Advertisement

Modern oral contraceptives and cardiovascular disease

      Abstract

      We reviewed evidence that bears on the cardiovascular safety of combined oral contraceptives containing second- and third-generation progestogens and <50 μg of estrogen. Recent epidemiologic studies indicate that current use of these formulations is associated with a smaller increase in the incidence of venous thromboembolism than earlier formulations. In some studies the increase for third-generation formulations containing desogestrel or gestodene was about 1.5 to 2 times that for second-generation formulations, but there is evidence that differences between users in underlying risk and likelihood of being diagnosed contributed to this result. Recent studies of myocardial infarction suggest a smaller increase in risk associated with modern formulations than with earlier ones; one study suggests a threefold increase for second-generation formulations and no increase for third-generation formulations, but the finding requires confirmation. Recent studies of stroke indicate little or no increase in risk for modern formulations among women without risk factors. We conclude that modern combined oral contraceptives are safer than earlier formulations with respect to cardiovascular disease, which occurs rarely in young women.

      Keywords

      Epidemiologic studies established that first-generation oral contraceptives—50 μg or more of estrogen combined with a progestogen—increased the risk of venous thromboembolism, stroke, and myocardial infarction.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      Subsequent formulations have contained <50 μg of estrogen, and new progestogens have been developed, such as levonorgestrel (second generation) and desogestrel, gestodene, and norgestimate (third generation).
      • Robinson GE.
      Low-dose combined oral contraceptives.
      • Newton JR.
      Classification and comparison of oral contraceptives containing new generation progestogens.
      Substantial epidemiologic evidence on the cardiovascular safety of second- and third-generation oral contraceptives has emerged only recently.
      In October 1995 the British Committee on Safety of Medicines, the equivalent of the U.S. Food and Drug Administration, wrote to all physicians and pharmacists in the United Kingdom about the safety of third-generation combined oral contraceptives. On the basis of the results of three then unpublished epidemiologic studies,
      • Poulter NR
      • Chang CL
      • Farley TMM
      • Meirik O
      • Marmot MG.
      Venous thromboembolic disease and combined oral contraceptives: results of international multicenter case-control study, World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      • Jick H
      • Jick SS
      • Gurewich V
      • Myers MW
      • Vasilakis C.
      Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptives with differing progestagen components.
      • Spitzer WO
      • Lewis MA
      • Heinemann LAJ
      • Thorogood M
      • MacRae KD.
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
      the committee warned that the risk of venous thromboembolism among users of formulations containing desogestrel or gestodene was twice that among users of second-generation oral contraceptives; the studies provided no information on formulations containing norgestimate. The committee recommended that oral contraceptives containing desogestrel or gestodene be used only by women who could not tolerate other formulations. The advisory was followed by a pill scare.
      • Anonymous
      Pill scares and public responsibility [editorial].
      Many British women switched from third- to second-generation oral contraceptives, others decided to use alternative methods of contraception or none at all,
      • Graham A
      • Grieve F
      • Davie J
      • Glasier A.
      Oral contraceptives and venous thromboembolism [letter].
      and several months later a rise in unplanned pregnancies and an increase in abortions was observed.
      • British Pregnancy Advisory Service
      Number of abortions; comparison over 3 months: December-February 1993/94 to 1995/96.

      Huge rise in abortions after government's warning: human cost of the pill scare. Daily Mail 1996 Apr 15.

      The studies on which the committee based its advisory have now been published, as have several other studies. The findings are important to American women, health professionals, and drug regulators. Third-generation oral contraceptives containing desogestrel have captured 15% of the U.S. market,
      • Food and Drug Administration
      Oral contraceptives and risk of blood clots.
      and second-generation formulations account for most of the remaining use. The recommendations of the committee are controversial
      • Anonymous
      Pill scares and public responsibility [editorial].
      • Johannisson E
      • International Committee for Research in Reproduction
      Safety of modern oral contraceptives.
      • MacRae K
      • Kay C.
      Third generation oral contraceptive pills: is the scare over the increased risk of thrombosis justified?.
      • McPherson K
      Third generation oral contraception and venous thromboembolism: the published evidence confirms the Committee on Safety of Medicine's concerns.
      : the U.S. Food and Drug Administration has not affirmed the recommendation, whereas Planned Parenthood has recommended that new users not receive oral contraceptives containing desogestrel.
      • Planned Parenthood Federation of America
      Update: thromboembolic impact of oral contraceptives.
      This article reviews the evidence bearing on the safety of low-estrogen second- and third-generation oral contraceptives.

      The evidence

      The major epidemiologic findings on first-generation oral contraceptives and cardiovascular disease were based on studies conducted mainly in the 1960s and 1970s.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      • Vessey MP.
      Female hormones and vascular disease: epidemiologic overview.
      We briefly review those results and their biologic plausibility, followed by a more detailed review of data from the recent studies of second- and third-generation oral contraceptives.

      Early studies

      Epidemiologic evidence

      Case-control and follow-up studies in the United States and the United Kingdom established that there was an increased risk of venous thromboembolism (deep vein thrombosis and pulmonary embolism) in current users of oral contraceptives. Risk was unrelated to duration of use and declined to the level of nonusers within a month after cessation.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      • Vessey MP.
      Female hormones and vascular disease: epidemiologic overview.
      Relative risk estimates ranged from about 2 to 11. Higher estrogen doses were associated with a higher incidence, but no clear evidence linked the dose or type of progestogen to risk.
      • Oral contraceptives, venous thrombosis, and varicose veins
      Royal College of General Practitioners' Oral Contraception Study.
      • Stolley PD
      • Tonascia JA
      • Tockman MS
      • Sartwell PE
      • Rutledge AH
      • Jacobs MP.
      Thrombosis and low-estrogen oral contraceptives.
      • Inman WHW
      • Vessey MP
      • Westerholm B
      • Engelund A.
      Thromboembolic diseases and the steroidal content of oral contraceptives: a report to the Committee on Safety of Drugs.
      • Meade TW
      • Greenberg G
      • Thompson SG
      Progestogens and cardiovascular reactions associated with oral contraceptives and a comparison of the safety of 50- and 30-μg oestrogen preparations.
      The risk of myocardial infarction was increased threefold to fourfold in current users and it declined to baseline within a month or so after cessation.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      • Vessey MP.
      Female hormones and vascular disease: epidemiologic overview.
      Limited evidence suggested that women who had used oral contraceptives for long durations might be at increased risk after cessation of use,
      • Slone D
      • Shapiro S
      • Kaufman DW
      • Rosenberg L
      • Miettinen OS
      • Stolley PD.
      Risk of myocardial infarction in relation to current and discontinued use of oral contraceptives.
      but subsequent studies ruled out that possibility.
      • Rosenberg L
      • Palmer JR
      • Lesko SM
      • Shapiro S.
      Oral contraceptive use and the risk of myocardial infarction.
      • Stampfer MJ
      • Willett WC
      • Colditz CA
      • Speizer FE
      • Hennekens CH.
      A prospective study of past use of oral contraceptive agents and risk of cardiovascular disease.
      The increased risk in current users was largely confined to cigarette smokers and older women.
      • Slone D
      • Shapiro S
      • Kaufman DW
      • Rosenberg L
      • Miettinen OS
      • Stolley PD.
      Risk of myocardial infarction in relation to current and discontinued use of oral contraceptives.
      It was unclear whether the estrogen content or the dose or type of progestogen affected the risk.
      • Inman WHW
      • Vessey MP
      • Westerholm B
      • Engelund A.
      Thromboembolic diseases and the steroidal content of oral contraceptives: a report to the Committee on Safety of Drugs.
      • Meade TW
      • Greenberg G
      • Thompson SG
      Progestogens and cardiovascular reactions associated with oral contraceptives and a comparison of the safety of 50- and 30-μg oestrogen preparations.
      The incidence of thrombotic and hemorrhagic strokes was increased among current oral contraceptive users, with the British follow-up studies indicating a relative risk of 5 for stroke overall.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      • Vessey MP.
      Female hormones and vascular disease: epidemiologic overview.
      Equivocal evidence suggested that the risk of hemorrhagic stroke might also be increased by past use.
      • Petitti DB
      • Wingerd J.
      Use of oral contraceptives, cigarette smoking, and risk of subarachnoid haemorrhage.
      The strokes associated with oral contraceptive use were concentrated among smokers and older women.
      Largely on the basis of British data, the incidence of venous thromboembolism per 100,000 women at risk per year was estimated to be 110 for current oral contraceptive users and 30 for nonusers, with a case fatality rate of 1% to 2%.
      • Vessey MP.
      Female hormones and vascular disease: epidemiologic overview.
      • Mortality among oral-contraceptive users
      Royal College of General Practitioners' Oral Contraception Study.
      • Vessey MP
      Steroid contraception, venous thromboembolism, and stroke: data from countries other than the United States.
      • Maguire MG
      • Tonascia J
      • Sartwell PE
      • Stolley PD
      • Tockman MS.
      Increased risk of thrombosis due to oral contraceptives: a further report.
      The corresponding figures for the incidence of myocardial infarction were 11 for current users and 4 for nonusers at ages 30 to 39 years and 89 for current users and 22 for nonusers at ages 40 to 49 years, with a case fatality rate of about 50%.
      • Mann JI
      • Vessey MP
      • Thorogood M
      • Doll R.
      Myocardial infarction in young women with special reference to oral contraceptive practice.
      • Mann JI
      • Inman WHW.
      Oral contraceptives and death from myocardial infarction.
      • Mann JI
      • Inman WHW
      • Thorogood M.
      Oral contraceptive use in older women and fatal myocardial infarction.
      For stroke, the estimates were 47 in users and 10 in nonusers, with a case fatality rate of 5% to 10%.
      • Vessey MP.
      Female hormones and vascular disease: epidemiologic overview.
      • Royal College of General Practitioners
      Oral contraceptives and health.
      • Vessey MP
      • McPherson K
      • Yeates D.
      Mortality in oral contraceptive users.

      Pharmacologic effects and biologic mechanisms

      Several lines of evidence pointed to thrombosis as the likely mechanism by which oral contraceptive use increased the risk. The epidemiologic findings of increased risks of venous thromboembolism, myocardial infarction, and stroke in current but not past users suggested an acute mechanism. Autopsy studies of women who died of myocardial infarction while using oral contraceptives found evidence of clots.
      • Engel HJ
      • Engel E
      • Lichtlen PR
      Coronary atherosclerosis and myocardial infarction in young women—role of oral contraceptives.
      Pharmacology studies indicated that oral contraceptives increased the tendency for both arterial and venous thrombosis, with some evidence that the effects were related to the dose of estrogen.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      Some oral contraceptives had marked adverse effects on carbohydrate and lipid metabolism, raising concern about increased atherosclerosis.
      • Meade TW
      • Greenberg G
      • Thompson SG
      Progestogens and cardiovascular reactions associated with oral contraceptives and a comparison of the safety of 50- and 30-μg oestrogen preparations.
      • Wynn V
      • Adams PW
      • Godsland I
      • Melrose J
      • Niththyananthan R
      • Oakley NW
      • et al.
      Comparison of effects of different combined oral contraceptive formulations on carbohydrate and lipid metabolism.
      • Tikkanen MJ
      • Nikkila EA.
      Oral contraceptives and lipoprotein metabolism.
      The early oral contraceptives with progestogens derived from 19-nortestosterone (e.g., norethisterone) tended to decrease levels of high-density lipoprotein, increase levels of low-density lipoprotein, and also impair glucose metabolism. (Oral contraceptives with progestogens derived from progesterone had little effect, but they were no longer used after suspicion was raised that these progestogens caused breast tumors in beagle dogs.
      • Weisz J
      • Ross GT
      • Stolley PD.
      Report of the public board of inquiry on Depo-Provera.
      • World Health Organization
      Facts about injectable contraceptives: memorandum from a WHO meeting.
      ) However, the epidemiologic evidence weighs heavily against an atherosclerotic mechanism insofar as past use and the duration of use are largely unrelated to risk.

      Recent studies

      Epidemiologic evidence

      Venous thromboembolism the world health organization collaborative study

      The World Health Organization (WHO) Collaborative Study is a hospital-based case-control study of premenopausal women of childbearing ages conducted during 1989 through 1993 in 17 countries.
      • Poulter NR
      • Chang CL
      • Farley TMM
      • Meirik O
      • Marmot MG.
      Venous thromboembolic disease and combined oral contraceptives: results of international multicenter case-control study, World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      Women with first episodes of idiopathic venous thromboembolism were identified and interviewed in the hospital—470 cases in Europe and 747 in Africa, Asia, and Latin America. For each case, up to three age-matched controls were also interviewed, for a total of 2998.
      The relative risk estimate for current oral contraceptive use relative to nonuse was increased about fourfold in Europe and threefold in the developing countries.
      • Poulter NR
      • Chang CL
      • Farley TMM
      • Meirik O
      • Marmot MG.
      Venous thromboembolic disease and combined oral contraceptives: results of international multicenter case-control study, World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      The major potential confounding variables considered, hypertension, parity, obesity, and cigarette smoking, did not materially influence the estimates.
      Separate analyses of second- and third-generation oral contraceptives were carried out.
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      Third-generation oral contraceptives were used in nine countries, but most of the use occurred among women from the Oxford center in the United Kingdom. On the basis of 158 cases and 360 hospital controls from Oxford, the relative risk estimate was significantly elevated for use of third-generation oral contraceptives relative to nonuse (relative risk 6.8) and for use of oral contraceptives containing levonorgestrel relative to nonuse (relative risk 3.1) (Table I). Users of third-generation oral contraceptives containing norgestimate (one case and two controls) were included with levonorgestrel users because norgestimate is metabolized to levonorgestrel or its metabolites.
      • Stanczyk FZ
      • Roy S.
      Metabolism of levonorgestrel, norethindrone and structurally related contraceptive steroids.
      • Shenfield GM
      • Griffin JM.
      Clinical pharmacokinetics of contraceptive steroids: an update.
      For third-generation use compared with levonorgestrel use, the relative risk estimate was 2.2 (95% confidence interval 1.1 to 4.2). Among users of desogestrel, the relative risk estimate was greater for users of preparations containing 20 μg of ethinyl estradiol than of those with 30 μg of ethinyl estradiol, but there were few users of the former. Estimates in the other centers, on the basis of smaller numbers and with wider confidence intervals, suggested a similar pattern to that observed in the United Kingdom.
      Table IResults on venous thromboembolism from recent studies
      StudyComparisonExposed cases (No.)Relative risk estimate with 95% confidence intervalNo. of cases per 100,000 woman years with 95% confidence interval
      Case-control studies
       WHO
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      *
        OxfordThird generation vs nonuse486.8 (3.5-13.4)
      Levonorgestrel† vs nonuse403.1 (1.7-5.5)
      Third generation vs levonorgestrel†2.2 (1.1-4.2)
        Other centersThird generation vs nonuse2320.7 (7.9-53.7)
      Levonorgestrel vs nonuse974.0 (2.7-5.8)
      Third generation vs levonorgestrel5.2 (2.0-13.7)
       Transnational
      • Spitzer WO
      • Lewis MA
      • Heinemann LAJ
      • Thorogood M
      • MacRae KD.
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
        United KingdomThird generation vs nonuse984.4 (3.0-6.6)
      Second generation† vs nonuse643.0 (1.9-4.5)
      Third generation vs second generation†1.5 (1.0-2.2)
        GermanyThird generation vs nonuse296.7 (3.4-13.0)
      Second generation† vs nonuse883.7 (2.2-5.2)
      Third generation vs second generation†1.8 (1.0-3.3)
       Transnational
      • Lewis MA
      • Heinemann LAJ
      • Mac Rae KD
      • Bruppacher R
      • Spitzer WO
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      The increased risk of venous thrombolism and the use of third generation progestagens: role of bias in observational research.
        United Kingdom and GermanyNorgestimate vs levonorgestrel191.85 (0.95-3.58)
      Desogestrel with 30 μg estrogen vs levonorgestrel641.86 (1.23-2.83)
      Desogestrel with 20 μg estrogen vs levonorgestrel151.57 (0.80-3.07)
      Gestodene vs levonorgestrel581.68 (1.10-2.56)
       Leiden
      • Bloemenkamp KWM
      • Rosendaal FR
      • Helmerhorst FM
      • Buller HR
      • Vandenbroucke JP.
      Enhancement of factor V Leiden mutation of risk of deep-vein thrombosis associated with oral contraceptives containing third-generation progestagen.
      Desogestrel vs nonuse378.7 (3.9-19.3)
      Levonorgestrel vs nonuse203.8 (1.7-8.4)
      Desogestrel vs levonorgestrel2.2 (0.9-5.4)
      Follow-up studies
       General Practice Research DatabaseDesogestrel3029.3 (20.5-41.9)
      Gestodene2228.1 (18.5-42.5)
      Levonorgestrel2316.1 (10.7-24.3)
       Meditel
      • Farmer RDT
      • Lawrenson RA
      • Thompson CR
      • Kennedy JG
      • Hambleton IR.
      Population-based study of risk of various thromboembolism associated with various oral contraceptives.
      Second generation2931.0 (20.8-44.5)
       Levonorgestrel2436.2 (23.2-53.8)
       Second generation other than levonorgestrel518.4 (6.0-42.9)
      Third generation5449.6 (37.3-64.7)
       Desogestrel plus 30 μg estrogen1939.9 (24.0-62.2)
       Desogestrel plus 20 μg estrogen13115.3 (61.4-197.1)
       Gestodene2244.1 (27.5-68.8)
      *Results based on hospital controls. †Includes norgestimate.

      transnational study.

      The Transnational Study,
      • Spitzer WO
      • Lewis MA
      • Heinemann LAJ
      • Thorogood M
      • MacRae KD.
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
      • Spitzer WO
      • Thorogood M
      • Heinemann L.
      Trinational case-control study of oral contraceptives and health.
      funded by the manufacturer of oral contraceptives containing gestodene, was conducted during 1993 through 1995 in the United Kingdom, Germany, and three other European countries with methods similar to those used in the WHO study. The study was halted after publication of the WHO results.
      On the basis of 283 cases of venous thromboembolism from the United Kingdom, the relative risk estimate was significantly elevated, 4.4-fold, for current use of third-generation oral contraceptives containing gestodene or desogestrel and threefold for use of second-generation oral contraceptives, relative to nonuse (Table I). For comparability with the WHO analysis,
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      oral contraceptives containing norgestimate (18 cases and 28 controls) were included with second-generation oral contraceptives. In a comparison of third-generation with second-generation oral contraceptives, the relative risk estimate was 1.5 (95% confidence interval 1.0 to 2.2). Allowance was made for age, alcohol use, cigarette smoking, study center, body mass index, and the duration of oral contraceptive use. Results from Germany were higher but showed a similar pattern. Results obtained separately with hospital and community controls were similar.
      In a further analysis,
      • Lewis MA
      • Heinemann LAJ
      • Mac Rae KD
      • Bruppacher R
      • Spitzer WO
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      The increased risk of venous thrombolism and the use of third generation progestagens: role of bias in observational research.
      individual formulations were compared separately with oral contraceptives containing levonorgestrel; in these comparisons, users of norgestimate were no longer included with levonorgestrel. The relative risk estimate, ranging from 1.57 to 1.86, was increased for every third-generation oral contraceptive formulation, including preparations containing norgestimate (Table I). It was also elevated for second-generation oral contraceptives containing progestagens other than levonorgestrel, 1.56, and for first-generation oral contraceptives, 1.94. In a subanalysis conducted among women aged 25 to 44 years, the relative risk was greater the more recently the formulation had been introduced (p < 0.001 for trend). Relative to oral contraceptives containing levonorgestrel, which were first marketed in early 1970s, the relative risk estimate increased from 1.46 for oral contraceptives containing desogestrel and 30 μg of estrogen, marketed in 1981, to 2.84 for oral contraceptives containing desogestrel and 20 μg of estrogen, introduced in 1988 in the United Kingdom and in 1992 in Germany.

      general practice research database.

      The General Practice Research Database contains computerized clinical records from general practices in the United Kingdom.
      • Jick H
      • Jick SS
      • Gurewich V
      • Myers MW
      • Vasilakis C.
      Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptives with differing progestagen components.
      In a study of idiopathic nonfatal venous thromboembolism, 80 women with evidence of treatment with anticoagulants and admission to the hospital were identified. The incidence of venous thromboembolism per 100,000 woman years at risk was estimated to be 16.1 for levonorgestrel users, 29.3 for desogestrel users, and 28.1 for gestodene users (Table I). In a nested case-control study carried out within this study, in which calendar year, smoking status, and body mass index were controlled together with two broad categories of age, the relative risk estimates for desogestrel use and for gestodene use relative to levonorgestrel use were both elevated twofold.

      meditel studies

      The Meditel Studies used a computerized database derived from general practice records. In one study 116 cases of venous thromboembolism with evidence of specific treatment (e.g., anticoagulants) or admission to the hospital with the diagnosis of pulmonary embolism were identified.
      • Farmer RDT
      • Preston TD.
      The risk of venous thromboembolism associated with low oestrogen oral contraceptives.
      The incidence of venous thromboembolism per 100,000 woman years at risk was estimated to be 11.4 in nonusers of oral contraceptives, 30.5 in users of combined oral contraceptives containing 30 to 35 μg of estrogen, 30.3 in users of progestogen-only oral contraceptives, and 59.1 in pregnant women. In a further report on formulation-specific risks,
      • Farmer R.
      Safety of modern oral contraceptives [letter].
      the incidence of venous thromboembolism per 100,000 woman years at risk was estimated to be 34.6 for women who used a formulation containing desogestrel (Marvelon); that estimate did not differ significantly from estimates for formulations containing levonorgestrel (Microgynon, Ovranette, Logynon) or gestodene (Femodene), but numbers were small.
      In a subsequent study based on 83 cases of venous thromboembolism,
      • Farmer RDT
      • Lawrenson RA
      • Thompson CR
      • Kennedy JG
      • Hambleton IR.
      Population-based study of risk of various thromboembolism associated with various oral contraceptives.
      the incidence of venous thromboembolism was about 60% greater in users of third-generation progestogens relative to second-generation progestogens, and the highest incidence was observed for users of desogestrel plus 20 μg of ethinyl estradiol (Table I). In a nested case-control study that controlled for exact year of age and several other confounding factors, the relative risk for the comparison of desogestrel plus 20 μg of ethinyl estradiol to second-generation oral contraceptives was 3.49 (95% confidence interval 1.21 to 10.12), and the estimate for other third-generation formulations was 1.18 (95% confidence interval 0.66 to 2.17).

      leiden thrombophilia study

      The Leiden Thrombophilia Study was a population-based study of first episodes of venous thrombosis conducted during 1988 through 1992 in the Netherlands.
      • Bloemenkamp KWM
      • Rosendaal FR
      • Helmerhorst FM
      • Buller HR
      • Vandenbroucke JP.
      Enhancement of factor V Leiden mutation of risk of deep-vein thrombosis associated with oral contraceptives containing third-generation progestagen.
      Each case identified a friend or acquaintance, or the partner of another case was chosen, matched on sex and age to the case. Earlier findings from this study were that factor V Leiden mutation was present in 6% of control women and that the risk of venous thrombosis among carriers was increased eightfold overall and thirtyfold among carriers who were receiving oral contraceptives.
      • Vandenbroucke JP
      • Koster T
      • Brier E
      • Reitsma PH
      • Bertina RM
      • Rosendaal FR.
      Increased risk of venous thrombosis in oral contraceptive users who are carriers of factor V Leiden mutation.
      Analysis of specific formulations, prompted by the results of the WHO study,
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      was based on 126 female cases and 159 female controls.
      • Bloemenkamp KWM
      • Rosendaal FR
      • Helmerhorst FM
      • Buller HR
      • Vandenbroucke JP.
      Enhancement of factor V Leiden mutation of risk of deep-vein thrombosis associated with oral contraceptives containing third-generation progestagen.
      For current oral contraceptive use relative to nonuse, the age-adjusted relative risk for formulations containing 30 μg of ethinyl estradiol together with desogestrel was 8.7; for formulations that contained levonorgestrel, lynestrenol, or norethisterone, the estimates ranged from 2.2 to 3.8. A comparison of desogestrel-containing formulations with levonorgestrel-containing formulations that both contained 30 μg of ethinyl estradiol yielded a relative risk of 2.2 (Table I). Adjustment for family history, factor V Leiden mutation, or history of pregnancy did not change the relative risk estimates.

      other studies

      Several earlier studies provided data on second-generation oral contraceptives but had no information on third-generation oral contraceptives.
      • Böttiger LE
      • Boman G
      • Eklund G
      • Westerholm B
      Oral contraceptives and thromboembolic disease: effects of lowering oestrogen content.
      • Vessey M
      • Mant D
      • Smith A
      • Yeates D.
      Oral contraceptives and venous thromboembolism: findings in a large prospective study.
      • Helmrich SP
      • Rosenberg L
      • Kaufman DW
      • Strom B
      • Shapiro S.
      Venous thromboembolism and oral contraceptive use.
      • Gerstman BB
      • Piper JM
      • Tomita DK
      • Ferguson WJ
      • Stadel BV
      • Lundin FE.
      Oral contraceptive estrogen dose and the risk of deep venous thromboembolic disease.
      • Thorogood M
      • Mann J
      • Murphy M
      • Vessey M.
      Risk factors for fatal venous thromboembolism in young women: a case-control study.
      Most suggest that the risk of venous thromboembolism is lower for oral contraceptives containing lower doses of estrogen.

      Myocardial infarction

      transnational study

      The Transnational Study of myocardial infarction compared 153 women with first myocardial infarctions, of whom 129 were from the United Kingdom or Germany, with 210 hospital and 288 community controls.
      • Lewis MA
      • Spitzer WO
      • Heinemann LAJ
      • MacRae KD
      • Bruppacher R
      • Thorogood M
      • et al.
      Third generation oral contraceptives and risk of myocardial infarction: an international case-control study.
      The methods were the same as those in the study of venous thromboembolism.
      • Spitzer WO
      • Lewis MA
      • Heinemann LAJ
      • Thorogood M
      • MacRae KD.
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
      With control for the duration of oral contraceptive use and cigarette smoking, the relative risk estimate for current use of third-generation oral contraceptives containing desogestrel or gestodene relative to nonuse was close to 1.0; for second-generation oral contraceptives relative to nonuse, the relative risk was significantly elevated, 3.1 (Table II). For the comparison of third-generation to second-generation use, the relative risk estimate was 0.36 (not statistically significant). In an updated report based on a larger sample,
      • Lewis MA
      • Heinemann LAJ
      • Spitzer WO
      • Bruppacher R
      • Thorogood M
      • MacRae KD.
      Results of the Transitional Case-Control Study on Oral Contraceptives.
      the relative risk estimate for third-generation users (7 cases) versus second-generation users (28 cases) was 0.3 (95% confidence interval 0.1 to 1.0).
      Table IIResults on myocardial infarction and stroke from recent studies
      StudyOutcomeComparisonExposed cases (No.)Relative risk estimate and 95% confidence interval
      Transnational
      • Lewis MA
      • Spitzer WO
      • Heinemann LAJ
      • MacRae KD
      • Bruppacher R
      • Thorogood M
      • et al.
      Third generation oral contraceptives and risk of myocardial infarction: an international case-control study.
      Myocardial infarctionThird generation vs nonuse61.1  (0.4-3.4)
      Second generation* vs nonuse233.1  (1.5-6.3)
      Third generation vs second generation*0.36 (0.1-1.2)
      WHO (Europe)
      • Poulter NR
      • Chang CL
      • Farley TM
      • Meirik O
      • Marmot MG.
      Ischaemic stroke and combined oral contraceptive: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      Ischemic strokeSecond and third generations vs nonuse201.53 (0.71-3.31)
      Second generation vs nonuse161.53 (0.69-3.39)
      Third generation vs nonuse41.76 (0.33-9.36)
      WHO (Europe)
      • Poulter NR
      • Chang CL
      • Farley TM
      • Meirik O
      • Marmot MG.
      Haemorrhagic stroke, overall stroke risk, and combined oral contraceptives: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      Hemorrhagic strokeSecond and third generations vs nonuse271.27 (0.70-2.32)
      Third generation vs second generationNo significant difference;  data not given
      *Includes norgestimate. †Formulations with <50 μg of estrogen.

      other studies

      In a study of myocardial infarction based on the General Practice Research Database,
      • Jick H
      • Jick SS
      • Meyers MW
      • Vasilakis C.
      Risk of acute myocardial infarction and low-dose combined oral contraceptives [letter].
      the relative risk estimates for the comparison of current use of oral contraceptives containing desogestrel or gestodene to use of oral contraceptives with levonorgestrel did not differ significantly, but the number of users was very small. A recent U.S. population-based case-control study of myocardial infarction
      • Sidney S
      • Petitti DB
      • Quesenberry Jr, CP
      • Klatsky AL
      • Wolf S.
      Myocardial infarction in users of low dose oral contraceptives.
      found that current use of oral contraceptives containing <50 μg of estrogen was related to a smaller increase in risk than higher dose formulations—relative risk 1.65 and not statistically significant. There was no information on the relative effects of second- and third-generation formulations. Earlier studies
      • Adam SA
      • Thorogood M
      • Mann JI.
      Oral contraception and myocardial infarction revisited: the effects of new preparations and prescribing patterns.
      • Thorogood M
      • Mann J
      • Murphy M
      • Vessey M.
      Is oral contraceptive use still associated with an increased risk of fatal myocardial infarction?.
      • Rosenberg L
      • Palmer JR
      • Lesko SM
      • Shapiro S.
      Oral contraceptive use and the risk of myocardial infarction.
      also suggested a lower risk for formulations with lower doses of estrogen.

      Stroke

      who collaborative study

      The WHO Collaborative Study assessed ischemic stroke (697 cases, 1962 controls) in its hospital-based study of cardiovascular disease.
      • Poulter NR
      • Chang CL
      • Farley TM
      • Meirik O
      • Marmot MG.
      Ischaemic stroke and combined oral contraceptive: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      The overall relative risk for current oral contraception users was increased about threefold both in Europe and in the developing countries, but estimates were lower for younger women and for formulations with lower doses of estrogen. In Europe the relative risk estimate for current use of formulations with <50 μg of estrogen was 1.54 and not statistically significant; the estimates were similar for second- and third-generation progestogens, but numbers were small (Table II). The risk was increased about fivefold or more for women who both smoked and used oral contraceptives and more than tenfold for users who gave a history of hypertension.
      The WHO Collaborative Study also assessed hemorrhagic stroke (1068 cases, 2910 controls), intracerebral hemorrhage (420 cases, 1173 controls), and subarachnoid hemorrhage (608 cases, 1644 controls).
      • Poulter NR
      • Chang CL
      • Farley TM
      • Meirik O
      • Marmot MG.
      Haemorrhagic stroke, overall stroke risk, and combined oral contraceptives: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      The overall relative risk for hemorrhagic stroke was <2 in both Europe and the developing countries, and the estimate for Europe, 1.38, was not statistically significant (Table II). Relative risk estimates were lower for younger women. Risk was considerably increased among users who smoked, about threefold, and among users with hypertension, more than tenfold. Results for intracerebral and subarachnoid hemorrhage were similar. It was reported that there were no significant differences according to the type of progestogen used, but no data were shown.

      other studies

      A U.S. population-based case-control study of strokes occurring during 3.8 million women-years of observation found no increase in risk of ischemic stroke among current users of oral contraceptives (relative risk 1.18).
      • Petitti DB
      • Sidney S
      • Bernstein A
      • Wolf S
      • Queensberry C
      • Ziel HK.
      Stroke in users of low-dose oral contraceptives.
      For hemorrhagic stroke the overall relative risk estimate was also close to 1.0, but there was a suggestion of an interaction to increase the risk by about 3.5-fold among smokers. Virtually all oral contraceptive use was of low estrogen dose (<50 μg) preparations. There was no information on the relative safety of second- and third-generation formulations. An earlier study of stroke also suggested a lower risk of stroke for users of lower estrogen dose oral contraceptive preparations.
      • Lidegaard O.
      Oral contraception and risk of cerebral thromboembolic attack: results of a case-control study.

      Pharmacologic effects and biologic mechanisms

      In an attempt to reduce adverse effects, oral contraceptive manufacturers have lowered estrogen doses to about one fifth of their original content and have reduced progestogen doses about tenfold.
      • Newton JR.
      Classification and comparison of oral contraceptives containing new generation progestogens.
      • Fotherby K
      • Caldwell ADS.
      New progestogens in oral contraception.
      The search for new progestogens has focused on developing compounds that are less androgenic and that have little or no impact on lipid metabolism or glucose tolerance. Indeed, clinical studies indicate that each generation of combined oral contraceptives has had smaller effects on lipid metabolism and glucose tolerance than the previous generation.
      • Robinson GE.
      Low-dose combined oral contraceptives.
      • Fotherby K
      • Caldwell ADS.
      New progestogens in oral contraception.
      • Speroff L
      • De Cherney A
      • Advisory Board for the New Progestins
      Evaluation of a new generation of oral contraceptives.
      • Godsland IF
      • Crook D
      • Simpson R
      • Proudler T
      • Felton C
      • Lees B
      • et al.
      The effect of different formulations of oral contraceptives on lipid and carbohydrate metabolism.
      Second- and third-generation oral contraceptives have small effects on the hemostatic system.
      • Newton JR.
      Classification and comparison of oral contraceptives containing new generation progestogens.
      • Fotherby K
      • Caldwell ADS.
      New progestogens in oral contraception.
      • Speroff L
      • De Cherney A
      • Advisory Board for the New Progestins
      Evaluation of a new generation of oral contraceptives.
      They tend to increase the tendency to coagulation, but they also increase fibrinolysis. Effects on the natural anticoagulation system vary, but the general tendency is to a reduction in factors that inhibit coagulation. Even after the changes, values of the hemostatic variables tend to be within normal ranges. These effects appear to be related to the estrogen dose and are smaller than those of older oral contraceptives that had higher doses of estrogen. It is not known how the changes relate to clinical events.
      In a pharmacokinetics study of 22 women published in 1989
      • Jung-Hoffmann C
      • Kuhl H.
      Interaction with the pharmacokinetics of ethinyl estradiol and progestogens contained in oral contraceptives.
      serum levels of ethinyl estradiol were considerably higher in gestodene users than in desogestrel users. Because adverse outcomes such as hypertension and venous thromboembolism have been related to the estrogen content, the results raised the concern that oral contraceptives with gestodene might carry a greater risk. The German regulatory authorities issued an alert, which was followed by the initiation of the Transnational Study. Subsequent studies have not detected differences in serum levels of ethinyl estradiol between women using gestodene- and desogestrel-containing oral contraceptives.
      • Himpel M
      • Tauber U
      • Kuhnz W
      • Pfeffer M
      • Brill K
      • Heitbecker R
      • et al.
      Comparison of serum ethinyl estradiol, sex-hormone-binding globulin, corticoid-binding globulin and cortisol levels in women using two low-dose combined oral contraceptives.
      • Kuhnz W.
      Pharmacokinetics of the contraceptive steroids levonorgestrel and gestodene after single and multiple oral administration to women.
      • Kuhnz W
      • Sostarek D
      • Gansau C
      • Louton T
      • Mahler M.
      Single and multiple administration of a new triphasic oral contraceptive to women: pharmacokinetics of ethinyl estradiol and free and total testosterone levels in serum.
      • Back DJ
      • Ward S
      • Orme MLE.
      Recent pharmacokenitic studies of low-dose oral contraceptives.
      • Dibbelt L
      • Knuppen R
      • Jütting G
      • Heinmann S
      • Klipping CO
      • Parikka-Olexik H
      Group comparison of serum ethinyl estradiol, SHBG and CBG levels in 83 women using two low-dose combination oral contraceptives for three months.
      • Orme M
      • Back DJ
      • Ward S
      • Green S.
      The pharmacokinetics of ethynylestradiol in the presence and absence of gestodene and desogestrel.

      Comment

      The recent evidence on myocardial infarction and stroke indicates that second- and third-generation oral contraceptives are safer than first-generation formulations. Formulations with <50 μg of estrogen had little or no influence on the risk of stroke in women without risk factors.
      • Poulter NR
      • Chang CL
      • Farley TM
      • Meirik O
      • Marmot MG.
      Ischaemic stroke and combined oral contraceptive: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      • Poulter NR
      • Chang CL
      • Farley TM
      • Meirik O
      • Marmot MG.
      Haemorrhagic stroke, overall stroke risk, and combined oral contraceptives: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      • Petitti DB
      • Sidney S
      • Bernstein A
      • Wolf S
      • Queensberry C
      • Ziel HK.
      Stroke in users of low-dose oral contraceptives.
      One study of myocardial infarction suggests an increase in risk associated with second-generation formulations and no increase for third-generation pills,
      • Lewis MA
      • Spitzer WO
      • Heinemann LAJ
      • MacRae KD
      • Bruppacher R
      • Thorogood M
      • et al.
      Third generation oral contraceptives and risk of myocardial infarction: an international case-control study.
      • Lewis MA
      • Heinemann LAJ
      • Spitzer WO
      • Bruppacher R
      • Thorogood M
      • MacRae KD.
      Results of the Transitional Case-Control Study on Oral Contraceptives.
      but the latter finding requires confirmation. There is no obvious biologic rationale for third-generation formulations to affect the risk of myocardial infarction less than second-generation pills do. Although third-generation formulations are associated with fewer adverse effects on lipid profiles, the epidemiologic data suggest that oral contraceptives affect risk through thrombosis rather than through atherosclerosis.
      With regard to venous thromboembolism, relative risk estimates in recent studies
      • Poulter NR
      • Chang CL
      • Farley TMM
      • Meirik O
      • Marmot MG.
      Venous thromboembolic disease and combined oral contraceptives: results of international multicenter case-control study, World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      • Jick H
      • Jick SS
      • Gurewich V
      • Myers MW
      • Vasilakis C.
      Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptives with differing progestagen components.
      • Spitzer WO
      • Lewis MA
      • Heinemann LAJ
      • Thorogood M
      • MacRae KD.
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
      • Lewis MA
      • Heinemann LAJ
      • Mac Rae KD
      • Bruppacher R
      • Spitzer WO
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      The increased risk of venous thrombolism and the use of third generation progestagens: role of bias in observational research.
      • Bloemenkamp KWM
      • Rosendaal FR
      • Helmerhorst FM
      • Buller HR
      • Vandenbroucke JP.
      Enhancement of factor V Leiden mutation of risk of deep-vein thrombosis associated with oral contraceptives containing third-generation progestagen.
      • Farmer RDT
      • Lawrenson RA
      • Thompson CR
      • Kennedy JG
      • Hambleton IR.
      Population-based study of risk of various thromboembolism associated with various oral contraceptives.
      • Farmer RDT
      • Preston TD.
      The risk of venous thromboembolism associated with low oestrogen oral contraceptives.
      • Farmer R.
      Safety of modern oral contraceptives [letter].
      have been similar to or smaller than those for first-generation pills, and the underlying incidence of venous thromboembolism was lower than in the early studies
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      (see below). Thus modern pills appear to cause fewer cases of venous thromboembolism than earlier formulations did.
      Some studies suggest that third-generation oral contraceptives are associated with an increase in the risk of venous thromboembolism about 1.5-fold to twofold greater than that associated with second-generation formulations. Does the evidence support a causal explanation?
      There is no convincing evidence that third-generation oral contraceptives containing gestodene or desogestrel affect hemostasis or other biologic systems in a way that would increase the risk of thrombosis relative to other formulations.
      • Newton JR.
      Classification and comparison of oral contraceptives containing new generation progestogens.
      • Fotherby K
      • Caldwell ADS.
      New progestogens in oral contraception.
      • Speroff L
      • De Cherney A
      • Advisory Board for the New Progestins
      Evaluation of a new generation of oral contraceptives.
      Nor is there convincing evidence for a greater increase in risk associated with desogestrel formulations containing 20 μg of estrogen than those containing 30 μg of estrogen, as has been observed in several studies.
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      • Lewis MA
      • Heinemann LAJ
      • Mac Rae KD
      • Bruppacher R
      • Spitzer WO
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      The increased risk of venous thrombolism and the use of third generation progestagens: role of bias in observational research.
      • Farmer RDT
      • Lawrenson RA
      • Thompson CR
      • Kennedy JG
      • Hambleton IR.
      Population-based study of risk of various thromboembolism associated with various oral contraceptives.
      In addition, the relative risk estimates for the comparison of third- to second-generation oral contraceptive users have been of a magnitude that might well be explained by biases. Thus it is necessary to assess whether there is evidence of biases that may have influenced the findings.
      Problems with diagnosis are inherent in any study of venous thromboembolism. The condition manifests with a variety of symptoms and may resolve spontaneously. Oral contraceptive users who have certain symptoms, such as swelling in the leg, may be more often referred and more rigorously worked up for venous thromboembolism than nonusers are.
      • Realini JP
      • Goldzieher JW.
      Oral contraceptives and cardiovascular disease: a critique of epidemiologic studies.
      • Heinemann LAJ
      • Lewis MA
      • Assmann A
      • Cravens L
      • Guggenmoos-Holzmann I.
      Could preferential prescribing and referral behavior of physicians explain the elevated thrombotic risk found to be associated with third generation oral contraceptives?.
      This would lead to a tendency to overestimate the incidence of venous thromboembolism for oral contraceptive users relative to nonusers and also to overestimate the risk attributable to oral contraceptive use. The results based on the German data in the Transnational Study are compatible with this bias: The relative risk estimates were higher than those based on the United Kingdom data, possibly because of earlier widespread media coverage in Germany of a study that suggested adverse pharmacokinetics of oral contraceptives containing gestodene.
      • Jung-Hoffmann C
      • Kuhl H.
      Interaction with the pharmacokinetics of ethinyl estradiol and progestogens contained in oral contraceptives.
      If the diagnosis had been related to the specific oral contraceptive formulation used, comparisons between formulations could also have been biased. A survey of physicians in Germany,
      • Heinemann LAJ
      • Lewis MA
      • Assmann A
      • Cravens L
      • Guggenmoos-Holzmann I.
      Could preferential prescribing and referral behavior of physicians explain the elevated thrombotic risk found to be associated with third generation oral contraceptives?.
      conducted after the issuance of the advisory of the United Kingdom Committee on Safety of Medicines, suggested that they had been more likely to investigate the symptoms of women perceived to be at higher risk, because of obesity or family history of thrombosis, for example. The physicians also reported that they were more likely to prescribe third-generation oral contraceptives to such women. Thus third-generation users may have had a higher likelihood of referral for investigation of symptoms of thrombosis.
      The comparison between second- and third-generation oral contraceptive use would have been biased in all of the studies if users of second- and third-generation oral contraceptives differed in their susceptibility to venous thromboembolism. Such a difference could have occurred by two mechanisms. One involves the depletion of susceptible women from the pool of users. Women who have a genetic susceptibility to venous thromboembolism, such as those positive for factor V Leiden mutation, may have venous thromboembolism soon after starting oral contraceptive use and they will be removed from the pool of users. More susceptible women may have been removed from among users of a formulation long on the market than from among users of a newer formulation. Another mechanism is differential prescribing. If particular oral contraceptives are perceived as being safer (third-generation oral contraceptives were promoted as being safer), women thought to be at greater risk of venous thromboembolism may be selectively given those formulations.
      Evidence for a greater attrition of susceptible women from among users of oral contraceptives longer on the market is found in the Transnational Study. In the analysis of individual formulations relative to formulations with levonorgestrel, which was on the market longest,
      • Lewis MA
      • Heinemann LAJ
      • Mac Rae KD
      • Bruppacher R
      • Spitzer WO
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      The increased risk of venous thrombolism and the use of third generation progestagens: role of bias in observational research.
      the relative risk estimate for oral contraceptives containing norgestimate was elevated, and similar in magnitude, to that for desogestrel and gestodene. Because norgestimate is converted in part to levonorgestrel and to its metabolites,
      • Stanczyk FZ
      • Roy S.
      Metabolism of levonorgestrel, norethindrone and structurally related contraceptive steroids.
      • Shenfield GM
      • Griffin JM.
      Clinical pharmacokinetics of contraceptive steroids: an update.
      there is no compelling biologic credibility for a difference in risk between formulations with levonorgestrel and norgestimate. In addition, among women aged 25 to 44 years, the magnitude of the relative risk estimate was greater the more recently the formulation had been marketed, a pattern consistent with greater attrition of women susceptible to venous thromboembolism among users of formulations that were in use longer. In accord with this, relative risk estimates in the WHO Study
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      and the most recent Meditel Study
      • Farmer RDT
      • Lawrenson RA
      • Thompson CR
      • Kennedy JG
      • Hambleton IR.
      Population-based study of risk of various thromboembolism associated with various oral contraceptives.
      were greater for women who used desogestrel with 20 μg of ethinyl estradiol than for those who used this progestagen with 30 μg of ethinyl estradiol. The duration of use of third-generation oral contraceptives was shorter than that of second-generation oral contraceptives,
      • Spitzer WO
      • Lewis MA
      • Heinemann LAJ
      • Thorogood M
      • MacRae KD.
      • Transnational Research Group on Oral Contraceptives and the Health of Young Women
      Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
      which is also compatible with differential attrition of susceptible women.
      Evidence suggestive of differential prescribing is found in surveys of prescribers. Surveys of British,
      • Dunn N
      • Heinemann LAJ
      • Mann RD.
      Are third-generation oral contraceptives really more thrombotic than second-generation: some potential biases examined and compared in German and English doctors.
      German,
      • Heinemann LAJ
      • Lewis MA
      • Assmann A
      • Cravens L
      • Guggenmoos-Holzmann I.
      Could preferential prescribing and referral behavior of physicians explain the elevated thrombotic risk found to be associated with third generation oral contraceptives?.
      • Dunn N
      • Heinemann LAJ
      • Mann RD.
      Are third-generation oral contraceptives really more thrombotic than second-generation: some potential biases examined and compared in German and English doctors.
      and French
      • Jamin C
      • de Mouzon J.
      Selective prescribing of third-generation oral contraceptives (OCs).
      physicians indicated a tendency to prescribe third-generation formulations for women at higher risk. Similar results have also been reported for Dutch prescribers.
      • Herings RMC
      • de Boer A
      • Urquliart J
      • Leufkens HGM.
      Non-causal explanations for the increased risk of venous thrombembolism among users of third-generation oral contraceptives.
      If this type of preferential prescribing had operated in the recent epidemiologic studies, the result would have been an upward bias in the relative risk estimate comparing third- with second-generation oral contraceptive users.
      In summary, there is evidence of biases that may have contributed to the apparently greater risk of venous thromboembolism in third-generation users compared with second-generation users.

      Risks and benefits of second- and third-generation oral contraceptives

      Cardiovascular disease rarely occurs in young women. On the basis of data from the Oxford component of the WHO study, the incidence of idiopathic venous thromboembolism per 100,000 woman years at risk was estimated to be 3.9 for nonusers, 10.5 for users of oral contraceptives containing levonorgestrel, 21.3 for users of formulations containing desogestrel or gestodene, and 12.3 for users of other oral contraceptives.
      • Farley TMM
      • Meirik O
      • Chang CL
      • Marmot MG
      • Poulter NR.
      Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
      Estimates based on the Meditel database
      • Farmer RDT
      • Preston TD.
      The risk of venous thromboembolism associated with low oestrogen oral contraceptives.
      were higher (e.g., 50% greater for users of oral contraceptives with levonorgestrel), but both the WHO and Meditel estimates are an order of magnitude lower than the estimates from epidemiologic studies conducted in the 1960s and 1970s.
      • Stadel BV.
      Oral contraceptives and cardiovascular disease.
      Some of the differences between the early and recent estimates may be explained by more precise diagnosis of venous thromboembolism in the recent studies. Whatever the explanation, the data suggest that the incidence of venous thromboembolism attributable to use of modern oral contraceptives is less than that associated with first-generation formulations. The incidence of venous thromboembolism in pregnant women has been estimated to be about 60 per 100,000.
      • Farmer RDT
      • Preston TD.
      The risk of venous thromboembolism associated with low oestrogen oral contraceptives.
      Thus use of second- or third-generation oral contraceptives carries less risk of venous thromboembolism than pregnancy does. The incidence of stroke per 100,000 women per year, based on European data in the WHO Study, was estimated to be 4.8 for nonusers and 6.7 for users of lower estrogen dose oral contraceptives; in a U.S. study the overall incidence was about 11 per 100,000 women per year, and it differed little according to oral contraceptive use. The incidence of myocardial infarction estimated from a recent U.S. study was about 5 per 100,000 woman years.
      • Sidney S
      • Petitti DB
      • Quesenberry Jr, CP
      • Klatsky AL
      • Wolf S.
      Myocardial infarction in users of low dose oral contraceptives.
      With regard to mortality, deaths from myocardial infarction among women <45 years old exceed those from venous thromboembolism and stroke.
      • Newton JR.
      Classification and comparison of oral contraceptives containing new generation progestogens.
      • MacRae K
      • Kay C.
      Third generation oral contraceptive pills: is the scare over the increased risk of thrombosis justified?.
      Thus the relative safety of second- and third-generation oral contraceptives with reference to mortality depends on their effect on mortality from myocardial infarction. The Transnational Study results suggest that third-generation oral contraceptives are safer than second-generation formulations in terms of the incidence of myocardial infarction, but they require confirmation.
      • Lewis MA
      • Heinemann LAJ
      • Spitzer WO
      • Bruppacher R
      • Thorogood M
      • MacRae KD.
      Results of the Transitional Case-Control Study on Oral Contraceptives.
      There are no informative data available on mortality.
      Aside from the prevention of pregnancy and its attendant risks, oral contraceptives containing ≥50 μg of estrogen had several unanticipated benefits, the most important being reductions in the incidence of ovarian and endometrial cancer. An analysis based on data from the Oxford Family Planning Association follow-up study of oral contraceptive users and nonusers indicated that the benefits in decreased morbidity and mortality outweighed the increased risks associated with use.
      • Vessey MP.
      Benefits and risks of combined oral contraceptives.
      The benefit-risk equation for second- and third-generation oral contraceptives is not established but, as already noted, there is evidence that these formulations are safer with respect to cardiovascular disease than earlier oral contraceptives were.

      Acknowledgements

      We thank Drs. Daniel Cramer, William Brown, Sonia Buist, Fritz Kernper, and Walter Spitzer for their comments.

      References

        • Stadel BV.
        Oral contraceptives and cardiovascular disease.
        N Engl J Med. 1981; 305 (672-7): 612-617
        • Robinson GE.
        Low-dose combined oral contraceptives.
        Br J Obstet Gynaecol. 1994; 101: 1036-1041
        • Newton JR.
        Classification and comparison of oral contraceptives containing new generation progestogens.
        Hum Reprod Update. 1995; 1: 231-263
        • Poulter NR
        • Chang CL
        • Farley TMM
        • Meirik O
        • Marmot MG.
        Venous thromboembolic disease and combined oral contraceptives: results of international multicenter case-control study, World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
        Lancet. 1995; 346: 1575-1582
        • Farley TMM
        • Meirik O
        • Chang CL
        • Marmot MG
        • Poulter NR.
        Effect of different progestagens in low oestrogen oral contraceptives on venous thromboembolic disease.
        Lancet. 1995; 346: 1582-1588
        • Jick H
        • Jick SS
        • Gurewich V
        • Myers MW
        • Vasilakis C.
        Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptives with differing progestagen components.
        Lancet. 1995; 346: 1589-1593
        • Spitzer WO
        • Lewis MA
        • Heinemann LAJ
        • Thorogood M
        • MacRae KD.
        • Transnational Research Group on Oral Contraceptives and the Health of Young Women
        Third generation oral contraceptives and risk of venous thrombolic disorders: an international case-control study.
        BMJ. 1996; 312: 83-88
        • Anonymous
        Pill scares and public responsibility [editorial].
        Lancet. 1996; 347: 1707
        • Graham A
        • Grieve F
        • Davie J
        • Glasier A.
        Oral contraceptives and venous thromboembolism [letter].
        Lancet. 1995; 346: 1430
        • British Pregnancy Advisory Service
        Number of abortions; comparison over 3 months: December-February 1993/94 to 1995/96.
        The Service, West Midlands (UK)1996
      1. Huge rise in abortions after government's warning: human cost of the pill scare. Daily Mail 1996 Apr 15.

        • Food and Drug Administration
        Oral contraceptives and risk of blood clots.
        FDA Talk Paper. 1995 Nov 14;
        • Johannisson E
        • International Committee for Research in Reproduction
        Safety of modern oral contraceptives.
        Lancet. 1996; 347: 258
        • MacRae K
        • Kay C.
        Third generation oral contraceptive pills: is the scare over the increased risk of thrombosis justified?.
        BMJ. 1995; 311: 1112
        • McPherson K
        Third generation oral contraception and venous thromboembolism: the published evidence confirms the Committee on Safety of Medicine's concerns.
        BMJ. 1996; 312: 68-69
        • Planned Parenthood Federation of America
        Update: thromboembolic impact of oral contraceptives.
        Planned Parenthood Federation of America, New York1996
        • Vessey MP.
        Female hormones and vascular disease: epidemiologic overview.
        Br J Fam Plann. 1980; 6: 1-12
        • Oral contraceptives, venous thrombosis, and varicose veins
        Royal College of General Practitioners' Oral Contraception Study.
        J R Coll Gen Pract. 1978; 28: 393-399
        • Stolley PD
        • Tonascia JA
        • Tockman MS
        • Sartwell PE
        • Rutledge AH
        • Jacobs MP.
        Thrombosis and low-estrogen oral contraceptives.
        Am J Epidemiol. 1975; 102: 197-208
        • Inman WHW
        • Vessey MP
        • Westerholm B
        • Engelund A.
        Thromboembolic diseases and the steroidal content of oral contraceptives: a report to the Committee on Safety of Drugs.
        BMJ. 1970; 2: 203-209
        • Meade TW
        • Greenberg G
        • Thompson SG
        Progestogens and cardiovascular reactions associated with oral contraceptives and a comparison of the safety of 50- and 30-μg oestrogen preparations.
        BMJ. 1980; 280: 1157-1161
        • Slone D
        • Shapiro S
        • Kaufman DW
        • Rosenberg L
        • Miettinen OS
        • Stolley PD.
        Risk of myocardial infarction in relation to current and discontinued use of oral contraceptives.
        N Engl J Med. 1981; 305: 420-424
        • Rosenberg L
        • Palmer JR
        • Lesko SM
        • Shapiro S.
        Oral contraceptive use and the risk of myocardial infarction.
        Am J Epidemiol. 1990; 131: 1009-1016
        • Stampfer MJ
        • Willett WC
        • Colditz CA
        • Speizer FE
        • Hennekens CH.
        A prospective study of past use of oral contraceptive agents and risk of cardiovascular disease.
        N Engl J Med. 1988; 319: 1313-1317
        • Petitti DB
        • Wingerd J.
        Use of oral contraceptives, cigarette smoking, and risk of subarachnoid haemorrhage.
        Lancet. 1978; 2: 234-235
        • Mortality among oral-contraceptive users
        Royal College of General Practitioners' Oral Contraception Study.
        Lancet. 1977; 2: 727-731
        • Vessey MP
        Steroid contraception, venous thromboembolism, and stroke: data from countries other than the United States.
        in: Risks, benefits, and controversies in fertility control. Harper & Row, Hagerstown (MD)1978: 113-121
        • Maguire MG
        • Tonascia J
        • Sartwell PE
        • Stolley PD
        • Tockman MS.
        Increased risk of thrombosis due to oral contraceptives: a further report.
        Am J Epidemiol. 1979; 110: 188-195
        • Mann JI
        • Vessey MP
        • Thorogood M
        • Doll R.
        Myocardial infarction in young women with special reference to oral contraceptive practice.
        BMJ. 1975; 2: 241-245
        • Mann JI
        • Inman WHW.
        Oral contraceptives and death from myocardial infarction.
        BMJ. 1975; 2: 245-248
        • Mann JI
        • Inman WHW
        • Thorogood M.
        Oral contraceptive use in older women and fatal myocardial infarction.
        BMJ. 1976; 2: 445-447
        • Royal College of General Practitioners
        Oral contraceptives and health.
        Pitman, London1974
        • Vessey MP
        • McPherson K
        • Yeates D.
        Mortality in oral contraceptive users.
        Lancet. 1981; 1: 549-550
        • Engel HJ
        • Engel E
        • Lichtlen PR
        Coronary atherosclerosis and myocardial infarction in young women—role of oral contraceptives.
        Eur Heart J. 1983; 4: 1-8
        • Wynn V
        • Adams PW
        • Godsland I
        • Melrose J
        • Niththyananthan R
        • Oakley NW
        • et al.
        Comparison of effects of different combined oral contraceptive formulations on carbohydrate and lipid metabolism.
        Lancet. 1979; 1: 1045-1049
        • Tikkanen MJ
        • Nikkila EA.
        Oral contraceptives and lipoprotein metabolism.
        J Reprod Med. 1986; 31: 898-905
        • Weisz J
        • Ross GT
        • Stolley PD.
        Report of the public board of inquiry on Depo-Provera.
        U.S. Food and Drug Administration, Rockville (MD)1984
        • World Health Organization
        Facts about injectable contraceptives: memorandum from a WHO meeting.
        Bull WHO. 1982; : 199-210
        • Lewis MA
        • Heinemann LAJ
        • Mac Rae KD
        • Bruppacher R
        • Spitzer WO
        • Transnational Research Group on Oral Contraceptives and the Health of Young Women
        The increased risk of venous thrombolism and the use of third generation progestagens: role of bias in observational research.
        Contraception. 1996; 54: 5-13
        • Bloemenkamp KWM
        • Rosendaal FR
        • Helmerhorst FM
        • Buller HR
        • Vandenbroucke JP.
        Enhancement of factor V Leiden mutation of risk of deep-vein thrombosis associated with oral contraceptives containing third-generation progestagen.
        Lancet. 1995; 346: 1593-1596
        • Farmer RDT
        • Lawrenson RA
        • Thompson CR
        • Kennedy JG
        • Hambleton IR.
        Population-based study of risk of various thromboembolism associated with various oral contraceptives.
        Lancet. 1997; 349: 83-88
        • Stanczyk FZ
        • Roy S.
        Metabolism of levonorgestrel, norethindrone and structurally related contraceptive steroids.
        Contraception. 1990; 42: 67-96
        • Shenfield GM
        • Griffin JM.
        Clinical pharmacokinetics of contraceptive steroids: an update.
        Clin Pharmacokinet. 1991; 20: 15-37
        • Spitzer WO
        • Thorogood M
        • Heinemann L.
        Trinational case-control study of oral contraceptives and health.
        Pharmacoepidemiol Drug Saf. 1993; 2: 21-31
        • Farmer RDT
        • Preston TD.
        The risk of venous thromboembolism associated with low oestrogen oral contraceptives.
        J Obstet Gynaecol. 1995; 15: 195-200
        • Farmer R.
        Safety of modern oral contraceptives [letter].
        Lancet. 1996; 347: 259
        • Vandenbroucke JP
        • Koster T
        • Brier E
        • Reitsma PH
        • Bertina RM
        • Rosendaal FR.
        Increased risk of venous thrombosis in oral contraceptive users who are carriers of factor V Leiden mutation.
        Lancet. 1994; 344: 1453-1457
        • Böttiger LE
        • Boman G
        • Eklund G
        • Westerholm B
        Oral contraceptives and thromboembolic disease: effects of lowering oestrogen content.
        Lancet. 1980; 1: 1097-1101
        • Vessey M
        • Mant D
        • Smith A
        • Yeates D.
        Oral contraceptives and venous thromboembolism: findings in a large prospective study.
        BMJ. 1986; 292: 526
        • Helmrich SP
        • Rosenberg L
        • Kaufman DW
        • Strom B
        • Shapiro S.
        Venous thromboembolism and oral contraceptive use.
        Obstet Gynecol. 1987; 69: 91-95
        • Gerstman BB
        • Piper JM
        • Tomita DK
        • Ferguson WJ
        • Stadel BV
        • Lundin FE.
        Oral contraceptive estrogen dose and the risk of deep venous thromboembolic disease.
        Am J Epidemiol. 1991; 133: 32-37
        • Thorogood M
        • Mann J
        • Murphy M
        • Vessey M.
        Risk factors for fatal venous thromboembolism in young women: a case-control study.
        Int J Epidemiol. 1992; 21: 48-52
        • Lewis MA
        • Spitzer WO
        • Heinemann LAJ
        • MacRae KD
        • Bruppacher R
        • Thorogood M
        • et al.
        Third generation oral contraceptives and risk of myocardial infarction: an international case-control study.
        BMJ. 1996; 312: 88-90
        • Poulter NR
        • Chang CL
        • Farley TM
        • Meirik O
        • Marmot MG.
        Ischaemic stroke and combined oral contraceptive: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
        Lancet. 1996; 348: 498-505
        • Poulter NR
        • Chang CL
        • Farley TM
        • Meirik O
        • Marmot MG.
        Haemorrhagic stroke, overall stroke risk, and combined oral contraceptives: results of an international, multicentre, case-control study, WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.
        Lancet. 1996; 348: 505-510
        • Lewis MA
        • Heinemann LAJ
        • Spitzer WO
        • Bruppacher R
        • Thorogood M
        • MacRae KD.
        Results of the Transitional Case-Control Study on Oral Contraceptives.
        in: Proceedings of the Twelfth International Conference on Pharmacoepidemiology; 1996 Aug 28; Amsterdam International Conference on Pharmacoepidemiology, Amsterdam1996
        • Jick H
        • Jick SS
        • Meyers MW
        • Vasilakis C.
        Risk of acute myocardial infarction and low-dose combined oral contraceptives [letter].
        Lancet. 1996; 347: 627-628
        • Sidney S
        • Petitti DB
        • Quesenberry Jr, CP
        • Klatsky AL
        • Wolf S.
        Myocardial infarction in users of low dose oral contraceptives.
        Obstet Gynecol. 1996; 88: 939-944
        • Adam SA
        • Thorogood M
        • Mann JI.
        Oral contraception and myocardial infarction revisited: the effects of new preparations and prescribing patterns.
        Br J Obstet Gynaecol. 1981; 88: 838-845
        • Thorogood M
        • Mann J
        • Murphy M
        • Vessey M.
        Is oral contraceptive use still associated with an increased risk of fatal myocardial infarction?.
        Br J Obstet Gynaecol. 1991; 98: 1245-1253
        • Rosenberg L
        • Palmer JR
        • Lesko SM
        • Shapiro S.
        Oral contraceptive use and the risk of myocardial infarction.
        Am J Epidemiol. 1990; 131: 1009-1016
        • Petitti DB
        • Sidney S
        • Bernstein A
        • Wolf S
        • Queensberry C
        • Ziel HK.
        Stroke in users of low-dose oral contraceptives.
        N Engl J Med. 1996; 335: 8-15
        • Lidegaard O.
        Oral contraception and risk of cerebral thromboembolic attack: results of a case-control study.
        BMJ. 1993; 306: 956-963
        • Fotherby K
        • Caldwell ADS.
        New progestogens in oral contraception.
        Contraception. 1994; 49: 1-32
        • Speroff L
        • De Cherney A
        • Advisory Board for the New Progestins
        Evaluation of a new generation of oral contraceptives.
        Obstet Gynecol. 1993; 81: 1034-1047
        • Godsland IF
        • Crook D
        • Simpson R
        • Proudler T
        • Felton C
        • Lees B
        • et al.
        The effect of different formulations of oral contraceptives on lipid and carbohydrate metabolism.
        N Engl J Med. 1990; 323: 1375-1381
        • Jung-Hoffmann C
        • Kuhl H.
        Interaction with the pharmacokinetics of ethinyl estradiol and progestogens contained in oral contraceptives.
        Contraception. 1989; 40: 299-312
        • Himpel M
        • Tauber U
        • Kuhnz W
        • Pfeffer M
        • Brill K
        • Heitbecker R
        • et al.
        Comparison of serum ethinyl estradiol, sex-hormone-binding globulin, corticoid-binding globulin and cortisol levels in women using two low-dose combined oral contraceptives.
        Horm Res. 1990; 33: 35-39
        • Kuhnz W.
        Pharmacokinetics of the contraceptive steroids levonorgestrel and gestodene after single and multiple oral administration to women.
        Am J Obstet Gynecol. 1990; 163: 2120-2127
        • Kuhnz W
        • Sostarek D
        • Gansau C
        • Louton T
        • Mahler M.
        Single and multiple administration of a new triphasic oral contraceptive to women: pharmacokinetics of ethinyl estradiol and free and total testosterone levels in serum.
        Am J Obstet Gynecol. 1991; 165: 596-602
        • Back DJ
        • Ward S
        • Orme MLE.
        Recent pharmacokenitic studies of low-dose oral contraceptives.
        Adv Contracept. 1991; 7: 164-179
        • Dibbelt L
        • Knuppen R
        • Jütting G
        • Heinmann S
        • Klipping CO
        • Parikka-Olexik H
        Group comparison of serum ethinyl estradiol, SHBG and CBG levels in 83 women using two low-dose combination oral contraceptives for three months.
        Contraception. 1991; 43: 1-21
        • Orme M
        • Back DJ
        • Ward S
        • Green S.
        The pharmacokinetics of ethynylestradiol in the presence and absence of gestodene and desogestrel.
        Contraception. 1991; 43: 305-316
        • Realini JP
        • Goldzieher JW.
        Oral contraceptives and cardiovascular disease: a critique of epidemiologic studies.
        Am J Obstet Gynecol. 1985; 152: 729-798
        • Heinemann LAJ
        • Lewis MA
        • Assmann A
        • Cravens L
        • Guggenmoos-Holzmann I.
        Could preferential prescribing and referral behavior of physicians explain the elevated thrombotic risk found to be associated with third generation oral contraceptives?.
        Pharmacoepidemiol Drug Saf. 1996; 5: 285-294
        • Dunn N
        • Heinemann LAJ
        • Mann RD.
        Are third-generation oral contraceptives really more thrombotic than second-generation: some potential biases examined and compared in German and English doctors.
        in: Proceedings of the Twelfth International Conference on Pharmacoepidemiology; 1996 Aug 28; Amsterdam International Conference on Pharmacoepidemiology, Amsterdam1996
        • Jamin C
        • de Mouzon J.
        Selective prescribing of third-generation oral contraceptives (OCs).
        Contraception. 1996; 54: 55-56
        • Herings RMC
        • de Boer A
        • Urquliart J
        • Leufkens HGM.
        Non-causal explanations for the increased risk of venous thrombembolism among users of third-generation oral contraceptives.
        Pharmacoepidemiol Drug Saf. 1996; 5: 588
      2. Faculty of Family Planning and Reproductive Health. Statement from the Clinical and Scientific Committee: risk of venous thromboembolism and the combined oral contraceptive pill. The Faculty, London1995
        • Vessey MP.
        Benefits and risks of combined oral contraceptives.
        Meth Inform Med. 1993; 32: 222-224