Cranial MR imaging with clinical correlation in preeclampsia and ...

N E U R O R A D I O L O G Y
Diagn Interv Radiol 2005; 11:189-194
© Turkish Society of Radiology 2005
O R I G I N A L A R T I C L E
Cranial MR imaging with clinical correlation in
preeclampsia and eclampsia
Özgür Demirtaş, Fazıl Gelal, Berna Dirim Vidinli, Leylant Ova Demirtaş, Engin Uluç, Ali Baloğlu
PURPOSE
Our aim was to determine the distribution and nature of
cranial MRI findings in preeclampsia/eclampsia, and also
to correlate them with clinical and laboratory data.
Preeclampsia is a disease characterized by hypertension, peripheral
edema, and proteinuria, which affects 4%-5% of all pregnancies.
Many of the patients with preeclampsia present with headache,
visual disorders, confusion, and depression of consciousness. The dis-
MATERIALS AND METHODS
ease is referred to as eclampsia when seizure is a component (1).
MR imaging was performed in 39 patients with
Pathological conditions, which present with headache, visual disor-
preeclampsia (n=30) and eclampsia (n=9), and the
ders, seizure, and depression of consciousness, and that regress both
distribution and signal patterns of the lesions were
clinically and neuroradiologically in a few weeks with the elimination
documented. Clinical findings, blood pressures, and
of etiological factors have different names such as “posterior reversible
laboratory data were compared statistically in pa-
tients with and without MR imaging findings.
encephalopathy syndrome” (PRES), “hypertensive encephalopathy”,
“reversible posterior cerebral edema syndrome” and “posterior reversi-
RESULTS: MR imaging was normal in 21 of the patients.
In 18 patients, cortical-subcortical lesions, which ap-
ble leukoencephalopathy syndrome” (PRLS). Recently, the term PRES (3)
peared iso-/hypointense on T1W and hyperintense on
has become preferable to the term PRLS (2), which was first described in
T2W images, were detected. The occipital lobe was in-
1996, because of the gray matter involvement accompanying white mat-
volved in all patients, followed by the parietal, frontal,
ter involvement in these patients. PRES can occur in conditions such as
and temporal lobes, and basal ganglia and pons. The
hypertension, preeclampsia/eclampsia, immunosuppressive treatment
lesions showed watershed distribution in 13 patients.
When the patients with and without MR imaging find-
(cyclosporin A, tacrolimus), and uremia (1, 2, 4-7). Vasogenic edema,
ings were compared, there was a statistically significant
most prominently at the posterior regions of the cerebral hemispheres,
difference regarding visual disturbances, depression
and less commonly, the brain stem, cerebellum, and basal ganglia, is
of consciousness, and seizures (p=0.042, p=0.006,
detected on computed tomography (CT) and magnetic resonance (MR)
p=0.000, respectively). Although patients with MR im-
aging findings showed higher blood pressures as com-
imaging examinations (1, 2, 4, 5, 7-9). Cytotoxic edema and irreversible
pared to those without MR imaging findings, there
brain damage occur if etiological factors remain or are not treated (10-
was no statistically significant difference (p=0.074). In
12).
patients with MR imaging findings, lactate dehydroge-
The cause of preeclampsia/eclampsia, which threatens the lives of
nase (LDH), uric acid, and creatinine levels were signifi-
mother and fetus, has not yet been explained. Investigation of the clini-
cantly higher than those without MR imaging findings
(p=0.006, p=0.010, p=0.005, respectively).
cal parameters associated with brain lesions detected in preeclampsia/
eclampsia cases might shed light to the pathogenesis of the disease (1,
CONCLUSION: Increased permeability of the blood-
2, 13). Schwartz et al. demonstrated that brain lesions detected in MR
brain-barrier related to endothelial injury plays a ma-
jor role in the pathogenesis of preeclampsia/eclamp-
examination are associated with endothelial damage indicators, not hy-
sia. Relatively minor increases in blood pressure may
pertension (1). The purpose of this study was to determine the distribu-
cause cerebral lesions. However, when the cerebral
tion and nature of cranial MR imaging findings in preeclampsia/eclamp-
autoregulation mechanism is considered, the distri-
sia, and also to correlate them with clinical and laboratory data.
bution of cerebral lesions in the posterior circulation
and watershed zones, which are relatively sparsely
innervated by sympathetic nerves, provides evidence
Materials and methods
that the main determinant of pathogenesis is acute
A total of 39 patients diagnosed with preeclampsia (n=30) and eclamp-
fluctuations in blood pressure.
sia (n=9) between 2000-2003 composed the study group. Ten of the pa-
Key words: ● preeclampsia ● eclampsia ● magnetic
tients were evaluated retrospectively and 29 of them prospectively. Age
resonance imaging
range was 18-39 years (mean, 26.2 years). Two of the patients were di-
agnosed with HELLP syndrome (hemolysis, elevated liver enzymes, and
From the Departments of Radiology (F.G.
fgelal@superonline.
thrombocytopenia). Two of the patients were admitted to the emergency
com, Ö.D., B.D.V., E.U.), Obstetrics and Gynecology (L.O.D.,
A.B.), İzmir Atatürk Training and Research Hospital, İzmir,
department with premature rupture of membranes and were diagnosed
Turkey.
with preeclampsia. Three of the patients had seizures in the postpartum
period. All of the patients were discharged from the hospital following
Received 11 April 2005; revision requested 29 June 2005;
revision received 09 July 2005; accepted 13 July 2005.
delivery and without sequela.
189

---

Age, systolic and diastolic arterial magnetic resonance imaging system. findings of the two HELLP cases were
blood pressure, mean arterial blood The examination protocol consisted of excluded from the statistical analyses
pressure, hematocrite (Hct), mean T1-weighted spin echo (axial and sag-
of laboratory values. All statistical anal-
eritrocyte volume (MEV), white blood ittal) and T2-weighted spin echo axial yses were performed with the SPSS 11.0
cell count (WBC), thrombocyte, lac-
sequences without intravenous con-
program for Windows. Mann-Whitney
tate dehydrogenase (LDH), aspartate trast material injection. Control MR U test and chi-square test were used for
transaminase (AST), alanine transami-
examinations were performed for two statistical evaluation and p<0.05 was
nase (ALT), urea, uric acid, albumin, of the patients two months after the accepted to be statistically significant
globulin, and calcium (Ca) values of first examination.
and p≥0.005 was accepted to be statis-
all patients were recorded. Mean arte-
Clinical and laboratory data of all tically insignificant. Data dependent
rial tension values were calculated with
patients, with or without pathological upon verbal explanations were depict-
1/3x (systolic pressure + 2 x diastolic findings on cranial MR imaging, were ed as frequency and %, data depend-
pressure) formula. Cranial MR exami-
compared statistically. As HELLP syn-
ent upon laboratory parameters were
nations were performed for all patients
drome is associated with fulminant he-
depicted as mean ±SD.
between the 1st and 7th day of the matological anomalies (14) and can er-
onset of clinical symptoms in a 0.5T roneously affect laboratory results, MR Results
Of our 39 patients, 37 (94.9%) had
headaches, 14 (35.9%) had visual dis-
Table 1. Clinical findings and distribution of lesions in patients with MR imaging findings
orders, 9 (23%) had seizures, and 9
(23%) had depression of conscious-
Patient No. Occipital
Parietal
Frontal
Temporal Basal ganglia
Pons Clinical findings
ness. Cranial MR imaging was found
1
+
+
+
+
+

-
HA, S, DC
to be normal in 21 patients (53.8%).
In the remaining 18 patients, cortical-
2
+
+
+
+
+

+
HA, S, DC
subcortical lesions that were iso-/hy-
3
+
+
-
-
+
-
HA, VD
pointense on T1-weighted images and
4
+
+
-
+
-
-
HA, VD
hyperintense on T2-weighted images
5
+
+
-
+
-
-
HA, VD, S, DC
were detected. Subcortical white mat-
ter was affected in only 8 of these 18
6
+
+
+
-
+
-
HA, VD
patients (44.4%) and pathological sig-
7
+
-
-
-
+
-
HA, DC
nals were present in both the cortex
8
+
+
+
+
+

-
HA, VD, DC
and the white matter in the remaining
10 patients (55.6%). Anterior, poste-
9
+
-
-
-
-
-
HA
rior, or parasagittal watershed zone
10
+
+
-
-
-
-
HA, VD
involvement was present in 13 pa-
11
+
-
-
-
-
-
HA
tients (72.2%). Clinical findings and
12
+
+
+
+
+

-
HA, S, DC
distribution of lesions in patients with
MR findings are presented in Table 1.
13
+
+
+
+
+
-
HA, S
Pathological findings regressed in two
14
+
+
+
+
+
-
HA, VD, S
patients who had control examina-
15
+
+
-
-
-
-
HA, VD
tions about two months after the first
16
+
+
+
+
+
+
HA, S, DC
examination (Figures 1-3).
Clinical findings and statistical anal-
17
+
-
-
-
-
-
HA, VD, S
yses of patients with or without MR
18
+
+
-
-
-
-
HA, VD, S, DC
findings are shown in Table 2. When
Total
100%
77.7%
50%
50%
55.5%
11.1%
the patients with or without MR im-
aging findings were compared, there
HA: headache, S: seizure, DC: depression of consciousness, VD: visual disorder
was a statistically significant difference
regarding visual disturbances, depres-
sion of consciousness, and seizures,
but there was no significant difference
Table 2. Clinical findings in patients with and without positive MR imaging findings, and
their statistical comparison
regarding headaches. There was no sta-
tistically significant difference between
MR imaging positive MR imaging negative
p
the mean age of patients with or with-
(n=18)
(n=21)
out MR imaging findings (Mann-Whit-
Headache
18
19
0.490
ney U test, p=0.097). Although patients
with MR imaging findings showed
Visual disorder
10
4
0.042*
higher blood pressures as compared
Depression of consciousness
8
1
0.006*
to those without MR findings, there
was no statistically significant differ-
Seizure
9
0
0.000*
ence between the two groups (Mann-
*Statistically significant p values with chi-square test
Whitney U test, p=0.074). In Table 3,
minimum and maximum mean blood
190 • December 2005 • Diagnostic and Interventional Radiology
Demirtaş et al.

---

pressure values, as well as their means, rogenic components, maintains stable which sympathetic innervation induc-
and statistical comparison of the pa-
blood perfusion in normal individuals es, moderately protects anterior circula-
tient groups with or without MR find-
(1). Effective functioning of neurogenic
tion areas from overperfusion. In PRES
ings are given. Biochemical data, lac-
mechanisms depends on sympathetic cases without hypertension, direct en-
tate dehydrogenase (LDH), uric acid, innervation. In PRES cases, direct toxic dothelial cell dysfunction, which in-
and creatinine levels were significantly effects on endothelium or vessel disten-
creases blood-brain barrier permeabil-
higher in patients with positive MR tion, which depends on elevated blood ity, is thought to be responsible for the
findings than those without MR find-
pressure, decreases the effect of myo-
pathogenesis (1, 8).
ings (Mann-Whitney U test, p=0.006, genic mechanisms (1, 4). In these cases,
In this study, various clinical and
p=0.010, p=0.005, respectively).
neurogenic mechanisms take over the laboratory parameters in preeclampsia/
regulation of cerebral perfusion; this eclampsia cases with and without posi-
Discussion
way, posterior circulation areas, which tive brain MR imaging findings were
Pathogenesis in PRES patients is are relatively sparsely innervated by compared. Brain lesions were found
not yet fully understood, but a major sympathetic nerves, become more sen-
to be associated with high LDH levels,
causative factor is thought to be the ex-
sitive to blood pressure elevations. In which is an indicator of hemolysis,
travasation of fluid (15). Brain lesions PRES cases with hypertension, serum and both high uric acid and creatinine
detected in these patients might be re-
extravasation occurs when the eleva-
levels, which are indicators of renal
lated to a disturbance of the cerebral tion in blood pressure passes beyond function disorder. Endothelial injury
autoregulation mechanism and impair-
the autoregulation capacity of brain yields to morphological disturbances in
ment of endothelial function (1, 8, 16).
blood vessels. Brain lesions are more erythrocytes and microvascular hemo-
The cerebral autoregulation mecha-
commonly demonstrated in posterior lysis; as a result, LDH level increases
nism, consisting of myogenic and neu-
areas in these cases. Vasoconstriction, (1). The cause of endothelial injury in
preeclampsia/eclampsia cases has not
yet been demonstrated, but circulat-
ing endothelial toxins or endothelium
Table 3. Comparison of biochemical data and mean blood pressure in patients without and
antibodies are thought to be responsi-
with positive MR imaging findings
ble (1). In previous studies, it has been
MR imaging negative
MR imaging positive
demonstrated that in preeclampsia/ec-
p
lampsia patients, high LDH levels oc-
Minimum Maximum
Mean±SD
Minimum Maximum
Mean±SD
cur before lesions appear in brain MR
Hct (%)
24.40
45.00
33.45±6.13
28.00
45.40
34.59±4.96
0.469
examination, and this finding shows
that high blood pressure does not lead
WBC (103/µl)
9.0
24.00
13.49±3.22
7.30
18.40
12.79±3.25
0.748
to endothelial injury (1, 17). Multiple
Thrombocyte
86.00
515.00
264.71±104.88
96.00
356.00
245.20±70.03
0.432
organ involvement in preeclampsia/
(103/µl)
eclampsia cases is thought to be as-
sociated with an endothelial function
MEV
62.40
96.90
86.76±7.33
78.10
93.80
89.76±4.25
0.123
disorder (17). Renal function disorder
LDH (U/l)
322
925
605.48±184.51
501
1012
792.40±186.08 0.006*
secondary to renal endothelial injury
results in uric acid and creatinine in-
AST (U/l)
15
75
35.76±15.67
20
39
31.93±6.78
0.797
crease.
ALT (U/l)
6
93
26.24±18.89
7
39
25.60±8.67
0.712
In our study, there was no statisti-
cally significant difference between
Uric acid
2.80
10
4.41±1.83
3.10
7.30
5.61±1.17
0.010*
(mg/dl)
blood pressure values of cases with or
without MR imaging evidence of brain
Urea (mg/dl)
12.9
128
30.45±24.26
6
58.0
33.40±12.31
0.077
lesions. But in cases of preeclampsia/
eclampsia, brain lesions might occur
Albumine
2.20
4.0
2.99±0.55
2
4
2.87±0.57
0.584
(g/dl)
although blood pressure values are
normal but still higher than a patient’s
Globuline
1.80
3.10
2.66±0.35
2.0
3.20
2.60±0.36
0.438
routine normal blood pressure (1).
(g/dl)
When blood pressure values and results
Ca
5.70
8.0
7.19±0.68
5.60
7.90
7.26±0.69
0.675
associated with laboratory parameters
(mEq/dl)
were evaluated together in our study,
brain edema detected in preeclampsia/
Creatinine
0.50 1.10
0.68±0.14
0.6
2.10 0.86±0.31
0.005*
(mg/dl)
eclampsia was thought to be secondary
to endothelial injury, rather than hy-
Mean BP
100
146
123.81±11.70
116
166
131.87±13.41
0.074
pertension. These findings correlated
(mmHg)
with the findings of Schwartz et al. (1).
SD: standard deviation, Hct: hematocrit, MEV: mean erythrocyte volume, WBC: white blood cell count,
But the distribution of brain lesions
LDH: lactate dehydrogenase, AST: aspartate transaminase, ALT: alanine transaminase, Ca: calcium, BP:
should also be evaluated in determin-
blood pressure.
ing whether hypertension or endothe-
*Statistically significant p values with Mann-Whitney U test
lial injury is more responsible for the
pathogenesis.
Volume 11 • Issue 4
Cranial MR imaging with clinical correlation in preeclampsia and eclampsia • 191

---

Figure 1. a, b. Eclampsia. A 23-year-old
a
b
patient, 36 weeks pregnant, with headache
for one week has been hospitalized for new
onset visual loss and generalized tonic clonic
convulsion. Axial T2-weighted MR image (a)
shows high-signal lesion in the right head of
caudate and partially the lentiform nucleus.
Axial T2-weighted image at a superior level
(b) shows bilateral parietooccipital high-signal
lesions involving the cortex and the white
matter.
Figure 2. a, b. Preeclampsia; cranial MR
a
b
imaging of a 19-year-old patient, 38 weeks
pregnant, with developing headache, visual
loss, and depression of consciousness.
Axial T2-weighted MR image (a) shows
bilateral frontoparietal and occipital lesions
involving the cortex and the white matter.
Two months later (b) the lesions have
disappeared except for minimal bilateral
parietal subcortical hyperintensities
(arrows).
a
b
c
Figure 3. a-c. Preeclampsia. Cranial MR imaging of a patient, 34 weeks pregnant, with headache for a few weeks and developing acute visual
loss. Axial T2-weighted MR images (a and b) show hyperintensities in bilateral frontal and parietooccipital lobes, quite symmetrical, involving
anterior and posterior watershed zones cortically and subcortically. Axial T1-weighted MR image (c) shows the lesions to be iso-/hypointense.
We considered that PRES cases might
cases, lesions occur only in cases of levels of blood pressure and vasogenic
be evaluated in three groups accord-
high blood pressure values, as there is edema occurs in posterior circulation
ing to pathogenesis. In the first group, no endothelial injury leading to fluid areas, especially in the occipital lobes
brain edema develops only secondary extravasation. Limits of cerebral au-
and watershed zones, which are rela-
to elevated blood pressure. In these toregulation pass beyond the normal tively sparsely innervated by sympa-
192 • December 2005 • Diagnostic and Interventional Radiology
Demirtaş et al.

---

thetic nerves. Arterial watershed zones follow this. The cerebellum and brain was no statistically significant differ-
have been demonstrated not be rich stem might be involved in more severe
ence between patients with or without
in sympathetic innervation as are pos-
cases (1, 24). The occipital lobe was in-
MR imaging findings regarding head-
terior circulation areas (18). Cases of volved in all of our cases. Lobar distri-
aches. The detection of visual disor-
hypertensive encephalopathy occur bution of lesions was similar to other ders, depression of consciousness, or
in this group. Distribution of lesions series in the literature, and watershed seizure in the follow-up of pregnant
in the second group is similar to the zone involvement was prominent in patients with preeclampsia/eclampsia
first group, but lesions might occur in our cases. Watershed zone-distrib-
should be a warning for possible brain
cases of lower blood pressure values, uted lesions evocate ischemia second-
lesions.
even those close to normal. Endothe-
ary to cerebral vasospasm, but as they
In conclusion, increased permeabil-
lial injury plays a major role in the are demonstrated not to be cytotoxic ity of the blood-brain barrier related to
pathogenesis of these cases; as a result edema with diffusion weighted images,
endothelial injury plays a major role in
of a failure of the blood-brain barrier, the possibility of ischemia is not an is-
the pathogenesis of preeclampsia/ec-
fluid might leak into the interstitium sue of concern (4).
lampsia. In cases where blood pressure
in milder blood pressure elevations. As
If radiological and clinical findings elevations do not reach very high lev-
lesions occur especially in areas of low occurring in PRES are easily recognized
els, intravascular fluid leaks into inter-
autoregulation capacity, acute blood and treated immediately, they might stitium and vasogenic edema occurs.
pressure fluctuations are thought to be totally reversible (25). All the pa-
However, when the cerebral autoregu-
be a major determinant in pathogen-
tients investigated in our study were lation mechanism is considered, the
esis. In experimental studies, acute discharged from the hospital without distribution of cerebral lesions in the
blood pressure elevations were shown neurological deficit within 1-2 weeks. posterior circulation and watershed
to impair the autoregulation mecha-
Control brain MR examination was zones, which are relatively sparsely in-
nism and cause fluid and blood leak-
performed in two of our patients two nervated by sympathetic nerves, there
age as a result of vasoconstriction and months after the first examination. To-
is evidence that the main determinant
vasodilatation, especially in arterial tal recovery was detected in one of the of pathogenesis is acute fluctuations in
watershed zones (19). Preeclampsia/ec-
patients and subcortical hyperintense blood pressure. When symptoms like
lampsia, cyclosporin-A, and tacrolimus
lesions were present in a very limited visual disorders, depression of con-
toxicity are placed in this group (1, area in the other patient. Ischemia, sciousness, or seizure develop during
5, 20, 21). In the third group, lesions massive infarct, and death might occur
the follow-up of pregnancy, or indica-
predominantly occur secondary to if the disease is not recognized. Diffu-
tors of endothelial injury are detected
endothelial injury; blood pressure el-
sion weighted MR imaging is used to to be elevated, appropriate precautions
evations do not play a major role in distinguish ischemia-cytotoxic edema to regulate blood pressure would be
pathogenesis. Symmetrical basal gan-
from vasogenic edema, which is present
beneficial.
glion involvement is typical in these in PRES cases (4, 11, 12, 16). Cases that
cases; thalamus, cerebellum, and brain develop cytotoxic edema-ischemia as References
stem involvement follow this. Occipi-
a complication of PRES have been re-
1. Schwartz RB, Feske SK, Polak JF, et al.
tal lobe and watershed zone involve-
ported (11, 12). It is accepted that mas-
Preeclampsia-eclampsia: clinical and neu-
ment is not as prominent as they are sive vasogenic edema might cause an
roradiographic correlates and insights into
the pathogenesis of hypertensive encepha-
in PRES cases with hypertension. Some
increase in tissue pressure and ischemia
lopathy. Radiology 2000; 217:371-376.
kinds of uremic encephalopathy, espe-
by impairing microcirculation (11).
2. Hinchey J, Chaves C, Appignani B, et al. A
cially hemolytic uremic syndrome, are
Visual disorders, depression of con-
reversible posterior leukoencephalopathy
placed in this group. (22). In uremic sciousness, and seizures were more fre-
syndrome. N Engl J Med 1996 22; 334:494-
500.
pathologies with renal hypertension, quently detected in cases of preeclamp-
3. Casey SO, Sampaio RC, Michel E, Truwit
lesion distribution might be similar to sia/eclampsia with cerebral edema in
CL. Posterior reversible encephalopathy
the first two groups (23). By evaluating
MR imaging than in patients with nor-
syndrome: utility of FLAIR imaging in
blood pressure measurements and the mal MR examinations. Brain lesions
the detection of cortical and subcorti-
distribution of brain lesions, an under-
were found to be significantly corre-
cal lesions. AJNR Am J Neuroradiol 2000;
21:1199-1206.
standing of pathogenesis might be pos-
lated with only seizure in the study of 4. Schwartz RB, Mulkern RV, Gudbjartsson
sible in PRES cases with the classifica-
Schwartz et al. (4). This diversity might
H, Jolesz F. Diffusion weighted MR imag-
tions we suggest.
be explained by the widespread dis-
ing in hypertensive encephalopathy: clues
Lesions are generally iso-/hypoin-
tribution of brain edema areas in our
to pathogenesis. AJNR Am J Neuroradiol
1998; 19:859-862.
tense on T1-weighted images, hyper-
study. Seizure is thought to be second-
5. Coley SC, Porter DA, Calamante F, et al.
intense on T2-weighted images, and ary to the irritative effect of fluid in the
Quantitative MR diffusion mapping and
petechial hemorrhage is rarely seen in cortex and white matter (1). Cerebral
cyclosporin-induced neurotoxicity. AJNR
patients with preeclampsia/eclampsia edema was present in our all cases with
Am J Neuroradiol 1999; 20:1507-1510.
(1, 13, 24). Lesions with similar signal seizures. Depression of consciousness 6. Oliverio PJ, Restrepo L, Mitchell SA, et al.
Reversible tacrolimus-induced neurotoxic-
changes were demonstrated in all of our
was demonstrated to be more frequent
ity isolated to the brainstem. AJNR Am J
cases, but no findings of hemorrhage in cases with basal ganglion involve-
Neuroradiol 2000; 21:1251-1254.
were detected. The occipital lobe is the ment in some studies (26). In 6 of the 7. Port J, Beauchamp N. Reversible intrac-
most affected region in preeclampsia/
10 (60%) patients with depression of
erebral pathologic entities mediated by
vascular autoregulatory dysfunction.
eclampsia; parietal, frontal, temporal consciousness in our study, basal gan-
Radiographics 1998; 18:353-367.
lobe, and basal ganglion involvement glion involvement was present. There
Volume 11 • Issue 4
Cranial MR imaging with clinical correlation in preeclampsia and eclampsia • 193

---

8. Keswani SC, Wityk R. Don't throw in the

15. Sheth RD, Riggs JE, Bodenstenier JB, et al.

21. Furukawa M, Terae S, Chu BC, Kaneko K,
towel! A case of reversible coma. J Neurol
Parietal occipital edema in hypertensive
Kamada H, Miyasaka K. MRI in seven cases
Neurosurg Psychiatry 2002; 73:83-84.
encephalopathy: a pathogenic mechanism.
of tacrolimus (FK-506) encephalopathy:
9. Karaköse S, Deniz E, Karabacakoğlu A,
Eur Neurol 1996; 36:25–28.
utility of FLAIR and diffusion-weighted
Sevgili M. Preeklampsi ve eklampside kran-

16. Covarrubias DJ, Luetmer PH, Campeau
imaging. Neuroradiology 2001; 43:615-
yal MRG bulguları. Tani Girisim Radyol
NG. Posterior reversible encephalopathy
621.
2001; 7:475-479.
syndrome: prognostic utility of quantita-

22. Steinborn M, Leiz S, Rudisser K, Griebel

10. Lamy C, Oppenheim C, Meder JF, Mas JL.
tive diffusion weighted MR images. AJNR
M, Harder T, Hahn H. CT and MRI in
Neuroimaging in posterior reversible en-
Am J Neuroradiol 2002; 23:1038-1048.
haemolytic uraemic syndrome with cen-
cephalopathy syndrome. J Neuroimaging

17. McCarthy AL, Woolfson RG, Raju SK,
tral nervous system involvement: distri-
2004; 14:89-96.
Poston L. Abnormal endothelial cell func-
bution of lesions and prognostic value

11. Ay H, Buonanno FS, Schaefer PW, et al.
tion of resistance arteries from women
of imaging findings. Pediatr Radiol 2004;
Posterior leukoencephalopathy without
with preeclampsia. Am J Obstet Gynecol
34:805-810.
severe hypertension: utility of diffusion
1993; 168:1323-1330.

23. Schmidt M, Sitter T, Lederer SR, Held E,
weighted MRI. Neurology 1998; 51:1369-

18. Sakaharova AV. Regional characteristics of
Schiffl H. Reversible MRI changes in a
1376.
the noradrenergic and cholinergic inner-
patient with uremic encephalopathy. J

12. Koch S, Rabinstein A, Falcone S, Forteza A.
vation of the vessels of the brain surface.
Nephrol 2001;14:424-427.
Diffusion-weighted imaging shows cytotox-
Bull Eksp Biol Med 1980; 89:141-143.

24. Sengar AR, Gupta RK, Dhanuka AK, Roy R,
ic and vasogenic edema in eclampsia. AJNR

19. Friedman SA, Taylor RM, Roberts JM.
Das K. MR imaging, MR angiography, and
Am J Neuroradiol 2001; 22:1068-1070.
Pathophysiology of preeclampsia. Clin
MR spectroscopy of the brain in eclamp-

13. Sanders TG, Clayman DA, Sanchez-Ramos
Perinatol 1991; 18:661-682.
sia. AJNR Am J Neuroradiol 1997; 18:1485-
L, Vines FS, Russo L. Brain in eclampsia: MR

20. Yamamoto A, Hayakawa K, Houjyou M.
1490.
imaging with clinical correlation. Radiology
CT and MRI findings of cyclosporine-re-

25. Dillon W, Rowley H. The reversible poste-
1991; 180:475-478.
lated encephalopathy and hypertensive
rior cerebral edema syndrome. AJNR Am J

14. Weinstein L. Syndrome of hemolysis, el-
encephalopathy. Pediatr Radiol 2002;
Neuroradiol 1998; 19:591.
evated liver enzymes and low platelet
32:340-343.

26. Koyama M, Tsuchiya K, Hanaoka H, et al.
count: a severe consequence of hyperten-
Reversible intracranial changes in eclamp-
sion in pregnancy. Am J Obstet Gynecol
sia demostrated by MRI and MRA. Eur J
1982; 142:159.
Radiol 1997; 25:44-46.
194 • December 2005 • Diagnostic and Interventional Radiology
Demirtaş et al.

---