Severe early-onset ovarian hyperstimulation syndrome with liver dysfunction in an IVF segmentation cycle

Abstract

Severe early-onset ovarian hyperstimulation syndrome (OHSS) with deranged liver function tests is an entity that cannot be eliminated wholly even after gonadotropin-releasing hormone (GnRH) agonist trigger without any luteal human chorionic gonadotropin (hCG) rescue in a GnRH antagonist protocol with the freeze-all approach. We describe a case of young polycystic ovary syndrome patient with prior history of severe early-onset OHSS in her last in vitro fertilisation cycle in which she received antagonist protocol followed by blastocyst transfer. Given her history, she was planned for agonist trigger and freeze all approach during the present cycle. Despite segmentation of the cycle without any luteal rescue hCG, she developed early-onset severe OHSS with markedly deranged liver function tests for which she underwent ascitic tapping and remained hospitalised for 8 days. Her symptoms improved with conservative management, and she was discharged satisfactorily. She underwent letrozole based frozen–thawed embryo transfer cycle after 4 months. One good quality blastocyst was transferred, and she conceived in the same cycle.

Background

The polycystic ovary syndrome (PCOS) is not a rare endocrine disorder in women of reproductive age, and it affects 5%–10% of women.¹ Nearly 70%–80% of women having PCOS are grieved with infertility.² In vitro fertilisation (IVF) is the only treatment option in PCOS women who do not respond to conventional infertility treatment modalities or who have coexisting infertility factors.³ PCOS patients undergoing controlled ovarian stimulation (COS) have a substantial risk of developing ovarian hyperstimulation syndrome (OHSS), which is a severe iatrogenic complication of ovarian stimulation. In its severe form it is a life-threatening condition, and hCG, either exogenous or endogenous, is the initiating factor. It is a self-limiting entity presented with clinical manifestations related to increased capillary permeability and fluid retention mediated by many inflammatory mediators, including vascular endothelial growth factor (VEGF).¹ Currently, the best-employed strategy to prevent OHSS is the ‘freeze-all strategy’, which comprises the use of a gonadotropin-releasing hormone (GnRH) agonist trigger for oocytes maturation in a GnRH antagonist protocol, followed by cryopreservation of all embryos.⁴ However, the freeze-all strategy might not wholly eliminate OHSS in all patients.⁵ We herein describe a rare case of polycystic ovarian syndrome who had severe early-onset OHSS with deranged liver function tests after a GnRH agonist trigger without any luteal hCG rescue in a GnRH antagonist protocol with the freeze-all approach.

Case presentation

A 32-year-old woman with a diagnosed case of polycystic ovarian syndrome presented to our infertility clinic with secondary infertility of 1 year and the previous one failed IVF cycle. Her menstrual cycles had been regular. Husband’s semen analysis was normal (count: 43 million/mL, motility 40%, morphology 4%).

Her obstetric history revealed that she conceived on ovulation induction 7 years back for which medical termination of pregnancy (MTP) was done for missed abortion. Again, in January 2018, she conceived on ovulation induction and intrauterine insemination for which laparoscopic salpingectomy was done for ectopic gestation. She underwent one cycle of COS for IVF with antagonist protocol in July 2018. During this cycle, she was stimulated for 11 days with a starting dose of recombinant Follicle Stimulating Hormone (rFSH) as 300 IU. Total 10 oocytes were retrieved and two fresh blastocysts were transferred. She had early-onset OHSS for which she remained hospitalised for 10 days, and ascitic fluid tapping was also done. However, no records of hospital stay were available with the patient, but she was discharged under satisfactory conditions.

She was enrolled in the IVF programme at our centre. She had a normal early follicular phase endocrine profile of Follicle Stimulating Hormone (FSH) 6.17 mIU/L, Luteinizing Hormone (LH) 0.64 mIU/L, AMH 5.69 ng/mL, TSH 1.72 uIU/mL, prolactin 16.8 ng/mL. Her body mass index was 21.8 kg/m² (weight: 55 kg). Her baseline scan showed bilateral polycystic ovarian morphology with antral follicle count of 15 follicles each ovary. Endometrial thickness was 8 mm, triple-layered. Before stimulation, she was informed about the OHSS risk, and agonist trigger and freeze-all strategy for OHSS prevention and informed consent was obtained.

Ovarian stimulation—the patient was stimulated in a flexible GnRH antagonist protocol. Stimulation with rFSH 225 IU started on day 2 of the menstrual cycle. On day 5 of stimulation, her serum oestradiol (E2) level was 2100 pg/mL and GnRH antagonist (cetrorelix 0.25 mg) was started on the same day. She was triggered with leuprolide acetate 2 mg subcutaneously on stimulation day 9 for final oocyte maturation. The total dose of gonadotropin received by the patient was 2000 IU. On the day of trigger, her E2 was 8000 pg/mL and progesterone (P4) was 1.7 ng/mL.

At oocyte retrieval 36 hours later, 11 oocytes were retrieved. She was kept prophylactically on dopamine agonist cabergoline 0.5 mg once a day. In total, six embryos were generated and cryopreserved as day three embryos.

On day 3 post retrieval, she presented in emergency with symptoms of nausea, vomiting and abdominal pain. There was no breathlessness and no claims of decreased urine output. On examination, her vitals were stable. There was no abdominal distension and rigidity.

Investigations

The initial blood count revealed haematocrit 39.5%, platelet count 116 × 10ˆ9/L, total leucocyte count 15.59 × 10ˆ9/L.

Ultrasound showed bilateral enlarged ovaries of 8 cm in diameter each with free fluid in pelvis and around ovaries. Following day, her haematocrit rose to 43% with normal renal function tests; electrolytes and liver function tests. On day 4 of hospital stay, haematocrit was 46% with deranged liver function tests (Total bilirubin: 3.6 mg% (normal range: 0.8–1.0 mg%), Serum glutamic-oxaloacetic transaminase (SGOT): 168 IU (normal range: up to 50 IU), Serum glutamic pyruvic transaminase (SGPT): 195 IU (normal range: up to 50 IU), alkaline phosphatase (ALP): 324IU (normal range: 80–240 IU), serum albumin: 3.89 g% (normal range: 4–5.5 g%)) with mild hyponatraemia. Ultrasound revealed a healthy liver with bilateral enlarged ovaries (maximum diameter 10 cm), ascites with fluid in Morrison’s pouch and mild right-sided pleural effusion. Her viral markers were repeated and found to be negative. Her coagulation profile and renal function tests remained normal throughout the hospital stay.

Treatment

She was administered intravenous fluids and was managed conservatively. Daily charting of vitals, urine output and abdominal girth was done. On day 4 of hospital stay, abdominal distension could be well appreciated. She was administered hydroxyethyl starch and saline. She was also given injection cetrorelix 0.25 mg. Over the next 2 days, a total of 3 L of clear ascitic fluid was drained and simultaneously replaced with albumin infusion.

Outcome and follow-up

She improved symptomatically after the paracentesis. Her haematocrit and liver enzymes also fell tremendously, although not normalised. At the time of discharge, her haematocrit was 36%, and liver function tests were still mildly deranged (total bilirubin: 0.3 mg%, SGOT: 39IU, SGPT: 60IU, ALP: 279IU), and she was discharged well after 8 days of hospitalisation. On her follow-up visit after 2 weeks, her liver function tests got normalised.

She underwent letrozole based frozen–thawed embryo transfer cycle after 4 months. One good quality blastocyst was transferred. Her urine pregnancy test came positive, and early ultrasound showed single live intrauterine gestation of 6 weeks.

Discussion

The manifestation of severe OHSS is the most threatening impediment of ovarian stimulation cycles.⁶ Young age, a history of elevated response to gonadotrophins, previous OHSS, polycystic ovary syndrome are the likely factors to intensify the response to ovarian stimulation.⁷ All these predisposing factors were there in the present case. OHSS is characterised by cystic enlargement of the ovaries and a fluid shift from the intravascular to the third space after increased capillary permeability and ovarian neoangiogenesis. β-hCG and its analogues, oestradiol, prolactin, histamine and prostaglandins have been considered to be associated in OHSS along with the vasoactive substances such as interleukins, tumour necrosis factor-α, endothelin-1 and VEGF secreted by the ovaries which have been implicated in increasing vascular permeability. Enlargement of the ovaries results in abdominal pain, nausea and vomiting. Leakage of fluid from follicles, increased capillary permeability or frank rupture of follicles can all cause ascites. After leakage of fluid through the impaired blood vessels, both within and outside the ovary, there is enormous fluid-shift from the intravascular bed to the third compartment resulting in intravascular hypovolaemia with associated oedema, ascites, hydrothorax and/or hydropericardium.⁸ VEGF by binding to VEGF receptor 2 is considered to be responsible for the raised capillary permeability during the hyperstimulation of the ovarian follicles.⁹ Dopamine agonist impedes the VEGF receptor 2 phosphorylation levels and associated vascular permeability partially without affecting luteal angiogenesis and therefore lowers the incidence of ‘early’ onset of OHSS.⁸ Cabergoline, a dopamine agonist, is recommended to effectively reduce the incidence of moderate OHSS, with no remarkable effect on clinical pregnancy rates and miscarriage rates.¹⁰

The use of GnRH antagonists in PCOS patients is considered to be a relatively safer way of performing ovarian stimulation for IVF.¹ Ovulation trigger by GnRH agonists in GnRH antagonist cycles appears to be a reasonable substitute because of their shorter half-life and luteolytic activity.⁵ However, clinically significant OHSS remains a restraint of multi follicular ovarian stimulation, even with the use of GnRH antagonist protocols.¹¹ Avoiding this iatrogenic complication has been a significant concern of clinicians, specifically in high-responder patients during IVF treatments. Therefore, a segmentation approach involving GnRH agonist trigger merged with freezing of all embryos has been advocated in altogether preventing OHSS. By this way, no exogenous or endogenous hCG exists, and it subsequently results in OHSS-free clinics.^{12 13}

La Marca et al proposed a detailed nomogram, based on patient’s age, day-3 serum FSH and AMH, to guide the starting dose of FSH in IVF cycles. They had included only the first IVF attempt cycles with agonist protocol with the exclusion of PCOS. Their findings indicated that the starting dose of FSH in IVF cycles should be individualised.¹⁴ In patients of PCOS, specifically with high AMH level, the nomogram does not appear adequate.¹⁵ In our patient, the dose was decided based on her response during previous IVF attempt, age and ovarian reserve markers.

There are various underlying causes of early-onset, severe OHSS mentioned in the literature. The deviant sensitivity to hCG is explained by the enhanced sensitivity of such high-responder patients triggered by FSH or LH receptor mutations. Augmented sensitivity to hCG and thyroid-stimulating hormone, jointly with an increase in basal activity was established by two mutants of rFSH (T449I, D567N).¹³ The mRNA expression of genes associated with OHSS and steroidogenesis in the granulosa cells of patients triggered by GnRH agonist had been explored and observed to be subordinate at the time of oocyte retrieval. This can lead to boosted FSH and LH receptor sensitivity of granulosa cells along with the enhanced size and numbers of ovarian follicle in a precise group of patients with subsequent development of severe early-onset OHSS.¹⁶ Fatemi et al backed the GnRH agonist receptor mutation concurrent with a longer duration of FSH and LH rise in cases of OHSS, even with a segmentation approach.⁵

The clinical implication of abnormalities in liver function, as seen in the present case, has not yet been thoroughly investigated. Only a few case reports have been quoted showing the evidence of hepatic disturbance in the most severe form of OHSS.^17–19 However, ambiguousness prevails over the mechanism of hepatic involvement. The proposed mechanisms of elevated liver function tests (LFT) include increased circulating sex steroid concentrations,¹⁷ hepatocellular damage from increased vascular permeability²⁰ and consequences of hepatic ischaemia/reperfusion as a result of circulatory dysfunction in severe OHSS. Liver biopsies performed by Ryley et al ¹⁸ and Sueldo et al ¹⁹ showed ultrastructural changes consistent with increased hepatic enzyme activity. However, as these changes are non-specific and detected in a wide range of circumstances, their clinical significance is undetermined. Association of elevated LFT with moderate OHSS has also been reported,²⁰ thereby demonstrating the importance of evaluation of LFT in even mild and moderate cases.

In the present case, early-onset severe OHSS was observed in a patient who underwent ovarian stimulation with GnRH antagonist protocol with agonist trigger and freeze-all strategy for OHSS prevention. Griesinger et al too witnessed early-onset OHSS with this approach.²¹ Gurbuz et al in their report on three cases stressed the enduring risk of severe OHSS even after segmentation of the IVF cycle.⁶ Fatemi et al reported two cases where segmentation approach resulted in severe OHSS requiring hospitalisation and peritoneal drainage.⁵ It accentuates that the occurrence of OHSS cannot be negated in all patients even with freeze-all strategy. The prevention of OHSS is based on its prediction. Strict vigilance for any signs suggesting the development of OHSS has to be kept in every case. OHSS cannot be abolished entirely by any method. However, its prevention is predominantly preferred over its treatment and can be lifesaving.²²

Rapid regression of the severe early OHSS had been witnessed with reinstating GnRH antagonist in the luteal phase.²³ In the index case, injection cetrorelix was administered, which might also contributed towards the resolution of the symptoms. Therefore, in addition to GnRH agonist trigger with a freeze-all strategy in GnRH antagonist protocol, a potential role exists for readministering GnRH antagonist from the day of oocyte retrieval to eliminate OHSS as we edge nearer towards an OHSS-free clinic concept.²⁴

Patient’s perspective.

I decided to undergo repeat IVF cycle at a tertiary care hospital as I had already experienced this complication in my previous cycle at a private Infertility Centre. Despite having been explained the repeated risk of OHSS, I consented to undergo repeat IVF. I was anxious from the very start of ovarian stimulation. I was told about all the possible measures that would be undertaken for OHSS prevention, including freeze all approach. Despite everything, I developed OHSS, which was a dreading experience for me as I had to hospitalise and undergo a battery of investigations daily with ascitic fluid tapping. I recovered well with conservative management. I underwent a FET cycle after four months and became pregnant.

Learning points.

• The use of in vitro fertilisation cycle segmentation approach is considered to be a safer practice for prevention of ovarian hyperstimulation syndrome (OHSS) in polycystic ovary syndrome patients. However, it does not eliminate the chances.

• Clinicians should always be vigilant to the enduring risk of early-onset OHSS even when treatment is segmented, although this risk is relatively rare.

• Using antagonist in the luteal phase may be a step forward as we tend to head towards an OHSS free era.

• Our case also depicts the association of elevated liver function tests in OHSS; however, its clinical implications need to be sought in future studies.