British Journal of Anaesthesia 95 (1): 3–19 (2005)

doi:10.1093/bja/aei063 Advance Access publication January 21, 2005

REVIEW ARTICLES
Perioperative myocardial infarction—aetiology and prevention
H.-J. Priebe*

University Hospital/Department of Anaesthesia, Hugstetter Strasse 55,

D-79106 Freiburg, Germany
*E-mail: [email protected]
Perioperative myocardial infarction (PMI) is one of the most important predictors of short- and
long-term morbidity and mortality associated with non-cardiac surgery. Prevention of a PMI is
thus a prerequisite for an improvement in overall postoperative outcome. The aetiology of PMI is
multifactorial. The perioperative period induces large, unpredictable and unphysiological
alterations in coronary plaque morphology, function and progression, and may trigger a mismatch
of myocardial oxygen supply and demand. With many diverse factors involved, it is unlikely that
one single intervention will successfully improve cardiac outcome following non-cardiac surgery.
A multifactorial, step-wise approach is indicated. Based on increasing knowledge of the nature of
atherosclerotic coronary artery disease, and in view of the poor positive predictive value of
non-invasive cardiac stress tests, and the considerable risk of coronary angiography and coronary
revascularization in high-risk patients, the paradigm is shifting from an emphasis on extensive
non-invasive preoperative risk stratification to a combination of selective non-invasive testing
and aggressive pharmacological perioperative therapy. Perioperative plaque stabilization by
pharmacological means may be as important in the prevention of PMI as an increase in myocardial
oxygen supply or a reduction in myocardial oxygen demand.
Br J Anaesth 2005; 95: 3–19
Keywords: complications, cardiac; recovery

Perioperative myocardial infarction (PMI) is one of the most composition, and biological activity of coronary plaques
important predictors of short- and long-term morbidity and appear to be closely linked.96 Plaque instability correlated
mortality associated with non-cardiac surgery.80 97 98 132 more with biological activity and cellular composition than
Prevention of a PMI is thus a prerequisite for the improve- with angiographical findings.90 The interaction between
ment in overall postoperative outcome. The design of morphological and functional factors is unpredictable.
effective preventive measures requires basic knowledge Exogenous factors (e.g. mechanical stress, vasomotor
of the aetiology of PMI. Unfortunately, the exact nature tone, infection, blood viscosity, coagulability) further
of PMI remains an area of uncertainty and the subject of modify such interaction, making the final outcome even
continued debate and controversy.80 97 132 Accordingly, the less predictable. Systemic or multi-focal arterial inflamma-
first part of this article will address the aetiology of PMI, tion may be independent risk factors for acute coronary
with consideration of the aetiology and pathophysiology of events,19 103 supporting the broader concept of the ‘vulner-
acute coronary syndromes, and of the diagnosis of MI able patient’.115
(recently reviewed in detail in reference132). The second
part will address strategies and means of preventing a PMI. Pathophysiology
Plaque progression is frequently abrupt, mostly unpredict-
able,19 and often related to episodes of thrombosis (which, in
Aetiology of PMI turn, are triggered by plaque rupture, erosion, endothelial
activation, or inflammation). In the absence of a hypercoa-
Acute coronary syndrome gulable state, thrombi may remain mural rather than become
Aetiology occlusive, and may thus produce few if any symptoms (un-
Acute coronary syndromes (ACS) are associated with struc- less they embolize).102 If subsequent lysis is incomplete and
turally as well as functionally complex plaques and coronary is followed by re-endothelialization, the plaque will grow.
artery stenoses, coronary endothelial lesions, and plaque The unpredictability of plaque progression is probably
inflammation.19 24 42 47 75 103 Structural morphology, cellular related to fluctuations in risk factors and triggers, for

# The Board of Management and Trustees of the British Journal of Anaesthesia 2005. All rights reserved. For Permissions, please e-mail: [email protected]
 Priebe

example physical activity, mental stress, environmental resulting in total interruption of nutrient blood flow to
temperature, smoking, infection, hydration, and arterial the myocardium. The situation may be worsened by
pressure; to heterogeneity of plaque histology; and to distal embolization of microthrombi and by coronary
differences in the physical forces to which plaques are vasoconstriction caused by local mediator release or
exposed.18 41 43 56 103 systemic sympathetic activation. If coronary blood flow is
It is impossible to predict the time it will take the vulner- interrupted for longer than 30 min, a MI may result. Per-
able plaque to become unstable, or the trigger that causes sistent coronary artery occlusion will cause a progressive
the plaque to rupture (i.e. mechanical stress, coronary increase in infarct size. Loss of functional myocardium
vasospasm, widespread acute inflammatory endothelial results in impaired left ventricular function, which may
activation, or the chronic inflammatory component of impair quality of life, and usually leads to premature death.13
atherosclerosis). In a substantial percentage of culprit Any discussion of the aetiology of PMI must take into
lesions, thrombosed plaques without detectable fissures account the extreme variations in clinical presentation of
were observed.29 In such cases, plaque vulnerability is prob- ACS in general. At one end of the spectrum are those
ably caused by thrombogenic or high-risk blood and/or local patients who suffer a sudden cardiac death or an MI without
pro-inflammatory cytokines that trigger thrombosis, some- any preceding episode of angina, and not followed by
times even in the absence of inflammatory cell infiltration recurrent instability. At the other end of the spectrum are
and a lipid core. those patients who develop an MI after episodes of unstable
Rupture of the intimal surface of a plaque is the result of angina over a period of days to weeks, and who often
a combination of cellular processes that promote plaque develop post-infarction angina and/or re-infarction. It is
instability, and of physical (haemodynamic) processes conceivable that the triggers of instability differ between
that influence the magnitude and distribution of stress on such groups of patients.
the plaque. The size of the thrombus that forms at the site of
plaque rupture and the clinical consequences will depend on Perioperative MI
several key factors: the degree of plaque disruption (ulcera- Aetiology
tion, fissure, or erosion) and substrate exposure as a major There is pathological and angiographic evidence that the
determinant of thrombogenicity at the local coronary artery aetiology of PMI resembles that in the non-surgical set-
site; the composition of the plaque; the magnitude of the ting.21 30 In PMI, acute plaque disruption and haemorrhage
stenosis; and the extent of platelet activation and intrinsic in the infarct-related coronary artery seems to be com-
fibrinolytic activity.47 mon,21 30 106 but the severity of underlying coronary artery
Plaque rupture is more common during various kinds of stenosis does not necessarily predict the infarct territory.30
strenuous physical activity and emotional stress.111 Activa- The high incidence of histologically confirmed transmural
tion of the sympathetic nervous system in these situations infarctions21 seems to be in contradiction to the ECG finding
leads to increased plasma concentrations of catecholamines, of almost exclusively non-Q-wave PMIs. On the other hand,
blood viscosity, and of arterial pressure and heart rate, which the presence of circumferential PMIs21 is consistent with a
are accompanied by detectable increases in platelet aggrega- myocardial oxygen supply/demand mismatch being the
tion and decreases in fibrinolytic activity that both tend to main trigger of myocardial injury. However, myocardial
favour thrombosis.60 This combination of increased pro- oxygen supply/demand mismatch and plaque rupture are
thrombotic and reduced fibrinolytic activity could initiate not mutually exclusive mechanisms, and MIs may develop
propagation and total occlusion of the coronary artery by a by different mechanisms at different locations in the
mural thrombus overlying a small plaque erosion that might same patient.
otherwise have been harmless. The perioperative period is Patients who experience a PMI have angiographic evi-
characterized by comparable adrenergic stimulation, and dence of extensive coronary artery disease (CAD).39
increased prothrombotic and reduced fibrinolytic activity. Various angiographic findings were consistent with perio-
In the event of plaque rupture, thrombus growth depends perative plaque rupture at sites other than critically narrowed
not only on the size and thrombogenicity of the fissured coronary artery stenoses, as well as with the possibility that
plaque, but also on the number and activation of exposed in some patients with severe but stable CAD, PMI might
inflammatory cells.54 Inflammatory activation of the endo- have developed primarily on the basis of prolonged
thelium can turn its physiological vasodilatory and anti- myocardial ischaemia.39 85
thrombotic properties into pathological vasoconstrictor Most (>80%) PMIs occur early after surgery, are asymp-
and prothrombotic properties. In addition, the inflammatory tomatic,10 of the non-Q-wave type (60–100%),10 46 81 83 and
response of the circulating blood may activate coagula- are most commonly preceded by ST-segment depression
tion.138 These many variables explain why coronary lesions rather than ST-segment elevation.80–82 95 107 124 127 Long-
that are angiographically fairly small may progress acutely duration (single duration >20–30 min or cumulative duration
to severe stenosis or total occlusion. >1–2 h) rather than merely the presence of postoperative
When intraluminal thrombi attach to a ruptured plaque, ST-segment depression, seems to be the important factor
total occlusion of an epicardial coronary artery may occur associated with adverse cardiac outcome.81 107 126 ECG

4
 Perioperative myocardial infarction

evidence of myocardial ischaemia was strongly associated stasis.52 53 Some patients with stenotic atherosclerotic
with an initial postoperative low troponin level and lesions may develop acute MI without evidence of plaque
conventional subsequent increases in serum troponin rupture and superimposed thrombus formation. This may
concentration.82 The frequent combination of increases in happen if there is a marked decrease in myocardial oxygen
heart rate preceding the ischaemic episodes, ST-segment supply (e.g. prolonged severe coronary vasospasm), or a
depression rather than elevation during all ischaemic epi- marked increase in myocardial oxygen demand (e.g. tachy-
sodes; non-Q-wave rather than Q-wave MIs in almost all cardia). It is thus conceivable that coronary thrombosis in the
cases; the lack of angiographically visible thrombus or rup- postoperative setting can be the consequence rather than the
tured plaques in some patients who underwent coronary cause of prolonged myocardial ischaemia and PMI.
angiography following PMI; and complete reversal of Most ischaemic episodes tend to start at the end of surgery
ECG changes to baseline in all but one of the patients and during emergence from anaesthesia.85 This period is
with ischaemia (including those with infarction),85 are characterized by increases in heart rate, arterial pressure,
highly suggestive that prolonged stress-induced myocardial sympathetic tone, and procoagulant activity.16 Increased
ischaemia is the likely primary cause of PMI. Repeated brief sympathetic tone can result in increases in arterial pres-
ischaemic episodes may well have a cumulative effect and sure, heart rate, contractility, coronary vasomotor tone, and
ultimately cause myocardial necrosis.55 coronary vascular shear stress. This, in turn, may trigger
Although ST-segment depression usually reflects suben- coronary vasospasm, plaque disruption, and coronary throm-
docardial ischaemia and is often regarded as reversible bosis. Increases in arterial pressure, heart rate, and cardiac
injury, it is not inconsistent with an MI. During the early contractility lead to subendocardial ischaemia by increasing
evolution of an MI, significant ST-segment elevation may be myocardial oxygen demands in the presence of limited or
lacking.74 For that reason, in current clinical practice, acute absent coronary vasodilator reserve as a result of underlying
MI is divided into ST-segment and non-ST-segment eleva- CAD. Surgery-induced simultaneous procoagulant and anti-
tion MI (which ultimately develop with little cross-over into fibrinolytic activity may trigger coronary artery thrombosis
Q-wave and non-Q-wave MI, respectively).74 In most stud- during low-flow conditions in the presence of underlying
ies on perioperative cardiac ischaemic events, the popula- stable CAD even in the absence of acute plaque disruption.
tions consisted largely of elderly patients. Thus, prolonged The ultimate fate of the thrombus and, thus, the extent of
ST-segment depression may reflect ongoing myocardial jeopardized myocardium will depend on the duration and
ischaemia (ultimately leading to MI), or it may reflect the degree of coronary occlusion. If the plaque disruption is
beginning of an evolving MI. major with extensive exposure of thrombogenic core mater-
Not all investigations found an association between post- ial to the blood stream, acute total coronary occlusion with
operative cardiac complications and long-duration ST- subsequent MI, or sudden death may develop. If the disrup-
segment depression, or between changes in heart rate and tion is minor, the forming thrombus can be non-occlusive
postoperative ST-segment changes or troponin release. Such and the patient may stay asymptomatic or develop unstable
findings would suggest either non-ischaemic causes of ST- angina or a non-Q-wave infarction. A concomitant increase
segment depression in the perioperative period (e.g. hyper- in coagulability and coronary vasoconstriction (as is com-
ventilation, electrolyte changes, drug effects, positional mon in the perioperative setting) may, however, transform a
changes), compensatory mechanisms in response to myo- non-occlusive thrombus to an occlusive thrombus. Ulti-
cardial ischaemia (e.g. preconditioning as a result of mul- mately, the balance between thrombosis and thrombolysis,
tiple brief episodes of myocardial ischaemia and coronary and the flow conditions (affected by coronary vasomotor
reperfusion), or functional collateral perfusion.46 137 tone, perfusion pressure, and rheological properties) are
The preponderance of non-Q-wave infarctions is clearly the decisive factors in determining whether the clinical
different from the non-surgical setting. This again might sug- outcome will be myocardial ischaemia or an MI.
gest that PMIs are more often the result of prolonged isch-
aemia than of thrombotic occlusion, similar to the presumed Diagnosis
pathophysiology of silent ischaemia.141 However, presence A satisfactory explanation of the aetiology of PMI depends
or absence of a Q-wave are not determined primarily by pres- greatly on reliable data on the association between various
ence or absence of an MI or by the transmural nature of the variables and the occurrence of PMI. Such data can only be
underlying MI but rather by the total size of the MI.7 112 The obtained if the detection of PMI is quantitatively and qua-
probability of a Q-wave infarction increases with MI size and litatively reliable. However, fundamental questions remain
the number of transmural segments. Transmural infarctions regarding the definition and diagnostic criteria of MI in
were of the non-Q-wave type in 29% of 100 consecutive general,5 and perioperatively in particular.97 According to
patients with documented previous MI.112 In addition, the the definition of the World Health Organization (WHO), at
Q-wave takes time to develop and, accordingly, does not least two of three criteria must be fulfilled to diagnose
figure strongly in present acute-management decisions. MI: (i) typical ischaemic chest pain; (ii) increased serum
In the presence of severe but stable CAD, coronary throm- concentration of creatine kinase (CK)-MB isoenzyme;
bosis may result from a decrease in coronary blood flow and and (iii) typical electrocardiographic findings, including

5
 Priebe

development of pathological Q-waves.152 160 Perioperative cTnT >0.1 ng ml 1), and 23% (low level cut-off values
MI is mostly silent, and the electrocardiogram (ECG) is of cTnI >0.6 and/or cTnT >0.03 ng ml 1). Only 5.6% of
often difficult to interpret and frequently does not exhibit patients fulfilled the revised definition of MI (presence of at
characteristic ST-segment elevation or Q-waves.17 85 There- least two of three criteria: prolonged chest pain, elevated
fore, if the diagnosis of MI is based solely on the classical CK-MB of cTn, ischaemic ECG changes). However, with-
triad, considerable under-reporting of the true incidence of out routine measurements of serum concentrations of bio-
PMI is to be expected, possibly obscuring the aetiology chemical markers and continuous ECG monitoring for 3
of PMI. postoperative days, MI would have been diagnosed in
The development of assays for the cardiac troponins T only those 3.6% of patients who experienced prolonged
(cTnT) and I (cTnI) that are highly specific and sensitive for chest pain or symptoms of congestive heart failure. Simi-
myocardial injury formed the basis of a revised definition of larly, whereas 12% of patients had increased cTnT concen-
MI as proposed by the European Society of Cardiology and trations during routine postoperative monitoring, only 3%
the American College of Cardiology.4 150 Either of the two had a PMI by the WHO definition.78
following criteria satisfy the diagnosis of an acute, evolving, The question remains whether a reported incidence of
or a recent MI: (1) typical rise and gradual fall in cardiac perioperative myocardial injury based on traditional defini-
troponin concentrations or more rapid rise and fall of CK- tion underestimates the true incidence of clinically relevant
MB concentration in combination with at least one of the myocardial injury, or whether a reported incidence based on
following: (a) typical ischaemic symptoms, (b) development serum concentrations of cardiac troponins overestimates it.
of pathological Q-waves in the ECG, (c) ECG changes When using exclusively biochemical markers, specificity
indicative of myocardial ischaemia (ST-segment elevation may be sacrificed for sensitivity.5 117 Another question is
or depression), or (d) coronary artery intervention; and (2) whether biochemical marker-defined myocardial injury car-
pathological findings of an acute MI. ries the same predictive value as traditionally defined infarc-
Debate continues as to the appropriate cut-off values of tions, and whether mechanisms and triggers are identical in
troponin concentrations for defining a clinically relevant MI. both cases. Irrespective of whether one refers to small
Initial cut-off values (cTnI >1.5 ng ml 1 and cTnT >0.1 ng increases in serum concentrations of troponin as ‘myocardial
ml 1 for certain assays) were derived from titration of tro- infarction’ or ‘subclinical myocardial injury’ or ‘at risk’,
ponin concentrations to a population of patients with clin- even minor increases in serum concentrations of troponins
ically diagnosed MI. However, even small increases in (cTnI >0.6 and/or cTnT >0.03 ng ml 1) and CK-MB
serum concentrations of cardiac troponins are associated (CK>170 iu and CK-MB/total CK >5%) during the first 3
with adverse cardiac outcome in patients with or without postoperative days were associated with approximately
ST-segment elevation ACS.73 92 Considering the high 50–100% increases in long-term mortality following
specificity of cardiac troponins for myocardial cell injury, major vascular surgery (follow-up period 1–5 years, mean
the recent consensus document of the European Society of 32 months).86 Larger, conventional increases in troponin
Cardiology and the American College of Cardiology concentration (cTnI >1.5 and/or cTnT >0.1 ng ml 1) and
Committe on the re-definition of MI states that in the CK-MB (CK >170 iu and CK-MB/total CK >10%) were
presence of documented myocardial ischaemia, even associated with 2-fold and almost 4-fold higher long-term
minor increases in troponin serum concentration to greater mortality, respectively.86 Postoperative increases in cTnT
than the 99th percentile of the normal population should >0.1 ng ml 1 correlated with postoperative cardiac events
be regarded as MI. As most troponin assays still lack (admission for unstable angina, non-fatal MI, cardiac death)
adequate precision at such low concentrations, slightly within the first 6 months following non-cardiac surgery;95
higher cut-off values based on <10% imprecision are postoperative increases in cTnT more than 0.02 ng ml 1
recommended.4 150 conferred a 15-fold increase in 1-yr mortality in elderly
In the frequent absence of typical symptoms and ECG patients undergoing non-cardiac surgery;123 and routine
signs of acute MI, the diagnosis of PMI has to rest heavily on cTnI measurements during the first 3 postoperative days
changes in biochemical markers. Cardiac troponins appear enabled prediction of all cause mortality within the first
to be better suited to identify PMI than the CK-MB isoen- 6 months following vascular surgery.78 Furthermore, pro-
zyme.1 85 88 110 The first study using cTnT in the diagnosis of longed postoperative myocardial ischaemia and increases in
PMI utilized a cut-off value for cTnT serum concentration postoperative cardiac troponin concentrations correlated
of 3.1 ng ml 1.1 Subsequent studies utilized cut-off values of strongly.85 86 Postoperative myocardial ischaemic episodes
0.2 and 0.1 ng ml 1,88 95 and as low as cTnI >0.6 and/or of more than 30 min and more than 60 min were, in turn,
cTnT >0.03 ng ml 1.82 The following example will dem- associated with 2.6- and 3.7-fold increases in long-term
onstrate the dilemma of defining the ‘correct’ incidence of mortality, respectively.86 Taken together, existing evidence
PMI. In the same study,82 depending on the biochemical clearly suggests that even small increases in serum concen-
marker and the cut-off values, the overall incidence of trations of cardiac troponins in the perioperative period
PMI varied between 2.8% (CK-MB >10%), 9% (conven- reflect clinically relevant myocardial injury with short-
tional cut-off values of cTnI >1.5 ng ml 1 and/or and long-term consequences on outcome. Perioperative

6
 Perioperative myocardial infarction

measurements in high-risk patients enable prompt initiation demonstrates the effectiveness of preoperative coronary
of appropriate diagnostic and therapeutic measures, which revascularization in improving short- and long-term cardiac
may affect long-term cardiac outcome. outcome and mortality in high-risk patients undergoing
high-risk surgery.27 58 67 104 109 146
Summary
Thus, the aetiology of PMI remains poorly Preoperative coronary angiography
understood.80 98 132 Existing data are inconclusive and do Coronary angiography in approximately 25% of patients
not allow a definitive decision on whether long-duration with positive dipyridamole-thallium scintigraphy (DTS)
subendocardial myocardial ischaemia or acute coronary followed by coronary revascularization in roughly a third
occlusion as a result of plaque disruption or thrombosis is of those undergoing coronary angiography was associated
the primary mechanism of perioperative MI in the individual with improved perioperative cardiac outcome following
patient. This uncertainty is to be expected considering the vascular surgery compared with patients not subjected to
enormous structural and functional diversity of coronary such invasive preoperative cardiac evaluation and treat-
atherosclerosis, the unpredictability of plaque progression ment.105 However, this benefit was entirely offset by two
and vulnerability, and the outstanding methodological pro- MIs and three deaths occurring during invasive cardiac
blems of reliably detecting and diagnosing perioperative evaluation. As a result, there was no statistically significant
myocardial ischaemia and infarction. Plaque transformation difference between patients who did and those who did not
from the stable to the vulnerable state can be acute. Wide- undergo extensive preoperative cardiac evaluation and
spread waxing and waning of coronary inflammation and/or coronary intervention in overall early and late non-fatal
of systemic blood thrombogenicity may contribute to the and fatal MI and cardiac death. These and similar findings
development of plaque vulnerability, in the absence or pres- suggest that in many patients with positive DTS undergoing
ence of underlying structurally vulnerable plaques. Some high-risk surgery, coronary angiography does not necessar-
patients may remain vulnerable for a period of weeks to ily confer additional useful information or benefit.
months. In such (chronically) inflamed patients it is possible
that plaques will suddenly flare up and become unstable, Preoperative percutaneous intervention
even in the absence of inflammatory cell infiltration and a Data from randomized controlled trials on the effectiveness
central lipid core. Plaque rupture may occur without clinical of PCI on postoperative cardiac outcome are lacking. All we
manifestations (silent plaque rupture). have are data from retrospective analyses.
Percutaneous transluminal coronary angioplasty. Follow-
Prevention of perioperative MI ing preoperative percutaneous transluminal coronary angio-
Two principal strategies have been used in an attempt to plasty (PTCA) in patients undergoing non-cardiac surgery,
reduce the incidence of PMIs and other cardiac events and the reported incidences of non-fatal PMI and perioperative
complications: preoperative coronary revascularization, and cardiac death were low.2 40 58 67 However, neither of these
pharmacological treatment. studies is of adequate size or design to allow any conclusion
regarding the effectiveness of preoperative PTCA. None
Preoperative coronary revascularization of these reports contained control groups of patients with
Controversy remains as to the appropriate management of CAD that were not subjected to preoperative PTCA. In a
patients identified preoperatively as having relevant but cor- retrospective, case-matched study, preoperative PTCA was
rectable CAD. The effectiveness of preoperative coronary associated with reduced overall perioperative cardiac events
revascularization in this population continues to be debated. when compared with patients with CAD that had not under-
Proponents of ‘prophylactic’ coronary revascularization in gone PTCA.131 However, the overall reduction in cardiac
selected patients argue that it improves both perioperative as events was because of a reduction in the incidence of angina
well as long-term outcome.25 37 48 64 82 118 134 Opponents of pectoris and congestive heart failure but not in non-fatal PMI
this approach point out that morbidity and mortality of per- and mortality. Furthermore, if PTCA had been performed
cutaneous coronary intervention (PCI) and coronary artery less than 90 days before surgery, any potential benefit
bypass surgery (CABG) in high-risk elderly vascular was lost.
patients are substantial and outweigh any benefit; that Coronary stenting. Patients who have recently been sub-
recovery from such major morbidity substantially delays jected to coronary stenting run a high risk of suffering a
and even prevents the surgery for which the intervention PMI and serious bleeding.72 156 162 Of 40 patients who un-
was undertaken; that it does not differentiate between derwent coronary stenting within 6 weeks of major non-
young and old age and between patients with symptomatic cardiac surgery, seven suffered a PMI, 11 experienced
CAD and those with CAD discovered by cardiac stress test- major bleeding, and eight died.72 All PMIs, deaths, and
ing only; that only survivors of coronary revascularization haemorrhages occurred in patients who had undergone stent-
are included in the various reports; and, most importantly, ing within 14 days of surgery. Fatal cardiac events
that no prospective randomized trial exists to date that were mostly caused by stent thrombosis (likely a result of

7
 Priebe

interruption of antiplatelet medication 1–2 days before sur- patients who survived the preoperative coronary evaluation
gery). In contrast, severe bleeding occurred in patients and revascularization.
whose antiplatelet medication was continued. In patients with peripheral vascular disease, in-hospital
Eight (4%) of 207 patients undergoing non-cardiac sur- and long-term outcomes (MI, transient ischaemic
gery in the 2 months following successful coronary stenting attack, stroke, bleeding complications) following PCI
suffered major cardiac events (non-fatal PMI one, fatal PMI were substantially worse and procedural success lower
two, deaths six).162 All of these eight patients were than in patients without peripheral vascular disease.142 In
among those 168 patients who underwent surgery within such a patient population, coronary angiography was asso-
6 weeks of stent placement. No major complication was ciated with an incidence of MI and mortality of 0.07–0.25%
observed in the 39 patients who had surgery 7–9 weeks and 1.0–2.5%, respectively.104 Percutaneous intervention
after stent placement. and CABG carry a 3–10% risk of MI, and a mortality of
1–2.5% and 2–8.5% (depending on urgency of CABG),
Preoperative surgical coronary revascularization respectively.104 121 Depending on the type and extent of
Numerous reports include a substantial number of patients CAD, vascular surgery alone is associated with an incidence
who underwent high-risk vascular surgery following of PMI of 0.5–15% and a mortality of 0.8–20%.104
CABG.37 48 64 84 134 Unfortunately, no prospective, rando- It is obvious from these numbers that in individual
mized trial exists on the effect of preoperative CABG on patients overall perioperative cardiac morbidity and mortal-
cardiac outcome. Several retrospective studies,25 37 45 48 118 ity may well be lower when undergoing vascular surgery
and one prospective non-randomized study82 have suggested without prior coronary revascularization. Decision analysis
that in survivors of preoperative CABG surgery, periopera- suggested that on average, vascular surgery without
tive morbidity and mortality of subsequent major non- preoperative coronary intervention results in better perio-
cardiac surgery is comparable with patients without clinical perative outcome.104 However, the overall long-term
evidence of CAD. In a retrospective analysis of 3368 opera- outcome might well be comparable because those patients
tions in patients enrolled in the Coronary Artery Surgery who did not undergo preoperative coronary angiography and
Study registry, prior CABG improved outcome in major coronary revascularization will be faced at some time post-
non-cardiac surgery (abdominal, thoracic, vascular, head, operatively with just those coronary interventions and their
and neck).37 Compared with medically treated patients, associated risks. In addition, there is evidence that the results
the perioperative event rate was lower in patients who of PCI in high-risk patients has improved in recent years.142
had previously undergone CABG (PMI 0.8 vs 2.7%, mor- This is likely a result of more routine coronary stenting,
tality 1.7 vs 3.3%). The event-lowering effect of CABG was increased use of drug-eluting stents and stents that are easier
most pronounced in patients with advanced angina and/or to deploy, and advances in pharmacotherapy.12 It is conceiv-
multi-vessel CAD. No difference in cardiac outcome was able that such advances might shift the risk/benefit balance
observed during minor surgery. Although these findings towards preoperative PCI.
seem to suggest a cardioprotective effect of preoperative
CABG in patients with CAD undergoing major non-cardiac Recommendations for preoperative coronary angiography
surgery, the analysis did not take into account the added risks In general, indications for preoperative coronary angiogra-
of coronary angiography and myocardial revascularization. phy are similar to those in the non-operative setting. The
Timing of non-cardiac surgery following CABG may be Class I recommendations for preoperative coronary angio-
crucial. In a retrospective, case-control study, patients who graphy are accordingly restrictive, apply only to patients
underwent high-risk vascular surgery within 1 month of with suspected or known CAD and include: (i) evidence
CABG had a higher mortality and a trend towards a higher for high risk of adverse outcome based on non-invasive
incidence of MI than those undergoing surgery at a later test results; (ii) angina pectoris unresponsive to adequate
date.15 This confirms previous findings of increased mortal- medical therapy; (iii) unstable angina, particularly when
ity associated with simultaneous CABG and vascular sur- facing intermediate or high risk non-cardiac surgery; and
gery,134 or with non-cardiac surgery within 1–6 months (iv) equivocal non-invasive test results in patients at high
of CABG compared with surgery performed later than clinical risk undergoing high risk surgery.34 35
6 months following CABG.26
Recommendations for preoperative coronary intervention
In view of the considerable risk of percutaneous intervention
Risk calculation in preoperative coronary
in high-risk patients, it is highly unlikely that prophylactic
revascularization
PCI to merely ‘get the patient through surgery’ will reduce
In the individual patient, the combined risk of preoperative
the incidence of PMI. PCI with or without stenting should
coronary interventions (coronary angiography, PCI, or
thus be reserved for patients who have a medical indication
CABG) and scheduled non-cardiac surgery may well exceed
for such intervention unrelated to surgery.
the perioperative risk of non-cardiac surgery alone in patients
without prior CABG. Any potential benefit of preoperative PTCA. The indications for preoperative PTCA are identical
coronary revascularization will be restricted to those to those in the non-operative setting.144 Following balloon

8
 Perioperative myocardial infarction

angioplasty without coronary stenting, surgery should be clopidogrel and aspirin) to be continued perioperatively, a
delayed for at least a week.34 35 This is beyond the time drug-eluting stent can be placed. If surgery (or the surgeon)
period of within hours to days of the intervention during does not allow perioperative continuation of dual antiplate-
which arterial recoil and acute thrombosis at the site of let therapy, a drug-eluting stent should probably not be used.
angioplasty are most likely to occur, and it allows healing In such a case, heparin- or phosphorylcholine-coated stents
of the vessel injury at the site of balloon treatment. may possibly reduce the risk of stent thrombosis in the
absence of clopidogrel. This possibility is, however, not
Coronary stenting. Stent thrombosis is a serious complica-
supported by any data. Under certain circumstances,
tion and mostly results in Q-wave infarction or death.28
PTCA without stent placement may be the most appropriate
In patients not undergoing surgery, stent thrombosis most
option. Needless to say, the choice for the type of stent (bare
commonly occurs within hours to days of stent placement.28
metal vs drug eluting vs coated) and the type of PCI (angio-
Dual antiplatelet medication with a thienopyridine (mostly
plasty with or without stenting) rests entirely with the inter-
clopidogrel) and aspirin reduces the incidence of early stent
ventionalist. However, as the anaesthetist will be involved in
thrombosis to less than 1%.163 The risk of stent thrombosis
the perioperative management of these high-risk patients,
diminishes with re-endothelialization of the initially com-
and as the perioperative period differs considerably from the
pletely denuded endothelial surface. Re-endothelialization
non-operative setting, the interventionalist may benefit from
occurs within approximately 8 weeks of placement of
the anaesthetist’s knowledge and feedback. Close preopera-
the stent.153 On the other hand, re-stenosis may develop
tive consultation between interventionalist and anaesthetist
by 6–8 weeks after stent placement. If, however, re-stenosis
is required to minimize the risk of perioperative cardiac and
has not occurred by 8–12 months after PCI (with or without
overall morbidity and mortality.
stent), it is unlikely to develop thereafter.
If a coronary stent is placed, elective non-cardiac surgery Recommendation for preoperative surgical
should be delayed for an absolute minimum of 2 weeks, but coronary revascularization
ideally for 4–6 weeks.34 35 162 This delay will allow comple- The indications for preoperative surgical coronary
tion of a course of dual antiplatelet medication (which revascularization are essentially identical to those in the
reduces the incidence of early stent thrombosis), and non-operative setting.36 62 They include patients with:
allow bare metal stents to be endothelialized. (i) acceptable coronary revascularization risk and suitable
Today, however, stents eluting antiproliferative drugs, viable myocardium with left main stenosis; (ii) three-vessel
which delay endothelialization are increasingly being CAD in conjunction with left ventricular dysfunction;
placed.44 As this may well increase the risk of early and (iii) two-vessel disease involving severe proximal left
late stent thrombosis, a 6–12-month period of antiplatelet anterior descending artery obstruction; and (iv) intractable
treatment has been recommended.61 Thus, conclusions coronary ischaemia despite maximal medical therapy.
drawn from, and recommendations based on, previous If major non-cardiac surgery is indicated following recent
reports72 156 162 apply to bare metal stents only. This caveat CABG, timing appears crucial. Limited data would
is emphasized by a recent Letter to the Editor.8 Twelve suggest postponing elective major surgery for at least
weeks after having simultaneously received a bare metal 4–6 weeks,15 134 151 possibly for even up to 6 months after
stent and two paclitaxel-eluting stents accompanied by anti- CABG.26
platelet therapy with clopidogrel and aspirin, antiplatelet
drugs were discontinued before knee surgery. Immediately Conclusions
postoperatively, the patient suffered a PMI. Coronary angio- Thus, prospective, randomized investigations on the effect
graphy revealed total occlusion of both paclitaxel-eluting of preoperative coronary revascularization on short- and
stents but an open bare metal stent. This single report long-term cardiac and overall outcome do not exist. Survi-
would suggest that following placement of a drug-eluting vors of coronary revascularization tend to have a better
stent, it might be safer to postpone elective surgery for perioperative and long-term cardiac outcome than patients
several months rather than weeks. with comparable CAD without preoperative coronary
The matter becomes even more complicated because revascularization.
there is preliminary evidence to suggest that the rate of However, in this analysis the high cardiac morbidity and
re-endothelialization differs between drugs being eluted mortality associated with preoperative coronary angiogra-
from the stent. Antiplatelet therapy may have to be contin- phy and coronary revascularization (PCI or CABG) in
ued for a longer period of time following placement of high-risk patients are not taken into account. In addition,
a paclitaxel-eluting stent (possibly for 6 months) than survivors of PCI face the perioperative risk of coronary (stent)
following placement of a sirolismus-eluting stent (possibly thrombosis or haemorrhage associated with discontinuation
2–3 months). or continuation of dual antiplatelet therapy, respectively.
If patients require elective non-cardiac surgery within Overall outcome may thus be comparable between
2 months of PCI, there are several options. If surgery preoperatively revascularized and non-revascularized
(and the surgeon) allows dual antiplatelet therapy (mostly patients—it may be even worse in individual revascularized

9
 Priebe

patients. The decision for or against preoperative coronary afterload, and contractility). b-Blockers have the potential
revascularization, and for or against PCI or CABG, should of reducing myocardial O2 consumption (thus improving
therefore be based entirely on universally accepted the myocardial O2 supply/demand balance) by decreasing
medical indications for coronary revascularization and the sympathetic tone and myocardial contractility, in turn
appropriate technique. The philosophy of performing pre- resulting in decreases in heart rate and arterial pressure.
operative coronary revascularization merely ‘to get the pa- Furthermore, they decrease b2-adrenoceptor-mediated
tient through surgery’ is contrary to all available evidence. If release of intracardiac norepinephrine during ischaemia
the decision for preoperative coronary revascularization is (reducing cardiac toxicity); they attenuate exercise-induced
made, timing with respect to the subsequent surgery appears coronary vasoconstriction (improving exercise capacity);
crucial. If these caveats are being observed, it is conceivable and they have antiarrhythmic properties (increasing the
that carefully selected patients might benefit from preopera- threshold for ventricular fibrillation during myocardial
tive coronary revascularization. However, only prospective ischaemia).
randomized trials can tell.
Effect on perioperative cardiac mortality. The effect of
perioperative b-blocker therapy on cardiac outcome has
Pharmacological treatment been assessed in two, much discussed studies.100 130 In a
Beta-blockers randomized, double-blind, placebo-controlled study, the
Perioperative b-blocker therapy has been listed as a ‘top- benefit of perioperative atenolol in patients with or at risk
tier’ patient safety practice by the Institute of Medicine.140 for CAD undergoing major non-cardiac surgery under
Several prospective and retrospective studies suggest that general anaesthesia was examined.100 Atenolol (n=99) or
perioperative b-blockade improves cardiac outcome in placebo (n=101) were started intravenously approximately
patients with or at risk CAD,100 157 and in patients with 30 min before induction of anaesthesia and continued until
documented inducible myocardial ischaemia undergoing hospital discharge or for up to 7 days postoperatively. Out-
non-cardiac surgery.14 129 130 It has been suggested that come variables included cardiac death (death because of MI,
b-adrenoceptor-antagonists (‘b-blockers’) should be admi- dysrhythmia, or congestive heart failure), and cardiac events
nistered to almost all patients with one or more factors that (non-fatal MI, unstable angina and/or congestive heart
are known to be associated with a higher perioperative failure requiring admission and treatment, myocardial revas-
cardiac risk.76 cularization) during the 2 yr following hospital discharge
(i.e. in-hospital cardiac morbidity and mortality were not
Rationale for the use of perioperative b-blocker therapy. included in the analysis). Over the 2-yr follow-up period,
Numerous cardiovascular and other effects (anti-arrhythmic, overall mortality after hospital discharge was significantly
anti-inflammatory, altered gene expression and receptor lower in the atenolol (10%) than in the placebo group (21%,
activity, protection against apoptosis) of b-blockers may P=0.019). This amounts to a relative risk reduction of 55%.
account for their cardioprotective effect in the operative The main reason for this difference was a reduction in
and non-operative setting.6 70 94 All b-blockers, except cardiac deaths during the first 6 months in the atenolol-
those with intrinsic sympathetic activity, reduce mortality treated patients. The combined cardiovascular outcomes
in both MI,51 57 143 and heart failure patients.68 149 Rando- were similarly reduced in the atenolol group.
mized clinical trials involving more than 24 000 patients The study has been criticized on numerous grounds.
have shown that b-adrenoceptor-antagonism (‘b-blockade’) (i) Only those adverse events were included in the analysis
reduces post-myocardial infarction mortality, probably by a that occurred after hospital discharge when patients had
reduction in infarct size and ventricular arrhythmias.147 stopped taking b-blockers. However, four patients in the
On the basis of such data it seems logical that perioperative atenolol, and two patients in the control group died during
b-blocker therapy should be beneficial during the period of hospitalization. It is inappropriate to exclude these
perioperative stress. in-hospital events from the overall analysis. If they are
Activation of the hypothalamus–pituitary–adrenal axis included, the difference in deaths between the atenolol
persists for at least 1 week following surgery. Adrenal (n=13) and the placebo group (n=23) loses statistical signi-
cortical stimulation is accompanied by sympathetic ficance (P=0.1). (ii) The potential for acute b-withdrawal
nervous system-induced adrenal medullary activation, symptoms in the control group cannot be excluded. Eight
resulting in the release of catecholamines with subsequent patients on chronic b-blocker medication were acutely taken
stimulation of adrenergic receptors. Adrenergic receptors off their b-blockers when they were randomized to the con-
are located in virtually every organ. In the human heart, trol group. Thus, acute b-withdrawal symptoms could pos-
they mediate numerous biological responses, including sibly have contributed to the less favourable outcome in the
inotropy, chronotropy, myocyte apoptosis, and direct placebo group. (iii) Approximately 40% of patients did not
myocyte toxicity. tolerate the full dose, and roughly 15% did not tolerate
Catecholamines increase each of the four determinants of atenolol at all. (iv) Female gender was under-represented.
myocardial oxygen consumption (i.e. heart rate, preload, (v) The exact number of patients with intermediate rather

10
 Perioperative myocardial infarction

than high risk for adverse perioperative cardiac outcome was to the control group) was not defined. (v) The study
not specified. (vi) There was a trend towards a more severe population was highly selective: of the 1351 initially
cardiac history (previous MI, angina, diabetes, coronary screened patients, only 112 (8%) were eventually included
revascularizations, advanced age) in the placebo group, in the actual study. Thus, the results are not necessarily
and a trend towards more effective cardiac therapy (i.e. representative of a broader patient population. (vi) Patients
b-blockers, angiotensin-converting enzyme inhibitors) at with severe CAD were excluded. (vii) Finally, the 34%
hospital discharge in the atenolol group. Given the many complication rate in the standard care group (nine cardiac
study limitations (mainly the overall small number of events, deaths, nine MIs) is rather high. A high complication rate in
the unrealistically high treatment effect of 55% and no the control group generally tends to favour the treatment
statistically significant difference between groups when group. Despite the various limitations, the accompanying
in-hospital deaths are included), one has to question the editorial87 stated that ‘. . . In the absence of major contra-
appropriateness of the recommendation for perioperative indications therapeutic doses of beta-adrenergic antagonists
b-blocker therapy by the American College of Physicians should be given to patients with an intermediate or high risk
that was based on this study.125 of cardiac complications’.
A subsequent study looked at the benefit of perioperative
bisoprolol in patients with documented CAD (diagnosed Unanswered questions
by new wall motion abnormalities on dobutamine stress The repeated recommendations for perioperative b-block-
echocardiography) undergoing major vascular surgery.130 ade in patients with suspected or documented CAD71 87 125 is
1351 patients undergoing elective major vascular surgery mainly based on the findings of those two prospective, ran-
were screened for cardiac risk factors (age over 70 yr, domized controlled trials in a little over 300 patients. Several
angina, prior MI, compensated or a history of congestive unanswered questions remain.
heart failure, current treatment for ventricular arrhythmias,
current treatment for diabetes mellitus, limited exercise Should b-blockers be administered together with other
capacity). 846 of the 1352 patients had at least one of sympatholytic therapies? The safety of simultaneously
these cardiac risk factors and were, in turn, screened for a administering b-blockers in patients receiving thoracic
positive dobutamine stress echocardiogram (DSE). Of the epidural anaesthesia or a2-adrenergic agonists has not
846 patients, 173 had a positive DSE. Of these, 61 were been established. It is conceivable that the interaction
excluded from further study because of either extensive between treatments causes an unacceptably high incidence
wall motion abnormalities on DSE, strong evidence on of bradycardia and hypotension, counteracting any potential
DSE for left main or severe three-vessel CAD, or because cardioprotective effect of b-blocker therapy. At present, it
they were already taking b-blockers. The remaining remains unknown whether it is necessary to add b-blockers
112 patients were randomized to receive either bisoprolol to treatments like a2-adrenergic agonists that have them-
(n=59) or ‘standard care’ (n=53). Bisoprolol was started on selves demonstrated cardioprotection in the perioperative
average 37 (range 7–89) days before surgery and was period.158 161
continued for 30 days postoperatively. Outcome variables Is there a b-blocker of choice for perioperative b-blocker
included cardiac death and non-fatal MI during the first therapy? Blocking or blunting the perioperative adrenergic
30 days following surgery. The authors reported a 10-fold stress response is most likely the key pathophysiological
lower rate of perioperative cardiac events in the bisoprolol intervention that associates perioperative b-blocker therapy
group compared with the ‘standard care’ group (3.4 vs with improved cardiac outcome. Therefore, although not
34%; P=0.001). proven yet, it is rather unlikely that pharmacological differ-
Although the results strongly suggest that patients with ences between b-blockers (e.g. in receptor selectivity and
documented CAD disease undergoing high-risk surgery affinity, lipophilicity, intrinsic sympathomimetic activity)
benefit from perioperative b-blockade, this investigation have any impact on efficacy and safety of treatment. Choice
also has several limitations: (i) the study included only of the b-blocker should be based on those, admittedly very
112 patients and a total of merely 20 events. In view of few, controlled randomized trials that have demonstrated
the small sample size, the possibility cannot be ruled out effectiveness of perioperative b-blocker therapy.100 130
that the observed 90% relative reduction in 30-day adverse Any cardioselective b-blocker (such as atenolol, bisoprolol,
outcome occurred by chance alone. The number of 3.2 or metoprolol) is probably an acceptable choice.
needed to treat for prevention of PMI and mortality is
10-fold lower than a meta-analysis-derived number of 42 When should perioperative b-blockade be started?
for secondary prevention after MI in the medical setting.9 51 Perioperative cardioprotection was demonstrated when the
(ii) The trial was terminated early because the interim medication had been initiated either weeks before the sched-
analysis had suggested a large treatment effect. However, uled surgery,130 or as late as during premedication148 and
unexpectedly large beneficial effects suggested by studies induction of anaesthesia.100 The recently revised ACC/AHA
that are terminated early are cause for scepticism.159 guidelines on perioperative cardiovascular evaluation for
(iii) Treatment was not blinded. (iv) Standard care (given non-cardiac surgery34 recommend that in patients with

11
 Priebe

Class I indications for perioperative b-blocker therapy (see small number of patients would suggest that discontinuation
below), b-blockers be started days or weeks before elective of b-blockers in vascular surgery patients may be associated
surgery. This makes sense as it will allow titration of the with an increased risk of postoperative morbidity and mor-
b-blocker to the targeted heart rate. tality.139 It thus seems advisable to discontinue b-blocker
therapy gradually (and only after a period of preferably
What should be the therapeutic goal? It is assumed that 30 days postoperatively) in those patients considered not
cardiovascular and sympathetic suppression is required to to have a clear indication for long-term therapy.
produce cardiac protection. The extent of such suppression
Is ‘routine’ chronic b-blocker therapy continued peri-
is difficult to assess clinically. Basically all studies on the
operatively as effective as acutely initiated, closely moni-
perioperative use of b-blockers have, therefore, taken heart
tored, heart rate-targeted perioperative b-blocker therapy?
rate as a physiological surrogate of sympathetic tone.100 130
Those 53 patients who were excluded from the bisoprolol
Preoperatively, b-blockers were titrated to achieve heart
study130 because they were already taking b-blockers,
rates between 50 and 60 beats min 1.130 Postoperatively,
subsequently underwent planned vascular surgery under
heart rates of less than 80 beats min 1 100 130 154 or 20%
continued but not specified b-blocker therapy. In this sub-
below the preoperative ischaemic threshold133 were
population, the 30-day perioperative cardiac mortality was
aimed at. The revised ACC/AHA Guidelines recommend
7.5%, which is twice as high as that reported in
that the preoperative dose is titrated to achieve a resting
the randomized part of the trial. These findings would
heart rate between 50 and 60 beats min 1.34
suggest that perioperative b-blocker therapy might be less
For how long should b-blocker therapy be continued post- effective when not closely monitored and strictly heart
operatively? In the two main controlled, randomized trials rate-targeted.
on the effectiveness of perioperative b-blocker therapy, The total cohort of 1351 consecutive patients initially
b-blockers were continued for up to a week100 and up to screened in the randomized trial on bisoprolol130 was
a month130 following surgery. In the latter study, following retrospectively re-analyzed.14 360 (27%) patients received
the initial study period of 30 postoperative days, the b-blockers perioperatively, whereas 991 (73%) did not.
101 survivors continued to receive either bisoprolol therapy Except for those 59 patients who were part of the rando-
(n=57) or standard care (n=44) according to their initial mized trial, b-blocker management was not specified in
randomization.129 In the bisoprolol group, the dose was those 360 patients. The perioperative cardiac event rate
adjusted to achieve a heart rate between 50 and 60 beats (non-fatal MI, cardiac death) was 2.2% (n=8) in the
min 1. Patients were followed for 11–30 months after sur- b-blocker-treated patients (which is comparable with
gery (median 22 months). Cardiac events (cardiac death the 3.4% event rate reported in the randomized part of
and non-fatal MI) occurred in seven (12%) patients in the the trial), and 3.7% (n=37) in the non-b-blocked patients
bisoprolol group and in 14 (32%) patients in the standard (which is almost 10-fold lower than the 34% event rate
care group (P=0.025). These results suggest that long-term reported in the randomized part of the trial). The finding
postoperative b-blockade reduces the incidence of late car- of a comparably low perioperative cardiac event rate in this
diac events, certainly among survivors of major vascular larger population of b-blocked patients (who, presumably,
surgery who had received perioperative b-blockade. were less intensively monitored than those patients who
It appears obvious that those patients with objective indi- participated in the prospective, controlled trial) could be
cations for the use of b-blockers should continue b-blocker interpreted as suggestive evidence that ‘routine’ chronic
therapy after hospital discharge. In patients without clear b-blocker therapy continued perioperatively is, in fact, as
indications for long-term b-blocker therapy, b-blockers effective as acute, closely monitored, heart rate-targeted
should probably be continued for at least the time of perioperative b-blocker therapy.
hospitalization, and preferably for up to 1 month post-
Who should receive perioperative b-blocker therapy? Of the
operatively. In those patients in whom b-blocker therapy
1351 patients initially screened in the randomized trial on
is going to be discontinued after discharge, the dose should
bisoprolol, 1118 (83%) had at most one or two clinical risk
be tapered slowly to avoid acute withdrawal symptoms.
factors (defined as age >70 yr, current angina, prior MI,
In those patients in whom b-blocker therapy is going to be
congestive heart failure, prior cerebrovascular event, dia-
continued after discharge, the dose should be adjusted as
betes mellitus, renal failure).14 Among this subgroup of
indicated.
patients with relatively low cardiac risk, those receiving
Is there a risk of discontinuing perioperative b-blockade? It b-blockers perioperatively had a significantly lower cardiac
is conceivable that acute withdrawal symptoms could de- event rate (2/263 patients, 0.8%) than those not receiving
velop when b-blockade is abruptly discontinued in patients b-blockers (20/855 patients, 2.3%). Amongst a further
at increased cardiac risk who were started preoperatively on subset of 375 patients with no clinical risk factor at all,
b-blockers. Although neither of the controlled randomized cardiac event rates were comparably low between those
trials reported such adverse effects of discontinuation of receiving (0/48 patients, 0%) and those not receiving
b-blockers,100 130 results of a retrospective analysis in a b-blockers (4/327 patients, 1.2%).

12
 Perioperative myocardial infarction

In contrast, in the subgroup of 233 (17%) patients with AHA/ACC Guidelines for secondary prevention of MI and
more than/equal to three clinical risk factors, those receiving death recommend to initiate b-blockade in all post-MI
b-blockers had a cardiac event rate of 6.2% (6/97 patients) patients and to continue such therapy indefinitely.143
compared with a cardiac event rate of 12.5% (17/136 They list as absolute contraindications for the use of
patients) in those not receiving b-blockers. Within this b-blockers symptomatic bradycardia (usually a heart rate
subgroup of patients with more than/equal to three clinical <50–60 beats min 1), symptomatic hypotension (usually a
risk factors, 207 patients had four or fewer new wall motion systolic arterial pressure <90–100 mm Hg), severe heart
abnormalities on dobutamine stress echocardiography. failure requiring i.v. diuretics or inotropes, cardiogenic
Those receiving b-blockers perioperatively had a lower car- shock, asthma or reactive airway disease requiring
diac event rate (2/86 patients, 2.3%) than those not receiving bronchodilator and/or steroids, and 2 or 3 AV block.
b-blockers (12/121 patients, 10.6%). However, in a further
subgroup of 26 patients with more than/equal to three clin- Proposed algorithm for the use of perioperative
ical risk factors and five or more new wall motion abnor- b-blocker therapy
malities on dobutamine stress echocardiography, there was An algorithm for the use of perioperative b-blocker therapy
no difference in the cardiac event rates between those based on various studies on perioperative b-blocker therapy
receiving b-blockers perioperatively (4/11 patients, 36%) and preoperative risk stratification has been suggested.9
and those not receiving b-blockers (5/15 patients, 33%). In high-risk patients with more than/equal to three major
These data are based on retrospective analysis, treatments clinical risk factors (high-risk surgical procedure, CAD,
were neither controlled nor randomized or blinded, and the cerebrovascular disease, insulin-dependent diabetes melli-
number of patients in some of the subgroups was too small to tus, chronic renal insufficiency) and a positive non-invasive
allow valid statistical analysis. Taking these limitations into cardiac stress test (DTS, DSE), additional coronary angio-
due consideration, the results would suggest that in patients graphy and coronary revascularization should be considered
undergoing vascular surgery, perioperative b-blocker ther- because the perioperative cardiac event rate will remain in
apy may possibly be beneficial in all but subsets of very low the 6.5–16% range even with perioperative b-blocker ther-
or very high risk patients. The findings would further suggest apy. In high-risk patients with negative non-invasive test
indirectly that aggressive b-blockade in high-risk patients results, and in clinically intermediate risk patients
undergoing high-risk surgery may reduce the need for [1–2 major clinical risk factors or any of two minor risk
additional preoperative non-invasive cardiac testing, and factors100 such as age >65 yr, hypertension, current smoker,
coronary angiography and revascularization. It is likely that serum cholesterol >6.18 mmol litre 1 (>240 mg dl 1), non-
the combined morbidity and mortality from the three insulin-requiring diabetes mellitus] with good functional
sequential procedures, coronary angiography, coronary capacity and without evidence of angina or peripheral vas-
revascularization and subsequent major vascular surgery, cular disease, b-blocker therapy is started preoperatively
is higher than the 3.4% incidence of major cardiac complica- and surgery is performed as planned.
tions in patients receiving perioperative bisoprolol.130 Only In clinically intermediate-risk patients with poor func-
in a subset of patients with extensive myocardial ischaemia, tional capacity and with evidence of angina or peripheral
may perioperative b-blocker therapy not be sufficiently vascular disease, additional therapies and/or interventions
protective.14 (coronary angiography and revascularization) should be
Based on the results of various studies, the revised ACC/ considered. Finally, in low-risk patients without clinical
AHA Guidelines34 list several conditions as Class I indica- risk factors, the perioperative cardiac event rate is low with
tions for perioperative b-blocker therapy (i.e. conditions for (0.4%) or without (0.4–1.0%) perioperative b-blockade, so
which there is evidence for and/or general agreement that the that perioperative b-blockade is deemed unnecessary.
therapy is useful and effective): (i) the need for b-blockers in In conclusion, although perioperative b-blocker therapy
the recent past to control symptoms of angina; (ii) patients has been designated as one of 11 specific practices with
with symptomatic arrhythmias or hypertension; and (iii) sufficient clinical-based evidence for patient safety to justify
patients at high risk for a perioperative cardiac event immediate and widespread implementation,140 before a
based on the finding of myocardial ischaemia on perioperat- final recommendation for a liberal use of perioperative
ive testing who are undergoing vascular surgery. Class IIa b-blockade can be made safely, several caveats have to
indications for perioperative b-blocker therapy (i.e. condi- be kept in mind. All studies that support use of perioperative
tions for which there is conflicting evidence and/or a diver- b-blocker therapy have included rather small numbers of
gence of opinion about the usefulness/efficacy of the patients (as few as 26154). Often, recruitment of patients was
performed therapy, with the weight of evidence/opinion highly selective and consecutive (recruitment rate as low as
in favour of usefulness/efficacy of the performed therapy) 8%130), excluding application of the results to an unselected
include preoperative identification of untreated hyperten- surgical population. Furthermore, the beneficial effects were
sion, known CAD, or major risk factors for CAD. probably not only because of a rather aggressive therapy
When it comes to defining the contraindications for the (targeted heart rates maximally 80 beats min 1), but also
use of b-blockers, it is helpful to remember that the 2001 (and perhaps even more importantly) because of continuous

13
 Priebe

close monitoring of the patient. This will ensure both undergoing non-cardiac surgery (2.13 vs 3.05% in non-
optimal cardioprotection and patient safety. More uncon- treated patients).93 These findings are consistent with results
trolled but equally aggressive postoperative administration of a cohort study in almost 20 000 patients with ACS.145
of b-blockers on ordinary surgical wards might well result in Patients who were taking statins when experiencing ACS
more adverse side-effects, possibly negating any beneficial had fewer MIs than patients not taking statins. Institution of
effects. Although current evidence suggests that selected aggressive statin therapy in patients with acute coronary
patients are likely to benefit from perioperative b-blocker syndrome resulted in reduced plaque volume at 6-month
therapy, we have to acknowledge that data on risks and follow-up.120 It thus appears that statin therapy may modu-
benefits of such therapy are still few and inconclusive.32 late early pathophysiological processes during ischaemic
A large definitive trial on perioperative b-blocker therapy cardiac events.
is needed.31 Until the results of such a trial are available, ‘Pleiotropic’ effects of statins independent of their lipid-
it seems fair to conclude: ‘Peri-operative b-blockade: a lowering action have been proposed as the mechanisms of
useful treatment that should be greeted with cautious their beneficial effects. These pleiotropic effects include
enthusiasm’.66 reversal of endothelial dysfunction,91 155 modulation of
macrophage activation,3 immunological effects,3 and
Alpha-2 adrenoceptor agonists anti-inflammatory,3 antithrombotic,155 and antiproliferative
Alpha-2 adrenoceptor agonists improve cardiovascular actions (possibly mediated by the induction of heme
morbidity and mortality following non-cardiac and oxygenase-1).89 The direct effect of statins on vascular func-
cardiac surgery.33 108 119 122 158 161 In a prospective, rando- tion may result in coronary plaque stabilization.
mized, double-blinded study in 190 patients undergoing
Miscellaneous preventive measures
non-cardiac surgery, prophylactic clonidine (0.2 mg orally
Postoperative myocardial ischaemia has been shown to be
and as dermal patch for 4 days) reduced perioperative
associated with postoperative anaemia,116 hypothermia,49 50
myocardial ischaemia and improved 30-day and 2-yr
and pain.11 101 All of them activate sympathetic tone with
mortality but had no effect on PMI.158 The long-term
adverse effects on cardiovascular function and coagulation.
beneficial effect could have been a result of the reduction
The result will be an increase in myocardial oxygen con-
in perioperative myocardial ischaemia.
sumption in the presence of a decrease in delivery. As peri-
The mechanism of the protective effect is likely to be
operative myocardial ischaemia is a predictor of adverse
manifold. Alpha-2 adrenoceptor agonists attenuate peri-
short- and long-term cardiac outcome, maintenance of an
operative haemodynamic instability,108 inhibit central sym-
appropriate haemoglobin concentration, normothermia,
pathetic discharge,114 reduce peripheral norepinephrine
and adequate pain control are essential preventive measures.
release,38 and dilate post-stenotic coronary vessels.65
Aspirin Conclusions
Early postoperative administration of aspirin improved
The aetiology of PMI is multifactorial. The perioperative
outcome following coronary artery bypass surgery.99
period induces large, unpredictable and unphysiological
Aspirin is known to reduce cardiac events in patients
changes in sympathetic tone, cardiovascular performance,
with ACS and in patients not known to have CAD.23 It
coagulation and inflammatory response (to name just a few).
eliminates the diurnal variation in plaque rupture.136 Com-
These changes induce, in turn, unpredictable alterations in
pared with controls, patients with unstable angina had more
plaque morphology, function and progression. Simultaneous
than twice the blood concentrations of interleukin-6, CRP,
perioperative alterations in homeostasis and coronary plaque
and macrophage colony-stimulating factor.69 Those con-
characteristics may trigger a mismatch of myocardial oxy-
centrations decreased after 6 weeks of aspirin treatment.
gen supply and demand by numerous mechanisms. If not
Aspirin will, of course, reduce platelet aggregability, but
alleviated in time, it will ultimately result in MI, irrespective
its ability to reduce future MI appears greatest in individuals
of its aetiology (morphologically, haemodynamically,
with serological evidence of increased inflammation.135
inflammatory, or coagulation induced). With these many
Thus, the anti-inflammatory effect of aspirin may be additive
and diverse factors involved, it is highly unlikely that one
to its antithrombotic effect in patients with plaque instabil-
single intervention will successfully improve cardiac out-
ity. This effect may be of particular relevance in the
come following non-cardiac surgery. A multifactorial, step-
perioperative setting.
wise approach is indicated.20 22 59 79 113
Statins Based on increasing knowledge of the nature of athero-
Perioperative use of statins may be associated with reduced sclerotic CAD, and in view of the poor positive predictive
perioperative mortality in patients undergoing major vascu- value of non-invasive cardiac stress tests and the consider-
lar surgery.77 128 In a multicentre observational study includ- able risk of coronary angiography and coronary revascular-
ing over 780 000 individuals, lipid-lowering therapy ization in high-risk patients, the paradigm is shifting from an
(primarily statins) during the first two days of hospitaliza- emphasis on extensive non-invasive preoperative risk strati-
tion was associated with decreased mortality in patients fication to an emphasis on a combination of selective

14
 Perioperative myocardial infarction

non-invasive testing (to reliably identify those patients who 10-year experience from an academic centre. Anaesthesia 2004;
truly benefit from preoperative intervention, such as cancel- 59: 422–7
lation of surgery, preoperative coronary revascularization, 16 Breslow MJ, Parker SD, Frank SM, et al. Determinants of cate-
cholamine and cortisol responses to lower extremity revascular-
initiation or optimization of cardioprotective medication),
ization: the PIRAT study group. Anesthesiology 1993; 79: 1202–9
and aggressive perioperative pharmacological therapy.59 63 17 Browner WS, Li J, Mangano DT, for the Study of Perioperative
Perioperative plaque stabilization by pharmacological Ischemia Research Group. In-hospital and long-term mortality in
means (statins, aspirin, b-blockers) may be as important male veterans following noncardiac surgery. JAMA 1992; 268:
in the prevention of PMI as an increase in myocardial oxy- 228–32
gen supply (by coronary revascularization), or a reduction in 18 Buja LM, Willerson JT. Role of inflammation in coronary plaque
myocardial oxygen demand (by b-blockers or a2-agonists). disruption. Circulation 1994; 89: 503–5
19 Casscells W, Naghavi M, Willerson JT. Vulnerable atherosclerotic
plaque. A multifocal disease. Circulation 2003; 107: 2072–5
20 Chassot PG, Delabays A, Spahn DR. Preoperative evaluation of
References patients with, or at risk of, coronary artery disease undergoing
1 Adams JE III, Sicard GA, Allen BT, et al. Diagnosis of perioperative non-cardiac surgery. Br J Anaesth 2002; 89: 747–59
myocardial infarction with measurement of cardiac troponin I. 21 Cohen MC, Aretz TH. Histological analysis of coronary artery
N Engl J Med 1994; 330: 670–4 lesions in fatal postoperative myocardial infarction. Cardiovasc
2 Allen JR, Helling TS, Hartzler GO. Operative procedures not Pathol 1999; 8: 133–9
involving the heart after percutaneous transluminal coronary 22 Cohn SL, Goldman L. Preoperative risk evaluation and
angioplasty. Surg Gynecol Obstet 1991; 173: 285–8 perioperative management of patients with coronary artery
3 Almog Y, Shefer A, Novack V, et al. Prior statin therapy is asso- disease. Med Clin N Am 2003; 87: 111–36
ciated with a decreased rate of severe sepsis. Circulation 2004; 23 Collaborative Group of the Primary Prevention Project. Low-
110: 880–5 dose aspirin and vitamin E in people at cardiovascular risk: a
4 Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarc- randomised trial in general practice. Lancet 2001; 357: 89–95
tion redefined—a consensus document of the Joint European 24 Corti R, Fuster V, Badimon JJ. Pathogenic concepts of acute
Society of Cardiology/American College of Cardiology Commit- coronary syndromes. J Am Coll Cardiol 2003; 41: 7S–14S
tee for the redefinition of myocardial infarction. J Am Coll Cardiol 25 Crawford ES, Morris GC jr, Howell JF, Flynn WF, Moorhead DT.
2000; 36: 959–69 Operative risk in patients with previous coronary artery bypass.
5 Alpert JS. Defining myocardial infarction: ‘Will the real myo- Ann Thorac Surg 1978; 26: 215–21
cardial infarction please stand up?’ (Editorial). Am Heart J 2003; 26 Cruchley PM, Kaplan JA, Hug CC jr, Nagle D, Sumpter R,
146: 377–9 Finucane D. Non-cardiac surgery in patients with prior myocar-
6 Anzai T, Yoshikawa T, Takahashi T, et al. Early use of beta- dial revascularization. Can Anaesth Soc J 1983; 30: 629–34
blockers is associated with attenuation of serum C-reactive pro- 27 Cutler BS, Leppo JA. Dipyridamole thallium 201 scintigraphy to
tein elevation and favorable short-term prognosis after acute detect coronary artery disease before abdominal surgery. J Vasc
myocardial infarction. Cardiology 2003; 99: 47–53 Surg 1987; 5: 91–100
7 Arai AE, Hirsch GA. Q-wave and non-Q-wave myocardial infarc- 28 Cutlip DE, Baim DS, Ho KKL, et al. Thrombosis in the modern era:
tions through the eyes of cardiac magnetic resonance imaging a pooled analysis of multicenter coronary stent clinical trials.
(Editorial Comment). J Am Coll Cardiol 2004: 44: 561–3 Circulation 2001; 103: 1967–71
8 Auer J, Berent R, Weber T, Eber B. Risk of noncardiac surgery in 29 Davies MJ. Stability and instability: two faces of coronary athero-
the months following placement of a drug-eluting coronary stent sclerosis. Circulation 1996; 94: 2013–20
(Letter to the Editor). J Am Coll Cardiol 2004; 43: 713 30 Dawood MA, Gutpa DK, Southern J, Walia A, Atkinson JB,
9 Auerbach DA, Goldman L. b-blockers and reduction of cardiac Eagle KA. Pathology of fatal perioperative myocardial infarction:
events in noncardiac surgery. Scientific review. JAMA 2002; 287: implications regarding pathophysiology and prevention. Int J Car-
1435–44 diol 1996; 57: 37–44
10 Badner NH, Knill RL, Brown JE, et al. Myocardial infarction after 31 Devereaux PJ, Leslie K, Yang H. The effect of perioperative beta-
noncardiac surgery. Anesthesiology 1998; 88: 572–8 blockers on patients undergoing noncardiac surgery—is the
11 Beattie WS, Buckley DN, Forrest JB. Epidural morphine reduces answer in? Can J Anesth 2004; 51: 749–55
the risk of postoperative myocardial ischaemia in patients with 32 Devereaux PJ, Yusuf S, Yang H, Choi PTL, Guyatt GH. Are the
cardiac risk factors. Can J Anaesth 1993; 40: 532–41 recommendations to use perioperative b-blocker therapy in
12 Bhatt DL. Peripheral arterial disease in the catheterization labor- patients undergoing noncardiac surgery based on reliable
atory: an underdetected and undertreated risk factor. Mayo Clin evidence? CMAJ 2004; 171: 245–7
Proc 2004; 79: 1107–9 33 Dorman BH, Zucker JR, Verrier ED, Gartman DM, Slachman FN.
13 Boersma E, Mercado N, Poldermans D, Gardien M, Vos J, Clonidine improves perioperative myocardial ischemia, reduces
Simoons ML. Acute myocardial infarction. Lancet 2003; 361: anesthetic requirement, and alters hemodynamic parameters in
847–58 patients undergoing coronary artery bypass surgery. J Cardiothorac
14 Boersma E, Poldermans D, Bax JJ, et al., for the DECREASE Vasc Anesth 1993; 7: 386–95
Study Group. Predictors of cardiac events after major vascular 34 Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guideline update
surgery: role of clinical characteristics, dobutamine for perioperative cardiovascular evaluation for noncardiac sur-
echocardiography, and beta-blocker therapy. JAMA 2001; 285: gery update: a report of the American College of Cardiology/
1865–73 American Heart Association Task Force on Practice Guidelines
15 Breen P, Lee J-W, Pomposelli F, Park KW. Timing of high-risk (Committee to Update the 1996 Guidelines on Perioperative
vascular surgery following coronary artery bypass surgery: a Cardiovascular Evaluation for Noncardiac Surgery). 2002.

15
 Priebe

American College of Cardiology Web site. Available at: www. 52 Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathophy-
acc.org/clinical/guidelines/perio/dirIindex.htm siology of coronary artery disease and the acute syndromes (I).
35 Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guideline update N Engl J Med 1992; 326: 242–50
for perioperative cardiovascular evaluation for noncardiac 53 Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathophy-
surgery—Executive Summary. A report of the American College siology of coronary artery disease and the acute syndromes (II).
of Cardiology/American Heart Association Task Force on Prac- N Engl J Med 1992; 326: 310–8
tice Guidelines (Committee to Update the 1996 Guidelines on 54 Fuster V, Fayad ZA, Badimon JJ. Acute coronary syndromes:
Perioperative Cardiovascular Evaluation for Noncardiac Surgery). biology. Lancet 1999; 353: (Suppl 2): S115–9
Circulation 2002; 105: 1257–67 55 Geft IL, Fishbein MC, Ninomyia K, et al. Intermittent brief periods
36 Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004 guideline of ischemia have a cumulative effect and may cause myocardial
update for coronary artery bypass graft surgery: summary article: necrosis. Circulation 1982; 66: 1150–3
a report of the American College of Cardiology/American Heart 56 Gertz SD, Roberts WC. Hemodynamic shear force in rupture of
Association Task Force on Practice Guidelines (Committee to coronary arterial atherosclerotic plaques. Am J Cardiol 1990; 66:
update the 1999 Guidelines on Coronary Artery Bypass Graft 1368–72
Surgery). J Am Coll Cardiol 2004; 44: 1146–54 57 Gheorghiade M, Goldstein S. b-Blockers in the post-myocardial
37 Eagle KA, Rihal CS, Mickel MC, Holmes DR, Foster ED, Gersh BJ. infarction patient. Circulation 2002; 106: 394–8
Cardiac risk of noncardiac surgery: influence of coronary disease 58 Gottlieb A, Banoub M, Sprung J, Levy PJ, Beven M, Mascha EJ.
and type of surgery in 3368 operations. Circulation 1997; 96: Perioperative cardiovascular morbidity in patients with coronary
1882–7 artery disease undergoing vascular surgery after percutaneous
38 Ellis JE, Drijvers G, Pedlow S, et al. Premedication with oral and transluminal coronary angioplasty. J Cardiothorac Vasc Anesth
transdermal clonidine provides safe and efficacious postoperative 1998; 12: 501–6
sympatholysis. Anesth Analg 1994; 79: 1133–40 59 Grayburn PA, Hillis LD. Cardiac events in patients undergoing
39 Ellis SG, Hertzer NR, Young JR, Brener S. Angiographic correlates noncardiac surgery: shifting the paradigm from noninvasive risk
of cardiac death and myocardial infarction complicating major stratification to therapy. Ann Intern Med 2003; 138: 506–11
nonthoracic vascular surgery. Am J Cardiol 1996; 77: 1126–8 60 Grignani G, Soffiantino F, Zuchella M, et al. Platelet activation
40 Elmore JR, Hallett JW jr, Gibbons RJ, et al. Myocardial revascu- by emotional stress in patients with coronary artery disease.
larization before abdominal aortic aneurysmorrhaphy: effect of Circulation 1991; 83 (Suppl II): 128–36
coronary angioplasty. Mayo Clin Proc 1993; 68: 637–41 61 Grube E, Silber S, Hauptmann KE, et al. TAXUS I: six- and twelve-
41 Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation month results from a randomized, double-blind trial on a slow-
1995; 92: 657–71 release paclitaxel-eluting stent for de novo coronary lesions.
42 Farb A, Burke AP, Tang AL, et al. Coronary plaque erosion with- Circulation 2003; 107: 38–42
out rupture into a lipid core. A frequent cause of coronary throm- 62 Guidelines and indications for coronary artery bypass graft
bosis in sudden coronary death. Circulation 1996; 93: 1354–63 surgery: a report of the American College of Cardiology/
43 Feldman CL, Stone PH. Intravascular hemodynamic factors American Heart Association on Assessment of Diagnostic and
responsible for progression of coronary atherosclerosis and Therapeutic Cardiovascular Procedures (Subcommittee on
development of vulnerable plaque. Curr Opin Cardiol 2000; 15: Coronary Artery Bypass Graft Surgery). J Am Coll Cardiol 1991;
430–40 17: 543–89
44 Finkelstein A, McClean D, Kar S, et al. Local drug delivery via a 63 Henke PK, Blackburn S, Proctor MC, et al. Patients undergoing
coronary stent with programmable release pharmacokinetics. infrainguinal bypass to treat atherosclerotic vascular disease
Circulation 2003; 107: 777–84 are under-prescribed cardioprotective medications: effect on
45 Fleisher LA, Eagle KA, Shaffer T, Anderson GF. Perioperative and graft patency, limb salvage, and mortality. J Vasc Surg 2004; 39:
long-term mortality rates after major vascular surgery: the rela- 357–65
tionship to preoperative testing in the Medicare population. 64 Hertzer NR, Beven EG, Young JR, et al. Coronary artery disease
Anesth Analg 1999; 89: 849–55 in peripheral vascular patients. A classification of 1000 coronary
46 Fleisher LA, Nelson AH, Rosenbaum SH. Postoperative angiograms and results of surgical management. Ann Surg 1984;
myocardial ischemia: aetiology of cardiac morbidity or manifesta- 199: 223–33
tion of underlying disease? J Clin Anesth 1995; 7: 1–6 65 Heusch G, Schipke J, Thamer V. Clonidine prevents the
47 Forrester JS. Role of plaque rupture in acute coronary syndromes. sympathetic initiation and aggravation of poststenotic myocardial
Am J Cardiol 2000; 86 (Suppl): J15–23 ischemia. J Cardiovasc Pharmacol 1985; 7: 1176–82
48 Foster ED, Davis KB, Carpenter JA, Abele S, Fray D. Risk of 66 Howell SJ, Sear JW, Foex P. Peri-operative b-blockade: a useful
noncardiac operations in patients with defined coronary disease: treatment that should be greeted with cautious enthusiasm
The Coronary Artery Surgery Study (CASS) registry experience. (Editorial). Br J Anaesth 2001; 86: 161–4
Ann Thorac Surg 1986; 41: 42–50 67 Huber KC, Evans MA, Bresnahan JF, Gibbons RJ, Holmes DR jr.
49 Frank S, Beattie C, Christopherson R, et al. Unintentional hypo- Outcome of noncardiac operations in patients with severe
thermia is associated with postoperative myocardial ischemia. coronary artery disease successfully treated preoperatively
The Perioperative Ischemia Randomized Anesthesia Trial Study with coronary angioplasty. Mayo Clin Proc 1992; 67: 15–21
Group. Anesthesiology 1993; 78: 468–76 68 Hunt SA, Baker DW, Chin MH, et al. ACC/AHA Guidelines for
50 Frank S, Fleisher L, Breslow M, et al. Perioperative maintenance the Evaluation and Management of Chronic Heart Failure in the
of normothermia reduces the incidence of morbid cardiac events. Adult: A Report of the American College of Cardiology/American
A randomized clinical trial. JAMA 1997; 277: 1127–34 Heart Association Task Force on Practice Guidelines (Committee
51 Freemantle N, Cleland J, Young P, Mason J, Harrison J. b-Blockade to Revise the 1995 Guidelines for the Evaluation and Management
after myocardial infarction: systematic review and meta regres- of Heart Failure). Available at: http://www.acc.org/clinical/
sion analysis. Br Med J (Clin Res Ed) 1999; 318: 1730–7 guidelines/failure/hf_index.htm

16
 Perioperative myocardial infarction

69 Ikonomidis I, Andreotti F, Economou E, et al. Increased 89 Lee T-S, Chang C-C, Zhu Y, Shyy JYJ. Simvastatin induces heme
proinflammatory cytokines in patients with chronic stable oxygenase-1. A novel mechanism of vessel protection. Circulation
angina and their reduction by aspirin. Circulation 1999; 100: 2004; 110: 1296–301
793–8 90 Libby P. Current concepts of the pathogenesis of the acute cor-
70 Jenkins NP, Keevil BG, Hutchinson IV, Brooks NH. Beta-blockers onary syndromes. Circulation 2001; 104: 365–72
are associated with lower C-reactive protein concentration in 91 Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis.
patients with coronary artery disease. Am J Med 2002; 112: Circulation 2002; 105: 1135–43
269–74 92 Lindahl B, Toss H, Siegbahn A, et al., the Fragmin during Instability
71 Jones KG, Powell JT. Slowing the heart saves lives: advantages of in Coronary Artery Disease (FRISC) Study Group. Markers of
perioperative b-blockade. Br J Surg 2000; 87: 689–90 myocardial damage and inflammation in relation to long-term
72 Kaluza GL, Joseph J, Lee JR, Raizner ME, Raizner AE. Catastrophic mortality in unstable coronary artery disease. N Engl J Med
outcomes of noncardiac surgery soon after coronary stenting. 2000; 343: 1139–47
J Am Coll Cardiol 2000; 35: 1288–94 93 Lindenauer PK, Pekow P, Wang K, Gutierrez B, Benjamin EM.
73 Kaul P, Newby LK, Fu Y, et al. Troponin T and quantitative ST- Lipid-lowering therapy and in-hospital mortality following major
segment depression offer complimentary prognostic information noncardiac surgery. JAMA 2004; 291: 2092–9
in the risk stratification of acute coronary syndrome patients. J Am 94 London MJ, Zaugg M, Schaub MC, Spahn DR. Perioperative
Coll Cardiol 2003; 41: 371–80 b-adrenergic receptor blockade. Physiologic foundations and
74 Keith A, Fox A. Management of acute coronary syndromes: an clinical controversies. Anesthesiology 2004; 100: 170–5
update. Heart 2004; 90: 698–706 95 Lopez-Jimenez F, Goldman L, Sacks DB, et al. Prognostic value of
75 Kereiakes DJ. The emperor’s clothes. In search of the vulnerable cardiac troponin T after noncardiac surgery: 6-month follow-up
plaque. Circulation 2003; 107: 2076–7 data. J Am Coll Cardiol 1997; 29: 1241–5
76 Kertai MD, Bax JJ, Klein J, Poldermans D. Is there any reason to 96 MacNeill BD, Jang I-K, Bouma BE, et al. Focal and multi-focal
withhold b blockers from high-risk patients with coronary artery plaque macrophage distributions in patients with acute and stable
disease during surgery? Anesthesiology 2004; 100: 4–7 presentations of coronary artery disease. J Am Coll Cardiol 2004;
77 Kertai MD, Boersma E, Westerhout CM, et al. Association 44: 972–9
between long-term statin use and mortality after successful 97 Mangano DT. Adverse outcomes after surgery in the year
abdominal aortic aneurysm surgery. Am J Med 2004; 116: 96–103 2001—a continuing odyssey (Editorial). Anesthesiology 1998; 88:
78 Kim LJ, Martinez EA, Faraday N, et al. Cardiac troponin I predicts 561–4
short-term mortality in vascular surgery patients. Circulation 2002; 98 Mangano DT. Perioperative cardiac morbidity. Anesthesiology
106: 2366–71 1990; 72: 153–84
79 Krupski WC. Update on perioperative evaluation and manage- 99 Mangano DT, for the Multicenter Study of Perioperative Ischemia
ment of cardiac disease in vascular surgery patients. J Vasc Surg Research Group. Aspirin and mortality from coronary bypass
2002; 36: 1292–308 surgery. N Engl J Med 2002; 347: 1309–17
80 Landesberg G. The pathophysiology of perioperative myocardial 100 Mangano DT, Layug EL, Wallace A, Tateo I, for the Multicenter
infarction: facts and perspectives. J Cardiothorac Vasc Anesth 2003; Study of Perioperative Ischemia Research Group. Effect
17: 90–100 of atenolol on mortality and cardiovascular morbidity after
81 Landesberg G, Luria MH, Cotev S, et al. Importance of noncardiac surgery: Multicenter Study of Perioperative Ischemia
long-duration postoperative ST-segment depression in cardiac Research Group. N Engl J Med 1996; 335: 1713–20
morbidity after vascular surgery. Lancet 1993; 341: 715–9 101 Mangano DT, Siliciano D, Hollenberg M, et al. Postoperative
82 Landesberg G, Mosseri M, Shatz V, et al. Cardiac troponin after myocardial ischemia therapeutic trials using intensive analgesia
major vascular surgery. The role of perioperative ischemia, pre- following surgery. Anesthesiology 1992; 76: 343–53
operative thallium scanning, and coronary revascularization. J Am 102 Mann J, Davies MJ. Mechanisms of progression in native coronary
Coll Cardiol 2004; 44: 569–75 artery disease: role of healed plaque disruption. Heart 1999; 82:
83 Landesberg G, Mosseri M, Wolf Y, et al. Perioperative myo- 265–8
cardial ischemia and infarction. Identification by continuous 103 Maseri A, Fuster V. Is there a vulnerable plaque? Circulation 2003;
12-lead electrocardiogram with online ST-segment monitoring. 107: 2068–71
Anesthesiology 2002; 96: 262–70 104 Mason JJ, Owens DK, Harris RA, Cooke JP, Hlatky MA. The role
84 Landesberg G, Mosseri M, Wolf YG, et al. Preoperative thallium of coronary angiography and coronary revascularization before
scanning, selective coronary revascularization, and long-term noncardiac surgery. JAMA 1995; 273: 1919–25
survival after major vascular surgery. Circulation 2003; 108: 105 Massie MT, Rohrer MJ, Leppo JA, Cutler BS. Is coronary
177–83 angiography necessary for vascular surgery patients who have
85 Landesberg G, Mosseri M, Zahger D, et al. Myocardial infarction positive results of dipyridamole thallium scans? J Vasc Surg
following vascular surgery: the role of prolonged, stress-induced, 1997; 25: 975–82
ST-depression-type ischemia. J Am Coll Cardiol 2001; 37: 1858–63 106 Master AM, Dack S, Jaffe H. Perioperative coronary artery
86 Landesberg G, Shatz V, Akopnik I, et al. Association of cardiac occlusion. JAMA 1938; 110: 1415–8
troponin, CK-MB, and postoperative myocardial ischemia with 107 McCann RL, Clements FM. Silent myocardial ischemia in patients
long-term survival after major vascular surgery. J Am Coll Cardiol undergoing peripheral vascular surgery: incidence and association
2003; 42: 1547–54 with perioperative cardiac morbidity and mortality. J Vasc Surg
87 Lee TH. Reducing cardiac risk (Editorial). N Engl J Med 1999; 341: 1989; 9: 583–7
1838–40 108 McSPI-Europe Research Group. Perioperative sympatholysis:
88 Lee TH, Thomas EJ, Ludwig LE, et al. Troponin T as a marker for beneficial effects of the a2-adrenoceptor agonist mivazerol on
myocardial ischemia in patients undergoing major noncardiac hemodynamic stability and myocardial ischemia. Anesthesiology
surgery. Am J Cardiol 1996; 77: 1031–6 1997; 86: 346–63

17
 Priebe

109 Mesh CL, Cmolik BL, Van Heekeren DW, et al. Coronary bypass myocardial infarction in patients undergoing peripheral vascular
in vascular patients: a relatively high-risk procedure. Ann Vasc Surg surgery. J Vasc Surg 1989; 10: 617–25
1997; 11: 612–9 128 Poldermans D, Bax JJ, Kertai MD, et al. Statins are associated with
110 Metzler H, Gries M, Rehak P, Lang T, Frühwald S, Toller W. a reduced incidence of perioperative mortality in patients under-
Perioperative myocardial cell injury: the role of troponins. Br J going major noncardiac vascular surgery. Circulation 2003; 107:
Anaesth 1997; 78: 386–90 1848–51
111 Mittleman MA, Maclure M, Tofler GH, et al. Triggering of acute 129 Poldermans D, Boersma E, Bax JJ, et al., for the Dutch Cardiac Risk
myocardial infarction by heavy physical exertion: protection Evaluation Applying Stress Echocardiography Group. Bisoprolol
against triggering by regular exercise. N Engl J Med 1993; 329: reduces cardiac death and myocardial infarction in high-risk
1677–83 patients as long as 2 years after successful major vascular surgery.
112 Moon JCC, De Arenaza DP, Elkington AG, et al. The pathological Eur Heart J 2001; 22: 1353–8
basis of Q-wave and non-Q-wave myocardial infarction. J Am Coll 130 Poldermans D, Boersma E, Bax JJ, et al., for the Dutch Echocar-
Cardiol 2004; 44: 554–60 diographic Cardiac Risk Evaluation Applying Stress Echocardio-
113 Mukherjee D, Eagle KA. Perioperative cardiac assessment for graphy Study Group. The effect of bisoprolol on perioperative
noncardiac surgery: eight steps to the best possible outcome. mortality and myocardial infarction in high-risk patients under-
Circulation 2003; 107: 2771–4 going vascular surgery. N Engl J Med 1999; 341: 1789–94
114 Muzi M, Goff DR, Kampine JP, et al. Clonidine reduces sympath- 131 Posner KL, Van Norman GA, Chan V. Adverse cardiac
etic activity but maintains baroreflex responses in normotensive outcomes after noncardiac surgery in patients with prior
humans. Anesthesiology 1992; 77: 864–71 percutaneous transluminal coronary angioplasty. Anesth Analg
115 Naghavi M, Libby P, Falk E, et al. From vulnerable plaque to vul- 1999; 89: 553–60
nerable patient. A call for new definitions and risk assessment 132 Priebe HJ. Triggers of perioperative myocardial ischaemia and
strategies: part I. Circulation 2003; 108: 1664–72 infarction. Br J Anaesth 2004; 93: 9–20
116 Nelson AH, Fleisher LA, Rosenbaum SH. Relationship between 133 Raby KE, Brull SDJ, Timimi F, et al. The effect of heart rate control
postoperative anaemia and cardiac morbidity in high-risk vascular on myocardial ischemia among high-risk patients after vascular
patients in the intensive care unit. Crit Care Med 1993; 21: 860–6 surgery. Anesth Analg 1999; 88: 477–82
117 Neumayer G, Gaenzer H, Pfister R, et al. Plasma level of cardiac 134 Reul GJ, Cooley DA, Duncan JM, et al. The effect of coronary
troponin I after prolonged strenuous exercise. Am J Cardiol 2001; bypass on the outcome on peripheral vascular operations in 1093
87: 369–71 patients. J Vasc Surg 1986; 3: 788–98
118 Nielsen JL, Page CP, Mann C, Schwesinger WH, Fountain RL, 135 Ridker PM, Cushman M, Stampfer MJ, et al. Inflammation, aspirin,
Grover FL. Risk of major elective operation after myocardial and the risk of cardiovascular disease in apparently healthy men.
revascularization. Am J Surg 1992; 164: 423–6 N Engl J Med 1997; 336: 973–9
119 Nishina K, Mikawa K, Uesugi T, et al. Efficacy of clonidine for 136 Ridker PM, Manson JE, Buring JE, et al. Circadian variation of acute
prevention of perioperative myocardial ischemia: a critical myocardial infarction and the effect of low-dose aspirin in a ran-
appraisal and meta-analysis of the literature. Anesthesiology domized trial of physicians. Circulation 1990; 82: 897–902
2002; 96: 323–9 137 Sabia PJ, Powers ER, Ragosta M, et al. An association between
120 Okazaki S, Yokoyama T, Miyauchi K, et al. Early statin treatment in collateral blood flow and myocardial viability in patients
patients with acute coronary syndrome. Demonstration of the with recent myocardial infarction. N Engl J Med 1992; 327:
beneficial effect on atherosclerotic lesions by serial volumetric 1825–31
intravascular ultrasound analysis during half a year after coronary 138 Sambola A, Osende J, Hathcock J, et al. Role of risk factors in the
event: The ESTABLISH Study. Circulation 2004; 110: 1061–8 modulation of tissue factor activity and blood thrombogenicity.
121 O’Rourke DJ, Quinton HB, Piper W, et al., for the Northern New Circulation 2003; 107: 973–7
England Cardiovascular Disease Study Group. Ann Thorac Surg 139 Shammash JB, Trost JC, Gold JM, et al. Perioperative b-blocker
2004; 78: 466–70 withdrawal and mortality in vascular surgical patients. Am Heart J
122 Oliver MF, Goldman L, Julian DG, Holme I. Effect of mivazerol on 2001; 141: 148–53
perioperative cardiac complications during non-cardiac surgery in 140 Shojania KG, Duncan BW, McDonald KM, Wachter RM. Safe but
patients with coronary heart disease: the European Mivazerol sound: patient safety meets evidence-based medicine. JAMA 2002;
Trial (EMIT). Anesthesiology 1999; 91: 951–61 288: 508–13
123 Oscarsson A, Eintrei C, Anskär S, et al. Troponin T-values provide 141 Singh N, Langer A. Current status of silent myocardial ischemia.
long-term prognosis in elderly patients undergoing non-cardiac Can J Cardiol 1995; 11: 286–9
surgery. Acta Anaesthesiol Scand 2004; 48: 1071–9 142 Singh M, Lennon RJ, Darbar D, Gersh B, Holmes DR, Rihal C.
124 Ouyang P, Gerstenblith G, Furman WR, et al. Frequency and Effect of peripheral arterial disease in patients undergoing cor-
significance of early postoperative silent myocardial ischemia in onary intervention with intracoronary stents. Mayo Clin Proc 2004;
patients having peripheral vascular surgery. Am J Cardiol 1989; 64: 79: 1113–8
1113–6 143 Smith SC Jr, Blair SN, Bonow RO, et al. AHA/ACC Scientific
125 Palda VA, Detsky AS. Perioperative assessment and management Statement: AHA/ACC guidelines for preventing heart attack
of risk from coronary artery disease. Clinical Guideline, Part II. and death in patients with atherosclerotic cardiovascular disease:
Ann Intern Med 1997; 127: 313–28 2001 update. A statement for healthcare professionals from the
126 Pasternak PF, Grossi EA, Baumann G, et al. Silent myocardial American Heart Association and the American College of
ischemia monitoring predicts late as well as perioperative cardiac Cardiology. Circulation 2001; 104: 1577–9
events in patients undergoing vascular surgery. J Vasc Surg 1992; 144 Smith SC jr, Dove JT, Jacobs AK, et al. ACC/AHA guidelines for
16: 171–80 percutaneous coronary intervention: a report of the American
127 Pasternak PF, Grossi EA, Baumann G, et al. The value of silent College of Cardiology/American Heart Association Task Force
myocardial ischemia monitoring in the prediction of perioperative on Practice Guidelines (Committee to Revise the 1993 Guidelines

18
 Perioperative myocardial infarction

for Percutaneous Transluminal Coronary Angioplasty). J Am Coll 158 Wallace AW, Galindez D, Salahieh A, et al. Effect of clonidine on
Cardiol 2001; 37: 2215–38 cardiovascular morbidity and mortality after noncardiac surgery.
145 Spencer FA, Allegrone J, Goldberg RJ, et al., for the Grace Inves- Anesthesiology 2004; 101: 284–93
tigators. Association of statin therapy with outcomes of acute 159 Wheatley KW, Clayton D. Be skeptical about unexpected large
coronary syndromes: the GRACE study. Ann Intern Med 2004; apparent treatment effects: the case of an MRC AML 12
140: 857–66 randomization. Control Clin Trials 2003; 24: 66–70
146 Steinberg JB, Kresowik TF, Behrendt DM. Prophylactic myocardial 160 WHO MONICA Project Principal Investigators. World
revascularization based on dipyridamole-thallium scanning before Health Organization MONICA Project (monitoring trends
peripheral vascular surgery. Cardiovasc Surg 1993; 1: 552–7 and determinants of cardiovascular disease): a major international
147 Stevenson LW. Beta-blockers for stable heart failure. N Engl J Med collaboration. J Clin Epidemiol 1988; 41: 105–14
2002; 346: 1346–7 161 Wijeysundera DN, Naik JS, Beattie S. Alpha-2 adrenergic agonists
148 Stone JG, Foex P, Sear JW, et al. Myocardial ischemia in untreated to prevent perioperative cardiovascular complications: a meta-
hypertensive patients: Effects of a single small dose of a beta- analysis. Am J Med 2003; 114: 742–52
adrenergic blocking agent. Anesthesiology 1988; 68: 495–500 162 Wilson HS, Fasseas P, Orford JL, et al. Clinical outcome of patients
149 The CIBIS-II Investigators. The Cardiac Insufficiency undergoing non-cardiac surgery in the two months following
Bisoprolol Study II (CIBIS-II): a randomized trial. Lancet 1999; coronary stenting. J Am Coll Cardiol 2003; 42: 234–40
353: 9–13 163 Wilson RH, Rihal CS, Bell MR, Velianou JL, Holmes DR jr,
150 The Joint European Society of Cardiology/American College of Berger PB. Timing of coronary stent thrombosis in patients
Cardiology Committee. Myocardial infarction redefined—a con- treated with ticlopidine and aspirin. Am J Cardiol 1999; 83:
sensus document of the Joint European Society of Cardiology/ 1006–11
American College of Cardiology Committee for the redefinition
of myocardial infarction. Eur Heart J 2000; 21: 1502–13
151 Thompson JP. Ideal peri-operative management of patients with
cardiovascular disease: the quest continues (Editorial). Anaesthesia Addendum
2004; 59: 417–21 Since preparation of the manuscript, the results of a randomized study
152 Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, et al. Myocardial were published which investigated the effect of preoperative coronary
infarction and coronary deaths in the World Health Organization revascularization (by either percutaneous coronary intervention or
MONICA Project. Registration procedures, event rates, and bypass surgery) on long-term outcome in 510 patients undergoing
case-fatality rates in 38 populations from 21 countries in four vascular surgery.1 Patients with severe coronary artery disease, poor
continents. Circulation 1994; 90: 583–612 left ventricular function or severe aortic stenosis were excluded from
153 Ueda Y, Nanto S, Komamura K, Kodama K. Neointimal coverage randomization. At a median follow-up time of 2.7 yr, there was no
of stents in human coronary arteries observed by angioscopy. J Am difference in mortality between the revascularized and non-revascu-
Coll Cardiol 1994; 23: 341–6 larized groups (22% and 23%, respectively). Twenty-four months after
154 Urban MK, Markowitz SM, Gordon MA, et al. Postoperative pro- randomization, the vast majority of patients were taking beta-blockers
phylactic administration of b-adrenergic blockers in patients at (approx. 80%), statins (approx. 70%), aspirin (approx. 85%) and angio-
risk of myocardial ischemia. Anesth Analg 2000; 90: 1257–61 tensin-converting-enzyme inhibitors (approx. 55%). The results sug-
155 Vaughan CJ, Gotto AM jr. Update on statins: 2003. Circulation gest that in patients with stable coronary artery disease and unimpaired
2004; 110: 886–92 left ventricular function who receive excellent perioperative medical
156 Vicenzi MN, Ribitsch D, Luha O, Klein W, Metzler H. Coronary therapy, coronary revascularization before vascular surgery does not
artery stenting before noncardiac surgery: more threat than improve long-term outcome.
safety? Anesthesiology 2001; 94: 367–8
157 Wallace AW, Layug B, Tateo I., et al., for the McSPI Research 1 McFalls EO, Ward HB, Moritz TE, et al. Coronary-artery
Group. Prophylactic atenolol reduces postoperative myocardial revascularization before elective major vascular surgery. N Engl J
ischemia. Anesthesiology 1998; 88: 7–17 Med 2004; 351: 2795–804

19