Serous Retinal Detachment in Hypertensive Posterior Reversible Encephalopathy Syndrome

Discussion

We have described 2 patients who developed binocular serous retinal detachments in the setting of accelerated hypertension and PRES. Considering that clinical, pathologic, and experimental evidence indicates choroidal vasoconstriction as the underlying cause of serous retinal detachments, we believe that these cases fortify the concept that vasoconstriction rather than vasodilation underlies the development of PRES.

Ample evidence supports a link between systemic hypertension and vasoconstrictive choroidopathy. Histopathologic studies of postmortem specimens drawn from the acute ischemic phase of hypertension show constriction of choroidal arterioles with obliteration of their lumina.⁶ Patchy filling of the choriocapillaris on fluorescein angiography has been documented in patients with hypertension⁸; and fibrinoid necrosis of choroidal arteries and arterioles with occlusion of the choriocapillaris has been noted on histopathogic studies of human eyes with accelerated hypertension.⁹ Primate models of renovascular hypertension have shown serous retinal detachment and RPE lesions associated with delayed or patchy choroidal filling.⁶

The vasodilation mechanism of PRES in accelerated hypertension is supported by studies showing increased perfusion in the edematous portions of the brain in patients with hypertensive encephalopathy.³ However, emerging evidence suggests that vasoconstriction is a better explanation for the PRES of accelerated hypertension. In patients with acute hypertension, Brubaker et al⁴ found significant posterior brain hypoperfusion, reduced cerebral blood volume, reduced cerebral blood flow, and a heterogeneous pattern of mean transit time on perfusion MR imaging. Other MR perfusion studies^10,11 of PRES have revealed that relative cerebral blood volume is reduced in high-signal-intensity areas of the cerebral cortex and white matter compared with normal-signal-intensity areas.

In a retrospective study of patients with PRES by using CCA and MRA, Bartynski and Boardman¹² observed that 87% of patients showed diffuse vasoconstriction, focal vasculopathy, or vessel pruning of the PCA. Vasospasm on CCA and MRA and watershed hypoperfusion on technetium-Tc99m-hexamethylpropyleneamine oxime single-photon emission tomography have also been demonstrated in women with toxemia of pregnancy, a condition with clinical features similar to PRES and hypertensive encephalopathy.¹³

The development of infarction is further evidence for vasoconstriction as a common mechanism of choroidopathy and PRES in our patients. In our case 1, the visual cortex on 1 side became infarcted. In our case 2, there were residual Elschnig spots in both eyes, reflecting multifocal RPE infarctions. Infarction has been previously reported in PRES,⁴ a phenomenon less likely to occur if vasodilation was the mechanism.

If serous retinal detachment and PRES share the same mechanism, why has serous detachment not been reported before in PRES? We speculate that this retinal abnormality has been overlooked because it may present asymptomatically.