NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women’s and Children’s Health (UK). Fertility: Assessment and Treatment for People with Fertility Problems. London: Royal College of Obstetricians & Gynaecologists; 2013 Feb. (NICE Clinical Guidelines, No. 156.)

Cover of Fertility

Fertility: Assessment and Treatment for People with Fertility Problems.

Show details

8Ovulation disorders

8.1. Introduction

Ovulation disorders, presenting as menstrual disturbance, are the cause of infertility in around 25% of couples who have difficulty conceiving. The World Health Organization (WHO) categorises ovulation disorders into three groups:

  • Group I ovulation disorders are caused by hypothalamic pituitary failure. This category includes conditions such as hypothalamic amenorrhea and hypogonadotrophic hypogonadism. Typically, women present with amenorrhoea (primary or secondary) which is characterised by low gonadotrophins and oestrogen deficiency. Approximately 10% of women with ovulation disorders have a group I ovulation disorder.
  • Group II ovulation disorders are defined as dysfunctions of the hypothalamic-pituitary-ovarian axis. This category includes conditions such as polycystic ovary syndrome and hyperprolactinaemic amenorrhoea. Around 85% of women with ovulation disorders have a group II ovulation disorder.
  • Group III ovulation disorders are caused by ovarian failure. Around 5% of women with ovulation disorders have a group III ovulation disorder.

This chapter focuses on the management of women with WHO group I or group II ovulation disorders. These two groups of disorders can be managed with drug treatments, lifestyle modifications and/or surgical interventions. Women with a group III ovulation disorder (‘ovarian failure’) can only conceive through oocyte donation and then IVF treatment (see Chapters 18 and 15, respectively).

8.2. WHO Group I ovulation disorders

Introduction

WHO Group I ovulation disorders, also known as hypogonadotrophic hypogonadism, are caused by hypothalamic pituitary failure. Women with these conditions typically present with amenorrhoea (primary or secondary), often called hypothalamic amenorrhoea, which is characterised by low gonadotrophins levels and oestrogen deficiency.

Hypogonadotrophic hypogonadism has usually an unknown cause. However, it may be congenital, for example when it is associated with anosmia it is known as Kallmann’s syndrome. Hypothalamic amenorrhoea commonly develops as a result of low body weight or excessive exercise. Hypopituitarism is uncommon and, as with all causes of infertility, must be appropriately investigated before ovulation induction is considered.

Treatment of WHO Group I ovulation disorders depends on the diagnosis. Treatment options include:

  • lifestyle interventions (normalising weight and exercise)
  • pulsatile gonadotrophin-releasing hormone (GnRH) (‘GnRH pump’)
  • gonadotrophins (human menopausal gonadotrophin [hMG]).

Review question

What is the effectiveness and safety of ovulation induction strategies in women with WHO Group I ovulation disorders?

Description of included studies

In the 2004 version of this guideline two studies were identified examining the value of pulsatile GnRH in women with WHO Group I ovulation disorders. One was a case series study which reported the use of pulsatile GnRH in women with WHO Group I ovulation disorders and a study comparing hMG with pulsatile GnRH. Evidence from these two studies is reported below.

No prospective comparative studies were found in the 2004 or 2013 reviews that reported on the use of gonadotrophins, GnRH analogues or lifestyle interventions for women with WHO Group I ovulation disorders.

Evidence profile

Five reviews were undertaken to answer this review question. These were a comparison of:

  • drugs compared with no treatment or placebo for women with WHO Group I ovulation disorders
  • different types of drugs for women with WHO Group I ovulation disorders
  • lifestyle interventions compared with no treatment or placebo for women with WHO Group I ovulation disorders
  • different lifestyle interventions for women with WHO Group I ovulation disorders
  • lifestyle interventions versus drugs.

Pulsatile gonadotrophin-releasing hormone

In case series studies, pulsatile GnRH induces ovulation, achieving cumulative pregnancy rates of up to 82% in women with hypogonadotrophic hypogonadism and 95% in women with weight-related amenorrhoea after 12 cycles. The corresponding figures for live birth rates were 65% and 85%, respectively.571573 [Evidence level 3]

A study comparing hMG with pulsatile GnRH reported no difference in multiple gestation rates (14.8% versus 8.3%) but a lower rate of triplets in the pulsatile GnRH group.575 [Evidence level 2b]

Evidence to recommendations

Relative value placed on the outcomes considered

Clinical pregnancies and live full-term singleton births were selected as the primary outcomes since they allow clinicians to inform women of their chances of conception and consequent live birth. However, both studies only reported live birth rates and not live full-term singleton live births. Secondary outcomes relating to adverse effects of the treatments were also searched for in the evidence as they provide women with information of the potential risks of treatment.

Trade-off between clinical benefits and harms

The evidence that was identified on pulsatile GnRH was of very low quality. The guideline development group (GDG) highlighted that the population in the case series data was not the same as the population considered in this question, but as this was the only data identified that considered GnRH it was included. The case series data suggested that pulsatile GnRH improves pregnancy and live birth rates and reduces the risk of triplets. There was no evidence identified for any of the other ovulation induction strategies covered by the clinical question. Consequently, the GDG did not make recommendations on interventions other than pulsatile GnRH. The evidence from case series concurred with the GDG members’ clinical opinions.

Quality of evidence

The quality of the evidence was very low. Nevertheless, the benefits of pulsatile GnRH identified in the case series data concurred with the clinical experience of the GDG members. Therefore, the GDG considered that the 2004 recommendation on pulsatile GnRH reflected standard practice and that, in the absence of any new evidence, it should remain unchanged in the guideline.

Other considerations

The GDG empasised that appropriate expertise is needed when using pulsatile GnRH.

The GDG’s clinical opinion was that a low body mass index (BMI), irregular menstruation or amenorrhea and/or a high level of exercise are associated with anovulation. To establish evidence for this would require studies that included women with these risk factors and were of sufficient power to undertake subgroup analyses. The GDG acknowledged that such studies were unlikely to be undertaken. In the absence of evidence, the GDG considered that advice to women with a low BMI to increase their weight and to moderate high levels of exercise was very unlikely to be harmful and could be beneficial. Therefore it should be considered as part of the initial advice offered to women seeking treatment for ovulation disorders (see Chapter 5). This might include information from a dietician, warnings of the potential risks in pregnancy and, if appropriate, the offer of access to exercise advice and psychosocial support.

Equalities

The people considered in this review were

  • People in same sex relationships who cannot have heterosexual intercourse.
  • Specific patient subgroups listed in the guideline Scope who may need specific consideration:
    • people in same-sex relationships who have unexplained infertility after donor insemination
    • people who are unable to, or would find it very difficult to, or who have been advised not to, have heterosexual intercourse
    • people with conditions or disabilities that require specific consideration in relation to methods of conception.
  • People who are preparing for cancer treatment who may wish to preserve their fertility.

There were no specific issues that needed to be addressed with respect to any of these subgroups for this review.

Recommendations

NumberRecommendation
90Advise women with WHO Group I anovulatory infertility that they can improve their chance of regular ovulation, conception and an uncomplicated pregnancy by:
  • increasing their body weight if they have a BMI of less than 19 and/or
  • moderating their exercise levels if they undertake high levels of exercise. [new 2013]
91Offer women with WHO Group I ovulation disorders pulsatile administration of gonadotrophin-releasing hormone or gonadotrophins with luteinising hormone activity to induce ovulation. [2013]

8.3. WHO Group II ovulation disorders

Introduction

Polycystic ovary syndrome (PCOS) is a heterogenous group of disorders affecting 5–10% of women of reproductive age and is the most commonly encountered type of WHO Group II ovulation disorder. Common clinical features of PCOS include oligo- or amenorrhoea, anovulatory infertility, obesity and hyperandrogenism. Insulin resistance plays an important role in the pathogenesis of the disorder. Ultrasound examination of the ovaries reveals characteristic appearances, with multiple (12 or more) small antral follicles present. In 2003, the Rotterdam consensus meeting (sponsored by European Society of Human Reproductive and Embryology [ESHRE/American Society for Reproductive Medicine [ASRM]) agreed a definition for PCOS, namely the presence of at least two of the following three criteria with the exclusion of other causes of menstrual cycle disturbance or androgen excess:

  • oligo-ovulation and/or anovulation,
  • hyperandrogenism (clinical and/or biochemical)
  • polycystic ovaries on ultrasound scan;.

Options for treatment include:

  • weight loss
  • medical treatment
  • second-line treatments including laparoscopic ovarian diathermy (LOD) and injectable gonadotrophin ovulation induction
  • assisted conception (usually in vitro fertilisation [IVF]).

Obesity is associated with increased insulin resistance and an exacerbation of PCOS. Weight loss is therefore often the first line treatment for obese PCOS patients.

Medical treatment of anovulatory infertility due to PCOS is often initially undertaken with the oral anti-oestrogen clomifene citrate and/or the oral insulin sensitising agent metformin hydrochloride (though metformin is unlicensed for this indication). Clomifene is associated with a multiple pregnancy rate of around 10% and so ultrasound follicular monitoring, particularly in the first cycle of treatment, is indicated. The requirement for access to scan monitoring may limit the ability for prescribing in primary care. Conventionally, clomifene is taken as a single daily dose for 5 days from early in the menstrual cycle. If ovulation is not achieved at the lowest dose (usually 50 mg) then in subsequent cycles the dose is escalated. If no ovulation occurs at doses of 100–150 mg daily then the term ‘clomifene resistance’ is used. Metformin is taken every day in divided doses and since the aim is to restore ‘normal’ mono-ovulation then arguably no scan monitoring is required. The most common side-effect is gastro-intestinal upset.

Potential advantages of laparoscopy include the ability to assess the pelvis for additional treatable causes of infertility, such as endometriosis and/or adhesions, and to assess tubal patency. An electrical current (diathermy) is applied to a number of points on each ovary. If successful, then mono-ovulation occurs which can continue for months and/or years without the need for ultrasound scan monitoring. Risks of LOD include those associated with surgery and general anaesthesia, and a low risk of causing ovarian damage and/or peri-ovarian adhesions.

Gonadotrophin ovulation induction involves sub-cutaneous injections once daily for around 10–20 days per cycle. Frequent ultrasound scan monitoring is required and risks include multiple pregnancy and, uncommonly, ovarian hyperstimulation syndrome (OHSS).

Assisted conception is the third-line treatment option for WHO Group II ovulation disorders. The most important risks of IVF are OHSS (particularly for women with PCOS) and multiple pregnancy (see Chapter 15).

Hyperprolactinaemic amenorrhoea is another, though much less common, WHO Group II ovulation disorder. Clinically, in addition to amenorrhoea and infertility, women with the condition have galactorrhoea. The most common source of the excess prolactin production is a pituitary microadenoma. Treatment is with dopamine agonists.

The evidence for the clinical effectiveness and safety of these interventions for WHO Group II ovulation disorders is reviewed in this section.

Growth hormone as an adjunct to ovulation induction therapy

For women with clomifene citrate-resistant PCOS, co-treatment with recombinant human growth hormone plus gonadotrophin-releasing hormone agonist (GnRHa), or growth hormone plus hMG, has no significant effect on the amount and duration of hMG used, ovulation (respectively: 93% versus 93%; 88% versus 100%) and pregnancy rates (respectively: 26% versus 20%; 25% versus 13%) when compared with GnRHa and hMG alone.569 [Evidence level 1b] It has been suggested that co-treatment with growth hormone may improve ovarian responses to exogenous gonadotrophins, thus reducing the overall gonadotrophin requirement.570

Pulsatile gonadotrophin-releasing hormone

A systematic review of three RCTs, one non-RCT and 18 uncontrolled case series studies found insufficient evidence for or against a beneficial effect of pulsatile GnRH in women with clomifene citrate-resistant PCOS when compared with other ovulatory agents (hMG, follicle-stimulating hormone [FSH], with and without pretreatment with GnRHa).574 [Evidence level 1a]

Review question

What is the effectiveness and safety of ovulation induction strategies in women with WHO Group II ovulation disorders?

Evidence profile

The evidence is presented separately for women receiving first line treatment for WHO Group II ovulation disorders and for those who are known to be clomifene resistant. Treatments were compared in three main groups:

  • drugs currently used as standard treatment compared with non-standard drugs
  • surgical interventions compared with drugs
  • lifestyle modifications (such as changes to diet and level of exercise) compared with drugs and/or surgery.

The evidence is presented in the following profiles:

  • Ovarian stimulation as first-line treatment in women with WHO Group II ovulation disorders:
    • clomifene citrate or tamoxifen compared with other drugs (see Table 8.2)
    • surgery compared with drugs (see Table 8.3)
    • lifestyle modification compared with drugs and/or surgery (see Table 8.4).
  • Ovarian stimulation treatment in women with WHO Group II ovulation disorders who are known to be clomifene citrate resistant:
    • metformin plus clomifene compared with other drugs (see Table 8.5)
    • surgery compared with drugs (see Table 8.6)
    • lifestyle compared with drugs and/or surgery (see Table 8.7).
Table 8.2. GRADE findings for comparison of clomifene citrate or tamoxifen with other drugs (first-line treatment for PCOS).

Table 8.2

GRADE findings for comparison of clomifene citrate or tamoxifen with other drugs (first-line treatment for PCOS).

Table 8.3. GRADE findings for surgery compared with drugs (first-line treatment for PCOS).

Table 8.3

GRADE findings for surgery compared with drugs (first-line treatment for PCOS).

Table 8.4. GRADE findings for comparison of lifestyle modification compared with drugs or surgery (first-line treatment for PCOS).

Table 8.4

GRADE findings for comparison of lifestyle modification compared with drugs or surgery (first-line treatment for PCOS).

Table 8.5. GRADE findings for comparison of other drugs with clomifene plus metformin (clomifene resistant PCOS).

Table 8.5

GRADE findings for comparison of other drugs with clomifene plus metformin (clomifene resistant PCOS).

Table 8.6. GRADE findings for comparison of surgery with drugs (clomifene resistant PCOS).

Table 8.6

GRADE findings for comparison of surgery with drugs (clomifene resistant PCOS).

Table 8.7. GRADE findings for comparison of lifestyle with drugs or surgery (clomifene resistant PCOS).

Table 8.7

GRADE findings for comparison of lifestyle with drugs or surgery (clomifene resistant PCOS).

Definitions

The studies used various definitions of PCOS and also of clomifene citrate resistance, particularly in studies conducted prior to 2003 when the Rotterdam consensus criteria regarding PCOS were published. These are outlined in Table 8.1.

Table 8.1. The definition of PCOS and clomifene citrate resistance variation across studies.

Table 8.1

The definition of PCOS and clomifene citrate resistance variation across studies.

First-line ovarian stimulation treatment for women with polycystic ovary syndrome (PCOS)

Clomifene citrate or tamoxifen compared with other drugs

Fourteen of the 29 papers reported on trials of clomifene citrate or tamoxifen compared with other drugs as first-line ovarian stimulation treatment in women with PCOS (Atay et al., 2006; Badawy et al., 2009; Bayar et al., 2006; Dasari et al., 2009; Dehbashi et al., 2009; Elsedeek et al., 2011; Johnson et al., 2010; Karimzadeh et al., 2010; Legro et al., 2007; Lopez et al., 2004; Moll et al., 2006; Palomba et al., 2005; Sahin et al., 2004; and Zain et al., 2009).

Surgery compared with drugs

No papers reported on trials of surgery compared with drugs for first-line ovarian stimulation treatment in women with PCOS.

Lifestyle modification compared with drugs or surgery

One paper reported on a trial comparing a low calorie diet with exercise compared with clomifene citrate as a first-line ovarian stimulation treatment in women with PCOS (Karimzadeh et al., 2010). Only women with a BMI of 25–29.9 were included in the study. Another paper reported on a trial comparing a low calorie diet to metformin (Qublan et al., 2007). Only women with a BMI of over 30 were included in the study.

Ovarian stimulation treatment in women who are clomifene citrate resistant

Metformin plus clomifene compared with other drugs

Nine papers reported on trials that compared metformin in combination with clomifene citrate with other drugs as ovarian stimulation treatment for women with PCOS who were resistant to clomifene citrate (Abu Hashim et al., 2010; Begum et al., 2009; Badawy et al., 2008; Cheng et al., 2010; George et al., 2003; Hwu et al., 2005; Malkawi & Qublan, 2002; Sohrabvand et al., 2006; Vandermolen et al., 2001).

Surgery compared with drugs

Six papers reported on trials that compared drugs with surgery as treatments to stimulate the ovaries in women with PCOS who were clomifene citrate resistant (Abdel et al., 1990; AbuHashim et al., 2010; Bayram et al., 2004; Farquhar et al., 2002; Kamel et al., 2004; Zakherah et al., 2010).

Lifestyle compared with drugs or surgery

No papers were found that reported trials of lifestyle modifications compared with drugs or surgery or other lifestyle modifications in women with PCOS who are clomifene citrate resistant.

Evidence statements

First line ovarian stimulation treatment for women with PCOS
Clomifene citrate or tamoxifen compared with other drugs
Live full-term singleton birth

There was no significant difference in the number of live full-term singleton births when comparing metformin, metformin plus clomifene citrate, letrozole or FSH to clomifene citrate alone.

There were significantly more live births with metformin plus clomifene citrate than metformin alone.

Clinical pregnancy

There was no significant difference in the number of clinical pregnancies with metformin compared with clomifene citrate alone.

There were significantly more clinical pregnancies with metformin plus clomifene citrate compared with clomifene citrate alone or metformin alone. There were significantly more clinical pregnancies with letrozole compared with clomifene citrate, and when using recombinant follicle-stimulating hormone (rFSH) compared with clomifene citrate.

Adverse pregnancy outcomes

There were no significant differences between metformin and clomifene citrate in the number of miscarriages, ectopic pregnancies, cases of gestational hypertension, cases of gestational diabetes, women with preterm labour or premature rupture of membranes, intrauterine fetal deaths, cases of placenta previa, cases of postpartum haemorrhage, placental abruptions, second or third trimester pregnancy losses, cervical incompetence or preterm labour, cases of severe pre-eclampsia, cases of HELLP syndrome (a severe form of pre-eclampsia comprising haemolysis, elevated liver enzymes and low platelets), or number of maternal deaths.

There were no significant differences between metformin plus clomifene citrate compared with clomifene citrate alone in the number of maternal deaths, preterm births, miscarriages, second or third trimester pregnancy losses, ectopic pregnancies or cases of: gestational diabetes, gestational hypertension, pre-eclampsia, severe pre-eclampsia, HELLP syndrome, preterm labour or premature rupture of membranes, preterm labour or cervical incompetence, placental abruption, placenta previa, or postpartum haemorrhage.

There were no significant differences between metformin compared with metformin plus clomifene citrate in the number of maternal deaths, miscarriages, ectopic pregnancies, second or third trimester pregnancy loss, or cases of: cervical incompetence of preterm labour, gestational hypertension, mild pre-eclampsia, severe pre-eclampsia, HELLP syndrome, gestational diabetes, pre-term labour or premature rupture of membranes, placental abruption, placenta previa or postpartum haemorrhage.

There were no significant differences in the number of miscarriages per woman or per pregnancy when comparing letrozole to clomifene citrate, or when comparing rFSH to clomifene citrate.

Multiple pregnancies

There were no significant differences in the number of multiple pregnancies when comparing metformin, metformin plus clomifene citrate, letrozole or rFSH to clomifene citrate alone. There was no significant difference in the number of multiple pregnancies when comparing metformin to metformin plus clomifene citrate.

Multiple births

No evidence was reported regarding multiple births.

OHSS

There were no significant differences in the number of cases of OHSS when comparing letrozole plus hCG to clomifene citrate plus hCG, or when comparing rFSH plus hCG to clomifene citrate plus hCG.

Congenital abnormalities

There were no significant differences in the number of congenital abnormalities when comparing metformin, metformin plus clomifene citrate, or letrozole to clomifene citrate alone. There was no significant difference in the number of congenital abnormalities when comparing metformin plus clomifene to metformin alone.

Patient satisfaction

No evidence was reported regarding patient satisfaction.

Health related quality of life

No evidence was reported regarding health related quality of life.

Anxiety and/or depression

There were no significant differences in the number of women with anxiety and/or depression when comparing metformin or metformin plus clomifene citrate with clomifene citrate alone. There was also no significant difference when comparing metformin plus clomifene citrate to metformin alone.

Surgery compared with drugs
Live full-term singleton birth

No evidence was reported regarding live births.

Clinical pregnancy

No evidence was reported regarding clinical pregnancy.

Adverse pregnancy outcomes

No evidence was reported regarding adverse pregnancy outcomes.

Multiple pregnancies

No evidence was reported regarding multiple pregnancies.

Multiple births

No evidence was reported regarding births from multiple pregnancies.

OHSS

No evidence was reported regarding cases of OHSS.

Congenital abnormalities

No evidence was reported regarding congenital abnormalities.

Patient satisfaction

No evidence was reported regarding patient satisfaction.

Health related quality of life

No evidence was reported regarding health related quality of life.

Anxiety and/or depression

No evidence was reported regarding the number of women with anxiety and/or depression.

Lifestyle modification compared with drugs or surgery
Live full-term singleton birth

No evidence was reported regarding live births.

Clinical pregnancy

There were no significant differences in the number of clinical pregnancies when comparing lifestyle modification (low calorie diet plus exercise) with clomifene citrate alone, metformin alone, or clomifene citrate plus metformin.

Adverse pregnancy outcomes

No evidence was reported regarding adverse pregnancy outcomes.

Multiple pregnancies

There were no significant differences in the number of multiple pregnancies when comparing lifestyle modification (low calorie diet plus exercise) with clomifene citrate alone, metformin alone, or clomifene citrate plus metformin.

Multiple births

No evidence was reported regarding births from multiple pregnancies.

OHSS

No evidence was reported regarding cases of OHSS.

Congenital abnormalities

No evidence was reported regarding congenital abnormalities.

Patient satisfaction

No evidence was reported regarding patient satisfaction.

Health related quality of life

No evidence was reported regarding health related quality of life.

Anxiety and/or depression

No evidence was reported regarding the number of women with anxiety and/or depression.

Ovarian stimulation treatment in women who have clomifene citrate resistance
Metformin plusclomifene compared with other drugs
Live full-term singleton birth

There were significantly more live full-term singleton births after metformin plus clomifene citrate compared with clomifene citrate alone. There were significantly more live full-term singleton births after letrozole plus metformin compared with metformin plus clomifene citrate. There was no significant difference when comparing hMG to metformin plus clomifene citrate.

Clinical pregnancy

There were significantly more clinical pregnancies after metformin plus clomifene citrate compared with clomifene citrate alone, and after uFSH compared with metformin plus clomifene citrate. There were no significant differences when comparing hMG to metformin plus clomifene citrate, letrozole to clomifene citrate, or letrozole plus metformin to metformin plus clomifene citrate.

Adverse pregnancy outcomes

There was no significant difference in the number of miscarriages per woman or per pregnancy when comparing clomifene citrate to metformin plus clomifene citrate.

There were no significant differences in the number of miscarriages, intrauterine deaths at 28 weeks, or the number of ectopic pregnancies per woman or per pregnancy when comparing hMG to metformin plus clomifene citrate.

There was no significant difference in the number of miscarriages when comparing letrozole or hMG to clomifene citrate.

There were no significant differences in the number of miscarriages when comparing metformin plus clomifene citrate to letrozole plus metformin or to uFSH.

Multiple pregnancies

There was no significant difference in the number of multiple pregnancies when comparing metformin plus clomifene citrate, letrozole, or hMG to clomifene citrate alone. There was no significant difference in the number of multiple pregnancies when comparing uFSH or letrozole to metformin plus clomifene citrate.

Multiple births

No evidence was reported regarding births from multiple pregnancies.

OHSS

There were no significant differences in the number of cases of OHSS when comparing metformin plus clomifene citrate, or hMG to clomifene citrate alone. There was no significant difference in the number of cases of OHSS when comparing metformin plus clomifene citrate to letrozole.

Congenital abnormalities

No evidence was reported regarding congenital abnormalities.

Patient satisfaction

No evidence was reported regarding patient satisfaction.

Health related quality of life

No evidence was reported regarding health related quality of life.

Anxiety and/or depression

No evidence was reported regarding the number of women with anxiety and/or depression.

Surgery compared with drugs
Live full-term singleton birth

There was no significant difference in the number of live full-term singleton births when comparing surgery to clomifene plus tamoxifen, hMG, FSH or rFSH.

Clinical pregnancy

There were no significant differences in the number of clinical pregnancies when comparing surgery to clomifene plus tamoxifen, metformin plus clomifene citrate, hMG, FSH or rFSH. There was also no significant difference in the number of clinical pregnancies when comparing surgery plus clomifene citrate to FSH.

Adverse pregnancy outcomes

There were no significant differences in the number of miscarriages when comparing surgery to clomifene citrate plus tamoxifen, metformin plus clomifene citrate, hMG or rFSH.

There was no significant difference per woman or per pregnancy in the number of preterm births when comparing surgery with rFSH.

Multiple pregnancies

There were significantly more multiple pregnancies per woman with FSH or rFSH compared with surgery. However, the difference was not significant per pregnancy.

There was no significant difference in the number of multiple pregnancies when comparing surgery to hMG, rFSH or metformin plus clomifene citrate.

Multiple births

No evidence was reported regarding the number of babies born from multiple pregnancies.

OHSS

There was no significant difference in the number of cases of OHSS after surgery compared with after hMG or rFSH.

Congenital abnormalities

There was no evidence reported regarding the number of congenital abnormalities.

Patient satisfaction

There was no evidence reported regarding patient satisfaction.

Health related quality of life

There was no evidence reported regarding health related quality of life.

Anxiety and/or depression

There was no evidence reported regarding the number of women with anxiety and/or depression.

Lifestyle compared with drugs or surgery
Live full-term singleton birth

No evidence was reported regarding the number of live full-term singleton birth

Clinical pregnancy

No evidence reported regarding the number of clinical pregnancies

Adverse pregnancy outcome

No evidence was reported regarding adverse pregnancy outcomes.

Multiple pregnancies

No evidence was reported regarding the number of multiple pregnancies.

Multiple births

No evidence was reported regarding the number of multiple pregnancies resulting in birth

OHSS

No evidence was reported regarding the number of cases of OHSS

Congenital abnormalities

No evidence was reported regarding the number of congenital abnormalities

Patient satisfaction

No evidence was reported regarding patient satisfaction

Health related quality of life

No evidence was reported regarding health related quality of life

Anxiety and/or depression

No evidence was reported regarding the number of women with anxiety and/or depression.

Body mass index (BMI)

Eight included studies set inclusion/exclusion criteria based on BMI (Bayar et al., 2006; Elsedeek et al., 2011; Farquahar et al., 2002; George et al., 2003; Johnson et al., 2010; Karimzadeh et al., 2010; Palomba et al., 2005; Qublan et al., 2007). For three of these studies of BMI restricted populations, the treatment regimens were unique to these studies and not found in the unrestricted studies reported above (Farquahar et al., 2002; George et al., 2003; Karimzadeh et al., 2010). Thus it was not possible to analyse the effect of BMI. For the five remaining studies of BMI restricted populations, while they used treatment regimens that were reported in the unrestricted populations above, the studies were of insufficient size to allow a confident comparison to be made.

Although a subgroup analysis by BMI was not undertaken, the GDG noted that the studies that only included women with a BMI of 32 or less (Johnson et al., 2010 [BMI 32 or less], Karimzadeh et al., 2010 [BMI 25 to 29.9], Palomba et al., 2005 [BMI 30 or less]) showed a trend towards the effectiveness of metformin over clomifene citrate for live birth and clinical pregnancy rates (although this was not significant). Of the two studies that did not restrict the entry of women according to their BMI, one found a significant advantage of clomifene over metformin for live birth but not clinical pregnancy (Zain et al, 2009) while the other found a significant advantage of clomifene over metformin for clinical pregnancy but not live birth (Legro et al., 2007), and both of the non-significant effects showed a trend towards favouring clomifene.

Health economics profile

A formal health economic profile was not undertaken for this review.

Evidence to recommendations

Relative value placed on the outcomes considered

Live full-term singleton birth is the most important outcome which allows clinicians to inform couples of their chances of having a baby. However, all of the studies in this review reported only live birth rates, which may have included pre-term births and/or births from multiple pregnancies, and were therefore downgraded for ‘indirectness’ as a consequence. Clinical pregnancy is the second most important measure as it reflects a woman’s ability to conceive. The other outcomes in this review relate to side-effects of the treatments and are important when informing women of potential risks of treatment.

Consideration of clinical benefits and harms

First-line treatment

The review found that metformin plus clomifene resulted in significantly more live full-term singleton births and clinical pregnancies than metformin alone, and that it was significantly more effective than clomifene citrate alone in terms of live full-term singleton births. The additional benefit of the drugs in combination was more marked in comparison with metformin than clomifene. The evidence showed that the standard UK first-line treatment (clomifene citrate) did not result in significantly more live births than the alternatives of metformin, letrozole or FSH. The GDG noted that there was not a large difference in the absolute number of clinical pregnancies or live births when comparing metformin, clomifene citrate and a combination of metformin and clomifene citrate. However, the GDG was aware from studies of women with lower BMI that metformin may be more effective than clomifene citrate alone in these women, while clomifene citrate may be more effective than metformin alone in other women. There was no significant difference in the number of adverse pregnancy outcomes or cases of OHSS for the different drugs. However, the GDG acknowledged that adverse effects, such as nausea, are more prevalent with metformin compared with clomifene citrate.

There are limited data comparing the number of cases of OHSS and the number of multiple pregnancies with letrozole alone to clomifene citrate alone. The GDG noted that there are concerns surrounding the safety of letrozole, and do not consider these to be outweighed by the limited evidence. The GDG also notes that letrozole is not used in standard practice in the UK.

No studies were found that compared surgery to drugs as first-line treatment.

Studies on lifestyle modification found no significant difference in the number of clinical pregnancies following a low calorie diet with exercise than clomifene citrate alone, metformin alone or clomifene citrate with metformin. However, the GDG noted that one of the two studies that reported evidence on lifestyle modification only included women with a BMI of 25 to 29.9, which may not be applicable to women with WHO Group II ovulatory infertility with higher BMIs. Also, the effect of diet and exercise on live birth rates was not reported. The GDG acknowledged the complexities of using diet and exercise advice to improve ovulation disorders, including patient compliance and the amount of time that may be required to reduce weight to a level that has a significant effect on ovulation. The GDG emphasised that losing weight should be considered as part of the fertility treatment for women with WHO Group II ovulatory infertility and, furthermore, that a woman’s BMI should not be considered a barrier to treatment.

Overall, the GDG’s considered view was that, as a first-line treatment for women with WHO Group II ovulatory disorders, clomifene citrate and metformin offer similar chances of live birth. It is biologically plausible that the addition of clomifene to metformin may increase the chances of live birth compared with the use of either drug alone but the evidence was not strong enough to make a recommendation that metformin should be used with clomifene to increase the chances of a singleton live birth.

Second-line treatment
Women with PCOS who are resistant to clomifene citrate

There were significantly more live full-term singleton births and clinical pregnancies after double treatment with metformin plus clomifene citrate compared with clomifene citrate alone. There was no significant difference in live births when comparing hMG with metformin plus clomifene citrate. There were significantly more clinical pregnancies after uFSH compared with metformin plus clomifene citrate, and no significant difference in the number of clinical pregnancies when comparing hMG to metformin plus clomifene citrate. These findings imply that gonadotrophins may be as effective in women with PCOS who are resistant to clomifene citrate as a combination of metformin plus clomifene citrate.

The GDG’s view was that gonadotrophins are used in second-line treatment when there is clomifene citrate resistance, and metformin in combination with clomifene citrate is less common practice for second-line treatment.

Surgery and drugs were equally effective in terms of live full-term singleton births or clinical pregnancies.

No evidence was reported comparing lifestyle modification (such as diet and exercise) to drugs and/or surgery in clomifene citrate resistant women.

Data reporting adverse pregnancy outcomes and OHSS was either not reported or found no significant difference between interventions.

Consideration of health benefits and resource use

No studies were identified that considered the relative cost effectiveness of interventions for women requiring ovarian stimulation. Lifestyle interventions, such as dietary advice and exercise, are likely to have lower cost to the NHS than medical or surgical intervention, but low-cost interventions are not necessarily the most cost effective. The time taken to provide counselling and advice to alter lifestyles takes time to provide by a healthcare professional. If it is not effective, it takes resources away from more cost-effective treatments.

The cost of metformin is relatively low compared with clomifene and results in fewer multiple pregnancies (which also increase the cost of birth). The cost of combination therapy is higher with limited evidence of improved effectiveness. However, the GDG noted that the cost of medical management includes resources other than the cost of the drugs themselves. Clomifene requires more scanning and monitoring than metformin due to the increased risk of multiple pregnancies (as acknowledged in the 2004 guidance), and this increases the cost of clomifene. On the other hand, general practitioners are unable to prescribe clomifene citrate, whereas they are able to prescribe metformin, so there is the additional cost of at least one outpatient visit for clomifene.

The GDG considered that, overall, there is a higher cost associated with treatment with clomifene. Nevertheless, clomifene is an established drug and is part of standard clinical practice. The GDG concluded that the evidence was not strong enough to change the existing recommendation that clomifene should be one of the drugs offered.

Quality of evidence

The quality of the evidence was mainly very low due to limitations of the studies, particularly the lack of reporting on blinding and power analysis, and wide confidence intervals. Clomifene citrate resistance is defined in this guideline as ovulation that is not induced after treatment of up to 3 cycles with dose escalation but the definition of clomifene citrate resistance and PCOS varied from study to study. Moreover, the included studies only reported on a PCOS population. Therefore, the conclusions may not be generalisable to all types of WHO Group II ovulation disorders.

Limited reporting on patient characteristics and outcomes in the studies included in the review meant that it was not possible to undertake all relevant analyses. For example, a sub-group analysis on BMI was not undertaken. No studies reported patient satisfaction and a limited number reported relevant adverse outcomes.

Other considerations

Gonadotrophin-releasing hormone analogues in ovulation induction therapy

The 2004 version of the guideline included a review comparing the use of gonadotrophins alone to the use of gonadotrophins in conjunction with gonadotrophin-releasing hormone (GnRH) agonists to achieve pituitary down-regulation and facilitate cycle control in ovarian stimulation. As the 2004 guideline recommends the use of clomifene citrate or tamoxifen for women with WHO Group II ovulation disorders, a review was undertaken for the 2013 update of the guideline to compare clomifene citrate and/or tamoxifen with other drugs, including gonadotrophins with or without GnRH agonists. The 2004 review comparing the use of gonadotrophins with and without GnRH agonists is therefore no longer relevant to the consideration of the evidence for ovulation induction therapy in women with WHO Group II disorders, and has been removed from the guideline text.

Lifestyle advice

The evidence base for weight loss was very limited. Also, the effect of diet and exercise on live birth rates was not evaluated. However, it did show that that weight loss was as effective as clomifene at achieving ovulation. The GDG acknowledged the complexities of using diet and exercise advice to improve ovulation disorders, including patient compliance and the amount of time required to reduce weight to a level that has a significant effect on ovulation. However, based on clinical experience, the considered view of the GDG was that overweight women should be counselled to lose weight because of the positive impact on conception rates and pregnancy outcomes and the negative impact of high BMI on pregnancy outcomes. The advice might include specific advice from a dietician, warnings of the potential risks in pregnancy and, if appropriate, the offer of access to exercise advice and psychosocial support.

Medical management

Metformin is currently not licensed for use in the treatment of PCOS (its license is for use in diabetes). The GDG took into account that metformin needs to be taken multiple times a day whereas clomifene citrate is taken 5 days per month, and that this could be a consideration when discussing the best treatment for each individual. In addition, clomifene citrate requires appropriate monitoring which, along with the duration of treatment, should be taken into consideration when discussing the most appropriate treatment for each woman. The GDG noted that clomifene citrate is licensed for use for up to 6 months at a time. The GDG believed 6 months use of clomifene citrate is an adequate amount of time to determine whether a woman will respond or is resistant to it, and so recommended that clomifene citrate should not be continued after this time.

The GDG took into account that gonadotrophins are often used in second-line treatment when the woman is resistant to clomifene citrate, and that metformin in combination with clomifene citrate is less common practice in England and Wales.

Surgical intervention

The GDG also considered laparoscopic ovarian drilling as a second-line treatment following clomifene resistance. A significant benefit is the elimination of the increased risk of multiple pregnancies and thus laparoscopic ovarian drilling could be an option that would be preferable for some women. Although no evidence was identified to support its use, the view of the GDG was that it should remain a treatment option depending on the individual woman’s clinical circumstances and preferences.

Equalities

The people considered in this review were:

  • People in same sex relationships who cannot have heterosexual intercourse.
  • Specific patient subgroups listed in the guideline Scope who may need specific consideration:
    • people in same-sex relationships who have unexplained infertility after donor insemination
    • people who are unable to, or would find it very difficult to, or who have been advised not to, have heterosexual intercourse
    • people with conditions or disabilities that require specific consideration in relation to methods of conception.
  • People who are preparing for cancer treatment who may wish to preserve their fertility.

There were no specific issues that needed to be addressed with respect to any of these subgroups for this review.

Recommendations

NumberRecommendation
In women with WHO Group II ovulation disorders receiving first-line treatment for ovarian stimulation:
92Advise women with WHO Group II anovulatory infertility who have a BMI of 30 or over to lose weight (see recommendation 26). Inform them that this alone may restore ovulation, improve their response to ovulation induction agents, and have a positive impact on pregnancy outcomes. [new 2013]
93Offer women with WHO Group II anovulatory infertility one of the following treatments, taking into account potential adverse effects, ease and mode of use, the woman’s BMI, and monitoring needed:
  • clomifene citrate or
  • metformin* or
  • a combination of the above. [new 2013]
94For women who are taking clomifene citrate, offer ultrasound monitoring during at least the first cycle of treatment to ensure that they are taking a dose that minimises the risk of multiple pregnancy. [2013]
95For women who are taking clomifene citrate, do not continue treatment for longer than 6 months. [2013]
96Women prescribed metformin* should be informed of the side effects associated with its use (such as nausea, vomiting and other gastrointestinal disturbances). [2004]
In women with WHO Group II ovulation disorders who are known to be resistant to clomifene citrate:
97For women with WHO Group II ovulation disorders who are known to be resistant to clomifene citrate, consider one of the following second-line treatments, depending on clinical circumstances and the woman’s preference:
  • laparoscopic ovarian drilling or
  • combined treatment with clomifene citrate and metformin* if not already offered as first-line treatment or
  • gonadotrophins. [new 2013]
98Women with polycystic ovary syndrome who are being treated with gonadotrophins should not be offered treatment with gonadotrophin-releasing hormone agonist concomitantly because it does not improve pregnancy rates, and it is associated with an increased risk of ovarian hyperstimulation. [2004]
99The use of adjuvant growth hormone treatment with gonadotrophin-releasing hormone agonist and/or human menopausal gonadotrophin during ovulation induction in women with polycystic ovary syndrome who do not respond to clomifene citrate is not recommended because it does not improve pregnancy rates. [2004]
100The effectiveness of pulsatile gonadotrophin-releasing hormone in women with clomifene citrate-resistant polycystic ovary syndrome is uncertain and is therefore not recommended outside a research context. [2004]
*

At the time of publication (February 2013), metformin did not have a UK marketing authorisation for this indication. The prescriber should follow relevant professional guidance, taking full responsibility for the decision. The patient should provide informed consent, which should be documented. See the General Medical Council’s Good practice in prescribing medicines – guidance for doctors for further information.

NumberResearch recommendation
RR 16What is the cost effectiveness and safety of using clomifene citrate or metformin or a combination of the two to induce ovulation in women with WHO Group II ovulation disorders?

8.4. Hyperprolactinaemic amenorrhoea – dopamine agonists

Introduction

Two RCTs (n = 306) comparing cabergoline to bromocriptine in women with hyperprolactinaemic amenorrhoea reported that cabergoline was more effective in restoring ovulation and increased pregnancy rates (72% and 72% with cabergoline versus 52% and 48% with bromocriptine, respectively).576,577 [Evidence level 1b] However, the manufacturer advises discontinuation of cabergoline at least one month before pregnancy.181 [Evidence level 4]

A systematic review of three RCTs found no improvement in pregnancy rates (OR 1.12, 95% CI 0.48 to 2.57) following treatment with bromocriptine versus placebo in couples with unexplained infertility.578 [Evidence level 1a]

Recommendations

NumberRecommendation
101Women with ovulatory disorders due to hyperprolactinaemia should be offered treatment with dopamine agonists such as bromocriptine. Consideration should be given to safety for use in pregnancy and minimising cost when prescribing. [2004]

8.5. Monitoring ovulation induction during gonadotrophin therapy

Ovarian monitoring provides information on ovarian response to ovulation induction agents by ascertaining the number and size of the developing follicles.

Ultrasonography is regarded as a safe, accurate and efficient method of monitoring follicular development in response to ovulation induction,579581 in helping to reduce multiple pregnancy rates, especially in women with PCOS571 when compared with oestrogen monitoring. [Evidence level 2b] Oestrogen monitoring provides no additional information compared with ovarian ultrasound.579 [Evidence level 2b] Ultrasonography was found to have good predictive value in the occurrence of OHSS which was associated with larger number of immature follicles at time of hCG administration.582 [Evidence level 3] An observational study reported that follicular sonography performed during ovarian stimulation predicted 88% of cycle decisions.583 [Evidence level 3]

Ovarian hyperstimulation syndrome

The aim of ovulation induction therapy is to stimulate the ovaries to produce more than one egg. This carries the risk of overstimulation and OHSS. OHSS is a potentially fatal condition when many follicles are stimulated, leading to ascites, pleural and pericardial effusion, haemoconcentration and coagulopathy.584

The exact incidence of severe OHSS when fertility drug therapy is used has not yet been determined. Available data suggest an incidence of 3% of cycles when hMG is used,585 and in 0.2% to 1.0% of all assisted reproduction cycles.586588 Results generated by the European Society for Human Reproduction and Embryology (ESHRE) on assisted reproductive technology in Europe in 1999 reported an incidence of OHSS of 0.9% (range 0.3 % to 2.7%; 1083 cases of OHSS after 114,628 cycles).589 [Evidence level 3]

Clinics that provide ovarian stimulation should have protocols in place for the prevention, diagnosis and management of OHSS (see Section 15.5).

Multiple pregnancy

Prevention of iatrogenic multifetal gestation involves judicious use of ovulation induction drugs and monitoring with ultrasound to chart follicular development. It is best carried out in a specialist clinic.

There is a strong correlation between the initial number of embryos, the final number and the risks of pregnancy loss and prematurity.590,591 [Evidence level 3] Multiple gestations are high-risk pregnancies associated with higher obstetric complications, perinatal, neonatal and infant mortality,592 as well as significant financial593,594 and psychological595 consequences. [Evidence level 3] However, assisted reproduction multiple pregnancies do not appear to be at any more risk of poor obstetric and neonatal outcomes than those conceived spontaneously.596,597 [Evidence level 3] Recent surveys have suggested that multiple pregnancies may not be viewed as an adverse outcome by women with fertility problems.598602 [Evidence level 3–4]

The exact numbers of multiple pregnancies arising from ovarian stimulation, with or without IUI, are unknown, as there are no national registers that record the outcome of controlled ovarian stimulation,603 as there are with IVF and ICSI, such as the register monitored by the HFEA. Multiple pregnancy occurs in 2–13% of women with all causes of infertility taking clomifene citrate.604 This compares with a spontaneous multiple pregnancy rate of about 1–2% of women in the North American and European populations.605,606 Women with clomifene citrate-resistant PCOS treated with conventional regimens of gonadotrophins have a 36% multiple pregnancy rate.607 [Evidence level 3] A one-year survey of triplets and higher-order pregnancies in the UK found that 31% of the triplet pregnancies were spontaneous, 34% were from various methods of ovulation stimulation and 35% were from IVF/GIFT. Triplet pregnancies accounted for 56% of all pregnancies attributable to clomifene citrate.608 [Evidence level 3]

The issue of multiple pregnancies arising from IVF is discussed in Chapter 15.

Multifetal pregnancy reduction refers to the termination of one or more normal fetuses in a multifetal pregnancy in order to improve the survival rates for the remaining fetuses and to decrease maternal morbidity.590 [Evidence level 4] For any initial number of embryos, reduction to twins has the highest survival rate.591 [Evidence level 3] Reduction to singletons rather than twins is associated with a higher gestational age at delivery but a lower survival rate.590 [Evidence level 3]

Recommendations

NumberRecommendation
102Women who are offered ovulation induction with gonadotrophins should be informed about the risk of multiple pregnancy and ovarian hyperstimulation before starting treatment. [2004]
103Ovarian ultrasound monitoring to measure follicular size and number should be an integral part of gonadotrophin therapy to reduce the risk of multiple pregnancy and ovarian hyperstimulation. [2004]
Copyright © 2013, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK327781

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (4.7M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...