Recurrent HIV-associated posterior reversible encephalopathy syndrome

PRACTICAL IMPLICATIONS

Consider the diagnosis of PRES in HIV-infected patients even in the absence of traditionally established risk factors.

Posterior reversible encephalopathy syndrome (PRES) typically presents with headache, visual disturbance, cognitive changes, and seizures. The characteristic neuroimaging appearances include white matter edema located predominantly in the posterior cerebral regions.¹ PRES has been commonly associated with severe hypertension and immunosuppressive therapy. However, there have been only a few previous reports of PRES related to HIV/AIDS patients, and all these cases were associated with multiple comorbidities. We present a patient with recurrent episodes of PRES, who had no other detectable risk factors apart from HIV.

Case

A 57-year-old man presented with a 3-day history of bilateral temporal and occipital headache with a gradual onset. He had nausea and photophobia. He noticed intermittent dark spots along with brightly colored, momentary flashes of light occurring across his entire visual field, which he described as “looking through a prism.” He had no other systemic or neurologic symptoms. These symptoms resolved by the second day in hospital. He had experienced at least 3 similar episodes over the past 6 years, each lasting only a few days. He had generalized seizures during a PRES episode in 2012, but had had no recurrence on phenytoin.

He was diagnosed with HIV infection in 1987 and had received several antiretroviral regimens with virologic suppression for most of his illness. He had had no AIDS-defining illness. His current combination antiretroviral therapy (cART) regimen comprised ritonavir-boosted darunavir, lamivudine with abacavir, and dolutegravir.

He was prescribed opiates and sodium valproate for his HIV-associated painful distal sensory neuropathy. Other comorbidities included an iliac artery thrombosis in 2001, treated by embolectomy and stenting followed by warfarin, as well as mild hypertension and dyslipidemia treated with lisinopril and pravastatin.

His temperature was 37°C, blood pressure 180/80 mm Hg, and heart rate 50 beats/min. His neurologic examination including cognitive functions was unremarkable.

He had normal neutrophil and lymphocyte counts. The tests of electrolytes, renal and liver function were normal. His CD4 cell count was 360 cells/mL. He had an HIV viral load of 1,077 copies/mL, and it was noted to have increased to 3,277 copies/mL, 2 months ago from a baseline of <20 copies/mL. Of interest, there had been a small increase in the viral load just a few months prior to 3 out of 4 of his PRES episodes (figure, E).

Figure. MRI and timeline of PRES and HIV viral load flares.

MRI brain in 2016 (A) showed bilateral confluent subcortical hyperintense lesions in the occipital lobes on FLAIR sequences. These abnormalities extended to the gray-white matter interface. MRI brain 6 weeks later (B) showed resolution of the occipital hyperintensities. A similar PRES episode in 2015 was demonstrated on MRI scans 8 weeks apart (C and D). (E) chronicled the timeline of PRES episodes (above the time bar) and major HIV viral load flares (below the time bar) between 2012 and 2018. No neuroimaging was available in 2013 or 2014. FLAIR = fluid-attenuated inversion recovery; PRES = posterior reversible encephalopathy syndrome.

An MRI brain showed bilateral confluent subcortical hyperintense lesions in the occipital lobes on fluid-attenuated inversion recovery sequences (figure, A). Diffusion-weighted imaging showed 2 punctate foci of restricted diffusion in the occipital lobes. There was no contrast enhancement. CSF examination showed 10 ×10^6/L mononuclear cells and protein of 52 mg/dL. PCR for cytomegalovirus, herpes simplex virus, enterovirus, and John Cunningham virus were negative, and cryptococcal antigen was not detected. He made a full recovery with symptomatic treatment without any additional antihypertensives. An MRI brain performed 6 weeks later showed complete resolution of the edema in the occipital lobes (figure, B). Review of his previous MRI scans performed during headache episodes showed typical changes of PRES with subsequent resolution (figure, C and D).

Discussion

In Louis Caplan's seminal article on PRES, severe hypertension (including eclampsia), nephropathy, and the use of immunosuppressive or cytotoxic drugs were recognized as being highly associated with, and potentially the causes of, the condition.² HIV/AIDS has not previously been linked to PRES,¹ although there have been several case reports of HIV-positive patients presenting with PRES.^3–5 A considerable number of these patients also had hypertension and/or end-stage renal disease. Most had low CD4 counts (<200 cells/mL), and some had concurrent disseminated opportunistic infections, all of which meant that the etiology of the PRES could not be clearly established.

Our case was unique in terms of both recurrent nature of his presentation and the relative stability of his CD4 status. Each presentation involved almost identical symptomatology with complete recovery, both clinically and radiologically. It is important that there seemed to be a temporal correlation between the HIV flares and PRES episodes.

Endothelial dysfunction has been identified as a major underlying factor in PRES, and endothelial dysfunction is known to occur in HIV infection.⁶ In addition, protease inhibitors have been associated with endothelial dysfunction.⁷ Our patient was taking protease inhibitors; however, these episodes were not temporally related to any changes to his cART regimen. Therefore, we postulate that the immune dysregulation from HIV and the transient viral escape may be the major risk factor (and potentially the cause) of these PRES episodes.

As HIV-positive patients are living longer in the cART era, clinicians should be aware of HIV-related PRES. The risk factors associated with this entity may differ from the established causes.

Appendix. Authors

Study funding

No targeted funding reported.

Disclosure

The authors report no disclosures relevant to the manuscript. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.