2.1 Definition and Pathophysiology

OHSS is a third-spacing phenomenon of intravascular fluids and proteins into the ovaries, peritoneal, pleural and, rarely, pericardial cavities. In normal, unstimulated ovarian cycles there is a natural increase in the volume of peritoneal fluid around the time of ovulation. Vascular Endothelial Growth Factor (VEGF) would appear to be central to this process. This cytokine is produced in response to the luteinising hormone (LH) surge and, in iatrogenic ovarian stimulation, to the human chorionic gonadotrophin (hCG) used to induce the final maturation of the follicle prior to oocyte retrieval [4]. Part of these peri-ovulatory changes is the neo-vascularisation of the granulosa cells in the follicle and part of that process is an increase in capillary permeability. When this process is exaggerated by the presence of a large number of follicles, the effects of the VEGF extend to the peritoneal capillaries resulting in an outpouring of a protein-rich exudate from the vascular compartment. OHSS will not occur in a stimulated cycle if there is no spontaneous LH surge or the hCG trigger is withheld. OHSS also appears to have two peaks of incidence: the first around 7 days after the ovulatory hCG injection and the second around 14 days later and thought to be in response to the hCG of pregnancy. The first group self-limits unless there is a pregnancy but in the second group the condition can persist for weeks.

The loss of protein-rich fluid from the vascular space results in significant haemoconcentration with a resultant rise in haematocrit and blood viscosity. This puts the patient at immediate risk of deep vein thrombosis, pulmonary embolism and, much more rarely, arterial thrombosis. In addition, the reduction in circulating volume results in a decrease in renal perfusion potentially leading to pre-renal failure. This induces an increase in anti-diuretic hormone, which lowers the serum sodium, and an activation of the renin–angiotensin–aldosterone system which results in hyperkalaemia.

The accumulation of the exudate in the peritoneal cavity can lead to considerable ascites and the abdomen gradually becomes increasingly tense, ultimately splinting the diaphragm and resulting in significant breathing difficulties. This is further compounded by a ‘Meig’s Syndrome’–like phenomenon with pleural effusion(s) and further compromise of respiratory efforts. As a result the patient can become markedly short of breath, even hypoxic at rest, ultimately developing Adult Respiratory Distress Syndrome which carries a high mortality rate.

2.2 Prediction and Prevention

In the past prediction relied on clinical history – age, PCOS, past history of OHSS etc. (Box 21.1). More recently it has become clear that patients with an elevated anti-mullerian hormone (AMH) level and/or a high antral follicle count are at markedly increased risk of developing the condition. As a result the stimulation protocol can be adjusted to decrease the risk. Metformin and dopamine agonists such as Cabergoline have also been reported to be helpful in reducing the incidence and severity of OHSS [5, 6]. However, the use of pure gonadotrophin releasing hormone (GnRH) antagonists rather than agonists has made the greatest inroad into reducing the incidence of the condition which is further decreased by using a GnRH agonist trigger instead of hCG. In a Cochrane review of 17 randomised controlled trials, Youssef et al reported that whilst trigger substitution resulted in a decrease in OHSS it also resulted in a decrease in live birth rate and an increase in miscarriage [7]. These poorer pregnancy outcomes were not seen in ovum recipients from such cycles.

Alternative strategies for decreasing the incidence of the condition include coasting – reducing or withdrawing the follicle-stimulating hormone (FSH) stimulation for a few days but with the attendant risk of collapsing estradiol levels and loss of the cycle; reducing the dose of hCG given or withholding it completely and abandoning the cycle; using progestogen/progesterone exclusively for endometrial support in the luteal phase instead of hCG – which ‘rescues’ the copora lutea by stimulation; and freezing all embryos to avoid pregnancy.

2.3 Diagnosis and Management

All patients should be warned to look out for the signs and symptoms of OHSS after oocyte retrieval. Ideally patients should weigh themselves daily: if their weight increases by more than 5kg, they are at high risk of OHSS. Equally, they may present with increasing abdominal bloating and discomfort or with increasing shortness of breath on even minimal exertion. Rarely they present with deep vein (or arterial) thrombosis and/or pulmonary embolism.

On presentation they should be weighed and their abdominal girth measured. Increases in both mark clinical deterioration. Apart from documenting a relevant history, the abdomen, chest and calf muscles must be examined and any shortness of breath noted. An ultrasound of the abdomen will confirm and quantify ascites and ovarian enlargement. A chest X-ray, an ECG and a ventilation perfusion scan may be ordered as appropriate. Blood should be drawn for measurement of serum sodium, potassium and albumin levels and also for evidence of renal or liver dysfunction. Similarly a full blood picture will show evidence of haemoconcentration, particularly with the haematocrit. The white cell count will also be raised and may be mistaken for evidence of an infection. Leucocytosis is part of OHSS; pyrexia is not. A baseline coagulation profile is also recommended at admission. If the case is thought to be in any way serious, a urinary catheter should be sited and fitted with a urinometer. The hourly renal output is key to the management of the condition.

The initial phase of management is rehydration to ensure that there is sufficient fluid in the vascular compartment to perfuse the kidneys and result in urine production. Only normal saline should be used for transfusion. There is no role for potassium containing fluids as the patients are likely to be (or to become) moderately hyperkalaemic. This will be corrected by normal saline infusion. The minimum renal output to ensure renal survival is 30mlhr⁻¹. To achieve this it may be necessary to infuse in excess of 200mlhr⁻¹.

Once rehydration has been achieved and the kidneys are working again, we enter the phase of fluid restriction. Clearly, with the capillaries leaking fluid so avidly, any excess fluid in the vascular compartment will leak into the third space. Therefore, fluids (usually intravenous) should be restricted to a rate sufficient to maintain renal function – i.e. to maintain the renal output between 30–50mlhr⁻¹. This maintenance phase is critical as the patient must not be over-transfused. It can also have quite a prolonged duration, particularly if the patient is pregnant. During this time it may be necessary to carry out paracentesis or a pleural tap as the clinical situation dictates. Eventually, there will be a spontaneous diuresis that heralds the resolution of the condition but not the resolution of the risk of deep venous thrombosis. Once the diuresis is underway, the intravenous fluids and fluid restriction can be stopped.

A key part of OHSS management is the prevention of thrombosis. Clearly, rehydration will help but it is mandatory to ensure that prophylactic doses of low molecular weight heparin are given and that compression stockings are worn. Where the patient is bedbound, intermittent pneumatic compression boots must be worn.

As the peritoneal fluid is an exudate it is rich in protein and patients may become albumin depleted. Albumin infusions may be given to correct this. In our own unit, if levels drop below 30gdl⁻¹ we give 100ml of 20% albumen intravenously and repeat as necessary.

One of the great temptations in the management of OHSS is to administer diuretics; this is particularly true of intensive care specialists when they become involved. Diuretics are contraindicated as the diuresis they induce is from fluid taken from the vascular space. Therefore, in fact they make the condition significantly worse. They are only indicated in the rare circumstance where clearly the haematocrit has dropped and exudation is being spontaneously physiologically reversed but this is not being reflected in the renal output. Here a one-off bolus may be given to nudge the kidneys into activity.

One of the other more unusual and unpleasant complications of OHSS is vulval oedema. In truth there is little that can be done about this. It will resolve spontaneously, with the condition, and in the interim management involves the liberal application of an emollient and possibly some sort of local pressure.