Posterior reversible encephalopathic syndrome is a clinical radiographic syndrome of heterogenous etiologies that are grouped together because of similar findings on neuroimaging studies. Its clinical features include seizures (74%), encephalopathy (28%), headache (26%), and visual disturbance (20%) [1]. Posterior cerebral white matter edema is the classic radiographic feature, but edema can be found elsewhere and may also affect gray matter. This vasogenic edema appears as regions of hypoattenuation on CT, hypointense regions on T1 MRI, and hyperintense regions on T2 MRI. Electroencephalogram (EEG) findings are diverse, with no specific or predominant pattern [2]. Its pathogenesis is not entirely understood, but a combination of cerebral autoregulatory failure and endothelial dysfunction are thought to play important roles. Risk factors for developing PRES include hypertension (61%), autoimmune diseases such as systemic lupus erythematosus (SLE) (45%), cytotoxic agents such as chemotherapy (19%) and sepsis (7%) [1]. Many chemotherapeutic agents have been associated with PRES, including the anti-vascular endothelial growth factor (VEGF) agent, Bevacizumab and the platinum-based agents such as cisplatin and oxaliplatin, which are known to cause neuropathy [3]. To date, PRES has been reported occurring after combination chemotherapy with doxorubicin, and in 2 cases after doxorubicin transarterial chemoembolization (TACE) of hepatic metastases, but there are no publications in the literature describing PRES after systemic monotherapy with doxorubicin or liposomal doxorubicin [4],[5],[6],[7],[8],[9],[10],[11],[12],[13],[14],[15] (Table 1). There have also been no previous reports of PRES occurring in patients with desmoid tumors. It is important to be aware of, as it is a significant complication that can result in serious morbidity and even mortality. As a result, it precluded our patient from receiving further cycles of doxorubicin chemotherapy for her desmoid tumor.

A 33-year-old female presented with a 6-week history of worsening nausea and abdominal pain, resulting in poor oral intake and weight loss.

She had a background of an abdominal desmoid tumor in the context of familial adenomatous polyposis (FAP), for which she underwent a prophylactic colectomy in 2007. The desmoid tumor was diagnosed in 2018 and was deemed to be unresectable due to vascular involvement. She had previous admissions to hospital due to secondary intra-abdominal infections and had grown resistant strains of Escherichia coli and Enterococcus faecium. However, at the time of her most recent outpatient review in 2019, her symptoms were stable on tamoxifen and non-steroidal anti-inflammatories (NSAIDs). Medical history was otherwise significant for migraines, obesity, and cholecystectomy. Of note, she did not have a prior history of hypertension. Her regular medications included regular tamoxifen and amitriptyline in addition to ibuprofen as required. She had allergies or drug intolerances to esomeprazole, meropenem, tetracyclines, vancomycin, and zolmitriptan. She lived at home with her husband and three children, was an ex-smoker having quit 15 years ago and denied alcohol and drug history.

Upon admission she was treated with analgesia, anti-emetics, and intravenous hydration, with complete symptomatic resolution. Computed tomography abdomen showed a 12.4 × 9.8 × 10.6 cm solid mesenteric mass encasing the super mesenteric artery (SMA), which had grown in size compared to previous imaging (Figure 1). Gastroscopy was performed, which identified a partial gastric outlet obstruction and complete obstruction at the 3rd part of the duodenum due to extrinsic tumor compression. Her case was discussed at a multi-disciplinary meeting and the consensus was to offer liposomal doxorubicin 20 mg/m² every two weeks. A percutaneous intravenous central catheter (PICC) was inserted and she received the first dose of chemotherapy on the same day.

On the day after chemotherapy was administered, the patient was reviewed after reporting a self-resolving episode of visual disturbance, namely a “swirly rainbow” pattern in her visual field. No neurological abnormalities were detected and she had a normal blood pressure at the time of symptoms were reported. Hours following this review, a blood pressure of 182/114 mmHg was recorded and she was noted to appear lethargic. On the following day, the patient was reviewed emergently for severe hypertension with a systolic blood pressure in excess of 200 mmHg, associated with acute distress and confusion. While no preceding seizures were witnessed, the clinical impression with the benefit of hindsight was that she may have been in a post-ictal state at the time of this review and that a major seizure had resulted in distress and hypertension as a result. To address her agitation, she was administered clonidine, fentanyl, and tramadol intravenously. This led to symptomatic improvement and reduction in systolic blood pressure to 115 mmHg over the following 30 minutes. Later that day, she was observed to have three generalized tonic-clonic seizures within the space of 2 hours. A venous blood gas taken at this time was significant for a lactate of 6.3, with a normal glucose and no electrolyte abnormalities. She failed to return to baseline neurological state in between episodes, consistent with status epilepticus. A loading dose of intravenous levetiracetam was administered and she was transferred to ICU for intubation and further investigation.

Computed tomography head showed focal areas of hypodensities and vasogenic edema (Figure 2A and Figure 2B). Magnetic resonance imaging of the brain showed patchy white matter T2/FLAIR hyperintensities in bilateral frontal, parietal, occipital lobes in addition to both cerebellar hemispheres, left corona radiata and pons (Figure 3A, Figure 3B, Figure 3C, Figure 3D). These radiographic findings were consistent with PRES. She was extubated and transferred back to the ward after two days in ICU and had no further seizures on regular levetiracetam. Electroencephalogram showed no focal or epileptiform abnormalities. Lumbar puncture was not performed due to clear radiological findings supporting the diagnosis of PRES. Septic screen was significant for a polymicrobial bacteremia, with both methicillin resistant Staphylococcus aureus (MRSA) and Streptococcus anginosus grown in multiple blood culture bottles. It was thought that the MRSA was related to the infection of the PICC and the S. anginosus was of gastrointestinal origin, related to the desmoid tumor. Two transthoracic echocardiograms (TTE) were negative for vegetations. The bacteremia was treated with daptomycin due to the patient’s allergy to vancomycin and she responded well.

The World Health Organisation Uppsala Monitoring Centre (WHO-UMC) causality assessment system was used to support a conclusion that PRES was likely the result of an adverse event to doxorubicin. As a result, the risk of seizure recurrence and exacerbation of PRES was felt to be unacceptably high with liposomal doxorubicin re-challenge. The patient was therefore referred for radiotherapy with no plans for further chemotherapy. She was followed up by Neurology, who successfully weaned levetiracetam. Unfortunately, the patient was readmitted shortly after receiving the first dose of radiotherapy to the desmoid tumor with intractable nausea and pain. Repeat imaging confirmed radiation-induced tumor necrosis. She was treated symptomatically and was later discharged home.

At follow-up three months later the patient reported stable symptoms and was generally well.

In a 1996 case series, a clinical syndrome now referred to as PRES was first described [16]. Since then, its recognition has improved as advanced neuroimaging modalities such as CT and MRI have become more readily accessible. However, it remains a rare diagnosis and its exact incidence is yet to be established. There are no standardized diagnostic criteria or nomenclature, with PRES being the most commonly used name among a handful of others including reversible posterior leukoencephalopathic syndrome (RPLS), reversible posterior cerebral edema syndrome, posterior leukoencephalopathy syndrome, hyperperfusion encephalopathy, and brain leak syndrome. None of these names accurately depicts the spectrum of the syndrome, as it not always posterior nor reversible [1]. Seizures are the most common clinical feature, but headache, encephalopathy, and visual disturbance are also often seen [1]. It is frequently associated with a number of conditions including hypertension, autoimmune disease, sepsis, and the use of immunomodulating agents [1]. The pathogenesis is not fully understood, but the main two theories are cerebral hyper-perfusion and endothelial dysfunction. The hyper-perfusion theory applies in the hypertension setting where elevations in blood pressure exceed the autoregulatory mechanism of cerebral blood flow. This leads to an increase in hydrostatic pressure, causing the blood brain barrier to breakdown and intravascular fluid to extravasate into the surrounding tissue, resulting in edema. Posterior circulation may be particularly susceptible to this given a relative lack of sympathetic supply in the posterior fossa, which could account for the predominance of imaging changes being found in this region [17]. Alternatively, the cytotoxic theory proposes that either endogenous or exogenous chemokines and/or vasoactive peptides can lead to blood brain barrier disruption via endothelial injury and vascular instability. This theory is supported by the relatively increased incidence of PRES among patients taking medications known to cause endothelium toxicity (e.g., chemotherapy) and those with comorbidities known to cause endothelial dysfunction (e.g., malignancy, sepsis, autoimmune disease) [1].

The chemotherapy agents that have more commonly been implicated include vasoactive agents, such as Bevacizumab, and the platinum based agents such as cisplatin and oxaliplatin [3]. Doxorubicin is an anthracycline that exerts its anti-tumor effect primarily via its inhibition of topoisomerase II, causing DNA damage and induction of apoptosis. However, it also results in the generation of free radicals, which could cause cerebral endothelial injury as described in the cytotoxic theory.

There have been 14 cases published to date of PRES occurring after doxorubicin, but in 12 cases it was part of combination therapy and 2 cases were trans-arterial chemoembolization [4],[5],[6],[7],[8],[9],[10],[11],[12],[13],[14],[15] (Table 1). To the author’s knowledge, there have been no published cases of PRES occurring after systemic doxorubicin monotherapy in the literature. However, 195 cases have been notified to WHO’s Vigibase and 262 cases to the FDA’s Adverse Events Reporting System (FAERS) at the time of writing [18],[19]. These databases do not differentiate liposomal from standard formulation, while the Eudravigilance Database makes specific reference to liposomal doxorubicin and lists a single report of epilepsy, but none of PRES [20]. This indicates that it is a phenomenon under reported in the literature.

Currently there are no established guidelines for the treatment of PRES, but it generally involves identification and removal of the cause, carefully managing blood pressure, and anti-seizure medication. Prognosis is generally favorable if identified and treated early, with symptom improvement or resolution within days to weeks [17]. Complications can be severe and include focal neurological deficits from ischemic injury, epilepsy, cerebellar herniation, and death [21].

In conclusion, we present a unique case of PRES occurring shortly after initiation of systemic doxorubicin monotherapy in a patient with an abdominal desmoid tumor. A small number of cases have been notified to pharmacovigilance databases, but there are minimal reports in the literature which may impede clinician awareness. It is important to think of PRES in the context of new seizures and to also be aware of it as a potential complication of chemotherapy. Understanding of the disease process is improving, but gaps in our knowledge remain. Consensus guidelines on diagnosis and treatment of PRES would be useful in addition to further studies examining disease prevalence.