Posterior reversible encephalopathy syndrome, preeclampsia or stroke? A diagnostic dilemma

Abstract

When a pregnant woman presents with headaches, visual disturbances, epigastric pain and nausea, preeclampsia quickly springs to mind. This case describes a primigravid 22-year-old female of 32 weeks gestation who presented with the symptoms described but was found to be apparently normotensive. Due to ongoing symptoms and diagnostic uncertainty in the absence of definitive evidence of preeclampsia, the patient was further investigated with an MRI brain scan, which was reported as either an acute stroke or an atypical presentation of posterior reversible encephalopathy syndrome. Together with blood results that showed heterozygosity for Factor V Leiden, we concluded that while the patient’s clinical diagnosis was certainly preeclampsia, her investigations also supported an unexpected diagnosis of silent brain infarction. This report outlines a diagnostic dilemma that required multidisciplinary working between obstetricians, neurologists, radiologists and stroke physicians to manage the patient who went on to make a full recovery and deliver a healthy baby.

Background

Preeclampsia is one of the most commonly made diagnoses in obstetric triage, and the symptom constellation of headaches, visual disturbances, epigastric discomfort, peripheral oedema and nausea is well described. At least one of either hypertension, proteinuria, brisk reflexes with clonus on neurological examination, deranged blood tests including elevated creatinine, transaminitis, thrombocytopaenia or deranged coagulation usually support the clinical diagnosis, and further investigations are rarely necessary or even helpful. This case report describes a patient presenting to obstetric triage with clinical symptoms highly suggestive of preeclampsia but with seemingly normal blood pressures and an absence of proteinuria, who deteriorated despite supportive management and was therefore further investigated. Investigations suggested that the patient was also suffering from either an acute stroke or posterior reversible encephalopathy syndrome (PRES), a rare acute neurological syndrome first described in 1996.¹ This case is being published as the authors wish to highlight the diagnostic challenges that the team faced in caring for the patient.

Case presentation

A fit and well 22-year-old primigravid South-Asian woman of 32 weeks gestation presented to obstetric triage with a 2-day history of headache, nausea and epigastric discomfort, with a more recent 1-day onset of blurred vision and diplopia at the extremes of lateral gaze, which prompted her to seek medical attention. The pregnancy had been unremarkable until her presentation, apart from a small-for-gestational age baby (<10th centile) on serial scans. General examination was unremarkable and in particular, reflexes were normal and clonus was absent. She was investigated for possible preeclampsia and a serial blood pressure profile (quarterly over a 1-hour period as per department protocol) recorded consistently normal blood pressures as highlighted in table 1, with a stable heart rate of 70–75 beats/min. Notably, the patient’s booking blood pressure (from an estimated 4 weeks gestational age) was 100/69.

Table 1.

Booking (estimated 4 weeks gestation) and admission (32 weeks gestation) blood pressure profile for the patient

	Blood pressure (mm Hg)
Booking	100/69
Admission 1 (0 min)	124/76
Admission 2 (15 min)	128/80
Admission 3 (30 min)	127/78
Admission 4 (45 min)	129/80
Admission 5 (60 min	124/75

Investigations

A standard panel of preeclampsia bloods, which included full blood count, liver function tests, urea and electrolytes and coagulation screen were sent and all returned within normal limits. Significantly, urinalysis did not show proteinuria (or any other pathology), and a formal urinary protein–creatinine ratio also came back as normal. The patient was admitted to the antenatal ward for observation and supportive care but continued to experience worsening symptoms, and hence was further investigated.

Following a surgical review for ongoing epigastric pain, an abdominal ultrasound was arranged and amylase levels were checked, both of which returned normal. A medical review was also sought, and an MRI of the brain was advised to exclude dural venous sinus thrombosis in light of ongoing headaches and visual disturbances. The MRI scan showed five supratentorial and infratentorial T2 and fluid-attenuated inversion recovery hyperintense lesions with restricted diffusion, four in the right frontal lobe and one in the right cerebellar hemisphere (figure 1), reported by the consultant radiologist as suggestive of either an acute stroke or less likely an atypical presentation of PRES.

Figure 1.

Fluid-attenuated inversion recovery MRI axial sequences demonstrating multiple supratentorial and infratentorial hyperintensities in the right cerebral and cerebellar hemispheres.

Differential diagnosis

The obstetricians continued to monitor and treat the patient for likely preeclampsia due to the convincing symptom constellation. Daily preeclampsia bloods, daily urinalysis and 4 hourly blood pressure monitoring were consistently unremarkable, reflecting either the atypical nature of the preeclampsia diagnosis (ie, convincing symptoms without supporting laboratory findings), or simply that blood pressures were being successfully controlled, and hence disease progression was prevented. Simultaneously, discussion between the stroke and neurology teams considered an acute stroke, or PRES, a rare clinicoradiological syndrome that pregnant patients are at increased susceptibility to. However, the MRI imaging pattern was felt to be very atypical for PRES, and this was quickly dismissed as a diagnosis on the advice of the consultant neurologist. The patient underwent a battery of tests arranged by the stroke team to investigate possible causes for early onset stroke including a vasculitic screen, echocardiogram, 24-hour cardiac rhythm monitor, ultrasound carotid Doppler, complement levels, thrombophilia screen, lipid profile and autoantibody screen; significantly, the patient was found to be heterozygous for Factor V Leiden (FVL). This knowledge of the patient’s prothrombotic tendency, taken together with the MRI findings, increasingly pointed towards ischaemic stroke as an additional diagnosis alongside preeclampsia.

Treatment

The patient was immediately started on Aspirin at a dose of 75 mg once a day. Our usual protocol of giving 300 mg of aspirin per day for the first 2 weeks following an acute stroke was not followed, as it was felt to have an unfavourable risk versus benefit profile after multidisciplinary discussion. Specifically, the significant bleeding risk if high-dose aspirin was used and preterm delivery became emergently necessary was felt to be unacceptable. Blood pressure was closely monitored and treated with 200 mg labetalol two times a day to achieve a target systolic pressure of 100–110 to prevent progression of the underlying preeclampsia or even PRES, for which hypertension is a common trigger. The epigastric pain and headaches were controlled with simple analgesia, and steroids were also given to aid fetal lung maturity in anticipation of potential early delivery. The patient started to report a gradual improvement in symptoms after day 5 of the admission. On day 7, however, a pathological cardiotocograph (CTG) with prolonged fetal bradycardia prompted an emergency category 1 caesarean section (EMCS), and a healthy live infant was delivered without complication at 33 weeks gestation. Intraoperatively, appearances suggestive of a small placental abruption were noted as the likely cause for the fetal compromise seen on CTG.

Outcome and follow-up

Both mother and baby had an uneventful recovery on the postnatal ward and in particular the patient’s symptoms, which had begun to resolve prior to the EMCS, had completely resolved by day 3 postnatally. She was followed up by the stroke team after 6 weeks and had remained symptom free.

Discussion

This case describes a 32-week pregnant woman who developed symptoms highly suggestive of preeclampsia without concurrent examination, blood pressure or laboratory findings. Initial diagnostic uncertainty in light of this prompted further investigations, which revealed that the patient had an underlying FVL deficiency, and had developed a stroke. In hindsight, although there was a lack of definitive evidence of preeclampsia, the patient’s eventual recovery after tight blood pressure control between 100 and 110 systolic on the antenatal ward pointed to this being a potential underlying diagnosis after all. Additionally, though our unit defines preeclampsia as blood pressures consistently over 140/90, our patient’s systolic blood pressure peaked at 29 mm Hg above their early pregnancy baseline (table 1). It is feasible that our patient developed preeclampsia at what might be considered ‘normotension’ due to her pre-existing physiological hypotension, and this represents an important learning point from this case.

PRES, also briefly considered as a diagnosis, is predominantly defined as a neurotoxic state secondary to dysfunctional endothelial autoregulation in the posterior circulation, usually due to acute hypertension.² This is thought to cause breakdown of the blood–brain barrier and subsequent vasogenic oedema in the parieto-occipital cortex on MRI, and this in turn causes the typical symptoms of headache, encephalopathy and even seizures.³ The absence of these MRI characteristics in our case meant that the neurologists quickly excluded PRES as a diagnosis. Additionally, although Marrone et al described a lower average systolic blood pressure trigger in pregnant patients with PRES (162 mm Hg), as compared with non-pregnant patients with PRES (185 mm Hg),⁴ our patient’s systolic blood pressure did not exceed 129 at any point, again making PRES an extremely unlikely diagnosis in the presence of MRI findings which were more in keeping with an acute stroke (figure 1).

The risk of stroke in pregnancy and the postpartum period is increased due to all components of Virchow’s triad; a procoagulant state, increased venous stasis and endothelial injury as a result of the mechanical process of delivery.⁵ Our patient was also heterozygous for FVL, the most common type of inherited thrombophilia with heterozygosity occurring in 3%–8% of the European population.⁶ Though FVL most commonly presents as venous thromboembolism (VTE), with the relative risk of VTE increased threefold to eightfold in these patients,^{7 8} it also to a lesser extent increases the risk of arterial thrombosis as seen in stroke, particularly in the presence of other thrombotic risk factors such as smoking,⁹ or in our case pregnancy. Considering our MRI findings, our patient had four lesions in the right frontal cortex and one in the right cerebellar cortex, which would not typically give visual disturbances, these being more consistent with a posterior cerebral artery infarct.¹⁰ It is likely therefore that the clinical presentation is not related to the radiological findings in our case. Instead, our patient’s MRI findings most likely indicate silent brain infarction, which is increasingly being identified incidentally on imaging in patients who have had no clinical history of transient ischaemic attack or stroke, with a prevalence in the general population ranging between 8% and 28%.¹¹ Although the acute event is unnoticed, the consequences of a ‘silent’ infarction should not be underestimated; it is associated with increased incidence of subsequent stroke, as well as cognitive and physical decline,^{12 13} and requires appropriate long-term management.

Heterozygosity for FVL was therefore central to solving the diagnostic uncertainty in this case. Indeed, pregnant carriers of FVL are at increased risk of preeclampsia,¹⁴ placental abruption⁶ and silent brain infarctions,¹⁵ particularly during the hypercoagulable state of pregnancy. In our case, FVL also helps to piece together why the patient demonstrated silent infarcts in multiple vascular distributions on imaging.¹⁵ Considering future pregnancies, the Royal College of Obstetricians and Gynaecologists recommend consideration of prophylactic low-molecular weight heparin (LMWH) in carriers of FVL who have suffered from previous VTE.¹⁶ No guidelines currently exist for the longer term management of FVL carriers who suffer silent brain infarctions during pregnancy, probably because this is seldom diagnosed. In our case, multidisciplinary discussion between stroke physicians and obstetricians about this unique presentation concluded that the patient should continue on 75 mg/day of aspirin, but in future pregnancies should be offered prophylactic LMWH.

In summary, this case describes a pregnant patient presenting with typical preeclampsia symptoms who was further investigated due to initial diagnostic uncertainty in the absence of definitive evidence of preeclampsia and found to also have suffered from silent brain infarction. With hindsight, heterozygosity to FVL was key to both diagnoses and to the authors' knowledge, this is the first case report of a pregnant patient heterozygous for FVL presenting with silent brain infarctions incidentally discovered due to investigation of unrelated symptomology. The novel association of FVL and silent brain infarction in pregnancy, an event which has likely gone unnoticed prior to this case report, poses a management challenge for future pregnancies which we have solved by offering the patient prophylactic LMWH. This case also highlights that while preeclampsia may be diagnosed without typical examination or laboratory findings when blood pressures are elevated above early pregnancy baselines, clinicians should also recognise the importance of considering alternative diagnoses, or even dual pathology, where diagnostic uncertainty exists. In our case, confirming a secondary diagnosis of hereditary thrombophilia and silent brain infarction enabled necessary and timely long-term management.

Learning points.

• Preeclampsia can be considered as a diagnosis when symptoms are highly suggestive and blood pressures are elevated from early pregnancy baselines, even if the patient is apparently normotensive and lacks definitive laboratory evidence of the diagnosis.

• Equally, alternative or even dual pathology can exist and clinicians should maintain a wide differential when diagnostic uncertainty exists.

• Silent brain infarction has a prevalence of up to 28% in some studies and is associated with long-term cognitive and physical health consequences, necessitating timely and long-term management.

• Heterozygosity for Factor V Leiden (FVL) can result in both preeclampsia and silent brain infarction during the prothrombotic state of pregnancy.

• Consideration should be given to prophylactic low-molecular weight heparin in pregnant carriers of FVL who have previously suffered multiple silent brain infarctions in an index pregnancy.