Current issues of ACP Journal Club are published in Annals of Internal Medicine


Nonfatal venous thromboembolism was associated with oral contraceptives that contain desogestrel and gestodene

ACP J Club. 1996 May-June;124:79. doi:10.7326/ACPJC-1996-124-3-079

Source Citation

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 Dec 16;346:1589-93.



To determine the risk for cardiovascular illness in women who take oral contraceptives (OCs) that contain estrogen combined with levonorgestrel, desogestrel, or gestodene. Study 1 evaluated unexpected idiopathic cardiovascular death, and study 2 evaluated venous thromboembolism (VTE) (pulmonary embolism or deep-venous thrombosis).


Cohort studies (studies 1 and 2) and a nested case-control study (study 2).


U.K. general practices (470 for study 1 and 365 for study 2).


Healthy women < 40 years old (n = 303 470 for study 1; 238 130 for study 2) who received OCs containing < 35 µg ethinylestridiol with either levonorgestrel, desogestrel, or gestodene. Exclusion criteria were history of VTE, acute myocardial infarction, stroke, cancer, epilepsy, diabetes, hypertension, hyperlipidemia, or cystic fibrosis. Control patients (study 2) were matched for age and current use of a study OC.

Assessment of Risk Factors

For study 1, person-years of exposure to each type of OC were calculated from computer records. For study 2, women were classified into users and past users of each type of OC. Smoking and body mass index were controlled in the analysis.

Main Outcome Measures

For study 1, death certificates and computer files were used to ascertain unexplained cardiovascular death. For study 2, computer records and discharge letters were used to ascertain VTE.

Main Results

15 unexpected cardiovascular deaths and 80 nonfatal VTEs occurred. For both studies, the relative risk (RR) for levonorgestrel was taken as 1.0. For study 1, the RR estimates for cardiovascular death adjusted for age (< 30 y and ≥ 30 y) and calendar time (1991 to 1992 and 1993 to 1994) did not differ for OCs combined with desogestrel (0.4, 95% CI 0.1 to 2.1) or gestodene (1.4, CI 0.5 to 4.5). For study 2, the age and calendar-time-adjusted RR for VTE was 1.9 (CI 1.1 to 3.2) in desogestrel-combined OCs, 1.8 (CI 1.0 to 3.2) for gestodene-combined OCs, and 0.2 (CI 0.1 to 0.6) for past use. Case-control analysis showed similar results for VTE: desogestrel, RR 2.2 (CI 1.1 to 4.4), and gestodene, RR 2.1 (CI 1.0 to 4.4).


Women who took combined OCs containing desogestrel or gestodene had an increased risk for nonfatal VTE when compared with women who took combined OCs containing levonorgestrel. The risk was reduced for past users.

Sources of funding: Berlex Laboratories, Bayer AG, Glaxo Wellcome Inc., Ciba-Geigy Corporation, Hoechst AG, Merck Research Laboratories, Pfizer Inc., and Sanofi Winthrop Pharmaceuticals.

For article reprint: Dr. H. Jick, Boston Collaborative Drug Surveillance Program, Boston University Medical Center, 11 Muzzey Street, Lexington, MA 02173, USA. FAX 617-862-1680.


Increased risk for VTE among women who take combined OCs is well documented and has been attributed to estrogen (1). Jick and colleagues show a 2-fold increased risk for VTE in women who take OCs containing desogestrel or gestodene when compared with women who take OCs containing levonorgestrel. The findings were unaffected when controlled for the potential selection bias that third-generation progestins may be more often prescribed to women at greater risk for VTE because of smoking, obesity, or first-time use. Stratification by certainty of diagnosis or duration of use or restricting the analysis to first-time OC use did not eliminate the association. 4 recent case-control studies showed similar or greater RRs for VTE among women who take desogestrel and gestodene OCs (2-5).

In the United States, OCs containing desogestrel have been available for several years and account for about 15% of the market. The RR for VTE with use of any of the low-dose estrogen pills is substantially lower than previously reported for OCs (1): about 16/100 000 woman-years among levonorgestrel OC users and 32/100 000 woman-years among desogestrel and gestodene users compared with approximately 60/100 000 woman-years for pregnancy and the post-partum period. Whether risks for other, rarer cardiovascular problems are reduced by third-generation progestins has not been established.

Clinicians should alter their prescribing practices for women who start taking OCs or who are otherwise at higher risk for VTE. Women who already take these OCs should be informed of the small but real increased risks. Before insisting that women discontinue taking OCs, clinicians should keep in mind the higher risks associated with pregnancy.

Joy Melnikow, MD, MPH
University of CaliforniaSacramento, California, USA


1. Stradel BV. N Engl J Med. 1981;305:612-7.

2. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet. 1995;346:1575-82.

3. World Health Organization Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Lancet. 1995;346:1582-8.

4. Bloemenkamp KW, Rosendaal FR, Helmerhorst FM, Buller HR, Vandenbroucke JP. Lancet. 1995;346:1593-6.

5. Spitzer WO, Lewis MA, Heinemann LA, et al. BMJ. 1996;312:83-8.