Outpatient management of severe early OHSS by administration of GnRH antagonist in the luteal phase: an observational cohort study

Ovarian hyperstimulation syndrome (OHSS) is a serious complication of ovarian stimulation in patients undergoing in-vitro fertilization (IVF) treatment, which is triggered by human chorionic gonadotrophin (hCG). There are two main clinical forms of OHSS, early and late OHSS, depending on the time of onset. Early OHSS is induced by exogenous hCG administered for final oocyte maturation, usually occurring within 3–7 days post hCG [1, 2]. Late OHSS is pregnancy-induced, occurs 12–17 days post hCG and is triggered by the endogenous hCG produced by an implanting blastocyst [1, 2]. OHSS is further distinguished in mild, moderate and severe forms, depending on the severity of symptoms [3]. Mild OHSS lacks clinical significance, moderate OHSS requires careful patient monitoring, while severe OHSS may prove to be critical or even life-threatening, characterized by massive ovarian enlargement, ascites, pleural effusion, oliguria, haemoconcentration, adult respiratory distress syndrome and thromboembolic phenomena, and may require hospitalization in an intensive care unit [4, 5].

Severe OHSS, although infrequent in the general IVF population, represents a really difficult situation for both patients and physicians. In high risk patients [6], the published incidence of severe OHSS after ovarian stimulation for IVF ranges from 10% to 38% [7‐9]. This high variation in the occurrence of OHSS is mainly due to the lack of a universally accepted criteria for diagnosis and classification of OHSS [3, 10].

It has been reported that in high-risk for OHSS PCOS patients, the use of GnRH antagonists is associated with a significantly decreased incidence of OHSS by 20% compared to the use of the long GnRH agonist protocol [11]. This observation was confirmed by a recent Cochrane Review [12], which reported a significant reduction in severe OHSS using the GnRH antagonist protocol in both PCOS patients and the general IVF population. However, the use of GnRH antagonists may only decrease the incidence of OHSS, which can still occur. According to the concept of an OHSS-free clinic [13], it has been proposed to trigger final oocyte maturation by replacing hCG with GnRH agonist in antagonist protocols, which appears to totally prevent the syndrome [14]. However, following GnRH agonist triggering, embryo transfer in the same cycle is associated with a significantly lower probability of pregnancy and therefore, embryo cryopreservation and transfer in a subsequent frozen-thawed cycle is usually performed [14, 15]. Despite the availability of agonist triggering in antagonist protocols, some patients at high risk for OHSS will still choose to proceed to oocyte retrieval and embryo transfer using a lower dose of hCG to trigger final oocyte maturation [16] and a proportion of them will eventually develop OHSS.

It should be emphasized that in patients downregulated with GnRH agonists, which currently represent the majority of IVF patients, hCG is the only way available for triggering final oocyte maturation and thus OHSS is more likely to occur.

Alternatively, the dopamine agonist cabergoline [17], and more recently quinagolide [18], have been shown to reduce the incidence and severity of OHSS [17‐21].

Currently, despite an extensive list of available prevention methods [6], if severe OHSS occurs there is no established way of management apart from conservative treatment, involving correction of fluid and electrolyte imbalance, prevention of thromboembolism, aspiration of the ascitic fluid etc. [5].

Recently, it has been suggested that GnRH antagonist administration in the luteal phase in patients with established severe early OHSS appears to prevent patient hospitalization and to result in quick regression of the syndrome on an outpatient basis. This intervention appears to be effective in both agonist and antagonist-treated patients. However, the existing published data are three small case series [22‐24], which although promising, require further evaluation.

The aim of the present study was to evaluate in a larger series of patients whether it is feasible to manage women with established severe early OHSS as outpatients by treating them with GnRH antagonists in the luteal phase and cryopreserving all their embryos.

From the cohort of 362 high risk patients included in the present study, triggering of final oocyte maturation was performed by administration of 5000 IU of hCG in 353 patients, or by administration of GnRH agonist in 9 patients (Figure 1).

Baseline characteristics of the 353 patients in whom hCG was injected for triggering final oocyte maturation and either developed (n = 40) or did not develop (n = 313) severe early OHSS are shown in Table 2, while ovarian stimulation and embryological data are shown in Table 3.

Patients who developed severe early OHSS had similar baseline characteristics compared to patients who did not develop severe early OHSS, apart from a significantly lower baseline FSH (p = 0.008) (Table 2).

Regarding ovarian stimulation and embryological data, patients who developed severe early OHSS had lower total FSH dose (p = 0.014), higher number of follicles (p < 0.001) and higher oestradiol levels (p < 0.001) on the day of triggering final oocyte maturation, higher number of oocytes retrieved (p < 0.001), higher number of mature oocytes (p < 0.001), and higher number of fertilized oocytes (p < 0.001) compared to patients who did not develop severe early OHSS (Table 3).

The proportion of patients at high risk for OHSS who developed severe early OHSS was 11.3% (95% CI 8.3%-15.0%) (40/353). Incidence of late OHSS in patients at high risk who were triggered with hCG, did not develop early OHSS and proceeded to embryo transfer was 1.1% (95% CI 0.36%-2.71%) (4/353) and was associated with pregnancy achievement.

All 40 patients with severe early OHSS had blastocyst cryopreservation in combination with GnRH antagonist administration.

In all 40 patients with severe early OHSS outpatient management was feasible and none required hospitalization following administration of GnRH antagonist 5 days after oocyte retrieval and embryo cryopreservation (0%, 95% CI: 0 to 8.8). Patient monitoring showed improvement of patients’ symptoms, ultrasound and laboratory findings (Figures 2, 3).

In the 40 patients with severe early OHSS diagnosed on day 5 post oocyte retrieval, the highest values of hematocrit, WBC count, ovarian volume, as well as oestradiol and progesterone were observed on the day of severe OHSS diagnosis (day 5 post oocyte retrieval) (Figure 2).

Compared to Day 5 post oocyte retrieval, hematocrit, ovarian volume, oestradiol and progesterone declined significantly (p < 0.01) two days (Day 7) after initiation of GnRH antagonist administration, while WBC count displayed a significant decrease on day 9, compared to day 5 (day of GnRH antagonist initiation) (Figure 2).

All 40 patients with severe OHSS were diagnosed with significant ascites on the day of GnRH antagonist initiation (27.5% massive/tense ascites; 72.5% moderate/marked ascites), which progressively declined to moderate levels (35%) or low/no detectable levels (65%) at the end of the monitoring period (p < 0.01) (Figure 3). Three patients (3/303) displayed marked ascites on day 11 of monitoring.

One patient with severe early OHSS developed mild respiratory problems on Day 5 post oocyte retrieval. The patient was examined by a pathologist, O₂ saturation was checked, and it was decided that admission to hospital was not necessary at that moment. The patient was instructed to contact the doctors in case she developed dyspnoea, dizziness, fainting, or general discomfort, for immediate admission to hospital. However, the mild respiratory problem disappeared within 24 hours, along with a general improvement of clinical symptoms.

In high risk patients who did not develop severe early OHSS and proceeded to embryo transfer (n = 303), biochemical, clinical and ongoing pregnancy rates were 70.3% (n = 213), 59.1% (n = 179) and 50.5% (n = 153), respectively (Table 4).

This is the first observational cohort study in the literature describing outpatient management of established severe early OHSS, using administration of GnRH antagonist in the luteal phase and cryopreservation of all embryos. None of the patients evaluated required hospitalization, suggesting that outpatient management is feasible using this approach. Luteal antagonist administration was associated with rapid regression of established severe early OHSS, improvement of patient symptoms and ultrasound and laboratory findings. This flexible approach allows high risk patients to safely proceed at least to oocyte retrieval using low dose hCG for triggering final oocyte maturation, and potentially to embryo transfer if severe OHSS does not occur, thus avoiding an unnecessary embryo transfer cancellation in the majority (88.7%) of high risk patients.

Administration of GnRH antagonist in the luteal phase appears to result in rapid resolution of severe OHSS as early as two days after initiation of GnRH antagonist, with a significant decline of ovarian volume, hematocrit and ascites, as well as oestradiol and progesterone concentrations, confirming previous reports published in three small case series [22‐24]. This rapid decline continued in a progressive manner until the end of the monitoring period.

The rapid decrease of ovarian volume, oestradiol and progesterone levels, observed in the patients with established severe OHSS, suggests a luteolytic effect of the GnRH antagonist, as recently proposed [23, 24]. It is believed that hCG administration for triggering of final oocyte maturation induces massive luteinization, elevated secretion of angiogenic factors (such as vascular endothelial growth factor, angiotensin II, interleukins, histamine, prolactin, prostaglandins, endothelin-1, selectins) from multiple corpora lutea of hyperstimulated ovaries, leading to an increase of vascular permeability, fluid shift to the third space and finally development of OHSS [33‐36].The luteolysis induced by GnRH antagonist possibly leads to a decrease of ovarian activity and to minimized secretion of locally produced angiogenic factors, resulting in regression of severe OHSS.

The results obtained in the current study might be explained by a direct action of GnRH antagonist on the ovary. The presence and function of extrapituitary GnRH receptors has been demonstrated in several tissues, including the human ovary [37]. In addition GnRH antagonists have been shown to inhibit the expression of locally produced ovarian angiogenic factors, such as VEGF [38], in human granulosa luteal cell cultures.

It seems unlikely that the luteolytic action of GnRH antagonist occurs by a decrease in LH secretion, since LH concentrations following ovarian stimulation for IVF are deeply suppressed in the luteal phase [39].

It can be assumed that severe OHSS resolution observed within the monitoring period of the current study may be due to the action of GnRH antagonist rather than natural course of the syndrome. It is recognized that OHSS is a self-limited disease, which however requires an extensive period for natural regression, with prolonged hospitalization ranging from 11 to 23 days [40], occasionally in intensive care units, accompanied by multiple ascites punctures, human albumin administration, correction of intravascular fluid volume and electrolyte imbalance [32]. On the contrary, our proposed approach resulted in rapid regression of severe OHSS as early as two days after initiation of GnRH antagonist, without any use of invasive treatment for the patients and avoiding the need for hospitalization.

It has been shown that while elective embryo cryopreservation can prevent pregnancy-induced late OHSS, it cannot completely eliminate early OHSS, which is induced by exogenous administration of hCG for triggering final oocyte maturation [3, 41‐43].

In addition, it was recently proposed that early OHSS can be equally severe as late OHSS, requiring a mean of 6.8 to 20 days of hospitalization depending on whether the women subsequently became pregnant or not [44], despite previous reports showing that late pregnancy-induced OHSS is associated with more severe symptoms [1, 2].

In the present study, only 11.3% (95% CI 8.3%-15.0%) (40/353) of high risk patients receiving hCG developed severe OHSS and required the intervention (i.e. GnRH antagonist administration and cryopreservation of all embryos). The occurrence of severe OHSS presented here is in agreement with the 12.6% rate shown for PCO patients [9], and is at the lower end (10%) of the incidence range previously reported in the literature for high risk patients (10-38%) [7, 8]. In addition, the extremely low percentage (1.1%) of late OHSS in the present study is significantly lower than the 10% rate of late OHSS previously reported for women with polycystic ovaries [2], probably indicating the importance of accurate identification and monitoring of high risk patients. Using the proposed approach, it is possible to minimize the occurrence of late OHSS by accurate diagnosis of early severe OHSS prior to the performance of embryo transfer.

Despite the prevention of late OHSS by elective embryo cryopreservation shown in the present study, the possible transfer of embryos combined with luteal administration of GnRH antagonist could potentially be problematic and lead to lower implantation rates due to a putative negative effect of the antagonist in the hormonal profile to sustain implantation.

The proposed approach offers a flexible alternative solution in cases when GnRH agonist for triggering final oocyte maturation (OHSS-free clinic [13]) is not feasible, i.e. in patients treated with a long protocol or patients not accepting embryo transfer cancellation. Also, administration of GnRH antagonist in the luteal phase may prove useful in patients who develop severe OHSS despite the use of preventive measures.

In conclusion, the current study suggests, for the first time, that successful outpatient management of severe OHSS with antagonist treatment in the luteal phase is feasible and is associated with rapid regression of the syndrome, challenging the dogma of inpatient management. The proposed management is a flexible approach that minimizes unnecessary embryo transfer cancellations in the majority of high risk for OHSS patients.