This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

Gold, incentives and meh

The latest evidence on COCs and POPs

In the first of a two-part series, Professor John Guillebaud details the latest advice on prescribing oral contraceptives

Combined oral contraceptives (COCs)

Current scientific evidence suggests only two prerequisites for the safe provision of COCs – a careful personal and family history with particular attention to cardiovascular risk factors including migraine2,8 and a well-taken blood pressure7. Routine screening by any blood test, and breast/bimanual examinations are not relevant to the COC per se – do them only if clinically indicated. It is impossible to list every known disease that might have a bearing on COC prescription, and for many the usage data does not exist. A working rule therefore is to ascertain whether or not a pre-existing condition might lead to summation with known major adverse effects of COCs, particularly with the risk of any circulatory disease. In the absence of such risk factors, COCs are broadly usable with alertness for the onset of new risk factors. Reliable protection from pregnancy is often particularly important in these patients.

Breast cancer

A major US study1 of 4,575 women aged 35-64 with breast cancer and matched controls shows no increased risk – whether for current users, past users, long-term users, users starting at a young age or before full-term pregnancy or users with a family history.

The study is reassuring, but it is not the last word. The study did not recruit COC-continuers right up to menopause, nor women with cancer developing before the age of 35. For the latter we still use the 1996 model, equating to one extra case per 1,000 women by age 45 if COC used up to age 352.

Cervical cancer

A good 2002 study3 supports the case for COC being a co-factor, once the HPV oncogen is acquired, in the progression of CIN through to more advanced stages (the risk of invasive cancer is increased fourfold after 10 years' COC use). Yet invasive cancer of the cervix remains a minimal risk provided there is three-yearly cervical cytology plus colposcopy as indicated.

Venous thrombo-embolism (VTE)

While levonorgestrel/norethisterone progestogens reduce VTE risk for a given ethinylestradiol dose, the difference in absolute risk between second and third generation products is tiny. CSM figures from 1999 suggest 100 extra cases of VTE per million users per year.

Assuming a 2 per cent mortality for VTE, that means the difference in annual VTE mortality between third-generation (desogestrel/gestodene) and second-generation (levonorgestrel/ norethisterone) products is two per million – the equivalent of a two-hour car journey a year.

So if a pill-taker chooses to switch from Microgynon to, say, Marvelon or Femodene, as she sensibly may to control a minor side-effect, she can avoid any increased overall VTE risk in the next year by avoiding one two-hour drive2!

Arterial disease

Recent research4 suggests pills with third-generation progestogens may actually be associated with a lower risk of myocardial infarction than levonorgestrel ones, contradicting the MICA study5. Pending more data, I consider we should continue to use 20µg desogestrel/gestodene products for risk-factor and migraine-free COC-users from age 35 to 51, with norethisterone (Loestrin 20) an alternative. Reassuringly, 10 years after use of COCs ceases, mortality in past users is indistinguishable from that in never-users6.

Yasmin (3mg drospirenone/30µg of ethinylestradiol)

Drospirenone differs from other progestogens in COCs in having diuretic properties that may help to oppose the salt- and fluid-retaining effects of ethinylestradiol and so reduce fluid-retention symptoms.

A suggestion that Yasmin may lower blood pressure is so far a unique finding from a small trial, and is not yet proven to be clinically significant.

Drospirenone is an anti-androgen, so Yasmin may be an alternative to Dianette and Marvelon for conditions like polycystic ovarian syndrome. The combination with ethinylestradiol is oestrogen-dominant, and so cannot be expected to have the lower risk of VTE associated with the functionally anti-oestrogenic levonorgestrel in, for example, Microgynon.

Yasmin should not be used – if COCs are usable at all – in anyone at risk of high potassium levels – those with severe renal insufficiency, hepatic dysfunction, and adrenal insufficiency.

Yasmin should be considered when the need for contraception is accompanied by a need for oestrogen/anti-androgen therapy (significant acne, PCOS) – but only if the first-choice (Marvelon or other oestrogen-dominant, ordinary COC) is judged unsatisfactory. Marvelon is best for maintenance treatment after Dianette.

Yasmin is also useful second-line for empirical control of a minor side-effect, especially one judged to be linked to fluid retention.

Weight is not usually an indication, given the lack of evidence of a true (non-fluid related) weight benefit and its extra cost. A levonorgestrel/ norethisterone COC would be preferred for BMIs of 30 or more and all ethinylestradiol-containing products (COCs, Evra, NuvaRing) are contraindicated at BMI 39 or higher.

Evra transdermal combined hormonal contraception

This innovative transdermal patch delivers ethinylestradiol with norelgestromin, the active metabolite of norgestimate, producing blood levels in the range of those after a tablet of Cilest.

Contraindications and most practical management advice are the same as for COCs. It appears to be relatively oestrogen dominant, and about 2 per cent of women in the trials had local skin reactions which led to discontinuation.

The patch has excellent adhesion, the incidence of detachment was 1.8 per cent (complete) and 2.9 per cent (partial). Failure rate for consistent users of Evra was similar to the oral pills – less than one per 100 woman-years.

Failure is more likely if body weight is greater than 90kg, so avoid in these women.

Each patch is worn for seven days, for three consecutive weeks followed by a patch-free week. There is a useful two-day margin for error for late patch changes – and users can request a weekly reminder by text message (by visiting

As with the COC it is essential never to lengthen the contraception-free (patch-free) interval. If this interval exceeds eight days for any reason (either through late application, or the first new patch detaching in the first week and this being identified late), extra precautions are required for the duration of the first freshly applied patch (ie seven days). This should be after immediate emergency contraception if there has been sexual exposure during the preceding patch-free time.

During any short-term enzyme inducer therapy, and for 28 days after this ends, additional contraception is advised plus elimination of any patch-free intervals during this time.

Progestogen-only pills (POPs)

Cerazette (desogestrel 75µg)

The first POP to be primarily an anovulant, Cerazette, blocks ovulation in about 97 per cent of cycles, in addition to its mucus effect, so this is a valuable alternative if COCs are contraindicated in a young woman. In low fertility states like lactation or women over 40, where POPs are already so effective, cheaper old-type POPs suffice.

It can be useful as a trial before inserting Implanon, though this risks Cerazette's usual early bleeding problems. It is an option where there is a high risk of thrombosis or (unlike other POPs) a past ectopic pregnancy. Often, but not entirely predictably, it benefits menstrual symptoms, such as dysmenorrhoea, menorrhagia, PMS and mittelschmerz.

Irregular bleeding may occur in early months, usually improving to offer 50 per cent oligo-amenorrhoea at one year. Extra precautions are still advised after a delay in pill-taking of three hours or more – but new WHO practice recommendations say these need only be for 48 hours for all POPs, including Cerazette.

John Guillebaud is professor emeritus of family planning and reproductive health, University College London


1 Marchbanks P et al. Oral contraceptives and the risk of breast cancer.

New Engl J Med 2002; 346:2025-32

2 Guillebaud J. Contraception Today. A Pocketbook for GPs (5th Ed) London: Martin Dunitz, 2004, pp 26-37, 42-5, 73-4, 86, 96-8, 102-5, 141

3 Moreno V et al. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002; 359:1085-92

4 Tanis B et al. Oral contraceptives and the risk of myocardial infarction.

New Engl J Med 2002; 345:1787-1793

5 Dunn N et al. Oral contraceptives and myocardial infarction: the results of the MICA case-control study. BMJ 1999; 319:795-6

6 Beral V et al. Mortality associated with oral contraceptive use: 25-year follow-up of cohort of 46,000 women from RCGP OC study. BMJ 1999;318:96-100

7 Hannaford P, Webb A. Evidence-guided prescribing of combined oral contraceptives: consensus statement. Contraception 1996; 54:125-9

8 Guillebaud J. Contraception: Your Questions Answered (4th Ed) Oxford: OUP, 2004, pp 199-212, 213-218, 535 and other pages expanding on those at ref [2].

Further reading

•Cooper A, Guillebaud J. Sexuality and Disability. London: Radcliffe Medical Press, 1999. [An often neglected subject]

•Kubba A et al (eds). Contraception and Office Gynaecology: choices in reproductive health care. London: Saunders, 1999

•McPherson A, Waller D (eds). Women's Health – Oxford General Practice Series. Oxford: Oxford University Press, 2003

This article represents the personal opinions of John Guillebaud, based wherever possible on published and sometimes unpublished evidence. Health care professionals must understand they take ultimate responsibility for their patient and ensure that any clinical advice they use from herein is applicable to the specific circumstances they encounter. The author has received payments for research projects, lectures, ad hoc consultancy work and related expenses from the manufacturers of contraceptive products.

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say