Content uploaded by Brian J Holloway
Author content
All content in this area was uploaded by Brian J Holloway on Jan 25, 2019
Content may be subject to copyright.
PICTORIAL REVIEW
The radiological manifestations of sickle
cell disease
G. Madani
a,
*, A.M. Papadopoulou
a
, B. Holloway
a
, A. Robins
b
,
J. Davis
c
, D. Murray
c
a
Department of Radiology, Royal Free Hospital NHS Trust, London, UK, Departments of
b
Paediatrics, and
c
Radiology, Whittington Hospital NHS Trust, London, UK
Received 7 July 2006; received in revised form 20 December 2006; accepted 2 January 2007
Sickle cell disease (SCD) is an inherited abnormality of the ß-globin chain, which causes a spectrum of haemolytic
anaemias. Clinical manifestations in SCD include anaemia, jaundice, recurrent vaso-occlusive crises, and infections
(particularly by encapsulated bacteria) due to functional asplenia and cerebrovascular accidents. Radiological inves-
tigations play a critical role both in the diagnosis and in the primary prevention of the complications of SCD.
ª2007 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction
Sickle cell disease (SCD) is an inherited abnorma-
lity of the ß-globin chain resulting in a spectrum
of haemolytic anaemias. The most common type of
SCD is sickle cell anaemia (SCA). Other types of SCD
are caused by combinations of haemoglobin (Hb) S
with Hb C or ß-thalassaemia.
As the heterozygous carriage of the sickle ß-
globin gene affords a survival advantage against
Plasmodium falciparum infection, Hb S is preva-
lent in areas where malaria is endemic: Africa,
the Middle East, Mediterranean countries, and India.
The slave trade carried SCD to North America, the
Caribbean, Central America, and a few countries
of South America. More recent migration from
Africa and the Caribbean has brought it to the
British Isles.
The clinical manifestations of SCD result from
the intracellular polymerization of the Hb S mole-
cule. In hypoxic conditions, the Hb S molecule is
deoxygenated causing polymerization and resulting
in loss of erythrocyte flexibility. Repeated cycles of
oxygenation and deoxygenation cause irreversible
membrane damage and formation of sickled cells.
The sickled erythrocytes are less deformable, thus
resulting in microvascular occlusion and haemo-
lytic anaemia, which are typical of the disease. The
exact mechanism of vaso-occlusion has not been
fully elucidated; many factors are implicated
including adhesion molecules, endothelial cell
abnormalities, cytokines, and coagulation factors.
The most common clinical manifestations in SCD
are anaemia, jaundice, recurrent vaso-occlusive
crises, and infections by encapsulated bacteria,
such as Streptococcus pneumoniae. This increased
susceptibility to bacterial infections is primarily
due to organic or functional asplenia.
Fetal haemoglobin (Hb F) normally accounts for
around 80% of the haemoglobin concentration at
birth, usually reducing to less than 1% by the age of
6 months and gives protection against the compli-
cations of SCD over the first few months of life.
1
Furthermore Hb F inteferes with Hb S polymeriza-
tion and an increase in serum Hb F concentration
in children and adults with SCD, directly correlates
with a reduction in severity of disease and improve-
ment of prognosis.
2
Hydroxyurea, which increases
Hb F concentration, reduces the incidence and
ameliorates the severity of painful crises in SCD.
3
* Guarantor and correspondent:G.Madani,Departmentof
Radiology, Royal Free Hospital NHS Trust, Pond Street, London
NW3 2QG, UK. Tel.: þ44 20 7794 0500x1467.
E-mail address: gittamadani@yahoo.com (G. Madani).
0009-9260/$ - see front matter ª2007 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.crad.2007.01.006
Clinical Radiology (2007) 62, 528e538
We review the radiological manifestations of
SCD and the role of imaging in the diagnosis and
prevention of its complications.
The lungs
Pulmonary complications are the most common
cause of death in SCD and a frequent cause of
hospital admission. Acute insults include pneumo-
nia and acute chest syndrome (ACS). Children with
SCD are 100 times more likely than the general
population to develop pneumonia.
4
Although the
radiographic manifestations are described below,
it is important to note that a normal chest radio-
graph does not exclude incipient ACS. ACS is
defined as a new focus of opacity on a chest radio-
graph, associated with a variable combination of
signs and symptoms including fever, leucocytosis,
hypoxia, and chest pain.
5
The reported incidence
of infection in ACS varies widely from 2e56%.
6e9
An infective aetiology is more common in children
who suffer a milder course of illness.
3
ACS affect-
ing adults has a more fulminant course and is
more likely to be fatal.
6,9
As infection is difficult
to exclude and given the increased susceptibility
to infection, ACS is nearly always treated with
antibiotics. The other underlying causes of ACS in-
clude infarction (due to microvascular occlusion)
or fat embolism.
10,11
Hypoventilation, pulmonary
oedema and thromboembolism are thought to in-
fluence the development of ACS.
12
Results from
the multicentre study of hydroxyurea in SCD,
show a reduction in ACS in patients treated with
hydroxyurea, suggesting that a substantial number
of cases of ACS are caused by vaso-occlusion.
2
The
relationship of these risk factors has yet to be es-
tablished and the wide variation in their reported
prevalence may reflect the multifactorial and
complex aetiology.
The plain film features of ACS are a new a focus
of consolidation (which may segmental, lobar or
multilobar) or collapse with or without a pleural
effusion (Fig. 1). High-resolution computed tomo-
graphy (CT) is more sensitive and often reveals
ground-glass density, which does not have a lobar
distribution and may have a scattered or mosaic
pattern. Consolidation is a less common finding.
11
The upper lobes are more frequently affected in
children: adults suffer from lower lobe or multilo-
bar ACS.
11
Microvascular occlusion causes a reduc-
tion in the vascular markings and infarction results
in linear scarring.
10,11
Repeated episodes of ACS
may lead to chronic interstitial lung disease and
pulmonary hypertension.
11,13
Interstitial lung dis-
ease manifests as interlobular septal thickening
(mainly in a basal distribution), parenchymal
bands, pleural tags, dilated secondary pulmonary
lobules, traction bronchiectasis and architectural
distortion
13
(Fig. 2). Pulmonary hypertension may
cause mosaic perfusion. Notably, there is no corre-
lation between the severity of ACS and the extent
of radiological findings.
11
However, there is a posi-
tive correlation between the severity of interstitial
lung disease and the number of episodes of ACS.
13
The presence of pulmonary hypertension is
associated with an increase in mortality.
14
Figure 1 Frontal chest radiograph in a 7-year-old girl
with SCD demonstrating cardiomegaly, left lower lobe
consolidation and osteonecrosis of the right humeral
head.
Figure 2 High-resolution CT of the chest in an adult
female with SCD. There is subpleural scarring with pa-
renchymal bands due to previous vaso-occlusive infarcts.
Radiological manifestations of SCD 529
Increasing age, markers of haemolytic anaemia,
and a history of renal and cardiovascular disease
and priapism are independent predictors for the
development of pulmonary hypertension. Chronic
thromboembolism is an important cause of pulmo-
nary hypertension and its early diagnosis and treat-
ment is critical.
14
The brain
Common neurological complications of SCD are
infarction, atrophy and cognitive impairment. Cra-
nial vasculopathy predisposes to infarcts, which
may be silent or have clinical manifestations.
15
The
largest randomized controlled trial to date, the
Cooperative Study of Sickle Cell Disease (CSSCD),
shows a 22% prevalence of ischaemia, atrophy or
infarction at 8 years of age.
16
Using standard mag-
netic resonance imaging (MRI) and MR angiography
(MRA), a more recent study showed a 44% preva-
lence of infarction, ischaemia and atrophy, and
a 55% prevalence of vasculopathy in children with
SCD.
17
Combining the MRI and MRA findings, only
32% of patients over 3-years-old had normal brain
imaging. The high prevalence of abnormalities,
compared with previous studies, was attributed
to improvement in imaging techniques.
Clinical infarcts (strokes) may present during
vaso-occlusive or sequestration crises but most
commonly occur as an isolated event.
18
Eleven
percent of patients with SCA have suffered from
a stroke by the age of 20 years.
15
Pergelow
et al.
19
found a 22% prevalence of silent infarcts
in children with SCD. Their presence predisposes
the patient to new silent infarcts and is the stron-
gest independent predictor of a future infarctive
stroke.
19,20
Infarctive strokes are more common
in children, whereas haemorrhagic strokes are
more frequent in adults; more than 90% of first
strokes in patients younger than 20 years are in-
farctive, whereas 52% of first strokes in adults
over 20 years are haemorrhagic.
16
Strokes and silent infarcts manifest as high-
signal intensity on T2-weighted and fluid attenua-
tion inversion recovery (FLAIR) sequences. Silent
infarcts are usually confined to the deep white
matter (usually at arterial border zones) whereas
strokes are more likely to affect the cortex and the
white matter and are larger (75% are larger than
1.5 cm in diameter).
16,21
The most common site of
stroke and silent infarct is the frontal and parietal
lobes.
6
However, children with a history of stroke
are more likely to also have involvement of the
basal ganglia and thalami (69%) than those with
silent infarcts (11%)
18
(Fig. 3).
Moyamoya is Japanese for a ‘‘hazy puff of
smoke’’. Moyamoya syndrome describes an angio-
graphic pattern of large vessel occlusion and
telangiectatic collateral circulation (Fig. 4), seen
in patients with SCD, as well as other conditions,
such as tuberose sclerosis, neurofibromatosis, vas-
culitides and infection.
22
On conventional MRI the
clues to the presence of Moyamoya syndrome are
the absence of flow voids in the occluded arteries
(usually the middle cerebral or internal carotid)
and the presence of flow voids within collateral
vessels in the territory of the occluded artery. In
a retrospective study of 43 patients with SCD,
who had suffered from clinical stroke before 18
years of age, 41% of patients had subsequent cere-
brovascular events. The presence of Moyamoya
collaterals increased the risk of recurrent stroke
or transient ischaemic attacks by a factor of two,
despite transfusion therapy.
22
Similarly, Moyamoya
syndrome in non-sickle patients is a predictor of
recurrent stroke.
22
The pathogenesis of stroke in SCD involves small
and large vessel disease. Brain injury can be
Figure 3 Mature infarct in a 3-year-old child who pre-
sented with a clinical stroke. Axial T2-weighed image of
the head showing high signal within the anterior striatal
tissue and the anterior limb of the internal capsule on
the left (arrow) as well as dilatation of the frontal
horn of the left lateral ventricle. The findings are in
keeping with a mature end-artery territory infarct.
530 G. Madani et al.
associated with vasculopathy in vessels too small
to be visualized by MRA.
23
Cerebral angiography is
indicated in cases of haemorrhagic stroke, in order
to rule out aneurysms.
23
Cerebral aneurysms
affecting the circle of Willis are more common in
older children with SCD.
15
They are frequently
multiple (57%) and up to 30% affect the posterior
circulation.
24
The haemorrhage is usually sub-
arachnoid, but as with the non-sickle population,
may be intraventricular or parenchymal. Non-
invasive imaging techniques have replaced conven-
tional angiography for the diagnosis of Moyamoya
syndrome and suspected vascular stenoses and
occlusions, which tend to affect the internal
carotid and middle cerebral arteries. The vertebro-
basilar and the extracranial carotid systems are
usually spared.
15
In a study of 21 patients with
SCD, the accuracy of MRA in the diagnosis of steno-
sis or occlusion was 85% of that of conventional
angiography (12/13).
25
Stenoses exceeding 6 mm,
which are also associated with a reduction of dis-
tal flow, correlate with subclinical infarction.
26
Conversely, short-segment abnormalities, which
commonly occur where vessels branch, are not
associated with MRI evidence of infarction.
26
In
a prospective study of 107 SCD patients with cere-
bral infarcts, 27 of whom had clinical strokes, the
combination of MRA and transcranial Doppler
abnormalities best predicted future stroke.
27
The
sensitivity of CT angiography appears to be similar
to MRA angiography in the non-SCD population.
28
However, there are no studies using this technique
in patients with SCD.
Transcranial Doppler (TCD) and TCD sonography
can detect stenoses in the circle of Willis by
identifying focal areas of high-velocity blood
flow.
15
Although it has been long established that
maintenance of HbS levels below 30% reduces the
incidence of first and recurrent strokes, the Stroke
Prevention Trial in Sickle Cell Anaemia (STOP)
study demonstrated the role of TCD in primary pre-
vention; 130 children with abnormal TCD results
(time-averaged velocity of greater than 200 cm/s
in the internal carotid or middle cerebral arteries)
were randomized to standard care versus transfu-
sion therapy. Transfusion greatly reduced the inci-
dence of stroke in this population and the trial was
terminated early.
23,29e31
The musculoskeletal system
Red marrow is normally present throughout the
skeleton at birth. In the healthy subject, red
marrow regresses centripetally and over the first
decade of life is replaced by fatty marrow in the
peripheral skeleton. The epiphyses contain fatty
marrow throughout life. In patients with SCD,
anaemia causes increased haematopoiesis result-
ing in persistence of, or reconversion to, red
marrow in the peripheral skeleton and possible
extension to the epiphyses. When anaemia is
severe bone marrow expansion (resulting in the
classic ‘‘hair on end appearance’’ in the skull;
Fig. 5) and extramedullary haematopoiesis may
also be observed. Extramedullary haematopoiesis
most commonly occurs in the paraspinal location
but may affect any organ containing pluripotent
stem cells including the liver, spleen, adrenal
glands, skin and breasts. When extramedullary
haematopoiesis presents as a focal mass T-99
sulphur colloid will confirm the diagnosis.
32
The
growth retardation observed in SCD is multi-
factorial but is in part attributable to marrow
hyperplasia.
33
Persistence of red marrow results in increased
susceptibility to osteomyelitis and bone marrow
infarction, as well as making the detection of
these entities more difficult
34
(Fig. 6). Infarction
may affect any marrow-containing bone, most fre-
quently presenting as a painful crisis (but may be
clinically silent) and is 50 times more common
than osteomyelitis in the paediatric SCD popula-
tion.
35
The presumed mechanism is marrow con-
gestion causing impedence of blood flow leading
to local hypoxia and sickling. Osteomyelitis most
commonly affects the diaphyses of the femur, tibia
or humerus (whereas the metaphyseal location is
more common in the non-sickle population).
36
Most studies state that salmonella is the most fre-
quent causative organism and is two to five times
more common than Staphylococcus aureus (which
Figure 4 Moyamoya in an adult with SCD. Magnetic
resonance angiogram of the circle of Willis showing ir-
regular and stenosed middle cerebral arteries bilaterally
with multiple surrounding collateral vessels giving the
‘‘puff of smoke’’ appearance (arrows).
Radiological manifestations of SCD 531
is the leading cause of osteomyelitis in the non-
SCD population).
32,37
Untreated osteomyelitis
may extend into the joint resulting in septic arthri-
tis. Septic arthritis may also arise de novo.
The clinical presentation of acute infarction and
osteomyelitis is similar. Radiological differentia-
tion of these conditions also presents a challenge.
Initially both conditions are radiographically
normal. The plain film features of periostitis and
an ill-defined radiolucency subsequently occur in
both infarction and osteomyelitis. Over a period of
months, sclerosis supervenes in both conditions.
Subperiosteal collections, seen on CT imaging,
were once thought to be diagnostic of osteo-
myelitis, but have also been demonstrated in
infarction.
38
Both osteomyelitis (Fig. 7) and infarc-
tion exhibit high T2-weighted signal intensity (as
does persistent red marrow which may further
confound the diagnosis). Periosteal and paraosteal
soft-tissue enhancement cannot differentiate
between these conditions unless communication
between the enhancing medullary cavity and the
enhancing soft tissue via a cortical defect is dem-
onstrated, which is diagnostic of osteomyelitis.
39
The pattern of enhancement on fat-suppressed
T1-weighted gadolinium-enhanced sequences
often provides the most useful clue to the differ-
ential diagnosis; infarcts demonstrate rim en-
hancement whereas osteomyelitis is more likely
Figure 6 Sagittal T1-weighted MRI image of the lum-
bar spine demonstrating heterogeneous low signal
throughout the marrow due to red marrow reconversion
(which could hamper the detectionofinfarctionor
infection). There is central depression of the vertebral
endplates (H-shaped vertebrae; arrow).
Figure 5 Marrow expansion of the skull in a 20-year-old
female. (a) Lateral skull radiograph shows a striated
‘‘hair-on-end’’ appearance of the skull due to diploic
space widening and trabercular prominence. (b) Delayed
Tc99m bone scintigraphy image of the same patient
demonstrating increased tracer uptake in the calvarium
due to marrow expansion.
532 G. Madani et al.
to exhibit irregular and geographic enhancement
of the infected marrow.
39
Chronic infarcts exhibit
low signal intensity on all sequences due to fibrosis
and sclerosis.
32
Although infarction can affect any marrow-
containing bone (Fig. 8), three characteristic pat-
terns are seen in SCD. Infarction and subsequent
central depression of the vertebral endplates re-
sults in H-shaped vertebrae, which are characteris-
tic of SCD.
10
(Fig. 9) H-shaped vertebrae may be
associated with increase in the height of the adja-
cent vertebra described as a tower vertebra.
40
Dactylitis is due to infarction of the bone marrow,
medullary trabeculae and inner cortical layer of
the bones in the hands and feet, and usually occurs
in children younger than 4 years old; it is fre-
quently the first clinical manifestation of SCD.
41
Early radiographs are normal. After around 10
days, periostitis, subperiosteal new bone, cortical
thinning and moth-eaten osteopenia are observed
on plain films (Figs. 10 and 11).
Epiphyseal infarction is more commonly called
osteonecrosis or avascular necrosis and tends to
affect the proximal femur and proximal humerus.
SCD is the most frequent cause of osteonecrosis in
children; around half of all patients with SCD have
suffered osteonecrosis of the femoral head by the
age of 35 years.
32
As with the non-sickle popula-
tion early radiographs are normal. Subsequent ra-
diographs show osteopenia and small subchondral
lucencies, which may coalesce forming a crescentic
luceny leading to flattening of the articular bone
and secondary osteoarthritis (Fig. 12). On T2-
weighted MRI sequences osteonecrosis may result
in the ‘‘double line’’ sign with an outer low-signal
intensity new bone formation and inner high-signal
intensity granulation tissue, which outline the
central infarcted marrow (Fig. 13).
Bone marrow infarcts in the skull may cause
a subdural haematoma.
42
Muscular infarction oc-
curs occasionally in SCD and may be demonstrated
with CT or MRI.
43
Leg ulcers are also seen and may
be associated with an underlying periosteal
reaction. Other less frequent musculoskeletal
complications of SCD include gout arthropathy,
impingement syndrome, stress fracture, subtalar
fusion, recurrent dislocation of the patella and
tibialis anterior tendonitis.
44
The genitourinary system
Renal manifestations of SCD include haematuria
(usually self-limiting), proteinuria, nephrotic
syndrome, as well as acute and chronic renal
failure.
45
Around half of patients with SCD have
large kidneys, which are attributed to increased
renal blood volume, glomerular hypertrophy, and
Figure 7 Osteomyelitis of the innominate bone in
a 12-year-old male patient. Coronal T2-weighted image
of the pelvis demonstrating a focus of high signal inten-
sity within the right femoral head as well as diffuse
increased signal intensity within the right innominate
bone (arrow) communicating with a loculated fluid col-
lection (arrowhead) in the adjacent soft tissues in keep-
ing with osteomyelitis and adjacent soft tissue abscess
formation. Bilateral joint effusions are also present.
Figure 8 Osteonecrosis of the humeral head in an
adult female patient. Coronal T1-weighted image of
the left shoulder showing a subchondral serpiginous
low signal area in the humeral head indicative of osteo-
necrosis (arrow). The high signal at the lateral aspect of
the humeral head represents fat.
Radiological manifestations of SCD 533
engorgement.
46,47
The prevalance of renal papil-
lary necrosis, which is often asymptomatic, may
be as high as 40%.
47
Although, the echo character-
istics of the kidneys is normal in the majority of
SCD patients, a generalized increase in echogenic-
ity (attributed to diffuse fibrosis) and an increase
in medullary echogenicity (attributed to nephro-
calcinosis or iron deposition) is seen in 3% and 5%
of patients, respectively.
48
Renal infarction secondary to microvascular
occlusion is well-recognized and may be associated
with perirenal haematoma.
49
Renal medullary car-
cinoma is a rare aggressive tumour associated with
sickle cell trait (but has never been reported in
homozygous SCD) and most frequently presents
with massive haematuria.
50
Priapism, due to occlusion of penile venous
drainage, affects 89% of male patients by the age
of 20 years.
32
MRI has been used in the evaluation
of priapism and erectile dysfunction, demonstrat-
ing intracorporeal fibrosis and haemosiderosis.
51
Testicular infarction in SCD has also been
described.
52
The spleen
The spleen is nearly always involved in SCD.
Microinfarcts result in loss of function, termed
hypo or asplenism; by the age of 5 years 94% of
patients are functionally asplenic resulting in in-
creased susceptibility to infection by encapsulated
bacteria.
53
Radiologically the spleen is present but
small and may be calcified. In sequestration
syndrome there is rapid pooling of a large volume
of blood in a solid organ (nearly always affecting
Figure 9 Lateral lumbar spine radiograph demon-
strating coarsened traberculation and square compres-
sion infarcts of the vertebral endplates, resulting in
H-shaped vertebrae, which are characteristic of SCD.
Figure 10 Dactylitis (handefoot syndrome) in an
18-month-old female patient. (a) Frontal radiograph of
the hands demonstrating periosteal reaction along the
second metacarpals (an early feature of infarction), as
well as soft tissue swelling. (b) Frontal radiograph of
the feet demonstrating periosteal reaction along the
third and fourth metatarsal shafts on the right and the
third on the left with associated soft-tissue swelling.
534 G. Madani et al.
the spleen but rarely the liver) causing organome-
galy and occasionally rupture. Splenic sequestra-
tion affects 10e30% of children with SCD usually
between the ages of 6 months and 3 years and
may recur.
53
The enlarged spleen appears hetero-
geneous on ultrasound and contrast CT imaging.
32
The spleen may also be enlarged due to extrame-
dullary haematopoiesis or haemosiderosis. The
coexistence of haemorrhagic sequestration and
haemosiderin deposition result in variable and po-
tentially confusing imaging appearances.
54
Splenic
abscess is a recognized complication of SCD. Is-
lands of functioning splenic tissue may be present
in a small spleen and are hypoechoic on ultrasound
and low in density on CT imaging; these may there-
fore be misdiagnosed as splenic abscess.
55
Resolu-
tion of splenic infarct or abscess may lead to
formation of pseudocysts.
The liver
Hepatic complications of SCD include fibrosis,
cirrhosis, and intrahepatic biliary duct stenoses
resulting in cholestasis (all believed to be due to
infarction), as well as hepatitis secondary to re-
peated transfusions. Pigmented gallstones affect
50% of adults and up to 20% of children with SCA.
56
Liver involvement is reported in 10e39% of
patients presenting with acute vaso-occlusive cri-
ses.
57,58
Acute hepatic sequestration is a rare com-
plication, accounting for less than 1% of acute
admissions due to vaso-occlusive crisis.
59
Iron overload
Repeated transfusions may cause haemosiderosis,
which manifests as reduced marrow signal intensity
on all MRI sequences.
60
The spleen and renal corti-
ces of non-transfusion-dependant patients may
exhibit a reduction in signal intensity, although
these patients do not have an increase in total
body iron (Fig. 14).
61
Conversely, haemosiderosis
of the liver and pancreas occurs in transfusion-
dependant patients in spite of iron chelation.
61
Miscellaneous complications
Ophthalmic complications of SCD include acute
retinal artery occlusion and proliferative retinon-
opathy. Infarcts of the bones of the orbit may
result in orbital haemorrhage causing orbital com-
pression syndrome.
62
CT imaging will identify the
Figure 11 Frontalradiographofthehandsina15-
year-old female patient showing the effects of previous
dactylitis with premature epiphyseal fusion, chevron
metaphyses and shortened right third and fourth meta-
carpals and proximal phalanges of the left middle and
little fingers. There is also minor shortening of the right
fifth metacarpal and proximal phalanx of the right little
finger.
Figure 12 Osteonecrosis of the femoral head in a
12-year-old male patient. (a) Frontal radiograph of the
pelvis shows advanced right femoral head osteonecrosis
with collapse of the articular surface. A recent acetabu-
lar osteotomy has been performed. (b) Fifteen months
later a pelvic radiograph demonstrates osteonecrosis of
the contralateral femoral head with subchondral
lucency and irregularity of the articular surface.
Radiological manifestations of SCD 535
collection but cannot exclude an orbital abscess.
MRI identifies the infarct and differentiates
between haemorrhagic and non-haemorrhagic
collections thus avoiding unnecessary surgical
intervention.
63
An association between SCD and sensorineural
hearing loss (SNHL) is well documented but the
reported prevalence varies widely.
64,65
SNHL is
hypothesized to be secondary to vaso-occlusive
phenomena. Whitehead et al. reported two cases
of labyrinthine haemorrhage (manifesting as high
signal intensity on T1-weighted MRI), which re-
sulted in SNHL and vestibular symptoms.
66
Cardiomegaly due to congestive cardiac failure
is a common finding. Myocardial ischaemia (which
may be irreversible on myocardial perfusion scin-
tigraphy) is also associated.
67
Conclusion
Radiology plays a crucial role in the management of
the complications of SCD. Judicious use of imaging
in the complications of SCD aids diagnosis and may
prevent unnecessary surgical intervention.
Acknowledgements
The authors thank the consultant radiologists and
haematology staff at the Whittington Hospital for
their invaluable assistance.
References
1. Murphy MF. Disease of the blood. In: Kumar P, Clark M,
editors. Clinical Medicine. 3rd edn. London: Balliere Tindall;
1994. p. 293e351.
2. Platt OS, Thorington BD, Brambilla DJ, et al. Pain in sickle
cell disease. Rates and risk factors. N Engl J Med 1991;
325:11e6.
3. Charache S, Terrin ML, Moore RD, et al. Effect of hydroxy-
urea on the frequency of painful crises in sickle cell anemia.
Investigators of the multicenter study of hydroxyurea in
sickle cell disease. N Engl J Med 1995;332:1317e22.
4. Stark P, Pfeiffer WR. Intrathoracic manifestations of sickle
cell disease. Radiologe 1985;25:33e5.
5. Charache S, Scott JC, Charache P. ‘‘Acute chest syndrome’’
in adults with sickle cell anemia. Microbiology, treatment,
and prevention. Arch Intern Med 1979;139:67e9.
6. Vichinsky EP, Styles LA, Colangelo LH, et al. Acute chest
syndrome in sickle cell disease: clinical presentation and
course. Cooperative Study of Sickle Cell Disease. Blood
1997;89:1787e92.
7. Martin L, Buonomo C. Acute chest syndrome of sickle cell
disease: radiographic and clinical analysis of 70 cases.
Pediatr Radiol 1997;27:637e41.
8. Quinn CT, Buchanan GR. The acute chest syndrome of sickle
cell disease. J Pediatr 1999;135:416e22.
9. Vichinsky EP, Neumayr LD, Earles AN, et al. Causes and out-
comes of the acute chest syndrome in sickle cell disease.
National Acute Chest Syndrome Study Group. N Engl J Med
2000;342:1855e65.
10. Leong CS, Stark P. Thoracic manifestations of sickle cell
disease. J Thorac Imaging 1998;13:128e34.
11. Bhalla M, Abboud MR, McLoud TC, et al. Acute chest
syndrome in sickle cell disease: CT evidence of microvascu-
lar occlusion. Radiology 1993;187:45e9.
Figure 13 Osteonecrosis of the femoral heads in a
10-year-old female patient. Coronal short tau inversion
recovery image of the pelvis showing heterogeneous
signal intensity within the right femoral epiphysis and
more advanced changes in the left with flattening of
the femoral articular surface, the ‘‘double line’’ sign
(arrow) of low signal intensity new bone formation,
and adjacent high signal intensity granulation tissue as
well as bilateral reactive joint effusions (arrowheads),
which resolved on follow-up imaging.
Figure 14 Iron deposition in an adult female with SCD.
T2-weighted MRI image shows low signal within the renal
cortex due to iron deposition (arrowheads). In addition,
there is hepatomegaly of uncertain aetiology and splenic
atrophy.
536 G. Madani et al.
12. Castro O, Brambilla DJ, Thorington B, et al. The acute chest
syndrome in sickle cell disease: incidence and risk factors.
Blood 1994;84:643e9.
13. Aquino SL, Gamsu G, Fahy JV, et al. Chronic pulmonary
disorders in sickle cell disease: findings at thin-section CT.
Radiology 1994;193:807e11.
14. Machado RF, Gladwin MT. Chronic sickle cell lung disease:
new insights into the diagnosis, pathogenesis and treatment
of pulmonary hypertension. Br J Haematol 2005;129(4):
449e64.
15. Adams RJ, Ohene-Frempong K, Wang W. Sickle cell and the
brain. Hematology Am Soc Hematol Educ Program 2001:
31e46.
16. Moser FG, Miller ST, Bello JA, et al. The spectrum of brain
MR abnormalities in sickle-cell disease: a report from the
Cooperative Study of Sickle Cell Disease. AJNR Am J Neuro-
radiol 1996;17:965e72.
17. Steen RG, Emudianughe T, Hankins GM, et al. Brain imaging
findings in pediatric patients with sickle cell disease. Radio-
logy 2003;228:216e25.
18. Moran CJ, Siegel MJ, DeBaun MR. Sickle cell disease: imag-
ing of cerebrovascular complications. Radiology 1998;206:
311e21.
19. Pegelow CH, Macklin EA, Moser FG, et al. Longitudinal
changes in brain magnetic resonance imaging findings
in children with sickle cell disease. Blood 2002;99:
3014e8.
20. Miller ST, Macklin EA, Pegelow CH, et al. Cooperative Study
of Sickle Cell Disease. Silent infarction as a risk factor for
overt stroke in children with sickle cell anemia: a report
from the Cooperative Study of Sickle Cell Disease. J Pediatr
2001;139:385e90.
21. Hoppe C. Defining stroke risk in children with sickle cell
anaemia. Br J Haematol 2005;128:751e66.
22. Dobson SR, Holden KR, Nietert PJ, et al. Moyamoya
syndrome in childhood sickle cell disease: a predictive fac-
tor for recurrent cerebrovascular events. Blood 2002;99:
3144e50.
23. Adams RJ, McKie VC, Hsu L, et al. Prevention of a first
stroke by transfusions in children with sickle cell anemia
and abnormal results on transcranial Doppler ultrasonogra-
phy. N Engl J Med 1998;339:5e11.
24. Preul MC, Cendes F, Just N, et al. Intracranial aneurysms
and sickle cell anemia: multiplicity and propensity for the
vertebrobasilar territory. Neurosurgery 1998;42:971e7.
25. Kandeel AY, Zimmerman RA, Ohene-Frempong K. Compari-
son of magnetic resonance angiography and conventional
angiography in sickle cell disease: clinical significance and
reliability. Neuroradiology 1996;38:409e16.
26. Gillams AR, McMahon L, Weinberg G, et al. MRA of the intra-
cranial circulation in asymptomatic patients with sickle cell
disease. Pediatr Radiol 1998;28:283e7.
27. Seibert JJ, Glasier CM, Kirby RS, et al. Transcranial Doppler,
MRA, and MRI as a screening examination for cerebrovascu-
lar disease in patients with sickle cell anemia: an 8-year
study. Pediatr Radiol 1998;28:138e42 [Erratum in: Pediatr
Radiol 1998;28:546].
28. Katz DA, Marks MP, Napel SA, et al. Circle of Willis: evalua-
tion with spiral CT angiography, MR angiography, and
conventional angiography. Radiology 1995;195:445e9.
29. Pegelow CH, Adams RJ, McKie V, et al. Risk of recurrent
stroke in patients with sickle cell disease treated with
erythrocyte transfusions. J Pediatr 1995;126:896e9.
30. Adams RJ. Lessons from the Stroke Prevention Trial in Sickle
Cell Anemia (STOP) study. J Child Neurol 2000;15:344e9.
31. Adams RJ. TCD in sickle cell disease: an important and
useful test. Pediatr Radiol 2005;35:229e34.
32. Lonergan GJ, Cline DB, Abbondanzo SL. Sickle cell anemia.
RadioGraphics 2001;21:971e94.
33. Almeida A, Roberts I. Bone involvement in sickle cell
disease. Br J Haematol 2005;129:482e90.
34. Rao VM, Mitchell DG, Rifkin MD, et al. Marrow infarction in
sickle cell anemia: correlation with marrow type and distri-
bution by MRI. Magn Reson Imaging 1989;7:39e44.
35. Keeley K, Buchanan GR. Acute infarction of long bones in
children with sickle cell anemia. J Pediatr 1982;101:170e5.
36. Stark JE, Glasier CM, Blasier RD, et al. Osteomyelitis in chil-
dren with sickle cell disease: early diagnosis with contrast-
enhanced CT. Radiology 1991;179:731e3.
37. Atkins BL, Price EH, Tillyer L, et al. Salmonella osteomyeli-
tis in sickle cell disease children in the east end of London.
J Infect 1997;34:133e8.
38. Frush DP, Heyneman LE, Ware RE, et al. MR features of
soft-tissue abnormalities due to acute marrow infarction
in five children with sickle cell disease. AJR Am J Roent-
genol 1999;173:989e93.
39. Umans H, Haramati N, Flusser G. The diagnostic role of
gadolinium enhanced MRI in distinguishing between acute
medullary bone infarct and osteomyelitis. Magn Reson
Imaging 2000;18:255e62.
40. Marlow TJ, Brunson CY, Jackson S, et al. ‘‘Tower vertebra’’:
a new observation in sickle cell disease. Skeletal Radiol
1998;27:195e8.
41. Weinberg AG, Currarino G. Sickle cell dactylitis: histopath-
ologic observations. Am J Clin Pathol 1972;58:518e23.
42. Ng WH, Yeo TT, Seow WT. Non-traumatic spontaneous acute
epidural haematoma dreport of two cases and review of
the literature. J Clin Neurosci 2004;11:791e4.
43. Howlett DC, Hatrick AG, Jarosz JM. The role of CT and MR in
imaging the complications of sickle cell disease. Clin Radiol
1997;52:821e9.
44. Bahebeck J, Atangana R, Techa A, et al. Relative rates and
features of musculoskeletal complications in adult sicklers.
Acta Orthop Belg 2004;70:107e11.
45. Pham PT, Pham PC, Wilkinson AH, et al. Renal abnormalities
in sickle cell disease. Kidney Int 2000;57:1e8.
46. Walker TM, Beardsall K, Thomas PW, et al. Renal length in
sickle cell disease: observations from a cohort study. Clin
Nephrol 1996;46:384e8.
47. Minkin SD, Oh KS, Sanders RC, et al. Urologic manifestations
of sickle hemoglobinopathies. South Med J 1979;72:23e8.
48. Walker TM, Serjeant GR. Increased renal reflectivity in
sickle cell disease: prevalence and characteristics. Clin
Radiol 1995;50:566e9.
49. Sickles EA, Korobkin M. Perirenal hematoma as a complica-
tion of renal infarction in sickle-cell trait. A case report. Am
J Roentgenol Radium Ther Nucl Med 1974;122:800e3.
50. de Santis Feltran L, de Abreu Carvalhaes JT, Sesso R. Renal
complications of sickle cell disease: managing for optimal
outcomes. Paediatr Drugs 2002;4:29e36.
51. Burnett AL, Allen RP, Tempany CM, et al. Evaluation of erec-
tile function in men with sickle cell disease. Urology 1995;
45:657e63.
52. Gofrit ON, Rund D, Shapiro A, et al. Segmental testicular in-
farction due to sickle cell disease. J Urol 1998;160:835e6.
53. Lane PA. Sickle cell disease. Pediatr Clin North Am 1996;43:
639e64.
54. Roshkow JE, Sanders LM. Acute splenic sequestration crisis
in two adults with sickle cell disease: US, CT, and MR imag-
ing findings. Radiology 1990;177:723e5.
55. Levin TL, Berdon WE, Haller JO, et al. Intrasplenic masses
of ‘‘preserved’’ functioning splenic tissue in sickle cell dis-
ease: correlation of imaging findings (CT, ultrasound, MRI,
and nuclear scintigraphy). Pediatr Radiol 1996;26:646e9.
Radiological manifestations of SCD 537
56. Krauss JS, Freant LJ, Lee JR. Gastrointestinal pathology in
sickle cell disease. Ann Clin Lab Sci 1998;28:19e23.
57. Koskinas J, Manesis EK, Zacharakis GH, et al. Liver involve-
ment in acute vaso-occlusive crisis of patients with sickle
cell disease: prevalence and predisposing factors. Scand J
Gastroenterol 2006; doi:10.1080/00365520600988212.
58. Koullapis NG, Kouroupi IG, Dourakis SP. Hepatobiliary man-
ifestations of sickle cell disease. Haema 2005;8:393e404.
59. Davies SC, Brozovic M. Acute admissions of patients with
sickle cell disease who live in Britain. BMJ 1987;294:1206e8.
60. Van Zanten TE, Statius van Eps LW, Golding RP, et al. Imag-
ing the bone marrow with magnetic resonance during a crisis
and in chronic forms of sickle cell disease. Clin Radiol 1989;
40:486e9 [Erratum in: Clin Radiol 1990;41:145].
61. Siegelman ES, Outwater E, Hanau CA, et al. Abdominal iron
distribution in sickle cell disease: MR findings in transfusion
and nontransfusion dependent patients. J Comput Assist
Tomogr 1994;18:63e7.
62. Curran EL, Fleming JC, Rice K, et al. Orbital compression
syndrome in sickle cell disease. Ophthalmology 1997;104:
1610e5.
63. Rebsamen SL, Bilaniuk LT, Granet D, et al. Orbital wall
infarction in sickle cell disease: MR evaluation. AJNR Am J
Neuroradiol 1993;14:777e9.
64. MacDonald CB, Bauer PW, Cox LC, et al. Otologic findings in
a pediatric cohort with sickle cell disease. Int J Pediatr
Otorhinolaryngol 1999;47:23e8.
65. Donegan JO, Lobel JS, Gluckman JL. Otolaryngologic mani-
festations of sickle cell disease. Am J Otolaryngol 1982;3:
141e4.
66. Whitehead RE, MacDonald CB, Melhem ER, et al. Spontane-
ous labyrinthine hemorrhage in sickle cell disease. AJNR Am
J Neuroradiol 1998;19:1437e40.
67. Acar P, Sebahoun S, de Pontual L, et al. Myocardial perfu-
sion in children with sickle cell anaemia. Pediatr Radiol
2000;30:352e4.
538 G. Madani et al.