contents TIB


Introduction
The pregnancy-related syndromes 

of pre-eclampsia, eclampsia, the

Haemolysis, Elevated Liver enzymes 
and Low Platelets or HELLP syndrome,
and the Thrombotic Thrombocytopenic
Purpura/Haemolytic Uraemic Syndrome
(TTP/HUS) are a group of disorders shar-

ing various clinical and pathogenetic fea-

tures, principally related to endothelial

damage and development of thrombotic

microangiopathy triggered by pregnancy.

The clinical abnormalities of these disor-

ders include neurological disturbances,

acute renal failure, hypertension, haemoly-

sis, deranged liver enzymes, and thrombo-

cytopenia. This review looks at the similar-

ities and differences between these various

pregnancy-related syndromes as well as

some of the relevant imaging features that

may sometimes be encountered by radiolo-

gists.

Definitions
Gestational hypertension is defined as

blood pressure (BP) elevation of 140/90

mmHg or higher in a woman previously

normotensive before 20 weeks’ gestation,

without proteinuria. This is seen in about

6% of all pregnancies.1,2

Pre-eclampsia is defined as gestational

hypertension (BP ± 140/90 mmHg) with

proteinuria > 0.3 g /24 hours. Pre-eclamp-

sia is seen in about 3 - 8% of all pregnan-

cies.1,2

Severe pre-eclampsia is defined as

occurring when the BP is higher than 160 -

180/110 mmHg and/or evidence of sys-

temic involvement or end-organ dysfunc-

tion.1,2

Eclampsia is defined as the occurrence

of seizures without any apparent neurolog-

ical aetiology in a woman who has or who

develops pre-eclampsia. About 2% of

women with pre-eclampsia will develop

seizures.1,2

The HELLP syndrome is defined as the

presence of haemolysis, elevated liver

enzymes and thrombocytopenia.

Although often referred to as being severe

pre-eclampsia with the additional features

mentioned above, hypertension does not

have to be severe or even present at all in

the HELLP syndrome. HELLP syndrome

occurs to some degree in about 4 - 39% of

women with pre-eclampsia. When the

platelet count is < 50 000/µl the syndrome

is considered severe.1-3

TTP/HUS is characterised by the pres-

ence of thrombocytopenia, haemolysis,

neurological abnormalities and oliguric

renal failure. Most pregnancy-related cases

occur in the last trimester or postpartum

period, and many have associated pre-

eclampsia. However, only 13% of cases in

women with TTP/HUS are pregnancy-

related.1,2

Pathology
The aetiology of these syndromes

remains largely unclear, with several pro-

posed theories including abnormalities of

the placento-maternal unit, auto-immune

disorders, endothelial cell damage and fur-

ther genetic and dietary influences.1,4 The

basic underlying process involves vascular

endothelial cell damage. This, in turn, may

be initiated by circulating factors released

by the placenta secondary to ischaemic

damage within the placenta itself.

Substances released by the damaged

endothelium such as ultra-large multimers

of Von Willebrand factor (VWF) and

thromboxane A2 cause platelet activation

and aggregation, further activating other

clotting factors and fibrin. These platelet-

fibrin microaggregates then lodge in the

microvessels (arterioles) either blocking

them completely leading to reduced tissue

perfusion, or partly leading to damage of

erythrocytes as they are forced through the

narrowed channels. This, in turn, leads to

haemolysis and is therefore termed

microangiopathic haemolytic anaemia or

MAHA. The endothelial cell damage also

results in an abnormal pressor response to

both circulating vasopressor agents  (e.g.

angiotensin II) and local endothelial ones

(e.g. thromboxane A2, nitric oxide). This

results in a generalised increase in vascular

tone with resulting hypertension and

vasospasm further contributing to end-

organ hypoperfusion and damage. Finally

the endothelial damage leads to capillary

leakage with the development of tissue

oedema, reduced circulating volume and

haemoconcentration. These are the basic

common pathomechanisms behind the

pregnancy-related thrombotic microan-

giopathies. The various clinical syndromes

are largely differentiated by the degree of

severity and specificity of the end-organ

damage, although the common pathome-

chanisms permit an explanation for the

considerable degree of clinical and patho-

genetic overlap between them.

Clinical features
These are largely related to the presence

of vasomotor abnormalities and reduced

tissue perfusion, as well as the effects of

thrombocytopenia, haemolysis and

anaemia, and mainly include neurological

manifestations, hepatic and renal dysfunc-

tion, and coagulapathies. Although pre-

eclampsia is diagnosed on the basis of the

elevated BP and proteinuria (minimal diag-

nostic criteria), there are a number of other

associated abnormalities that when present

make the diagnosis of pre-eclampsia more

certain. These include mild thrombocy-

topenia with a platelet count of < 100

000/ml, haemolysis with an abnormal

blood smear and increased lactate dehydro-

genase (LDH) levels, elevated liver function

REVIEW ARTICLE

9 SA JOURNAL OF RADIOLOGY • October 2005

Pre-eclampsia, eclampsia 
and the thrombotic 

microangiopathic syndromes
of pregnancy

I C Duncan
FFRad(D) SA

Unitas Interventional Unit 
Centurion

pg9-12  9/28/05  1:17 PM  Page 9



tests (ALT, AST), epigastric pain and nau-

sea due to liver distension, and neurological

disturbances such as persistent headache or

visual disturbances. Whereas mild throm-

bocytopenia is common in women with

pre-eclampsia (15 - 29% of cases),

haemolytic anaemia is not and occurs in

only about 2% of pre-eclamptic women.

Haemolysis, if significant enough, is more

likely to be part of HELLP or TTP/HUS.

HELLP is characterised by low platelets 

(< 100 000/ml), haemolysis and elevated

liver enzymes with associated swelling of

the liver producing epigastric or right

upper quadrant pain, nausea and vomiting.

In rare cases spontaneous hepatic rupture

may occur due to massive liver swelling.

Hypertension may or may not be present

with HELLP, and cerebral manifestations

are rare.3 On the other hand, renal and

cerebral complications are more frequently

encountered in patients with TTP/HUS,2,5,6

whereas liver function disturbances are not,

although these patients may still develop

abdominal pain, nausea and vomiting. The

neurological features associated with

TTP/HUS include generalised mental sta-

tus abnormalities (such as confusion, stu-

por or coma), focal neurological abnormal-

ities (such as visual disturbances, aphasia,

dysarthria, hemiparesis) and seizures.

Hypertension is not a feature of TTP/HUS

unless there is associated pre-eclampsia.2

Put simplistically, pre-eclampsia/severe

pre-eclampsia may have some or all of the

abovementioned clinical features plus

hypertension, whereas HELLP mainly

affects the liver and TTP/HUS preferential-

ly involves the kidneys and central nervous

system.

Many cases of antepartum pre-eclamp-

sia, HELLP or TTP/HUS will resolve spon-

taneously. Some may persist, worsen or

even change following delivery, e.g. from

pre-eclampsia to HELLP or TTP/HUS.

Some cases develop de novo in the postpar-

tum period. Eclampsia developing within

48 hours after delivery occurs in approxi-

mately 25% of all eclamptics. Late postpar-

tum eclampsia (beyond 48 hours after

delivery) is very rare and can occur up to 23

days after delivery.7,8 Mortality due to

eclampsia can occur in up to 13% of cases,

while in the HELLP syndrome mortality

ranges from 2% to 24%.4 Between 50% and

65% of maternal deaths due to eclampsia

occur as a result of cerebral haemorrhage,

with other causes including hypertensive

encephalopathy, cardiac arrhythmias, con-

gestive heart failure and disseminated

intravascular coagulapathy. Other compli-

cations that can occur in eclamptics include

pulmonary oedema and retinal detach-

ment.4

Obstetric complications related to

these syndromes include placental abrup-

tion and foetal distress and death.

More about
TTP/HUS

TTP was first described in 1925, being

characterised by fever, fluctuating neuro-

logical manifestations, renal dysfunction,

thrombocytopenia and haemolytic

anaemia. HUS was first described in 1955,

being described primarily as a disease of

childhood and also being characterised by

fever, oliguric renal failure, thrombocy-

topenia and haemolytic anaemia. HUS

occurs mainly in young children due to

infection by Shiga-toxin producing entero-

haemorrhagic strains of Escherichia coli.

These children invariably do well clinically

with supportive therapy only.2,5,6More

recently both TTP and HUS have been

recognised as being different clinical pre-

sentations of a common pathological spec-

trum of thrombotic microangiopathy, and

are now referred to jointly as TTP/HUS.

There are idiopathic and familial forms of

TTP/HUS, with other cases of TTP/HUS

occurring secondary to cancer and

chemotherapy, transplantation and

immunosuppressive therapy, systemic

infection, autoimmune disease, drug reac-

tions and pregnancy. Sixty-six per cent of

all patients with TTP/HUS are female. Of

these cases only 13% are pregnancy related,

meaning that unlike HELLP which is total-

ly pregnancy-induced, only 8.5% of all

TTP/HUS cases are pregnancy related.

Some patients with TTP/HUS have been

found to be severely deficient in the activi-

ty of the metalloprotease ADAMTS-13 

(< 5% normal activity).9,10 ADAMTS-13

cleaves ultra-large multimeric molecules of

Von Willebrand factor (VWF) released by

the endothelium. VWF is a potent activa-

tor of platelet activation and aggregation.

Therefore a severe deficiency of ADAMTS-

13 allows uncontrolled platelet aggregation

resulting in microvascular occlusion,

thrombocytopenia and consumption coag-

ulopathy. As not all patients with

TTP/HUS have ADAMTS-13 deficiency

there are clearly other pathogenetic factors

involved that have yet to be elucidated.

Pregnancy has been shown to provoke

acute TTP/HUS in women with familial

ADAMTS-13 deficiency,2,5 particularly in

homozygotes. Without  treatment TTP is

fatal in 90% of patients. The only effective

treatment is plasma exchange, thought to

work by removing autoantibodies or other

circulating pathogenetic trigger factors, as

well as by replenishing absent levels of

ADAMTS-13. It is therefore important to

differentiate cases with TTP/HUS from

those with severe pre-eclampsia and

HELLP as plasma exchange is of no benefit

in the treatment of the latter pathologies.2,5,9

Imaging features
These are generally related to the neu-

rological manifestations of these patholo-

gies. Cerebral vasospasm has been demon-

strated in eclampsia and severe pre-

eclampsia using a variety of imaging meth-

ods including transcranial Doppler (TCD),

magnetic resonance angiography and con-

ventional catheter angiography (Figs 1a,

b).7,11-16 The true incidence of cerebral

vasospasm in eclamptics and severe pre-

eclamptics is unknown. Similar findings

have also been reported in patients with

HELLP,3,8 and we have seen a single case of

cerebral vasospasm associated with post-

partum TTP/HUS. In addition to the

abnormal pressor response of blood vessels

relating to endothelial cell damage, the

develop of cerebral vasospasm may also

result in part due to high circulating levels

of oxyhaemoglobin following breakdown

of red blood cells due to the MAHA, with

irritation of the vessels by the free oxy-

haemoglobin itself.3 This spasm can affect

intracranial vessels in an interrupted or

continuous manner, and affects both large

and medium-sized vessels. It mimics cere-

bral vasculitis angiographically.12,16,17 As

REVIEW ARTICLE

10 SA JOURNAL OF RADIOLOGY • October 2005

pg9-12  9/28/05  1:17 PM  Page 10



with the clinical symptoms the angiograph-

ic findings can spontaneously resolve with-

in weeks of presentation.17,18 Although most

cases are eventually reversible, some can

develop associated cerebral infarction or

haemorrhage.13,18 The other major radiolog-

ical abnormality seen in these cases is cere-

bral oedema. Hypodense areas are seen on

CT scans and hyperintense areas are seen

on T2-weighted and FLAIR MR images in

such cases. These areas tend to be subcor-

tical but can include the cortex, basal gang-

ha and brainstem. Associated infarction

and haemorrhage are uncommon.19 Two

key differentiating features of cerebral

oedema related to the pregnancy syn-

dromes are firstly that the changes are com-

moner in the occipital and parietal lobes,

and secondly that they are reversible.7,14,15,19,20

These features have led to coining of the

term ‘Posterior Reversible Encephalopathy

Syndrome’ (PRES). PRES is, however, not

unique to pregnant patients, and has also

been described in cases receiving immuno-

suppressive and chemotherapeutic drugs,

after transplantation, in patients with

hyperintensive encephalopathy, and in

patients with TTP/HUS due to causes other

than pregnancy.10,21-23 Despite a predilection

for the posterior circulation territories, the

deep structures and anterior circulation

territories are also frequently involved (Figs

2a-c). The density/signal changes are said

REVIEW ARTICLE

11 SA JOURNAL OF RADIOLOGY • October 2005

Fig. 1a. 36-year-old woman with last trimester
pre-eclampsia with severe postpartum hyper-
tension and intracerebral haemorrhage 11
days after delivery. Note the generalised spas-
tic appearence of the central cerebral arteries
at arteriography.

Fig. 1b. Mirror oblique projection, left internal
carotid artery showing diffuse vasospasm on
this side as well.

Fig. 2a. 30-year-old woman who developed
HELLP syndrome and eclampsia shortly after
delivery at 39 weeks. This FLAIR MR image
shows areas of increased signal in the cere-
bellum and pons.

Fig. 2b. Similar areas of increased signal are
seen in the occipital lobes and midbrain.

Fig. 2c. This image shows further involvement
of the deep cerebral structures. There is rela-
tive sparing of the frontal and anterior parietal
and temporal regions. The MR appearance is
compatible with PRES, predominantly affect-
ing the posterior brain structures.

Fig. 2d. A gradient-echo sequence shows sub-
tle early haemorrhage in both hemicerebellar
hemispheres. This suggests the development
of true infarctions with early haemorrhagic
transformation, which is uncommonly seen in
PRES.There is also severe thrombocytopenia
present at this stage.

Fig. 2e. A CT scan obtained 2 days later after
further neurological deterioration now shows
further haemorrhagic transformation of the
cerebellar lesions, leading to greater brain
swelling and eventual fatal outcome.

pg9-12  10/8/05  11:16 AM  Page 11



REVIEW ARTICLE

12 SA JOURNAL OF RADIOLOGY • October 2005

to be due to vasogenic oedema, showing up

on T2 and FLAIR images but with negative

diffusion scans. Where present diffusion

abnormalities suggest progression from

vasogenic to cytotoxic oedema with cell

death and infarction and thus poorer clini-

cal outcome. Subclinical haemorrhages

may be seen using the appropriate gradi-

ent-echo or other blood/iron sensitive

sequences (Figs 2d,e). The pathophysiolo-

gy of these reversible changes is incom-

pletely understood and is felt to be due to a

failure in cerebral autoregulation, possibly

secondary to endothelial damage coupled

with hypertension, vasospasm and micro

circulatory ischaemic damage.21-23 The dif-

ferential diagnosis for cerebral oedema

and/or haemorrhage in pregnancy must

include central venous/ dural sinus throm-

bosis and venous infarction, haemorrhage

from a vascular malformation or

aneurysm, arterial infarction and trau-

ma.1,24

Treatment
The mainstay of prevention of eclamp-

sia remains administration of magnesium

sulphate. Nimodipine has been shown to

be less effective in this regard.25,26

Supportive therapy such as platelet and

blood transfusions, antihypertensive

agents, dialysis and early delivery may all be

of benefit in severe pre-eclampsia. High-

dose intravenous corticosteroid therapy is

useful in HELLP syndrome.27 There is a

single documented case28 of percutaneous

angioplasty used to treat eclampsia-related

vasospasm with good angiographic and

clinical results.

Conclusion
In conclusion, the role of the radiolo-

gist in the diagnosis and management of

these conditions is somewhat limited, but it

is useful to have at least some understand-

ing of their pathogenesis and in particular

of the potential neurological complications

and cause and recognition thereof.

References
1. Sloan MA, Stern BJ. Cerebrovascular disease in

pregnancy. Current Treatment Options in
Neurology 2003; 5: 391-407.

2. McMinn JR, George JN. Evaluation of women with
clinically suspected thrombotic thrombocy-
topoenic purpura - hemolytic uremic syndrome. J
Clin Apheresis 2001; 16: 202-209.

3. Knopp U, Kehler U, Rickmann H, Arnold H,
Gilemroth J. Cerebral haemodynamic pathologies
in HELLP syndrome. Clin Neurol Neurosurg 2003;
105: 256-261.

4. Lipstein H, Lee CC, Crupi RS. A current concept of
eclampsia. Am J Emerg Med 2003; 21: 223-226.

5. George JN. The association of pregnancy with
thrombotic thrombocytopoeinic purpura -
hemolytic uremic syndrome. Curr Opin Hematol
2003; 10: 339-344.

6. Kakishita E. Pathophysiology and treatment of
thrombotic thrombocytopoenic purpura/
hemolytic uremic syndrome (TTP/HUS). Int J
Hematol 2000; 71: 320-327.

7. Bartynski WS, Sanghaui A. Neuroimaging of
delayed eclampsia. J Comput Assist Tomogr 2003;
27: 699-713.

8. Gilemroth J, Knopp U, Kehler U, Felberbaum R,
Nowak G. HELLP syndrome with hemoglobin
vasospasm. J Clin Neurosci 2000; 7: 58-62.

9. Vesely SK, George JN, Lemmie B, et al. ADAMTS-
13 activity in thrombotic thrombocytopoenic pur-
pura - haemolytic uraemic syndrome: relation to
presenting features and clinical outcomes in a
prospective cohort of 142 patients. Blood 2003;
102: 60-68.

10. Lattuada A, Rossi E, Calzarossa C, Candolfi R,
Marinucci PM. Mild to moderate reduction of Von
Willebrand factor cleaving protease (ADAMTS-
13) in pregnant women with HELLP microangio-
pathic syndrome. Haematologica 2003; 88: 1029-
1034.

11. Qureshi BL, Frankel MR, Ottenlips JR, Stern BJ.
Cerebral vasospasm and eclampsia. Stroke 1988;
19: 326-329.

12. Trommer BL, Homer D, Mikhael MA. Cerebral
vasospasm and eclampsia. Stroke 1988; 19: 326-
329.

13. Geraghty JJ, Hoch DB, Robert ME, Vinters HV.
Fatal puerperal cerebral vasospasm and stroke in a
young woman. Neurology 1991; 41: 1145-1147.

14. Sengar AR, Gupta RK, Dhanuka AK, Das K. MR
imaging, MR angiography and MR spectroscopy of
the brain in eclampsia. Am J Neuroradiol 1997; 18:

1485-1490.

15. Lewis LK, Hinshow DB, Will AD, Hasso AN,
Thompson JR. CT and angiographic correlation of
severe neurological disease in toxemia of pregnan-
cy. Neuroradiology 1988; 30: 59-64.

16. Ito T, Sakai T, Inagawa S, Utsu M, Bun T. MR
angiography of cerebral vasospasm in pre-eclamp-
sia. Am J Neuroradiol 1995; 16: 1344-1346.

17. Garner BF, Burns P, Bunning RD, Laureno R. Acute
blood pressure elevation can mimic arteriographic
appearance of cerebral vasculitis. J Rheumatol
1990; 17: 93-97.

18. Ursell MR, Marras CL, Farb R, Rowed DW, Black
SE, Perry JR. Recurrent intracranial hemorrhage
due to postpartum cerebral angiopathy. Stroke
1998; 29: 1995-1998.

19. Dahmus MA, Barton JR, Sibai BM. Cerebral imag-
ing in eclampsia: Magnetic resonance imaging ver-
sus computed tomography. Am J Obstet Gynecol
1992; 157: 935-941.

20. Raroque HG, Orrison WW, Rosenberg GA.
Neurologic involvement in toxemia of pregnancy:
Reversible MRI lesions. Neurology 1990; 40: 167-
169.

21. Hinchey J, Chaves C, Appignani B, et al. A
reversible posterior leukoecephalopathy syn-
drome. N Engl J Med 1996; 334: 494-500.

22. Provenzale JM, Petrella JR, Cruz LCH, Wong JC,
Engelter S, Barboriak DP. Quantitative assessment
of diffusion abnormalities in posterior reversible
encephalopathy syndrome. Am J Neuroradiol 2002;
23: 1038-1048.

23. Covarrubias DJ, Luetmer PH, Campeau NG.
Posterior reversible encephalopathy syndrome:
Prognostic utility of quantitative diffusion-weight-
ed MR images. Am J Neuroradiol 2002; 23: 1038-
1048.

24. Mantello MT, Schwartz RB, Jones KM, Ahn SS,
Tice HM. Imaging of neurologic complications
associated with pregnancy. Am J Roentgenol 1993;
150: 843-847.

25. Naidu S, Payne AJ, Moodley J, Hoffmann M,
Gouws E. Randomised study assessing the effect of
phenyton and magnesium sulphate on maternal
cerbebral circulation in eclampsia using transcra-
nial doppler ultrasound. Br J Obstet Gynaecol 1996;
103: 111-116.

26. Belfort MA, Anthony J, Saade GR, Allen JC. A com-
parison of magnesium sulfate and nimodepine for
the prevention of eclampsia. N Engl J Med 2003;
348: 304-311.

27. Martin JN jun., Thigpen BD, Rose CH, Cushman J,
Moore A, May WL. Maternal benefit of high-dose
intravenous corticosteroid therapy for HELLP syn-
drome. Am J Obstet Gynecol 2003; 189: 830-834.

28. Ringer AJ, Quereshi AI, Kim S, Fessler RD,
Guterman LR, Hopkins LN. Angioplasty for cere-
bral vasospasm from eclampsia. Surg Neurol 2001;
56: 373-379.

pg9-12  9/28/05  1:17 PM  Page 12