Oestrogen patches for PMS: small benefit, weak evidence

Premenstrual syndrome brings irritability, low mood, mood swings, bloating, breast tenderness and disrupted sleep in the luteal phase, easing as menstruation ends, and about 3% to 10% of affected women also meet criteria for premenstrual dysphoric disorder. Because the symptoms arise from ovulation, one long-standing treatment logic is simply to stop ovulating. Transdermal oestradiol by patch, gel or implant does that at lower doses than oral contraceptives, but it requires a short monthly course of progestogen to protect the endometrium — and progestogen can itself reproduce PMS-type symptoms. Whether this trade-off nets out in women's favour is the question here.

This Cochrane review searched a wide set of databases and trial registries in March 2016, including the Cochrane Gynaecology and Fertility Group register, MEDLINE, Embase, PsycINFO, CINAHL and international trial registries. Eligible studies were randomised placebo or active controlled trials of non-contraceptive oestrogen-containing preparations in women of reproductive age whose PMS had been confirmed prospectively over at least two cycles, excluding women with a current psychiatric disorder. From 524 potentially relevant articles, five RCTs covering 305 women were included, using oral tablets, transdermal patches and implants; no trial used gels. Two authors independently assessed risk of bias and extracted symptom and adverse-effect data, with GRADE used to rate certainty.

Three studies comparing continuous oestrogen plus progestogen with placebo showed a small to moderate benefit on global symptom scores (SMD -0.34, 95% CI -0.59 to -0.10, P = 0.005, 158 women, I² = 63%), rated very low quality. One small crossover trial in 11 women testing oral luteal-phase oestrogen produced data too poor to analyse, though the authors reported it was ineffective, possibly aggravated PMS, and caused no adverse events. Evidence on withdrawals due to adverse effects was too imprecise to separate the groups (RR 0.64, 95% CI 0.26 to 1.58). A dose comparison of 100 versus 200 µg patches was inconclusive for symptoms but hinted the lower dose carried less overall adverse-event risk.

The review gives qualified support to the ovulation-suppression rationale for PMS, but the certainty is so low that it cannot settle how well continuous oestrogen works or for whom. Heterogeneity between trials was substantial and attrition bias was a recurring problem, so the pooled effect size should be read as a rough signal. Crucially, no study reported long-term outcomes such as endometrial or breast cancer, and trial durations of 2-8 months leave safety largely unexamined. The authors conclude that dose and delivery route may reasonably come down to individual preference, effectiveness and tolerability — an admission of how thin the comparative evidence is.

If you are weighing non-contraceptive oestrogen for premenstrual symptoms, this is a reason to think continuous patches or implants with progestogen may help modestly, while recognising the evidence base is small and rated very low quality. The one clear negative signal is for luteal-phase oral oestrogen, which appeared not to work and may have made symptoms worse, though that came from a handful of women. Longer-term safety simply has not been tested in these trials. This is prescription territory, so the practical route is a conversation with a clinician rather than a conclusion drawn from pooled effect sizes.