**What is meant by Perioperative?** "Perioperative" refers to th...
**What is meant by infraction?** An infraction is a violation or...
**What is atrial fibrillation?** Atrial Fibrillation, often abbr...
**What is ischaemic?** The term "ischaemic" or "ischemic" comes ...
**What is atherosclerotic?** "Atherosclerotic" pertains to or is...
**What is extrcranial carotid stenosis?** A medical condition in...
**What is obstructive sleep apnoea?** Obstructive sleep apnea (O...
What is an anticoagulant?
www.thelancet.com/neurology Vol 22 October 2023
Lancet Neurol 2023; 22: 946–58
Department of Health Research
Methods, Evidence, and
Impact, McMaster University,
Hamilton, ON, Canada
(M Marcucci MD,
P J Devereaux PhD); Department
of Medicine, McMaster
University, Hamilton, ON,
Canada (M Marcucci,
P J Devereaux); Population
Health Research Institute,
Hamilton, ON, Canada
(M Marcucci, P J Devereaux);
The Chinese University of
Hong Kong, Hong Kong Special
Administrative Region, China
(M T V Chan PhD); Hotchkiss
Brain Institute, University of
Calgary, Calgary, AB, Canada
(E E Smith MD); Department of
Anaesthesiology, University
Medical Center Groningen,
University of Groningen,
Groningen, Netherlands
(A R Absalom MBChB)
Correspondence to:
Dr P J Devereaux
Population Health Research
Institute, Hamilton, ON L8L 2X2,
Prevention of perioperative stroke in patients undergoing
non-cardiac surgery
Maura Marcucci, Matthew T V Chan, Eric E Smith, Anthony R Absalom, P J Devereaux
About 300 million adults undergo non-cardiac surgery annually. Although, in this setting, the incidence of
perioperative stroke is low, the absolute number of patients experiencing a stroke is substantial. Furthermore, most
patients with this complication will die or end up with severe disability. Covert brain infarctions are more frequent
than overt strokes and are associated with postoperative delirium, cognitive decline, and cerebrovascular events at
1 year after surgery. Evidence shows that traditional stroke risk factors including older age, hypertension, and atrial
fibrillation are also associated with perioperative stroke; previous stroke is the strongest risk factor for perioperative
stroke. Increasing evidence also suggests the pathogenic role of perioperative events, such as hypotension, new atrial
fibrillation, paradoxical embolism, and bleeding. Clinicians involved in perioperative care should be aware of this
evidence on prevention strategies to improve patient outcomes after non-cardiac surgery.
Worldwide, about 300 million adults undergo non-
cardiac surgery annually.
Acute stroke is an infrequent
but devastating complication of non-cardiac surgery,
which often leads to death or major disability.
Perioperative covert brain infarctions (panel 1) are more
frequent than overt strokes and are associated with
postoperative delirium and long-term neurocognitive
These sequelae can have a substantial eect
on patients, caregivers, and health-care systems.
This Review covers current evidence on the
epidemiology, risk factors, and potential mechanisms of
perioperative stroke and covert brain infarction in the
setting of non-cardiac surgery. We also review evidence-
based perioperative strategies for prevention of these
complications, and discuss future directions of
Incidence and clinical relevance
The analyses of a large US administrative database showed
that the incidence of perioperative death and myocardial
infarction decreased between 2004 and 2013; however, the
incidence of perioperative ischaemic stroke increased over
the same time period, across all age groups.
A minority
of perioperative strokes occur intraoperatively; up to
30% of strokes occur by post operative day 2, about
50% occur by postoperative day 5; and 20–25% occur after
the initial hospital stay but within 30 days of surgery.
The incidence of overt stroke in the non-cardiac surgery
setting ranges from 0·03% to 4∙00%.
The Vascular
Events in Noncardiac Surgery Patients Cohort Evaluation
(VISION) study followed up a representative sample of
40 004 patients (aged ≥45 years) who underwent non-
cardiac surgery at 28 centres in 14 countries.
stroke occurred in 132 patients (0·3% [95% CI 0·3–0·4]),
Panel 1: Glossary of terms
We refer to non-cardiac surgery for surgeries that do not
involve the heart (ie, gastrointestinal, gynaecological,
orthopaedic, thoracic, vascular, urological, etc). The evidence
that we cover in this Review comes from studies of patients
undergoing major non-cardiac surgeries that can result in major
physiological insults and are most commonly characterised by
requiring patients to stay in the hospital at least overnight after
their surgery.
We assign the term perioperative to conditions, events, and
actions during the time window from the beginning of surgery
to after surgery (typically 30 days after surgery, if not otherwise
specified). In some instances, we use the term to include also
the period that precedes surgery, to identify a clinical setting, or
a period that is defined by its relation with surgery (ie, around
We use the term preoperative for conditions, events, and
actions that involve the time before surgery. When not
otherwise specified, this is usually the days close to the surgery
or in preparation for surgery.
We use the term intraoperative for conditions, events, and
actions that occur during surgery or, more generically, the time
spent in the operating room.
We use the term postoperative for conditions, events, and
actions that occur after the completion of surgery. In general,
researchers have tended to investigate a 30-day period after
surgery to evaluate the risk of surgery-related events. However,
large nationwide cohort studies have suggested that the risk of
acute cardiovascular and cerebrovascular events remains
increased, compared with matched individuals in control
groups that did not receive surgery, up to 6 weeks after non-
cardiac surgeries.
We refer to overt stroke as a clinically diagnosed acute event of
a neurological deficit lasting more than 24 h, with
neuroimaging detecting a brain infarct or haemorrhage.
We refer to covert strokes as those that are detected on imaging
but not recognised clinically, sometimes due to unappreciated,
subtle, or misclassified manifestations.
www.thelancet.com/neurology Vol 22 October 2023
with higher incidences in patients undergoing vascular
surgery (0·9%) or neurosurgery (1·5%) compared with
patients undergoing other types of non-cardiac surgery.
Perioperative strokes are more commonly ischaemic
than haemorrhagic. The PeriOperative ISchemic
Evaluation-1 (POISE-1) trial included 8351 patients (aged
≥45 years) at 190 hospitals in 23 countries who had, or
were at risk of, atherosclerotic disease, and underwent
non-cardiac surgery.
Among the 60 patients who had a
perioperative overt stroke, more than 80% had an
ischaemic stroke.
Evidence suggests that more strokes are fatal or cause
severe deficits when they occur during or after non-
cardiac surgery, compared with strokes that occur in the
non-operative setting.
More than 20% of patients with
an acute perioperative stroke will die within 30 days of
surgery, whereas less than 2% of the patients who
undergo non-cardiac surgery and do not have a
perioperative stroke will die within 30 days.
More than
50% of people who survive a perioperative stroke will
have severe disability and be discharged to a location
other than home.
Research shows that perioperative
strokes account for a substantial proportion (16–77%) of
in-hospital stroke cases. Furthermore, compared with
community-onset strokes, perioperative strokes show:
greater severity; large vessels or cardio-embolic origin,
with involvement of the anterior circulation; longer times
to neuroimaging and treatment; and lower rates of
intravenous thrombolysis (due to contraindications or
being outside of the therapeutic time window) and of
admission to stroke units.
In a retrospective study of
39 patients with perioperative stroke, 10% (n=4) had a
new focal deficit at emergence from anaesthesia.
Recognition of stroke symptoms was commonly delayed,
with about 15% of the patients presenting with changes
in mental status only.
In the non-operative setting, covert strokes are more
frequent than overt strokes and they are associated with
cognitive decline and functional psychomotor
complications, and an increased risk of subsequent
overt stroke.
Using perioperative neuro imaging,
early studies had suggested that perioperative covert
brain infarcts are much more frequent than
perioperative overt strokes, with incidences of up to 40%
among patients undergoing carotid stenting and
10% among patients undergoing non-vascular
The Neurological Impact of CerebroVascular Events In
Noncardiac Surgery Patients (NeuroVISION) was an
international cohort study of 1114 people (aged ≥65 years)
who underwent elective non-cardiac surgery in 12 centres
in nine countries.
Perioperative covert stroke, which was
detected by diusion-weighted MRI sequences performed
between postoperative day 2 and 9, occurred in all types of
non-cardiac surgeries, with an overall incidence of 7%
(78 of 1114 patients [95% CI 6–9]).
This incidence is
substantially higher than that in the non-operative setting,
where population-based studies reported incidences of
About 130 million people (aged ≥65 years),
without a previous diagnosis of dementia,
undergo non-cardiac surgery worldwide annually
300 million adults undergo non-cardiac surgery
worldwide annually
About 1·3 million have a
perioperative stroke
About 9·1 million have a
perioperative covert brain
1 year
since surgery
30 days
since surgery
1·00% (range 0·03–4·00) 7% (95% CI 6–9)
20% (95% CI 14–28) 80% (95% CI 72–86) 10% (95% CI 4–19)
64% (95% CI 35–87) 81% (95% CI 72–88)
42% (95% CI 30–54)
22% (95% CI 18–27)
89 (95% CI 87–90)
23% (95% CI 21–26)
About 260
000 people will
die due to perioperative stroke
About 1·04 million
people will be alive at 30 days
About 910
000 patients will
have delirium
000 will be alive at 1 year
About 1·4 million patients will have severe
disability or require institutionalisation
3·8 million will have a ≥2 point MoCA decline
About 836
000 will have a loss in autonomy in
one or more activities of daily living
About 213
000 patients will
get a diagnosis of dementia
Figure 1: Epidemiology and the impact of perioperative stroke in patients undergoing non-cardiac surgery
Percentage estimates were obtained from the studies of Ng and colleagues,
Mrkobrada and colleagues,
Spence and colleagues,
Devereaux and colleagues,
Gleason and colleagues,
Pendlebury and Rothwell,
and Koton and colleagues.
Estimates at 1 year assume that 1-year outcomes after perioperative stroke occur
with the same proportions and timing as in the non-operative population. MoCA=Montreal Cognitive Assessment.
www.thelancet.com/neurology Vol 22 October 2023
acute or subacute covert strokes of less than 1% on brain
MRI diusion-weighted images that were performed
randomly or for reasons dierent from clinically
suspected stroke or transient ischaemic attack.
NeuroVISION was designed to investigate the
association of perioperative covert stroke with cognitive
changes 1 year after surgery. 1 year after surgery, 42% of
participants who had a perioperative covert stroke versus
29% of participants who did not have a perioperative
covert stroke had cognitive decline, defined as a at least
a two-point decrease in the Montreal Cognitive
Assessment compared with baseline (adjusted odds
ratio [OR] 1·98 [95% CI 1·22–3·20]).
Also at 1 year, more
patients with a perioperative covert stroke experienced
a decline in their performance on the Digit-Symbol
Substitution Test and the Trail-Making Test Part B.
Delirium in the first 3 days after surgery occurred in
10% of participants with a perioperative covert stroke
versus in 5% of participants without a perioperative
covert stroke (adjusted hazard ratio [HR] 2·24
[95% CI 1·06–4·73]). In a secondary analysis of a subset
of NeuroVSION participants, participants with a peri-
operative covert stroke had higher peak neuro filament
light levels and higher delirium severity scores than
participants without a perioperative covert stroke.
stroke was also associated with an increased risk of overt
stroke or transient ischaemic attack in the subsequent
year (HR 4·13 [1·14–14·99]).
In figure 1, we present a summary of the epidemiology
and impact of perioperative stroke in the setting of non-
cardiac surgery.
Risk factors and pathogenesis
Unless otherwise specified, in this section, we will refer
to overt strokes exclusively, as it is uncertain whether
perioperative covert brain infarctions share the same
mechanisms and risk factors. The table provides a
summary of the findings from large cohort studies of
patients undergoing various non-cardiac surgeries and
in which multivariable analyses identified predictors of
perioperative stroke.
Age and other risk factors
Older patients (aged ≥65 years) are more likely to have
major cardiovascular complications after non-cardiac
surgery, including stroke, than younger patients.
have consistently shown that a history of stroke or
transient ischaemic attack is an independent predictor of
perioperative stroke (table). Chronic kidney disease,
hypertension, diabetes, coronary artery disease, and
smoking are also associated with an increased risk of
perioperative stroke.
There is a paucity of data on the risk of perioperative
stroke in people with extracranial carotid stenosis.
Two retrospective studies in non-cardiac, non-carotid
surgery (one including 224 patients and the other
including 2110 patients) did not find an association
between the severity of the carotid artery stenosis and the
risk of perioperative stroke.
Patients with symptomatic
intracranial artery stenosis have about a 15% risk of stroke
over the subsequent year.
Although it might be assumed
that these patients would have a higher risk of stroke
than 15% if they undergo non-cardiac surgery, their actual
risk is unknown.
Type of surgery and anaesthetic techniques
The risk of perioperative stroke is higher after emergency
surgery than after elective surgery. Certain types of
surgery (eg, neurosurgery and vascular surgery) carry a
higher risk than other types of surgery, independent of
the increased risk associated with the medical history of
patients under going these surgeries (table).
large randomised controlled trials have investigated the
eects of dierent anaesthetic techniques on the
incidence of perioperative cardiovascular events in
patients undergoing non-cardiac surgery. The ENIGMA II
trial randomly assigned 7112 patients to receive an
anaesthetic mixture with nitrous oxide—which is known
to irreversibly inhibit vitamin B12 and its metabolism—at
an inspired concentration of 70%, or to a nitrous oxide-
free general anaesthesia; the primary outcome was a com-
posite of death and cardiovascular complications within
30 days of surgery.
The BALANCED Anaesthesia
study com pared the eect of light anaesthesia (bispectral
index target 50) versus deep anaesthesia (bispectral index
target 35) on 1-year all-cause mortality in 6644 patients
(aged ≥60 years) undergoing surgeries with expected
durations of at least 2 h (6531 [98%] of the 6644 patients
who underwent non-cardiac surgery).
Both trials did not
show a dierence on their primary outcome.
In both
trials, perioperative stroke was evaluated as a secondary
outcome and occurred with an incidence of about 1%,
with no dierence between the two study groups.
Two trials—the REGAIN trial,
including 1600 patients
and 12 stroke events, and the RAGA trial,
950 patients and one stroke event—that compared hip
fracture surgery under regional versus general
anaesthesia did not show a dierence in perioperative
stroke events between the two groups.
Atrial fibrillation
A history of atrial fibrillation has been identified as a risk
factor for perioperative stroke but,
plausible, it remains uncertain whether atrial fibrillation
is associated with silent brain infarctions. The incidence
of new perioperative atrial fibrillation in patients
undergoing non-cardiac surgery varies according to the
type of surgery and the definition of perioperative
atrial fibrillation.
In the VISION cohort, 370 of
40 004 patients (0·9% [95% CI 0·8–1·0]) had new atrial
fibrillation that resulted in angina, congestive heart
failure, symptomatic hypotension, or that required
pharmacological treatment or cardioversion.
In patients
who undergo non-cardiac surgery with major
www.thelancet.com/neurology Vol 22 October 2023
Study design, source of data Number of patients,
number of stroke events
Risk factors* OR† (95% CI)
Platzbecker and colleagues
Retrospective study, data
from a registry of the
Massachusetts General
Hospital and two community
hospitals in Massachusetts,
USA, for surgeries performed
between 2007 and 2015
249 480,‡ 2257 events
(within 1 year after
Age (eg, >70 years vs ≤40 years);
Sex (male vs female);
Race or ethnicity (Black vs other);
History of ischaemic stroke or TIA ≤3 months before surgery;
History of ischaemic stroke or TIA >3 months before surgery;
Carotid artery stenosis;
Patent foramen ovale;
Atrial fibrillation;
Valvular heart disease;
ASA physical status (ASA ≥2 vs ASA <2);
Emergency vs elective surgery;
Neurosurgery (cranial) vs other surgery;
Vascular surgery vs other surgery
3·03 (2·34–3·93);
1·22 (1·08–1·38);
1·51 (1·16–1·93);
47·94 (36·97–61·56);
12·18 (9·30–15·96);
3·22 (2·46–4·18);
2·48 (1·92–3·25);
1·93 (1·26–2·97);
1·31 (1·12–1·54);
1·32 (1·16–1·52);
1·39 (1·15–1·70);
2·32 (1·97–2·69);
1·63 (1·28–2·10);
4·31 (3·71–4·95);
1·68 (1·35–2·10)
Woo and colleagues (2021)
Retrospective study, data
from the ACS-NSQIP database
for surgeries performed
between 2007 and 2010
1 165 750,‡ 2005 events
(within 30 days after
Age (per year);
History of stroke;
Preoperative low haematocrit (0-27% vs >27%);
Preoperative high creatinine (>1·8 mg/dL vs ≤1·8 mg/dL);
ASA physical status (eg, ASA 3 vs ASA 1);
Emergency vs elective surgery;
Thoracic endovascular aneurysm repair§ vs other abdominal surgery;
Abdominal endovascular aneurysm repair§ vs other abdominal surgery;
Carotid endarterectomy§ vs other abdominal surgery;
Neurosurgery (cranial)§ vs other abdominal surgery;
Liver, pancreas, or spleen surgery§ vs other abdominal surgery
1·03 (1·02–1·03);
2·30 (2·08–2·55);
1·29 (1·08–1·55);
1·41 (1·24–1·60);
8·51 (4·20–17·25);
1·89 (1·68–2·12);
15·11 (10·98–20·79);
2·17 (1·59–2·96);
5·19 (4·36–6·17);
11·91 (8·88–15·96);
2·41 (1·82–3·19)
Mashour and colleagues
Retrospective study, data
from the ACS-NSQIP database
for surgeries performed
between 2005 and 2008
523 059,‡ 714 events
(within 30 days after
Age (≥62 years vs <62 years);
BMI (≥35 kg/m
vs <35 kg/m);
Current smoker;
History of stroke;
History of TIA;
Myocardial infarction within 6 months;
Acute renal failure;
3·9 (3·0–5·0);
0·6 (0·4–0·9);
1·5 (1·1–1·9);
2·9 (2·3–3·8);
1·9 (1·3–2·6);
3·8 (2·4–6.0);
3·6 (2·3–5·8);
2·3 (1·6–3·4);
2·0 (1·6–2·6);
1·8 (1·4–2·4)
POISE Study Group (2008)
Prospective cohort based on
an RCT population, data from
patients recruited between
October, 2002, and July, 2007,
across 23 countries
8351, 60 events (within
30 days after surgery)
History of stroke or TIA;
Use of clopidogrel or ticlopidine in 24 h before surgery;
Clinically significant hypotension;
Clinically significant bleeding;
New clinically significant atrial fibrillation
2·80 (1·66–4·73);
3·12 (1·43–6·77);
2·14 (1·15–3·96);
2·18 (1·06–4·49);
3·51 (1·45–8·52)
Bateman and colleagues
Retrospective study, data
from the US Nationwide
Sample hospital database for
surgeries performed between
2000 and 2004
371 641,‡ 1601 events
(in hospital)
Age (per 10 years);
Sex (female vs male);
History of stroke;
Atrial fibrillation;
Congestive heart failure;
Renal disease;
Valvular disease
1·43 (1·35–1·51);
1·21 (1·07–1·36);
1·64 (1·25–2·14);
1·95 (1·69–2·26);
1·18 (1·01–1·39);
1·44 (1·21–1·70);
2·98 (2·52–3·54);
1·54 (1·25–1·90)
ACS-NSQIP=American College of Surgeons National Surgical Quality Improvement Program. ASA=American Society of Anesthesiologists physical status classification. COPD=chronic obstructive pulmonary
disease. RCT=randomised controlled trial. TIA=transient ischemic attack. *Only statistically significant predictors are shown. †Measure of association in multivariable analyses. ‡Include both derivation and
validation cohorts. §The study also reported risk estimates for
other types of surgery (we selected to report the most common and relevant types of surgery [and with statistically significant effect]).
Table: Large cohort studies of patients undergoing non-cardiac surgery, that have evaluated the risk factors associated with stroke through multivariable analyses
cardiovascular risk factors or disease, the incidence of
new atrial fibrillation is higher (ie, 2–3%).
were substantial in studies that used continuous electro-
cardiogram monitoring.
Regardless of whether atrial
fibrillation is a true perioperative event or a newly
detected paroxysmal event, perioperative atrial fibrillation
has been associated with perioperative stroke, and a
long-term increased risk of incident stroke, similar to
the risk associated with atrial fibrillation detected in a
non-operative setting.
In the POISE-1 trial, patients
with new perioperative atrial fibrillation had an adjusted
OR for stroke at 30 days of 3·51 (95% CI 1·45–8·52).
cardioembolic aetiology might explain the severity of
many perioperative strokes.
In NeuroVISION, silent
perioperative infarcts were found in the cortex of 35 (45%)
of 78 patients and 10 (13%) patients with covert brain
infarcts had multiple acute brain infarcts.
These features
suggest an embolic mechanism.
www.thelancet.com/neurology Vol 22 October 2023
Patent foramen ovale
Non-cardiac surgery can increase the risk of paradoxical
embolism in the presence of patent foramen ovale
through changes in pulmonary and systemic pressures
favouring right-to-left shunting and an increased risk of
thrombotic and non-thrombotic (eg, fat) embolic
A systematic review pub lished in
2022 summarised the evidence on the association
between patent foramen ovale and peri operative stroke,
and reported an adjusted OR of 5·49 (95% CI 1·62–18·6),
but there was substantial hetero geneity (I²=100%) across
studies. These studies also found that patients with
patent foramen ovale had an increased risk of stroke,
with higher National Institute of Health Stroke Scale
(NIHSS) scores and greater in-hospital mortality
compared with patients with perioperative stroke without
patent foramen ovale.
All three studies included in the
systematic review included diagnosed patent foramen
ovale based on International Classification of Diseases
One study, that included only patients who had
an echocardiogram, showed a much higher proportion of
patients with stroke (35·1% in patients with and 6·0% in
patients without patent foramen ovale or atrial septal
than in the other two studies (eg, in a study by
Ng and colleagues,
stroke occurred in 3·2% of patients
with and 0·5% of patients without patent foramen ovale).
Due to their retrospective methodology and the use of
administrative data, these studies were susceptible to
selection and ascertainment biases.
Perioperative haemodynamics
Perioperative hypotension is common. In the VISION
study, 28% of patients had clinically significant
hypotension (ie, a systolic blood pressure <90 mm Hg,
requiring an intervention) during non-cardiac surgery
and 19% of patients had clinically significant hypotension
after non-cardiac surgery.
In a Dutch nested case-
control study that used data from the electronic health
records of patients undergoing non-cardiac surgery, and
included 42 patients with stroke and 252 controls
matched by age and type of surgery, the time during
surgery spent with a mean blood pressure lower than
30% of the patient baseline was associated with the
occurrence of postoperative stroke, after adjusting for
potential confounders.
A post-hoc analysis of the
POISE-1 trial showed that clinically significant
postoperative hypotension was the strongest predictor of
perioperative stroke, with an adjusted OR of 2·14
(95% CI 1·15–3·96), and an attributable risk of 15%.
Two retrospective analyses of data from two large US
cohorts (>350 000 non-cardiac surgeries) based on
hospital administrative databases, included 1553 patients
who had a perioperative stroke within 1 week of surgery
and 2482 perioperative strokes within 30 days after
surgery, respectively.
In multivariable analyses, these
studies did not find an association between intraoperative
or post operative hypotension and perioperative ischaemic
By contrast, a case-control study using US
admini strative data (122 confirmed patients with stroke
and 496 patients without stroke in the control group)
showed that intraoperative hypotension—defined as time
and reduction below key mean arterial blood pressure
thresholds, according to the total area under the curve
approach—was significantly associated with postoperative
stroke within 30 days after surgery.
Potential explanations
for the inconsistency of these results include the use of
dierent hypotension definitions, dierent timing and
modality of blood pressure monitoring, and dierent
types of data and data collection.
Three randomised controlled trials have evaluated
intraoperative blood pressure targets in patients at
high-risk of perioperative complications undergoing
non-cardiac surgery; however, none of these trials was
adequately powered to detect dierences in major
vascular complications including stroke.
Perioperative bleeding and transfusions
Perioperative bleeding might lead to cerebrovascular
events through a decline in haemoglobin concentration
and a negative eect on haemodynamics. Perioperative
bleeding might also delay restarting chronic
antithrombotic drugs, which could further increase the
risk of stroke. In an analysis of data from 651 775 adults
who underwent non-cardiac surgery at several US
centres (1321 stroke events), Kamel and colleagues
found that patients who had a major haemorrhage
(ie, bleeding necessitating transfusion of >4 units of
blood) had an adjusted HR of 2·7 (95% CI, 2·1–3·4) for
stroke within 30 days of surgery, compared with patients
who did not suer a major haemorrhage. There was a
dose–response relationship between the number of units
transfused intraoperatively and the risk of perioperative
stroke. In more than two-thirds of patients, bleeding
preceded vascular events by at least 1 day.
significant bleeding was also one of the independent
predictors of perioperative stroke in the POISE-1 trial
(adjusted OR 2·18 [95% CI 1·06–4·49]).
In another
study, that included 12 786 patients undergoing elective
carotid endarterectomy, receiving at least one blood
transfusion intraoperatively was associated with an
adjusted OR of 4·75 (1·75–12·86) for stroke within
30 days (stroke incidence 6·1% in patients who received
a blood transfusion vs 1·4% in patients who did not
receive a blood transfusion).
Indeed, the link between perioperative bleeding and
stroke might be mediated by the requirement for blood
transfusions, at least in part. Perioperative red-blood-cell
transfusions have been associated with thrombotic
A Cochrane systematic review of 48 trials
(21 433 participants) evaluating transfusion strategies in
surgical and non-surgical settings did not find an eect
on the risk of stroke (risk ratio [RR] 0·84 [95% CI
However, a systematic review including
only trials in patients with hip fracture (3371 patients and
www.thelancet.com/neurology Vol 22 October 2023
who had 41 strokes) found a protective eect for
perioperative stroke with a restrictive transfusion strategy
(RR 0·40 [95% CI 0·19–0·86]).
Other potential pathogenic mechanisms or risk factors
Many patients undergo non-cardiac surgery to treat
cancer or its complications, and cancer is associated with
a hypercoagulability state. A secondary analysis of the
Reasons for Geographic and Racial Dierences in Stroke
(REGARDS) study (n=6602) found that, com pared with
participants without cancer, participants who were newly
diagnosed with cancer were at increased risk of incident
cerebrovascular events during follow-up (0·52% vs 0·05%,
adjusted HR 6·6 [95% CI 2·7–16·0]). This association
was only present in the first 30 days after diagnosis
(ie, when patients more often receive active treatments,
including surgery) and not beyond 30 days.
Anecdotal case reports have described patients with
perioperative stroke secondary to acute vertebral artery
dissection, possibly due to perioperative procedures
involving the manipulation of cervical structures
(eg, intubation, positioning, central venous cannulation,
and spine surgery) in people who might be anatomically
susceptible or underwent particularly arduous
Risk stratification and prevention strategies
Prediction models can help estimate an individual’s risk
of perioperative stroke. Two models have been developed
using data from large administrative databases. The
model by Woo and colleagues
provides risk adjusted
estimates for the absolute risk of stroke within 30 days of
non-cardiac surgery; however, this model does not include
potentially important risk factors like atrial fibrillation,
carotid stenosis, and patent foramen ovale among the
predictors. These predictors are included in the model by
Platzbecker and colleagues;
however, their model
predicts the risk of stroke in the year following surgery
and the appropriateness of its use for a perioperative risk
evaluation is unclear. Caution is recommended with the
use of both these models, because they have not been
validated externally nor prospectively.
A preoperative 12-lead electro cardiogram is
recommended in many guidelines and can be used to
screen for preoperative atrial fibrillation. The most recent
guideline regarding screening for carotid stenosis in
asymptomatic patients was published by the US
Preventive Services Task Force in 2021. Based on the
most current evidence showing uncertain benefit of
revascularisation in patients who are asymptomatic and
the risks associated with revascularisation, the guideline
recommends against screening such patients.
In the
absence of specific evidence for non-cardiac surgery, this
indirect evidence provides the best insights regarding
screening before non-cardiac surgery. Scientific
guidelines vary in their recom mendation to use
echocardiography for preoperative cardiovascular risk
and there is no current evidence or
guidance regarding the eectiveness and cost-
eectiveness of screening for patent foramen ovale for
primary prevention in patients at high risk of stroke
through paradoxical embolism. Taken as a whole, there is
insucient evidence to support preoperative screening
of carotid stenosis or patent foramen ovale. However,
many patients will have had a previous echocardiogram
or carotid ultrasound, which should be reviewed at
preoperative assessment (together with an assessment of
the reasons for which those exams had been done).
Time of surgery
Patients with a previous stroke are at increased risk of
perioperative stroke; however, the risk changes
substantially based on the time elapsed from the previous
stroke. In a large Danish cohort study (481 183 elective
non-cardiac surgeries done in adults in 2005–11, with
7137 previous strokes), patients who had a stroke within
the 3 months before surgery had an adjusted
OR of 67·6 (95% CI 52·3–87·4) for postoperative
ischaemic stroke within 30 days after non-cardiac surgery
compared with patients who have never had a previous
The OR became 24·0 (15·0–38·4) for patients
who had a stroke between 3 months and 6 months before
non-cardiac surgery and 10·4 (6·2–17·4) for patients who
had a stroke between 6 months and 12 months before
non-cardiac surgery.
Cubic spline analyses showed that
the increased risk of postoperative stroke levelled o for
patients with strokes that occurred 9 months or more
before non-cardiac surgery.
Using data from 5·8 million Medicare beneficiaries
(aged >65 years) between 2011 to 2018, a study by Glance
and colleagues
confirmed that the risk of perioperative
stroke varies with the time of surgery from the previous
stroke; however, they found much lower incremental
risks (ie, patients with a history of stroke within 30 days
of surgery had an 8-fold higher risk of perioperative
stroke than patients without a previous stroke); moreover,
the risk associated with a previous stroke levelled o
when the previous stroke occurred more than 90 days
before, suggesting that delaying elective surgery until
6–9 months after a previous stroke, as per the American
Heart Association statement based on the Danish study,
might appear too conservative.
Improvements in
secondary stroke prevention might explain the dierent
findings in the more recent US cohort; however,
dierences between the two studies remain mostly
unexplained. In the clinical management of a patient
who had a stroke, any decision to delay elective surgery
until 3 months, 6 months, or 9 months after a previous
stroke, should always be balanced against the expected
harms resulting from delaying the planned surgery.
Atherosclerosis, anticoagulation, and bleeding
In the POISE-1 trial (n=8351), more patients randomly
assigned to receive a perioperative beta-blocker
www.thelancet.com/neurology Vol 22 October 2023
(ie, metoprolol succinate) had a stroke (41 [1·0%]
vs 19 [0·5%] patients; HR 2·17 [95% CI, 1·26–3·74]) than
those who received placebo.
In the POISE-2 trial
(n=10 004), the continuation or initiation of aspirin for
patients with known or at risk of atherosclerotic disease
who were undergoing non-cardiac surgery did not reduce
the risk of major cardiovascular complications at 30 days
after surgery, including stroke (0·3% of patients in the
aspirin group vs 0·4% of patients in the placebo group;
HR 0·84 [95% CI 0·43–1·64]), but aspirin did increase
major bleeding (4·6% vs 3·8%; 1·23 [1·01–1·49])
compared with placebo.
The results were consistent
whether or not patients were taking aspirin before the
Several small and heterogeneous randomised con-
trolled trials have evaluated the eects of statin on
perioperative cardiovascular complications in patients
undergoing non-cardiac surgery. Two meta-analyses of
these trials (the most comprehensive one including
12 trials and 4707 patients) found that perioperative statin
initiation or loading reduced the incidence of post-
operative myocardial infarction, but had no significant
eect on stroke (nine events).
Observational data
suggest that preoperative hyperglycaemia increases the
risk of perioperative stroke in people with diabetes
undergoing non-cardiac surgery;
however, trials on
intense glucose control found no benefit in preventing
perioperative myocardial infarction or stroke (although
all data on stroke events come from trials in cardiac
All these findings reveal that current evidence
does not support the initiation or intensification of these
pharmacological therapies in the perioperative period to
prevent stroke in non-cardiac surgery. However, when
patients have to wait for weeks or months till their
Panel 2: Case study: a woman with perioperative overt stroke
A 99-year-old woman with dementia was admitted to our
hospital for repeat surgery of her right hip, after a failed open
reduction and internal fixation for a hip fracture 3 months
earlier. She had a medical history of atrial fibrillation
-VASc score of 2 for hypertension and age), for which
she was on warfarin (international normalised ratio [INR]
range 2·0–3·0). She also had obstructive sleep apnoea but was
non-compliant with treatment by use of continuous positive
airway pressure. Even after the previous surgery for her hip
fracture, she was ambulatory with a walker, requiring assistance
only with instrumental activities of daily living.
After an uneventful new surgery, the patient was admitted to
the cardiac care unit (CCU) for oxygen and cardiac monitoring.
She was discharged from the CCU on the first postoperative
day to the general ward. Her haemoglobin was 106 g/L on
postoperative day 1, 96 g/L on postoperative day 2, and
remained stable at approximately 80 g/L while in hospital. Her
postoperative troponin measurements were normal. The
patient was mobilised and her rehabilitation progressed well.
Warfarin was resumed on postoperative day 2, with the plan of
continuing prophylaxis with low molecular weight heparin
until the INR reached a value at least 1·8. On postoperative
day 6, when her INR was still 1·4, the patient was found drowsy
in her hospital bed, with conjugate sight deviation toward the
right. A CT showed a large infarct of the right middle cerebral
artery consistent with an ASPECTS
of 5 (figure 2). Compared
with previous scans, an increased prominence of a focal
hyperdensity was noted at the branch point of the M2 arteries,
suspicious for thrombus. As per her family’s request, the
patient was discharged home on postoperative day 9 for
palliative care and she died 2 days later.
Figure 2: Postoperative head CT of the patient described in panel 2
(A) Non-contrast head CT showing large acute infarction of the right middle cerebral artery territory. (B) Thrombus is evident in the right middle cerebral artery
(dot sign; arrow). (C) Possible thrombus of the left middle cerebral artery (dot sign; arrow).
www.thelancet.com/neurology Vol 22 October 2023
surgery (in particular, when a decision to delay surgery
is made), it is reasonable to consider this time as an
opportunity to optimise the control of atherosclerotic
risk factors that have been associated with an increased
incidence of stroke also perioperatively (eg, hypertension,
diabetes, obesity, and smoking).
Between 3% and 8% of patients in cohorts undergoing
major non-cardiac surgery are on chronic therapeutic
oral anticoagulation for conditions that increase the risk
of thromboembolic stroke (eg, atrial fibrillation).
Most surgeries discussed in this Review are those
typically classified as at high risk of bleeding,
most current risk estimates for major or prognostically
important bleeding at 30 days from surgery above 5%;
some (eg, laparoscopic procedures) will follow under the
low to moderate risk category (ie, ≤2% major bleeding
For patients scheduled to undergo these types
of surgeries, scientific guidelines recommend temporary
interruption of oral anticoagulation based on algorithms
specific to the type of oral anticoagulant, which, in
addition to the surgery-related risk of bleeding, considers
the patient-specific risk of thromboembolic compli-
cations; other relevant factors are patient renal function
and the type of anaesthesia.
About one fourth to one third of patients on chronic
oral anticoagulation and scheduled for major non-
cardiac surgery would receive vitamin K antagonists (ie,
The pharmacological properties of these
agents add challenges to their perioperative discon-
tinuation and resumption, which has led to the use and
evaluation of bridging anticoagulation strategies.
The BRIDGE trial (n=1884), done at 108 centres in
Canada and the USA, was the first placebo-controlled
randomised trial to evaluate perioperative low-
molecular weight heparin bridging inpatients with
atrial fibrillation who had their chronic therapy with
warfarin interrupted 5 days before non-cardiac
Not bridging was non-inferior to bridging
for the incidence of arterial thromboembolic events
(4 patients [0·4%] in the placebo group vs 3 patients
[0·3%] in the bridging group; p=0·01 for non-
inferiority). By contrast, the incidence of major bleeding
was 1·3% in the placebo group and 3·2% in the
bridging group (RR 41 [95% CI 0·20–0·78]).
PERIOP2 trial, including 1471 patients with atrial
fibrillation and mechanical heart valves (21%), found
similar results.
This trial assessed a composite of
arterial and venous thromboembolic endpoints, but the
study was stopped early due to insucient funding.
On the basis of this evidence, routine perioperative
Figure 3: Postoperative brain MRI of the patient described in panel 3
Arrows indicate three foci of acute infarction, (A) two within the frontal lobes bilaterally and (B) one in the left
parietal lobe.
Panel 3: Case study: a woman with perioperative covert brain infarction
A 73-year-old woman, who was a retired teacher, was admitted to
our hospital for right periprosthetic hip fracture secondary to a
fall. She had a history of dyslipidaemia and gastroesophageal
reflux disease, and she was taking atorvastatin and pantoprazole.
She was independent for basic and instrumental activities of daily
living and had never smoked. Her surgery was uncomplicated. As
planned during the preoperative evaluation, she was transferred
to the ward after surgery with continuous oxygen monitoring,
based on her high risk of obstructive sleep apnoea (STOP-Bang
score of 5).
Postoperatively, the patient remained haemodynamically stable
and started physiotherapy. During routine bedside assessment
on postoperative day 4, the patient had some word-finding
difficulties—which her family noted as unusual—with no clear
attention deficit. A brain MRI showed three small foci of acute
infarction within the bilateral frontal lobes and left parietal lobe
(figure 3). The lesions were not visible on a CT angiography
performed later on the same day. The CT angiography showed a
calcified atherosclerotic plaque at the right carotid bifurcation
resulting in a 50% stenosis, but without stenosis of the left
carotid; the proximal vessels of the circle of Willis were patent.
The word-finding difficulty resolved in the subsequent hours,
and no other neurological deficit was noted. Telemetry was
ordered, which demonstrated no rhythm or rate abnormalities
in the following 3 days in hospital. An echocardiogram showed
normal global and regional left ventricular systolic function and
a normal right ventricle. The left atrium was severely enlarged.
The contrast study with agitated saline at rest and post-Valsalva
manoeuvre showed no interatrial shunting, and Doppler
ultrasound showed no interatrial or interventricular
communication. The neurologist suggested monotherapy with
an antiplatelet agent, with indication for 2-week Holter
monitoring after discharge. Her length of stay was further
prolonged by COVID-19 infection. She was discharged home
with her husband, deconditioned but stable, more than
2 months after her surgery.
www.thelancet.com/neurology Vol 22 October 2023
anti coagulation bridging is not recommended.
However, the surgeries that patients underwent in
these trials were mostly low-risk, non-cancer ambu-
latory procedures. Moreover, most of the patients had a
-VASc score of 2 or less and only 9% of
patients in the BRIDGE trial and 12% of patients in the
PERIOP2 trial
had a previous stroke. Finally, both
trials had challenges recruiting patients, in part due to
the increased use of direct oral anticoagulants over
warfarin. It should be noted that direct oral
anticoagulants oer the advantages of a more rapid and
relatively more predictable onset and washout of their
Patient requiring non-cardiac surgery
1. Preoperative stroke risk stratification*
• Sociodemographics (eg, age, ethnicity, and smoking)
Type of surgery
• Medical history (eg, cerebrovascular disease, atrial fibrillation, diabetes, myocardial infarction, obesity, hypertension, renal disease, and valvular disease)
• Blood results (eg, haemoglobin or haematocrit, creatinine, glucose, and glycated haemoglobin)
• Electrocardiogram (eg, atrial fibrillation and signs of ischaemic or hypertensive cardiomyopathy)
• Review echcocardiogram if available (eg, for ischaemic or hypertensive cardiomyopathy, valcular disease, and patent foramen ovale)†
• Review carotid ultrasound if available†
High risk of perioperative stroke
2. Risk optimisation
Preoperative Intraoperative Postoperative
If patient had previous stroke and delaying surgery
is feasible and acceptable, postpone surgery to
3–6 months from stroke‡
Minimise haemodynamic perturbations||
Minimise bleeding**
Monitoring for new atrial fibrillation††
Weeks or months to
<1 week to surgery
Optimise control
of atherosclerotic
risk factor§
Start perioperative
management in
patients receiving long-
term anticoagulation
Discontinue aspirin¶
Continue perioperative anticoagulation or resume aspirin, as appropriate, according to daily reassessment of
bleeding and thromboembolic risk
Optimise control of atherosclerotic risk factors§
Figure 4: Evidence-based clinical approach to perioperative stroke in non-cardiac surgery
*Risk factors listed in the figure are those found associated with an increased risk of perioperative stroke in large cohorts of patients undergoing inpatient non-
cardiac surgery, in multivariable analyses (see table 1). †Currently there is no evidence to support the implementation of preoperative screening with
echocardiography or carotid ultrasonography. However, if a recent echocardiogram or carotid ultrasound is available in the patient records, knowledge of findings
that have been associated with an increased risk of stroke or perioperative stroke might inform patient risk stratification and optimisation. ‡An approach supported
by findings from two large retrospective cohort studies based on administrative data showing a higher risk of perioperative stroke in patients with a previous stroke
who undergo elective non-cardiac surgery earlier than 3–6 months from the event.
No randomised controlled trial has addressed the question of surgery timing in
patients with previous stroke. §Current experimental evidence does not support the initiation or intensification of pharmacological cardiovascular therapies in the
immediate perioperative period for lack of efficacy or safety concerns. However, in the weeks leading to the surgery date, it is reasonable to apply the evidence from
the non-operative setting showing the efficacy and safety of stroke prevention strategies and assume that they might help reduce the baseline risk of perioperative
stroke. The same suggestion applies to the postoperative period, once the surgery-related risks of bleeding, hypotension, and hypoglycaemia subside, given the fact
that the risk of perioperative stroke remains high even days after surgery. This suggestion intends also to highlight the significance of the perioperative moment as an
opportunity for risk stratification and long-term risk optimisation. ¶An approach supported by the results of the large POISE-2 trial
showing no reduction of
cardiovascular complications and increase of bleeding with continuing aspirin perioperatively. ||Multivariable analyses from a large prospective cohort and case-
control studies support the association between perioperative hypotension and stroke.
Randomised trials on different intraoperative blood pressure targets were
not powered to show differences in cardiovascular outcomes.
**An approach supported by observational evidence on the association between bleeding or
transfusions and perioperative stroke in large prospective and retrospective cohorts;
and by a systematic review of randomised trials in patients with hip fracture
showing a reduction in perioperative stroke with a restrictive transfusion strategy.
††An approach supported by substantial observational evidence on the
association between new perioperative atrial fibrillation and an increased risk of stroke at 30 days and 1 year after non-cardiac surgery.
www.thelancet.com/neurology Vol 22 October 2023
anticoagulation eect than warfarin, and allow for
possible reversal, which are advantages that are
particularly relevant in the perioperative setting.
Comprehensive updated guidelines for perioperative
anticoagulation management were published in 2022.
We emphasise that decisions on if and when to interrupt
and resume anticoagulation perioperatively should be
made on the basis of the careful assessment of patient-
specific and surgery-specific thromboembolic and
bleeding risks. Crucially, these risks should be reassessed
daily after surgery. This daily assessment requires the
participation of the surgeon, whenever feasible, and the
consideration of patient’s values.
The potential association of perioperative bleeding and
transfusions with stroke requires the adoption of
strategies to minimise these complications. In an
inter national randomised trial (POISE-3), among
9535 patients undergoing non-cardiac surgery at
increased risk of cardiovascular events and bleeding, the
admini stration of 2 g of intravenous tranexamic acid
(1 g bolus given at the beginning and 1 g bolus given at
the end of surgery) was associated with 25% relative risk
reductions in major bleeding outcomes and transfusion
require ment, compared with placebo.
Tranexamic acid
as an antifibrinolytic drug has been suggested to be a
pro-thrombotic drug; however, in POISE-3, tranexamic
acid did not increase the risk of stroke and other
perioperative cardiovascular outcomes.
Conclusions and future directions
Although a small proportion of patients undergoing
non-cardiac surgery will have a stroke, this complication
can have a substantial eect on their prognosis (case
studies in panel 2 and figure 2 and panel 3 and
figure 3).
Considering the volume of patients undergoing non-
cardiac surgery, and especially those undergoing surgery
at an old age, the projected burden of perioperative
stroke at the population level is high. In these
circumstances, general internists, neurologists,
geriatricians, anaes thesi o logists, surgeons, and every
clinician involved in perioperative care should be
equipped with up-to-date evidence regarding the
epidemiology, pathogenesis, risk stratification, and
optimisation of perioperative stroke. In figure 4, we
suggest a clinical evidence-based approach for an adult
patient undergoing non-cardiac surgery that includes
stroke prevention strategies through risk stratification
and risk optimisation.
Given the relatively low incidence of perioperative
stroke, large prospective cohorts and randomised trials
are paramount to fill the current gaps in knowledge
about the pathogenic role of perioperative events
(eg, atrial fibrillation, paradoxical embolism, and
haemodynamic abnormalities) and to inform clinical
practice regarding the screening, prevention, and
treatment of perioperative stroke. The findings on a
relatively high perioperative incidence of silent but
Search strategy and selection criteria
We searched PubMed using MeSH terms, free text, and keywords. We did not limit the
dates of our searches but gave preference to articles published since 2016. We did not
apply language restrictions to our search, but we planned to selectively look for help with
the translation of non-English full texts based on the content of the abstract, if provided
in English. Our search involved several combinations of the following terms: “(non-
cardiac/noncardiac) surgery”, “perioperative/postoperative stroke”, “covert/silent brain
infarct/infarction”, “covert/silent stroke”, “neurological complications/disorders”.
Additional terms were added to build ad-hoc searches related to specific topics, including
“(perioperative/postoperative) atrial fibrillation”, “hypotension”, “hypertension”, “patent
foramen ovale”, “carotid stenosis”, “diabetes”, “hyperglycaemia”, “hypoglycaemia”,
“bleeding”, “transfusions”, “bridging anticoagulation”, “(intraoperative/perioperative/
postoperative) monitoring”. We also reviewed references from original research articles
and reviews to identify additional studies. This Review is not a systematic review and
might not be inclusive of all published articles. The final list of included references was
selected to accommodate the objectives of our Review.
clinically relevant brain infarctions might pave the way
for a better understanding of the nature and role of
perioperative cerebrovascular events and might trigger
further research in the field. The discovery of blood
(eg, a brain troponin) or digital biomarkers (eg, subclinical
changes in psychomotor functions) of perioperative
brain infarctions would facilitate screening. Furthermore,
more research is needed on the preoperative, intra-
operative, and postoperative strategies that could reduce
the eects of perioperative brain infarctions on cognition
and functional decline.
MM searched and selected the references and wrote the first draft of the
Review. PJD, MTVC, EES, and ARA helped to retrieve relevant
references, established the content and structure of the Review, and
contributed to its writing and revision. All authors approved the final
version of the paper.
Declaration of interests
MM has been co-investigator in investigator-initiated research projects
supported by grants or in-kind contributions from Cloud DX and Roche
Diagnostics. ARA’s research group has received grants and funding
from The Medicines Company and Masimo and has performed
sponsor-initiated work for The Medicines Company and Rigel. ARA has
received honoraria and grants from Paion, The Medicines Company,
Terumo, BD (Carefusion), Philips, Ever Pharma, and Johnson &
Johnson (Janssen Pharmaceutica). PJD is a member of a research group
with a policy of not accepting honorariums or other payments from
industry for their own personal financial gain. PJD’s research group do
accept honorariums or payments from industry to support
researchendeavours and costs to participate in meetings. Based on
study questions PJD originated andgrants he has written, he has
received grants from Abbott Diagnostics, AstraZeneca, Bayer,
Boehringer Ingelheim, Bristol-Myers-Squibb, Cloud DX, Coviden,
Octapharma, Philips Healthcare, Roche Diagnostics, Siemens, and
Stryker. PJD has also participated in an advisory board meeting for
GlaxoSmithKline, Bayer, Quidel Canada, and Trimedic and an expert
panel meeting with AstraZeneca, Boehringer Ingelheim, and Roche.
MTVC and EES declare no competing interests.
MM holds a Physicians’ Services Incorporated Foundation Mid-Career
Clinical Research Award for her work on perioperative vascular and
neurocognitive outcomes. PJD holds a Tier-1 Canada Research Chair in
Perioperative Care.
www.thelancet.com/neurology Vol 22 October 2023
1 Weiser TG, Haynes AB, Molina G, et al. Estimate of the global
volume of surgery in 2012: an assessment supporting improved
health outcomes. Lancet 2015; 385 (suppl 2): S11.
2 Devereaux PJ, Yang H, Yusuf S, et al. Eects of extended-release
metoprolol succinate in patients undergoing non-cardiac surgery
(POISE trial): a randomised controlled trial. Lancet 2008;
371: 1839–47.
3 Vascular Events in Noncardiac Surgery Patients Cohort Evaluation
(VISION) Study Investigators, Spence J, LeManach Y, et al.
Association between complications and death within 30 days after
noncardiac surgery. CMAJ 2019; 191: E830–E 7.
4 Lalmohamed A, Vestergaard P, Cooper C, et al. Timing of stroke in
patients undergoing total hip replacement and matched controls:
a nationwide cohort study. Stroke 2012; 43: 3225–29.
5 Lalmohamed A, Vestergaard P, Klop C, et al. Timing of acute
myocardial infarction in patients undergoing total hip or knee
replacement: a nationwide cohort study. Arch Intern Med 2012;
172: 1229–35.
6 NeuroVISION Investigators. Perioperative covert stroke in patients
undergoing non-cardiac surgery (NeuroVISION): a prospective
cohort study. Lancet 2019; 394: 1022–29.
7 Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P,
van Gool WA. Delirium in elderly patients and the risk of
postdischarge mortality, institutionalization, and dementia:
a meta-analysis. JAMA 2010; 304: 443–51.
8 Steinmetz J, Christensen KB, Lund T, Lohse N, Rasmussen LS,
Group I. Long-term consequences of postoperative cognitive
dysfunction. Anesthesiology 2009; 110: 548–55.
9 Smilowitz NR, Gupta N, Ramakrishna H, Guo Y, Berger JS,
Bangalore S. Perioperative major adverse cardiovascular and
cerebrovascular events associated with noncardiac surgery.
JAMA Cardiol 2017; 2: 181–87.
10 Banco D, Dodson JA, Berger JS, Smilowitz NR. Perioperative
cardiovascular outcomes among older adults undergoing in-hospital
noncardiac surgery. J Am Geriatr Soc 2021; 69: 2821–30.
11 Ng JL, Chan MT, Gelb AW, Warner DS. Perioperative stroke in
noncardiac, nonneurosurgical surgery. Anesthesiology 2011;
115: 879–90.
12 Lisabeth LD, Brown DL, Zahuranec DB, et al. Temporal trends in
ischemic stroke rates by ethnicity, sex, and age (2000–2017):
The Brain Attack Surveillance In Corpus Christi Project. Neurology
2021; 97: e2164–72.
13 Kim J, Thayabaranathan T, Donnan GA, et al. Global stroke
statistics 2019. Int J Stroke 2020; 15: 819–38.
14 Woo SH, Marhefka GD, Cowan SW, Ackermann L. Development
and validation of a prediction model for stroke, cardiac, and
mortality risk after non-cardiac surgery. J Am Heart Assoc 2021;
10: e018013.
15 Bateman BT, Schumacher HC, Wang S, Shaefi S, Berman MF.
Perioperative acute ischemic stroke in noncardiac and nonvascular
surgery: incidence, risk factors, and outcomes. Anesthesiology 2009;
110: 231–38.
16 Dong Y, Cao W, Cheng X, et al. Risk factors and stroke characteristic
in patients with postoperative strokes. J Stroke Cerebrovasc Dis 2017;
26: 1635–40.
17 Saltman AP, Silver FL, Fang J, Stamplecoski M, Kapral MK. Care
and outcomes of patients with in-hospital stroke. JAMA Neurol
2015; 72: 749–55.
18 Emmett ES, Douiri A, Marshall IJ, Wolfe CDA, Rudd AG, Bhalla A.
A comparison of trends in stroke care and outcomes between
in-hospital and community-onset stroke—The South London
Stroke Register. PLoS One 2019; 14: e0212396.
19 Cummings S, Kasner SE, Mullen M, et al. Delays in the
identification and assessment of in-hospital stroke patients.
J Stroke Cerebrovasc Dis 2022; 31: 106327.
20 Vlisides PE, Mashour GA, Didier TJ, et al. Recognition and
management of perioperative stroke in hospitalized patients.
A A Case Rep 2016; 7: 55–56.
21 Debette S, Beiser A, DeCarli C, et al. Association of MRI markers of
vascular brain injury with incident stroke, mild cognitive
impairment, dementia, and mortality: the Framingham Ospring
Study. Stroke 2010; 41: 600–06.
22 Longstreth WT Jr, Dulberg C, Manolio TA, et al. Incidence,
manifestations, and predictors of brain infarcts defined by serial
cranial magnetic resonance imaging in the elderly:
the Cardiovascular Health Study. Stroke 2002; 33: 2376–82.
23 Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ,
Breteler MM. Silent brain infarcts and the risk of dementia and
cognitive decline. N Engl J Med 2003; 348: 1215–22.
24 Vermeer SE, Longstreth WT Jr, Koudstaal PJ. Silent brain infarcts:
a systematic review. Lancet Neurol 2007; 6: 611–19.
25 Conen D, Rodondi N, Müller A, et al. Relationships of overt and
silent brain lesions with cognitive function in patients with atrial
fibrillation. J Am Coll Cardiol 2019; 73: 9–99.
26 Gargiulo G, Sannino A, Stabile E, Perrino C, Trimarco B,
Esposito G. New cerebral lesions at magnetic resonance imaging
after carotid artery stenting versus endarterectomy: an updated
meta-analysis. PLoS One 2015; 10: e0129209.
27 Mrkobrada M, Hill MD, Chan MT, et al. Covert stroke after
non-cardiac surgery: a prospective cohort study. Br J Anaesth 2016;
117: 191–97.
28 Arnold Fiebelkorn C, Vemuri P, Rabinstein AA, et al. Frequency of
acute and subacute infarcts in a population-based study.
Mayo Clin Proc 2018; 93: 300–06.
29 Yamada K, Nagakane Y, Sasajima H, et al. Incidental acute infarcts
identified on diusion-weighted images: a university hospital-based
study. AJNR Am J Neuroradiol 2008; 29: 937–40.
30 Batool S, O’Donnell M, Sharma M, et al. Incidental magnetic
resonance diusion-weighted imaging-positive lesions are rare in
neurologically asymptomatic community-dwelling adults. Stroke
2014; 45: 2115–17.
31 Taylor J, Eisenmenger L, Lindroth H, et al. Perioperative
ischaemic brain injury and plasma neurofilament light: a secondary
analysis of two prospective cohort studies. Br J Anaesth 2022;
130: e361–69.
32 Spence J, LeManach Y, Chan MTV, et al. Vascular Events in
Noncardiac Surgery Patients Cohort Evaluation Study I. Association
between complications and death within 30 days after noncardiac
surgery. CMAJ 2019; 191: E830–37.
33 Gleason LJ, Schmitt EM, Kosar CM, et al. Eect of delirium and
other major complications on outcomes after elective surgery in
older adults. JAMA Surg 2015; 150: 1134–40.
34 Pendlebury ST, Rothwell PM, Oxford Vascular S. Incidence and
prevalence of dementia associated with transient ischaemic attack
and stroke: analysis of the population-based Oxford Vascular Study.
Lancet Neurol 2019; 18: 248–58.
35 Koton S, Pike JR, Johansen M, et al. Association of ischemic stroke
incidence, severity, and recurrence with dementia in the
atherosclerosis risk in communities cohort study. JAMA Neurol
2022; 79: 271–80.
36 Platzbecker K, Grabitz SD, Raub D, et al. Development and external
validation of a prognostic model for ischaemic stroke after surgery.
Br J Anaesth 2021; 127: 713–21.
37 Mashour GA, Shanks AM, Kheterpal S. Perioperative stroke and
associated mortality after noncardiac, nonneurologic surgery.
Anesthesiology 2011; 114: 12–96.
38 Evans BA, Wijdicks EF. High-grade carotid stenosis detected before
general surgery: is endarterectomy indicated? Neurology 2001;
57: 1328–30.
39 Sonny A, Gornik HL, Yang D, Mascha EJ, Sessler DI. Lack of
association between carotid artery stenosis and stroke or myocardial
injury after noncardiac surgery in high-risk patients. Anesthesiology
2014; 121: 922–29.
40 Chimowitz MI, Lynn MJ, Howlett-Smith H, et al. Comparison of
warfarin and aspirin for symptomatic intracranial arterial stenosis.
N Engl J Med 2005; 352: 1305–16.
41 Nunn JF. Clinical aspects of the interaction between nitrous oxide
and vitamin B12. Br J Anaesth 1987; 59: 3–13.
42 Myles PS, Leslie K, Chan MT, et al. The safety of addition of nitrous
oxide to general anaesthesia in at-risk patients having major non-
cardiac surgery (ENIGMA-II): a randomised, single-blind trial.
Lancet 2014; 384: 1446–54.
43 Short TG, Campbell D, Frampton C, et al. Anaesthetic depth and
complications after major surgery: an international, randomised
controlled trial. Lancet 2019; 394: 1907–14.
www.thelancet.com/neurology Vol 22 October 2023
44 Neuman MD, Feng R, Carson JL, et al. Spinal anesthesia or general
anesthesia for hip surgery in older adults. N Engl J Med 2021;
385: 2025–35.
45 Li T, Li J, Yuan L, et al. Eect of regional vs general anesthesia on
incidence of postoperative delirium in older patients undergoing
hip fracture surgery: the RAGA randomized trial. JAMA 2022;
327: 50–58.
46 Kikura M, Oikawa F, Yamamoto K, et al. Myocardial infarction and
cerebrovascular accident following non-cardiac surgery: dierences
in postoperative temporal distribution and risk factors.
J Thromb Haemost 2008; 6: 742–48.
47 Sharifpour M, Moore LE, Shanks AM, Didier TJ, Kheterpal S,
Mashour GA. Incidence, predictors, and outcomes of perioperative
stroke in noncarotid major vascular surgery. Anesth Analg 2013;
116: 424–34.
48 McIntyre WF, Vadakken ME, Rai AS, et al. Incidence and
recurrence of new-onset atrial fibrillation detected during
hospitalization for non-cardiac surgery: a systematic review and
meta-analysis. Can J Anaesth 2021; 68: 1045–56.
49 Conen D, Alonso-Coello P, Douketis J, et al. Risk of stroke and
other adverse outcomes in patients with perioperative atrial
fibrillation 1 year after non-cardiac surgery. Eur Heart J 2020;
41: 645–51.
50 Butt JH, Olesen JB, Havers-Borgersen E, et al. Risk of
thromboembolism associated with atrial fibrillation following
noncardiac surgery. J Am Coll Cardiol 2018; 72: 2027–36.
51 Higuchi S, Kabeya Y, Matsushita K, et al. Incidence and
complications of perioperative atrial fibrillation after non-cardiac
surgery for malignancy. PLoS One 2019; 14: e0216239.
52 Siontis KC, Gersh BJ, Weston SA, et al. Associations of atrial
fibrillation after noncardiac surgery with stroke, subsequent
arrhythmia, and death : a cohort study. Ann Intern Med 2022;
175: 1065–72.
53 Smith EE, Shobha N, Dai D, et al. Risk score for in-hospital
ischemic stroke mortality derived and validated within the
Get With the Guidelines-Stroke Program. Circulation 2010;
122: 1496–504.
54 Elgendy AY, Saver JL, Amin Z, et al. Proposal for updated
nomenclature and classification of potential causative mechanism
in patent foramen ovale-associated stroke. JAMA Neurol 2020;
77: 878–86.
55 Yan C, Li H. Preliminary investigation of in situ thrombus within
patent foramen ovale in patients with and without stroke. JAMA
2021; 325: 2116–18.
56 Hobbes B, Akseer S, Pikula A, et al. Risk of perioperative stroke in
patients with patent foramen ovale: a systematic review and meta-
analysis. Can J Cardiol 2022; 38: 11–200.
57 Ng PY, Ng AK, Subramaniam B, et al. Association of preoperatively
diagnosed patent foramen ovale with perioperative ischemic stroke.
JAMA 2018; 319: 452–62.
58 Villablanca PA, Lemor A, So CY, et al. Increased risk of
perioperative ischemic stroke in patients who undergo noncardiac
surgery with preexisting atrial septal defect or patent foramen ovale.
J Cardiothorac Vasc Anesth 2020; 34: 2060–68.
59 Smilowitz NR, Subashchandran V, Berger JS. Atrial septal defect
and the risk of ischemic stroke in the perioperative period of
noncardiac surgery. Am J Cardiol 2019; 124: 1120–24.
60 Roshanov PS, Rochwerg B, Patel A, et al. Withholding versus
continuing angiotensin-converting enzyme inhibitors or
angiotensin II receptor blockers before noncardiac surgery:
an Analysis of the Vascular events In noncardiac Surgery patIents
cOhort evaluatioN Prospective Cohort. Anesthesiology 2017;
126: 16–27.
61 Bijker JB, Persoon S, Peelen LM, et al. Intraoperative hypotension
and perioperative ischemic stroke after general surgery: a nested
case-control study. Anesthesiology 2012; 116: 658–64.
62 Wongtangman K, Wachtendorf LJ, Blank M, et al. Eect of
intraoperative arterial hypotension on the risk of perioperative
stroke after noncardiac surgery: a retrospective multicenter cohort
study. Anesth Analg 2021; 133: 1000–08.
63 Gregory A, Stapelfeldt WH, Khanna AK, et al. Intraoperative
hypotension is associated with adverse clinical outcomes after
noncardiac surgery. Anesth Analg 2021; 132: 1654–65.
64 Smischney NJ, Shaw AD, Stapelfeldt WH, et al. Postoperative
hypotension in patients discharged to the intensive care unit after
non-cardiac surgery is associated with adverse clinical outcomes.
Crit Care 2020; 24: 682.
65 Khanna AK, Shaw AD, Stapelfeldt WH, et al. Postoperative
hypotension and adverse clinical outcomes in patients without
intraoperative hypotension, after noncardiac surgery. Anesth Analg
2021; 132: 1410–20.
66 Vlisides PE, Mentz G, Leis AM, et al. Carbon dioxide, blood