REVIEWS

Silent cerebral infarcts associated

with cardiac disease and procedures
Mariëlla E. C. Hassell, Robin Nijveldt, Yvo B. W. Roos, Charles B. L. Majoie, Martial Hamon, Jan J. Piek
and Ronak Delewi
Abstract | The occurrence of clinically silent cerebral infarcts (SCIs) in individuals affected by cardiac disease
and after invasive cardiac procedures is frequently reported. Indeed, atrial fibrillation, left ventricular thrombus
formation, cardiomyopathy, and patent foramen ovale have all been associated with SCIs. Furthermore,
postprocedural SCIs have been observed after left cardiac catheterization, transcatheter aortic valve
implantation, CABG surgery, pulmonary vein isolation, and closure of patent foramen ovale. Such SCIs are
often described as precursors to symptomatic stroke and are associated with cognitive decline, dementia, and
depression. Increased recognition of SCIs might advance our understanding of their relationship with heart
disease and invasive cardiac procedures, facilitate further improvement of therapies or techniques aimed at
preventing their occurrence and, therefore, decrease the risk of adverse neurological outcomes. In this Review,
we provide an overview of the occurrence and clinical significance of, and the available diagnostic modalities
for, SCIs related to cardiac disease and associated invasive cardiac procedures.
Hassell, M. E. C. et al. Nat. Rev. Cardiol. 10, 696–706 (2013); published online 29 October 2013; doi:10.1038/nrcardio.2013.162

Introduction
Cardiac disease and related invasive interventions are indicators of cerebral small-vessel diseases, including
potential causes of cerebral embolic events. Cardio­ brain white matter abnormalities and lacunar stroke.11,12
embolic stroke is the subtype of stroke with the highest Indeed, accumulating evidence implicates SCIs in cogni­
in-hospital mortality, approximately 20%.1,2 Improve­ tive decline, dementia, and depression.13–16 Furthermore,
ments in anticoagulation therapy and interventional several studies have demonstrated that SCIs have impor­
cardiology techniques have led to a decrease in the inci­ tant prognostic implications for the risk of future stroke.
dence of acute ischaemic stroke; however, silent cerebral For example, in the Rotterdam Scan Study,17 the pres­
infarcts (SCIs) are increasingly observed and are much ence of SCIs was associated with a more than threefold
Department of
Cardiology
more prevalent than symptomatic stroke.3 increase in the risk of stroke, independently of other risk
(M. E. C. Hassell, SCIs are described as parenchymal lesions that have factors. SCIs can, therefore, be considered precursors of
J. J. Piek, R. Delewi), MRI characteristics of a previous infarct event, but are ischaemic stroke, and potentially enable the identification
Department of
Neurology not associated with acute clinical signs or symptoms of of patients at high risk of this condition. Consequently,
(Y. B. W. Roos), stroke or transient ischaemic attack (TIA).4 Advancements detection of SCIs and treatment of these high-risk patients
and Department
of Neuroradiology in MRI have resulted in increased detection and aware­ might prevent or reduce the large burden of cardio­embolic
(C. B. L. Majoie), ness of such lesions.3 The majority of SCIs are located in stroke. The relationships between cardiac diseases,
Academic Medical
Center, University
the s­ubcortex—the region of the brain directly below the cardiac procedures, SCIs, and neurological c­onditions are
of Amsterdam, c­erebral cortex that contains the thalamus, hypothalamus, s­ummarized in Figure 1.
Meibergdreef 9, cerebellum, and brain stem—and are commonly called In this Review, we introduce the imaging technologies
1105 AZ, Amsterdam,
Netherlands. asymptomatic lacunar infarcts.5 SCIs have been detected that can be used to diagnose SCIs. However, we focus our
Department of in 8–28% of the general population, 6,7 and in 38% of discussion on the association between SCIs and various
Cardiology, VU
University Medical
patients with ischaemic stroke,8,9 with the incidence of SCI cardiovascular diseases and invasive procedures used to
Center Amsterdam, strongly increasing with age.10 treat cardiac diseases. We also comment on the clinical
De Boelelaan 1117, SCIs have been associated with atrial fibrillation (AF), significance of these relationships.
1081 HV, Amsterdam,
Netherlands cardiomyopathies, patent foramen ovale (PFO), catheter­
(R. Nijveldt). Clinical ization, transcatheter aortic valve implant­ation (TAVI), Diagnostic neuroimaging modalities
Research Department,
University Hospital of
CABG surgery, pulmonary vein isolation (PVI), and PFO SCIs are characterised by an infarction in the territory
Caen, Avenue Côte de closure. In addition to the settings of cardiac disease and of one perforating arteriole with a threshold size of
Nacre, Caen,
the interventional procedures used to manage them, SCIs ≥3 mm. Such infarcts can be visualized on CT images as
Normandy, France
(M. Hamon). have been observed in the context of hypertension and hypodense lesions, and when using MRI as hyperintense
lesions on T2‑weighted images. Improvements in MRI
Correspondence to:
R. Delewi Competing interests techniques, including stronger field magnets, thinner
r.delewi@amc.nl The authors declare no competing interests. slices, and modified pulse sequences, have resulted in

696 | DECEMBER 2013 | VOLUME 10  www.nature.com/nrcardio

© 2013 Macmillan Publishers Limited. All rights reserved

a higher sensitivity for the detection of acute cerebral
ischaemia than with CT.18 Indeed, a number of MRI
techniques are now available to detect SCIs, includ­
ing diffusion-weighted MRI (DWI) for identification
of acute ischaemic lesions, and T2‑weighted imaging
and fluid-attenuation inversion recovery (FLAIR) MRI
for visualization of chronic SCIs.19,20 A wide variety of
diagnostic criteria can be used for the evaluation of SCIs
depending on the MRI sequences used. For example,
some studies classified lesions with a diameter ≥3 mm
as potential SCIs, whereas others did not take diam­
eter into account. Moreover, definitions in MRI signal
­characteristics are not standardized. These differences
in diagnostic ­criteria limit the capacity to compare data
between studies.19
DWI is the most-sensitive technique for the visual­iz­
ation and quantification of acute cerebral ischaemia.18
Acute ischaemic cerebral lesions are detectable by MRI
owing to changes in the free motion of H2O molecules
(diffusion restriction) caused by acute ischaemia,
which is observed as a hyperintense, bright signal on
­diffusion-weighted images, and as low values on appar­
ent diffusion coefficient maps.21 These signals usually
d­isappear within 14 days after the onset of ischaemia.22
When assessing SCIs, these lesions must be distin­
guished from dilated Virchow–Robin spaces, which
can be challenging using neuroimaging, particularly in
instances when these MRI features coexist.11 Virchow–
Robin spaces are fluid-filled perivascular canals that
typically follow the course of a blood vessel through the
grey or white matter, have a similar signal intensity to
cerebrospinal fluid on all MRI pulse sequences, and are
generally <3 mm in diameter. Old SCI lesions (>14 days)
and cerebrospinal fluid both have hyperintense signals
by T2‑weighted MRI and DWI. According to a recent
consensus statement by Wardlaw and colleagues aimed
at providing universal defin­itions to be used in both
research and clinical settings, the minimum neuro­
imaging examination for assessment of SCIs should
include DWI, FLAIR MRI, and T2‑weighted imaging.11
Cardiovascular disease and SCIs
Atrial fibrillation
Left atrial thrombus formation as a result of AF is the
most-common cause of thromboembolic stroke and
accounts for >45% of cardiogenic thromboemboli.23 AF
is associated with a twofold increase in the incidence of
stroke, an association observed in patients with either
paroxysmal or chronic AF.24 In addition, in a large
(n = 966), population-based, post-mortem study, AF was
identified as an independent predictor of SCIs (OR 2.46,
95% CI 1.07–5.68).25 This finding was confirmed in the
Framingham Offspring Study,10 in which AF was associ­
ated with an increased risk of SCIs detected using MRI
(OR 2.16, 95% CI 1.07–4.40). Moreover, various stu­d­
ies have shown that AF is associated with an increased
­incidence of SCIs (Table 1).26–32
In a longitudinal, observational study, patients aged
<60 years and with type 2 diabetes mellitus and subclini­
cal AF had an increased prevalence of SCIs at baseline
NATURE REVIEWS | CARDIOLOGY
© 2013 Macmillan Publishers Limited. All rights reserved
REVIEWS
Key points
■■ Silent cerebral infarcts (SCIs) are increasingly observed in patients with cardiac
disease and in individuals who have undergone invasive cardiac procedures
■■ Cardiac diseases associated with the occurrence of SCIs include atrial
fibrillation, cardiomyopathies, and atrial septal abnormalities
■■ Postprocedural SCIs have been detected using MRI after left cardiac
catheterization, transcatheter aortic valve implantation, CABG surgery,
pulmonary vein isolation, and closure of patent foramen ovale
■■ SCIs do not always result in acute symptoms, but have been associated with
a threefold increase in the risk of stroke, and can be considered a precursor of
ischaemic stroke
■■ Accumulating evidence suggests that SCIs might have a role in the development
of dementia and depression, and in cognitive decline
■■ Increased recognition of SCIs could advance our understanding of their links to
cardiac and neurological disorders, and facilitate the development of preventative
therapeutic approaches
Silent cerebral infarct
Cardioembolic heart Cardiac procedures
disease ■ Left heart
■ Atrial fibriliation catheterization
■ Left ventricular ■ CABG surgery
thrombus ■ Transcatheter aortic
■ Cardiomyopathy valve implantation
■ Patent foramen ■ Pulmonary vein
ovale isolation
■ Closure of patent
foramen ovale
Stroke Cognitive decline Dementia Depression
Figure 1 | Summary of cardioembolic heart diseases and cardiac procedures that
have been associated with silent cerebral infarcts. The central image provides an
example of small, hyperintense lesions that indicate silent cerebral infarcts
(arrows), visualized using fluid-attenuated inversion recovery MRI. Such lesions
have been associated with forms of cardioembolic heart disease and cardiac
intervention procedures. Silent cerebral infarcts are increasingly recognized as a
contributing factor in various neurological conditions. Therefore, approaches to
prevent or treat silent cerebral infarcts in patients with cardiovascular disease
might reduce the risk of these outcomes.
MRI assessment compared with patients who had type 2
diabetes and no diagnosed sub­clinical AF (61% versus
29%; P <0.01).27 Furthermore, the patients with subclini­
cal AF had a higher incidence of ischaemic stroke than
those without silent AF (17.3% versus 5.9%; P <0.01).27 In
a prospective pilot study by Neumann and colleagues, SCIs
were observed in 12.3% of the patients with symptomatic
and drug-refractory paroxys­mal or persistent AF (with no
history of stroke) before pulmonary vein isolation.29
The correlation between AF and cognitive impair­
ment, dementia, and Alzheimer disease has been assessed
in several studies.33–35 In a meta-analysis, AF was inde­
pendently associated with an increased risk of inci­
dent dementia (HR 1.42, 95% CI 1.17–1.72, P <0.001).36
Further­more, in a large, community-based cohort, the
cumulative rate of dementia was 2.7% at 1 year and 10.5%
at 5 years after a diagnosis of AF in patients without evi­
dence of cognitive dysfunction or stroke at the time of
onset. 37 Subsequently, SCIs have been postulated to
underlie pathophysiological mechanisms related to these
observations, based on their a­ssociation with both AF
and dementia.37
VOLUME 10 | DECEMBER 2013 | 697
REVIEWS

Table 1 | Associations between cardiac diseases and SCIs

Diagnosis Reference Study design (n) Anticoagulation Imaging Rate of SCIs (%) Rate of stroke
therapy (%) modality (%)
AF
Paroxysmal and Gaita et al. Case–control (90 paroxysmal AF, Aspirin: 24 MRI Paroxysmal AF: 89 Patients with
persistent AF (2013)26 90 persistant AF, 90 controls) OAC: 66 Persistent AF: 92 previous CVA or
None: 11 Controls: 46 TIA were excluded
Silent AF in patients Marfella et al. Case–control (176 with silent AF, Aspirin: 99 MRI Silent AF: 61 Silent AF: 17
with type 2 diabetes (2013)27 288 without silent AF) No silent AF: 29 No silent AF: 6
mellitus (3‑year follow-up)
Nonvalvular AF Kobayashi et al. Case–control (71 with AF, Aspirin: 55 MRI AF: 80 Patients with
(2012)28 71 controls) OAC: 31 Controls: 69 previous CVA or
TIA were excluded
Symptomatic and Neumann et al. Prospective pilot study Aspirin: 25 MRI 12.3 Patients with
drug-refractory (2011)29 (45 cryoballoon ablation, 44 VKA: 51 previous CVA were
paroxysmal or radiofrequency energy technique) Clopidogrel: 1 excluded
persistent AF None: 25
Nonvalvular AF Sato et al. Case–control (212 with AF, None in the MRI AF: 86 Patients with
(2004)30 78 controls) first year Controls: 54 previous CVA were
At 12‑month follow-up, excluded
21% of the patients with
AF had new SCIs
Nonrheumatic AF with EAFT Study Double-blind RCT of OAC NA CT 14 (at baseline, of whom 20
history (<3 months) of Group (1996)31 treatment versus aspirin or 14% had additional SCIs
ischaemic stroke or TIA placebo (n = 985) at 2‑year follow-up)
Nonrheumatic AF Ezekowitz et al. RCT of warfarin versus placebo NA CT 15 (on baseline CT) 7*
without history of (1995)32 (n = 516) (3‑year follow-up)
stroke or TIA
Cardiomyopathy
Ischaemic and Kozdag et al. Case–control (46 with ischaemic Aspirin: 78 MRI Ischaemic DCM: 39 Patients with
nonischaemic DCM (2008)44 DCM, 26 with nonischaemic Nonischaemic DCM: 27 previous CVA or
DCM, 56 controls) Controls: 4 TIA were excluded
Cardiomyopathy with Siachos et al. Observational cohort (n = 117) Aspirin: 26 CT or 34 Patients with
LVEF <20% being (2005)41 VKA: 27 MRI previous CVA or
evaluated for heart None: 47 TIA were excluded
transplantation
PFO
PFO Di Tullio et al. Case–control (n = 360) Aspirin: 2 MRI PFO: 17 PFO: 9
(2013)57 Controls: 14 Controls: 10
(in entire cohort of
1,100 participants)
Pulmonary embolism Clergeau et al. Case–control (15 with PFO, NA MRI PFO: 33 2
with or without a PFO (2009)50 45 without PFO) No PFO: 2
*In the 307 patients for whom baseline and follow-up scans were available. Abbreviations: AF, atrial fibrillation; CVA, cerebrovascular accident; DCM, dilated cardiomyopathy; LVEF, left ventricular
ejection fraction; NA, not available; OAC, oral anticoagulant; PFO, patent foramen ovale; RCT, randomized controlled trial; SCIs, silent cerebral infarcts; TIA, transient ischaemic attack; VKA,
vitamin K antagonist.

Left ventricular thrombus formation
Left ventricular (LV) thrombus formation after myo­
cardial infarction carries the risk of systemic thrombo­
embolism, particularly in the cerebral circulation. In
a meta-analysis of 11 studies, including a total of 856
patients with anterior myocardial infarction, an odds ratio
of 5.5 (95% CI 3.0–9.8) for embolic events was reported.38
Although LV thrombus formation is known to be associ­
ated with an increased embolic risk, no data are currently
available on the occurrence of SCIs in patients with this
condition. Nonetheless, in the ongoing, randomized, con­
trolled LV‑THROMBUS study,39 two anticoagulant regi­
mens for the treatment of LV thrombus after myocardial
infarction are being compared. The occurrence of SCIs is
being recorded over time using serial MRI measurements,
enabling investigation of the potential role of SCIs as
p­recursors to stroke and cognitiv­e dysfunction.
Cardiomyopathy
Heart failure is known to be associated with an increased
risk of thromboembolism, with the reported rate of
stroke in trials of heart-failure therapies varying from
1.8% to 2.4% per year.40 Siachos and colleagues were the
first to report a 34% prevalence of SCIs in patients with
advanced heart failure (LV ejection fraction <20%) being
evaluated for heart transplantation (Table 1), suggest­
ing that the rate of subclinical infarcts is far higher than
that of clinically manifest stroke.41 The causes of heart
failure can be stratified into ischaemic and non­ischaemic
cardio­myopathies, with dilated cardio­myopathy being the

698 | DECEMBER 2013 | VOLUME 10  www.nature.com/nrcardio

© 2013 Macmillan Publishers Limited. All rights reserved

most-common form of nonischaemic cardio­myopathy.
The underlying mechanisms predisposing patients with
dilated cardiomyopathy to an increased risk of emboli
formation include low cardiac output, with subsequent
stasis of blood in the dilated chamber, and an altered
coagulation status.42
In a case–control study, patients (n = 72) with ischae­
mic or nonischaemic dilated cardiomyopathy, but no
history of stroke or TIA, had a 35% prevalence of MRI-
detected SCIs.43 This prevalence of SCIs was significantly
higher than that reported in the control group compris­
ing healthy individuals (35% versus 3.6%; P <0.01).43
In a subsequent publication, the same investigators
reported that 27% of the 26 patients with nonischaemic
dilated cardio­myopathy assessed in this study had SCIs
(Table 1).44 Ischaemic cardiomyopathy was also associ­
ated with an increased risk of thromboembolism; SCIs
were detected in 39% of the 46 patients with ischae­
mic cardiomyopathy, which was markedly higher than
the frequency observed in patients with nonischaemic
cardio­myopathy and the age-matched controls (27% and
3.6%, respectively).44 In ischaemic cardiomyopathy, the
mechanisms leading to an increased risk of thrombo­
embolism are similar to those described in dilated
cardiomyopathy, but with the additional risk factor of
atherosclerosis. Independent risk factors associated with
SCIs in patients with cardio­myopathy include impaired
LV function, restrictive diastolic filling patterns on echo­
cardiography, left atrial and aortic spontaneous echo con­
trast, and complex or c­alcified a­therosclerotic lesions in
the aorta.43,44
Patent foramen ovale
Currently, whether a causal relationship exists between
paradoxical emboli resulting from a PFO and crypto­
genic stroke, a subtype (30–40%) of ischaemic strokes for
which no well-defined underlying pathological mecha­
nism is found, is a subject of debate.45,46 Nevertheless,
investigators in various observational studies have
reported an increased frequency of PFO in patients with
cryptogenic stroke.47,48 Interestingly, a cross-sectional
analysis of patients with cryptogenic stroke (or TIA)
and PFO included in the Tufts PFO registry 49 showed
a 17% prevalence of SCIs on MRI, suggesting that an
a­ssociation between PFO, SCIs, and stroke might exist.
Several pathophysiological mechanisms are postu­
lated to contribute to cerebral embolus formation in
patients with PFO. One hypothesis is that paradoxical
emboli result from venous emboli travelling through the
right–left shunt of the PFO into the left atrial circula­
tion, thereby avoiding filtration in the lungs. This theory
is supported by the findings of a prospective study by
Clergeau and colleagues, who identified PFO as an inde­
pendent predictor of SCIs in patients with pulmonary
embolism.50 The reported prevalence of SCIs in patients
with pulmonary embolism and PFO was significantly
higher than in patients with pulmonary embolism
without a PFO (33.3% versus 2.2%; P = 0.003; Table 1).50
However, only one patient in the study, who was found
to have SCIs, experienced a stroke.50 Large studies with
NATURE REVIEWS | CARDIOLOGY
© 2013 Macmillan Publishers Limited. All rights reserved
REVIEWS
a long duration of follow-up are needed to investigate
this association further.
Other speculative mechanisms for cerebral emboli in
patients with PFO include thrombus formation within
the redundant interatrial tissue, particularly in patients
with atrial septal aneurysm, and atrial arrhythmias
resulting from PFO.51,52 Indeed, the results of case–
control studies suggest that the prevalence of PFO and
atrial septal aneurysm is increased in patients suffering
from cryptogenic stroke.47,53 In particular, the combina­
tion of a PFO and atrial septal aneurysm is thought to
increase the risk of paradoxical emboli owing to com­
pounded right–left shunt. Contrary to these findings,
prospective and population-based studies have shown
that the presence of PFO alone was not associated with
ischaemic stroke, suggesting that concomitant venous
thromboembolism would have to be present.54–56 In a
study by Di Tullio and colleagues, no significant dif­
ference in the rates of SCI and stroke were observed in
patients with a PFO compared with control individuals.57
Cardiac procedures and SCIs
The heart–brain relationship and the outcomes of inva­
sive procedures used in interventional cardiology have
been investigated in an increasing number of studies.
In general, such studies have shown an increased post­
procedural prevalence of SCIs (Table 2). In addition to
DWI, noninvasive transcranial Doppler (TCD) sono­
graphy has been used in a research setting to assess
the occurrence of microembolisms in real-time during
cardiac procedures. TCD sonography allows the detec­
tion of both gaseous and solid microemboli entering
the intracranial vessels of the circle of Willis, which are
v­isualized as high-intensity transient signals.58
Left cardiac catheterization
Coronary angiography and percutaneous coronary
intervention (PCI) are standard invasive procedures
in patients with ischaemic coronary artery disease.
Ischaemic stroke has been described as an infrequent
complication of these interventions, with incidences
varying from 0.2% to 0.4% for cardiac catheterizations
and 0.07% to 0.4% for PCI procedures.59,60 The frequency
of SCIs in patients who have undergone such proce­
dures has been shown higher than for ischaemic stroke,
ranging from 2% to 35% (Table 2).59–71
Numerous studies with periprocedural TCD monitor­
ing have shown that catheterization causes gaseous as
well as solid cerebral microemboli. Importantly, patients
with SCIs were reported to have a considerably higher
number of periprocedural solid microemboli than
patients without SCIs, as measured using MRI.63 Blood
clots formed on the tip of the catheter must be consid­
ered as a possible origin of cerebral emboli. Furthermore,
solid microemboli can result from catheter manipulation
in the aortic root, causing the release of small pieces of
atherosclerotic debris. This mechanism is of particular
relevance in the setting of extensive atheroma, which
is an important consideration in patients undergoing
cardiac catheterization who usually have generalized
VOLUME 10 | DECEMBER 2013 | 699
REVIEWS

Table 2 | Associations between interventional cardiac procedures and SCIs

Procedure and disease Reference Study design (n) Access route Imaging Rate of SCIs (%) Rate of
modality stroke (%)
Catheterization
CAD Ohi et al. (2013)64 Prospective comparative study Radial or MRI or 18 0
(111 left cardiac catheterization, femoral MDCT
56 MDCT)
Aortic valve stenosis Hamon et al. (2012)65 RCT (83 radial approach, Radial or MRI 14 0
77 femoral approach) femoral
CAD Kim et al. (2012)66 Observational (272 in phase I Radial or MRI 17 0
study, 102 in phase II study) femoral
CAD Kim et al. (2011)67 Observational (197) Radial or MRI 10 0
femoral
PPCI-treated patients Murai et al. (2008)68 Observational (75) Radial or MRI 35 0
with ACS femoral
Aortic valve stenosis Hamon et al. (2007)69 Observational (41) Radial MRI 5 0

Aortic valve stenosis Hamon et al. (2006)70 Observational (46) Femoral MRI 2 0
Patients with CAD Büsing et al. (2005)61 Observational (48) Femoral MRI 15 0
undergoing diagnostic
and interventional
catheterization
CAD Lund et al. (2005)63 Observational (47) Radial or MRI 9 0
femoral
Aortic valve stenosis Omran et al. (2003)62 RCT (cardiac catheterisation Femoral MRI 19 and 3, respectively 3
with [101] or without [51]
passage through the aortic valve)
TAVI
Aortic valve stenosis Ghanem et al. (2013)78 Observational pilot study (61) Transfemoral MRI 72* 7
Aortic valve stenosis Fairbairn et al. (2012)79 Observational (31) Femoral or MRI 77 6
subclavian
Aortic valve stenosis Rodés-Cabau et al. Observational (29 transfemoral, Transapical or MRI Transfemoral: 71 3
(2011)77 31 transapical) transfemoral Transapical: 62
Aortic valve stenosis Astarci et al. (2011)80 Observational (21 transfemoral, Transapical or MRI Transfemoral: 90 0
14 transapical) transfermoral Transapical: 93
Aortic valve stenosis Ghanem et al. (2010)81 Observational (22) Transfemoral MRI 73 4
Aortic valve stenosis Kahlert et al. (2010)76 Case–control (31 TAVI, 21 Transfemoral MRI TAVI: 84 0
surgical valve replacement) Surgical replacement: 48
Aortic valve stenosis Arnold et al. (2010)82 Observational (25) Transapical MRI 64 4
CABG surgery
CAD Ito et al. (2012)88 Observational (449) NA MRI 35 15
CAD Knipp et al. (2008)87 Observational (39) NA MRI 51 0
CAD Gerriets et al. (2008)89 Observational (106) NA MRI 15 (in MRI cohort, NA
n = 86)
CAD Bendszus et al. (2002)90 Observational (35) NA MRI 26 0
Pulmonary vein isolation
AF Haeusler et al. (2013)101 Observational (37) Trans-septal MRI 41 0
AF Neumann et al. (2011) 29
Prospective pilot study Trans-septal MRI 8 0
(45 cryoballoon ablation, 44
radiofrequency energy technique)
AF Herrera Siklódy et al. Observational Trans-septal MRI Irrigated radiofrequency 0
(2011)102 (27 irrigated radiofrequency energy: 7
energy, 23 cryoballoon, Cryoballoon: 4
24 multielectrode phased Multielectrode phased
radiofrequency) radiofrequency: 38
AF Schwarz et al. (2010)103 Case–control (23 AF, 23 controls) Trans-septal MRI 10 5
AF Gaita et al. (2010)104 Observational (232) Trans-septal MRI 14 0.4
AF Schrickel et al. (2010) 105
Observational (53) Trans-septal MRI 11 0

700 | DECEMBER 2013 | VOLUME 10  www.nature.com/nrcardio

© 2013 Macmillan Publishers Limited. All rights reserved

Table 2 (Cont.) | Associations between interventional cardiac procedures and SCIs
Procedure and disease Reference
PFO closure
Patients with PFO and Skowasch et al. (2010)108
history of cryptogenic
stroke
PFO Dorenbeck et al. (2007)109
*Of 39 patients who completed the imaging protocol. Abbreviations: ACS, acute coronary syndrome; AF, atrial fibrillation; CAD, coronary artery disease; MDCT, multidetector computed
tomography; NA, not available; PFO, patent foramen ovale; PPCI, primary percutaneous coronary intervention; RCT, randomized controlled trial; SCI, silent cerebral infarct; TAVI, transcatheter
aortic valve implantation.
atherosclerosis. In a study of 1,000 patients undergoing
PCI, 24–65% (depending upon the shape of the catheter
used) had aortic atheromatous material retrieved from
blood aspirated via the catheter.72 However, no associa­
tion was found between the presence of this material and
in-hospital ischaemic complications, which might par­
tially be explained by the sufficient withdrawal of blood
containing the debris before injection of contrast.72 In this
study, no routine neuroimaging was performed before or
after the procedure and, consequently, the potential rela­
tionship between atheromatous debris retrieved from the
catheter and small thromboemboli resulting in SCIs could
not be determined. Indeed, reported in-hospital ischae­
mic complications might be only the tip of the iceberg of
procedure-related thromboemboli.
Total duration of the procedure and procedural fluoro­
scopy time have also been described as independent
predictors of the occurrence of cerebral infarctions in
patients undergoing angiography or PCI.61 Additionally,
the frequency of postprocedural SCIs seems to vary
according to the catheterization procedure used. In par­
ticular, retrograde catheterization of severe aortic valve
stenosis is associated with a high rate of SCIs (22%).62 In
patients undergoing retrograde catheterization of severe
aortic valve stenosis, patient height (OR 8.24, 95% CI
2.71–25.02, P <0.0001) and a low transvalvular gradi­
ent (OR 0.96, 95% CI 0.93–0.99, P = 0.027) were asso­
ciated with an increased risk of periprocedural SCIs.65
Furthermore, on the basis of previous observational data,
the radial arterial access site for catheterization was asso­
ciated with an higher rate of cerebral embolic compli­
cations than femoral arterial access.73 However, a 2012,
multicentre, randomized trial, the SCIPION study, 65
showed that the incidence of new SCIs after catheteriza­
tion did not differ significantly between the femoral and
radial arterial approaches (11.7% versus 17.5%; OR 0.85,
95% CI 0.62–1.16, P = 0.31) in patients with severe aortic
stenosis scheduled for valve surgery. In addition to DWI,
periprocedural TCD assessment of high-intensity tran­
sient signals was performed in a subgroup of patients and
showed similar results for each access site.65 However,
after completion, this study seemed to be underpowered.
Therefore, although the arterial access site is unlikely to
influence the frequency of SCIs after catheterization,
large studies are required to confirm this finding. Such
research is of particular importance given that coronary
angiography is in­creasingly performed via the radial
arterial access route.
NATURE REVIEWS | CARDIOLOGY
© 2013 Macmillan Publishers Limited. All rights reserved
REVIEWS
Study design (n) Access route Imaging Rate of SCIs (%) Rate of
modality stroke (%)
Observational (63) Trans-septal MRI 3 2
Observational (35) Trans-septal MRI 6 3
Transcatheter aortic valve implantation
TAVI is an alternative treatment for patients with severe
symptomatic aortic stenosis considered to be at high
risk from conventional surgical valve replacement,
mostly elderly individuals with advanced atherosclerotic
disease. Estimates of the risk of postprocedural stroke
associated with TAVI vary from 1.5% to 10.0%.74,75 Using
MRI, SCIs have been observed more frequently than
stroke, with the prevalence ranging from 62% to 93%
(Table 2).76–82 Kahlert and colleagues compared the rate
of SCIs detected using MRI in patients undergoing trans­
femoral TAVI with historical controls who underwent
surgical aortic valve replacement.76 Postprocedural SCIs
were significantly more frequent in patients who under­
went TAVI than in control individuals (84% versus 48%;
P = 0.011).76 Nonetheless, stroke was reported in only one
patient who had undergone surgical aortic valve replace­
ment, which necessitates further study of the relationship
between the increased frequency of SCIs after TAVI and
neurological complications.
Cerebral microemboli that occur after TAVI are prob­
ably caused during device positioning and implantation
in the stenotic aortic valve. TCD studies provide impor­
tant insight into the mechanisms of cerebral emboli
during these procedures and have shown that deployment
of the valve prosthesis during TAVI is associated with the
highest frequency of high-intensity transient signals.83–85
In addition, hypoperfusion after rapid right ventricular
pacing during deployment of the valve p ­ rosthesis might
also be related to ischaemic brain injury.86
The two most-widely used approaches for TAVI are the
transfemoral and transapical routes. During the more-
commonly used transfemoral approach, a large cath­
eter (18–24 F) containing the valve is advanced through
the aortic arch and crossed retrograde over the severely
diseased native aortic valve. The alternative trans­apical
approach is preferred in patients with diseased or severely
calcified iliofemoral arteries, and involves direct punc­
turing of the ventricular apex through a small left lateral
thoracotomy. A catheter is then inserted through the
ventricular apex in the mid portion of the ventricular
cavity, through which the valve is advanced and placed in
the native aortic valve. Given that manipulation of large
catheters in the aorta and the retrograde crossing of the
native aortic valve are avoided using the transapical pro­
cedure, this technique was assumed to cause a lower rate
of procedure-related SCIs than the transfemoral approach.
However, in a prospective, multicentre study, in which the
VOLUME 10 | DECEMBER 2013 | 701
REVIEWS
incidence of SCIs in patients who underwent either trans­
femoral or transapical TAVI was compared, no significant
difference was observed between the two routes (66% and
71%; P = 0.78).77 Of the new cerebral microinfarcts identi­
fied using DWI, 91% were <1 cm, 76% were multiple in
number, and 73% involved both cerebral hemispheres.77
Most lesions found on DWI were clinically silent, except
in two patients, one from each treatment group, who expe­
rienced symptomatic cerebral emboli within 24 h of the
procedure.77 In addition to DWI, cognitive function was
assessed before and 6 days after the procedure, with no
significant differences recorded between patients with or
without new cerebral lesions.77 However, this result might
be attributable to the short evaluation time or a lack of
sensitivity of the cognitive assessment. Unfortunately, no
long-term data are currently available on the incidence
of SCIs or stroke after TAVI, and the possible effects on
co­gnitive decline and dementia.
CABG surgery
Interest in cerebral complications, such as symptomatic
cerebral infarcts, SCIs, and cognitive decline, after CABG
surgery is increasing. Indeed, the occurrence of SCIs
after cardiac surgery has been assessed in several studies,
showing an incidence between 15% and 51% (Table 2).71,87–90
Additionally, postoperative cognitive decline is a frequent
cerebral complication after CABG surgery, detected in
14–48% of patients at postoperative follow-up.91 The clini­
cal consequences of cognitive decline are numerous and
result in increased use of health-care resources.92
The aetiology of postoperative cognitive decline is
probably multifactorial.93,94 However, studies using TCD
imaging have detected showers of emboli periprocedurally
during CABG surgery, particularly during cannulation and
clamping of the aorta.95 Furthermore, in patients undergo­
ing CABG surgery, the extent of atheromatous disease of
the ascending aorta and the aortic arch has been associated
with the microembolic load during periprocedural TCD
monitoring and on post­procedural DWI.96 Moreover,
the occurrence of p ­ ostprocedural SCIs is postulated to be
higher after CABG surgery combined with aortic valve
replacement than after CABG surgery alone.71,97
Investigators in various studies have assessed whether
SCIs detected using DWI are associated with cognitive
dysfunction after CABG surgery. However, in most of
these studies, cognitive function was assessed early post­
operatively, and variable results have been reported.98 A
fairly small study (n = 39) with a 3‑year follow-up showed
a two-stage course of cognition after CABG surgery: early
cognitive decline during the initial days after surgery
(until hospital discharge) that improved at 3 months,
followed by a second cognitive decline observed at
3 years after surgery.87 In this study, SCIs reported post­
operatively using DWI were not associated with early or
late ­cognitive decline.87
Off-pump CABG surgery without cardiopulmonary
bypass and reduced aortic manipulation was assumed to
result in fewer new SCIs than on-pump CABG surgery.
However, a large randomized trial (n = 281) showed
no significant difference in the frequency of cognitive
702 | DECEMBER 2013 | VOLUME 10  www.nature.com/nrcardio
© 2013 Macmillan Publishers Limited. All rights reserved
dysfunction for on-pump and off-pump CABG surgery
at the 3‑month or 12‑month follow-up.99 Unfortunately,
no MRI was performed to assess the occurrence of SCIs
in the individuals enrolled in this study. The findings from
the aforementioned randomized trial were supported by a
study by Lund and colleagues, who observed no significant
difference between off-pump and on-pump CABG surgery
in the rate of new post­operative SCIs lesions and cognitive
decline at 3 months after surgery.91
Pulmonary vein isolation
International clinical guidelines have established PVI an
important treatment strategy in patients with AF who
remain symptomatic despite optimal medical therapy, and
in patients in whom the potential benefit is sufficient to
justify an ablation procedure.100 Various pulmonary vein
ablation strategies, including segmental PVI and circum­
ferential antral PVI with or without linear lesions, and the
radiofrequency energy or cryoballoon techniques, can be
used. Thromboembolic complications after PVI in patients
with AF have been described, typically occurring within
the first 24 h after pulmonary vein ablation and with an
increased risk in the first 2 weeks after the procedure.
Various studies have shown that the incidence of SCIs
after PVI is between 4% and 41%.29,101–105
In the prospective, pilot MEDAFI‑TRIAL,29 the inci­
dence of cerebral emboli detected using MRI was assessed
in patients undergoing PVI with either the cryo­balloon
or the radiofrequency ablation technique. New SCIs were
found in 7.9% of the patients within 1 day after pulmon­
ary vein ablation (Table 2), with no significant differences
between the two treatment modalities, and none of the
patients developing symptomatic cerebral infarcts after
the procedure.29 Furthermore, the occurrence of SCIs
and their relationship with postprocedural cognitive
functioning were assessed in patients with symptomatic
paroxysmal AF undergoing left atrial catheter ablation
in the MACPAF study.101 New postprocedural, MRI-
detected SCIs were reported in 41% of patients, although
these ischaemic lesions were not associated with cogni­
tive impairment immediately after the procedure or at
6–9 months follow-up.101
Patent foramen ovale closure
Currently, whether patients who experience a cryptogenic
stroke and have a PFO should be treated with closure of the
PFO, or whether medical therapy with anticoagulants is
sufficient for secondary prevention of ischaemic events,
is heavily debated. Intention-to-treat analyses have not
revealed a substantially reduced rate of recurrent stroke,
TIA, or death in patients treated with PFO closure com­
pared with medical therapy alone.106,107 However, Meier
and colleagues noted that, after completion, their study
seemed to be underpowered, making it difficult to detect
a clinical benefit of PFO closure.106 Observational studies
have shown a 3%108 and 6%109 incidence of SCIs after PFO
closure in patients who had suffered from a cryptogenic
stroke (Table 2). Nevertheless, the relationships between
PFO closure, SCIs, and neurological complications remain
unclear and should be the subject of future studies.

Clinical implications of SCIs
Improvements in anticoagulation therapy and invasive
cardiac interventions have led to a decrease in the preva­
lence of cardioembolic stroke. Despite these improve­
ments, advancements in neuroimaging have resulted in
an increased awareness of subclinical SCIs, which are
much more prevalent than symptomatic stroke. SCIs
detected in daily clinical practice are often found by
chance, as a consequence of neuroimaging performed
to address some other clinical question. Owing to the
lack of symptoms associated with SCIs, these incidental
findings are frequently disregarded. However, whether
these lesions are ‘innocent’ bystanders or contributors
to neurological conditions remains unclear.
Large, observational, population-based studies, such
as the Rotterdam Scan Study 17 and the Cardiovascular
Health Study,110 were among the first investigations of
the occurrence of SCIs and their association with the
risk of future stroke, cognitive decline, and dementia.
In the Cardiovascular Health study,110 participants with
SCIs detected using MRI had an increased incidence of
stroke during the 4‑year follow-up period compared
with individuals without SCI lesions (18.7 versus 9.5 per
1,000 persons per year). Furthermore, the adjusted rela­
tive risk of stroke increased in individuals with multiple
SCIs (HR 1.9, 95% CI 1.2–2.8).110 In the Rotterdam Scan
Study,17 the presence of SCIs increased the risk of stroke
more than threefold, independently of other risk factors
(adjusted HR 3.9, 95% CI 2.3–6.8). The increased risk
of stroke associated with SCIs might be attributable to
the underlying conditions, such as cardiac disease, that
also caused the SCIs and are conceivably still influenc­
ing the patient. However, in the setting of SCIs caused
by an external event, such as invasive cardiac interven­
tion, whether the procedure is associated with the risk
of subsequent stroke beyond the initial period after the
p­rocedure remains unclear and warrants future study.
In addition to their association with an increased
risk of stroke, MRI-detected SCIs have been reported
to more than double the risk of dementia and, in par­
ticular, Alzheimer disease in the general population.13
Furthermore, MRI studies in patients with vascular
dementia have supported the concept that the cumula­
tive burden of ischaemic brain injury resulting from
multiple SCIs contributes to cognitive decline. 110,111
Interestingly, the decline in various cognitive domains
has been found to be associated with the location of SCIs
detected using MRI.13 Although the pathophysiological
mechanism underlying dementia is likely to be hetero­
geneous, various mechanisms can be postulated to explain
the observed association between SCIs and Alzheimer
disease. For example, SCIs occurring in a brain already
affected by Alzheimer disease might further impair cogni­
tion, resulting in the final diagnosis of dementia. However,
such lesions might also trigger the development of neuro­
fibrillary tangles and senile plaques, which potentiates the
abnormalities associated with Alzheimer disease.112
Several studies have also shown that major depression
occurring for the first time during or after the presenile
period might be related to SCIs, an event that might be
NATURE REVIEWS | CARDIOLOGY
© 2013 Macmillan Publishers Limited. All rights reserved
REVIEWS
Type of cerebral infarct Frequency Method of detection
Clinically apparent 0.2–0.4% Neurological examination
Subclinical Neuropsychological testing
Subtle cerebral infarct Cerebral imaging
(5–23%)
Silent
Subtle cerebral injury Biological
(100%) markers?
Figure 2 | Schematic overview of silent to clinically apparent cerebral infarcts.
Clinically silent cerebral injury is thought to occur in all individuals. Silent cerebral
infarcts might affect around a quarter of these individuals, whereas subclinical
cerebral infarcts (detected using neurophysiological testing) are observed in an
even smaller proportion. Clinically apparent cerebral infarcts are observed in
<0.5% of the population. Therefore, a large number of patients potentially harbour
cerebral injuries that might influence the development of various neurological
conditions. Imaging modalities that can or might confirm the presence of cerebral
injury and infarcts at each stage are indicated.
comparable to depression after a stroke.113–116 When the
clinical outcome in patients with depression aged >50 years
was investigated over a 3‑year period, MRI-detected SCIs
were found to be associated with an increased frequency
and longer duration of hospital admission owing to depres­
sion.115 Furthermore, Fujikawa and colleagues found that
SCIs identified using MRI were more frequently observed
in senile (individ­uals aged >65 years) depression than in
presenile (individuals aged 50–60 years) depression (93.7%
versus 65.9%; P <0.01).114
Importantly, the increases in the risk of future stroke,
cognitive decline, dementia, and depression associated
with SCI lesions discussed above were not observed in the
setting of cardiac disease or interventional cardiology pro­
cedures. Nonetheless, these results support the potential
clinical influence of SCIs. Moreover, AF has been indepen­
dently associated with an increased risk of dementia,36 and
SCIs might be the underlying mechanism; however, this
possibility needs to be further investigated. Additionally,
several studies have investigated the relationship between
SCIs and cognitive decline after invasive cardiac inter­
ventions, such as TAVI, CABG surgery, PVI, and PFO
closure. These studies had different durations of follow-
up and reported contradictory results, which complicates
comparisons between studies. Therefore, future studies
are required to clarify the associations between cardiac
p­rocedures, SCIs, and neurological conditions.
Whether SCIs are truly asymptomatic is still a subject
of debate. Evidently, whether a cerebral infarct detected
on neuroimaging is ‘silent’ depends on the vantage point
of both the patient and physician and might even differ
between the two (Figure 2). Some patients might not
be aware that the symptoms that they have experienced
were caused by a cerebral infarct, or clinical evaluations
VOLUME 10 | DECEMBER 2013 | 703
REVIEWS

might not have been performed at the time, so a stroke
was never diagnosed. Several studies have used other
terms for ‘silent’, such as ‘prior’, ‘covert’, or ‘subclinical’
cerebral microinfarcts. Indeed, the variation in terms and
definitions of SCIs used among the studies performed to
date limits cross-study comparisons, which are impor­
tant for interpretation of the pathological correlation
and clinical consequences of such lesions. Therefore,
future studies should describe the criteria used to define
‘silent’ infarcts, as well as the types of neurological
examinations performed.
associated with an increased risk of cognitive impairment,
subsequent dementia, and depression. Therefore, detec­
tion of SCIs might facilitate the management of patients
with cardiac disease and those undergoing invasive
cardiac interventions. Given the high incidence of SCIs
in cardiac diseases and after the interventional cardiology
procedures used to treat them, further research assess­
ing the prognostic implications of these lesions, as well as
their possible prevention or management, are warranted.
Review criteria

Conclusions We searched the PubMed and EMBASE databases for

The occurrence of SCIs in cardiac disease and after inva­
sive cardiac procedures is frequently observed. AF 26–32
and cardiomyopathy 41,44 have been associated with an
increased occurrence of SCIs. Other cardiac diseases
likely to be related to an increased incidence of SCIs, but
which require further investigation, are LV thrombus for­
mation and PFO. Furthermore, the development of new
SCI lesions after cardiac interventions is gaining research
interest. Left heart catheterization, 66 TAVI, 76 CABG
surgery,87 and PVI101 have all been linked with a high
frequency of SCIs, with TAVI having the highest inci­
dence of postprocedural SCIs (62–84% of patients).76–82
Although PFO closure in relation to stroke has been
extensively investigated, the occurrence of SCIs after this
i­ntervention has not been widely studied.
Additionally, various large, population-based studies
have shown that SCIs double the risk of future stroke
and can be considered a prodromal symptom before
the development of clinical stroke. Moreover, increasing
evidence suggests that these ‘silent’ cerebral lesions are
articles published between 1980 and September 2013.
The following search terms were used alone or in
combination: “brain ischemia”, “intracranial embolism”,
“silent brain ischemia”, “silent cerebral ischemia”,
“silent brain infarct”, “silent cerebral infarct”, “silent
stroke”, “silent brain injury”, “silent intracranial emboli”,
“silent cerebral emboli”, “silent lacunar infarct”, “silent
ischemic lesion”, “microemboli”, “magnetic resonance
imaging”, “dementia”, “Alzheimer disease”, “cognitive
decline”, “cognitive impairment”, “cognitive dysfunction”,
“depressive disorder”, “depression”, “atrial fibrillation”,
“thrombosis”, “left ventricular thrombus”, “mural
thrombus”, “cardiomyopathy”, “heart failure”, “patent
foramen ovale”, “percutaneous coronary intervention”,
“percutaneous coronary revascularization”, “cardiac
catheterization”, “heart catheterization”, “coronary
angiography”, “aortic valve implantation”, “transcatheter
aortic valve implantation”, “coronary bypass surgery”,
“cardiac surgery”, “coronary bypass graft surgery”,
“catheter ablation”, and “pulmonary vein isolation.” The
relevant, full-text papers published in English were selected
from those identified. Furthermore, we reviewed the
reference lists of selected articles for additional studies.

1. Hamon, M., Baron, J. C., Viader, F. & Hamon, M.
Periprocedural stroke and cardiac catheterization.
Circulation 118, 678–683 (2008).
2. Arboix, A & Alió, J. Cardioembolic stroke:
clinical features, specific cardiac disorders and
prognosis. Curr. Cardiol. Rev. 6, 150–161 (2010).
3. Bendszus, M. & Stoll, G. Silent cerebral
ischaemia: hidden fingerprints of invasive medical
procedures. Lancet Neurol. 5, 364–372 (2006).
4. Sacco, R. L. et al. An updated definition of stroke
for the 21st century: a statement for healthcare
professionals from the American Heart
Association/American Stroke Association.
Stroke 44, 2064–2089 (2013).
5. Arboix, A. & Martí-Vilalta, J. L. Lacunar stroke.
Expert Rev. Neurother. 9, 179–196 (2009).
6. Vermeer, S. E., Koudstaal, P. J., Oudkerk, M.,
Hofman. A. & Breteler, M. M. Prevalence and risk
factors of silent brain infarcts in the population-
based Rotterdam Scan Study. Stroke 33, 21–25
(2002).
7. Price, T. R. et al. Silent brain infarction on
magnetic resonance imaging and neurological
abnormalities in community-dwelling older
adults: the Cardiovascular Health Study. Stroke
28, 1158–1164 (1997).
8. Kase, C. S. et al. Prevalence of silent stroke
in patients presenting with initial stroke: the
Framingham Study. Stroke 20, 850–852 (1989).
9. Ricci, S. et al., Silent brain infarctions in patients
with first-ever stroke: a community-based study
in Umbria, Italy. Stroke 24, 647–651 (1993).
10. Das, R. R. et al., Prevalence and correlates
of silent cerebral infarcts in the Framingham
Offspring Study. Stroke 39, 2929–2935 (2008).
11. Wardlaw, J. M. et al. Neuroimaging
standards for research into small vessel
disease and its contribution to ageing and
neurodegeneration. Lancet Neurol. 12,
822–838 (2013).
12. Kloppenborg, R. P. et al. Cerebral small-vessel
disease and progression of brain atrophy: the
SMART‑MR study. Neurology 79, 2029–2036
(2012).
13. Vermeer, S. E. et al., Silent brain infarcts
and the risk of dementia and cognitive decline.
N. Engl. J. Med. 348, 1215–1222 (2003).
14. Fujikawa, T., Yamawaki, S. & Touhouda, Y.
Incidence of silent cerebral infarction in
patients with major depression. Stroke 24,
1631–1634 (1993).
15. Yamashita, H. et al. Long-term prognosis of
patients with major depression and silent
cerebral infarction. Neuropsychobiology 62,
177–181 (2010).
16. Debette, S. et al., Association of MRI markers
of vascular brain injury with incident stroke, mild
cognitive impairment, dementia, and mortality:
the Framingham Offspring Study. Stroke 41,
600–606 (2010).
17. Vermeer, S. E. et al. Silent brain infarcts and
white matter lesions increase stroke risk in the
general population: the Rotterdam Scan Study.
Stroke 34, 1126–1129 (2003).
18. Mullins, M. E. et al. CT and conventional and
diffusion-weighted MR imaging in acute stroke:
study in 691 patients at presentation to the
emergency department. Radiology 224,
353–360 (2002).
19. Zhu, Y. C., Dufouil, C., Tzourio, C. & Chabriat, H.
Silent brain infarcts: a review of MRI diagnostic
criteria. Stroke 42, 1140–1145 (2011).
20. Marks, M. P. et al. Acute and chronic stroke:
navigated spin-echo diffusion-weighted MR
imaging. Radiology 199, 403–408 (1996).
21. Kucharczyk, J., Mintorovitch, J., Asgari, H. S.
& Moseley, M. Diffusion/perfusion MR imaging
of acute cerebral ischemia. Magn. Reson. Med.
19, 311–315 (1991).
22. Burdette, J. H., Ricci, P. E., Petitti, N. & Elster, A. D.
Cerebral infarction: time course of signal
intensity changes on diffusion-weighted MR
images. AJR Am. J. Roentgenol. 171, 791–795
(1998).
23. Kelley, R. E. & Minagar, A. Cardioembolic stroke:
an update. South. Med. J. 96, 343–349 (2003).
24 Friberg, L., Hammar, N. & Rosenqvist, M. Stroke
in paroxysmal atrial fibrillation: report from the
Stockholm Cohort of Atrial Fibrillation. Eur. Heart J.
31, 967–975 (2010).
25. Shinkawa, A. et al. Silent cerebral infarction
in a community-based autopsy series in Japan:
the Hisayama Study. Stroke 26, 380–385 (1995).
26. Gaita, F. et al. Prevalence of silent cerebral
ischemia in paroxysmal and persistent atrial
fibrillation and correlation with cognitive

704 | DECEMBER 2013 | VOLUME 10  www.nature.com/nrcardio

© 2013 Macmillan Publishers Limited. All rights reserved

function. J. Am. Coll. Cardiol. http://dx.doi.org/
10.1016/j.jacc.2013.05.074.
27. Marfella, R. et al. Brief episodes of silent atrial
fibrillation predict clinical vascular brain disease
in type 2 diabetic patients. J. Am. Coll. Cardiol.
62, 525–520 (2013).
28. Kobayashi, A., Iguchi, M., Shimizu, S.
& Uchiyama, S. Silent cerebral infarcts and
cerebral white matter lesions in patients with
nonvalvular atrial fibrillation. J. Stroke
Cerebrovasc. Dis. 21, 310–317 (2012).
29. Neumann, T. et al. MEDAFI‑Trial (Micro-
Embolization During Ablation of Atrial Fibrillation):
comparison of pulmonary vein isolation using
cryoballoon technique vs. radiofrequency energy.
Europace 13, 37–44 (2011).
30. Sato, H. et al. Aspirin attenuates the incidence
of silent brain lesions in patients with
nonvalvular atrial fibrillation. Circ. J. 68,
410–416 (2004).
31. EAFT Study Group. Silent brain infarction in
nonrheumatic atrial fibrillation: European Atrial
Fibrillation Trial. Neurology 46, 159–165
(1996).
32. Ezekowitz, M. D. et al. Silent cerebral infarction
in patients with nonrheumatic atrial fibrillation.
Circulation 92, 2178–2182 (1995).
33. Ott, A. et al. Atrial fibrillation and dementia in a
population-based study: the Rotterdam Study.
Stroke 28, 316–321 (1997).
34. Knecht, S. et al. Atrial fibrillation in stroke-free
patients is associated with memory impairment
and hippocampal atrophy. Eur. Heart J. 29,
2125–2132 (2008).
35. Mielke, M. M. et al. Vascular factors predict rate
of progression in Alzheimer disease. Neurology
69, 1850–1858 (2007).
36. Santangeli, P. et al. Atrial fibrillation and the risk
of incident dementia: a meta-analysis. Heart
Rhythm 9, 1761–1768 (2012).
37. Miyasaka, Y. et al. Risk of dementia in stroke-free
patients diagnosed with atrial fibrillation: data
from a community-based cohort. Eur. Heart J. 28,
1962–1967 (2007).
38. Vaitkus, P. T. & Barnathan, E. S. Embolic
potential, prevention and management of mural
thrombus complicating anterior myocardial
infarction: a meta-analysis. J. Am. Coll. Cardiol.
22, 1004–1009 (1993).
39. US National Library of Medicine. ClinicalTrials.gov
[online], http://clinicaltrials.gov/ct2/show/
NCT01556659 (2012).
40. Dries, D. L., Rosenberg, Y. D., Waclawiw, M. A.
& Domanski, M. J. Ejection fraction and risk of
thromboembolic events in patients with systolic
dysfunction and sinus rhythm: evidence for
gender differences in the studies of left
ventricular dysfunction trials. J. Am. Coll. Cardiol.
29, 1074–1080 (1997).
41. Siachos, T. et al. Silent strokes in patients with
heart failure. J. Card. Fail. 11, 485–489 (2005).
42. Koniaris, L. S. & Goldhaber, S. Z. Anticoagulation
in dilated cardiomyopathy. J. Am. Coll. Cardiol. 31,
745–748 (1998).
43. Kozdag, G. et al. Silent cerebral infarction in
patients with dilated cardiomyopathy:
echocardiographic correlates. Int. J. Cardiol. 107,
376–381 (2006).
44. Kozdag, G. et al. Silent cerebral infarction in
chronic heart failure: ischemic and nonischemic
dilated cardiomyopathy. Vasc. Health Risk Manag.
4, 463–469 (2008).
45. Bal, S. et al. High rate of magnetic resonance
imaging stroke recurrence in cryptogenic
transient ischemic attack and minor stroke
patients. Stroke 43, 3387–3388 (2012).
46. Foulkes, M. A., Wolf, P. A., Price, T. R., Mohr, J. P.
& Hier, D. B. The Stroke Data Bank: design,
NATURE REVIEWS | CARDIOLOGY
methods, and baseline characteristics. Stroke
19, 547–554 (1988).
47. Overell, J. R., Bone, I. & Lees, K. R. Interatrial
septal abnormalities and stroke: a meta-
analysis of case–control studies. Neurology
55, 1172–1179 (2000).
48. Cabanes, L. et al. Atrial septal aneurysm and
patent foramen ovale as risk factors for
cryptogenic stroke in patients less than 55 years
of age: a study using transesophageal
echocardiography. Stroke 24, 1865–1873
(1993).
49. Kitsios, G. D., Lasker, A. Singh, J. & Thaler, D. E.
Recurrent stroke on imaging and presumed
paradoxical embolism: a cross-sectional
analysis. Neurology 78, 993–997 (2012).
50. Clergeau, M. R. et al. Silent cerebral infarcts in
patients with pulmonary embolism and a patent
foramen ovale: a prospective diffusion-weighted
MRI study. Stroke 40, 3758–3762 (2009).
51. Ning, M. et al. The brain’s heart—therapeutic
opportunities for patent foramen ovale (PFO) and
neurovascular disease. Pharmacol. Ther. 139,
111–123 (2013).
52. Cabanes, L. et al. Atrial septal aneurysm and
patent foramen ovale as risk factors for
cryptogenic stroke in patients less than 55 years
of age. A study using transesophageal
echocardiography. Stroke 24, 1865–1873
(1993).
53. Mas, J. L. et al. Recurrent cerebrovascular
events associated with patent foramen ovale,
atrial septal aneurysm, or both. N. Engl. J. Med.
345, 1740–1746 (2001).
54. Davis, D. et al. Patent foramen ovale, ischemic
stroke and migraine: systematic review and
stratified meta-analysis of association studies.
Neuroepidemiology 40, 56–67 (2013).
55. Handke, M., Harloff, A., Bode, C. & Geibel, A.
Patent foramen ovale and cryptogenic stroke:
a matter of age? Semin. Thromb. Hemost. 35,
505–514 (2009).
56. Lethen, H. et al. Frequency of deep vein
thrombosis in patients with patent foramen
ovale and ischemic stroke or transient ischemic
attack. Am. J. Cardiol. 80, 1066–1069 (1997).
57. Di Tullio, M. R. et al. Patent foramen ovale,
subclinical cerebrovascular disease, and
ischemic stroke in a population-based cohort.
J. Am. Coll. Cardiol. 62, 35–41 (2013).
58. Russell, D. & Brucher, R. Online automatic
discrimination between solid and gaseous
cerebral microemboli with the first multifrequency
transcranial Doppler. Stroke 33, 1975–1980
(2002).
59. Fuchs, S. et al. Stroke complicating percutaneous
coronary interventions: incidence, predictors,
and prognostic implications. Circulation 106,
86–91 (2002).
60. Segal, A. Z., Abernethy, W. B., Palacios, I. F.,
BeLue, R. & Rordorf, G. Stroke as a complication
of cardiac catheterization: risk factors and
clinical features. Neurology 56, 975–977 (2001).
61. Büsing, K. A. et al. Cerebral infarction:
incidence and risk factors after diagnostic
and interventional cardiac catheterization—
prospective evaluation at diffusion-weighted MR
imaging. Radiology 235, 177–183 (2005).
62. Omran, H. et al. Silent and apparent cerebral
embolism after retrograde catheterisation of the
aortic valve in valvular stenosis: a prospective,
randomised study. Lancet 361, 1241–1246
(2003).
63. Lund, C. et al. Cerebral emboli during left heart
catheterization may cause acute brain injury.
Eur. Heart J. 26, 1269–1275 (2005).
64. Ohi, Y. et al. Cerebral microembolism following
coronary angiography—a prospective
© 2013 Macmillan Publishers Limited. All rights reserved
REVIEWS
comparative study between left cardiac
catheterization and multidetector computed
tomography. Intern. Med. 52, 1869–1874 (2013).
65. Hamon, M. et al. Silent cerebral infarcts after
cardiac catheterization: a randomized
comparison of radial and femoral approaches.
Am. Heart J. 164, 449–454 (2012).
66. Kim, B. J. et al. Insufficient platelet inhibition
is related to silent embolic cerebral infarctions
after coronary angiography. Stroke 43, 727–732
(2012).
67. Kim, I. C. et al. Incidence and predictors of silent
embolic cerebral infarction following diagnostic
coronary angiography. Int. J. Cardiol. 148, 179–182
(2011).
68. Murai, M. et al. Asymptomatic acute ischemic
stroke after primary percutaneous coronary
intervention in patients with acute coronary
syndrome might be caused mainly by
manipulating catheters or devices in the
ascending aorta, regardless of the approach to
the coronary artery. Circ. J. 72, 51–55 (2008).
69. Hamon, M. et al. Risk of acute brain injury
related to cerebral microembolism during
cardiac catheterization performed by right upper
limb arterial access. Stroke 38, 2176–2179
(2007).
70. Hamon, M. et al., Cerebral microembolism during
cardiac catheterization and risk of acute brain
injury: a prospective diffusion-weighted magnetic
resonance imaging study. Stroke 37, 2035–2038
(2006).
71. Sun, X., Lindsay, J., Monsein, L. H., Hill, P. C.
& Corso, P. J. Silent brain injury after cardiac
surgery: a review: cognitive dysfunction and
magnetic resonance imaging diffusion-weighted
imaging findings. J. Am. Coll. Cardiol. 60, 791–797
(2012).
72. Keeley, E. C. & Grines, C. L. Scraping of aortic
debris by coronary guiding catheters: a
prospective evaluation of 1, 000 cases. J. Am.
Coll. Cardiol. 32, 1861–1865 (1998).
73. Karalis, D. G. et al. Risk of catheter-related
emboli in patients with atherosclerotic debris in
the thoracic aorta. Am. Heart J. 131, 1149–1155
(1996).
74. Webb, J. G. et al. Transcatheter aortic valve
implantation: impact on clinical and valve-related
outcomes. Circulation 119, 3009–3016 (2009).
75. Grube, E. et al. Percutaneous aortic valve
replacement for severe aortic stenosis in high-
risk patients using the second- and current third-
generation self-expanding CoreValve prosthesis:
device success and 30-day clinical outcome.
J. Am. Coll. Cardiol. 50, 69–76 (2007).
76. Kahlert, P. et al. Silent and apparent cerebral
ischemia after percutaneous transfemoral aortic
valve implantation: a diffusion-weighted
magnetic resonance imaging study. Circulation
121, 870–878 (2010).
77. Rodés-Cabau, J. et al. Cerebral embolism
following transcatheter aortic valve implantation:
comparison of transfemoral and transapical
approaches. J. Am. Coll. Cardiol. 57, 18–28
(2011).
78. Ghanem, A. et al. Prognostic value of cerebral
injury following transfemoral aortic valve
implantation. EuroIntervention 8, 1296–1306
(2013).
79. Fairbairn, T. A. et al. Diffusion-weighted MRI
determined cerebral embolic infarction following
transcatheter aortic valve implantation:
assessment of predictive risk factors and the
relationship to subsequent health status. Heart
98, 18–23 (2012).
80. Astarci, P. et al. Magnetic resonance imaging
evaluation of cerebral embolization during
percutaneous aortic valve implantation:
VOLUME 10 | DECEMBER 2013 | 705
REVIEWS
comparison of transfemoral and trans-apical
approaches using Edwards Sapiens valve. Eur. J.
Cardiothorac. Surg. 40, 475–479 (2011).
81. Ghanem, A. et al. Risk and fate of cerebral
embolism after transfemoral aortic valve
implantation: a prospective pilot study with
diffusion-weighted magnetic resonance imaging.
J. Am. Coll. Cardiol. 55, 1427–1432 (2010).
82. Arnold, M. et al. Embolic cerebral insults after
transapical aortic valve implantation detected by
magnetic resonance imaging. JACC Cardiovasc.
Interv. 3, 1126–1132 (2010).
83. Erdoes, G. et al. Transcranial Doppler-detected
cerebral embolic load during transcatheter aortic
valve implantation. Eur. J. Cardiothorac. Surg. 41,
778–783 (2012).
84. Reinsfelt, B. et al. Transcranial Doppler
microembolic signals and serum marker
evidence of brain injury during transcatheter
aortic valve implantation. Acta Anaesthesiol.
Scand. 56, 240–247 (2012).
85. Kahlert, P. et al. Cerebral embolization during
transcatheter aortic valve implantation: a
transcranial Doppler study. Circulation 126,
1245–1255 (2012).
86. Hynes, B. G. & Rodés-Cabau, J. Transcatheter
aortic valve implantation and cerebrovascular
events: the current state of the art. Ann. NY
Acad. Sci. 1254, 151–163 (2012).
87. Knipp, S. C. et al. Cognitive outcomes three years
after coronary artery bypass surgery: relation to
diffusion-weighted magnetic resonance imaging.
Ann. Thorac. Surg. 85, 872–879 (2008).
88. Ito, A. et al. Postoperative neurological
complications and risk factors for pre-existing
silent brain infarction in elderly patients
undergoing coronary artery bypass grafting.
J. Anesth. 26, 405–411 (2012).
89. Gerriets, T. et al. Evaluation of methods to
predict early long-term neurobehavioral outcome
after coronary artery bypass grafting. Am. J.
Cardiol. 105, 1095–1101 (2010).
90. Bendszus, M. et al. Brain damage after coronary
artery bypass grafting. Arch. Neurol. 59,
1090–1095 (2002).
91. Lund, C. et al. Cerebral ischemic injury and
cognitive impairmentafter off-pump and on-pump
coronary artery bypass grafting surgery.
Ann. Thorac. Surg. 80, 2126–2131 (2005).
92. Wimo, A. et al. The economic impact of dementia
in Europe in 2008-cost estimates from the
Eurocode project. Int. J. Geriatr. Psychiatry 26,
825–832 (2011).
93. Blum, S. et al. Memory after silent stroke:
hippocampus and infarcts both matter.
Neurology 78, 38–46 (2012).
706 | DECEMBER 2013 | VOLUME 10  www.nature.com/nrcardio
94. Arvanitakis, Z., Leurgans, S. E., Barnes, L. L.,
Bennett, D. A. & Schneider, J. A. Microinfarct
pathology, dementia, and cognitive systems.
Stroke 42, 722–727 (2011).
95. Gerriets, T. et al. Protecting the brain from
gaseous and solid micro-emboli during
coronary artery bypass grafting: a randomized
controlled trial. Eur. Heart J. 31, 260–268
(2010).
96. Djaiani, G. et al. Mild to moderate atheromatous
disease of the thoracic aorta and new ischemic
brain lesions after conventional coronary artery
bypass graft surgery. Stroke 35, e356–e358
(2004).
97. Floyd, T. F. et al. Clinically silent cerebral
ischemic events after cardiac surgery: their
incidence, regional vascular occurrence, and
procedural dependence. Ann. Thorac. Surg. 81,
2160–2166 (2006).
98. Selnes, O. A. & McKhann, G. M. Neurocognitive
complications after coronary artery bypass
surgery. Ann. Neurol. 57, 615–621 (2005).
99. Van Dijk, D. et al. Cognitive outcome after
off‑pump and on-pump coronary artery bypass
graft surgery: a randomized trial. JAMA 287,
1405–1412 (2002).
100. European Heart Rhythm Association. Guidelines
for the management of atrial fibrillation: the Task
Force for the Management of Atrial Fibrillation of
the European Society of Cardiology (ESC). Eur.
Heart J. 31, 2369–2429 (2010).
101. Haeusler, K. G. et al. 3 Tesla MRI-detected
brain lesions after pulmonary vein isolation for
atrial fibrillation: results of the MACPAF study.
J. Cardiovasc. Electrophysiol. 24, 14–21
(2013).
102. Herrera Siklódy, C. et al. Incidence of
asymptomatic intracranial embolic events
after pulmonary vein isolation: comparison of
different atrial fibrillation ablation technologies
in a multicenter study. J. Am. Coll. Cardiol. 58,
681–688 (2011).
103. Schwarz, N. et al. Neuropsychological decline
after catheter ablation of atrial fibrillation. Heart
Rhythm. 7, 1761–1767 (2010).
104. Gaita, F. et al., Radiofrequency catheter
ablation of atrial fibrillation: a cause of silent
thromboembolism? Magnetic resonance
imaging assessment of cerebral
thromboembolism in patients undergoing
ablation of atrial fibrillation. Circulation 122,
1667–1673 (2010).
105. Schrickel, J. W. et al. Incidence and predictors of
silent cerebral embolism during pulmonary vein
catheter ablation for atrial fibrillation. Europace
12, 52–57 (2010).
© 2013 Macmillan Publishers Limited. All rights reserved
106. Meier, B. et al. Percutaneous closure of patent
foramen ovale in cryptogenic embolism. N. Engl.
J. Med. 368, 1083–1091 (2013).
107. Carroll, J. D. et al. Closure of patent foramen
ovale versus medical therapy after cryptogenic
stroke. N. Engl. J. Med. 368, 1092–1100 (2013).
108. Skowasch, D. et al. Silent and apparent cerebral
embolism after interventional closure of
symptomatic patent foramen ovale. Int. J. Cardiol.
145, 401–402 (2010).
109. Dorenbeck, U. et al. Cerebral embolism with
interventional closure of symptomatic patent
foramen ovale: an MRI-based study using
diffusion-weighted imaging. Eur. J. Neurol. 14,
451–454 (2007).
110. Bernick, C. et al. Silent MRI infarcts and the
risk of future stroke: the Cardiovascular Health
Study. Neurology 57, 1222–1229 (2001).
111. Choi, S. H. et al. Diffusion-weighted MRI in
vascular dementia. Neurology 54, 83–89
(2000).
112. Vermeer, S. E., Longstreth, W. T. Jr
& Koudstaal, P. J. Silent brain infarcts:
a systematic review. Lancet Neurol. 6, 611–619
(2007).
113. Fujikawa, T., Yanai, I. & Yamawaki, S.
Psychosocial stressors in patients with major
depression and silent cerebral infarction. Stroke
28, 1123–1125 (1997).
114. Fujikawa, T., Yamawaki, S. & Touhouda, Y.
Incidence of silent cerebral infarction in patients
with major depression. Stroke 24, 1631–1634
(1993).
115. Yanai, I., Fujikawa, T., Horiguchi, J., Yamawaki, S.
& Touhouda, Y. The 3‑year course and outcome
of patients with major depression and silent
cerebral infarction. J. Affect. Disord. 47, 25–30
(1998).
116. Fujikawa, T., Yamawaki, S. & Touhouda, Y.
Background factors and clinical symptoms of
major depression with silent cerebral infarction.
Stroke 25, 798–801 (1994).
Acknowledgements
This work was supported by grants from the
Dutch Heart Foundation (2011 T022) and National
Health Insurance Board/ZonMw, Netherlands
(40‑00703‑98‑11629) to R. Delewi.
Author contributions
M. E. C. Hassell researched the data for the article.
M. E. C. Hassell, R. Delewi, and R. Nijveldt
contributed substantially to discussion of the content.
M. E. C. Hassell and R. Delewi contributed
substantially to writing the article. All the authors
reviewed/edited the manuscript before submission.