Contraception and cardiovascular disorders.

PIP: Prescribing contraceptives for women with underlying medical conditions requires careful attention from practitioners. This article reviews current knowledge on the metabolic effects and cardiovascular risks associated with use of combined oral contraceptives (OCs). OCs exert effects on lipids, high- and low-density lipid cholesterol, serum triglycerides, hemostasis, insulin resistance and hyperinsulinemia, and hypertension, all of which may have implications for ischemic heart disease, cerebrovascular accidents, and venous thromboembolism. Also discussed are alternative contraceptive methods for women with contraindications to OC use. Preconception counseling is especially important to provide women with information on the likely impact of their disease on pregnancy outcome and of pregnancy on their disease.^ieng

A randomized cross-over study on various hormonal parameters of two triphasic oral contraceptives.

The effect of two triphasic oral contraceptives (Triquilar [TRQ] and Trisiston [TRS]) containing ethinyl estradiol (EE) and levonorgestrel (LNG) on various hormonal parameters was investigated in 26 women during a cross-over study. TRS consisted of 0.03 mg EE + 0.05 mg LNG (six tablets), 0.04 mg EE + 0.075 mg LNG (six tablets), and 0.03 mg EE + 0.15 mg LNG (nine tablets), whereas TRQ was different in the second phase (five tablets) and third phase (10 tablets). Blood samples were taken on days 6, 11, 21, and 28 of the control and washout cycles and the third treatment cycle. Both formulations inhibited ovulation reliably and decreased the serum levels of gonadotropins, free testosterone, and dehydroepiandosterone sulfate in a time-dependent manner, whereas estradiol and testosterone were already suppressed on day 6, indicating a direct suppressive effect on ovarian steroid synthesis. Prolactin, which rose sporadically in some women, was not significantly changed. In contrast, the levels of sex hormone binding globulin, corticosteroid binding globulin, and cortisol were significantly elevated by 100%. During the hormone-free interval of 7 days, all parameters returned at least partly to baseline. There was no significant difference between the effects of both formulations. The results suggest the possibility of a direct inhibitory effect of contraceptive steroids on ovarian steroid synthesis.

PIP: A randomized crossover study involving 26 women in Germany investigated the effect of two triphasic oral contraceptives (OCs) on selected hormonal parameters. The first triphasic, Trisiston, contained 0.03 mg of ethinyl estradiol (EE) and 0.05 mg of levonorgestrel (LNG) (6 tablets), 0.04 mg EE and 0.075 mg LNG (6 tablets), and 0.03 mg EE and 0.15 mg LNG (9 tablets). The second, Triquilar, differed from the first in the second (5 tablets) and third (10 tablets) phases. Serum samples were collected on days 6, 11, 21, and 28 of the control and washout cycles and the third treatment cycle. There were no significant differences in the hormonal effects of the two formulations. Both triphasics inhibited ovulation reliably and decreased serum levels of gonadotropins, free testosterone, and dehydroepiandrosterone sulfate in a time-dependent manner. Estradiol and testosterone were already suppressed on day 6. Prolactin rose sporadically in some women, but was not significantly changed. In contrast, levels of sex hormone binding globulin, corticosteroid binding globulin, and cortisol were significantly elevated by 100%. During the 7-day hormone-free interval, all parameters returned at least partly to baseline. These findings suggest a direct inhibitory effect of OC steroids on ovarian steroid synthesis.

Effect of two oral contraceptives containing ethinyl estradiol and gestodene or norgestimate on different lipid and lipoprotein parameters.

The effect of a triphasic oral contraceptive containing ethinyl estradiol and gestodene (EE/GSD) on various lipid and lipoprotein parameters was compared with that of a monophasic formulation containing 35 micrograms ethinyl estradiol and 250 micrograms norgestimate (EE/NGM). Blood samples were collected from 46 women on days 2, 11, and 21 of the preceding control cycle and of the third, sixth, and twelfth treatment cycles. There was no significant difference between formulations with regard to the influence on any measured parameter. As compared with controls, a significant increase was observed in the plasma levels of total triglycerides (24-78%), total phospholipids (7-20%), very low density lipoprotein (VLDL) triglycerides (61-76%), VLDL-phospholipids (14-60%), low density lipoprotein (LDL) triglycerides (8-35%), LDL-phospholipids (28-30%), high density lipoprotein (HDL) cholesterol (8-16%), HDL 3-cholesterol (11-20%), HDL-triglycerides (17-66%), HDL-phospholipids, HDL 3-phospholipids (7-11%), apolipoprotein (apo) A-I (5-20%) and apo A-II (10-40%) during treatment with both formulations. In contrast, the LDL-cholesterol levels were significantly decreased. These changes in lipid metabolism appear to reflect a predominance of the effect of the estrogen component. The results indicate that both low dose oral contraceptives containing different progestins and different amounts of EE do not exert a deleterious effect on lipoprotein metabolism, as high HDL-cholesterol and low LDL-cholesterol levels are known as low risk factors of cardiovascular disease. In contrast to endogenous hypertriglyceridemia, an EE-induced rise in triglyceride levels does not appear to increase cardiovascular risk if LDL is not increased.

PIP: Oral contraceptives (OCs) that contain a progestogen with high androgenic activity have been shown to have an atherogenic effect on lipid and lipoprotein metabolism. The present study compared the effect of a triphasic OC containing ethinyl estradiol and gestodene on selected lipid and lipoprotein parameters with that of a monophasic OC containing 35 mcg of ethinyl estradiol and 250 mcg of norgestimate. 46 healthy volunteers from Frankfurt, Germany, were enrolled and randomly assigned to receive one of the two OCs. Serum samples were collected on days 2, 11, and 21 of the control cycle and treatment cycles 3, 6, and 12. No significant differences between formulations were observed for any of the measured parameters. Significant increases were recorded during OC use in plasma levels of total triglycerides (24-78%), total phospholipids (7-20%), very low density lipoprotein (VLDL) triglycerides (61-76%), VLDL phospholipids (14-60%), low density lipoprotein (LDL) triglycerides (8-35%), LDL phospholipids (28-30%), high density lipoprotein (HDL) cholesterol (8-16%), HDL 3-cholesterol (11-20%), HDL triglycerides (17-66%), HDL phospholipids (7-11%), apolipoprotein (apo) A-I (5-20%), and apo A-II (10-40%). In contrast, LDL-cholesterol levels were significantly decreased during treatment with both formulations. These changes appear to reflect a predominance of the effect of the estrogen component.

The effects on mortality of the use of combined oral contraceptives.

OBJECTIVE: To assess, using a computerised model, the effects on mortality of the use of combined oral contraceptives (COCs). DESIGN: Two hypothetical cohorts of one million women each, identical except for their use of contraception. The impact of COC use was explored by applying, to each cohort, death rates which were adjusted according to a series of assumptions about the risks associated with COC use. The model also explored the effects of a number of different patterns of COC use. SETTINGS AND SUBJECTS: Women aged 16, followed through to ages 50 and 75, exposed to 1994 UK death rates. MAIN OUTCOME MEASURES: Numbers of deaths from various cancers and cardiovascular diseases attributable to COC use. RESULTS: Based on the standard pattern of use, there were 1.7 per cent more deaths in the COC cohort to age 50. The important effects on mortality of different patterns of use and of different assumptions about risks in ex-users were illustrated. CONCLUSIONS: The results confirm the findings of earlier work and provide some reassurance about the likely adverse effects of COC use.

PIP: A computer model that takes account of state-of-the-art knowledge about the health risks and benefits of combined oral contraceptive (OC) use was applied to two hypothetical cohorts of 1 million women each aged 16 years and followed through to ages 50 and 75 years, exposed to 1994 UK death rates, who were identical except for contraceptive use. The effects of OC use were examined in relation to deaths from five cancers (ovarian, breast, cervical, endometrial, and liver), cardiovascular diseases (acute myocardial infarction, cerebral thrombosis, venous thromboembolism, and subarachnoid hemorrhage), and deaths associated with IUD use and sterilization. Among the five patterns of OC use included in the model, the standard was OC use from 16 to 29 years, IUD use from 30 to 39 years, and sterilization at age 40 years. Based on the standard pattern of use and standard risk assumptions, the number of deaths in the OC cohort at age 50 years was 1.7% higher than that in the control group, corresponding to an extra 465 deaths by age 50 years in a group of 1 million 16 year olds. To age 75 years, the number of deaths in the OC cohort was 0.5% lower than in the control group, corresponding to a benefit of 735 lives. The total number of deaths in the OC cohort compared to the control group ranged from an increase of 0.2-12.1% to age 50 years and from a decrease of 1.3% to an increase of 0.5% to age 75 years. Analysis of the standard set of assumptions against the standard pattern of use resulted in 22 fewer maternal deaths in the OC cohort than the control group. Although more research is needed on the implications of OC use well into the fifth decade of life, these results provide some reassurance about the likely adverse effects of OC use.

Metabolic interrelationships, cardiovascular disease, and sex steroids.

Future studies of the pharmacodynamics of oral contraceptives should encompass interactions between various areas of physiology rather than concentrate on single metabolic processes. Changes in one area of metabolism may affect other areas. Insulin plays a central role in metabolic control and, in addition to profound effects on carbohydrate and lipid metabolism, also affects the hematological system. Insulin has been proposed as a major physiological regulator of plasminogen activator inhibitor, and hyperinsulinemia is associated with increased blood coagulability and decreased fibrinoloysis. There is a close relationship between insulin and triglyceride metabolism, and this may affect factor VII activity in blood. There are many interactions between lipid metabolism and hematological factors. The apoprotein of lipoprotein (a) is structurally similar to plasminogen. Tissue factor pathway inhibitor, a regulator of coagulation, circulates in blood bound to LDL and HDL. Some fatty acids may act as a contact surface for activation of hematological factors. Dietary factors also need to be considered. Changes not only in the quantity but also in the composition of dietary fat influence lipid metabolism and also blood levels of a number of hematological factors. Other aspects of pharmacodynamic studies that require consideration and other factors that affect metabolic interrelationships are discussed.

PIP: Previous studies of the pharmacodynamics of sex steroids in vivo have tended to focus on single metabolic processes, with scant attention to their integration. This paper reviews the human in vivo research evidence on the effects of diet on metabolism, the central role of insulin in metabolic control, and the interaction between lipid metabolism and hematologic factors. There is a need for more attention to events occurring within vascular tissue itself, especially at the site of atherosclerosis and thrombus formation. Insulin has been proposed as a major physiologic regulator of plasminogen activator inhibitor, and hyperinsulinemia is associated with increased blood coagulability and decreased fibrinolysis. There is a close association between insulin and triglyceride metabolism, and this may affect factor VII activity in blood. Changes in both the quantity and composition of dietary fat influence lipid metabolism as well as blood levels of a number of hematologic factors. Sex steroid-induced changes in aspects of metabolism such as blood lipid concentrations may not be as central to cardiovascular disease risk as originally believed. Changes in carbohydrate metabolism and in hematologic parameters induced by the currently used doses of sex steroids appear to be minor, although more intensive studies are recommended. Evidence suggests that genetic and early life influences are more important to the development of insulin resistance than later acquired causes.

Safety evaluation of modern oral contraceptives. Effects on lipoprotein and carbohydrate metabolism.

An ideal oral contraceptive should either be neutral as regards metabolic risk markers for arterial disease or should only change them in directions that would be expected to reduce risk. Depending on their formulation, modern low dose oral contraceptives affect systems such as hemostasis, lipoprotein metabolism, and glucose and insulin metabolism. Some of these actions would be expected to decrease the risk of arterial disease and some might be expected to increase risk. Despite these associations there is at present no justification for widespread metabolic screening as a strategy to further improve oral contraceptive safety. Recent developments in atherosclerosis research support the introduction of progestogens such as desogestrel that allow the estrogenic increase in high density lipoprotein levels to persist and that may cause less of an elevation in plasma insulin responses to glucose. The predicted benefit of these formulations in terms of arterial disease is difficult to demonstrate in an epidemiological setting because of the rarity of the disease in young women.

PIP: Depending on their formulation, modern low-dose oral contraceptives (OCs) affect systems such as hemostasis, lipoprotein metabolism, and glucose and insulin metabolism. This paper reviews the research literature on these effects and assesses their clinical implications for arterial disease. At greatest risk are women with high plasma levels of the potentially atherogenic lipoproteins (low, intermediate, and very low density lipoproteins) and low plasma levels of high density lipoprotein. Recent studies suggest that the quality of lipoproteins, especially their ability to resist oxidative modification, may be as important as their plasma levels. At present, there is no justification for widespread metabolic screening as a strategy to further improve OC safety. Recent developments in atherosclerosis research support the introduction of progestogens such as desogestrel that allow the estrogenic increase in high density lipoprotein levels to persist and that may cause less of an elevation in plasma insulin responses to glucose. The predicted benefit of these formulations in terms of arterial disease is difficult to demonstrate in an epidemiologic setting, however, because of the rarity of the disease in young women.

The risks of venous thromboembolic disease among German women using oral contraceptives: a database study.

This study investigated the risk of venous thromboembolic disease (VTE) between second and third generation combined oral contraceptives, using the German MediPlus database of patient records. Women studied included 42 patients between the ages of 18 and 49 years, with a diagnosis of VTE treated with an anticoagulant, who were exposed to an oral contraceptive (OC). Four controls per patient (168), matched by year of birth and exposure to an OC on the even day, were identified. More women were users of second generation than third generation OC, and none were using progestogen-only pills. There was no significant difference between patients and control subjects with respect to the type of OC used on the event day (unadjusted odds ratio for third versus second generation users was 0.77; 95% confidence interval [CI] 0.38-1.57). There was no significant age difference between second and third generation users among patients or control subjects. Between January 1 and the event date, there was no significant difference between the patients and control subjects in terms of the number of oral contraceptive prescriptions, number of consultations for psychotherapeutic complaints, or mixed physical and psychotherapeutic consultations; however, patients did demonstrate significantly more consultations for purely physical complaints compared with control subjects (p < 0.0001). There were no significant consultation differences between patients with pulmonary emboli (n = 6) and other VTE patients (n = 36). No significant differences with respect to VTE risk between users of second and third generation oral contraceptives were found in this study. Consultations (physical) for patients were higher than for control subjects before the VTE event. If consultation rate relates to the general health status of a person, this might indicate that VTE risk is higher among women of poorer health, but that this is not related to the type of progestogen in the oral contraceptive that they use.

PIP: The German MediPlus database of patient records from 451 practices was used to investigate the risk of venous thromboembolism (VTE) in users of second- and third-generation combined oral contraceptives (OCs). Cases included 42 women 18-49 years of age with a diagnosis in 1992-95 of a VTE treated with an anticoagulant and with a history of OC use. Also enrolled were 168 controls (4 per case), matched to cases by year of birth and exposure to an OC on the event day. 64.3% of cases and 53.0% of controls had used a second-generation OC; use of a third-generation OC was reported by 35.7% of cases and 38.1% of controls. No significant differences in terms of VTE risk factors were identified between users of second- and third-generation OCs. The odds ratio for VTE among users of third-generation compared to second-generation OCs was 0.77 (95% confidence interval, 0.38-1.57). There were no significant differences between cases and controls in terms of the type of OC used, age, the total number of OC prescriptions issued, number of consultations for psychotherapeutic complaints, and number of consultations for mixed psychotherapeutic and physical complaints. Although cases had more consultations for physical complaints before the VTE (presumed to be a proxy for poor general health), this was not related to the type of progestogen in the OC.

Gabapentin does not interact with a contraceptive regimen of norethindrone acetate and ethinyl estradiol.

Anticonvulsants that induce hepatic metabolism increase clearance of oral contraceptive hormones and thereby cause contraceptive failure. Gabapentin is not metabolized in humans and has little liability for causing metabolic-based drug-drug interactions. In healthy women receiving 2.5 mg norethindrone acetate and 50 microg ethinyl estradiol daily for three consecutive menstrual cycles, concurrent gabapentin administration did not alter the steady-state pharmacokinetics of either hormone. Thus, gabapentin is unlikely to cause contraceptive failure.

PIP: Anticonvulsants that induce hepatic metabolism increase the clearance of synthetic estrogens and progestogens used in oral contraceptives (OCs), thereby potentiating contraceptive failure. In contrast, the anticonvulsant drug gabapentin is not metabolized in humans and has little liability for metabolic-based drug interactions. The present study sought to confirm whether concurrent administration of gabapentin would alter the pharmacokinetics of norethindrone acetate (2.5 mg) and ethinyl estradiol (50 mcg) in healthy US women. A total of 13 women were enrolled for three menstrual cycles each. Pharmacokinetic values did not change appreciably as a result of the addition of gabapentin. The rate and extent of absorption of both hormones were unaffected by the anticonvulsant. Gabapentin plasma concentration time profiles and pharmacokinetic values from this study were similar to historical values after administration of gabapentin alone. The observed lack of interaction between gabapentin and norethindrone acetate or ethinyl estradiol is consistent with the fact that gabapentin is not metabolized, is not an inducer or inhibitor of hepatic drug metabolizing enzymes, is absorbed via a specific transport system for amino acids, and is not bound to plasma proteins. Anticonvulsant drugs that do not interact with OCs should be considered for the treatment of epileptic women of childbearing age who are using this method of fertility control.

Dimeric inhibin serum values as markers of ovarian activity in pill-free intervals.

Levels of inhibin A and B as well as other hormones in serum samples obtained during the pill-free interval in women taking combined oral contraceptives (OC) were measured to asses the extent of ovarian activity during that period. Type of pill and day of pill-free interval were recorded during routine gynecologic check-ups, if patients were in the pill-free period and had taken their pills regularly in the previous cycle. In addition to inhibin A and B, serum levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), and progesterone were also quantified. Inhibin B levels rise significantly in parallel with rising levels of FSH, LH, and E2. Progesterone levels were completely suppressed and inhibin A levels rose slightly but insignificantly. Inhibins are sensitive biochemical markers of ovarian activity in pill-free intervals.

PIP: Serum values of dimeric inhibin A and B were measured to assess the restoration of pituitary and ovarian activity during the pill-free interval in women taking combined oral contraceptives. 175 healthy women 18-35 years of age from five areas in Belgium were enrolled and monitored during routine gynecologic examinations. During the 7 day pill-free interval, inhibin B levels rose significantly in parallel with rising levels of follicle-stimulating hormone, luteinizing hormone, and estradiol. Progesterone levels were completely suppressed. Inhibin A levels rose slightly but insignificantly, reflecting an absence of development of preovulatory follicles. These findings indicate that inhibins are sensitive biochemical markers of ovarian activity in pill-free intervals. Inhibin B appears to be predominantly a product of the cohort of developing primary and subsequent early antral follicles, while inhibin A secretion is more indicative of dominant follicular and corpus luteum function.

Effects of oral contraceptives on free fatty acid metabolism in women.

These studies examined whether women using oral contraceptives have abnormalities in free fatty acid (FFA) metabolism compared with women not using oral contraceptives. Plasma palmitate kinetics ([3H]palmitate) were measured at rest, following glucose ingestion, and during epinephrine infusion in 13 oral contraceptive users and 13 matched women not using oral contraceptives. Oral contraceptive users had significantly greater plasma triglyceride concentrations and glucose responses to oral glucose tolerance testing. No differences in basal (2.1 +/- 0.1 v 1.8 +/- 0.2 micromol x kg fat-free mass x FFM(-1) x min[-1]), glucose-suppressed (0.6 +/- 0.1 v 0.5 +/- 0.1 micromol x kg FFM(-1) x min[-1]), or epinephrine-stimulated (3.3 +/- 0.1 v 3.6 +/- 0.2 micromol x kg FFM(-1) x min[-1]) palmitate flux were detected between women using and not using oral contraceptives. The respiratory quotient (RQ) also was not different between groups. We conclude that the increase in plasma triglycerides and the mild glucose intolerance seen with oral contraceptive use is not associated with significant abnormalities of FFA metabolism.

PIP: Some researchers have reported differences in plasma free fatty acid (FFA) concentrations between oral contraceptive (OC) users and nonusers, suggesting an influence of the pill on adipose tissue lipolysis. To confirm this finding, the present study examined FFA flux in 13 OC users and 13 controls matched for age, weight, and body composition under overnight postabsorptive conditions, after glucose ingestion to suppress FFA release, and during an epinephrine infusion to simulate lipolysis. No significant differences in FFA kinetics were observed between OC users and nonusers under any of these three conditions. In addition, the respiratory quotient under all three conditions was comparable in the two groups of women, suggesting no effect of OCs on fatty acid oxidation. The results suggest that the hypertriglyceridemia and slightly greater plasma glucose response to glucose tolerance testing seen in OC users are not associated with significant abnormalities of effective adipose tissue lipolysis.

Can oral contraceptive steroids influence the elimination of nifedipine and its primary pryidine metabolite in humans?

OBJECTIVE: To investigate the influence of oral contraceptives on cytochrome P450 3A4 (P450NF) activity. METHODS: In 23 healthy women, the pharmacokinetics of nifedipine and its main metabolite dehydronifedipine in plasma were assessed after a single oral dose, prior to and after intake of one of two oral contraceptive formulations, one containing 2 mg dienogest and 0.03 mg ethinylestradiol (group A) and the other containing 0.125 mg levonorgestrel and 0.03 mg ethinylestradiol (group B). RESULTS: While the intake of two oral contraceptives for 21 days did not influence the plasma concentration-time curve of unchanged nifedipine, mean AUC0-23.5 h and the mean Cmax values of dehydronifedipine were significantly lower in both groups tested/(24% in group A and 25% in group B). This observation may indicate a reduced formation rate of metabolites and reflects an inhibition of cytochrome P450 3A4 activity. The activation of the same or other metabolic degradation mechanism(s) could explain this result. CONCLUSION: The investigation presented demonstrates the importance of metabolite measurement when in vivo studies are undertaken to investigate different influences on drug metabolizing ability.

PIP: The influence of combined oral contraceptives (OCs) on cytochrome P450 3A4 activity was investigated in a study of 23 healthy women. The pharmacokinetics of nifedipine and its main metabolite, dehydronifedipine, in plasma were measured after a single oral dose, before and after intake of 1 or 2 OC formulations. The first contained 2 mg of dienogest and 0.03 mg of ethinyl estradiol (group A); the second contained 0.125 mg of levonorgestrel and 0.03 mg of ethinyl estradiol (group B). OC use for 21 days did not influence the plasma concentration-time curve of unchanged nifedipine. However, the mean area-under-curve values at 0-23.5 hours after nifedipine administration were significantly lower than baseline (by 24% in group A and 25% in group B). This finding may indicate a reduced formation rate of metabolites and reflects an inhibition of cytochrome P450 3A4 activity. Activation of the same or other metabolic degradation mechanisms could explain this finding. Assessment of its clinical importance requires a longer period of nifedipine use. This study demonstrates the importance of metabolite measurement when in vivo studies are undertaken to investigate different influences on drug metabolizing activity.

Effect of desogestrel-containing oral contraceptives on vascular reactivity and catecholamine levels.

The modifications induced by new oral contraceptives (OC) on blood pressure, great vessel vascular reactivity by color Doppler, and catecholamine levels were investigated. Young healthy women not taking OC (n = 22; controls) or receiving, for > or = 6 months, OC containing desogestrel with either 30 micrograms (n = 14) or 20 micrograms of EE (n = 8) were enrolled. Blood pressure measured at rest in supine position was similar between controls and OC users. The pulsatility index (PI), an indirect index of resistance to blood flow, of axillary artery was significantly higher (p < 0.05) in 30 micrograms than in 20 micrograms EE OC users or controls. A similar trend, albeit not significant, was observed for the internal carotid artery PI. Norepinephrine (p < 0.01) and dopamine (p < 0.05) but not epinephrine levels, were lower in 30 micrograms EE OC users than in 20 micrograms EE OC users or controls. Thus, both 20 micrograms and 30 micrograms EE OC had no negative effect on blood pressure, but the 30 micrograms EE OC tended to increase great vessel resistance to blood flow, independently of catecholamine levels.

PIP: The effects on blood pressure of oral contraceptives (OCs) containing desogestrel plus either 20 or 30 mcg of ethinyl estradiol were investigated in 22 women who had been using one of these formulations for 6 months or more and 22 matched controls. There were no significant differences between both groups of cases and controls in blood pressure measured at rest in supine position. However, subtle differences were recorded in vascular reactivity, as evaluated by color Doppler ultrasound investigation. Compared to controls and users of OCs containing 20 mcg of ethinyl estradiol, cases taking OCs consisting of 30 mcg of ethinyl estradiol had a significantly higher pulsatility index in the axillary artery, indicating increased vessel resistance to blood flow. A similar, although not significant, trend was found in the internal carotid artery pulsatility index. On the other hand, catecholamine (dopamine and norepinephrine) levels were reduced by both OCs in a dose-dependent manner. Overall, these findings suggest that third-generation OCs, especially those containing 20 mcg of ethinyl estradiol, exert minimal effects on circulatory parameters. It is hypothesized that the cardiovascular effects of catecholamine reduction are antagonized by other mechanisms, among them the stimulus on the renin-angiotensin system.

Normalized activated protein C ratio itself not associated with increased risk of venous thromboembolism.

Recently, discussions focused on the question whether acquired activated (APC) resistance is a clue to the observed association between venous thromboembolism (VTE) risk and oral contraceptive (OC) use, especially with the so-called third-generation OC. The objective of our study was to check the validity of acquired APC resistance regarding VTE risk in a case-control study. Sixty-seven women with confirmed VTE diagnosis (n = 67) were consecutively ascertained in primary health care settings, interviewed and blood samples taken (at the earliest 6 months after VTE). Cases were age-matched to 290 population controls. Acquired APC resistance was measured as normalized APC ratio (APCRN). The effect of APC on tissue factor initiated thrombin generation was measured in plasma using alpha 2-macroglobulin attached thrombin activity as an endpoint. Higher risk (odds) ratio with 95% CI) of VTE for carriers of heterozygote Factor V Leiden mutation was confirmed [OR = 2.72 (CI:1.51-4.92)]. However, there is no association between VTE and the level of APCRN OR 0.65 (CI:0.35-1.22). We conclude that acquired APC resistance, measured with a tissue factor initiated test, is unlikely to have a direct association to the clinical outcome of venous thromboembolism.

PIP: The observed association between venous thromboembolism (VTE) risk and third-generation oral contraceptive (OC) use may reflect an OC-related activated protein C (APC) resistance. The validity of this hypothesis was assessed in a case-control study of 67 reproductive age women (mean age, 34.2 years) from southeast Germany diagnosed with VTE in 1995-97 and 290 age-matched population controls. The normalized APC ratio was measured as the effect of APC on tissue factor-initiated thrombin generation in plasma using alpha 2-macroglobulin attached thrombin activity as an end point. To minimize the influence of acute phase proteins, blood was drawn at least 6 months after VTE. The risk of VTE was significantly increased in carriers of heterozygote Factor V Leiden (odds ratio (OR), 2.72; 95% confidence interval (CI), 1.51-4.92). There was no association between VTE and the normalized APC ratio (OR, 0.65; 95% CI, 0.35-1.22). In women who were not current OC users and did not have the Factor V Leiden mutation, there was no association between the normalized APC ratio and VTE risk. Thus, Factor V Leiden mutation and OC use may be mild confounders of the association between the normalized APC ratio assay for acquired APC resistance and VTE risk. Overall, these findings suggest that acquired APC resistance does not have a direct association with the clinical outcome of VTE. However, this lack of association should be confirmed in a cohort study capable of taking time-dependent influences into account.

Estrogen and progestin components of oral contraceptives: relationship to vascular disease.

Recently, new information has been published about: a) the relationship between combination oral contraceptives (OCs), estrogen dose, cigarette smoking, and the risk of myocardial infarction (MI) and stroke; and b) the effect of different progestins on the risk of venous thromboembolism (VTE). We review the epidemiologic data. Regardless of age, in the absence of smoking, use of sub-50 micrograms OCs is not associated with any meaningful increase in risk of MI or stroke. If the small, statistically nonsignificant elevations in risk for these diseases are assumed (for the sake of argument) to be causal, then the incidence of MI and stroke associated with use of OCs containing less than 50 micrograms ethinyl estradiol (EE) would be approximately 2 per 100,000 per year. For women less than 35 years of age who do not smoke or do not have a history of hypertension, the risk would be even lower. Any woman over the age of 35 who smokes should be advised to use a non-estrogen or nonhormonal contraceptive. There are now two reports, from jick et al. and Lewis et al., that demonstrate that the relative risk of MI is certainly no greater for users of OCs containing desogestrel or gestodene than for users of OCs containing older progestins. In fact, both show reduced relative risks for the newer progestins compared to the older ones. With respect to progestins, four recent epidemiologic studies have indicated a twofold increased risk of nonfatal VTE with use of OCs containing desogestrel or gestodene compared with levonorgestrel. A fifth report, which showed an increased relative risk for norgestimate, is based on use among only 19 cases and 31 controls and is not statistically significant. As the authors themselves caution and as subsequent follow-up analyses and editorials conclude, these studies do not provide evidence for a cause-and-effect relationship between OCs containing desogestrel or gestodene, and VTE. The recommendation with respect to desogestrel- and gestodene-containing OCs is that no change in prescribing practices is warranted for either current or new-start patients. There is a growing body of evidence demonstrating that OCs containing 30 or 35 micrograms of EE have lower risks of MI, stroke, and VTE than higher dose OCs. However, there is no epidemiologic study that demonstrates a greater risk of vascular events among women using OCs containing 30 or 35 micrograms EE compared with preparations containing 20 micrograms EE. Users of sub-50 micrograms OCs of any age have no clinically meaningful increase in incidence of MI or stroke compared with non-OC users. This is also true for smokers under the age of 35 years who use OCs. However, smokers over the age of 35 years who use OCs still have an unacceptably high incidence rate of MI and stroke and should not use combination OCs. Sub-50 micrograms OCs of all types are associated with a small excess risk of VTE, about 15 per 100,000 events per year. Until there is biologic explanation of the twofold greater risk of VTE in users of OCs containing desogestrel or gestodene compared with users of those containing older progestins, this association should not be accepted as one of cause and effect.

PIP: A review of the recent epidemiologic evidence indicates that use of low-dose combined oral contraceptives (OCs) is not associated with any clinically significant increase in risk of myocardial infarction (MI) or stroke, including for smokers under 35 years of age. Even if the small elevation in risk for these diseases is assumed to be causal, the incidence of both MI and stroke associated with use of OCs containing under 50 mcg of ethinyl estradiol would be only 2 per 100,000 events per year. This risk would be even lower for women under 35 years of age who do not smoke and have no history of hypertension. Two new reports have identified even lower relative risks of MI and stroke among users of OCs containing the progestins desogestrel and gestodene compared with users of second-generation OCs. However, four additional epidemiologic studies have revealed a two-fold increased risk of non-fatal venous thromboembolism for OCs containing desogestrel or gestodene compared to levonorgestrel; the excess risk is about 15 per 100,000 events per year. Until there is a biologic explanation of the greater thromboembolism risk in users of third-generation OCs, this association should not be viewed as causal and no change in OC prescribing practices is warranted for either current or new acceptors. However, smokers over 35 years of age should not use any combination OCs.

Factor V Leiden mutation and the risks for thromboembolic disease: a clinical perspective.

BACKGROUND: A single point mutation in the gene coding for coagulation factor V results in a form of factor Va that is resistant to degradation by activated protein C and leads to a relative hypercoagulable state. This mutation, factor V Leiden, is found in 4% to 6% of the U.S. population. PURPOSE: To review clinical data on factor V Leiden mutation, with emphasis on prevalence of and risks for thromboembolism and implications for screening and management. DATA SOURCES: A MEDLINE search of the English-language literature published between 1993 and April 1997 and an extensive bibliography review. STUDY SELECTION: Case-control and prospective cohort studies were reviewed if clinical features of thromboembolic disease associated with factor V Leiden mutation or resistance to activated protein C were presented. Original research articles were reviewed if they addressed the identification of the laboratory abnormality of activated protein C or factor V Leiden mutation. Case reports and case series were reviewed when no analytic data were available. DATA EXTRACTION: Review of the identified articles. DATA SYNTHESIS: Factor V Leiden mutation is associated with three- to sixfold increases in risks for primary and recurrent venous thromboembolism, especially in patients without transient risk factors, such as surgery or trauma. Risks for venous thromboembolism in genetically affected persons are substantially higher among patients with coexistent predispositions for thrombosis, such as advanced age, use of oral contraceptives, hyperhomocystinemia, and deficiencies of protein C and protein S. Factor V Leiden mutation does not seem to increase risks for arterial thrombosis. Whether patients with the mutation would benefit from more intense or prolonged anticoagulation is unknown. CONCLUSIONS: The presence of factor V Leiden mutation predisposes patients to venous thromboembolism, but screening for this disorder is of uncertain utility. Decisions about whether to screen for the mutation will depend on the results of clinical trials designed to evaluate the benefit-to-risk ratio of long-term anticoagulation in the secondary prevention of venous thromboembolism in patients with resistance to activated protein C.

PIP: The factor V Leiden mutation, present in 4-6% of the US population, makes the activated form of factor V relatively resistant to degradation by activated protein C, in turn producing resistance to activated protein C. Clinical studies have suggested that factor V Leiden mutation increases the risk of venous thrombosis during pregnancy and in oral contraceptive (OC) users, but the benefit-to-risk ratio of screening for this mutation is unclear. This paper reviews English-language articles published in 1993-97 on resistance to activated protein C or the factor V Leiden mutation with regard to laboratory diagnosis, prevalence, risks for thromboembolic disease, screening, and management. Included were case-control studies, prospective cohort studies, and case reports. The literature suggests that factor V Leiden mutation is associated with 3- to 6-fold increases in risks for primary and recurrent venous thromboembolism. In genetically affected persons, this risk is substantially higher among those with co-existent predispositions for thrombosis, including advanced age, OC use, hyperhomocystinemia, and deficiencies of proteins C and S. Any decisions about screening for factor V Leiden must await clinical trials designed to evaluate the benefit-to-risk ratio of long-term anticoagulation in the secondary prevention of venous thromboembolism in patients with resistance to activated protein C.

Haemostatic changes and the oral contraceptive pill.

Oral contraceptives have been linked to an increased incidence of thrombovascular disease. This may be mediated by their effects on the haemostatic system. An increase in the activity of coagulation Factors VII, X and fibrinogen occur with pill usage. Increased Factor VII levels are dependent on both the oestrogen and progestogen component of the oral contraceptive. A reduction in antithrombin III levels has also been observed in some but not all studies. Increased fibrinolysis has also been shown in oral contraceptive users which should balance the changes in the coagulation pathway. The increase in fibrinolytic potential is thought to be due to a decrease in the levels of plasminogen activator inhibitor I combined with an increase in the levels of plasminogen; tissue plasminogen activator antigen is decreased in most studies. The increased levels of endpoints of coagulation and fibrinolysis in pill users indicate that enhanced activity of both systems is occurring in vivo. The increased coagulation activity appears to be balanced by the rise in fibrinolytic activity, so preserving haemostatic balance. Enhanced platelet activity has also been shown in women taking oral contraceptives. Thrombus formation can result, however, when local vascular wall damage exists, or when other risk factors for thrombo-embolism, such as older age and smoking, coexist and create a local activation resulting in a thrombus. In these situations, the small differences in levels of coagulation factors in women taking different oral contraceptive formulations may be important. Pills containing the lowest doses of oestrogen (20 micrograms ethinyloestradiol) have shown the least changes in haemostatic factors. The progestogen component of the pill modifies the effect of oestrogen on the haemostatic system.

PIP: Numerous studies have investigated the mechanisms whereby, in some healthy women, oral contraceptives (OCs) can cause thrombosis. In healthy OC users, the coagulation and fibrinolytic systems are in dynamic balance and control fibrin deposition within the vascular compartment. The production of fibrin can be initiated by either the intrinsic or extrinsic coagulation pathway. Factors such as light smoking in older women or an asymptomatic borderline deficiency in a coagulation inhibitor may cause thrombogenic changes in the hemostatic system in women with a normal blood flow and a healthy vascular endothelium. OC usage is associated with an increase in the activity of coagulation factors VII, X, and fibrinogen; a reduction in antithrombin III levels has been shown in some studies. The Factor V Leiden mutation explains 20-50% of cases of venous thromboembolism. Low doses of ethinyl estradiol combined with less androgenic third-generation progestogens were thought to represent little or no risk of thromboembolic or cardiovascular disease, which may have led to overprescribing of these formulations to women with possible risk factors for thromboembolic disease. Pills containing 20 mcg of ethinyl estradiol have shown the least changes in hemostatic factors. The progestogen component of the OC modifies the effect of estrogen on the hemostatic system. The recent epidemiologic reports suggest that all pills carry an increased thrombotic risk, and each woman requesting OCs should undergo careful clinical screening.

The pill and thrombosis: epidemiological considerations.

The oral contraceptive is one of the most widely taken medications in the healthy population. The clinically important side-effects are venous and arterial thrombosis. Accurate estimates of incidence of these side-effects have proven to be difficult. Diagnostic modalities for thrombosis are sub-optimal and the problems of study methodology, primarily a reliance on non-experimental studies, have limited the ability to define the attributable risk of thrombosis from oral contraception. Pharmacological attempts to further decrease venous thrombotic side-effects by the use of third-generation oral contraceptives have failed. This places a greater emphasis on the selection of patients to help avoid giving medication to those patients with underlying thrombotic risk factors. An example of this approach has been the clear confirmation of the adverse effects of cigarette smoking and arterial thrombosis in oral contraceptive users. At the biochemical level, hypercoagulability testing may be useful. Screening for high-frequency prothrombotic abnormalities, such as the Factor V Leiden genotype, represents an important addition to the process by which patients are selected, and may be prototypic of further advances.

PIP: Precise definition of the attributable risk of thrombosis from oral contraceptive (OC) use has been limited by a reliance on nonexperimental studies and suboptimal diagnostic modalities. The consensus of studies conducted up to 1995 was that there is a small but real increase in the incidence of deep venous thrombosis in OC users that can be overcome by careful patient selection and use of formulations containing low doses of estrogen. The subsequent finding of an increased thrombotic risk associated with third-generation OCs containing the progestins desogestrel and gestodene placed an even greater emphasis on the need to select patients without underlying thrombotic risk factors. At the biochemical level, hypercoagulability may be useful. Screening for high-frequency thrombotic abnormalities such as Factor V Leiden genotype may be the prototype of further advances. In general, however, nonexperimental studies will continue to dominate, with the concomitant risk of referral bias.

The role of triphasic levonorgestrel in oral contraception: a review of metabolic and hemostatic effects.

Epidemiological studies suggesting a possible association between the use of combined oral contraceptives and an increased risk of cardiovascular disease have led to extensive investigations into the effect of oral contraceptives on lipid and carbohydrate metabolism, and on hemostasis. Since this association was originally suggested, the steroid dose in oral contraceptives has been significantly reduced and new progestogens have been developed. Also, triphasic formulations have been introduced which offer a well-balanced estrogen/progestogen ratio, allowing a further reduction in the progestogen dose per cycle, and thus helping to minimize unwanted metabolic and hemostatic effects. The metabolic interactions of triphasic levonorgestrel, the first triphasic formulation to be introduced, have received particular attention. Lipid metabolism appears to be largely unaffected by triphasic levonorgestrel, most studies reporting no significant change in high- (HDL-C) or low-density lipoprotein-cholesterol (LDL-C) levels. Several studies have reported a decrease in the lipoprotein subfraction HDL2-C levels, but in most cases measurements of the LDL-C/HDL-C and apolipoprotein A-1/B ratios reveal no clinically significant effects. Concerning lipids, most studies suggest that triphasic levonorgestrel has less metabolic impact than the monophasic formulation. In common with all currently available oral contraceptives, triphasic levonorgestrel appears to have some effect on carbohydrate metabolism. The study results vary, however; some investigators have found an impairment of glucose tolerance, whereas others have not detected any significant effect. Compared with lipid and carbohydrate metabolism, fewer studies have investigated the effect of triphasic levonorgestrel on hemostasis. In common with all estrogen-containing oral contraceptives, levonorgestrel appears to stimulate some procoagulant activity, elevating the levels of factors VII and X, and fibrinogen. However, the effect of triphasic levonorgestrel appears to be balanced, with most studies reporting a corresponding increase in anti-coagulant-fibrinolytic activity. Although most of the studies reviewed here reported some statistically significant metabolic interactions, many authors comment that the changes are probably not clinically relevant in terms of an altered risk of cardiovascular disease. The true risk of vascular disease associated with modern low-dose oral contraceptives remains to be confirmed when sufficient epidemiological data eventually become available.

PIP: This review of studies published since 1985 on the metabolic effects of triphasic levonorgestrel in oral contraceptives (OCs) reveals that triphasic levonorgestrel tends not to affect lipid metabolism. Specifically, it does not significantly change the level of the lipids associated with increased vascular risk (low density lipoprotein-cholesterol (LDL-C), high density lipoprotein-cholesterol (HDL-C), and lipid ratios). Overall, triphasic levonorgestrel OCs do not affect triglycerides, while other OCs increase the level of triglycerides. They also have a lower effect on lipid metabolism than do monophasic levonorgestrel OCs. About 50% of the studies suggest that triphasic levonorgestrel OCs do not affect carbohydrate metabolism, while the remaining studies suggest that they cause a significant increase in carbohydrates. Some studies found that triphasic levonorgestrel OCs impair glucose tolerance. Studies on the effect of triphasic levonorgestrel on hemostasis are less numerous than those on lipid and carbohydrate metabolism. Unlike all combined monophasic OCs, triphasic levonorgestrel OCs appear to have a balanced effect on hemostasis. In fact, they stimulate both the coagulant and anti-coagulant-fibrinolytic pathways. Many of the researchers who found statistically significant metabolic interactions found that the changes fell within the normal clinical range and therefore did not increase the risk of cardiovascular disease. Triphasic levonorgestrel OCs tend to have less of a metabolic effect than its monophasic counterpart and a metabolic effect similar to other low-dose OCs.

Lack of interaction between orlistat and oral contraceptives.

OBJECTIVES: Orlistat, a potent and selective inhibitor of gastrointestinal lipases, is designed for the treatment of obesity. A double-blind, randomised, placebo-controlled, 2-way crossover study investigated the possible influence of orlistat on the ovulation-suppressing action of combination oral contraceptives (OC). METHODS: After an 8-day run-in prior to the first of two consecutive menstrual cycles (Day 1 was the first day of menstruation), two groups of 10 healthy women, 20-27 years of age and on a stable regimen with OCs, received either 120 mg orlistat t.i.d. or placebo t.i.d. on Days 1-23 of the first cycle, and, separated by a placebo washout period on Days 24-28, the alternative treatment on Days 1-23 of the second cycle. In both cycles, serum luteinizing hormone (LH) was measured on Days 12-16 and progesterone on Days 12, 16, 19-23. RESULTS: The geometric means of time averaged concentrations (Days 12-16 for LH and Days 19-23 for progesterone) in the cycles with orlistat and placebo, respectively, and the one-sided 95% confidence region for the mean in the cycle with orlistat were 1.92, 2.03 and < 2.23 IU 1-1 for LH and 0.147, 0.145 and < 0.176 micrograms 1-1 for progesterone. The one-sided 95% confidence region for the ratio (orlistat/placebo) of geometric means was < 1.06 for LH and < 1.11 for progesterone. CONCLUSION: During normal ovulation the peak serum concentration of LH is above 30 IU 1-1 around Day 14 of the cycle, and that of progesterone exceeds 3 micrograms 1-1 around Day 21. The 95% confidence regions for the means, as well as all individual concentrations, were below these limits. It was concluded that orlistat did not influence the ovulation suppressing action of oral contraceptives.

PIP: In the Netherlands, a double-blind, randomized, placebo-controlled, two-way crossover study was conducted to determine whether administration of the inhibitor of gastrointestinal lipases, Orlistat, concomitantly with combined oral contraceptives (OCs) inhibits the ovulation-suppressing action of OCs. The 20 subjects, 20-27 years old, were healthy and had a body mass index between 22 and 27 kg m-2. All subjects completed the study. Most adverse events were mild and related to the pharmacological effect of Orlistat (fatty or oily stool, flatus with discharge, or abdominal pain). The geometric means of time-averaged serum concentrations in the cycles with Orlistat and the placebo and the 1-sided 95% confidence region for the mean in the cycle with Orlistat were 0.147, 0.145, and less than 0.176 mcg l-1 for progesterone and 1.92, 2.03, and less than 2.23 IU l-1 for luteinizing hormone (LH), respectively. These figures were well below the peak concentrations during normal ovulation (3 mcg l-1 for progesterone and 30 IU l-1 for LH). The plasma concentration of Orlistat was either close to the limit of quantification (1 mcg l-1) or below this limit. These findings suggest that Orlistat had no effect on the ovulation-suppression capabilities of the OCs.

[Pharmacology of oral contraceptives]. / Pharmacologie des contraceptifs oraux.

Oral contraceptives include two types of steroids; ethinyl-estradiol as the main estrogenic component which dose vary from 20 to 50 micrograms per tablet (mostly 30 to 35 micrograms) and progestins essentially derivatives of 19 nortestosterone. Derivatives of 19 norprogesterone such as nomegestrol acetate or ST 1435 are not used as oral contraceptives but are being evaluated through parenteral administration, e.g. implants or transdermal systems. The assessment of the pharmacological properties of these progestins indicate a high antigonadotropic and a high antiestrogenic properties for levonorgestrel and for the newer gestagens as well. Therefore very low doses are being used in the current oral contraceptives. However, there is a lower margin of security with the low dose contraceptives than with previous standard combinations and especially when concomitant medications are ingested such as enzyme-inducing agents. Selection of contraceptive methods should be discussed when specific co-medications are necessary.

PIP: Oral contraceptives (OCs) contain two types of steroids: ethinyl estradiol (the estrogenic component [20-50 mcg/pill]) and progestins derived essentially from 19-nortestosterone. The progestins derived from 19-norprogesterone (e.g., nomegestrol acetate or ST 1435) are not yet used in OCs, but are being evaluated for parenteral administration. The study of pharmacologic properties of progestins suggests that levonorgestrel and its derivatives have a powerful antigonadotropic and antiestrogenic effect, which allow them to be effective at very low doses in new OCs. These derivatives are desogestrel, gestodene, and norgestimate. The margin of safety for levonorgestrel and its derivatives is limited, especially when taken with other medication, however. Drugs that reduce the efficacy of OCs include rifampicin, antiepileptic drugs (e.g., phenobarbital), antibiotics (neomycin, lincomycin, and ampicillin), griseofulvin (an enzyme-inducing drug), antimalarials, and antacids. Thus, selection of contraceptive methods must be discussed when certain concomitant medications are required.