Prevention of ovarian hyperstimulation syndrome

INTRODUCTION — Ovarian hyperstimulation syndrome (OHSS) is the most serious complication of controlled ovarian hyperstimulation (COH) for assisted reproduction technologies (ART). It is characterized by a broad spectrum of signs and symptoms that includes abdominal distention and discomfort, enlarged ovaries, ascites, and other complications of enhanced vascular permeability. The syndrome can be strictly defined as the shift of serum from the intravascular space to the third space, mainly to the abdominal cavity, in the context of enlarged ovaries due to follicular stimulation. In its very severe form, OHSS is a life-threatening condition.

The prevention of OHSS will be reviewed here. The pathogenesis, clinical manifestations, and management of established OHSS are discussed separately. (See "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome" and "Management of ovarian hyperstimulation syndrome".)

BACKGROUND — Ovarian hyperstimulation syndrome (OHSS) is the most serious complication of controlled ovarian hyperstimulation (COH) for assisted reproduction technologies (ART). It is a broad spectrum of signs and symptoms that include abdominal distention and discomfort, enlarged ovaries, ascites, and other complications of enhanced vascular permeability [1,2]. OHSS is an iatrogenic and potentially life-threatening condition that affects young, healthy women. In addition, there is an important economic burden associated with OHSS due to absence from work, bed rest, or hospitalization and intensive medical management of severe cases.

The pathophysiology of OHSS is not fully understood, but increased capillary permeability with the resulting loss of fluid into the third space is its main feature (see "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome"). In the susceptible patient, human chorionic gonadotropin (hCG) administration for final follicular maturation and triggering of ovulation is the pivotal stimulus for OHSS, leading to overexpression of vascular endothelial growth factor (VEGF) in the ovary, release of vasoactive-angiogenic substances, increased vascular permeability, loss of fluid to the third space, and full-blown OHSS (algorithm 1).

There are two clinical forms of OHSS, both hCG related: the early-onset form (occurring in the first eight days after hCG administration) and the late-onset form (occurring nine or more days after hCG administration, related to pregnancy-induced hCG production) [3]. (See "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome", section on 'Onset'.)

PREVENTION OF OHSS — Ovarian hyperstimulation syndrome (OHSS) remains a common complication among women undergoing assisted reproductive technologies (ART); reported frequencies are approximately 20 to 33 percent for mild cases, 3 to 6 percent for moderate cases, and 0.1 to 2 percent for severe cases [4-6]. (See "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome", section on 'Epidemiology'.)

Keys to prevention — The keys to preventing OHSS are to identify the potential risk for the individual patient and to plan appropriate strategies to avoid its occurrence. The main steps include [7]:

●Recognition of risk factors for OHSS (such as previous episode of OHSS or exuberant ovarian response to gonadotropins, and polycystic ovarian syndrome [PCOS]). (See "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome", section on 'Risk factors'.)

●Extensive clinical experience with drugs used for ovarian stimulation (for ovulation induction, but more importantly for assisted reproduction, as the ovarian stimulation protocols are more aggressive and OHSS is more common). (See 'Ovarian stimulation protocol' below.)

●Use of individualized ovarian stimulation regimens for assisted reproduction, using the minimum dose and duration of gonadotropin therapy necessary to achieve the treatment goal (particularly in those at risk for OHSS). (See 'Ovarian stimulation protocol' below.)

•Other interventions such as pretreatment with metformin and addition of a gonadotropin-releasing hormone (GnRH) antagonist are suggested in women with PCOS, a high-risk group. (See 'Pretreatment with metformin' below and 'Addition of GnRH agonist or antagonist' below.)

●Modifying treatment when indicators for increasing OHSS risk develop:

•Serum estradiol (E2) concentration >3500 pg/mL (12,850 pmol/L)

•Development of many intermediate-sized follicles (more than 20 follicles >10 mm) [8]

●Withholding gonadotropin therapy while continuing pituitary suppression with a GnRH agonist or antagonist until serum E2 levels fall into a range acceptable for human chorionic gonadotropin (hCG) administration (coasting) if the risk for OHSS is high. (See 'Coasting' below and 'Withholding hCG (cycle cancellation)' below.)

●Using an alternative to standard-dose hCG for final oocyte maturation (lower dose hCG or gonadotropin-releasing hormone [GnRH] agonist). (See 'Ovulatory triggers' below.)

Ovarian stimulation protocol

Gonadotropin dose and type — We suggest the use of individualized ovarian stimulation regimens, using the minimum dose and duration of gonadotropin therapy necessary to achieve follicular development and pregnancy, while avoiding the risk of OHSS.

Gonadotropin doses correlate with OHSS severity as indicated by ovarian size and severity of ascites and pleural effusion [9]. Therefore, in patients at risk, the starting dose of gonadotropins should be decreased (to 100 to 150 international units) [10]. The current approach to ovarian stimulation emphasizes an approach to dosing based upon patient variables such as age, body mass index (BMI), baseline serum anti-müllerian hormone concentrations, antral follicle count, and previous ovarian response. This approach typically uses lower doses than fixed ovarian stimulation protocols.

A clinical trial based upon this approach reported that clinical pregnancy rates were similar with a standard and individualized approach, but that the cancellation and OHSS rates were lower with the individualized approach [11]. Of note, there is currently no consensus on the ideal dosing approach that optimizes pregnancy rates yet minimizes OHSS risk. (See "In vitro fertilization: Overview of clinical issues and questions".)

In contrast to dose, the type of gonadotropin preparation (eg, human menopausal gonadotropins [hMG] versus recombinant follicle-stimulating hormone [rFSH] versus urinary follicle-stimulating hormone [uFSH]) does not appear to affect the risk of OHSS [12-14].

Monitoring — We suggest monitoring patients using both transvaginal ultrasound (TVUS) (for follicular number and size) and serum E2 concentrations. Since multiple follicle development [15] and high E2 levels [16] are important risk factors, early detection of either helps prevent OHSS by withholding the ovulatory dose of hCG in high-risk cycles.

One meta-analysis of eight studies of women undergoing in vitro fertilization (IVF) concluded that TVUS alone was adequate for patient monitoring, as clinical pregnancy and OHSS rates were similar in women monitored with TVUS alone or TVUS combined with serum E2 monitoring [17]. However, the overall quality of the evidence was low. Given the potential risks and important consequences of OHSS, we continue to suggest monitoring with both TVUS and serum E2.

Addition of GnRH agonist or antagonist — In the setting of controlled ovarian stimulation with exogenous gonadotropins for IVF, a GnRH agonist or antagonist is also administered to prevent the endogenous luteinizing hormone (LH) surge. hCG is then given to induce final oocyte maturation prior to oocyte retrieval, when follicles are judged to be mature based upon size and serum E2 concentrations. (See "In vitro fertilization: Overview of clinical issues and questions".)

We suggest the use of GnRH antagonists rather than GnRH agonists in women at high risk for OHSS. The use of GnRH agonists is associated with a higher incidence of OHSS, probably due to enhanced follicular recruitment. In contrast, prospective randomized trials and two meta-analyses have reported a lower incidence of OHSS when GnRH antagonists are used, including in women with PCOS [18-25]. However, these same analyses suggest that the use of GnRH antagonists may result in lower pregnancy rates when compared with GnRH agonist therapy.

The use of GnRH antagonists followed by a GnRH agonist for final oocyte maturation for women donating oocytes is discussed below. (See 'GnRH agonist trigger' below.)

Coasting — Coasting refers to withholding gonadotropin therapy while continuing pituitary suppression with a GnRH agonist or antagonist until serum E2 levels fall into a range acceptable for hCG administration (eg, associated with a lower OHSS risk). This approach is less common now because of the availability of GnRH antagonists. The larger follicles can continue their growth and maturation when follicle-stimulating hormone (FSH) is stopped; the smaller follicles have a greater FSH requirement and therefore undergo atresia. Although we agree with trying this approach in high-risk cycles, available data are conflicting on its impact on preventing OHSS [26,27].

We typically start coasting when the dominant follicles are ≥16 mm and serum E2 levels are >3500 pg/mL (12,850 pmol/L) [26]. Once initiated, daily TVUS and serum E2 measurements should be performed; administration of hCG should be withheld until serum E2 falls below 3500 pg/mL (12,850 pmol/L). Coasting for greater than three days (but not up to three days) has a modest adverse effect on pregnancy rates [28]. Therefore, we consider cycle cancellation if E2 levels have not fallen by the fourth day of coasting.

Based on one review, this approach results in an acceptably low incidence of severe OHSS (<2 percent) with satisfactory fertilization and pregnancy rates (55 to 71 and 37 to 63 percent, respectively) [29].

Withholding hCG (cycle cancellation) — Cycle cancellation before administration of exogenous hCG is an effective strategy to prevent OHSS, ie, postponing the treatment cycle until the ovaries have been rendered quiescent [30]. However, cycle cancellation has financial and emotional implications, frustrates both patient and clinician, and results in cancellation of a high percentage of cycles that would not have progressed to clinical OHSS. For patients undergoing agonist cycles at high-risk for severe OHSS, this approach is still a valid and safe alternative that prevents both early and late-onset OHSS forms. As noted, we consider cycle cancellation if E2 levels have not fallen by the fourth day of coasting. (See 'Coasting' above.)

Pretreatment with metformin — We suggest metformin pretreatment for women with PCOS undergoing IVF. Metformin is an antidiabetic drug that has been extensively investigated in the management of PCOS. Its indications in the management of PCOS have diminished, but it does appear to be useful for pretreatment prior to controlled ovarian stimulation prior to IVF for reducing the risk of OHSS. (See "Metformin for treatment of the polycystic ovary syndrome".)

In two studies, metformin pretreatment beginning four to five weeks before starting gonadotropin therapy for IVF reduced the risk of OHSS [31,32]. In a meta-analysis of eight trials (n = 798 women), metformin pretreatment decreased the risk of OHSS when compared with placebo (odds ratio [OR] 0.29, 95% CI 0.18-0.49) [33]. The authors calculated that for a woman with a 27 percent risk of having OHSS without metformin, the corresponding chance using metformin treatment would be between 6 and 15 percent. Similar results were seen in a second meta-analysis of 10 trials of over 800 women; metformin pretreatment reduced the risk of OHSS (OR 0.27, 95% CI 0.16-0.46) [34]. No difference in live birth rate was seen with metformin compared with placebo in either meta-analysis [33,34].

Luteal phase support — We suggest progesterone regimens over hCG regimens for luteal phase support because they are associated with a lower risk of OHSS (see "In vitro fertilization: Overview of clinical issues and questions"). Progesterone supplementation is generally initiated on the day of oocyte retrieval or at the time of embryo transfer. An alternative approach has been to administer intermittent, low doses of hCG. In a meta-analysis of seven trials comparing luteal phase progesterone and hCG regimens, the risk of OHSS was significantly lower with progesterone (OR 0.45, 95% CI 0.26-0.79) [35].

Ovulatory triggers

Low versus standard-dose hCG — hCG, which mimics the LH surge, is the standard drug administered for inducing final oocyte maturation. As noted, it is thought to be the pivotal stimulus for OHSS; if withheld, the most severe types of OHSS do not occur (table 1). The effect of hCG is related to its high biological activity, which is six to seven times that of endogenous LH (because of hCG's longer half-life) [36].

In patients at risk, it had been assumed that lower doses of hCG would be associated with lower risks of OHSS by decreasing vascular endothelial growth factor (VEGF) secretion by granulosa cells [37,38]. Evidence to date, including a retrospective study [39], a randomized trial [40], and a systematic review [41], suggests that lowering the dose of hCG to as low as 2500 to 3300 international units results in successful oocyte maturation with similar fertilization and pregnancy rates as doses of 5000 or 10,000 international units. However, it is not clear that the risk of OHSS is reduced, because both the studies and the number of cases are small.

Recombinant LH/recombinant hCG — The risk of OHSS appears to be similar with recombinant LH (rLH), recombinant hCG (rhCG) and urinary hCG. Recombinant hCG is a purer preparation of hCG that can be administered subcutaneously, as opposed to urinary hCG which is administered intramuscularly. The risk of OHSS appears to be similar with the two preparations [42]. The biologic half-life of rLH is approximately 10 hours, which is significantly shorter than that of urinary hCG preparations (24 to 36 hours). Therefore, it had been thought that the risk of OHSS may be lower with rLH. However, pooled data from three trials comparing rLH and urinary hCG showed no difference in achieving final follicular maturation in IVF, with similar pregnancy and OHSS rates [42].

GnRH agonist trigger — In GnRH antagonist cycles, the administration of a GnRH agonist at the end of ovarian stimulation induces an endogenous rise in both LH and FSH concentrations and effectively triggers oocyte maturation [43-47]. This is a useful approach in high-risk patients, preventing both early and late forms of OHSS. In a meta-analysis of 13 studies that assessed fresh autologous IVF cycles and four that assessed donor-recipient cycles, the use of GnRH agonist, when compared with hCG for final oocyte maturation, lowered the risk of OHSS but had a negative impact on live birth rates in the autologous IVF group but not the donor-recipient cycles. We therefore suggest this approach in oocyte donors, a group of healthy young women in whom minimizing OHSS risk is a particularly high priority [48,49]. In IVF patients, we typically use leuprolide 0.5 to 1 mg or triptorelin 0.1 to 0.2 mg subcutaneously. We agree with the Society of Obstetricians and Gynecologists of Canada who suggest using a GnRH antagonist protocol that includes a GnRH agonist as a substitute for hCG for final oocyte maturation in two settings [49]:

●Women at high risk for OHSS (development of multiple follicles during ovarian stimulation [>20 follicles over 10 mm]) (table 2)

●Women undergoing ovarian stimulation who plan to donate oocytes to recipients, or women undergoing fertility preservation cycles (eg, women planning to freeze their oocytes or embryos for future use)

Other interventions

In vitro oocyte maturation — In vitro maturation (IVM) is an experimental technique that consists of the in vitro conversion of oocytes at the germinal vesicle stage to oocytes at the metaphase II stage. To be successful, this technology must include nuclear and cytoplasmic maturation of the oocyte and give rise to embryos that have a developmental potential that is similar to embryos obtained from standard IVF or from spontaneously in vivo matured oocytes. Initial studies with IVM resulted in low fertilization rates and suboptimal embryonic quality [50]. However, preliminary studies in women with PCOS suggest that IVM is associated with a lower OHSS rate but also a lower live birth rate than IVF using a GnRH antagonist protocol [51,52]. This technique is not yet used clinically.

Embryo cryopreservation — Although elective cryopreservation is often considered in cycles at risk for OHSS, its efficacy is not well established. Observational studies [53-55] and one prospective study [56] suggest that elective cryopreservation prevents severe OHSS and that embryo replacement in subsequent natural cycles results in good pregnancy rates. In contrast, in a review of two clinical trials of embryo cryopreservation (one compared with intravenous albumin therapy and one compared with fresh embryo transfer), no differences were observed between groups in the incidence of OHSS, pregnancy rates, or live birth rates [57]. Although data are conflicting, we currently suggest embryo cryopreservation in the following settings:

●Patients at risk for OHSS (>20 follicles larger than 10 mm) who received a GnRH agonist trigger. In these patients, the risk of moderate/severe OHSS is virtually zero, but implantation rates are lower due to impairment of endometrial receptivity. Until luteal support is well established for these cases, cryopreservation of all oocytes/embryos is the best approach.

●Patients at high risk for OHSS who received hCG. Those patients should have their oocytes/embryos cryopreserved to avoid pregnancy and late OHSS. They are still at risk for early OHSS, and dopamine agonists should be considered. (See 'Dopamine agonists' below.)

Women with polycystic ovary syndrome — Women with PCOS who undergo IVF have lower rates of OHSS with the transfer of frozen rather than fresh embryos. In the largest trial to date, frozen embryo transfer was associated with a lower incidence of OHSS (1 versus 7 percent) compared with fresh embryo transfer [58]. (See "In vitro fertilization: Overview of clinical issues and questions".)

Intravenous albumin — We do not suggest the routine use of intravenous albumin or other volume expanders for the prevention of OHSS. The potential mechanism of action of prophylactic intravenous fluids is to increase plasma oncotic pressure, maintain intravascular volume, and bind to and inactivate ovarian mediators (eg, VEGF) involved in the development of OHSS [59]. In a meta-analysis of eight randomized trials of human intravenous albumin versus placebo, there was only a borderline statistically significant decrease in the incidence of severe OHSS with albumin (OR 0.67, 95% CI 0.45-0.99).

These results are consistent with two previously published meta-analyses, both of which found no significant benefit of intravenous albumin for reducing the occurrence of severe OHSS in high-risk women [60,61]. In one analysis, pregnancy rates were lower in women receiving albumin compared with no treatment (relative risk [RR] 0.85, 95% CI 0.74-0.98) [61]. Thus, we do not suggest this approach as there is no evidence of benefit and a possible negative impact on pregnancy rates.

A second volume expander, hydroxyethyl starch, was associated with a reduction in OHSS rates compared with placebo in three trials (OR 0.12, 95% CI 0.04-0.40), but the safety of this product has not been well established.

Intravenous albumin has also been used for intravenous hydration in the management of severe OHSS. However, available evidence suggests that intravenous albumin provides no additional benefit when compared with crystalloid solutions. (See "Management of ovarian hyperstimulation syndrome", section on 'Severe and critical OHSS' and "Treatment of severe hypovolemia or hypovolemic shock in adults", section on 'Normal saline (crystalloid)'.)

Dopamine agonists — We suggest dopamine agonist administration in women at high risk for OHSS who have received hCG. Dopamine agonists inhibit VEGF receptor phosphorylation and thereby decrease vascular permeability, and a number of studies have reported that either dopamine or dopamine agonists reduce the risk of OHSS in women undergoing controlled ovarian stimulation [62-68].

●Cabergoline – In a pooled analysis [69] of two placebo-controlled trials of women at high risk of OHSS undergoing ovarian stimulation who were randomly assigned to receive cabergoline (0.5 mg daily beginning on the day of hCG administration [66] or the day of oocyte retrieval [68]) for prevention of OHSS, the rate of moderate OHSS was significantly lower with cabergoline than placebo (20 and 44 percent, respectively). In addition, pregnancy rates do not appear to be compromised by the addition of cabergoline [67].

●Quinagolide – Quinagolide, another dopamine agonist, appears to reduce the risk of early onset severe OHSS. This was demonstrated in a multicenter trial of 182 IVF patients randomized to receive quinagolide (50, 100, or 200 mcg/day beginning on the day of hCG administration) or placebo [8].

The incidence of moderate/severe early-onset OHSS was 23 percent in the placebo group and 12 percent, 13 percent, and 4 percent in the quinagolide 50, 100, and 200 mcg/day groups, respectively. The overall rate of moderate/severe OHSS was lower with all quinagolide groups combined compared with the placebo group (OR 0.28, 95% CI 0.09-0.81). No severe OHSS cases were reported in the 200 mcg/day group. Among women who conceived and who were therefore at risk for late-onset OHSS, quinagolide had no beneficial effect.

Quinagolide is available in some countries, but not the United States, and may be more difficult to tolerate than cabergoline.

Low-dose aspirin — Two studies suggest that low-dose aspirin may be associated with a lower risk of OHSS [70,71]. However, we do not suggest its use at this time.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Ovarian hyperstimulation syndrome".)

SUMMARY AND RECOMMENDATIONS

●The keys to preventing ovarian hyperstimulation syndrome (OHSS) are to identify the potential risk for the individual patient and to plan appropriate strategies to avoid its occurrence (table 2).

●Risk factors include a previous episode of OHSS or exuberant ovarian response to gonadotropins, and polycystic ovarian syndrome (PCOS) (table 2). (See "Pathogenesis, clinical manifestations, and diagnosis of ovarian hyperstimulation syndrome", section on 'Risk factors'.)

●We suggest the use of individualized ovarian stimulation regimens, using the minimum dose and duration of gonadotropin therapy necessary to achieve the treatment goal (particularly in those at risk for OHSS) (Grade 2B). (See 'Ovarian stimulation protocol' above.)

●In the setting of controlled ovarian stimulation with exogenous gonadotropins for IVF, a GnRH agonist or antagonist is also administered to prevent the endogenous luteinizing hormone (LH) surge. For women at high risk for OHSS, we suggest the use of GnRH antagonists rather than GnRH agonists (Grade 1B). (See 'Addition of GnRH agonist or antagonist' above.)

●We suggest metformin pretreatment for women with PCOS undergoing IVF (Grade 2B). (See 'Pretreatment with metformin' above.)

●During gonadotropin therapy, indicators of increasing OHSS risk that require treatment modification include:

•Serum estradiol (E2) concentration >3500 pg/mL (12,850 pmol/L).

•Development of many intermediate sized follicles (>20 follicles over 10 mm diameter).

●Treatment modification options in high-risk cycles to lower OHSS risk include:

•Withholding gonadotropin administration (coasting) until serum E2 falls below 3500 pg/mL (12,850 pmol/L). (See 'Coasting' above.)

•Using an alternative to standard-dose human chorionic gonadotropin (hCG) for final oocyte maturation (GnRH agonist). (See 'Ovulatory triggers' above.)

•Administration of dopamine agonist if hCG has already been given as the ovulatory trigger.

ACKNOWLEDGMENT — The editorial staff at UpToDate would like to acknowledge Bruno Lunenfeld, MD and the late Vaclav Insler, MD, who contributed to earlier versions of this topic review.