Publications

Published: July 1998

The Safety of Combined Oral Contraceptives - Ischaemic and Haemorrhagic Stroke, Breast and Cervical Cancer

Prescriber Update 15, August 1997
July 1998

Medsafe Editorial Team

Recent studies have confirmed that small increases in the risk of ischaemic and haemorrhagic stroke, and breast and cervical cancer occur with the use of combined oral contraceptives. The risk of stroke is increased in smokers, hypertensives and those older than 35 years. The relative risk of breast cancer falls to unity within 10 years of discontinuing oral contraceptives. Most of the elevation in risk of cervical cancer is associated with recent long-term use. While these studies provide reassuring evidence of the low risk of these events with combined oral contraceptives, they also give reason to carefully assess the benefits and risks of the options in relation to the woman's risk factors before making a choice of contraceptive.

Ischaemic and Haemorrhagic Stroke

The results of the WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception with respect to ischaemic and haemorrhagic stroke and combined oral contraceptives (OCs) have now been published,1,2following publication of the results related to venous thromboembolism in December 1995.3

The study included 703 cases of ischaemic stroke and 1083 of haemorrhagic stroke who were women aged 20-44 years without a history of thromboembolic events, pregnancy in the previous 3 weeks, natural or surgical menopause, recent surgery, or illness causing bed rest of  > 1 week. The type of stroke was confirmed by computed tomography, magnetic resonance imaging, cerebral angiography, or lumbar puncture within 3 weeks of the clinical event. Up to three controls were matched to each case for age, hospital and time of admission.

Combined OC use increases ischaemic and haemorrhagic stroke risk

The adjusted odds ratio for ischaemic stroke in users of combined oral contraceptives was 2.99 (95% confidence interval 1.65-5.40) in Europe and 2.93 (2.15-4.00) in developing countries. The corresponding figures for haemorrhagic stroke were 1.38 (0.84-2.25) and 1.76 (1.34-2.30). There was no increase in risk of haemorrhagic stroke with oral contraceptive use in younger non-smoking women.

The authors calculated the incidence rates of haemorrhagic stroke in women aged 20-44 years in European countries: 4.8 per 100 000 woman-years among non-users of oral contraceptives; 6.7 among users of low-oestrogen dose (< 50mcg) combined contraceptives; and 12.9 among users of the higher dose (≥ 50mcg) contraceptives. A similar calculation was not performed for ischaemic stroke. In Europe the background incidence of ischaemic stroke is 1-3 per 100 000 woman-years for women < 35 years, and 10 per 100 000 woman-years for those > 35 years.

Risk factors are smoking, hypertension and age ≥ 35 years

Current smoking ( ≥ 10 cigarettes per day) and a history of hypertension increased the risk of both ischaemic and haemorrhagic stroke, as did age ≥ 35 years (odds ratios presented in table). In addition, women who had had their blood pressure checked (based on user recall) before using oral contraceptives were at reduced risk of ischaemic stroke compared with those in whom blood pressure had not been checked. For ischaemic stroke the association between smoking and combined oral contraceptive use appeared to be synergistic. For haemorrhagic stroke, a family history of stroke was an additional risk factor.

Odds ratios (95% confidence interval) for ischaemic or haemorrhagic stroke in combined oral contraceptive users in the presence of various risk factors

Region and type of user Europe Developing countries
Ischaemic stroke
Current smoker 7.20 (3.23-16.1) 4.83 (2.76-8.43)
History of hypertension 10.7 (2.04-56.6) 14.5 (5.36-39.0)
Age ≥ 35/BP check 2.50 (0.80-7.79)* 2.73 (1.17-6.35)*
Age ≥ 35/no BP check 5.94 (1.89-18.7)* 5.87 (3.29-10.5)*
Haemorrhagic stroke
Current smoker 3.10 (1.65-5.83) 3.73 (2.43-5.71)
History of hypertension 10.3 (3.27-32.3) 14.3 (6.72-30.4)
Age ≥ 35/non-smoker 2.39 (0.89-6.43)* 2.31 (1.43-3.71)*
Age ≥ 35/current smoker 3.91 (1.54-9.89)* 5.36 (2.87-10.0)*

* The comparative groups for these odds ratios were confined to women ≥ 35 years, hence these odds ratios cannot be directly compared with those in the other rows of the table.

The results were in accord with those of another recent study4 that included 408 cases in a managed health-care population of 1.1 million women aged 15-44 years. This American study found very small but non-significant increases in the risk of ischaemic (1.18; 0.54-2.59) and haemorrhagic (1.14; 0.60-2.16) stroke with the use of oral contraceptives. The risk was increased among smokers, but there were insufficient users of oral contraceptives with hypertension to calculate the effect of this risk factor. In this study, the overall association appears to be weaker because the women were more carefully screened for cardiovascular risk factors before commencing oral contraceptive use than in the WHO study.

Check BP before prescribing an OC

These results suggest that the risk of ischaemic and haemorrhagic stroke is low in young women using a low-oestrogen combined oral contraceptive. But the risk of both kinds of events is increased in users of combined oral contraceptives who smoke ≥ 10 cigarettes per day, have hypertension and are aged ≥ 35 years. Every woman should be checked for hypertension before an oral contraceptive is prescribed so that risks associated with this condition can be discussed before making the choice of contraceptive.

Breast Cancer

Breast cancer is another issue related to the safety of oral contraceptives that has been examined in a recent significant study.5 The study was a combined analysis of 54 epidemiological studies that each included ≥ 100 cases of breast cancer. The studies were conducted in 25 countries (including New Zealand) and included a total of 53 297 women with breast cancer and 100 239 women without breast cancer. This study provides the best information that is available at present about the risk of breast cancer in users of oral contraceptives.

Small increased risk of breast cancer not a life-long effect

A relative risk of breast cancer of 1.24 (1.15-1.33) in current users of combined oral contraceptives was found. The relative risk of breast cancer was also slightly elevated in the time periods 1-4 years and 5-9 years after stopping oral contraceptive use, but the study found the reassuring result that there was no increased risk of breast cancer after 10 years (relative risk 1.01; 0.96-1.05). No life-long or dose effect was found.

Cancers of the breast were more localised in those who had used combined oral contraceptives than in those who had never used them. This effect persisted for more than 10 years after discontinuation.

The authors concluded that it is not possible to distinguish whether the small increase in risk in current and recent users of oral contraceptives is due to earlier diagnosis, the biological effects of hormonal contraceptives or a combination of both.

Increase in incidence of breast cancer in older women taking OCs

Another way of presenting the risk is to examine absolute incidence. In developed countries, the cumulative incidence of breast cancer at age 35 is 16 per 10 000 women and at age 45 it is 100 per 10 000 women.6 For women who stop using oral contraceptives at the age of 25, the cumulative incidence at 35 and 45 years is virtually unchanged from the baseline incidence. However, women who stop using oral contraceptives at the age of 40 enter a group with a measurable increase in the incidence of breast cancer, because the baseline incidence is higher. At age 50 their cumulative risk is 199 per 10 000, which is 19 above the baseline rate, and at age 60 their cumulative risk is 394 per 10 000, which is 14 above baseline.

Cervical Cancer

Another study7 examined the risk of cervical cancer associated with oral contraceptive use and employed the Oxford Family Planning Association contraceptive study cohort, 22 years after the women were recruited. In the cohort of 17 000 women aged under 45 years, there were 33 cases of invasive carcinoma, 121 of carcinoma in situ and 159 of dysplasia. The overall relative risk for all types of cervical neoplasia in women who had used oral contraceptives was only slightly above unity at 1.40 (1.00-1.96). The greatest elevation in risk was observed for invasive carcinoma (4.44; 1.04-31.6).

Current or recent long-term use may increase risk

The increased risk of cervical neoplasm with oral contraceptive (combined and progestogen-only) use was largely associated with current or recent (past 2 years) long-term use. Odds ratios for current or recent use were 3.34 (1.96-5.67) for 49-72 months of use, 1.69 (0.97-2.95) for 73-96 months and 2.04 (1.34-3.11) for ≥ 97 months. The data were not able to be analysed for the potential confounding factors of age at first intercourse and number of sexual partners. All the women in the cohort were married and 69% reported having only one sexual partner. The authors concluded that oral contraceptives may have an effect on the later stages of cervical carcinogenesis.

Cervical screening is important in users of OCs

The study provides further evidence of the need for cervical screening in women using oral contraceptives, particularly those on long-term contraception by this means.

Consider all benefits and risks when prescribing OCs

The conclusion of the authors of the WHO Collaborative Study2 gives balance to these risks associated with oral contraceptive use:

Any estimated risk attributable to OC use should be considered in the context of the risks and benefits associated with other forms of contraception together with the effects of OCs on other cardiovascular endpoints, on protection against certain forms of neoplasia, on quality of life, and, ultimately, on overall morbidity and mortality.

Overall these studies provide reassuring evidence of the safety of combined oral contraceptives; the results confirm that the absolute risks of ischaemic and haemorrhagic stroke, and breast and cervical cancer associated with use are small.

References
  1. WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Ischaemic stroke and combined oral contraceptives: results of an international, multicentre, case-control study. Lancet 1996;348:498-505.
  2. WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Haemorrhagic stroke, overall stroke risk, and combined oral contraceptives: results of an international, multicentre, case-control study. Lancet 1996;348:505-10.
  3. WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Venous thromboembolism and combined oral contraceptives: results of international multicentre case-control study. Lancet 1995;346:1575-82.
  4. Petitti DB, Sidney S, Bernstein A et al. Stroke in users of low-dose oral contraceptives. NEJM 1996;335:8-15.
  5. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713-27.
  6. Meirik O. The pill and breast cancer: new information. International Planned Parenthood Federation Medical Bulletin 1996;30(6):1-2.
  7. Zondervan KT, Carpenter LM, Painter R, Vessey MP. Oral contraceptives and cervical cancer - further findings from the Oxford Family Planning Association contraceptive study. Brit J Cancer 1996;73:1291-97.