Case-control study of oral contraceptives and risk of thromboembolic stroke: results from international study on oral contraceptives and health of young women

BMJ 1997; 315 doi: (Published 06 December 1997) Cite this as: BMJ 1997;315:1502
  1. Lothar A J Heinemann, director and professora,
  2. Michael A Lewis, directorb,
  3. Margaret Thorogood, readerc,
  4. Walter O Spitzer, professor emeritusd,
  5. Irene Guggenmoos-Holzmann, director and professore,
  6. Rudolf Bruppacher, professorf
  1. a ZEG-Centre for Epidemiology and Health Research Berlin, D-16341 Zepernick, Germany
  2. b EPES Epidemiology Pharmaco-epidemiology and Systems Research, D-12165 Berlin, Germany
  3. c Health Promotion Sciences Unit, London School of Hygiene and Tropical Medicine, London WC1E 7HT
  4. d Department of Epidemiology and Biostatistics, McGill University Montreal, QC, Canada H3A 1A2
  5. e Institute of Medical Statistics, Epidemiology and Informatics, Free University Berlin, D-12200 Berlin, Germany
  6. f University Basle, CH-4310 Rheinfelden, Switzerland
  1. Correspondence to: Professor Heinemann
  • Accepted 29 July 1997


Objective: To determine the influence of oral contraceptives (particularly those containing modern progestins) on the risk for ischaemic stroke in women aged 16–44 years.

Design: Matched case-control study.

Setting: 16 centres in the United Kingdom, Germany, France, Switzerland, and Austria.

Subjects: Cases were 220 women aged 16–44 who had an incident ischaemic stroke. Controls were 775 women (at least one hospital and one community control per case) unaffected by stroke who were matched with the corresponding case for 5 year age band and for hospital or community setting. Information on exposure and confounding variables were collected in a face to face interview.

Main outcome measures: Odds ratios derived with stratified analyses and unconditional logistic regression to adjust for potential confounding.

Results: Adjusted odds ratios (95% confidence intervals) for ischaemic stroke (unmatched analysis) were 4.4 (2.0 to 9.9), 3.4 (2.1 to 5.5), and 3.9 (2.3 to 6.6) for current use of first, second, and third generation oral contraceptives, respectively. The risk ratio for third versus second generation was 1.1 (0.7 to 2.0) and was similar in the United Kingdom and other European countries. The risk estimates were lower if blood pressure was checked before prescription.

Conclusion: Although there is a small relative risk of occlusive stroke for women of reproductive age who currently use oral contraceptives, the attributable risk is very small because the incidence in this age range is very low. There is no difference between the risk of oral contraceptives of the third and second generation; only first generation oral contraceptives seem to be associated with a higher risk. This small increase in risk may be further reduced by efforts to control cardiovascular risk factors, particularly high blood pressure.

Key messages

  • This study shows a slightly increased relative risk of thromboembolic stroke in women currently using oral contraceptives compared with women not using them. The high dose oestrogen pills carry a higher risk than the low dose formulations, irrespective of the type of progestin

  • The absolute risk of occlusive stroke for women who currently use modern oral contraceptives is very small—that is, the incidence in this age range is very low and not different between second and third generation oral contraceptives

  • This small increase in risk can be controlled by avoiding prescription of oral contraceptives to women who have evidence of cardiovascular disease, in particular high blood pressure


The transnational case-control study on oral contraceptives and the health of young women was launched in 1991. There were three substudies for cardiovascular events (venous thromboembolism, myocardial infarction, and thromboembolic stroke). The results for venous thromboembolism1 and myocardial infarction2 have been reported. We report the results of the evaluation of the relation between the use of three generations of oral contraceptives and the occurrence of ischaemic (thromboembolic) stroke.

Subjects and methods

The detailed description of the methods used has been published elsewhere3; the methods were similar to those of the recently published World Health Organisation study.4 In summary, women aged 16–45 years were enrolled in 16 centres in five countries (Austria, France, Germany, Switzerland, and the United Kingdom) between July 1993 and February 1996. At least one hospital and one community control in the same 5 year age band as the case was matched to each incident case with an average of three controls per case. The procedure for control selection with exclusion criteria was defined in the protocol.3 Cases were accepted as ischaemic stroke if the symptoms lasted longer than 24 hours and if there was evidence of (pre)cerebral arterial stenosis or occlusion, cerebral thrombosis, or embolism.3 The diagnosis of thromboembolic stroke was based on computed tomography, magnetic resonance imaging, or angiography within 3 weeks of the event.3 Current use of oral contraceptives was defined as use within 3 months before the diagnosis of the case or the admission date of the hospital control or the interview date of the community control. Oral contraceptives were divided into three categories: first generation (high dose: ≥50 g ethinyloestradiol), second generation (low dose: <50 μg with other gestagens), and third generation (low dose with either gestodene or desogestrel). We report unmatched odds ratios with 95% confidence intervals adjusted for potential confounders and matched odds ratios (matched by age and centre) as a sensitivity verification (with STATA software5). To be consistent with the WHO study6 we classified norgestimate as second generation in our analysis but report it in both ways in the table.


In total 220 women with thromboembolic stroke (cases) and 336 hospital and 439 community controls were recruited. Of these, 67 cases were enrolled in the United Kingdom, 82 in Germany, 37 in France, 19 in Switzerland, and 15 in Austria. The table shows the adjusted unmatched odds ratios for occlusive stroke in women currently using oral contraceptives: 4.4 (2.0 to 9.9) for first generation, 3.4 (2.1 to 5.5) for oral contraceptives with second generation progestins, and for 3.9 (2.3 to 6.6) third generation. The point estimates derived from the matched analysis were slightly lower. The risk ratio for third versus second generation in the unmatched analysis was 1.1 (0.7 to 2.0) and similar in the matched analysis (1). Stratified by region, the fully adjusted odds ratios in the United Kingdom for second generation oral contraceptives versus no current use were 5.0 (1.5 to 17.1) and 6.2 (2.1 to 18.8) for third versus no use. The figures were 3.9 (2.2 to 7.0) and 3.7 (1.9 to 7.1), respectively, for the same comparisons in continental Europe. The adjusted odds ratios of second generation versus no use were 7.0 (3.8 to 12.8) for hospital controls and 2.6 (1.5 to 4.6) for community controls; and the estimates for third generation versus no use were 5.8 (3.0 to 11.3) for hospital controls and 3.4 (1.8 to 6.3) for community controls. The odds ratios for stroke were lower if the blood pressure was checked before prescription (odds ratios for checked versus not checked were 1.8 (1.0 to 3.0) versus 4.5 (2.1; 9.6) for second generation use versus non-current use and 2.5 (1.4; 4.4) versus 4.6 (2.0; 10.6) for third generation use, respectively).

Oral contraceptive uses and risk of ischaemic stroke. Use refers to current use. Norgestimate coded as second generation oral contraceptive

View this table:


Our study confirms the well established, increased relative risk for thromboembolic stroke in women who use oral contraceptives; the median value of the published relative risk estimates is between 2 and 4.7 The risk estimates for high dose oestrogen oral contraceptives (first generation) are higher than those for the low dose pills, but there was no difference between those containing second or third generation progestagens. The recently published WHO study, which used almost identical methods and classifications, found an adjusted odds ratio (conditional analysis) for ischaemic stroke of 2.99 (1.65 to 5.40) associated with current use of any oral contraceptive in Europe.6 With the same analytic approach our data resulted in an odds ratio of 2.9 (2.0 to 4.0). Our results agree with the findings of the WHO study (Europe) with respect to the odds ratios for first generation progestins but yield slightly higher estimates for second and third generation progestins. There was no significant difference between the estimated risk of thrombotic stroke in users of second versus third generation pills. The United Kingdom and the continental countries had similar findings. These odds ratios should be assessed against the backdrop of the small absolute risk they entail and in the context of the clear benefits of use of oral contraceptives for women of reproductive age. The annual event rate is between 1-1.6 stroke events per 10 000 women aged 25-44—that is, 1 stroke per 12 000 women. Three strokes per 100 000 women per year may be attributable to the use of oral contraceptives. This risk could be controlled by avoiding prescription of the pill in women who have important cardiovascular risk factors such as high blood pressure and might be lessened by appropriate management of these risk factors.


The investigators were accountable only to the Scientific Reference Board (members listed in reference 1).

Transnational case-control study. Principal investigators: WO Spitzer (Montreal, 1993-5), LAJ Heinemann (Berlin, 1996 to date). Scientific review board: U Bergman (Stockholm), M Breckwoldt (Freiburg), J Collins (Toronto), F Kemper (Munster), J LeLorier (Montreal), S MacLeod (Toronto), K MacRae (London, until May 1995), W Ray (Nashville), J Schlesselman (Pittsburgh). Centre directors: Austria—H Concin (Bregenz), G Stark (Graz), K Pfeiffer (Innsbruck), R Weitgasser (Salzburg); France—B Begaud (Bordeaux), B Boissel (Lyons). Germany—J Haerting, F Lautenschlaeger (Halle), K Püschel (Hamburg), F Hoffmann (Jena), BP Robra (Magdeburg), M Lustermann (Nordhausen), I Schulzki (Schwerin), S Boethig (Zwickau). Switzerland—F Gutzwiller (Zurich). United Kingdom—L de Caestecker (Glasgow), C McCollum (till August 1995), P Hannaford (from August 1995) (Manchester), R Mann (Southampton). Publication committee: I Guggenmoos-Holzmann (Berlin); R Bruppacher (Basle); M Lewis (Berlin); M Thorogood (London), LAJ Heinemann (Berlin). Data management centre: M Lewis, D Kühl-Habich, C Klindworth, A Assmann (Potsdam) (for more extensive list of collaborators see reference 1).

Funding: Unconditional grant from Schering AG, Berlin.

Conflict of interest: Funding by Schering AG.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
View Abstract