Catheter Angiography, MR Angiography, and MR

ORIGINAL
Perfusion in Posterior Reversible Encephalopathy
RESEARCH Syndrome
W.S. Bartynski BACKGROUND AND PURPOSE: The cause of posterior reversible encephalopathy syndrome (PRES) is
J.F. Boardman unknown. Two primary hypotheses exist: 1) hypertension exceeding auto-regulatory limits leading to
forced hyper-perfusion and 2) vasoconstriction and hypo-perfusion leading to ischemia with resultant
edema. The purpose of this study was to evaluate the catheter angiography (CA), MR angiography
(MRA), and MR perfusion (MRP) features in PRES in order to render further insight into its mechanism
of origin.

MATERIALS AND METHODS: In 47 patients with PRES, 9 CAs and 43 MRAs were evaluated for
evidence of vasculopathy (vasoconstriction and vasodilation), and 15 MRP studies were evaluated for
altered relative cerebral blood volume (rCBV) in PRES lesions and regions. Visualization of vessels on
MRA and toxicity blood pressures were compared with the extent of hemispheric vasogenic edema.

RESULTS: Vasculopathy was present in 8 of 9 patients on CA (direct correlation to MRA in 3/6

patients). At MRA, moderate to severe vessel irregularity consistent with vasoconstriction and vaso-
dilation was present in 30 of 43 patients and vessel pruning or irregularity in 7 patients, with follow-up
MRA demonstrating reversal of vasoconstriction or vasodilation in 9 of 11 patients. Vasogenic edema
was less in patients with hypertension compared with patients who were normotensive. Preserved
normal length of the posterior cerebral artery (PCA) was commonly seen in patients with severe
hypertension despite diffuse or focal vasoconstriction or vasodilation. In these patients, lengthier
visualization of the distal PCA correlated with a lower grade of hemispheric edema (P ⫽ .002). Cortical
rCBV was significantly reduced in 51 of 59 PRES lesions and regions compared with a healthy
reference cortex (average 61% of reference cortex) with mild decrease in the remainder.
CONCLUSION: Vasculopathy was a common finding on CA and MRA in our patients with PRES, and
MRP demonstrated reduced cortical rCBV in PRES lesions. Vasogenic edema was reduced in patients
with hypertension, and superior distal PCA visualization correlated with reduced hemispheric edema in
patients with PRES and severe hypertension.

BRAIN
N eurotoxicity with development of posterior reversible en-
cephalopathy syndrome (PRES) is commonly seen in as-
sociation with cyclosporine and FK-506 immune suppression
observations on CA, MRA, and MRP could render further
insight into the state of brain perfusion in PRES. Therefore,
the purpose of this study was to retrospectively evaluate the

ORIGINAL RESEARCH
after transplantation (allogeneic bone marrow transplanta- CA, MRA, and MRP features in a large group of patients with
tion [allo-BMT], solid organ transplantation); preeclampsia PRES.
and eclampsia; infection, sepsis, and shock; nonspecific med-
ical renal disease; and in autoimmune conditions as well as Materials and Methods
after high-dose chemotherapy.1-13 The mechanism behind the We searched the radiology report data base at our institution (January
development of PRES is yet unproved. Two broad theories 1998 –October 2006) for patients in whom PRES, cyclosporine,
have generally been considered. tacrolimus, and FK-506 neurotoxicity, hypertensive encephalopathy,
Severe hypertension with autoregulatory failure and hy- systemic lupus erythematosus, Wegener granulomatosis, preeclamp-
perperfusion is often cited as the underlying mechanism. Al- sia and eclampsia, or scleroderma were cited on brain MR imaging.
ternatively, vasospasm has been demonstrated (catheter an- MR imaging studies were reviewed in identified patients for features
giography [CA], MR angiography [MRA]), decreased cerebral consistent with PRES. Criteria included complete or partial expres-
blood flow noted (MR perfusion [MRP], single-photon emis- sion of the typical PRES pattern, reversibility on follow-up imaging,
sion CT [SPECT]) and the imaging appearance typically re- or vasogenic edema as demonstrated by MR diffusion-weighted im-
sembles a watershed distribution suggesting a mechanism re- aging (DWI).
lated to brain hypoperfusion.1-3,5,9,13-20 A total of 116 patients were identified with clinical neurotoxicity
Given these opposing views, it was our opinion that parallel and imaging features consistent with PRES (infection, sepsis, and
shock, 31.4%; transplantation/cyclosporine and FK-506, 31.4%; au-
toimmune disease, 11.4%; postchemotherapy, 5.7%; and eclampsia
Received February 2, 2007; accepted after revision August 13.
or delayed eclampsia, 11.4%). The remaining cases included isolated
From the Department of Radiology, Division of Neuroradiology, University of Pittsburgh,
Presbyterian University Hospital, Pittsburgh, Penn. hypertension and undetermined associations.
Please address correspondence to Walter S. Bartynski, MD, Department of Radiology,
MRA or CA, or both, were available in 47 of these 116 patients,
Division of Neuroradiology, University of Pittsburgh, Presbyterian University Hospital, D132, and in 20 of 47, MRP was also obtained. These 47 patients form the
200 Lothrop St, Pittsburgh, PA 15213; e-mail: [email protected] basis of this retrospective evaluation and report. The Institutional
Indicates article with supplemental on-line tables. Review Board approved this retrospective study.
DOI 10.3174/ajnr.A0839 Inpatient and outpatient records on each patient were compre-

AJNR Am J Neuroradiol 29:447–55 兩 Mar 2008 兩 www.ajnr.org 447

hensively reviewed with relevant clinical information identified in- Table 1: PCA pruning scale
cluding underlying cause/association, clinical presentation, imaging
Pruning
time point, and baseline/toxicity blood pressures (BP). Mean arterial
Grade PCA Distal Visualization*
pressures (MAP) were calculated in a standard fashion (MAP ⫽ 2/3
1 Normal length of the distal calcarine artery to the occipital pole
diastolic ⫹ 1/3 systolic pressure). 2 Moderately reduced length of the distal calcarine artery (length
halfway between occipital pole and bifurcation with the
Imaging Techniques parieto-occipital artery)
MR imaging was performed at 1.5T (GE Healthcare, Milwaukee, Wis) 3 Severely reduced length of the calcarine artery (length reduced
including sagittal and axial T1-weighted images (TR, 600 ms; TE, to the level of the bifurcation with the parieto-occipital artery)
min; section thickness, 5 mm; number of acquisitions, 1) fast spin- Note:—PCA indicates posterior cerebral artery.
* Visualized length of the calcarine artery to the tip of the occipital pole.
echo axial proton-density (TR, 2000 –2500 ms; TE, min; section
thickness, 5 mm; number of acquisitions, 1), T2-weighted (TR, 2500 – Distal Branch Visualization (Pruning). We assessed the extent of
3000 ms; TE, 84 –102 ms; section thickness, 5 mm; number of acqui- distal branch visualization on MRA studies by evaluating the distal
sitions, 1). We obtained MRA using 3D time-of-flight (TOF) tech- branches of the PCA. Reduced branch visualization was characterized
nique (TR, default; TE, min; FA, 45°; FOV, 18 –22 cm; Matrix, 226 ⫻ by PCA pruning and appeared as a distinct and separate observation
224; number of acquisitions, 1) with multiple overlapping slab from PCA spasm. PCA pruning was recognized by reduced PCA
reconstruction. length (independent of vessel irregularity). Grade of PCA visualiza-
We performed MRP using dynamic susceptibility contrast MR tion was based on the length of calcarine artery visualization. PCA
imaging with gradient-echo echo-planar sequence during dynamic pruning was graded in severity on a 3-point scale and reviewed in
bolus contrast administration (TR, 2040 ms; TE, 60 ms; flip angle, 60°; Table 1.
matrix, 96 ⫻ 128; number of acquisitions, 1). Standard dose of 0.1 MCA pruning was also noted if typically identified MCA branches
mmol/kg of gadolinium dimeglumine (Magnevist; Bayer HealthCare demonstrated tapering and reduced visualization. The distal MCA
Pharmaceuticals, Wayne, NJ) or gadopentetate (ProHance; Bracco branches were not consistently included on the 3D TOF sequence
Diagnostics, Princeton, NJ) bolus injection (antecubital; 5 mL/s) was because of choice of placement of the upper margin of the slab. For
begun 10 seconds after scan initiation with a total of 40 data collection this reason, observations of MCA pruning were made if available, but
points (phases) acquired over 12 section locations. the MCA was not used for grading.
We obtained contrast-enhanced T1-weighted images after the Vasculopathy. The 2 primary features of vasculopathy were tab-
MRP sequence or independently with a standard dose (0.1 mmol/kg) ulated separately: 1) diffuse vessel constriction or narrowing, 2) focal
injection using typical T1-weighted parameters, as described above. vessel irregularity of first-, second-, and third-order branch vessels
DWI (single-shot echo-planar, 10,000 ms/min/5 mm/128 [TR/TE/ (focal vasoconstriction, focal vasodilation, and beaded or string-of-
section/matrix]) sequence was available in most patients. In 9 pa- bead appearance).21,22 Diffuse vasoconstriction (narrowing or con-
tients, CA was performed in standard fashion with common carotid/ striction) was considered present if significant diffuse vessel caliber
vertebral injection and digital subtraction imaging acquisition. reduction was observed in 2 or more major branch groups (ACA,
MCA, and PCA). We judged focal vessel irregularity using the tradi-
Imaging Assessment-Edema Grading tional features of vasospasm and arteritis and graded on a 4-point
PRES locations were tabulated (occipital, parietal, frontal, temporo- scale (0, normal; 1, possibly abnormal [mild vessel irregularity]; 2,
occipital, and cerebellar). Atypical locations (brain stem, thalamus, moderate vessel irregularity; and 3, severe vessel irregularity).
and basal ganglia) were also noted. Judgment of vessel abnormality (pruning, vasoconstriction, vaso-
MR imaging was independently graded by 2 neuroradiologists for dilation, string-of-bead appearance) was referenced relative to typical
the extent and severity of hemispheric cortex white matter edema benchmark normal MRA features obtained at our institution. Objec-
(grade summary: 1, limited cortex white matter edema; 2, white mat- tive reference criteria were used to confidently grade vessel irregular-
ter cortex edema with some deep white matter extension; 3, white ity as vasculopathy: 1) a distinct change in artery caliber/morphology
matter cortex edema with limited ventricle surface extension; 4, white between 2 MRA studies, 2) correlation with catheter angiogram, 3)
matter cortex edema, diffuse, widely confluent, extensive ventricle significant difference in vessel caliber/morphology relative to an in-
contact; 5, severe white matter cortex edema, diffuse confluence, ven- ternal patient reference, 4) age-inappropriate vessel caliber or irregu-
tricle deformity) as previously described.13 Graders were blinded to larity in a young patient, and 5) obvious intrinsically abnormal ACA,
the patients’ blood pressures and MRA findings. Grade differences MCA, and PCA branches with focal spasm and pruning or diffuse
were resolved by consensus. vasoconstriction.
MR Perfusion. We performed postprocessing using the Func-
Vessel and Perfusion Assessment Tools (GE Healthcare) workstation. Relative cerebral blood volume
Vascular studies (47 patients; MRA, 43; CA, 9) were independently (rCBV) was obtained from quantification of the area under the neg-
and blindly assessed by 2 neuroradiologists with differences resolved ative enhancement integral generated. Negative enhancement inte-
by consensus. Studies were graded for 2 distinct features: 1) extent of gral was obtained in a standard fashion computing the area under the
distal second- and third-order branch visualization on MRA imaging negative enhancement curve generated during bolus contrast arrival,
studies (best assessed in the PCA branches) and 2) presence or ab- with brain enhancement beginning at initiation of loss of signal in-
sence of vasculopathy (vasoconstriction, vasodilation and/or string- tensity and continuing to the point of re-establishment of the post-
of-bead appearance) at CA as traditionally recognized in vasospasm enhancement baseline signal intensity. Regional rCBV region-of-in-
or arteritis as well as the features consistent with vasculopathy of the terest measurements were obtained in a healthy-appearing cortex and
branches of the anterior cerebral artery (ACA), middle cerebral artery white matter as well as areas affected with PRES as identified on MR
(MCA), and posterior cerebral artery (PCA) on MRA. imaging (FLAIR sequence and T2*).

448 Bartynski 兩 AJNR 29 兩 Mar 2008 兩 www.ajnr.org

 Table 2: Presentation blood pressure and clinical associations
patient), and unstable hypertension or previous renal cell car-
(number of PRES patients) cinoma (1 patient).
Presentation at toxicity included headache, change in vi-
Blood Pressure ISS Tx E dE AI Chemo Unk
sion, and change in mental status alone (19 patients) or fol-
Normotensive 8 2 – 1 – – –
Moderate hypertension 1 2 1 1 – – –
lowed by seizure (21 patients), with seizure only in 7 patients.
Severe hypertension 8 5 7 1 5 2 3 In 11 patients, toxicity BP was either normal (baseline) or
Note:—PRES indicates posterior reversible encephalopathy syndrome; ISS, infection,
demonstrated minor increase in systolic pressure (average
sepsis, and shock; Tx, transplantation; E, eclampsia; dE, delayed eclampsia; AI, autoim- MAP, 94 mm Hg; range, 86 –104 mm Hg). Other than minor
mune disease; Chemo, post-cancer chemotherapy; Unk, nonspecific association.
fluctuations in BP as is commonly seen in patients in the in-
tensive care unit, no significant episodes of hypertension were
Establishing a baseline healthy cortex rCBV was challenging be- noted in the pretoxicity or peritoxicity period in these patients.
cause of 1) altered brain blood flow in a normal-appearing brain Toxicity BP was moderately elevated in 5 patients (average
(possibly because of intrinsic vasculopathy) and 2) potential volume MAP, 114 mm Hg; range, 110 –115 mm Hg, typically with
averaging with cortical vessels. Computed rCBV was often mildly moderate diastolic pressure elevation) and severely elevated in
reduced in normal-appearing regions of the brain, in particular, the 31 patients (average MAP, 133 mm Hg; range, 119 –177 mm
typical watershed and regions immediately adjacent to the PRES Hg; significant systolic pressure elevation, diastolic pressure
lesions. elevation, or both). BP in these patients was typically unstable
Therefore, regions were chosen as representative of normal that immediately before and after toxicity and was often difficult to
had the consistently greatest rCBV in a healthy-appearing brain (typ- control after development of neurotoxicity and PRES.
ically medial frontal and parietal cortex, frontal operculum, and lat- Vasogenic edema was identified in typical PRES locations
eral superior temporo-occipital junction). Regions of interest were (occipital, 44 patients; parietal, 45 patients; frontal, 36 pa-
chosen to avoid visibly linear high flow from surface blood vessels that tients; temporooccipital junction, 22 patients; and cerebellum,
would inappropriately increase the rCBV measurements in a healthy 19 patients) and atypical locations (brain stem [pons, 5 pa-
cortex. tients; medulla, 2 patients; midbrain, 3 patients]); basal gan-
Normal rCBV was established averaging 10 to 20 region-of-inter- glia, 4 patients; and thalamus, 6 patients). Of 32 patients, DWI
est maximum measurements (2– 4 mL voxel) with a healthy-appear- results were normal or demonstrated T2 shinethrough in 16.
ing cortex. The regions of interest were chosen to avoid visibly linear In 7 patients, DWI results were normal except for a single
high flow from surface blood vessels that would inappropriately in- focus of restricted diffusion (infarction, 7 patients; hemor-
crease the rCBV measurements. rhage, 3 patients). Minimal stippled cortical enhancement was
In the PRES lesions, 2 to 8 cortex measurements were obtained noted in 3 patients. Early follow-up imaging results demon-
over moderate-sized focal lesions or over representative confluent strated improvement in vasogenic edema in 12 patients with
regions, with careful attention not to alter measurements by extend- complete reversal of the PRES pattern in 19 patients. Fol-
ing into adjacent white matter or including dominant surface vessels. low-up imaging was not obtained in 16 patients, but DWI
All lesion and healthy-appearing brain measurements were agreed on results in these patients were normal or demonstrated T2
by consensus. Average lesion rCBV was referenced to average healthy shinethrough.
brain rCBV obtained for that patient.
Vasogenic Edema and MAP
Statistics Greater hemispheric edema was present in patients who were
We evaluated the statistical significance of edema grade relative to normotensive (grade average, 2.45) with reduced edema in
blood pressure at toxicity using the Student t test and Kruskal-Wallis patients with moderate to severe hypertension (grade aver-
extension of the Wilcoxon test (PROCAPABILITY, SAS software ages, 1.6 and 1.74, respectively). The difference between pa-
package release 8.2; SAS Institute, Cary, NC). Observation timing tients who were normotensive and those with severe hyperten-
effect on edema and PCA grades were tested by Kruskal-Wallis exten- sion was statistically significant (P ⫽ .045, Student t test).
sion of the Wilcoxon test and analysis of variance. Interobserver dif- Patients with eclampsia and autoimmune disease primarily
ferences for edema and PCA grades were tested by percentage agree- demonstrated severe hypertension and a low grade of vaso-
ment. The statistical significance of edema grade relative to PCA genic edema, whereas patients with infection, sepsis, and
pruning and visualization grades was evaluated with the linear-by- shock were commonly normotensive with more severe brain
linear association test for row and column independence, in which edema (Table 2 and On-line Table 1).
rows and columns have natural ordering.23 No statistically significant difference was detected in edema
grade relative to MR imaging timing from recognized neuro-
Results toxicity. Interobserver agreement of edema grade was 60.4%
The clinical characteristics of the 47 patients with PRES on CA, before consensus resolution.
MRA, and MRP are reviewed in Table 2 and On-line Table 1.
There were 8 patients who were men and 39 who were women, Angiographic Features (CA and MRA)
with an average age of 44.6 years (range, 19 –79 years) with In 46 of 47 patients, intracranial vessels could be assessed by
clinical associations as tabulated (eclampsia or delayed CA (9 patients) or MRA (43 patients). In 1 patient with severe
eclampsia,11; transplantation, 9 [cyclosporine and FK-506]; hypertension, MRA could not be assessed because of signifi-
autoimmune disease, 5; postchemotherapy, 2; and infection, cant motion artifact. The MRA and CA features in patients
sepsis, and shock, 17). PRES was associated with drug use for with PRES are summarized in Table 3 and On-line Table 1,
hypertension (1 patient), recently developed hypertension (1 and in Figs 1–5.

AJNR Am J Neuroradiol 29:447–55 兩 Mar 2008 兩 www.ajnr.org 449

 Table 3: Summary of MRA features and toxicity blood pressure
Overall MRA and CA Observations
At MRA, diffuse vasoconstriction and focal vasculopathy were
Overall Focal Vessel
more frequently noted in patients with moderate to severe
Features Features
hypertension with vessel pruning, mild irregularity, or healthy
# # Ir FS
appearance noted in patients who were normotensive (Table 3
Blood Pressure pts nl P DS vessels nl (Ir/P) (FS/P)
and On-line Table 1 and Figs 1B, 2C, 3B, 3E, 4B, and 5B). Focal
Normotensive 10 7 3 – 30 16 12 (6) 2
Moderately hypertensive 5 2 – 3 15 4 4 (2) 7
vasoconstriction (grade 2–3) was present at MRA in 30 of 43
Severely hypertensive 28 8 2 18 84 13 22 (9) 49 (5) patients and varied in each of the major branches (ACA [16
Note:—MRA indicates MR angiography; # pts indicates number of patients; nl, normal; P,
patients], MCA [18 patients], and PCA [23 patients]).
pruned; DS, diffuse vasoconstriction; # vessels, number of vessels; Ir, irregularity grade 1; Overall, in 40 (87%) of 46 patients, MRA and CA results
Ir/P, irregular and pruned; FS, focal vasoconstriction/vasodilation grade 2–3; FS/P, focal
vasoconstriction/vasodilation and pruned. demonstrated vessel abnormality consistent with diffuse vaso-
constriction, focal vasculopathy (grade 2–3), or vessel prun-
ing. First-order branch spasm was identified in 5 patients with
Conventional Angiography in PRES possible mild first-order spasm in an additional 6 patients.
Vasculopathy was identified in 8 of 9 patients with CA dem- Second- and third-order focal branch spasm was identified by
onstrating focal vasoconstriction, focal vasodilation, a string- CA and MRA in 33 of 46 patients.
of-bead appearance in at least 1 major arterial territory, or all 3
of these features. Involvement was most commonly in the sec- MRA and Vasogenic Edema Comparison in Patients Who
ond- and third-order branches but was also identified in Were Severely Hypertensive
fourth-order vessels and was frequently associated with re- Extent of PCA pruning relative to the extent of hemispheric
gions of PRES vasogenic edema. Delay or reduced capillary edema was compared in the patients who were severely hyper-
blush was noted in all 9 patients (typically interhemispheric tensive and are summarized in Table 4. Vasogenic edema was
region, parietal-occipital lobes, or in areas of PRES vasogenic less in patients who demonstrated lengthy visualization of the
edema). PCA and calcarine artery (Fig 5) with more severe edema in
MRA was available in 6 of 9 patients with CA. In 3 of 6 those patients with moderate to severe PCA pruning (Fig 4).
patients, severe vessel irregularity consistent with vasculopa- Therefore, hemispheric edema trended downward with
thy (grade 2–3) was present on MRA and corresponded to the greater visualized length of the PCA and calcarine arteries,
CA findings (Figs 1–3). In one of these patients (Fig 3), fol- which was statistically significant (P ⫽ .002). The observation
low-up MRA demonstrated reversal of diffuse and focal vaso- remained statistically significant when only the 25 patients
constriction and vasodilation with worsening noted in the sec- imaged within 1 day of toxicity (P ⫽ .003) were considered.
ond patient. No statistically significant difference was detected in the PCA
In the other 3 patients, initial MRA demonstrated moder- grade relative to the timing of the MR image from recognized
ate to severe vessel irregularity and diffuse vasoconstriction, neurotoxicity. Interobserver agreement of the PCA grade was
with follow-up CA demonstrating partial reversal with subtle 86% before consensus resolution.
residual vasculopathy in 2 and delayed or reduced interhemi-
MRP in PRES
spheric capillary blush in 1.
In 15 of 20 patients, the MRP studies were of good quality, and
reliable measurements of rCBV could be made with 5 studies
MRA in PRES
not evaluated because of an irregular bolus or motion artifact.
MRA was obtained in the immediate toxicity period (0 –3
There were 70 abnormal-appearing lesions or regions present
days) in 40 patients and at 4, 8, and 10 days after development
in the 15 patients, with reliable rCBV measurements made in
of PRES in 3. Follow-up MRA was available in 11 of 43
59 of these 70 locations (On-line Table 2 and Fig 6).
patients.
In 51 (86%) of 59 PRES lesions and regions, rCBV was
markedly diminished (ⱕ80%; average, 61% ⫾ 12%; range,
MRA with Follow-Up 31%– 80%) compared with the healthy reference cortex. In 8
In 7 of 11 patients, follow-up MRA demonstrated reversal of dif- studied lesions, rCBV was between 84% and 99% of the
fuse vasoconstriction with a distinct, and often marked, increase healthy cortex rCBV (single lesion in 5 patients, 3 lesions in 1
in vessel size and diameter along with improved visualization of patient studied 3 days after toxicity). Overall average rCBV in
the distal branch (On-line Table 1). Improved distal PCA visual- all measured PRES lesions and regions relative to a reference
ization only was noted in 2 patients, and in 2 patients (each normal cortex rCBV was 65.2% ⫾ 14.8% (range, 31%–99%).
studied again at 4 days), follow-up MRA demonstrated wors-
ening vessel irregularity and reduced branch visualization. Discussion
In patients demonstrating improvement, complete or The cause of PRES is not established. Classically, neurotoxicity
near-complete normalization of vessel irregularity and vascu- with PRES is associated with cyclosporine and FK-506 (trans-
lopathy was identified in 5 (follow-up imaging at 3, 11, 30, 90, plantation), preeclampsia and eclampsia, autoimmune dis-
and 162 days; Fig 3), improvement but still-obvious vessel ease, and postchemotherapy.1-12,24
irregularity in 3 (follow-up imaging at 2, 3, and 4 days), and The mechanism of PRES remains controversial. Severe hy-
significantly improved distal PCA visualization with im- pertension with autoregulatory failure and forced hyperperfu-
proved, but persistent, vessel irregularity in 1 patient with sion remains widely accepted. Alternatively, vasospasm is re-
sickle cell disease followed up at 33 days. ported (MRA or CA), and reduced perfusion has been

450 Bartynski 兩 AJNR 29 兩 Mar 2008 兩 www.ajnr.org

Fig 1. Patient 43 is a 53-year-old woman with a baseline blood pressure of 131/74 mm Hg who was receiving oral and skin patch narcotic pain control for a recently sustained pelvic
fracture in a motor vehicle crash. She developed nausea, vomiting, and altered mentation then progressed to generalized seizure with a blood pressure at toxicity of 149/105 mm Hg. A,
Axial MR imaging (FLAIR sequence) obtained 1 day after acute toxicity demonstrates vasogenic edema in the occipital region (open arrows) consistent with PRES. Similar areas of vasogenic
edema consistent with PRES was present in the frontal and parietal regions (not shown). B, MR angiogram obtained on the same day demonstrates areas of focal vasodilation in the
branches of MCA bilaterally with focal vasoconstriction also noted on the left (arrows). C, Frontal view from the left common carotid catheter angiogram obtained the same day demonstrates
an identical area of focal vasodilation and vasoconstriction in the branch of the left MCA (arrow). An identical finding was present at angiography in the right MCA consistent with the
MRA observation on the right. Similar findings were present on both MRA and CA in the PCA bilaterally (not shown).

Fig 2. Patient 10 is a 27-year-old man with a history of drug and alcohol abuse with a necrotic pneumonia and empyema
growing Pseudomonas and Klebsiella. He developed altered mentation and an asymmetric neurologic examination with blood
pressures ranging between 130/77 and 130/100 mm Hg. A, Axial MR imaging (FLAIR sequence) demonstrates extensive
vasogenic edema in the occipital lobes (open arrows) and temporooccipital junction (arrowheads) consistent with PRES.
Additional extensive areas of vasogenic edema consistent with PRES judged vasogenic edema grade 4 were present in the
frontal and parietal lobes (not shown). B, Vertebral catheter angiogram obtained after development of occipital hemorrhage
(5 days after initial toxicity) demonstrates areas of vessel narrowing and dilation along with a string-of-bead appearance and areas of focal spasm in parietal and occipital branches of
the PCA. Right common carotid catheter angiogram (not shown) also demonstrated extensive vasospasm in branches of the ACA and MCA. C, MR angiogram obtained 5 days after catheter
angiogram (10 days after initial toxicity) demonstrates persistent vessel irregularity in the posterior cerebral branches (arrows) with a pattern nearly identical to the appearance of the
catheter angiogram. Follow-up MR imaging demonstrated improvement of the extensive vasogenic edema.

demonstrated (MRP, technetium Tc99m hexamethylpropyl- MAP, 140 –160 mm Hg in patients who are normoten-
eneamine oxime single-photon emission CT with a watershed sive).25-29 This seems to be regulated by the endothelium
lesion distribution suggesting hypoperfusion.1-3,5,9,13-20 through effects of elevated transmural pressure and release of
The autoregulatory response is intended to maintain stable thromboxane A2 and endothelin.26 The upper limit is in-
cerebral blood flow in the face of blood pressure fluctua- creased in the setting of essential hypertension and sympa-
tions.25,26 With acute severe hypertension, precapillary arte- thetic stimulation as may exist in patients with cyclosporine
riolar vasoconstriction occurs in small (30 –300 ␮) resistance toxicity.25,27 Above this upper-threshold MAP, passive vaso-
vessels that limit cerebral blood flow (upper-limit-of-response dilation ensues with subsequent increase in CBF and endothe-

AJNR Am J Neuroradiol 29:447–55 兩 Mar 2008 兩 www.ajnr.org 451

 Fig 3. Patient 16 is a 39-year-old woman status post bowel resection and appendectomy for Crohn disease being maintained
on antibiotics and steroids with baseline blood pressure of 100/60 mm Hg. Brain imaging was obtained when she developed
headache, vertigo, and moderate elevation in blood pressure (145/95 mm Hg). A, Axial MR imaging (FLAIR sequence) results
demonstrate vasogenic edema in occipital lobes (open arrows) consistent with PRES. Additional areas of vasogenic edema
consistent with PRES were present in the frontal and parietal regions, and abnormality of signal intensity with restricted
diffusion was also present in the medial right cerebellar hemisphere (not shown). B, MR angiogram obtained 1 day after
toxicity demonstrates marked vessel irregularity in the PCAs bilaterally (arrows), greater on the right than on the left. C, Left
vertebral catheter angiogram (selected posterior magnified lateral view) demonstrates a beaded appearance (arrow) and focal
vasoconstriction and vasodilation (arrowhead) of right third- and fourth-order branches of the PCA along with a beaded
appearance of the branches of the distal medial posterior inferior cerebellar artery on the left (open arrow). D Results of
follow-up MR angiogram obtained 3 days after initial study demonstrate normalization of the vessel irregularity and marked
improvement in vessel caliber (arrows) consistent with reversal of the vasospasm/arteritis confirmed on conventional
angiography. Follow-up MR imaging demonstrated reversal of the PRES pattern.

Fig 4. Patient 34 is a 56-year-old woman with long-standing

sickle cell disease and a baseline blood pressure of 140/60
mm Hg in a sickle cell crisis with new-onset pneumonia. She
developed altered mental status and a seizure with blood
pressure at toxicity of 170/110 mm Hg. A, Axial MR imaging
(FLAIR sequence) demonstrates typical vasogenic edema in
the occipital pole regions bilaterally (open arrows) consistent
with PRES and judged edema, grade 3. Occipital lobe rCBV
relative to a healthy reference cortex was 63% on the right
and 54% on the left. B, Collapsed view of the 3D TOF MRA
sequence demonstrates both focal vasospasm (arrows) and
pruning of the PCAs bilaterally. Foreshortening of the PCAs
and only partial visualization to the midportion of the calcar-
ine arteries are noted bilaterally (arrowheads) and were
judged PCA pruning, grade 2. Follow-up MR imaging and
MRA (not shown) obtained 1 month after toxicity demon-
strated complete resolution of the vasogenic edema in the
occipital poles with marked improvement in distal PCA flow
bilaterally along with reversal of the pruning and vasospasm.

lial fluid transudation.26,27 Vasoconstriction and vasodilation range or is only minimally elevated in 20% to 30% of patients
have been noted in animal studies of acute severe hyperten- with PRES.5
sion, but these changes typically require MAP to exceed 180 to Our results at catheter angiography confirm a high incidence
200 mm Hg.28,29 Although this theory remains popular, tox- of vascular abnormality in PRES (focal vasoconstriction, focal
icity MAP is identified at this critical level (MAP ⬎ 140 –160 vasodilation, and string-of-bead appearance) and demonstrate
mm Hg) in only a small subset of patients who develop PRES. that these observations are reflected in the pattern seen at MRA.
In addition, blood pressure at toxicity is within the normal Results of follow-up MRA demonstrated reversibility of this vas-

452 Bartynski 兩 AJNR 29 兩 Mar 2008 兩 www.ajnr.org

 Fig 5. Patient 33 is a 23-year-old pregnant woman with
baseline blood pressure of 125/75 mm Hg who developed
eclampsia. Blood pressure at toxicity was 200/103 mm Hg. A,
Axial MR imaging (FLAIR sequence) demonstrates vasogenic
edema consistent with PRES in the parietal lobes bilaterally
(curved arrows) and was judged edema, grade 1. Vasogenic
edema consistent with PRES was also present in the frontal
lobes, occipital region, and cerebellum (not shown). B, Col-
lapsed view of the 3D TOF MRA sequence demonstrates PCA
spasms bilaterally without PCA foreshortening (arrows) and
was judged PCA pruning, grade 1. The ACA spasm present is
also partially visible on this collapsed view (arrowheads).

Table 4: Correlation of MRA pruning grade (PCA-calcarine artery)

that is considered responsible for the systemic effects that lead
vs vasogenic edema: 28 patients with severe hypertension to organ hypoperfusion in these conditions (endothelial acti-
(MAP-136) vation, white cell trafficking, cerebral vasoconstriction). In the
PCA/Calcarine Artery Pruning Grade setting of significant hypoxemia, endothelial cells become ac-
tivated and vascular endothelial growth factor mRNA (vascu-
Vasogenic Edema Grade 1 Grade 2 Grade 3
lar endothelial growth factor (VGEF), previously known as
Grade 4 0 1 1
Grade 3 3 1 0
vascular permeability factor) can upregulate with increased
Grade 2 6 4 0 tissue levels of VGEF within 6 to 24 hours.40 Sustained hypo-
Grade 1 12 0 0 perfusion and hypoxemia could result in expression of VEGF
Note:—MRA indicates magnetic resonance angiography; PCA, posterior cerebral artery; leading to endothelial leakage and vasogenic edema in PRES.
MAP, mean arterial pressure. The features of vasoconstriction and vasodilation were
identified in medium and small arteries (⬎300 ␮) even in the
culopathy with normalization of diffuse vasoconstriction, rever- absence of significant hypertension. Vessel imaging appear-
sal of focal vessel irregularity, and reduced vessel pruning consis- ance at CA and MRA might be reflecting the combined effects
tent with previous reports.13,14 In addition, MRP results of both increased vessel tone (diffuse and focal vasoconstric-
demonstrated significantly reduced rCBV in most regions of tion) and decreased tone (focal vasodilation) as can occur with
PRES imaging abnormality compared with a reference healthy induced endothelial injury or dysfunction.26 PRES also oc-
cortex. The observations in our patients are consistent with the curred in patients with autoimmune diseases (ie, systemic lu-
vasculopathy and hypoperfusion in PRES. pus erythematosus), conditions with a known tendency to de-
In most of our patients, PRES developed after transplanta- velop what is traditionally labeled arteritis.
tion (cyclosporine and FK-506), infection, sepsis, shock, and Although only a small portion of our patients with PRES
eclampsia, all associated with systemic inflammation, endo- were studied with MRP, measurements of rCBV parallel re-
thelial activation and injury (upregulation of E-selectin, P- sults of other studies. In 51 (86%) measured lesions, rCBV was
selectin, and intercellular adhesion molecule), and significant significantly reduced (average, 61%) compared with a healthy
release of cytokines (tumor necrosis factor-alpha [TNF-␣] cortex. This is consistent with a case reported by Engelter et
and interleukin-1 [IL-1]).30-35 Endothelial activation and in- al19 in which a rCBV lesion was 69% compared with healthy
jury could independently inhibit flow or promote increased white matter). Brubaker et al20 reports a somewhat lower av-
white cell and platelet adherence with subsequent blood flow erage rCBV (28%) comparing the affected posterior part with
effects at the capillary or venule level.36,37 TNF-␣ and IL-1 the unaffected anterior part of the brain. This difference might
promote endothelin-1 production and platelet degranulation be related to differing methodology or relatively small num-
could affect vessel tone, resulting in vasoconstriction.38,39 bers of patients in both of our study groups.
These conditions also develop endothelial injury with hemo- Pruning and tapering of the PCA seems distinct from the
lysis (red cell fragmentation, increased lactate dehydroge- observation of diffuse constriction and focal spasm of the
nase), capillary leakage with peripheral edema, and organ hy- PCA, though overlap was noted in many of our patients. Re-
poperfusion. Vascular instability is common in preeclampsia, sistance to organ flow is primarily at the arteriole, capillary, or
eclampsia, sepsis, and shock with both vasoconstrictive effects venule. If vasculopathy is present and resistance to brain blood
(platelet degranulation with thromboxane release, endothe- flow is increased, arterial flow velocity will likely decrease with
lin-1, angiotensin, vasopressin, and central sympathetic stim- reduced vessel visualization on the 3D TOF MRA sequence.
ulation) and vasodilatory effects (nitric oxide, prostacyclin) This could explain the pruned and tapered appearance.
noted.30,31 Although severe vessel irregularity noted at MRA may ap-
Brain vasculature may be experiencing a similar alteration pear nonspecific, these features, at least in part, reflect vascu-

AJNR Am J Neuroradiol 29:447–55 兩 Mar 2008 兩 www.ajnr.org 453

 Fig 6. Patient 41 is a 34-year-old woman with baseline blood
pressure of 104/60 mm Hg who at 33 weeks of pregnancy
developed preeclampsia (toxicity blood pressure of 153/102
mm Hg) followed by seizures and status epilepticus. She
ultimately required cesarean delivery for seizure control. A,
Axial MR imaging (FLAIR sequence) demonstrates typical
PRES vasogenic edema in the parietal regions (curved arrows)
bilaterally and was judged edema, grade 2. Frontal and
occipital involvement was also present bilaterally (not shown).
B, rCBV color map demonstrates decreased perfusion in the
region of the right parietal lesion (curved arrow). Lesion rCBV
was 65% relative to a reference healthy cortex.

lopathy by their confirmation on CA and reversal as docu- vasoconstriction (through autoregulation) and, like a biologic
mented on follow-up MRA studies. Reduced cerebral blood feedback loop, function to worsen hypoperfusion, endothelial
flow as suggested by rCBV could further contribute to per- dysfunction, or endothelial injury.
ceived vessel irregularity on the flow-sensitive 3D TOF MRA Several limitations of our study existed because of the ret-
sequence. More widespread use of advanced technology such rospective nature of this evaluation. Standardized imaging
as 3T MR imaging and intracranial MRA performed during timing and techniques (MRA, MRP, and brain imaging), com-
contrast infusion could markedly improve vessel margin visu- puter modeling of vasogenic edema, broader clinical data re-
alization, reduce flow-related effects, and help augment detec- cording (cytokines), and treatment may be of benefit. It is
tion of these subtle but important vascular features. without doubt that a prospective assessment of PRES would be
important.
PRES and Hypertension
PRES developing and spontaneously reversing in patients who Conclusion
are normotensive suggests a mechanism other than hyperten- CA, MRA, and MRP demonstrate evidence of vasculopathy
sion for the development of vasogenic edema. As graded in our with focal and diffuse vasoconstriction, focal vasodilation, and
study, vasogenic edema in patients with moderate and severe a string-of-bead pattern along with reduced rCBV suggesting a
hypertension was not greater but actually was reduced com- state of brain hypoperfusion in PRES. Vasogenic edema was
pared with patients who were normotensive (P ⫽ .045 be- lower in patients who were hypertensive compared with those
tween patients who were normotensive and those who were who were normotensive (P ⬍ .05). In the patients who were
severely hypertensive). Most patients who were severely hy- severely hypertensive, better PCA visualization correlated
pertensive demonstrated long-segment visualization of the with reduced vasogenic edema (P ⫽ .002).
PCA (despite diffuse vasoconstriction and focal spasm), and in These features suggest that the primary abnormality in
these patients, more lengthy visualization of the PCA (PCA PRES could be related to reduced cerebral blood flow and that
pruning grade 1) was associated with reduced brain edema (P hypertension may, at some level, exert a protective effect lim-
⫽ .002). If failed autoregulation with passive vasodilation and iting the developing of PRES and improving compromised
hyperperfusion were the mechanism in PRES (with or without cerebral perfusion.
endothelial injury), one would expect the opposite observa-
tions including 1) a greater degree of edema in the severe hy-
pertensive group compared with the normotensive group (ac- Acknowledgments
companied by long-segment PCA visualization) and 2) a The authors thank Marcia Kurs-Lasky, MS, for her support
greater degree of vasogenic edema within the severe hyperten- with statistical analysis and Eric Jablonowski for his assistance
sive group in those with greater PCA visualization. with image preparation.
Increased perfusion pressure might, at some point in the
evolving systemic process, help maintain brain blood flow in
the face of increased vascular resistance and hypoxemia, sug- References
gesting a protective effect. Elevated blood pressure and sys- 1. Raroque HG, Orrison WW, Rosenberg GA. Neurologic involvement in tox-
emia of pregnancy: reversible MRI lesions. Neuroradiology 1990;40:167– 69
temic vasoconstriction could be a reactive response (Cushing 2. Truwit CL, Denaro CP, Lake JR, et al. MR imaging of reversible cyclosporin
effect, biocontrol mechanism) designed to maintain adequate A-induced neurotoxicity. AJNR Am J Neuroradiol 1991;12:651–59
organ perfusion at the microvascular level. Of potential im- 3. Bartynski WS, Grabb BC, Zeigler Z, et al. Watershed imaging features and
clinical vascular injury in cyclosporin A neurotoxicity. J Comput Assist Tomogr
portance, severe systemic hypertension developing in the face 1997;21:872– 80
of an increased state of vasoreactivity might augment cerebral 4. Schaefer PW, Buonanno FS, Gonzalez RG, et al. Diffusion-weighted imaging

454 Bartynski 兩 AJNR 29 兩 Mar 2008 兩 www.ajnr.org

 discriminates between cytotoxic and vasogenic edema in a patient with 22. Osborne A. Nonatheromatous vasculopathy. In: Osborne A. Diagnostic cere-
eclampsia. Stroke 1997;28:1082– 85 bral angiography, 2nd ed. Philadelphia: Lippincott Williams and Wilkins;
5. Bartynski WS, Zeigler Z, Spearman MP, et al. Etiology of cortical and white 1999:341–58
matter lesions in cyclosporin-A and FK-506 neurotoxicity. AJNR Am J Neuro- 23. Agresti A, Mehta CR, Patel NR. Exact inference for contingency tables with
radiol 2001;22:1901–14 ordered categories. J Am Stat Assoc 1990;85:453–58
6. Hinchey J, Chaves C, Appignani B, et al. A reversible posterior leukoencepha- 24. Gijtenbeek JM, van den Bent MJ, Vecht CJ. Cyclosporine neurotoxicity: a re-
lopathy syndrome. N Engl J Med 1996;334:494 –500 view. J Neurol 1999;246:339 – 46
7. Provenzale JM, Petrella JR, Cruz LC, et al. Quantitative assessment of diffusion 25. Cerebral blood flow, cerebrospinal fluid, and brain metabolism. In: Guyton
abnormalities in posterior reversible encephalopathy syndrome. AJNR Am J AC, Hall, JE. Textbook of medical physiology. Philadelphia: Elsevier; 2006:761– 68
Neuroradiol 2001;22:1455– 61 26. Zwienenberg-Lee M, Muizelaar JP. Clinical pathophysiology of traumatic
8. Covarrubias DJ, Leutmer PH, Campeau NG. Posterior reversible encephalop- brain injury. In: Winn HR, ed. Youmans neurological surgery, 5th ed.
athy syndrome: prognostic utility of quantitative diffusion-weighted MR im- Philadelphia: Saunders; 2004:5039 – 63
ages. AJNR Am J Neuroradiol 2002;23:1038 – 48 27. Busija DW, Heistad DD, Marcus ML. Effects of sympathetic nerves on cerebral
9. Rippe DJ, Edwards MK, Schrodt JF, et al. Reversible cerebral lesions associated vessels during acute, moderate increases in arterial pressure in dogs and cats.
with tiazofurin usage: MR demonstration. J Comput Assist Tomogr Circ Res 1980;46:696 –702
1988;12:1078 – 81 28. Kontos HA, Wei EP, Navari RM, et al. Responses of cerebral arteries and arte-
10. Vaughn DJ, Jarvik JG, Hackney D, et al. High-dose cytarabine neurotoxicity: rioles to acute hypotension and hypertension. Am J Physiol 1978;234:H371– 83
MR findings during the acute phase. AJNR Am J Neuroradiol 1993;14:1014 –16 29. MacKenzie ET, Strandgaard S, Graham DI, et al. Effects of acutely induced
11. Ito Y, Arahata Y, Goto Y, et al. Cisplatin neurotoxicity presenting as reversible hypertension in cats on pial arteriolar caliber, local cerebral blood flow, and
posterior leukoencephalopathy syndrome. AJNR Am J Neuroradiol the blood brain barrier. Circ Res 1976;39:33– 41
1998;19:415–17 30. Cunningham FG, Gant NF, Leveno KJ, et al. Hypertensive disorders in preg-
12. Bartynski WS, Zeigler ZR, Shadduck RK, et al. Pretransplantation condition- nancy. In: Cunningham FG, Gant NF, Leveno KJ, et al, eds. Williams Obstetrics,
ing influence on the incidence of cyclosporine or FK-506 neurotoxicity in 21st ed. 2001:567– 618
allogeneic bone marrow transplantation. AJNR Am J Neuroradiol 31. Munford RS. Sepsis, severe sepsis, and septic shock. In: Mandell GL, Bennett
2004;25:261– 69 JE, Dolin R, eds. Principles and practice of infectious disease. Philadelphia:
13. Bartynski WS, Boardman JF, Zeigler ZR, et al. Posterior reversible encephalop- Elsevier; 2005:906 –26
athy syndrome in infection, sepsis, and shock. AJNR Am J Neuroradiol 32. Ferrara JL. Pathogenesis of acute graft-versus-host disease: cytokines and cel-
2006;27:2179 –90 lular effectors. J Hematother Stem Cell Res 2000;9:299 –306
14. Lin JT, Wang SJ, Fuh JL, et al. Prolonged reversible vasospasm in cyclosporin 33. Schots R, Kaufman L, Van Riet I, et al. Proinflammatory cytokines and their
A-induced encephalopathy. AJNR Am J Neuroradiol 2003;24:102– 04 role in the development of major transplant-related complications in the
15. Ito T, Sakai T, Inagawa S, et al. MR angiography of cerebral vasospasm in early phase after allogeneic bone marrow transplantation. Leukemia 2003;17:
preeclampsia. AJNR Am J Neuroradiol 1995;16:1344 – 46 1150 –56
16. Sengar AR, Gupta RK, Dhanuka AK, et al. MR imaging, MR angiography, and 34. Bartynski WS, Zeigler ZR, Shadduck RK, et al. Variable incidence of cyclospor-
MR spectroscopy of the brain in eclampsia. AJNR Am J Neuroradiol ine and FK-506 neurotoxicity in hematopoietic malignancies and marrow
1997;18:1485–90 conditions after allogeneic bone marrow transplantation. Neurocrit Care
17. Bartynski WS, Sanghvi A. Neuroimaging of delayed eclampsia. Report of 3 2005;3:33– 45
cases and review of the literature. J Comput Assist Tomogr 2003;27:699 –713 35. Daly AS, Xenocostas A, Lipton JH. Transplantation-associated thrombotic
18. Naidu K, Moodley J, Corr P, et al. Single photon emission and cerebral com- microangiopathy: twenty-two years later. Bone Marrow Transplant
puterised tomographic scan and transcranial Doppler sonographic findings 2002;30:709 –15
in eclampsia. Br J Obstet Gynaecol 1997;104:1165–72 36. Parent C, Eichacker PQ. Neutrophil and endothelial cell interactions in sepsis.
19. Engelter ST, Petrella JR, Alberts MJ, et al. Assessment of cerebral microcircu- The role of adhesion molecules. Infect Dis Clin North Am 1999;13:427– 47, x
lation in a patient with hypertensive encephalopathy using MR perfusion im- 37. McCuskey RS, Urbaschek R, Urbaschek B. The microcirculation during endo-
aging. AJR Am J Roentgenol 1999;173:1491–93 toxemia. Cardiovasc Res 1996:752– 63
20. Brubaker LM, Smith JK, Lee YZ, et al. Hemodynamic and permeability changes 38. Wanecek M, Weitzberg E, Rudehill A, et al. The endothelin system in septic and
in posterior reversible encephalopathy syndrome measured by dynamic sus- endotoxin shock. Eur J Pharmacol 2000;407:1–15
ceptibility perfusion-weighted MR imaging. AJNR Am J Neuroradiol 39. Mantovani A, Bussolino F, Dejanna E. Cytokine regulation of endothelial cell
2005;26:825–30 function. FASEB J 1992;6:2591–99
21. Ferris EJ. Arteritis. In: Newton TH, Potts DG. Radiology of the skull and brain: 40. Schoch HJ, Fischer S, Marti HH. Hypoxia-induced vascular endothelial
Angiography. Reprinted by Great Neck, New York: MediBooks, 1986; original growth factor expression causes vascular leakage in the brain. Brain 2002;125:
printing by Mosby 1974:2566 –97 2549 –57

AJNR Am J Neuroradiol 29:447–55 兩 Mar 2008 兩 www.ajnr.org 455