bensheim, Germany). The inter-and intra-assay of variation were 9.3% and 8.9%, respectively, and the sensitivity of the assay was 0.6 mmol/L.
A.O. Mueck et al. / Contraception 64 (2001) 357–362 Table 1 Basic demographic data (mean ⫾ SD) of the women treated with oral contraceptives
Age (years) Height (cm) Weight (kg)
Leios (n ⫽ 34)
Stediril 30 (n ⫽ 33)
27.2 ⫾ 5.1 167.8 ⫾ 7.4 62.7 ⫾ 9.2
27.3 ⫾ 5.4 168.6 ⫾ 7.5 65.3 ⫾ 7.4
The accuracy of the assay was 95–107%, cross-reactivity with atrial natriuretic peptide, brain natriuretic peptide, and C-type natriuretic peptide was ⬍0.07%. 2.2. Data analyses Demographic data were compared by Student’s t test. Within-group analyses of the absolute values for the marker excretions were done by a two-sided, one-paired t test. Between-group analyses of the relative changes of the marker excretions were calculated by a two-factorial ANOVA, using the logarithmic ratios of treatment values to pretreatment values. This transformation proved to be a more reliable indication compared to the absolute values because the individual absolute values are subject to large variations for all the markers. Within-group analyses were done by a two-sided, one-paired t test for the relative logarithmic changes of the markers. The level of significance was set at ␣ ⫽ 5%. All three urine samples from each study participant were included in the same assay.
3. Results In Table 1, the basic demographic data of the patients enrolled in the study are depicted. As can be seen, no significant differences were found for age, height, and weight between both treatment groups. The pretreatment values of cGMP excretion in both study groups were comparable, the values being 0.29 mol/8 h for Leios and 0.33 mol/8 h for Stediril 30 (Table 2). After taking the contraceptive pills for 3 months, no significant changes in the absolute values were observed for
359
both regimens. After 12 months the excretion of cGMP increased significantly to 0.39 mol/8 h (p ⫽ 0.014) in the group taking Leios, whereas for Stediril 30, only a tendency to an increase was found (p ⫽ 0.11). The pretreatment prostacyclin metabolite (DNPGF) excretion values were also comparable for both study groups, 0.51 g/8 h in the group taking Leios and 0.45 g/8h in the group taking Stediril 30. These values increased to 0.57 and 0.53 g/8 h, respectively, after 3 cycles and to 0.65 and 0.71 g/8 h, respectively, after 12 cycles, but were not significantly different from the pretreatment value. The pretreatment values of the thromboxane metabolite (DHTxB) were identical (1.31 g/8 h) in both treatment groups. These values did not change significantly after 3 months of treatment in the Leios and Sediril 30 groups, the values being 1.46 and 1.28 g/8 h, respectively. After 12 cycles of treatment, however, a significant decrease of DHTxB excretion was found in both groups, with 0.47 and 0.50 g/8 h, respectively (p ⫽ 0.001). The pretreatment values of the ratio of DNPGF to DHTxB differed slightly between the groups (0.31 and 0.23 g/8 h, respectively), but this was not significant. After 3 cycles of treatment, no significant change compared to pretreatment values was found. After 12 cycles of treatment, the mean values increased significantly in the Leios group to 1.09 (p ⫽ 0.002) and in the Stediril 30 group to 0.99 (p ⬍ 0.001). The mean values before treatment for the serotonin metabolite (5-HIAA) were comparable in the Leios and Stediril 30 groups (8.15 and 9.09 mmol/8 h, respectively). After 3 cycles in both groups, a tendency to an increase was observed. After 12 cycles of treatment, a significant increase to 11.81 mmol/8 h (p ⫽ 0.04) was found in the group treated with Leios. The mean values of urodilatin were 235 ng/8 h in the group taking Leios and 208 ng/8 h in the group taking Stediril 30. After 3 cycles of treatment, there was a tendency to a decrease in both groups. After 12 cycles of treatment, urodilatin excretion was significantly enhanced to 538 ng/8 h in the Leios group (p ⫽ 0.011) and to 770 ng/8 h in the group taking Stediril 30 (p ⫽ 0.008). The percent changes of the urinary excretion of the investigated markers compared to the pretreatment values
Table 2 Absolute values in renal excretion of various markers following treatment with Leios (n ⫽ 34) or Stediril 30 (n ⫽ 33); mean values (⫾SD) Leios
Stediril 30
Marker
Pretreatment
3 cycles
cGMP (mol/8 h) Prostacyclin metabolite (g/8 h) Thromboxane metabolite (g/8 h) Ratio, prostacyclin metabolite to thromboxane metabolite Serotonin metabolite (mmol/8 h) Urodilatin (ng/8 h)
0.29 ⫾ 0.12 0.51 ⫾ 0.29 1.31 ⫾ 1.08 0.31 ⫾ 0.46
0.33 ⫾ 0.16 0.57 ⫾ 0.40 1.46 ⫾ 0.89 0.20 ⫾ 0.16
8.15 ⫾ 4.96 235 ⫾ 192
10.02 ⫾ 7.96 202 ⫾ 146
12 cycles 0.39 ⫾ 0.18* 0.65 ⫾ 0.45 0.47 ⫾ 0.56** 1.09 ⫾ 1.29** 11.81 ⫾ 8.92* 538 ⫾ 648*
* Significantly difference with respect to pretreatment values, p ⬍ 0.05; ** p ⬍ 0.01.
Pretreatment
3 cycles
12 cycles
0.33 ⫾ 0.16 0.45 ⫾ 0.31 1.31 ⫾ 1.02 0.23 ⫾ 0.18
0.32 ⫾ 0.17 0.53 ⫾ 0.39 1.28 ⫾ 1.01 0.32 ⫾ 0.50
0.40 ⫾ 0.20 0.71 ⫾ 0.74 0.50 ⫾ 0.68** 0.99 ⫾ 0.49**
9.09 ⫾ 6.18 208 ⫾ 125
9.61 ⫾ 6.84 190 ⫾ 128
9.69 ⫾ 7.75 770 ⫾ 1180**
360
A.O. Mueck et al. / Contraception 64 (2001) 357–362
Table 3 Percent changes in renal excretion of various markers following treatment with Leios (n ⫽ 34) or Stediril 30 (n ⫽ 33) in relation to the values before treatment (100%); mean values (95% CI) Marker
cGMP Prostacyclin metabolite Thromboxane metabolite Ratio, prostacyclin metabolite to thromboxane metabolite Serotonin metabolite Urodilatin
Without Leios treatment 3 cycles
Stediril 30 12 cycles
3 cycles
100 100 100 100
⫹12.6 (⫺5.0/⫹33.6) ⫹1.7 (⫺14.9/⫹21.6) ⫹19.6 (⫺0.8/⫹44.1) ⫹15.2 (⫺41.0/⫹18.4)
⫹29.9 (⫹4.4/⫹61.6)* ⫹16.7 (⫺14.1/⫹58.4) ⫺67.9 (⫺79.3/⫺53.2)** ⫹336 (⫹307/⫹706)**
⫺6.2 (⫺24.6/⫹18.0) ⫹9.6 (⫺14.2/⫹40.1) ⫺8.3 (⫺34.2/⫹27.2) ⫹0.2 (⫺20.3/⫹55.9)
100 100
⫹3.3 (⫺19.8/⫹40.9) ⫺14.4 (⫺46.4/⫹26.6)
⫹23.7 (8.5/⫹91.6) ⫺6.5 (⫺26.3/⫹34.7) ⫹119 (⫹33.7/⫹197)** ⫺23.9 (⫺43.3/⫹12.7)
12 cycles ⫹22.4 (⫹2.0/⫹46.9)* ⫹37.2 (⫺1.4/⫹91.0) ⫺69.9 (⫺79.1/⫺55.5)** ⫹415 (⫹366/⫹724)** ⫺10.7 (⫺41.3/⫹35.8) ⫹74.9 (⫹4.5/⫹192)**
* Significant difference, p ⬍ 0.05; ** p ⬍ 0.01.
are summarized in Table 3. After 12 cycles, cGMP significantly increased in both groups by 30% (Leios) and 22% (Stediril 30). The prostacyclin metabolite tended to increase with both pills after 12 cycles. For the thromboxane metabolite, a significant decrease of 68% (Leios) and 69% (Stediril 30) after 12 cycles was observed. Thus, the ratio of prostacyclin to thromboxane metabolites increased significantly in both groups. The enhancement was 336% for Leios and 415% for Stediril 30 after 12 cycles. For the serotonin metabolite, 5-HIAA, no significant changes were observed in both groups. For urodilatin, Leios showed a significant increase to 120% and Stediril 30 an increase of 75% after 12 cycles. The between-group analysis showed no statistical difference for all markers investigated.
4. Discussion In the present study, the effect on various biochemical markers has been assessed of two OCs with different ethinyl estradiol and levonorgestrel concentrations. The urinary excretion of the vasoactive marker cGMP was significantly increased by both OCs, however, only after 12 cycles of treatment. The measurement of cGMP can be adduced as a measure of NO production. NO, which is formed in the vascular endothelium, possesses strong vasodilatory and antithrombotic properties and triggers the formation of cGMP. Although this process takes place intracellularly, the measurement of urinary cGMP permits conclusions to be drawn concerning the influence of the intracellular NO/cGMP system [13–15]. The advantage of measuring cGMP over measuring the oxidative products of NO, i.e., nitrite and nitrate, may be that cGMP reflects the total production of active NO, whereas nitrite and nitrate reflect the sum of active NO and of NO that has been degraded by radical reactions, e.g., superoxide ions. Other primary messengers such as ANP are known to stimulate cGMP; however, no data that demonstrate direct hormonal effects on this messenger exist. 17-estradiol has been shown to stimulate NO synthesis [16]. As yet, no in vivo
study on the effect of ethinyl estradiol on the production of NO exists. However, in vitro, ethinyl estradiol has been shown to increase the bioavailability of NO, but not the expression of NO synthase [17]. Independently from different possible primary messengers, cGMP, stable in the systemic circulation, can elicit strong vasodilative effects. The influence of two OCs containing 0.03 mg ethinyl estradiol on cGMP excretion was investigated in a previous study comparing the progestins norethisterone acetate and dienogest [18]. Although no effect of the contraceptives on cGMP excretion was observed after 3 cycles, a tendency to an increase was observed. For the prostanoids prostacyclin and thromboxane significantly strong increases in the ratio of these vasoactive compounds were observed for both contraceptives after 12 cycles. Prostacyclin has a vasodilatory and antithrombotic effect, whereas thromboxane possesses antagonistic effects, i.e., vasoconstrictory and thrombotic properties. The ratio is most meaningful for changes of vascular tone. Because both substances are very unstable, the stable renal metabolites were measured, which reflect the production of these two parameters. The increase of the ratio by the investigated OCs was mainly achieved by a decrease of thromboxane production after 12 cycles. It remains unclear which component of the contraceptive may be responsible for this effect. During in vitro experiments, inconsistent results on the effect of ethinyl estradiol on prostanoid production were observed. In our own studies we could not find an effect of ethinyl estradiol on the synthesis of prostacyclin and thromboxane in cultures of human umbilical vein endothelial cells and leg vein endothelial cells [19], whereas David et al. observed a decrease of prostacyclin production in ATPstimulated human umbilical vein endothelial cells after the addition of ethinyl estradiol [20]. The reason for this discrepancy remains unknown, but may be attributed to cell conditions and to the use of no stimulation or of stimulation of prostacyclin synthesis, respectively. Only three clinical studies have been conducted so far investigating the influence of OCs on prostacyclin and thromboxane metabolites. Ylikorkala et al. [21] found no
A.O. Mueck et al. / Contraception 64 (2001) 357–362
changes after treatment with 0.03 mg ethinyl estradiol and 0.150 mg levonorgestrel for 3 months, but did find a decrease in serum levels of the prostacylin metabolite with no changes in the thromboxane metabolite after 2 years. The latter findings differ from the present one and may be attributed to the longer treatment time or to the measurement in different biologic fluids, i.e., serum versus urine. Stanczyk et al. [22] observed no significant changes in the urinary excretion of 6-keto-prostaglandin F1␣ and thromboxane B2 as well as in the ratio of these compounds after 3 months of treatment with norethindrone acetate combined with 20 or 35 g ethinyl estradiol. These results are coincident with the present observations. Our own published investigations showed an increase in the ratio of prostacyclin to thromboxane metabolite only for the contraceptive containing the progestin dienogest, indicating a possible negative influence of the progestin norethisterone acetate [18]. Serotonin production seems not to be influenced by both OCs. Serotonin secretion affects not only the psychological state, with the general feeling of well-being, but also the pathogenesis of cardiovascular disease by means of effects on the blood vessel system [23,24]. Serotonin causes vasodilation in human coronary arteries where the endothelium is intact and vasoconstriction where the endothelium is pathologically altered. Thus, the present results indicate that Leios and Stediril 30 may not lead to abnormal vasoconstrictions in patients with injured endothelium by up-regulation of serotonin synthesis or release. On the other hand, the psychological state may not be further improved by these OCs. In the literature, only one study was found reporting on the influence of synthetic estrogens on serotonin levels. Gonzales et al. [25] were able to demonstrate that ethinyl estradiol, like estradiol, could cause an increase in plasma serotonin values. In our previous study, no significant changes in serotonin excretion were found for both investigated contraceptives [18]. Urodilatin is a peptide mainly synthesized in the kidney [26]. Its structure is homologous to the ANP. The physiologic importance of urodilatin seems to involve electrolyte and water balance in the kidney resulting in a vasodilating effect. In a previous study we were able to show that estradiol replacement therapy in postmenopausal women can increase urodilatin production in the kidney [27]. In the present study, both OCs significantly increased the production of urodilatin, but only after 12 cycles. This may be attributed to a vasodilatory effect of the synthetic estrogen or by a counter-regulatory mechanism of the kidney caused by a vasoconstrictory action of levonorgestrel. It is known that counterbalancing actions by the kidney may need at least 2–3 months. In our previous investigation comparing norethisterone acetate with dienogest a significant increase of urodilatin excretion was found after 3 cycles with both contraceptives [18]. Thus, a possible effect of the progestogenic component may be present with respect to the influence on urodilatin excretion.
361
Almost all experimental studies using estrogens or progestins indicate dose dependency of biologic or pharmacodynamic action. With regards to possible vascular effects of ethinyl estradiol/levonorgestrel preparations, it is obvious that possible beneficial vascular effects persist when the dosage is reduced, as long as the ratio of estrogen to progestin is held constant. However, in general the intent of dosage reduction, especially for ethinyl estradiol, is to reduce the risk of venous thromboembolic events. Thus, if ethinyl estradiol is able to elicit cardiovascular protection in the arterial system, this effect may also be present in the case of dosage reduction if the progestin is reduced according to the same ratio, at least when levonorgestrel is used. It can be concluded that the OCs Leios and Stediril 30 are able to influence the renal excretion of various vasoactive markers. There seems to be no significant difference between these two preparations, which may be attributed to the same ratio of estrogen to progestin. The resulting effect of both formulations on the cardiovascular system may be improvement of vascular tone, retardation of atherogenesis and of arterial thrombosis, and a positive influence on water and electrolyte balance. However, these actions occur only after 12 cycles. The reason for this phenomenon remains unclear and should be investigated in further clinical studies.
Acknowledgments We thank Dr. Hiramatsu, Ono Pharmaceutical Co., Japan, for kindly providing reagents for measuring urinary prostacyclin metabolite.
References [1] Collins P. Vascular aspects of oestrogen. Maturitas 1996;23:217–26. [2] Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47– 63. [3] Mendelsson ME, Karas RH. The protective effects of estrogen on the cardiovascular system. New Engl J Med 1999;340:1801–11. [4] Mueck AO, Seeger H, Wiesner J, Korte K, Lippert TH. Urinary prostanoids in postmenopausal women after transdermal and oral oestrogen. J Obstet Gynaecol 1994;14:341–5. [5] Mueck AO, Seeger H, Korte K, Haasis R, Lippert TH. Is there any effect of estradiol on the vasodilating NO/cGMP-system? Investigations of the mechanism of cardiovascular protection by estrogen substitution in postmenopausal women. Zentralbl Gyna¨ kologie 1994; 116:507–11. [6] Hillard TC, Bourne TH, Whitehead MI, Crayford TB, Collins WP, Campbell S. Differential effects of transdermal estradiol and sequential progestogens on impedance to flow within the uterine arteries of postmenopausal women. Fertil Steril 1992;58:959 – 63. [7] Cagnacci A, Arangino S, Angiolucci M, Melis GB, Volpe A. Oral contraceptives and vascular reactivity of great vessels in women. Eur J Contracep Reprod Health Care 1999;4:61–5. [8] Seeger H, Mueck AO, Teichmann AT, Lippert TH. Effect of sequential estrogen/progestin-treatment on biochemical vasoactive markers
362
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16] [17]
A.O. Mueck et al. / Contraception 64 (2001) 357–362 in postmenopausal women comparing oral and transdermal application. Clin Exp Obst Gyn 2000;27:17–20. Mueck AO, Seeger H, Kaßpohl-Butz S, Teichmann AT, Lippert TH. Urinary cGMP excretion after hormone replacement therapy in postmenopausal women. Exp Clin Endocrinol Diabetes 1996;104:392–5. Mueck AO, Seeger H, Kaßpohl-Butz S, Teichmann AT, Lippert TH. Influence of norethisterone acetate and estradiol on serotonin metabolism of postmenopausal women. Horm Metab Res 1997;29:80 –3. Mueck AO, Seeger H, Armbruster FP, Lippert TH. The influence of norethisterone acetate on urinary urodilatin excretion in postmenopausal women. Clin Exp Obstet Gynecol 1998;25:76 – 8. Hiramatsu M, Hayashi Y, Yamamoto S, et al. Application of an alpha-sidechain length-specific monoclonal antibody to immunoaffinity purification and enzyme immunoassay of 2,3-dinor-6-ketoprostaglandin F1 alpha from human urine. Prostaglandins Leukot Essent Fatty Acids 1994;50:69 –73. Conrad KP, Vernier K. Plasma level, urinary excretion, and metabolic production of cGMP during gestation in rats. Am J Physiol 1989; 257:R847–R853. Broadus AE, Hardman JG, Kaminsky NI, Ball JH, Sutherland EW, Liddle GW. Extracellular cyclic nucleotides. Ann NY Acad Sci 1971;185:50 – 66. Ohashi M, Fujio N, Nawata H, Kato K, Ibayashi H. Alpha-human atrial natriuretic polypeptide-induced rise of plasma and urinary 3⬘5⬘guanosine monophosphate concentration in human subjects. Clin Exper Theory Pract 1986;A8:67–73. Kauser K, Rubanyi GM. 17-Estradiol and endothelial nitric oxide synthase. Endothelium 1994;2:203– 8. Arnal JF, Clamens S, Pechet C, et al. Ethinylestradiol does not enhance the expression of nitric oxide synthase in bovine endothelial cells but increases the release of bioactive nitric oxide by inhibiting superoxide anion production. Proc Natl Acad Sci USA 1996;93: 4108 –13.
[18] Seeger H, Lu¨ dtke R, Gra¨ ser T, Wallwiener D, Mueck AO. Effect of oral contraceptives on the urinary excretion of biochemical markers indicating vasoactive action. J Clin Pharmacy Therap 2000;25:221– 6. [19] Mu¨ ck AO, Seeger H, Korte K, Dartsch PC, Lippert TH. Natural and synthetic estrogens and prostacyclin production in human endothelial cells from umbilical cord and leg veins. Prostaglandins 1993;45:517– 25. [20] David M, Greismacher A, Mu¨ ller MM. 17alpha-Ethinylestradiol decreases production and release of prostacyclin in cultured human umbilical vein endothelial cells. Prostaglandins 1989;38:431– 8. [21] Ylikorkala O, Puolakka J, Viinikka L. The effect of oral contraceptives on antiaggregatory prostacyclin and proaggregatory thromboxane A2 in humans. Am J Obstet Gynecol 1982;142:573– 6. [22] Stanczyk FZ, Ploszaj S, Gentzschein E, Qian D, Mishell DR Jr. Effect of oral contraceptives containing 20 and 35 micrograms ethinyl estradiol on urinary prostacyclin and thromboxane metabolite levels in smokers and nonsmokers. Contraception 1999;59:17–23. [23] Golino P, Piscione F, Willerson JT, et al. Divergent effects of serotonin on coronary-artery dimensions and blood flow in patients with coronary atherosclerosis and control patients. N Engl J Med 1991; 324:641– 8. [24] McFadden EP, Clarke JG, Davies GJ, Kaski JC, Haider AW, Maseri A. Effect of intracoronary serotonin on coronary vessels in patients with stable angina and patients with variant angina. N Engl J Med 1991;324:648 –54. [25] Gonzales GF, Carrillo C. Blood serotonin levels in postmenopausal women: effects of age and serum oestradiol levels. Maturitas 1993; 17:23–9. [26] Meyer M. Urodilatin—Von Entdeckung zu klinischer Anwendung. I. Mu¨ nchen: Holzapfel Verlag, 1997. [27] Seeger H, Armbruster FP, Mueck AO, Lippert TH. The effect of estradiol on urodilatin production in postmenopausal women. Arch Gynecol Obstet 1998;262:65– 8.
Original research article
Effect of two oral contraceptives with different ethinyl estradiol and levonorgestrel concentrations on the urinary excretion of biochemical vasoactive markers Alfred O. Muecka,*, Harald Seegera, Gudula Petersenb, Ernst Schulte-Wintropb, Diethelm Wallwienera a
Section of Gynecological Endocrinology and Menopause, Department of Obstetrics and Gynecology, University of Tuebingen, Tuebingen, Germany b Wyeth-Pharma GmbH, Muenster, Germany Received 6 December 2000; received in revised form 13 September 2001; accepted 10 October 2001
Abstract In the present study the effect on the urinary excretion of vasoactive markers of two oral contraceptives (OCs), i.e., Leios, containing 0.02 mg ethinyl estradiol and 0.1 mg levonorgestrel, and Stediril 30, containing 0.03 mg ethinyl estradiol and 0.15 mg levonorgestrel, was investigated. cGMP, prostacyclin and its antagonist thromboxane, serotonin, and urodilatin, a natriuretic and diuretic peptide formed in the kidney, were measured as markers. In a comparative, double-blind, randomized, parallel group study, 34 women received Leios and 33 women Stediril 30. Nocturnal urine was collected before treatment and during cyclic treatment after 3 and 12 cycles. Both contraceptives significantly enhanced cGMP excretion after 12 cycles. The prostacyclin metabolite remained unchanged for both formulations, but the excretion of the thromboxane metabolite was significantly decreased after 12 cycles. Thus, the ratio of prostacyclin to thromboxane, crucial for the resulting effect on vascular tone, increased significantly. For the serotonin metabolite, no changes were observed for both contraceptives. The excretion of urodilatin significantly increased for both preparations after 12 cycles compared to the pretreatment values. These results indicate that the low-dose OCs Leios and Stediril 30 may stimulate the production of some vasoactive markers, at least after 12 cycles of treatment. The positive influence of these contraceptives on the various markers investigated may improve vascular tone, impede development of atherosclerosis and arterial thrombosis, and improve water and electrolyte homeostasis. These effects most likely can be attributed to the estrogenic component. Levonorgestrel may elicit no impact on these estrogen-induced changes that, however, seem only to be manifested after a longer treatment period. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Contraceptive pills; Vasoactive markers; Urinary excretion
1. Introduction The natural estrogen, 17-estradiol, elicits vasodilation in different human vessels as demonstrated by studies using noninvasive techniques such as Doppler sonography and plethysmography [1]. This effect may contribute to the reduction of morbidity and mortality of cardiovascular diseases [2]. The mechanism, however, by which estradiol exerts its vasoactive actions is not fully understood, but up-regulation of the synthesis of potent vasodilators, such as nitric oxide (NO) or prostacyclin, and concomitantly down* Corresponding author. Tel.: ⫹49-7071-29-80770; fax: ⫹49-7071-294801. E-mail address: [email protected] (A.O. Mueck).
regulation of vasoconstrictors, such as endothelin, may play a crucial role [3]. In previous studies we have demonstrated that estradiol replacement therapy in postmenopausal women can influence the urinary excretion of vasoactive markers such as cGMP and the ratio of prostacyclin to thromboxane, indicating a vasodilative effect of estradiol [4,5]. Progestin addition to estrogen replacement can lead to vasoconstriction and/or antagonize beneficial vasodilating estrogen effects, as was shown by blood flow measurements with hormone replacement therapy as well as with contraceptives [6,7]. By measuring biochemical vasoactive markers with estrogen replacement in postmenopausal women combined with sequential norethisterone acetate, we were able to show a change of the excretion of prostacyclin and
0010-7824/01/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 1 0 - 7 8 2 4 ( 0 1 ) 0 0 2 8 7 - 6
358
A.O. Mueck et al. / Contraception 64 (2001) 357–362
thromboxane metabolites, indicating a vasoconstrictory progestin action [8]. We also were able to observe effects of progestin addition on other vasoactive markers [9 –11]; however, the progestin effect on the prostacyclin to thromboxane ratio seems to be the most sensitive marker reflecting vasoactive action. As yet, little is known about the effect of the synthetic estrogen, 17␣-ethinyl estradiol, on the synthesis of these markers, especially when combined with progestins as usually used in oral contraceptives (OCs). Therefore, in the present study, the effect on the renal excretion of vasoactive markers of 2 levonorgestrel-containing OCs, i.e., Leios, containing 0.02 mg ethinyl estradiol and 0.1 mg levonorgestrel, and Stediril 30, consisting of 0.03 mg ethinyl estradiol and 0.15 mg levonorgestrel, were investigated. The following biochemical markers, which can modulate vascular tone by direct vascular action, were assessed: cGMP, which can reflect the systemic production of nitric oxide (NO) and/or atrial natriuretic peptide (ANP); prostacyclin and thromboxane; serotonin; as well as urodilatin, a diuretic and natriuretic peptide mainly produced in the kidney.
2. Patients and methods In a comparative, double-blind, randomized, parallel group study design, 34 women were treated for 12 cycles with Leios, i.e., 0.02 mg ethinyl estradiol and 0.1 mg levonorgestrel, and 33 women with Stediril 30, i.e., 0.03 mg ethinyl estradiol and 0.15 mg levonorgestrel. As usual with OC administration, one cycle is defined as 21 days with treatment, followed by a pill-free interval of 7 days. Women who had taken medication that could influence the production of the various markers under investigation were excluded from the study. Other medication was allowed to be continued during the study. The intake of drugs such as caffeine and tobacco and of foods containing serotonin, such as chocolate and bananas, which possibly can interfere with the excretion of the markers investigated, was reduced as much as possible. Also excluded were patients with neoplasias, thromboembolia, and diseases of the liver or kidneys. The urine excreted between 10 PM and 6 AM was collected before treatment (collected once during Days 17– 21) and during cyclic treatment after 3 and 12 cycles (on Days 17–21). The urine was stabilized by the addition of hydrochloric acid. Aliquots were frozen at ⫺20°C until measured. The advantages of nocturnal collection of urine are that a convenient and reliable collection of sample is possible and that definite, comparable periods of time free of stress are involved. Furthermore, urine collection is a noninvasive method that is important in this study because invasive procedures such as blood collection can influence the production of vasoactive markers.
2.1. Measurement of the markers in the urine 2.1.1. Prostacyclin/thromboxane metabolites Because prostacyclin and thromboxane are unstable compounds, the stable metabolites 2,3-dinor-6-keto-prostaglandin-F1␣ (DNPGF) and 11-dehydro-thromboxane-B2 (DHTxB), respectively, were measured. Measurement of these metabolites took place following solid phase extraction by using enzyme immunoassays (EIAs), as recommended by the assay manufacturers. The extraction was conducted with Bond-Elut-C18 columns. After pre-conditioning of the columns with water and ethanol, the samples were applied and eluted with methylformiate after washing with hexane. Recovery rate of the separation was about 65–70% as evaluated by an internal standard (3H-6-ketoprostaglandin-F1␣). The prostacyclin metabolite was measured by EIA developed by Hiramatsu et al. [12]. The interand intra-assay coefficients of variation were 10.1% and 8.9%, respectively, and the sensitivity of the assay was 10 ng/L. The accuracy of the assay was 101–112%, and crossreactivity with other important urinary prostanoids such as 6-keto-prostaglandin F1␣ was ⬍0.01%. The thromboxane metabolite was measured by EIA with the use of a commercial kit from Cayman, USA. The inter- and intra-assay coefficients of variation were 9.3% and 8.2%, respectively, and the sensitivity of the assay was 10 ng/L. The accuracy of the assay was 96 –108%; the cross-reactivity for thromboxane B2 was 0.05% and for 2,3-dinor thromboxane B2 ⬍0.01%. 2.1.2. cGMP cGMP was measured directly in the urine by using a commercial EIA kit (Cayman). The inter- and intra-assay coefficients of variation were 8.1% and 9.9%, respectively, and the sensitivity of the assay was 5 nmol/L. The accuracy of the assay was 98 –107%, the cross-reactivity for GMP, AMP, and cAMP was ⬍0.001%. 2.1.3. Serotonin Because serotonin is also unstable, its stable metabolite 5-hydroxyindole acetic acid (5-HIAA) was measured. It was measured directly in the urine by using a commercial EIA (IBL, Hamburg, Germany). The inter- and intra-assay coefficients of variation were 9.2% and 8.1%, respectively, and the sensitivity of the assay was 0.6 mmol/L. The accuracy of the assay was 99 –117%, cross-reactivity for HIAA was ⬍0.01%. 2.1.4. Urodilatin Urodilatin was measured directly in the urine by radioimmunoassay (Immundiagnostik, Bensheim, Germany). The inter-and intra-assay of variation were 9.3% and 8.9%, respectively, and the sensitivity of the assay was 0.6 mmol/L.
A.O. Mueck et al. / Contraception 64 (2001) 357–362 Table 1 Basic demographic data (mean ⫾ SD) of the women treated with oral contraceptives
Age (years) Height (cm) Weight (kg)
Leios (n ⫽ 34)
Stediril 30 (n ⫽ 33)
27.2 ⫾ 5.1 167.8 ⫾ 7.4 62.7 ⫾ 9.2
27.3 ⫾ 5.4 168.6 ⫾ 7.5 65.3 ⫾ 7.4
The accuracy of the assay was 95–107%, cross-reactivity with atrial natriuretic peptide, brain natriuretic peptide, and C-type natriuretic peptide was ⬍0.07%. 2.2. Data analyses Demographic data were compared by Student’s t test. Within-group analyses of the absolute values for the marker excretions were done by a two-sided, one-paired t test. Between-group analyses of the relative changes of the marker excretions were calculated by a two-factorial ANOVA, using the logarithmic ratios of treatment values to pretreatment values. This transformation proved to be a more reliable indication compared to the absolute values because the individual absolute values are subject to large variations for all the markers. Within-group analyses were done by a two-sided, one-paired t test for the relative logarithmic changes of the markers. The level of significance was set at ␣ ⫽ 5%. All three urine samples from each study participant were included in the same assay.
3. Results In Table 1, the basic demographic data of the patients enrolled in the study are depicted. As can be seen, no significant differences were found for age, height, and weight between both treatment groups. The pretreatment values of cGMP excretion in both study groups were comparable, the values being 0.29 mol/8 h for Leios and 0.33 mol/8 h for Stediril 30 (Table 2). After taking the contraceptive pills for 3 months, no significant changes in the absolute values were observed for
359
both regimens. After 12 months the excretion of cGMP increased significantly to 0.39 mol/8 h (p ⫽ 0.014) in the group taking Leios, whereas for Stediril 30, only a tendency to an increase was found (p ⫽ 0.11). The pretreatment prostacyclin metabolite (DNPGF) excretion values were also comparable for both study groups, 0.51 g/8 h in the group taking Leios and 0.45 g/8h in the group taking Stediril 30. These values increased to 0.57 and 0.53 g/8 h, respectively, after 3 cycles and to 0.65 and 0.71 g/8 h, respectively, after 12 cycles, but were not significantly different from the pretreatment value. The pretreatment values of the thromboxane metabolite (DHTxB) were identical (1.31 g/8 h) in both treatment groups. These values did not change significantly after 3 months of treatment in the Leios and Sediril 30 groups, the values being 1.46 and 1.28 g/8 h, respectively. After 12 cycles of treatment, however, a significant decrease of DHTxB excretion was found in both groups, with 0.47 and 0.50 g/8 h, respectively (p ⫽ 0.001). The pretreatment values of the ratio of DNPGF to DHTxB differed slightly between the groups (0.31 and 0.23 g/8 h, respectively), but this was not significant. After 3 cycles of treatment, no significant change compared to pretreatment values was found. After 12 cycles of treatment, the mean values increased significantly in the Leios group to 1.09 (p ⫽ 0.002) and in the Stediril 30 group to 0.99 (p ⬍ 0.001). The mean values before treatment for the serotonin metabolite (5-HIAA) were comparable in the Leios and Stediril 30 groups (8.15 and 9.09 mmol/8 h, respectively). After 3 cycles in both groups, a tendency to an increase was observed. After 12 cycles of treatment, a significant increase to 11.81 mmol/8 h (p ⫽ 0.04) was found in the group treated with Leios. The mean values of urodilatin were 235 ng/8 h in the group taking Leios and 208 ng/8 h in the group taking Stediril 30. After 3 cycles of treatment, there was a tendency to a decrease in both groups. After 12 cycles of treatment, urodilatin excretion was significantly enhanced to 538 ng/8 h in the Leios group (p ⫽ 0.011) and to 770 ng/8 h in the group taking Stediril 30 (p ⫽ 0.008). The percent changes of the urinary excretion of the investigated markers compared to the pretreatment values
Table 2 Absolute values in renal excretion of various markers following treatment with Leios (n ⫽ 34) or Stediril 30 (n ⫽ 33); mean values (⫾SD) Leios
Stediril 30
Marker
Pretreatment
3 cycles
cGMP (mol/8 h) Prostacyclin metabolite (g/8 h) Thromboxane metabolite (g/8 h) Ratio, prostacyclin metabolite to thromboxane metabolite Serotonin metabolite (mmol/8 h) Urodilatin (ng/8 h)
0.29 ⫾ 0.12 0.51 ⫾ 0.29 1.31 ⫾ 1.08 0.31 ⫾ 0.46
0.33 ⫾ 0.16 0.57 ⫾ 0.40 1.46 ⫾ 0.89 0.20 ⫾ 0.16
8.15 ⫾ 4.96 235 ⫾ 192
10.02 ⫾ 7.96 202 ⫾ 146
12 cycles 0.39 ⫾ 0.18* 0.65 ⫾ 0.45 0.47 ⫾ 0.56** 1.09 ⫾ 1.29** 11.81 ⫾ 8.92* 538 ⫾ 648*
* Significantly difference with respect to pretreatment values, p ⬍ 0.05; ** p ⬍ 0.01.
Pretreatment
3 cycles
12 cycles
0.33 ⫾ 0.16 0.45 ⫾ 0.31 1.31 ⫾ 1.02 0.23 ⫾ 0.18
0.32 ⫾ 0.17 0.53 ⫾ 0.39 1.28 ⫾ 1.01 0.32 ⫾ 0.50
0.40 ⫾ 0.20 0.71 ⫾ 0.74 0.50 ⫾ 0.68** 0.99 ⫾ 0.49**
9.09 ⫾ 6.18 208 ⫾ 125
9.61 ⫾ 6.84 190 ⫾ 128
9.69 ⫾ 7.75 770 ⫾ 1180**
360
A.O. Mueck et al. / Contraception 64 (2001) 357–362
Table 3 Percent changes in renal excretion of various markers following treatment with Leios (n ⫽ 34) or Stediril 30 (n ⫽ 33) in relation to the values before treatment (100%); mean values (95% CI) Marker
cGMP Prostacyclin metabolite Thromboxane metabolite Ratio, prostacyclin metabolite to thromboxane metabolite Serotonin metabolite Urodilatin
Without Leios treatment 3 cycles
Stediril 30 12 cycles
3 cycles
100 100 100 100
⫹12.6 (⫺5.0/⫹33.6) ⫹1.7 (⫺14.9/⫹21.6) ⫹19.6 (⫺0.8/⫹44.1) ⫹15.2 (⫺41.0/⫹18.4)
⫹29.9 (⫹4.4/⫹61.6)* ⫹16.7 (⫺14.1/⫹58.4) ⫺67.9 (⫺79.3/⫺53.2)** ⫹336 (⫹307/⫹706)**
⫺6.2 (⫺24.6/⫹18.0) ⫹9.6 (⫺14.2/⫹40.1) ⫺8.3 (⫺34.2/⫹27.2) ⫹0.2 (⫺20.3/⫹55.9)
100 100
⫹3.3 (⫺19.8/⫹40.9) ⫺14.4 (⫺46.4/⫹26.6)
⫹23.7 (8.5/⫹91.6) ⫺6.5 (⫺26.3/⫹34.7) ⫹119 (⫹33.7/⫹197)** ⫺23.9 (⫺43.3/⫹12.7)
12 cycles ⫹22.4 (⫹2.0/⫹46.9)* ⫹37.2 (⫺1.4/⫹91.0) ⫺69.9 (⫺79.1/⫺55.5)** ⫹415 (⫹366/⫹724)** ⫺10.7 (⫺41.3/⫹35.8) ⫹74.9 (⫹4.5/⫹192)**
* Significant difference, p ⬍ 0.05; ** p ⬍ 0.01.
are summarized in Table 3. After 12 cycles, cGMP significantly increased in both groups by 30% (Leios) and 22% (Stediril 30). The prostacyclin metabolite tended to increase with both pills after 12 cycles. For the thromboxane metabolite, a significant decrease of 68% (Leios) and 69% (Stediril 30) after 12 cycles was observed. Thus, the ratio of prostacyclin to thromboxane metabolites increased significantly in both groups. The enhancement was 336% for Leios and 415% for Stediril 30 after 12 cycles. For the serotonin metabolite, 5-HIAA, no significant changes were observed in both groups. For urodilatin, Leios showed a significant increase to 120% and Stediril 30 an increase of 75% after 12 cycles. The between-group analysis showed no statistical difference for all markers investigated.
4. Discussion In the present study, the effect on various biochemical markers has been assessed of two OCs with different ethinyl estradiol and levonorgestrel concentrations. The urinary excretion of the vasoactive marker cGMP was significantly increased by both OCs, however, only after 12 cycles of treatment. The measurement of cGMP can be adduced as a measure of NO production. NO, which is formed in the vascular endothelium, possesses strong vasodilatory and antithrombotic properties and triggers the formation of cGMP. Although this process takes place intracellularly, the measurement of urinary cGMP permits conclusions to be drawn concerning the influence of the intracellular NO/cGMP system [13–15]. The advantage of measuring cGMP over measuring the oxidative products of NO, i.e., nitrite and nitrate, may be that cGMP reflects the total production of active NO, whereas nitrite and nitrate reflect the sum of active NO and of NO that has been degraded by radical reactions, e.g., superoxide ions. Other primary messengers such as ANP are known to stimulate cGMP; however, no data that demonstrate direct hormonal effects on this messenger exist. 17-estradiol has been shown to stimulate NO synthesis [16]. As yet, no in vivo
study on the effect of ethinyl estradiol on the production of NO exists. However, in vitro, ethinyl estradiol has been shown to increase the bioavailability of NO, but not the expression of NO synthase [17]. Independently from different possible primary messengers, cGMP, stable in the systemic circulation, can elicit strong vasodilative effects. The influence of two OCs containing 0.03 mg ethinyl estradiol on cGMP excretion was investigated in a previous study comparing the progestins norethisterone acetate and dienogest [18]. Although no effect of the contraceptives on cGMP excretion was observed after 3 cycles, a tendency to an increase was observed. For the prostanoids prostacyclin and thromboxane significantly strong increases in the ratio of these vasoactive compounds were observed for both contraceptives after 12 cycles. Prostacyclin has a vasodilatory and antithrombotic effect, whereas thromboxane possesses antagonistic effects, i.e., vasoconstrictory and thrombotic properties. The ratio is most meaningful for changes of vascular tone. Because both substances are very unstable, the stable renal metabolites were measured, which reflect the production of these two parameters. The increase of the ratio by the investigated OCs was mainly achieved by a decrease of thromboxane production after 12 cycles. It remains unclear which component of the contraceptive may be responsible for this effect. During in vitro experiments, inconsistent results on the effect of ethinyl estradiol on prostanoid production were observed. In our own studies we could not find an effect of ethinyl estradiol on the synthesis of prostacyclin and thromboxane in cultures of human umbilical vein endothelial cells and leg vein endothelial cells [19], whereas David et al. observed a decrease of prostacyclin production in ATPstimulated human umbilical vein endothelial cells after the addition of ethinyl estradiol [20]. The reason for this discrepancy remains unknown, but may be attributed to cell conditions and to the use of no stimulation or of stimulation of prostacyclin synthesis, respectively. Only three clinical studies have been conducted so far investigating the influence of OCs on prostacyclin and thromboxane metabolites. Ylikorkala et al. [21] found no
A.O. Mueck et al. / Contraception 64 (2001) 357–362
changes after treatment with 0.03 mg ethinyl estradiol and 0.150 mg levonorgestrel for 3 months, but did find a decrease in serum levels of the prostacylin metabolite with no changes in the thromboxane metabolite after 2 years. The latter findings differ from the present one and may be attributed to the longer treatment time or to the measurement in different biologic fluids, i.e., serum versus urine. Stanczyk et al. [22] observed no significant changes in the urinary excretion of 6-keto-prostaglandin F1␣ and thromboxane B2 as well as in the ratio of these compounds after 3 months of treatment with norethindrone acetate combined with 20 or 35 g ethinyl estradiol. These results are coincident with the present observations. Our own published investigations showed an increase in the ratio of prostacyclin to thromboxane metabolite only for the contraceptive containing the progestin dienogest, indicating a possible negative influence of the progestin norethisterone acetate [18]. Serotonin production seems not to be influenced by both OCs. Serotonin secretion affects not only the psychological state, with the general feeling of well-being, but also the pathogenesis of cardiovascular disease by means of effects on the blood vessel system [23,24]. Serotonin causes vasodilation in human coronary arteries where the endothelium is intact and vasoconstriction where the endothelium is pathologically altered. Thus, the present results indicate that Leios and Stediril 30 may not lead to abnormal vasoconstrictions in patients with injured endothelium by up-regulation of serotonin synthesis or release. On the other hand, the psychological state may not be further improved by these OCs. In the literature, only one study was found reporting on the influence of synthetic estrogens on serotonin levels. Gonzales et al. [25] were able to demonstrate that ethinyl estradiol, like estradiol, could cause an increase in plasma serotonin values. In our previous study, no significant changes in serotonin excretion were found for both investigated contraceptives [18]. Urodilatin is a peptide mainly synthesized in the kidney [26]. Its structure is homologous to the ANP. The physiologic importance of urodilatin seems to involve electrolyte and water balance in the kidney resulting in a vasodilating effect. In a previous study we were able to show that estradiol replacement therapy in postmenopausal women can increase urodilatin production in the kidney [27]. In the present study, both OCs significantly increased the production of urodilatin, but only after 12 cycles. This may be attributed to a vasodilatory effect of the synthetic estrogen or by a counter-regulatory mechanism of the kidney caused by a vasoconstrictory action of levonorgestrel. It is known that counterbalancing actions by the kidney may need at least 2–3 months. In our previous investigation comparing norethisterone acetate with dienogest a significant increase of urodilatin excretion was found after 3 cycles with both contraceptives [18]. Thus, a possible effect of the progestogenic component may be present with respect to the influence on urodilatin excretion.
361
Almost all experimental studies using estrogens or progestins indicate dose dependency of biologic or pharmacodynamic action. With regards to possible vascular effects of ethinyl estradiol/levonorgestrel preparations, it is obvious that possible beneficial vascular effects persist when the dosage is reduced, as long as the ratio of estrogen to progestin is held constant. However, in general the intent of dosage reduction, especially for ethinyl estradiol, is to reduce the risk of venous thromboembolic events. Thus, if ethinyl estradiol is able to elicit cardiovascular protection in the arterial system, this effect may also be present in the case of dosage reduction if the progestin is reduced according to the same ratio, at least when levonorgestrel is used. It can be concluded that the OCs Leios and Stediril 30 are able to influence the renal excretion of various vasoactive markers. There seems to be no significant difference between these two preparations, which may be attributed to the same ratio of estrogen to progestin. The resulting effect of both formulations on the cardiovascular system may be improvement of vascular tone, retardation of atherogenesis and of arterial thrombosis, and a positive influence on water and electrolyte balance. However, these actions occur only after 12 cycles. The reason for this phenomenon remains unclear and should be investigated in further clinical studies.
Acknowledgments We thank Dr. Hiramatsu, Ono Pharmaceutical Co., Japan, for kindly providing reagents for measuring urinary prostacyclin metabolite.
References [1] Collins P. Vascular aspects of oestrogen. Maturitas 1996;23:217–26. [2] Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47– 63. [3] Mendelsson ME, Karas RH. The protective effects of estrogen on the cardiovascular system. New Engl J Med 1999;340:1801–11. [4] Mueck AO, Seeger H, Wiesner J, Korte K, Lippert TH. Urinary prostanoids in postmenopausal women after transdermal and oral oestrogen. J Obstet Gynaecol 1994;14:341–5. [5] Mueck AO, Seeger H, Korte K, Haasis R, Lippert TH. Is there any effect of estradiol on the vasodilating NO/cGMP-system? Investigations of the mechanism of cardiovascular protection by estrogen substitution in postmenopausal women. Zentralbl Gyna¨ kologie 1994; 116:507–11. [6] Hillard TC, Bourne TH, Whitehead MI, Crayford TB, Collins WP, Campbell S. Differential effects of transdermal estradiol and sequential progestogens on impedance to flow within the uterine arteries of postmenopausal women. Fertil Steril 1992;58:959 – 63. [7] Cagnacci A, Arangino S, Angiolucci M, Melis GB, Volpe A. Oral contraceptives and vascular reactivity of great vessels in women. Eur J Contracep Reprod Health Care 1999;4:61–5. [8] Seeger H, Mueck AO, Teichmann AT, Lippert TH. Effect of sequential estrogen/progestin-treatment on biochemical vasoactive markers
362
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16] [17]
A.O. Mueck et al. / Contraception 64 (2001) 357–362 in postmenopausal women comparing oral and transdermal application. Clin Exp Obst Gyn 2000;27:17–20. Mueck AO, Seeger H, Kaßpohl-Butz S, Teichmann AT, Lippert TH. Urinary cGMP excretion after hormone replacement therapy in postmenopausal women. Exp Clin Endocrinol Diabetes 1996;104:392–5. Mueck AO, Seeger H, Kaßpohl-Butz S, Teichmann AT, Lippert TH. Influence of norethisterone acetate and estradiol on serotonin metabolism of postmenopausal women. Horm Metab Res 1997;29:80 –3. Mueck AO, Seeger H, Armbruster FP, Lippert TH. The influence of norethisterone acetate on urinary urodilatin excretion in postmenopausal women. Clin Exp Obstet Gynecol 1998;25:76 – 8. Hiramatsu M, Hayashi Y, Yamamoto S, et al. Application of an alpha-sidechain length-specific monoclonal antibody to immunoaffinity purification and enzyme immunoassay of 2,3-dinor-6-ketoprostaglandin F1 alpha from human urine. Prostaglandins Leukot Essent Fatty Acids 1994;50:69 –73. Conrad KP, Vernier K. Plasma level, urinary excretion, and metabolic production of cGMP during gestation in rats. Am J Physiol 1989; 257:R847–R853. Broadus AE, Hardman JG, Kaminsky NI, Ball JH, Sutherland EW, Liddle GW. Extracellular cyclic nucleotides. Ann NY Acad Sci 1971;185:50 – 66. Ohashi M, Fujio N, Nawata H, Kato K, Ibayashi H. Alpha-human atrial natriuretic polypeptide-induced rise of plasma and urinary 3⬘5⬘guanosine monophosphate concentration in human subjects. Clin Exper Theory Pract 1986;A8:67–73. Kauser K, Rubanyi GM. 17-Estradiol and endothelial nitric oxide synthase. Endothelium 1994;2:203– 8. Arnal JF, Clamens S, Pechet C, et al. Ethinylestradiol does not enhance the expression of nitric oxide synthase in bovine endothelial cells but increases the release of bioactive nitric oxide by inhibiting superoxide anion production. Proc Natl Acad Sci USA 1996;93: 4108 –13.
[18] Seeger H, Lu¨ dtke R, Gra¨ ser T, Wallwiener D, Mueck AO. Effect of oral contraceptives on the urinary excretion of biochemical markers indicating vasoactive action. J Clin Pharmacy Therap 2000;25:221– 6. [19] Mu¨ ck AO, Seeger H, Korte K, Dartsch PC, Lippert TH. Natural and synthetic estrogens and prostacyclin production in human endothelial cells from umbilical cord and leg veins. Prostaglandins 1993;45:517– 25. [20] David M, Greismacher A, Mu¨ ller MM. 17alpha-Ethinylestradiol decreases production and release of prostacyclin in cultured human umbilical vein endothelial cells. Prostaglandins 1989;38:431– 8. [21] Ylikorkala O, Puolakka J, Viinikka L. The effect of oral contraceptives on antiaggregatory prostacyclin and proaggregatory thromboxane A2 in humans. Am J Obstet Gynecol 1982;142:573– 6. [22] Stanczyk FZ, Ploszaj S, Gentzschein E, Qian D, Mishell DR Jr. Effect of oral contraceptives containing 20 and 35 micrograms ethinyl estradiol on urinary prostacyclin and thromboxane metabolite levels in smokers and nonsmokers. Contraception 1999;59:17–23. [23] Golino P, Piscione F, Willerson JT, et al. Divergent effects of serotonin on coronary-artery dimensions and blood flow in patients with coronary atherosclerosis and control patients. N Engl J Med 1991; 324:641– 8. [24] McFadden EP, Clarke JG, Davies GJ, Kaski JC, Haider AW, Maseri A. Effect of intracoronary serotonin on coronary vessels in patients with stable angina and patients with variant angina. N Engl J Med 1991;324:648 –54. [25] Gonzales GF, Carrillo C. Blood serotonin levels in postmenopausal women: effects of age and serum oestradiol levels. Maturitas 1993; 17:23–9. [26] Meyer M. Urodilatin—Von Entdeckung zu klinischer Anwendung. I. Mu¨ nchen: Holzapfel Verlag, 1997. [27] Seeger H, Armbruster FP, Mueck AO, Lippert TH. The effect of estradiol on urodilatin production in postmenopausal women. Arch Gynecol Obstet 1998;262:65– 8.