OBJECTIVE: To investigate associations between combined hormonal contraception and progestogen-only contraception and risks of venous thromboembolism by progestogen and carriership of genetic hemostatic variations.
METHODS: This was a case–control study in Sweden carried out between 2003 and 2009, which included 948 patients with venous thromboembolism and 902 individuals in a control group, all aged 18–54 years. Information was obtained by telephone interviews and DNA analyses of blood samples. Radiologic referrals were used for case ascertainment. For comparisons, odds ratios were estimated by unconditional logistic regression analysis adjusting for smoking, body mass index (BMI), and immobilization.
RESULTS: The odds ratio (OR) for current use of combined hormonal contraception was 5.3 (95% confidence interval [CI] 4.0–7.0). Desogestrel combinations had the highest OR (11.4, 95% CI 6.0–22.0). The OR for injection of medroxyprogesterone acetate was 2.2 (95% CI 1.3–4.0). In users of combined hormonal contraception with the factor V Leiden mutation, the OR was 20.6 (95% CI 8.9–58). In women who used progestogen-only contraception and carried the factor V Leiden mutation, the OR was 5.4 (95% CI 2.5–13).
CONCLUSION: Risks of venous thromboembolism in association with combined hormonal contraception vary by type of progestogen and independently of BMI and smoking. Thrombophilic genotypes such as factor V Leiden increase risks of venous thromboembolism in users of combined hormonal contraception. Except for injection of medroxyprogesterone acetate, progestin-only contraception seems to be the least thrombogenic hormonal contraception for women carrying genetic hemostatic variations.
LEVEL OF EVIDENCE: II
Risk of venous thromboembolism associated with hormonal contraception differs by type and dose of the progestogen and by thrombophilic genotypes.
Department of Medicine, Karolinska Institutet, the Centre for Pharmacoepidemiology, Department of Medicine, Solna, Karolinska Institutet, the Coagulation Unit, Hematology Centre, and the Division of Clinical Pharmacology, Karolinska University Hospital, Stockholm, the Medical Products Agency, Uppsala, and the Department of Clinical Sciences/Medical Radiology, Diagnostic Centre of Imaging and Functional Medicine, Lund University, Lund, Sweden; and the Department of Medicine, McMaster University, Hamilton, Ontario, Canada.
Corresponding author: Annica Bergendal, MD, PhD, Centre for Pharmacoepidemiology, Department of Medicine Solna, Karolinska Institutet SE-171 76 Stockholm, Sweden; e-mail: firstname.lastname@example.org.
Supported by unrestricted grants from Janssen-Cilag, Novartis, Organon, Schering, Wyeth, AFA Insurance, Center for Gender Medicine Karolinska Institutet, and the Medical Products Agency.
The authors thank Ebba Hallberg, Elisabet Stjernberg, and Annika Åkerström for data collection and study coordination; Anette Silvan for preparing DNA; and Kristina Holmberg for generating the genotyping data. Genotyping was performed at the KTH genotyping facility supported by the Wallenberg Consortium North.
Financial Disclosure The authors did not report any potential conflicts of interest.
Combined hormonal contraception is associated with an increased risk of venous thromboembolism. However, the effect of the type of progestogen is still debated.1,2 A recent review and meta-analysis concluded that combined hormonal contraceptives containing desogestrel, cyproterone acetate, or drospirenone seems to hold a higher of risk of venous thromboembolism as compared with those containing levonorgestrel or norgestimate.3
Some of the previous studies did not consider important confounders such as body mass index (BMI, calculated as weight (kg)/[height (m)]2), smoking, and family history of venous thromboembolism.4–6 Furthermore, importance of exposure time on risks and the effect of thrombophilic genetic variations in users of combined hormonal contraceptives have remained uncertain.7 For women carrying the thrombophilic genetic factor V Leiden and the prothrombin gene mutation, the reported risks are based on few cases and risks vary greatly.8 Data on other genetic variants affecting hemostasis are even more limited.9,10 Currently, thrombophilia screening is considered as not cost-effective before prescribing combined hormonal contraceptives.11 Progestogen-only contraception might constitute a safer alternative to combined hormonal contraception, but only a few studies have addressed risk on venous thromboembolism and the results are ambiguous.12–14
The aims of this nationwide case–control study (Thrombo Embolism Hormone Study) were to investigate associations between hormonal contraception and risks of venous thromboembolism with a focus on type of progestogen while considering important confounders such as BMI and smoking and further to study the influences of carriership of targeted genetic variants.
MATERIALS AND METHODS
Details of the study have previously been published elsewhere.15 The Thrombo Embolism Hormone Study was a nationwide case–control study carried out between January 1, 2003, and March 31, 2009, in Sweden (Fig. 1). Cases were women with a first episode of deep venous thrombosis located in the leg or pelvis or pulmonary embolism. Both inpatients and outpatients were recruited at 43 large-sized and middle-sized hospitals geographically spread throughout Sweden. All thromboembolic events were examined by objective radiologic methods. We only included patients who had initiated anticoagulant treatment. Women in the control group were frequency-matched by birth year to the women in the case group and randomly selected from the general Swedish Population Register. Each woman was assigned an index date, which was the date of diagnosis for women in the case group and the date of the telephone interview for the women in the control group. Women who had experienced a previous thrombosis, were or had been pregnant during the last 3 months before the index date, or had a current malignancy were not included. Women with a previous malignancy were included only if they had been free of disease and cancer treatment for at least 5 years before the index date. All women who fulfilled the inclusion criteria and accepted participation were scheduled for a telephone interview. A letter containing a written consent form to sign and a kit for blood sampling were sent to all participants.
We obtained information on dose and route of administration for specific substances of combined contraceptives and progestogen-only contraceptives during 3 months preceding the index date (current use). Length of use was estimated from date of treatment start and stop. The combined hormonal contraceptions included oral tablets, dermal patches, and intravaginal devices and the progestogen-only contraceptives were tablets, depot injections, intrauterine devices, and implants.
Combined hormonal contraception was categorized by type of progestogen and progestogen-only contraception was divided by dose and potency of the progestogen into four categories: very low (levonorgestrel intrauterine device), low (oral levonorgestrel, lynestrenol, or norethisterone), medium (implants and oral desogestrel), and high (injectable medroxyprogesterone).12 All current users of combined hormonal contraception and progestogen-only contraception were categorized by length of use and divided into three categories: 3 months or less, 3–12 months, and more than 12 months. Immobilization was divided into five mutually exclusive categories: no immobilization, minor trauma or bed rest more than 1 day, surgery only, plaster cast only, and both surgery and plaster cast, all occurring during the 3 months preceding the index date. To categorize BMI, we used the World Health Organization cutoff levels.16 Smoking was defined as current smokers and nonsmokers. Former smokers were categorized as nonsmokers.
All women were asked to donate a blood sample to ascertain common single nucleotide polymorphisms in genes coding for blood coagulation proteins. DNA was prepared using the QIAGEN FlexiGene DNA kit. Genotyping for factor V Leiden G1691A and the prothrombin gene mutation G20210A, methyl tetrahydroflate reductase C677T, endothelial nitric oxide synthase G894T, plasminogen activator inhibitor-1-675 4G/5 G, factor XIII G163T, and glycoprotein IIIa C1565T was performed by pyrosequencing.17
Analyses included women 18–54 years of age. Risks of venous thromboembolism associated with use of combined hormonal contraceptives or progestogen-only contraceptives were estimated by unconditional logistic regression analysis. We computed crude odds ratios (ORs) and adjusted ORs with 95% confidence intervals (CIs). Adjustments were made for age (matching factor), smoking, BMI, and immobilization. Risk of venous thromboembolism associated with combined hormonal contraception was analyzed by type of progestogen and by length of current use. For progestogen-only contraception, the risks for venous thromboembolism were assessed by dose and by length of current use. To investigate the influence of hemostatic single nucleotide polymorphisms, analyses were performed for venous thromboembolism risk associated with combined hormonal contraception and progestogen-only contraception use according to carriership (homozygotes and heterozygotes). For the analyses of the combined effects of hormonal contraception and carriership of hemostatic single nucleotide polymorphisms, we computed crude ORs with exact 95% and 99% CIs. Analyses were repeated after excluding women with BMIs of 30 or higher and those who had surgery or plaster cast. Among users of combined hormonal contraception, we calculated risks for venous thromboembolism by type of progestogen using combinations with levonorgestrel as a reference category. For users of progestogen-only contraception, we performed analyses calculating risks for venous thromboembolism by dose using the category “very low dose” (intrauterine device) as the reference group. Finally, we assessed risks associated with being a first-time user of combined hormonal contraception by calculating risks for venous thromboembolism among the women who reported use of combined hormonal contraception for the first time during the index period and who had used it for no longer than 12 months. Statistical analyses were performed using STATA 10.0. The study was approved by the regional research ethics committees in Stockholm (01-255, 04-469), Linköping, (01-453), Göteborg (M088-01), Uppsala (01-277), and Umeå (01-198), Sweden.
A total of 1,850 women were included in the study (Table 1). Among the 948 women in the case group, 640 (67.5%) had a deep vein thrombosis of the leg or the pelvis, 292 (30.8%) had a pulmonary embolism, and 16 (1.7%) had been diagnosed with both. Obesity, surgery, plaster cast, and smoking were more common among women in the case group than among women in the control group (Table 1).
Current use of combined hormonal contraception was reported by 311 women in the case group (32.8%) and 107 (11.9%) women in the control group. For progestogen-only contraception, the corresponding figures were 145 (15.3%) and 177 (19.6%), respectively (Table 1). Eight women (six in the case group and two in the control group) had used both types of contraception during the 3-month period. Carriership of factor V Leiden was more frequent among women in the case group than women in the control group (23.8% compared with 8.2%; Table 2) and so was carriership of the prothrombin gene mutation (5.9% compared with 3.2%). The proportion of methyl methyl tetrahydrofolate reductase carriers was slightly higher among women in the control group.
Current use of combined hormonal contraception was associated with a fivefold increased risk of venous thromboembolism (adjusted OR 5.3, 95% CI 4.0–6.9; Table 3). Combinations with the progestogen desogestrel yielded the highest risk estimate (adjusted OR 11.4, 95% CI 6.0–22.0) followed by drospirenone (adjusted OR 8.4, 95% CI 4.2–17.0). The adjusted OR could not be calculated for lynestrenol because there were no exposed women in the control group. In the analyses with users of combined hormonal contraceptions containing levonorgestrel as the reference category, the users of desogestrel combinations had a more than twofold increased risk (adjusted OR 2.6, 95% CI 1.3–5.4; Table 4). Users of norethisterone containing combined hormonal contraceptions had a decreased risk (adjusted OR 0.4, 95% CI 0.2–0.9).
With the exception of use of high-dose progestogen-only contraception, there was no increased risk of venous thromboembolism associated with this type of contraception (adjusted OR 0.9, 95% CI 0.7–1.2; Table 3). High-dose users had a doubled risk (adjusted OR 2.2, 95% CI 1.3–4.0) and when using very low-dose progestogen users as a reference category, the risk increase was threefold (adjusted OR 3.0, 95% CI 1.5–6.0; Table 5).
Women who had used combined hormonal contraception for 3 months or less had the highest risk of venous thromboembolism (adjusted OR 8.5, 95% CI 4.1–17.0; Table 6). The corresponding risk for more than 12 months of use was lower (adjusted OR 4.4, 95% CI 3.2–6.1). Among users of progestogen-only contraception, the length of use had no effect on risk. A risk estimate for first-time users of combined hormonal contraception could not be generated because only eight women in the case group were recorded as such.
Users of combined hormonal contraception who were carriers of factor V Leiden or of the prothrombin gene mutation had very high risks for venous thromboembolism (OR 20.6, 95% CI 8.9–58.0 and 17.6, 95% CI 4.4–158, respectively) as compared with nonusers and noncarriers (Table 7). For carriers of factor XIII, the risk of venous thromboembolism associated with combined hormonal contraception use was lower (OR 2.8, 95% CI 1.9–4.1) than in noncarriers (OR 4.9, 95% CI 3.3–7.3). The point estimate of risk associated with factor V Leiden in combined hormonal contraception users was almost doubled (OR 37.5, 95% CI 15–121) when excluding women who had surgery, a plaster cast, or both and obese women.
Carriers of factor V Leiden who used progestogen-only contraception had a fivefold risk of venous thromboembolism (OR 5.4, 95% CI 2.5–13; Table 8). When excluding women who had surgery, a plaster cast, or both and obese women, the risk attenuated slightly (OR 4.7, 95% CI 1.9–12). Women with carriership of factor V Leiden and using injectable medroxyprogesterone acetate had the highest risk (OR 16.7, 95% CI 2.4–714). Women carrying the factor V Leiden and using the very low dose progestogen-only contraception (intrauterine device) had a risk (OR 3.2, 95% CI 1.2–10.4) in line with that of carriers without hormonal contraception (OR 2.6, 95% CI 1.8–3.7). There were only six women in the case group and 10 women in the control group who were carriers of the prothrombin gene mutation and users of progestogen-only contraception, which generated a reduced but imprecise risk estimate (OR 0.73, 95% CI 0.22–2.23).
Our results suggest a fivefold increased risk for venous thromboembolism among users of combined hormonal contraception with approximately two times higher risks for combined hormonal contraceptions containing desogestrel compared with levonorgestrel. In contrast to high-dose preparations, use of low- and medium-dose progestogen-only contraception did not confer an increased risk for venous thromboembolism. Venous thromboembolism risk increased further in current combined hormonal contraception users being carriers of factor V Leiden and the prothrombin gene mutation and in users of high-dose progestogen-only contraception being carriers of factor V Leiden.
The fivefold risk increase for venous thromboembolism among users of combined hormonal contraception is somewhat higher than what has been reported previously, but in accordance with the results from recent studies.4,6,18 As previously described, the risk for venous thromboembolism was most pronounced during the first 3 months of treatment and declined by time.6,19,20 The higher risks in combined hormonal contraceptions containing desogestrel and drospirenone as compared with levonorgestrel are also in agreement with most previous studies.3,4,6 We had the possibility to address important confounding factors and the increased differential risks appear to be independent of BMI, smoking, and immobilization.3
Factor V Leiden and the prothrombin gene mutation are associated with thrombophilia and a moderate increase of the risk for venous thromboembolism. Use of combined hormonal contraceptions increases the risk of venous thromboembolism multiplicatively.21–24 The finding of a 20-fold increased risk in carriers of factor V Leiden and set in relation to a threefold increase in noncarriers is in accordance with combined estimates reported in meta-analyses.8,25 As for carriership of prothrombin gene mutation, our data suggesting an almost 20-fold increased risk contribute to the growing knowledge concerning the implication of genetic variations in women.8,26
High doses of progestogens may effect hemostasis, which has been noted after use of emergency contraception and in connection with treatment of dysfunctional bleeding.4,12,13,27,28 In line with what has previously been reported, we found an increased risk for venous thromboembolism in association with use of “high-dose” progestogen-only contraception (depot injectables).29 Despite limited knowledge, progestogen-only contraception is recommended as contraception in women with mild thrombophilia.13,30 We found that women who were users of progestogen-only contraception and carriers of factor V Leiden had a fivefold increased risk of venous thromboembolism when compared with noncarriers and not using progestogen-only contraception. The increased risk was confined to users of injectables.
Major strengths of this study include the population-based design and the possibility to adjust for acquired risk factors and to assess several potential thrombophilic genotypes. Potential limitations are the retrospective assessment of exposure and the limited sample size, leading to imprecise risk estimates in some analyses. Furthermore, women in the control group were randomly selected from the general population, which should preclude selection bias among the women in the control group.6 However, we could ascertain length of use only in those reporting current use of combined hormonal contraception; consequently, earlier use in some women was not accounted for. Furthermore, because all first-time-ever users during the index period were in the case group, risk estimates for being a first-time-ever user could not be calculated. In combined hormonal contraception, risk of venous thromboembolism depends on dose of the estrogen but as a result of fixed doses of 30–40 micrograms ethinyl estradiol in all combined hormonal contraceptions used, we could not assess this effect.4,31,32
The results from our study support the differential risks of venous thromboembolism by type of progestogen for women using combined hormonal contraceptions. Carriers of factor V Leiden and the prothrombin gene mutation should be prescribed combined hormonal contraceptions with caution and low- and medium-dose progestogen-only contraceptions seem to be the least thrombogenic alternative. Accordingly, identification of carriers before the decision on choice of hormonal contraception should be useful.
1. Shapiro S, Dinger J. Risk of venous thromboembolism among users of oral contraceptives: a review of two recently published studies. J Fam Plann Reprod Health Care 2010;36:33–8.
2. Dinger J, Shapiro S. Combined oral contraceptives, venous thromboembolism, and the problem of interpreting large but incomplete datasets. J Fam Plann Reprod Health Care 2012;38:2–6.
3. Martínez F, Ramírez I, Pérez-Campos E, Latorre K, Lete I. Venous and pulmonary thromboembolism and combined hormonal contraceptives. Systematic review and meta-analysis. Eur J Contracept Reprod Health Care 2012;17:7–29.
4. Lidegaard Ø, Løkkegaard E, Svendsen AL, Agger C. Hormonal contraception and risk of venous thromboembolism: national follow-up study. BMJ 2009;339:b2890.
5. Lidegaard Ø, Nielsen LH, Skovlund CW, Skjeldestad FE, Løkkegaard E. Risk of venous thromboembolism from use of oral contraceptives containing different progestogens and oestrogen doses: Danish cohort study, 2001-9. BMJ 2011;343:d6423.
6. van Hylckama Vlieg A, Helmerhorst FM, Vandenbroucke JP, Doggen CJ, Rosendaal FR. The venous thrombotic risk of oral contraceptives, effects of oestrogen dose and progestogen type: results of the MEGA case-control study. BMJ 2009;339:b2921.
7. Grimes DA, Stuart GS, Levi EE. Screening women for oral contraception: can family history identify inherited thrombophilias? Obstet Gynecol 2012;120:889–95.
8. Dayan N, Holcroft CA, Tagalakis V. The risk of venous thrombosis, including cerebral vein thrombosis, among women with thrombophilia and oral contraceptive use: a meta-analysis. Clin Appl Thromb Hemost 2011;17:E141–52.
9. van Hylckama Vlieg A, Middeldorp S. Hormone therapies and venous thromboembolism: where are we now? J Thromb Haemost 2011;9:257–66.
10. van Vlijmen EF, Veeger NJ, Middeldorp S, Hamulyák K, Prins MH, Büller HR, et al.. Thrombotic risk during oral contraceptive use and pregnancy in women with factor V Leiden or prothrombin mutation: a rational approach to contraception. Blood 2011;118:2055–61.
11. van Sluis GL, Söhne M, El Kheir DY, Tanck MW, Gerdes VE, Büller HR. Family history and inherited thrombophilia. J Thromb Haemost 2006;4:2182–7.
12. Bergendal A, Odlind V, Persson I, Kieler H. Limited knowledge on progestogen-only contraception and risk of venous thromboembolism. Acta Obstet Gynecol Scand 2009;88:261–6.
13. Blanco-Molina MA, Lozano M, Cano A, Cristobal I, Pallardo LP, Lete I. Progestin-only contraception and venous thromboembolism. Thromb Res 2012;129:e257–62.
14. Mantha S, Karp R, Raghavan V, Terrin N, Bauer KA, Zwicker JI. Assessing the risk of venous thromboembolic events in women taking progestin-only contraception: a meta-analysis. BMJ 2012;345:e4944.
15. Bergendal A, Bremme K, Hedenmalm K, Lärfars G, Odeberg J, Persson I, et al.. Risk factors for venous thromboembolism in pre-and postmenopausal women. Thromb Res 2012;130:596–601.
16. World Health Organization. Technical report series 894: obesity: preventing and managing the global epidemic. Geneva (Switzerland): World Health Organization; 2000. Available at: http://whqlibdoc.who.int/trs/WHO_TRS_894_(part1).pdf
. Retrieved July 7, 2014.
17. Ronaghi M, Uhlén M, Nyrén P. A sequencing method based on real-time pyrophosphate. Science 1998;281:363, 365.
18. Lidegaard O, Nielsen LH, Skovlund CW, Løkkegaard E. Venous thrombosis in users of non-oral hormonal contraception: follow-up study, Denmark 2001-10. BMJ 2012;344:e2990.
19. Suissa S, Blais L, Spitzer WO, Cusson J, Lewis M, Heinemann L. First-time use of newer oral contraceptives and the risk of venous thromboembolism. Contraception 1997;56:141–6.
20. Dinger JC, Heinemann LA, Kühl-Habich D. The safety of a drospirenone-containing oral contraceptive: final results from the European Active Surveillance Study on oral contraceptives based on 142,475 women-years of observation. Contraception 2007;75:344–54.
21. Lijfering WM, Brouwer JL, Veeger NJ, Bank I, Coppens M, Middeldorp S, et al.. Selective testing for thrombophilia in patients with first venous thrombosis: results from a retrospective family cohort study on absolute thrombotic risk for currently known thrombophilic defects in 2479 relatives. Blood 2009;113:5314–22.
22. Martinelli I, Taioli E, Bucciarelli P, Akhavan S, Mannucci PM. Interaction between the G20210A mutation of the prothrombin gene and oral contraceptive use in deep vein thrombosis. Arterioscler Thromb Vasc Biol 1999;19:700–3.
23. Legnani C, Palareti G, Guazzaloca G, Cosmi B, Lunghi B, Bernardi F, et al.. Venous thromboembolism in young women; role of thrombophilic mutations and oral contraceptive use. Eur Heart J 2002;23:984–90.
24. Vandenbroucke JP, Koster T, Briët 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;344:1453–7.
25. Wu O, Robertson L, Langhorne P, Twaddle S, Lowe GD, Clark P, et al.. Oral contraceptives, hormone replacement therapy, thrombophilias and risk of venous thromboembolism: a systematic review. The Thrombosis: Risk and Economic Assessment of Thrombophilia Screening (TREATS) Study. Thromb Haemost 2005;94:17–25.
26. Ravin AJ, Edwards RP, Krohn M, Kelley JR, Christopherson WA, Roberts JM. The factor V Leiden mutation and the risk of venous thromboembolism in gynecologic oncology patients. Obstet Gynecol 2002;100:1285–9.
27. van Rooijen M, Silveira A, Thomassen S, Hansson LO, Rosing J, Hamsten A, et al.. Rapid activation of haemostasis after hormonal emergency contraception. Thromb Haemost 2007;97:15–20.
28. Vasilakis C, Jick H, del Mar Melero-Montes M. Risk of idiopathic venous thromboembolism in users of progestagens alone. Lancet 1999;354:1610–1.
29. van Hylckama Vlieg A, Helmerhorst FM, Rosendaal FR. The risk of deep venous thrombosis associated with injectable depot-medroxyprogesterone acetate contraceptives or a levonorgestrel intrauterine device. Arterioscler Thromb Vasc Biol 2010;30:2297–300.
30. World Health Organization. Medical eligibility criteria for contraceptive use. 4th ed. Geneva (Switzerland): World Health Organization; 2009.
31. Vessey M, Mant D, Smith A, Yeates D. Oral contraceptives and venous thromboembolism: findings in a large prospective study. Br Med J (Clin Res Ed) 1986;292:526.
© 2014 by The American College of Obstetricians and Gynecologists.
32. Stegeman BH, de Bastos M, Rosendaal FR, van Hylckama Vlieg A, Helmerhorst FM, Stijnen T, et al.. Different combined oral contraceptives and the risk of venous thrombosis: systematic review and network meta-analysis. BMJ 2013;347:f5298.