|
INTRODUCTION
In this topic, we shall review human ovarian physiology, clinical reproductive
Endocrinology and their application for ovarian stimulation for in vitro fertilization
(IVF). We will focus on a practical clinical approach as used daily in a
large IVF program.
HUMAN OVARIAN PHYSIOLOGY
The initiation of follicular growth is a continuous process that is believed to not depend on hormonal stimulation. About 50 follicles each day begin to grow in the human ovaries. Most follicles undergo rapid atresia. This total growth phase occurs over approximately 90 days or over three ovarian cycles. However, under the hormonal changes in the late luteal phase and the beginning of the new cycle, a group of follicles will respond to small rises in serum follicle-stimulating hormone (FSH) by processing to the preantral stage (Fig. 1) . At this stage, the follicle is about 200 mm with multiple granulosa cell layers. Under the influence of FSH, the number of FSH
receptors on the granulosa cell increases to about 1500 receptors per cell, and, at the same time, the granulosa cells start to produce estradiol 17b (E2) by aromatizing androgens that are supplied by the theca cells under luteinizing hormone (LH) stimulation. E2 and FSH together cause proliferation of granulosa cells and increase the number of FSH receptors on the granulosa cell plasma membrane. Production of follicular fluid increases.
Its accumulation in the intracellular space eventually forms a cavity—the
antrum. The size of the antral follicle is now about 500 mm in diameter.By days 5–7, the dominant follicle is selected because of its ability to convert androgens to estrogens. The other follicles will not continue to grow and will undergo atresia. The dominant follicle continues to grow and secretes E2 and inhibin B that exert negative feedback on FSH production, causing a decline in serum FSH levels. FSH induces the appearance of LH receptors on granulosa cells, a process which is augmented by the concomitant presence of high estrogen levels. E2 production gradually increases, and plasma E2 levels achieve a threshold concentration required for the generation of the LH surge that starts 14–24 hr after the serum E2 reaches peak concentration.
Intra-ovarian peptides play important roles in modulating gonadotropin effects on ovarian function. At least 33 putative paracrine–autocrine regulators of follicular growth and atresia are identified . FSH enhances the secretion of most of them by granulosa cells. Insulin-like growth factor I (IGF-I) augments FSH-mediated aromatization, granulosa cell mitogenesis, and the induction of LH receptors. Inhibin, in addition to its endocrine negative effect on FSH secretion, inhibits aromatization and stimulates LH-induced androgen production by theca cells. Activin has a positive effect on aromatization, granulosa cell mitogenesis
, and a negative paracrine action on LH-induced androgen production by theca
cells . Activin is also involved in the regulation of apoptosis in the ovary .
Follistatin, the third member in the inhibin/activin family, is antagonistic to activin . Vascular endothelial growth factor (VEGF) and growth factors such as epidermal growth factor (EGF) and transforming growth factors a (TGFa)
also play important roles in modulating gonadotropin effects on ovarian
function .
The LH surge initiates utilization and the beginning of progesterone production by the granulose cells of the dominant follicle. It is also responsible for the resumption of meiosis in the oocyte . Actinic promotes and inhibin inhibits the LH surge and superovulation in a rat model . LH stimulates the synthesis of cytokines, the best known of which is interleukin-1 (IL-1) which modulates activation of prostaglandins and the proteolysis Figure 1 a schematic representation of the ovarian and menstrual cycles.
Abbreviations: FSH, follicle-stimulating hormone; LH, luteinizing hormone.
Basics of Ovarian Stimulation 45 cascade that is essential for follicular rupture . Ovulation occurs 24–36 hr after the onset of the LH surge, when the follicle, which is about 20 mm, ruptures and the oocyte, is released from the ovary.
After ovulation, the dominant follicle becomes the corpus luteum (Fig. 1), producing progesterone, E2, and inhibin, which suppress the growth of new follicles in the ovary. At the end of the cycle, luteolysis causes decline in both steroids and inhibin, allowing the elevation in serum LH and FSH that is necessary for the initiation of a new cycle.
Figure 1 A schematic representation of the ovarian and menstrual cycles.
Abbreviations: FSH, follicle-stimulating hormone; LH, luteinizing hormone.
<<Top>>
Intra-Ovarian Growth Factors
As introduced above, ovarian follicles produce a number of TGFa-related
proteins. Anti mullerian hormone, TGFbs, activins, and inhibins are produced by
granulosa cells. Both bone morphogenetic protein 15 (BMP15) and growth
differentiation factor 9 (GDF9) are expressed exclusively by the oocyte of
several species BMP15 and GDF9 stimulate granulose cell mitogenesis . BMP15 is a potent inhibitor of FSH-receptor expression and participates in negative feedback influencing granulose cell mitosis . BMP6 is also expressed in the oocyte and inhibits FSH action, probably by downregulation of adenylate cyclase . There is a rapid decrease in BMP6 concentration in granulose cells around the time of dominant follicle selection.
BMP Receptors Members of the TGFbasuperfamily signal through the activin/TGFa and/ or BMP pathways
. The BMP receptors (BMPRIA/ALK3, BMPRIB/ ALK6, and BMPR2) are transmembrane
serine/threonine kinases closely related to the transforming growth factor
beta receptors (TGFBRI/ ALK5, TGFBR2) and activin receptors (ACVR1, ACVR1B,
ACVR2, and ACVR2B). BMP receptors are expressed in granulose cells and oocytes and the BMPs exert their biological actions by forming heteromeric complexes with type I and II receptors . Ligands bind to the type II receptors leading to transphosphorylation of the type I receptor. The type
I kinase activates proteins which migrate to the nucleus and together with other proteins regulate expression of target genes. GDF9, BMP15, BMP4, and BMp7 all use BMP2 as a binding receptor . BMP15 signals through interaction of BMPR1B and BMPR2 activating the SMAD1/5/8 pathway . Consequently, BMP proteins appear to interact with a limited number of receptors to activate two downstream Smad pathways. Moreover, several high-affinity binding proteins including follistatin, noggin, and gremlin antagonize BMP signaling . How granulose cells and other cell types in the ovary differentiate between signals from multiple ligands in this pathway remains unclear.
46 Healy et al. Species Differences in BMP Signaling there are both similarities and important differences between species in regulation in BMP signaling and consequences of gene knockouts in this pathway. Both mice and sheep with knockout mutations in GDF9 are infertile. In contrast to the dramatic consequences of homozygous mutations of BMP15 in sheep, targeted deletion of the second exon of Bmp15 in the mouse has little effect on folliculogenesis . However, the knockout mice are sub-fertile due to defects in the ovulation process and early embryo development. Reasons for the difference are not clear, but may relate to the relative importance of BMP15 and GDF9 in regulating follicle development in the different species. Another possibility is differential effects of the gene disruptions on processing and formation of GDF9/BMP15 homo- and heterodimers . BMP15 seems essential for human fertility . Two sisters with hypergonadotropic ovarian failure due to ovarian dysgenesis carry a non-conservative amino acid substitution in the promoter region of BMP15 (tyr235Cys) which acts in a dominant negative fashion by altering processing of BMP15.
<<Top>>
CLINICAL REPRODUCTIVE ENDOCRINOLOGY
Stimulation Protocols
The first IVF baby was born as a result of an oocyte picked up in a natural cycle. However, the success rate of this protocol was very low, and the Monash group first reported large numbers of eggs and improved pregnancy rates using a stimulation protocol of clomiphene citrate and human menopausal gonadotropin (HMG) together
. Several other regimens using these two medicines were subsequently reported.
The common problems with these protocols were that endogenous gonadotropins
led to premature luteinization in 30–40% of the cases, and, in others,
ovulation occurred at an inconvenient time of the day. The major step in
simplifying IVF induction of ovulation protocols and preventing these unwanted
phenomena came with the introduction of GnRH agonists. GnRH analogs were
created by a series of modifications in the GnRH molecule that led to the
availability of new agonists and antagonists. The agonists initially enhance
gonadotropin released from the pituitary but, with continuing administration,
caused downregulation of the pituitary and reduced LH and FSH secretion for as
long as the analog was given. This effect was a powerful tool with which to
control the stimulated IVF cycle. GnRH antagonists are now available for
routine clinical use.
Initially, the use of GnRH agonists was confined to women with a history of unsatisfactory response to other stimulation protocols, a premature LH surge, or elevated plasma LH and FSH levels . There are now many reports advocating their use in all IVF cycles including those women Basics of Ovarian Stimulation 47
With normal basal gonadotropins . Tan et al. have reported life table analysis of conception and live birth rates in 3000 women undergoing IVF-ET with and without different regimens of the GnRH agonist buserelin.
They found that the cumulative conception rate and cumulative live birth rate
were significantly higher in those patients treated with a downregulation
regimen. The mechanism by which GnRH analogs improve follicular response is
not yet known. Palermo et al. reported improvement in synchronization of follicular development leading to a larger cohort of developing follicles, whereas others related it to the longer gonadotropin stimulation in the GnRH analog cycles.
Urinary gonadotropin preparation, HMG, which contains FSH and LH, was used since 1980 for ovarian stimulation for IVF . In 1991, a highly purified FSH preparation was introduced, suggesting that endogenous LH was sufficient for satisfactory folliculogenesis
. A meta-analysis of randomized trials of FSH versus HMG performed on 599
patients indicated that the use of FSH was associated with a significantly
higher clinical pregnancy rate than HMG.
Further meta-analyses by NICE have addressed cost-effective studies relating to gonadotropins. The National Collaborating Centre for Women’s and Children’s Health found that HMG, urinary FSH, and recombinant FSH were equally effective in achieving a live birth when used following pituitary downregulation as part of IVF treatment. NICE recommended that consideration should be given to minimizing cost when prescribing.
In the United Kingdom, use of urinary FSH products could represent a potential cost saving of £15 million/year.
Typical protocols of ovarian stimulation in a typical IVF program are:
1. GnRH agonist/FSH—long downregulation protocol.
2. GnRH agonist/FSH—short (‘‘boost’’ or ‘‘flare-up’’) protocol.
3. Clomiphene citrate/HMG.
The objective of the protocols is to obtain a cohort of codominant follicles in both ovaries with adequate E2 levels suitable for oocyte retrieval on days 11–15.
<<Top>>
GnRH Agonist/FSH—Long Downregulation Protocol
This protocol (Fig. 2) is currently the protocol of choice for first cycle patients and previously normal or high responders in many IVF programs.
A GnRH agonist [Leuprolide, 1mg SC, Buserelin 600 mg SC; or Nafarelin acetate (Synarel)], 400 mg nasal spray, is given for 10–14 days before starting gonadotropin treatment (‘‘long-protocol’’), and may be commenced in the mid-luteal phase of the previous cycle or on day 2 of the cycle. We, in addition to other researcher found that pituitary suppression was more effective when the therapy commenced in the mid-luteal phase of the previous cycle rather than in the early follicular phase of the treatment
. Urbancsek and Witthaus found that clinical pregnancy and live birth rates were better when a GnRH agonist was started in the mid-luteal phase rather than in the early follicular phase. Another advantage in commencing the analog in the mid-luteal phase was the decreased occurrence of ovarian cysts compared with commencing the GnRH agonist at the beginning of the follicular
Phase. These cysts, especially if functional, should be aspirated before commencing gonadotropin treatment.
Suggested criteria for pituitary–ovarian downregulation are E2 levels <180 pmol/L, LH levels <2 IU/L, and P4 levels <2 nmol/L. FSH commences after achievement of adequate downregulation and continues by daily injections according to individual endocrine and ovarian ultrasonic response until the day before the human chorionic gonadotropin (hCG) injection. A 5000 IU IM injection is given when the follicular cohort consists of at least three follicles of 17–20mm in diameter and serum E2 level is
Appropriately rising. Egg pick-up follows 36 hr later.
Disadvantages of this protocol include a greater consumption of GnRH agonist and FSH and, therefore, greater patient cost, as well as the uncertainty of possible pregnancy upon initiation of GnRH agonist treatment if the GnRH analog is started on day 21 of the previous cycle.
Figure 2 A representation of the luteal-phase commencement long downregulation
Protocol using FSH for ovarian stimulation for assisted reproductive technology.
Abbreviations: HCG, human chorionic gonadotropin; GnRH, gonadotrophin-releasing
Hormone; FSH, follicle-stimulating hormone.
For this reason, we ask our patients to avoid unprotected sexual intercourse
In the cycle where GnRH analog treatment begins. However, published data
Suggest that pregnancy outcome is not adversely affected by GnRH agonistAdministration in the luteal phase of the conception cycle (43). Luteal support
is important in these patients to avoid early abortion.
Patient age and follicular-phase serum FSH values are two useful predictors of gonadotropin dosage. FSH starting dose for the different patient groups in our program are listed in Table 1. The duration and daily doses of FSH are adjusted according to individual patient age and response as judged by serum E2 levels and follicular number and size in an ultrasound scan of the ovary.
The National Collaborating Centre for Women’s and Children’s Health have reviewed ovarian monitoring during gonadotropin therapy for IVF.
NICE found that serum estradiol monitoring provided no additional information compared with ovarian ultrasound and, therefore, do not recommend it.
This is different to the above guidelines for basic ovarian stimulation and monitoring. We are comfortable with this contradiction. We consider that serum E2 measurement does assist inexperienced units towards effective multifollicular ovulation induction. It also helps minimize ovarian hyperstimulation syndrome (OHSS). The above endocrinology guidelines may assist those units less experienced in IVF stimulation and monitoring.
<<Top>>
GnRH Agonist/FSH—Short Boost Protocol
This protocol (Fig. 3) is typically reserved for poor responders, namely patients with day 3 FSH >10 IU/L and patients with a previous IVF cycle in which they have been cancelled or less than three oocytes were retrieved (‘‘poor responders’’) or women who previously had a live delivery with a boost cycle. This protocol ‘‘boosts’’ the exogenous gonadotropin injected into the patient by a rise in serum concentration of her endogenous FSH. In this protocol, the GnRH agonist administration begins on day 2 of the cycle and FSH starts on day 3 of the cycle. Both medicines continue on the same regimen as the downregulation protocol. HCG administration criteria and dosage are the same as for the downregulation protocol.
The most important disadvantage of this protocol is the high P4 levels during the early follicular phase. This is most probably caused by rescue of the preceding corpus luteum. If the serum P4 is >6 nmol/L in the late follicular phase, we recommend cancellation of the cycle. NICE guidelines confirm a decreased clinical pregnancy rate per cycle using the short protocol compared with the long GnRH agonist protocol. The relative cost of agonists in the various regimens required for IVF require further economic studies to ascertain whether there is any true difference in cost.
Figure 3 A diagram of the procedures, monitoring, and medicines used for short or
Boost protocol commencing on day 2 of the menstrual cycle. Abbreviations: GnRH,
Gonadotropin-releasing hormone; FSH, follicle-stimulating hormone; HCG, human chorionic onadotrophin.
Into the patient by a rise in serum concentration of her endogenous FSH. In this protocol, the GnRH agonist administration begins on day 2 of the cycle and FSH starts on day 3 of the cycle. Both medicines continue on the same regimen as the downregulation protocol. HCG administration criteria and dosage are the same as for the downregulation protocol. The most important disadvantage of this protocol is the high P4 levels during the early follicular phase. This is most probably caused by rescue of the preceding corpus luteum. If the serum P4 is >6 nmol/L in the late follicular phase, we recommend cancellation of the cycle. NICE guidelines confirm a decreased clinical pregnancy rate per cycle using the short protocol compared with the long GnRH agonist protocol. The relative cost of agonists in the various regimens required for IVF require further economic studies to ascertain whether there is any true difference in cost.
<<Top>>
Clomiphene Citrate and FSH/HMG Protocol
This protocol used to be the preferred protocol in most IVF units prior to development of GnRH analogs. As mentioned, it was abandoned in our program as the dominant superovulatory regimen because of the high cancellation rate due to untimely LH surges and some reports of the possible adverse effect of clomiphene citrate on the endome-trium. In our unit, this Figure 3 A diagram of the procedures, monitoring, and medicines used for short or Boost protocol commencing on day 2 of the menstrual cycle. Abbreviations: GnRH, gonadotropin-releasing hormone; FSH, follicle-stimulating hormone; hCG, human chorionic gonadotrophin.
Basics of Ovarian Stimulation 51 protocol is used only for patients who have responded well (>5 oocytes retrieved) to clomiphene citrate and HMG in the past and who did not respond well to the downregulation or boost protocols, or to patients who did not want to use GnRH agonists and who previously responded well to the combination of clomiphene citrate and FSH/HMG. In this protocol, clomiphene citrate 100 mg/day commences on day 2 of the cycle and proceeds for 5 days. HMG begins on day 3 and continues, as do the other protocols, until 1 day before the hCG injection. hCG (5000 IU in IM injections) is given when the estimated leading follicle is 18–19mm in diameter with serum E2 levels >1800 pmol/L.
Natural Cycle
In natural cycle IVF, one aspirates the oocyte that has
matured during the natural follicular growth just after the natural LH surge. In
the first successful IVF-ET treatment described by Edwards and Steptoe in 1978,
the mature oocyte was retrieved from such a natural cycle. However, the success
rate of this protocol was low and with some exceptions it is currently not in
use in most IVF units.
Cycle Monitoring
NICE guidelines recommend ultrasound monitoring only to measure follicle size and follicle number for patient management during multi follicular folliculogenesis for IVF. Specifically, serum E2 monitoring is not recommended. We have minimized the use of serum E2, P4, and LH measurements during cycle monitoring in our IVF program and still find these helpful in our IVF program for basic ovarian stimulation. We do accept that experienced IVF units may have no difficulty in preventing ovarian hyperstimulation in all their patients without using endocrine measurements. However, we find OHSS still unpredictable to prevent and do recommend serum endocrinology to assist clinical IVF decision-making and minimize OHSS risk. That advice is summarized below. As clinical knowledge about the nature of IVF cycles accumulates, cycle monitoring has become easier and less complicated. In our IVF program, cycle monitoring initially comprised daily serum E2, LH, and P4 estimations and ovarian ultrasound determinations. Currently, cycle monitoring consists of blood sampling for E2 levels at days 7 and 9 of the cycle and ultrasound scan of the ovary only on day 9. According to the results on day 7, a decision about FSH dose is made using the following criteria: 1. 1000 > E2 < 4500 pmol/L—continues with the same dose.2. E2 < 1000 pmol/L increase FSH dose [except polycystic ovarian syndrome (PCOS)]. 3. E2 > 4500 pmol—decrease FSH dose.
52 Healy et al.
On the second visit (day 9), if six or more follicles are >16mm and E2 is >4000 pmol/L, HCG will be given the following day, followed by oocyte retrieval 36 hr later. If the criteria for HCG administration have not been met, a third blood sample is necessary to determine the right time for HCG injection and egg pick-up.
About 10% of patients on the IVF program are discharged before egg pick-up. Cancellation criteria in our unit are:
1. Serum P4 >6 nmol/L on day 9.
2. Fewer than three follicles >16mm and E2 less than 3000 pmol/L on day 9 or 10.
3. Serum LH >30 IU/L during a clomiphene citrate cycle.
4. Abnormal finding by ultrasonography.
5. Falling serum E2 levels despite increasing FSH dose,
6. Serum E2 >25,000 pmol/L.
It is uncertain why all women with regular menstrual cycles do not respond to ovarian hyperstimulation protocols in a satisfactory fashion.
We investigated spontaneous menstrual cycles after failure of multiple IVF olliculogenesis in 131 consecutive cycles. Thirty-six percent of the cycles were endocrinologically normal, whereas the rest had a range of endocrine abnormalities including low luteal phase P4, premature P4 elevation, occult ovarian failure, and hyperprolactinemia.
Although the importance of plasma P4 levels at the time of hCG administration is still controversial, most authors concluded that the subtle P4 elevation during GnRH agonist/FSH ovulation induction for IVF/ET does not predict IVF outcome . We do not measure serum P4 in downregulation/ FSH cycles. Repetition of the same superovulation regimen after inadequate stimulation in an IVF program results in improved folliculogenesis in only 10% of our patients. A review of patient history and the previous response to IVF folliculogenesis is therefore necessary before a decision is made about the next stimulation cycle protocol.
<<Top>>
POLYCYSTIC OVARIAN SYNDROME
PCOS is the most common cause of anovulatory infertility. This syndrome was first described by Stein and Leventhal in 1935. The PCOS Rotterdam Consensus Conference recently defined PCOS as a clinical syndrome comprising any two of the three features: amenorrhea or oligomenorrhea; clinical or biochemical hyperandrogenism, and bilateral polycystic ovaries on ultrasound . Although no single biochemical test is diagnostic of PCOS, most patients showed a characteristic ovarian ultrasonographic appearance, namely the presence of >10 follicles between 2 and 10mm in diameter.
On the other hand, ultrasonographically identified polycystic ovaries are a common finding in apparently normal women, with a prevalence of Basics of Ovarian Stimulation 53 20–23%. On careful review, menstrual irregularities and evidence of hyperandrogenism are frequently associated with the presence of polycystic ovaries in these apparently healthy women . Infertile women with PCOS represent a difficult therapeutic problem for assisted reproductive techniques because they have a higher incidence of OHSS (refer following section). On transvaginal ultrasound scanning, the prevalence of polycystic ovaries has been reported to be 25–33% in an assisted reproduction treatment population . These data underline the fact that polycystic ovaries are prevalent in infertile women who often lack the clinical characteristics of classical PCOS.
It has been proposed that a dysfunction of cytochrome P-450c 17a in PCOS leads to an increased 17-hydroxyprogesterone (17-OHP) response to a GnRH agonist-induced gonadotrophin rise . We investigated 106 IVF patients undergoing a boost stimulation regimen by correlating the ovarian ultrasound pattern with serum testosterone, 17-OHP, androstenedione, and estradiol responses and with the clinical outcome. There was a significantly higher prevalence of ovarian hyperandrogenism (serum testosterone >0.5 nmol/L after dexamethasone administration) in patients with polycystic ovaries (23%) compared with normal ovaries (7%). Patients with polycystic ovaries had approximately double the 17-OHP and estradiol responses to GnRH agonist. The number of oocytes retrieved was positively correlated with the estradiol responses. Although there was no difference in the total amount of FSH used between the patients with polycystic and normal ovaries, the median peak estradiol concentration was 1.6 times greater and the oocyte yield 2.3 times greater in patients with polycystic ovaries. The overall pregnancy rate per transfer was 32% and did not differ between patients with or without polycystic ovaries. We concluded from this study that the presence of polycystic ovaries on a vaginal ultrasound scan was the single most important marker for ovarian abnormality in an assisted reproduction patient population. Detection of polycystic ovaries predicts a subset of patients with abnormal ovarian androgen metabolism, exaggerated 17-OHP and estradiol responses to a GnRH agonist, and a higher oocyte yield. The ultrasonographic changes characteristic of PCO should be sought in all women undergoing assisted reproduction. In such cases, the ovarian-stimulating protocol is modified by reducing FSH starting dose (Table 1) and careful follow-up of serum E2 levels and follicular number and size.
<<Top>>
IVF-ET is an effective therapy for PCOS patients who are refractory
To ovulation induction in vivo even in the absence of other associated infertility factors. However, laparoscopic diathermy of the ovarian surface and hilum should precede an IVF attempt in these patients, as this procedure can reduce serum androgen concentration and normalize ovarian function. Following this procedure, spontaneous ovulation rate is about 90% and conception rates are 40–70% (53). Thus, many authors recommend 54 Healy et al.
This operation even before gonadotrophin treatment. Furthermore, it should be
considered for patients who had difficulties in multifolliculogenesis or had
severe OHSS after IVF treatment (refer following section). The overall
conception rates per cycle in PCOS patients compared favorably to conception
rates in patients with other infertility diagnoses . The use of GnRH agonist is
associated with a significant reduction in the incidence of pregnancy loss and
improvement in the cumulative pregnancy rates . Salat-Baroux et al. found that the longer the period of pituitary desensitization (the long protocol vs. shorter ones), the lower the levels of circulating androgens. The current protocol for PCOS patients in Monash IVF is the long down regulation protocol with starting FSH dosage of 112 IU/day.
OVARIAN HYPERSTIMULATION SYNDROME
OHSS is the major serious and potentially life-threatening complication of ovulation induction in IVF-ET treatment. It is characterized by transudation of protein-rich fluid from the vascular space into the peritoneal cavity and to a less extent, pleural and pericardial cavities. The basic path physiologic event in OHSS is an acute increase in capillary permeability; however, the exact factors responsible for this phenomenon have, until recently, not been clear.
Because intensity of the OHSS is related to the degree of ovarian
Response to ovulation induction therapy, OHSS is probably an exaggeration of normal ovarian physiology. Part of the angiogenic response, which occurs in the follicle at the time of ovulation, is increased vascular permeability VEGF. VEGF stimulates endothelial cell mitogenesis and renders capillaries highly permeable to high-molecular-weight protein.
VEGF has been identified in rat and primate ovaries predominantly after the LH surge. Luteal-phase treatment with GnRH agonist, to suppress LH secretion, decreased VEGF messenger-RNA expression, implying such expression is dependent on LH. We first reported the role of VEGF in OHSS
. We have demonstrated that VEGF is the major capillary permeability factor in
OHSS ascites. Although other capillary permeability factors may not have been
detected, 70% of the capillary permeability activity in OHSS as cites was
neutralized by recombinant human VEGF antiserum. Abramov et al. subsequently followed the kinetics of VEGF in the plasma of seven patients with severe OHSS from the time of admission to the hospital until clinical resolution. They found that compared with normo-responders, IVF patients suffering OHSS had VEGF plasma levels which were significantly higher and correlated to the clinical picture. Ascitic fluid obtained from OHSS patients contained high VEGF levels. The incidence of OHSS after induction of ovulation varies between 1% and 30%, as reported in various publications. This variation is probably due Basics of Ovarian Stimulation 55 to the difference in the definition of OHSS. OHSS is classically divided into three categories:
1. Mild OHSS, with high serum E2 levels, mild abdominal distension, and large ovaries of about 5 cm due to the presence of multiple follicular and corpus luteal cysts.
2. Moderate OHSS, characterized by more abdominal distension, mild as cites, and gastrointestinal symptoms such as nausea, vomiting, and less frequently, diarrhea, some gain in weight, and ovaries enlarged up to 12 cm.
3. Severe OHSS which can be a life-threatening situation, characterized by pronounced abdominal distension, as cites, pleural effusion, hem concentration, electrolyte imbalance, oliguria/ anuria, and sometimes disseminated intravascular coagulation and hypovolemic shock. The ovaries are enlarged to more than 12 cm in diameter.
The treatment of OHSS is conservative. Bed rest and symptomatic relief are usually sufficient for mild and moderate OHSS. In mild cases, symptoms subside usually within a few days, whereas in moderate cases, symptoms can require up to 3 wk to subside. When pregnancy occurs, OHSS will last longer.
Severe OHSS can be life-threatening and patients, therefore, should be hospitalized and monitored closely. Patients are put on bed rest; daily body weight and fluid balance monitoring are necessary; hematocrit, coagulation and kidney functions, serum electrolyte, and albumin studies are obtained daily, as well as ultrasound scan of the pelvis and chest; oxygen saturation measurements (using an oximeter while breathing air and if less than 85% using blood gases) should be performed if dyspnoea is notified; and pelvic examination should be avoided because of the fragility of the enlarged ovaries and ovarian surgery is relatively contraindicated.
Hypovolemia and hem concentration need immediate corrections.
When necessary, intravenous crystalloid infusion or plasma expanders are given to maintain central venous pressure. Replacement may require central venous pressure monitoring. Diuretics should not be given with, as the fluids in the abdominal and thoracic cavity are not responsive to diuretics and the further intravascular depletion can cause hypotension, shock, and thrombosis. Administration of low-dose heparin to prevent thromboembolism needs to be balanced against the risk of ovarian bleeding, possible need for paracentesis, and the possibility of heparin-induced thrombocytopenia. Most of our hospitalized patients are commenced on heparin 5000U_2/day subcutaneously on admission and reviewed after 24 hr. Heparin should be continued until discharge from the hospital if the patient has risk factors or past history of thrombosis, family history of thromboembolism, smoking, obesity, or varicose veins.
Heparin should also be continued if there is hem concentration, tense as cites with inferior vena cava (IVC) compression, and late presentation with possible 56 Healy et al.
Multiple pregnancies. Abdominal paracentesis under ultrasonographic guidance
is recommended where there are as cites with IVC compression and edema of the
lower limbs. Indomethacin has been shown to have beneficial effects on
hyperstimulation syndrome, but its safety in early pregnancy is unknown.
The acute clinical manifestation of the severe form of OHSS usually disappears within several days if there are no complications. If pregnancy does not occur, the ovarian enlargement subsides usually within 30–40 days.
If, however, pregnancy occurs, the high serum E2 levels will continue to the end of the first trimester.
Prevention of OHSS is vital. Low body mass index and PCOS are two clinical
predictors and gonadotropin dosage should be reduced in such patients. When
OHSS is suspected before HCG injection (as when serum E2 >25,000 nmol/L), HCG
should be withheld. If, however, OHSS is diagnosed after oocyte retrieval,
embryos should be frozen in order to avoid conception and replaced later in a
natural cycle (64) or with estrogen/progesterone therapy. This approach also
allows GnRH agonist administration to be renewed immediately after oocyte
retrieval and maintained until the onset of subsequent menstruation, further
reducing the risk of significant OHSS.
We use hCG at a dose of 5000 IU rather than 10,000 IU for final oocyte maturation before retrieval in all our IVF patients to help prevent OHSS.
In 1993, Asch et al. claimed that prevention of severe OHSS was possible by administration of intravenous albumin at a dose of 5% in ringer lactate in doses of 500 ml during oocytes retrieval and 500 ml immediately thereafter. Subsequent prospective randomized trials by Shoham et al., Shalev et al.
, and Shaker et al. all showed that the frequency and degree of OHSS developed in patients treated with intravenous albumin were not different than in control patients. However, the pregnancy rate was significantly higher in patients who had all embryos crypreserved to be transferred subsequently in hormonally manipulated cycle than in patients treated with albumin and fresh ET.
<<Top>>
POOR RESPONDERS
Poor responders to exogenous gonadotropins require markedly increased quantities of gonadotropins to obtain multifollicular development. Moreover, the number of oocytes, fertilization rates, embryo quality, and pregnancy rates are all decreased in this patient population. The incidence of poor response varies from 5% to 18% of unselected IVF patients . Poor response to gonadotropins is a result of diminished ovarian reserve. The most important cause is advanced female age. An age-related decline in fecundity, from 35 years in particular, has been observed in IVF. Other causes are prior chemotherapy, irradiation, ovarian surgery, severe endometriosis, autoimmune disease, or idiopathic. Some patients present with unexplained infertility and regular ovulatory cycles.
Basics of Ovarian Stimulation 57
It is ideal to identify these patients before initiation of
hormonal treatment for IVF. In this way, the patients can be counseled
concerning the lower chances for pregnancy and stimulating protocols can be
modified to lower the risk of cancellation and improve pregnancy rates. In 1987,
Navot et al. described the clomiphene citrate challenge test, a test measuring serum FSH levels before (day 3) and after day 10, the administration of 100 mg of clomiphene citrate on cycle days 5–9. They found that patients with an exaggerated FSH response, despite normal E2 response, had diminished ovarian reserve. They suggested that diminished capacity of the granulose cells to secrete inhibin could explain the discrepancy between E2 and FSH response. This test was found to be a reliable method for predicting IVF-ET outcome. Later, several studies have shown that elevated day 3 serum concentrations of FSH (<6.5–15 IU/L) and/orE2 (<80 pg/ml) have poor pregnancy outcomes with fertility therapy
. Martin et al. evaluated 1868 cycles and found that no pregnancies occurred if
day 3 FSH was >20 IU/L at least in two cycles, but if it happened just once, the
pregnancy rate was 5.6%.Recently, Seifer et al. found that women with low day 3 seruminhibin B concentrations (<45 pg/ml) demonstrated a poorer response to ovulation induction and IVF-ET treatment relative to women with high day 3 inhibin B.
<<Top>>
ORAL CONTRACEPTIVE PRETREATMENT
The role of GnRH antagonists in human IVF is discussed in . Oral contraceptive pre-treatment for ovarian stimulation in a GnRH agonist or a GnRH antagonist cycle has been recently investigated. Oral contraceptive scheduling of a GnRH agonist or a GnRH antagonist protocol results in follicular growth and hormone profile are similar to those observed in GnRH agonist protocols. The number of premature LH rises remains low. Similar numbers of oocyte and high-quality embryos are obtained. This is significant because the use of the oral contraceptive pre-treatment method significantly improves scheduling in a typical IVF program operating Monday to Friday.
The greater convenience of oral contraceptive pre-treatment scheduling appears off-set by the need for longer stimulation protocols and more FSH than with a nonscheduled regimen.
OVARIAN STIMULATION AND OVARIAN CANCER
Concern has been expressed that exposure to fertility drugs
might be associated with a risk of ovarian cancer. In particular, pooled
analysis of three case-controlled studies by Whittemore et al. suggested an odds ratio of 2.8 (95% CI 1.3–6.1) for invasive ovarian cancer infertile women treated with fertility drugs compared with women with no diagnosis of infertility or infertility drug treatment. This study, suggesting an apparent trebling 58 Healy et al.
Of risk in infertile women treated with fertility drugs, has been controversial and widely criticized for various biases in the study design.
In 1994, Rossing et al. undertook a study that yielded the best data, to this time, examining the role of clomiphene citrate and ovarian cancer. Rossing et al. used record linkage with a population-based cancer register. They identified an increased incidence of ovarian cancer comprising both borderline malignant tumors and invasive disease. They found a standardized incidence rate (SIR) of 2.5 with 95% confidence intervals (1.3–4.7) in a cohort of infertile women compared with age-standardized general population rates . This, of course, is a similar rate of increase found by Whittemore et al. in the previously mentioned study. In the Rossing data, an increased relative risk was also found in women who had been treated with clomiphene citrate for more than 1 year when compared with infertile women who had not taken clomiphene citrate. This increased risk of ovarian cancer was seen regardless of whether the patients did or did not have diagnosed ovarian abnormalities and were treated with clomiphene citrate. There was no increased risk when patients were prescribed clomiphene citrate for less than 1 year.
<<Top>>
There are significant clinical differences between the prescriptions of
Clomiphene citrate to a U.S. population of women compared with the administration of clomiphene citrate in other parts of the world. In Australia, e.g.
Clomiphene citrate is only available on prescription from specialist gynecologists and obstetricians rather than from general practitioners as in the United States. Furthermore, clomiphene citrate can only be prescribed in Australia for patients with polycystic ovary syndrome or related disorders and not for women who are infertile but who are spontaneously ovulating. Moreover, our general advice is to use clomiphene citrate in order to achieve four to six ovulatory cycles but not more in the treatment of an infertile woman with polycystic ovary syndrome and chronic an ovulation. The use of clomiphene citrate for more than 1 year in an infertile woman would be highly usual and not recommended in our experience.
Previous studies of cancer after infertility have been limited by low statistical power and difficulties in distinguishing possible effects of fertility drug exposure from the underlying ovulation disorder they were used to treat. We have completed a large study looking at the incidence of breast and ovarian cancer after infertility and IVF . The great majority of these patients received HMG or ovarian stimulation for IVF. In this study, we examined the incidence of breast and ovarian cancer and a cohort of 10,358 women referred for IVF treatment in Victoria, Australia, between 1978 and 1992.
The exposed group (n¼5564) had ovarian stimulation to induce multiple folliculogenesis for IVF. The unexposed group (n¼4794) had been referred for IVF but was untreated or had a natural cycle treatment without ovarian stimulation. The duration of follow-up in both groups ranged from 1 to 15 years.
Basics of Ovarian Stimulation 59
Cases of cancer were determined by record linkage with data from population- based cancer registries. We observed 34 cases of invasive breast cancer and six cases of invasive ovarian cancer. A comparison with the expected numbers derived by applying age-standardized general population rates to the cohort gave SIR for breast cancer of 0.89 (95% CI 0.62–1.56) in the unexposed group. For ovarian cancer, the SIRs were 1.70 (0.55–5.27) and 1.62 (0.52–5.02), respectively. In this large study, we were able to look at the rates of all cancers in these two groups of individuals and not only their rates of breast and ovarian cancers. The rates of all cancers in these two groups of women were not significantly different from the general population rates in Victoria, Australia.
We found that the risk of cancer of the body of the uterus, particularly endometrial adenocarcinoma, was increased in both these groups of infertile women combined [SIR 2.84 (1.18–6.18)]. Women with unexplained infertility, regardless of whether they had gonadotropin treatment for IVF or not, also had a significantly increased risk of ovarian cancer [RR¼19.19 (2.23–165.0)] and cancer of the uterus [RR¼6.34 (1.06–38.0)] compared with women with known causes of infertility.
This relatively short-term follow-up of patients after HMG treatment for IVF
indicates that ovarian stimulation with gonadotropins is not associated with any
increased risk of breast cancer. We also found no significant increased risk of
ovarian cancer after ovarian stimulation for IVF with gonadotropins. However,
the small number of cases limits the conclusions that can be drawn. Longer-term
follow-up of large cohorts of women who have been prescribed gonadotropins for
ovulation induction or IVF programs will be necessary. NICE guidelines suggest
that medical practitioners should confine the use of ovulation induction agents
to the lowest effective dose and shortest duration of use.
ACKNOWLEDGEMENTS
The assistance of the BUPA Foundation in the preparation of this manuscript can be gratefully acknowledged. The authors are indebted to their clinical colleagues and all of their patients.
<<Top>>
|