Review question
What are the benefits and risks of using medication known as gonadotropin-releasing hormone (GnRH) agonists and antagonists in the management of premenstrual syndrome (PMS)?
Key messages
• We found reliable evidence that GnRH agonists without add-back improve PMS symptoms compared to placebo, but also that menopausal side effects are common, and women taking GnRH agonists without add-back are more likely to stop their treatment than women taking placebo.
• We are moderately confident that PMS symptoms may improve when comparing GnRH agonists with add-back versus placebo. There was not enough evidence to decide if GnRH agonists with add-back improves PMS symptoms compared to GnRH alone, or if dosage of add-back hormones has an effect on PMS symptoms.
• Further studies known as 'randomised controlled trials' of GnRH agonists with add-back and long-term follow-up are needed.
What is premenstual syndrome?
Premenstrual syndrome (PMS) comprises a range of physical, psychological and behavioural symptoms that occur after release of an egg from the ovary (ovulation), disappear by the end of a woman's period and cause substantial distress or impairment to daily life, including work, school, social activities, hobbies and interpersonal relationships. Stopping ovulation using GnRH analogues may suppress those symptoms, but the downside is that they may cause menopause-like side effects, such as hot flushes, and - in the long term - osteoporosis. To counteract these side effects, another hormone (most often oestrogen or progestogen) can be added to the treatment; this is known as add-back.
What did we want to find out?
Gonadotropin-releasing hormone (GnRH) analogues are peptides (proteins) with a structure related to GnRH that are widely used to prevent ovulation. They are a group of drugs that influence the part of the brain containing the hypothalamus and pituitary gland. GnRH analogues have two types: agonists and antagonists. GnRH agonists initially stimulate GnRH production but with longer use suppress GnRH production. GnRH antagonists immediately suppress GnRH production. We wanted to find out if they were effective and safe. We searched a type of study known as a 'randomised controlled trial', in which people are assigned to one of two (or more) groups randomly, with less risk of bias in the results.
Study characteristics
We found 11 studies (randomised controlled trials) of the use of GnRH analogues in the management of PMS, enroling a total of 265 women who were clinically diagnosed with PMS. We used these RCTs in four comparisons: GnRH agonists without add-back versus placebo (9 studies, 173 women), GnRH agonists with add-back versus placebo (1 study, 31 women), add-back versus placebo during GnRH agonist treatment (2 studies, 60 women) and a comparison of different doses of add-back (1 study, 15 women). No study reported quality of life or long-term risks such as osteoporosis.
Key results
We found reliable evidence that GnRH agonists without add-back improve overall premenstrual symptoms, compared to placebo. However, women using GnRH agonists are more likely to withdraw from treatment due to negative side effects (such as menopause-like symptoms).
We found low-certainty evidence that GnRH agonists with add-back may improve global symptoms compared to placebo, while insufficient evidence was found about negative side effects.
The evidence for add-back versus placebo during GnRH agonist treatment and about different doses of add-back hormones was too imprecise to decide which worked best.
None of the included studies reported on quality of life or on long-term risks such as osteoporosis.
What are the limitations of the evidence?
The main limitations of the evidence are the small number of women included in most of the studies, and the women in the studies sometimes being aware of the treatment they were getting. We are confident about the results of GnRH agonists without add-back compared to placebo, but have little confidence in the results of the other three comparisons.
How up to date is the evidence?
The evidence is based on searches of healthcare databases that were run up to May 2023.
This review found that GnRH agonists without add-back improved global symptoms of premenstrual syndrome. However, the induced menopausal side effects and potential complications preclude long-term use. We found insufficient evidence to suggest that side effects can be reduced by ‘add-back’ without decreasing the global efficacy of GnRH agonists. Further RCTs of GnRH agonists with add-back and long-term follow-up are therefore needed to provide firm conclusions about long-term use. Until data are available to confirm or refute the safety of this combination, GnRH agonists with or without add-back can be administered to provide a short-term break from PMS symptoms. Future studies should attempt to assess the risk of osteoporosis.
Premenstrual syndrome (PMS) is a psychological and somatic disorder affecting 20% to 30% of women of reproductive age. PMS results from ovulation: symptoms recur during the luteal phase of the menstrual cycle and remit by the end of menstruation. Premenstrual dysphoric disorder (PMDD) is a severe form of PMS experienced by three to eight per cent of menstruating women. In this review, we use the term PMS to cover all core premenstrual disorders, including PMDD. The symptoms of all types are severe enough to affect daily functioning, interfering with work, school performance or interpersonal relationships.
Gonadotropin-releasing hormone (GnRH) analogues are a pharmacological treatment to suppress ovulation. They can be administered as GnRH-agonists or GnRH-antagonists, though currently, GnRH-antagonists are not generally used to treat PMS.
Suppressing ovarian function induces a hypo-oestrogenic state that can cause menopausal side effects such as hot flushes and mood changes. Having menopausal side effects instead of PMS symptoms can be distressing and can confuse clinical management. Longer-term GnRH therapy carries the risk of osteoporosis. To counteract these adverse effects, oestrogen or progestogen can be added to the PMS treatment; this is known as 'add-back' therapy or simply 'add-back'. Add-back may reduce menopausal side effects, allowing GnRH therapy to be used for a longer period without reducing efficacy.
To evaluate the therapeutic effectiveness and safety (adverse effects) of GnRH analogues (agonists or antagonists), with or without add-back, in the management of PMS.
A Cochrane Information Specialist searched the Cochrane Gynaecology and Fertility (CGF) Group trials register, CENTRAL, MEDLINE, Embase, PsycINFO and two trials registers on 29 May 2023. We also checked reference lists and contacted study authors and subject experts to identify additional studies.
We included randomised controlled trials (RCTs) of GnRH analogues used in the management of PMS in women of reproductive age with PMS diagnosed by at least two prospective menstrual cycles and no current psychiatric disorder. Control conditions could be no treatment, placebo, another type of GnRH, another dosage of GnRH or add-back.
We used the standard methodological procedures recommended by Cochrane. Our primary outcomes were overall severity of PMS symptoms (global symptoms), quality of life and adverse events.
The review found 11 RCTs that analysed results from 275 women. The evidence is of very low to high certainty. The evidence is limited by serious imprecision due to low sample sizes and a high risk of bias related to blinding and attrition.
Quality of life and long-term effects on bones were not reported in the studies. Most studies did not report on all our adverse events of interest; some did not report on any of them.
We found no RCTs that evaluated GnRH antagonists.
GnRH agonists (without add-back) versus placebo
GnRH agonists (without add-back) improve global symptoms compared to placebo (SMD −1.23, 95% CI −1.76 to −0.71; 9 RCTs, 173 women, effective sample size 278; I2 = 72%; high-certainty evidence).
GnRH agonists may increase the risk of menopausal side effects compared to placebo (RR 1.93, 95% CI 0.83 to 4.48; 2 RCTs, 23 women, effective sample size 31; low-certainty evidence). If the risk of menopausal side effects with placebo is 21%, the risk with GnRH agonist without add-back would be between 18% and 96%.
Seven RCTs reported on withdrawals from the study due to adverse events (though data extraction was not possible for one of these studies). Women using GnRH agonists have a higher risk of withdrawing due to adverse events than women using placebo (RR 4.24, 95% CI 1.10 to 16.36; 6 RCTs, 140 women, effective sample size 252; high-certainty evidence). If the risk of withdrawal from the study is 0.8% with placebo, the withdrawal risk with GnRH agonist without add-back would be between 0.8% and 13%.
GnRH agonists (with add-back) versus placebo
GnRH agonists with add-back may improve global symptoms compared to placebo (MD −3.89, 95% CI −6.19 to −1.59; 1 RCT, 31 women; low-certainty evidence).
We are very uncertain of the effect on withdrawal due to adverse events (RR 2.86, 95% CI 0.12 to 66.44; 1 RCT, 41 women; very low-certainty evidence).
Add-back versus placebo add-back during GnRH agonist treatment
The evidence of add-back versus placebo during GnRH agonist treatment was too imprecise to decide if there was an effect on global symptoms. The analysis was stratified by type of add-back (tibolone or oestrogen/progesterone). One RCT investigated the effect of tibolone as add-back and found insufficient information to decide if there was an effect on global symptoms (SMD −0.37, 95% CI −0.11 to 0.38; 1 RCT, 28 women; very low-certainty evidence). One RCT investigated the effect of oestrogen plus cyclical progestogen as add-back and found evidence that it may worsen global symptoms compared to placebo (SMD 0.90, 95% CI 0.17 to 1.63; 1 RCT, 32 women; low-certainty evidence).
We are very uncertain of the effect of tibolone on withdrawal due to adverse events (RR not estimable; 1 RCT, 28 women; very low-certainty evidence), and of oestrogen plus cyclical progestogen (RR 0.90, 95% CI 0.06 to 13.48; 1 RCT, 40 women; very low-certainty evidence).
Add-back dose comparison (low dose versus standard dose)
The evidence was too imprecise to determine if a lower add-back dose during GnRH agonist treatment affected global PMS symptoms (MD −11.90, 95% CI −28.51 to 4.71; 1 RCT, 15 women; low-certainty evidence).