Neuroradiology (2004) 46: 272–276
DOI 10.1007/s00234-003-1053-6

S. O. Casey
A. McKinney
M. Teksam
H. Liu
C. L. Truwit

Received: 5 January 2003
Accepted: 22 April 2003
Published online: 26 March 2004
Ó Springer-Verlag 2004

S. O. Casey (&) Æ A. McKinney
M. Teksam Æ H. Liu Æ C. L. Truwit
Department of Radiology,
University of Minnesota Medical School,
420 Delaware Street SE, Box 292
Minneapolis, MN 55455, USA
E-mail: Casey021@tc.umn.edu
Tel.: +1-612-2736031
Fax: +1-612-6248495

PAEDIATRIC NEURORADIOLOGY

CT perfusion imaging in the management
of posterior reversible encephalopathy

Abstract A 13-year-old girl with a
renal transplant presented with
hypertension and seizures. CT and
MRI demonstrated typical bilateral
parietal, occipital and posterior
frontal cortical and subcortical
edema, thought to represent
posterior reversible encephalopathy
syndrome. The cause was presumed
to be hypertension. Antihypertensive
therapy was started, lowering of the
blood pressure in the range of
110–120 mmHg systolic. However,
stable xenon (Xe) CT perfusion
imaging revealed ischemia within the
left parietal occipital region. The
antihypertensive was adjusted which
increased both the systolic and
diastolic blood pressure by 31 mm
Hg. The patient was re-imaged with
Xe CT and was found to have
resolution of the ischemic changes
within the left parietal occipital
region. In this report, we present a
case in which stable Xe CT was used

Introduction
Posterior reversible encephalopathy has been proposed
to denote a diverse group of conditions with similar
clinical and imaging features, with systemic hypertension
and bilateral posterior cerebral edema [1]. Posterior
reversible encephalopathy is most commonly associated
with hypertensive encephalopathy, eclampsia, and
cyclosporin-A neurotoxicity. Although the lesions on
imaging are typically reversible with treatment, some
patients have complications of infarction, predominantly

to monitor the degree of cerebral
perfusion and guide titration of
antihypertensive therapy. Such brain
perfusion monitoring may have
helped to prevent infarction of our
patient.
Keywords Xenon CT Æ Posterior
reversible encephalopathy Æ Brain
perfusion

within the posterior border zones of the cerebral hemispheres [2, 3]. The pathophysiology of these disorders has
been debated, but most workers believe it is a disorder of
cerebral autoregulation [4, 5, 6, 7, 8]. The underlying
abnormality is thought to be hyperperfusion, leading to
reversible edema, usually without infarction [5, 6, 9,
10, 11]. Most patients are markedly hypertensive on
presentation and therapy is typically directed at lowering
systemic blood pressure. It has been empirical, without
direct monitoring of its effects on cerebral perfusion.
However, in autoregulatory disturbances, overly rapid

273

correction of blood pressure may lead to hypoperfusion
and ischemia of the posterior border zones. We used
perfusion CT with stable Xenon (Xe CT) to monitoring
cerebral perfusion during antihypertensive therapy in a
patient with posterior reversible encephalopathy.

Case report
A 13-year-old girl presented with generalized tonic/clonic seizures.
She patient had chronic renal failure and had undergone a renal
transplant 3 months earlier, but had not complied with antihypertensive medication. She had been treated with cyclosporin-A
but this had been changed to oral steroids 2 weeks prior to presentation. She had a headache 2 days prior to presentation and,
on the day of presentation, had three generalized tonic/clonic
between 3:45 and 4:15 am). In the emergency department, her
blood pressure was 205/134 mm Hg. Laboratory tests were
unrevealing, except for serum creatinine, 7.2 mg/dl. The patient
was postictal, with a decreased level of consciousness, disoriented
(but rousable) and weak, although able to move her limbs.
Antihypertensive therapy was commenced immediately, in the
form of intravenous nitroprusside, initially at 0.5 mcg/kg/h. The
patient underwent CT at 4:30 am (Fig. 1A), which revealed subtle
bilateral subcortical low density in both parietal lobes. This was
interpreted as edema, and in the setting of severe hypertension, a
presumptive diagnosis of posterior reversible encephalopathy
was made.
Over the next 4 h, the nitroprusside infusion was incrementally
increased to 0.75 mcg/kg/h; the blood pressure at 8.30 am was 163/
104 mm Hg. Over this period, the patient developed flaccidity of
the right limbs, with a pronator drift, but still rousable to voice.
The clinicians continued to decrease the blood pressure with a
diastolic goal of 80–90 mm Hg. Over the next 4 h the nitroprusside
infusion was titrated up to 1.75 mcg/kg/h, ter blood pressure at
12.30 pm being 117/67 mm Hg, and consulted with the neuroradiologists about the onset of a right hemiplegia while the patient
was receiving increasing nitroprusside.
At 1.00 pm she underwent Xe perfusion CT on a scanner in
which an independent system for stable Xe gas delivery (27%
Xe/73% O2 mixture) and calculation of cerebral blood flow (CBF).
Perfusion images 1 cm thick were obtained at four levels. The
initial Xe CT, performed at a blood pressure of 119/67 mm Hg,
revealed evidence of ischemia in the left (CBF 12.2 cc/100 g/min)
and oligemic perfusion in the right parietal-occipital borderzone
(Fig. 1B). The blood pressure was then raised by slowing the nitroprusside infusion from 1.75 to 0.81 mcg/kg/hr over 30 min until
it reached 150/96 mm Hg at 1:30 pm.
Xe CT then demonstrated significant improvement in perfusion
in the left parieto-occipital region (27.8 cc/100 g/min) and normal
values in the right (Fig. 1C). The patient underwent MRI at
2.15 pm, confirming edema in the cortex and subcortical white
matter of the parietal, occipital, posterior frontal and posterior
temporal lobes, as typically encountered in posterior reversible
encephalopathy (Fig. 1D). There was also mild edema in both
cerebellar hemispheres. Diffusion-weighted imaging and single
voxel MR proton spectroscopy (of the left parietal-occipital signal
abnormality) revealed no definite abnormality. Dynamic susceptibility perfusion imaging with echoplanar MRI and a rapid bolus of
IV contrast medium revealed minimal cortical reduction of CBF
and cerebral blood volume with a mildly increased mean transit
time in the left parieto-occipital region, confirming that cerebral
perfusion was now only minimally reduced.
The patient’s left side was still flaccidity at 3.00 pm. The
nitroprusside infusion was titrated slowly down to 0.5 mcg/kg/h by
9 pm, when the blood pressure was 158/97 mm Hg. The right
hemiplegia had improved (there was no longer a pronator drift),

and the patient was awake but drowsy, with improved orientation
to self and place. The nitroprusside was given at the same rate
overnight, and at 8 am the next morning, when the blood pressure
was 148/92 mm HG, the patient had a mild residual pronator drift,
and still complained of headache. Over the next 24 h, her blood
pressure was controlled at just above 150 mm Hg systolic and
afterwards slowly reduced. The nitroprusside infusion was discontinued on the afternoon of the second hospital day and IV
hydralazine was commenced. The hemiplegia and headaches
completely resolved over the next 3 days, and the patient was
discharged on day 7. She patient returned for MRI 3 months later,
which showed total resolution of all brain lesions, with no evidence
of infarcts (Fig. 1E).

Discussion
Posterior reversible encephalopathy is typically heralded
by headache and altered mental functions, often with
seizures and visual symptoms. Diagnosis is made based
on symptoms and signs, typical imaging findings and,
usually, identification of a risk factor. Our patient had
several risk factors, including hypertension, seizures,
steroid use and renal insufficiency. In more severe cases,
the typical bilateral subcortical and cortical edema can
spread to the posterior frontal and temporal lobes,
corona radiata, pons, cerebellum and other sites. There
has been disagreement over the name reversible posterior leukoencephalopathy, since the condition involves
both gray and white matter, occasionally exclusively
gray matter [12, 13, 14]. We thus prefer the newer term,
posterior reversible encephalopathy.
The pathophysiology of hypertensive encephalopathy
and posterior reversible encephalopathy has been studied extensively. The favored pathophysiologic theory
suggests autoregulatory disturbance with hyperperfusion, resulting in blood-brain barrier disruption and
reversible edema, usually without infarction [5, 6, 7, 9].
Since the brain attempts to maintain a constant blood
flow by autoregulation, when systemic blood pressure
rises subacutely, the brain’s arterioles constrict to create
compensatory resistance, to avoid hyperperfusion [7, 15,
16]. Animal studies show an upper limit to cerebral
autoregulation, termed a ‘‘breakthrough’’ point, past
which constricted arterioles are forced to dilate in
response to the increasing blood pressure [8]. Past this
breakthrough point, perfusion pressure is sufficient to
overcome the blood-brain barrier, allowing extravasation of fluid (and even macromolecules and red blood
cells) into the brain parenchyma [8, 17]. Pathologic
studies in animals and humans typically show little or no
evidence of infarction, revealing only fibrinoid necrosis
within the walls of the arterioles, interstitial edema and
petechial microhemorrhage [3, 18, 19].
It is not known why the posterior circulation is
preferentially affected. However, perivascular sympathetic nerves protect the brain from marked increases
in intravascular pressure in the setting of extreme

274

Fig 1A–E A 13-year-old girl with headache, seizures and blood
pressure 205/134 mm Hg. A CT reveals possible low density
bilaterally in the parietal-occipital cortex, left more than right.
There is also subtle obscuration of the gray-white matter
distinction (black arrows). These findings suggest posterior reversible encephalopathy due to hypertension. B Stable Xe CT perfusion
examination. Baseline study, following institution of IV nitroprusside antihypertensive therapy, with a lower systemic blood
pressure, reveals bilateral posterior borderzone hypoperfusion,
more on the left. Region of interest (ROI) measurements are
12.2 cc/100 g brain/min from the area of lowest cortical perfusion
on the left (red arrows) and 25.8 cc/100 g/min on the right. C Xe
CT at the same level, 30–45 min later, following slowing of
nitroprusside administration, with blood pressure 150/96 mm Hg,
reveals improved perfusion in the posterior border zones (red
arrows): 27.8 cc/100 g/min on the left and 81.0 cc/100 g/min on the
right. D Fluid-attenuated inversion-recovery (FLAIR) image
immediately after Xe CT, confirms bilateral parietal cortical and
less marked subcortical edema (white arrows), typical of hypertensive and posterior reversible encephalopathy. E FLAIR image
3 months later shows resolution of the vasogenic edema and no
evidence of ischemic brain injury

hypertension [8, 20]. The vertebrobasilar system possesses relatively less sympathetic innervation than the
anterior circulation [21]. In the event of excessive prolonged hypertension (which has been reported to be
triggered by cyclosporine), perfusion increases past the
breakthrough point in the posterior temporal, posterior
parietal, occipital lobes, and posterior cranial fossa
structures [21, 22, 23, 24].
Although typically reversible, posterior reversible
encephalopathy can be complicated by ischemia, with or
without vasospasm, and infarcts, typically in a posterior
borderzone distribution between the middle and
posterior cerebral arteries. This ischemic mechanism is
less well understood, but may be related to an autoregulatory disturbance resulting from aggressive antihypertensive therapy. It has been suggested that, in
the setting of subacute hypertension, the upper limit
of cerebral autoregulation is shifted towards higher

275

pressures, presumably as a protective measure to prevent
excessive blood flow [25, 26, 27]. We postulate that if the
blood pressure is lowered too rapidly during treatment
in the setting of autoregulatory disturbance, it may no
longer maintain the caliber of the constricted arteries,
which may result in vasospasm and paradoxical
hypoperfusion (as we presume to have occurred in our
patient). Ischemia can even occur at relatively normal
blood pressures, because with subacute hypertension,
the lower limit of cerebral autoregulation gets shifted
upwards and takes time to adjust back down to normal
levels. There have been reports of death from too rapid
reduction of severe hypertension, with borderzone
infarcts due to hypoperfusion on autopsy, despite no
clinical evidence of systemic hypotension [2, 28]. In the
absence of treatment, hypoperfusion and ischemia
probably also eventually occur due to severe edema and
severely increased intracranial pressure approaching or
exceeding the brain perfusion pressure.
Perfusion CT with stable Xe is a quantitative measure
of cerebral perfusion [29]. Repeated Xe CT studies, with
interval blood pressure manipulation by adjustment of
IV pressor delivery rates, have been used for investiga-

tion of cerebral vasospasm following subarachnoid
hemorrhage [30]. In our patient, the initial Xe study
demonstrated posterior borderzone hypoperfusion with
left parietal-occipital perfusion values in the ischemic
range. There was relative normalization of perfusion on
the second CT, following upward adjustment of the
blood pressure, but well below the presenting level. This
suggests that the blood pressure may have been below
the lower limits of autoregulation. Alternatively, the
reduced blood pressure may have been inadequate for
perfusion of the edematous parieto-occipital regions.
Although seizure-induced brain edema is in the
differential diagnosis for the brain changes, these have
been reported to give high signal on diffusion-weighted
imaging and to show postictal regional hyperperfusion
[31]. Our patient did not show these characteristics.
Regardless of the exact pathophysiology of the
hypoperfusion, we believe that close monitoring of
cerebral perfusion with Xe CT and subsequent adjustment of therapy based on the findings decreased the risk
of infarcts in the early stages of treatment of the
hypertensive encephalopathy.

References
1. Hinchey J, Chaves C, Appignani B, et al
(1996) A reversible posterior leukoencephalopathy syndrome. N Engl J Med
334: 494–500
2. Haas DC, Anderson GH, Streeten DPH
(1985) Role of angiotensin in lethal
cerebral hypoperfusion during treatment of acute hypertension. Arch Intern
Med 145: 1922–1924
3. Ay H, Buonanno FS, Schaefer PW, et al
(1998) Posterior leukoencephalopathy
without severe hypertension: utility of
diffusion-weighted MRI. Neurology
51: 1369–1376
4. Byrom F (1954) The pathogenesis of
hypertensive encephalopathy and its
relation to the malignant phase of
hypertension. Lancet 2: 201–211
5. Johansson B (1974) Regional cerebral
blood flow in acute experimental
hypertension. Acta Neurol Scand 50:
366–372
6. Haggendal E, Johansson B (1972) On
the pathophysiology of the increased
cerebrovascular permeability in acute
arterial hypertension in cats. Acta
Neurol Scand 48: 265–270

7. Auer L (1978) The sausage-string
phenomenon in acutely induced hypertension—arguments against the
vasospasm theory in the pathogenesis of
acute hypertensive encephalopathy. Eur
Neurol 17: 166–173
8. Mackenzie ET, Strandgaard S, Graham
DI, Jones JV, Harper AM, Farrar JK
(1976) Effects of acutely induced
hypertension in cats on pial arteriolar
caliber, local cerebral flow, and the
blood-brain barrier. Circ Res 39: 33–41
9. Ekstrom-Jodal B, Haggendal E, Linder
LE, Nillson NJ (1971) Cerebral blood
flow autoregulation at high arterial
pressures and different levels of carbon
dioxide tension in dogs. Eur Neurol 6:
6–10
10. Tamaki K, Sadoshima S, Baumbach
GL, Iadecola C, Reis DJ, Heistad DD
(1984) Evidence that disruption of the
blood-brain barrier precedes reduction
in cerebral blood flow in hypertensive
encephalopathy. Hypertension
6: I75–181
11. Johansson B (1974) The blood brain
barrier and cerebral blood flow in acute
hypertension. Acta Neurol Scand 50:
366–372
12. Schwartz RB (1996) A reversible
posterior leukoencephalopathy
syndrome. N Engl J Med 334: 1743

13. Dillon WP, Rowley H (1998) The
reversible posterior cerebral edema
syndrome. AJNR 19: 591
14. Casey SO, Sampaio RC, Michel E,
Truwit CL (2000) Posterior reversible
encephalopathy syndrome: utility of
fluid-attenuated inversion recovery MR
imaging in the detection of cortical and
subcortical lesions. AJNR 21: 1199–206
15. Standgaard S, Paulson O (1984) Cerebral autoregulation. Stroke 15: 413–416
16. Paulson OB, Strandgaard S, Edvinsson
L (1990) Cerebral autoregulation.
Cerebrovasc Brain Metab Rev
2: 161–192
17. Ijima T, Kubota Y, Kuroiwa T,
Sankawa H (1994) Blood-brain barrier
opening following transient reflex
sympathetic hypertension. Acta
Neurochir [Suppl] 60: 142–144
18. Lanzino G, Cloft H, Hemstreet MK,
West K, Alston S, Ishitani M (1997)
Reversible posterior leukoencephalopathy following organ transplantation.
Description of two cases. Clin Neurol
Neurosurg 99: 222–226
19. Pavlakis SG, Frank Y, Kalina P,
Chandra M, Lu D (1997) Occipitalparietal encephalopathy: a new name
for an old syndrome. Pediatr Neurol 16:
145–148

276

20. Beausang-Linder M, Bill A (1981)
Cerebral circulation in acute arterial
hypertension—protective effects of
sympathetic nervous activity. Acta
Physiol Scand 111: 193–199
21. Edvinsson L, Owman C, Sjoberg NO
(1976) Autonomic nerves, mast cells,
and amine receptors in human brain
vessels: histochemical and pharmacologic study. Brain Res 115: 377–393
22. Scherrer U, Vissing SF, Morgan BJ,
et al (1990) Cyclosporin-induced sympathetic activation and hypertension
after heart transplantation. N Engl
J Med 693–699
23. Mark AL (1990) Cyclosporine, sympathetic activity, and hypertension. N
Engl J Med 323: 748–750

24. Schwartz RB, Bravo SM, Klufas RA,
et al (1995) Cyclosporine neurotoxicity
and its relationship to hypertensive
encephalopathy: CT and MR findings
in 16 cases. Am J Roentgenol 165: 627–
631
25. Yanagisawa M, Kurihara H, Kimura
S, et al (1988) A novel potent
vasoconstrictor peptide produced by
vascular endothelial cells. Nature 332:
411–415
26. Lerman A, Hildebrand FL Jr, Margulies KB, et al (1990) Endothelin: a new
cardiovascular regulatory peptide.
Mayo Clin Proc 65: 1441–1445
27. Lee RM, Owens GK, Scott-Burden T,
Head RJ, Mulvany MJ, Schiffrin EL
(1995) Pathophysiology of smooth
muscle in hypertension. Can J Physiol
Pharmacol 73: 574–584
28. Haas DC, Streeten DHP, Kim, RC,
Naalbandian AN, Obeid AI (1983)
Death from cerebral hypoperfusion
during nitroprusside treatment of acute
angiotensin-dependent hypertension.
Am J Med 75: 1071–1076

29. Hughes RL, Yonas H, Gur D, Latchaw
R (1989) Cerebral blood flow determination with the first 8 hours of cerebral
infarction using stable xenon-enhanced
computed tomography. Stroke 20: 754–
760
30. Darby JM, Yonas H, Marks EC,
Durham S, Snyder RW, Nemoto EM
(1994) Acute cerebral blood flow response to dopamine-induced hypertension after subarachnoid hemorrhage.
J Neurosurg 80: 857–864
31. Kassem-Moussa H, Provenzale JM,
Petrella JR, Lewis DV (1994) Early
diffusion-weighted MR imaging
abnormalities in sustained seizure
activity. Am J Roentgenol
174: 1304–1306