Advances in nuclear cardiology: Preoperative risk ... · Advances in nuclear cardiology: Preoperative risk stratiﬁcation Kenneth A. Brown, MD, FACC The determination of cardiac

MAJOR ACHIEVEMENTS IN NUCLEAR CARDIOLOGY: VI

tionessgoinucimpwilthisandconrese

forschsevdeacaraloespperhistwitmorep

cardiseare(CAwelthis

ass

FromF

RepUC

J Nu107Copdoi:

Advances in nuclear cardiology: Preoperative riskstratification

Kenneth A. Brown, MD, FACC

cohcritGoothtionreflgenclincohsurimaasyGodiaaorin cratevasbyDriYeaMIclinpatingatteimpthepatSuctheas s

in gcordiaatioco-diseoptwasstreas a

The determination of cardiac risk and the formula-of risk-management strategies have become an

ential part of the perioperative care of patients under-ng noncardiac surgery. Over the past 2 decades,lear cardiology techniques have developed into anortant component of such management. This review

l explore the major advances of nuclear cardiology inclinical setting, examining the historical backgroundcontext, as well as the evolution of current data and

cepts, and will present a hint of future areas ofarch.

HISTORICAL BACKGROUND

The development of nuclear cardiology techniquesassessing cardiac risk and guiding therapy in patientseduled for noncardiac surgery has been driven byeral factors. First, perioperative cardiac morbidity andth comprise a major medical and economic healthe problem, consuming $12 billion in the United Statesne.1 Furthermore, perioperative cardiac events carryecially poor outcomes. Mortality rates associated withioperative myocardial infarction (MI) have rangedorically from 26% to 70%, averaging 50%.2 Evenh decades of medical advances, the perioperative MIrtality rate remains very substantial, with recentorts of 15% to 26%.3,4

Second, the expense and outcome of perioperativediac events are primarily related to coronary heartase, as the majority of perioperative cardiac eventsrelated to underlying coronary artery disease

D).5-8 Nuclear cardiology techniques are particularlyl suited for diagnostic and prognostic evaluation inpopulation.Third, there was a growing recognition that clinical

essment of cardiac risk was inadequate in surgical

the Cardiology Unit, University of Vermont College of Medicine,letcher Allen Medical Center, Burlington, Vt.rint requests: Kenneth A. Brown, MD, FACC, Cardiology Unit,niversity of Vermont College of Medicine, Fletcher Allen Medicalenter, Burlington, VT 05401; [email protected] Cardiol 2004;11:335-48.

1-3581/$30.00yright © 2004 by the American Society of Nuclear Cardiology.

cian10.1016/j.nuclcard.2004.03.002

orts that had a high prevalence of CAD. The clinicaleria for cardiac risk determination developed byldman et al,9 Detsky et al,10 Dripps et al,11 anders12-14 were derived from general surgical popula-s in whom the prevalence of CAD is quite low,ecting the 1% to 10% prevalence of CAD in theeral population, based on age. The application ofical risk criteria generated from low-prevalence CADorts may not have relevance in high-prevalence CADgical cohorts such as vascular surgery, where approx-tely 60% of patients have CAD, many of whom aremptomatic.5 In fact, Jeffrey et al15 found that theldman index greatly underestimated the risk of car-c events in patients undergoing elective abdominaltic aneurysm surgery. The cardiac event rate was 8%lass I patients, who were predicted to have an eventof less than 1%. Similarly, Lette et al16 showed that

cular surgery patients who were identified as low riskthe clinical criteria of Goldman et al, Detsky et al,pps et al/American Society of Anesthesiologists, orger et al13 actually had perioperative cardiac death orrates of 5% to 8%, far higher than predicted by theical indices and significantly higher than observed in

ients with normal stress myocardial perfusion imag-(MPI) results (Figure 1). Although more recent

mpts to modify these classic scoring systems haveroved the predictive value of clinical risk indices,

y remain insensitive for identifying cardiac risk inients who undergo high-risk vascular procedures.17

h inadequacies of clinical criteria led to an interest inpredictive value of cardiac imaging techniques suchtress nuclear MPI in this setting.Fourth, although it was recognized that stress testingeneral could be helpful in assessing the response of

onary heart disease patients to the stresses of noncar-c surgery, many potential candidates for such evalu-n were unable to perform exercise stress testing. Age,morbidity, and especially lower-extremity vascularase frequently precluded exercise stress testing as an

ion. Fortunately, interest in a nuclear imaging optiondeveloping at the same time that pharmacologic

ss (in particular, vasodilator stress) became availablen alternative to exercise stress.Thus stress nuclear MPI was poised to assist physi-
s in making rational recommendations when asked to
335

evaasspro

EV

Gen

MPpertiensurdic16carrevnucmacon

ibleeveativinstto tcar

takofelecanaativsiondefwitsubofout

Ext

largjeodefa wCAthenuc(yeabnuumreflsmaonl

Figpatistre

Figfunhist

FiginindiSoc

336 Brown Journal of Nuclear CardiologyAdvances in nuclear cardiology: Preoperative risk stratification May/June 2004

luate a patient before noncardiac surgery and to helpess the risk/benefit ratio of deciding whether toceed with such planned surgery.

OLUTION OF CURRENT DATA AND CONCEPTS

eral Predictive Value

Preoperative risk stratification with stress nuclearI was first reported by Boucher et al18 in 1985. Theyformed dipyridamole–thallium 201 MPI in 48 pa-ts with evidence of CAD before peripheral vascular

gery. Perioperative cardiac events could not be pre-ted by clinical factors. However, they found that 8 ofpatients (50%) with reversible Tl-201 defects had

diac events compared with 0 of 32 patients withoutersible defects (P � .0001), and a new growth area inlear cardiology was set in motion. Since this land-rk study, many other subsequent studies19-42 havefirmed the central element of these findings: revers-

ure 2. Perioperative (Periop) cardiac death/MI rate inents with or without reversible defects (RD) on preoperative

ure 1. Perioperative (PERIOP) cardiac death (CD)/MI ratepatients defined as low risk by various clinical scoringces and dipyridamole (DP) stress MPI. ASA, Americaniety of Anesthesiologists.

impss nuclear MPI.

defects are the best predictor of perioperative cardiacnts, confirmed by a large meta-analysis of perioper-e stress imaging studies.42 However, it remainsructive to examine the evolution of data as it relateshe maturation of clinical thinking about the utility ofdiac imaging in the preoperative setting.

The next logical step in the development of data wasen by Leppo et al41 by comparing the predictive valuedipyridamole–Tl-201 MPI with clinical stress andtrocardiographic parameters using a multivariatelysis. They found that the best predictor of perioper-e death or MI was the presence of reversible perfu-

defects: 14 of 42 patients (33%) with reversibleects died or had MI compared with only 1 of 47 (2%)hout reversible defects (P � .001) (Figure 2). In agroup of patients undergoing exercise, the presencereversible defects remained the best predictor ofcome.

ent of Jeopardized Viable Myocardium

These findings, of course, are consistent with a verye body of work demonstrating that the presence of

pardized viable myocardium, reflected in reversibleects on stress MPI, has powerful prognostic value inide spectrum of coronary heart disease from chronicD to acute coronary syndromes.39,40 However, one offundamental tenets of the prognostic value of stresslear MPI is that cardiac risk is not simply categoricals/no based on presence or absence of perfusionormalities) but rather is best understood as a contin-, proportional to the extent of myocardium at risk

ected in the extent of reversible defects.39,40 Thus all degree of jeopardized viable myocardium reflects

y a small risk of cardiac events. Therefore the next

ure 3. Perioperative (Periop) cardiac death/MI rate as action of reversible defects on vasodilator stress MPI andory of diabetes mellitus.

ortant advancement in the thinking and evolution of

nuconsviathaor hnumBrodiajeooffinddetass

Int

gat

nucfordictAltcleapateveThequain msmaevebetNevcritpreto a

Tab

VasBCFSEYMLHKBBKMT(2

OthCSBYSVT

N/A

Journal of Nuclear Cardiology Brown 337Volume 11, Number 3;335-48 Advances in nuclear cardiology: Preoperative risk stratification

lear MPI preoperative risk stratification was to dem-trate a similar link between the extent of jeopardizedble myocardium and cardiac risk. Lette et al43 showedt patients could be classified as being at low, medium,igh risk for perioperative cardiac events based on theber of segments with reversible defects. Similarly,

wn and Rowen35 showed that the perioperative car-c death/MI rate was directly related to the extent ofpardized viable myocardium reflected in the numbersegments with reversible defects (Figure 3). Theseings allowed clinicians to have more latitude in

ermining perioperative patient care, as cardiac riskessment could be more “fine-tuned.”

egration With Clinical Data

Contemporaneous with these developments, investi-

le 1. Pharmacologic MPI for preoperative assessmen

Study

Thalliumredistribution

(%)

Perioevede

cular surgery onlyoucher et al18 (1985) 33utler and Leppo19 (1987) 47letcher et al20 (1988) 22achs et al21 (1988) 31agle et al22 (1989) 41ounis et al23 (1990) 36angano25 (1991) 37

ette et al26 (1992) 45endel et al27 (1992) 51resowik et al28 (1993) 39aron et al29 (1994) 35ry et al30 (1994) 46outelou et al31 (1995) 44arshall et al24 (1995) 47

otal (weighted average)417 total patients)

42

er surgeryoley et al33 (1992) 36haw et al34 (1992) 47rown et al35 (1993) 33ounis et al36 (1994) 31tratmann et al37 (1996) 29an Damme et al38 (1997) 34otal (weighted average)(923 total patients)

33

, Not available.

ors were recognizing that although preoperative stress

lear MPI had very powerful negative predictive valueperioperative cardiac events (99%), its positive pre-ive value was more limited (10%-15%) (Table 1).18-38

hough a 10% to 15% perioperative death or MI rate isrly “high risk,” nevertheless, the large majority of

ients with positive test results did not have cardiacnts, leading to inefficiencies of resource utilization.se findings are in part explained by the lack of anyntification of provokable ischemia seen on stress MPIost of the studies. As just discussed, patients with

ll degrees of ischemia are at low risk for cardiacnts, and most of these studies did not distinguishween small, medium, or large ischemia seen on MPI.ertheless, it was recognized that whereas clinical

eria alone may be inadequate to risk-stratify patientsoperatively, combining them with imaging could lead

more selective, cost-effective use of MPI.14

ardiac risk

iveI/

%)

Ischemia:Positive

predictivevalue (%)

Normal scan:Negative

predictive value(%)

19 10020 10037 10014 10016 9815 100

5 9517 9914 994 984 96

11 1006 100

16 9712 99

11 9921 10013 9918 986 99

N/A N/A13 99

t of c

peratnts Math (

61044875893573

107

410

59426

Eagle et al used a multivariate analysis of clinical

andprerevperofmemuscrestreintefac15%of30%risktest3 oevereq

cenrateand201impgroof t85%

catecantheformariskevefromwerdef

Figpresstre


imaging data in 200 vascular patients undergoingoperative dipyridamole–Tl-201 imaging. Althoughersible Tl-201 defects were the best predictor ofioperative cardiac events, 5 clinical variables (historyangina, Q waves, ventricular ectopy requiring treat-nt, diabetes, age �70 years) were also significantltivariate predictors. Using these clinical variables forening, the investigators showed that the benefit ofss MPI for risk stratification was greatest in thermediate-risk subgroup that had 1 or 2 clinical risk

tors (Figure 4). In this subgroup with an overall risk of, stress MPI provided a further 10-fold stratification

risk: 3% event rate without reversible defects versuswith reversible defects. In patients with no clinical

factors, the perioperative risk was low and additionaling was not required. On the other hand, in those withr more clinical risk factors, the risk of cardiacnts was high and additional testing was also notuired.L’Italien et al44 analyzed pooled data from several

ters and found that clinical information could sepa-patients into low-risk (3%), intermediate-risk (8%),high-risk (19%) subgroups. When dipyridamole–Tl-MPI data were applied, there was no significant

act on the low- or high-risk patients. However, in theup at intermediate clinical risk, which comprised 51%he total population, MPI data reclassified more than

Figure 4. Interaction of clinical risk factors andpatients undergoing noncardiac surgery. RD, Rev

of this cohort into either low-risk or high-risk defi

gories. Thus these studies suggest that clinical databe used as a first line of risk stratification, reservingmore sensitive but more expensive stress MPI testingthe intermediate-risk subgroup. However, stress MPIy still have value also in the patients at low clinical. In the study of Eagle et al,14 the risk of cardiacnts in patients without clinical risk factors increased

3% overall to 12% when reversible MPI defectse present, compared with 0% when no reversibleects were present. Similarly, Hendel and Leppo45

ure 5. Perioperative cardiac event rate as a function ofence or absence of reversible defects (RD) on preoperativess MPI in patients classified as low risk by clinical criteria

12 14

1 stress imaging data on outcome in 200defects.

Tl-20ersible

ned by Eagle et al index A or index B.

foucritapponwh

fromHeaical

noncomdiathre(2)diausinme

sion.


nd that in patients considered to be at low risk by theeria of Eagle et al, the cardiac event rate wasroximately 10% when reversible defects were presentpreoperative stress MPI compared with 1% or loweren reversible defects were absent (Figure 5).

These data have led to a consensus recommendationthe American College of Cardiology/American

rt Association (ACC/AHA) taking into account clin-

Figure 6. ACC/AHA guidelines for evaluating patheart failure; Med Rx, medical therapy; fct capac, fsurg, surgical; intermed, intermediate; OR, operhistory of cerebrovascular accident; HT, hyperten

predictors, functional capacity, and the type/risk of aor

cardiac surgery46 (Figure 6). Although somewhatplicated, the indications for noninvasive stress car-

c imaging boil down to identification of at least two ofe considerations: (1) intermediate clinical predictors,poor functional capacity, and/or (3) high-risk noncar-c surgery (Table 2). Bartels et al47 showed that byg a protocol modeled after the ACC/AHA recom-

ndations, in a series of 203 patients scheduled for

efore noncardiac surgery. CHF, Congestivenal capacity; METS, metabolic equivalents;room; EKG, electrocardiogram; Hx CVA,

ients bunctioating

tic vascular surgery, the cardiac mortality rate for the

cohdiaFroprolineThecoraftewertestrev

cosperverperBeftestrisk.29patat h


ort was only 1%, presumably reflecting the impact ofgnostic testing on perioperative care. More recently,ehlich et al48 described the impact of an educationalgram designed to implement the ACC/AHA guide-s before elective abdominal aortic vascular surgery.y examined the use of preoperative stress testing,

onary angiography, and revascularization before andr implementation of the ACC/AHA guidelines. Theree marked reductions in rates of preoperative stressing (88% to 47%), angiography (24% to 11%), and

Figure 6. Continued

ascularization (24% to 2%), as well as preoperative pro

t ($1087 to $171), that were maintained over a 2 yeariod. Importantly, perioperative outcome was not ad-sely affected. In fact, there was a trend toward a lowerioperative death/MI rate (11% to 4%, P � .08).ore implementation, extensive preoperative stressing was used similarly in both patients at low clinicaland patients at high clinical risk (84% vs 91%, P �

). After implementation, the stress testing rate inients at low clinical risk was only half that of patientsigh clinical risk (31% vs 61%, P � .005). These data
vide compelling evidence that a rational integration of

clinachcom

Lon

sucwileveevetialclusurlatethepre

stabprevasearamdefdeatorconwereveraterevofgrowit

Tabprestre

1. Ia

2. P3. H

opa

CHF Figfundefe


ical data and preoperative noninvasive testing can beieved to optimize resource utilization and patient out-e.

g-Term Risk

When the prevalence of underlying CAD is high,h as in peripheral vascular disease subgroups, patientsl remain at risk for cardiac events in the long termn if they survive the immediate perioperative periodnt-free. Because long-term outcome can also poten-ly influence perioperative management decisions, in-ding whether to proceed with the planned noncardiacgery, the ability of preoperative stress MPI to predictcardiac events can have important clinical value. Onbasis of work in other cohorts of patients with a highvalence of CAD, such as after MI or with chronic

le 2. Simplified approach to ACC/AHAoperative evaluation guidelines46: Proceed toss imaging if two or more factors present

ntermediate clinical predictors are present (mildngina, prior MI, compensated CHF, or diabetes)oor functional capacity (�4 METs)igh-risk surgical procedure (emergency major

perations, aortic repair or peripheral vascular,rolonged surgical procedures with large fluid shiftsnd/or blood loss)

, Congestive health failure; METs, Metabolic equivalents.

Figure 8. Long-term postoperative MI-free surv(DIP)–Tl-201 MPI results. (Used with permissioPerioperative and long-term prognostic value of i
patients with peripheral vascular disease. Am Heart J 1
le angina,40,41 it is not surprising that stress MPIdicts long-term cardiac events in the postoperativecular population as well. Hendel et al49 examined thely and late predictive value of preoperative dipyrid-ole–Tl-201 MPI before vascular surgery. Reversibleects were the best predictor of early postoperativeth or MI, whereas fixed defects were the best predic-of late cardiac events (30 months) (Figure 7). Intrast, Younis et al50 found that reversible defectse the best predictor of both early and late cardiacnts (Figure 8). Importantly, the overall death or MIwas approximately 50% by 2 years in patients with

ersible defects, which comprised approximately 20%the cohort. Identification of such a sizable high-riskup can influence decisions about whether to proceedh planned noncardiac surgery. A longer follow-up

ure 7. Perioperative (PERIOP) and late death/MI rate as action of preoperative stress MPI results. RD, Reversiblects; FD, fixed defects; F/U, follow-up.

a function of preoperative dipyridamoleYounis LT, Aguirre F, Byers S, et al.

ous dipyridamole thallium scintigraphy in

ival asn from

ntraven
990;119:1287-92.)

studendpatstreevewit(Figpreper

ImpOu

abidatundpatporgiepatnonlon

amtivesholowsubabdcas�-bperwitverpata ratermevehosdiscthe�-bcuronnotwitnot�-bdefporin 1hadpatexcwasonl� .

sugnona sanedioaddtien

Figfunresuposreve

Tab

TI-2His


y has been reported for patients undergoing carotidarterectomy.51 Although perioperative risk was nil,ients with reversible defects on MPI with a positivess electrocardiogram showed a high late cardiacnt rate that approached 50% by 7 years comparedh only 2% in patients without provokable ischemiaure 9). Thus preoperative stress MPI continues to

dict cardiac events beyond the early perioperativeiod and can influence therapeutic strategies.

act of Therapy and Strategy on Perioperativetcome

The previous studies outlined thus far establish thelity of stress nuclear MPI (in concert with clinicala) to distinguish low-risk patients who can go on toergo scheduled noncardiac surgery from high-risk

ients who require additional consideration. Two im-tant issues remain: (1) Are there management strate-s that can reduce the cardiac event rate in high-riskients? (2) Does this entire approach of preoperativeinvasive imaging actually lead to a better short- andg-term outcome?

ure 9. Long-term postoperative event-free survival as action of exercise MPI and exercise electrocardiographylts. positive RD/ExECG, Reversible defect present with

itive exercise electrocardiography; negative RD/ExECG, norsible defects with negative exercise electrocardiography.

le 3. Perioperative outcome as a function of �-block

Periope

�-Blockers

01 reversible defects 0/48 (0%)tory of MI 0/51 (0%)

�-Blockers. Several studies have suggested aneliorating effect of �-blockers when used periopera-ly. In a case-controlled study, Pasternack et al52

wed that the use of metoprolol was associated with aer perioperative MI rate (3%) compared with controljects (17%) (P � .05) in 83 patients undergoingominal aortic aneurysm repair. In a much largere-controlled series of 2088 vascular surgery patients,lockers were associated with a 50% reduction inioperative MI.4 Metoprolol has also been associatedh a lower frequency of perioperative ischemia (2%sus 28%, P � .05) in a randomized trial of 200ients undergoing noncardiac surgery.53 More recently,ndomized trial by Mangano et al54 showed a long-

benefit from bisoprolol over a 2-year follow-up,n though the drug was administered only during thepitalization. Perhaps more relevant to the currentussion, Froehlich et al55 retrospectively examinedoutcome of high-risk patients as a function of

locker treatment (Table 3). Perioperative death oc-red in 11% to 12% of patients with reversible defectspreoperative stress MPI or with prior MI if they were

treated with �-blockers perioperatively comparedh 0% if they were treated with �-blockers. Though

a randomized trial, these data do suggest thatlockers can reduce the risk associated with reversibleects on preoperative MPI. This conclusion is sup-ted by a randomized trial of bisoprolol versus placebo12 patients scheduled for major vascular surgery whopositive dobutamine echocardiography.56 Although

ients with the highest-risk echocardiograms wereluded, the 30-day postoperative cardiac death/MI rate

34% in patients receiving placebo compared withy 3% in patients receiving perioperative bisoprolol (P001) (Figure 10).Coronary revascularization. Accumulating data

gest that revascularization of high-risk patients beforecardiac surgery can reduce their preoperative risk. In

eries of 500 patients scheduled for abdominal aorticurysm surgery, 240 had symptomatic or electrocar-graphic evidence of CAD and were referred foritional testing, including stress MPI.57 Of 212 pa-ts who were considered to be at low risk and were not

atment

e cardiac death

Pvalue

No �-blockers

8/65 (12%) .029/84 (11%) .01

er tre

rativ

refehadpatcorpatthacarstreRevwerAmundrevtivepatfore

whbefdeacaronlizatbefnotculpat

revpatstudcormesupSurwitbyppatgervasbetbypratecarsur12)patof byea

Figin perat

Figpativers

Fignon(Megraf


rred for coronary revascularization, 12 (6%) died orMI perioperatively. A smaller subgroup of 28

ients was considered to be at high risk and underwentonary revascularization before surgery. None of theseients had a perioperative cardiac event, suggestingt coronary revascularization can reduce perioperativediac risk. Subsequently, Elliot et al58 used dobutaminess Tl-201 MPI before scheduled vascular surgery.ersible defects were seen in 42 patients, of whom 15e denied vascular surgery based on imaging results.ong 18 patients with MPI ischemia who went on toergo vascular surgery without preoperative coronaryascularization, 5 (18%) died or had MI periopera-ly. However, no cardiac events occurred in the 9

ients who underwent coronary revascularization be-vascular surgery.Younis et al59 described the outcome of 72 patients

ure 10. Thirty-day postoperative (POST-OP) death/MI rateatients receiving bisoprolol versus standard (STD) periop-ive care.

ure 11. Perioperative death/MI and all–cardiac event rate inents who received preoperative therapeutic intervention

advus those who did not receive it.

o had abnormal dipyridamole–Tl-201 MPI resultsore major noncardiac surgery. The perioperativeth/MI rate was 31% in 36 patients undergoing non-diac surgery without any intervention compared withy 6% in 36 patients undergoing coronary revascular-ion or intensification of anti-ischemic medicationore noncardiac surgery (P � .01) (Figure 11). Ofe, none of the 6 patients undergoing coronary revas-arization had a cardiac event compared with 2 of 30ients (7%) with medical intensification alone.

These studies suggest that preoperative coronaryascularization can turn high-risk patients into low-riskients. Although the number of patients was small, they of Younis et al59 also raised the possibility that

onary revascularization may be more effective thandical intervention for lowering risk. This concern isported by a more recent analysis of Coronary Arterygery Study (CASS) data.60 In this population, patientsh CAD were randomized to medical treatment orass surgery. The authors looked at the subgroup of

ients who subsequently underwent noncardiac sur-y. They found that for high-risk surgery (abdominal,cular, thoracic, and head/neck), the outcome waster in patients who had undergone prior coronaryass surgery than those treated medically (death/MI, 2.5% versus 6.0%, respectively; P � .002). The

diac event rate of the patients who underwent bypassgery approached that of patients without CAD (Figure. The benefit of bypass surgery appeared greatest inients with multivessel CAD, and the protective effectypass surgery appeared to remain intact for at least 6rs. Although these latter two studies suggest an

ure 12. Perioperative death/MI rate in patients undergoingcardiac surgery who had no CAD, CAD treated medicallyd Rx), and CAD treated with coronary artery bypassting (CABG).

antage of revascularization over medical therapy for

redmeexameciu

impCAthewascomtion�-b

terMPlonthatialApreof 5mocorthaHolargtionrevshoacrthewhFinbyp� .

aretannonet aperperlarobsPospatwitsubrevundnor(incgrotheautlacthavnar

Figpatior(moto ltiondefepermPreizatCirc

FigpatimodcoroeratculaversMoseleafte


ucing perioperative risk, one must recognize thatdical treatment was not optimized in either study. Formple, in the study of Younis et al, intensification ofdical treatment included the use of nitrates and cal-m channel blockers, which have not been shown to

ure 13. Long-term survival after noncardiac surgery inents with preoperative stress MPI. Group I: Patients with nosmall defects; group II: patients with moderate to larged-lg) fixed defects (FD); group III: patients with moderatearge transient defects (TD) without coronary revasculariza-

(RV); group IV: patients with moderate to large transientcts who underwent coronary revascularization. (Used withission from Landesberg G, Mosseri M, Wolf YG, et al.

operative thallium scanning, selective coronary revascular-ion, and long-term survival after major vascular surgery.ulation 2003;108:177-83.)

ure 14. Long-term survival after noncardiac surgery inents with preoperative stress MPI. Group III: Patients witherate to large (mod-lg) transient defects (TD) withoutnary revascularization (RV); group IV: patients with mod-

e to large transient defects who underwent coronary revas-rization with coronary artery bypass grafting (CABG)us PTCA. (Used with permission from Landesberg G,

sseri M, Wolf YG, et al. Preoperative thallium scanning,ctive coronary revascularization, and long-term survival

desr major vascular surgery. Circulation 2003;108:177-83.)

rove outcome perioperatively. With regard to theSS data, the medical treatment was not specified andpercentage of patients who were taking �-blockers

not reported.60 Thus more data are needed topare the relative benefit of coronary revasculariza-versus an intense medical regimen, including full

lockade.Impact of coronary revascularization on long-

m outcome. As discussed above, preoperative stressI can determine not only perioperative risk but alsog-term outcome after surgery. A recent study suggestst preoperative coronary revascularization can substan-ly improve long-term outcome in high-risk patients.61

series of 407 major vascular surgery patients hadoperative stress MPI and were followed up for a mean5 months.61 The long-term outcome of patients with

derate to large reversible defects who did not undergoonary revascularization was substantially worse thant of patients with no or small defects (Figure 13).wever, the mortality rate of patients with moderate toe reversible defect who did undergo revasculariza-was reduced to that of the group with no or small

ersible defects (Figure 13). Propensity score analysiswed that the benefit of revascularization was presentoss a wide spectrum of clinical and MPI data. Fur-rmore, the benefit of revascularization was similaren left ventricular function was normal or reduced.ally, there was a trend toward a better outcome withass surgery compared with coronary angioplasty (P05) (Figure 14).Coronary angioplasty/stenting. Available datalimited regarding the impact of preoperative percu-

eous coronary intervention (PCI) on outcomes aftercardiac surgery. In a nonrandomized study Gottliebl62 examined the outcome of 194 patients undergoingcutaneous transluminal coronary angioplasty (PTCA)formed 3 to 49 days (median, 11 days) before vascu-surgery. A low postoperative cardiac event rate waserved: 1 death (0.5%) and 1 nonfatal MI (0.5%).ner et al63 compared the outcome of a series ofients undergoing PTCA before noncardiac surgeryh a case-controlled matched population of normaljects and patients with CAD who did not undergoascularization before noncardiac surgery. Patientsergoing PTCA had twice the cardiac event rate asmal subjects. However, the overall cardiac event rateluding angina and heart failure) was half that for theup with CAD without revascularization, althoughre was no difference in MI or mortality rate. Thehors concluded that there was no benefit for prophy-ic preoperative PTCA. Furthermore, recent studiese raised special concerns regarding the use of coro-y stents before noncardiac surgery. Kaluza et al64

cribed the outcome of 40 consecutive patients under-

goi(medeaandas alessthisdayappdinrecuntitorPCal65

non

sten168PCnonweesurcarsten�-bPTC

comdenin pgieare

Figvasmencom

FigguinmanCAB

FigasCorto n

FigsurgCABtien


ng coronary intervention with stents less than 6 weeksan, 13 days) before noncardiac surgery. There were 8ths (20%) (6 fatal MIs and 2 fatal bleeding episodes)1 nonfatal MI. The cardiac events were time-related,ll occurred in patients undergoing noncardiac surgerythan 14 days after PCI, with a 34% mortality rate insubgroup. Patients undergoing noncardiac surgery 1after PCI had an 80% mortality rate (4/5). Most MIs

eared to be related to cessation of aspirin or ticlopi-e at the time of noncardiac surgery. The authorsommended deferring noncardiac surgery after PCIil the adenosine 5-diphosphate (ADP)-receptor inhib-

is stopped. A similar concern about preoperativeI/stenting was raised by a recent report from Wilson et

describing the outcome of 207 patients undergoingcardiac surgery 2 months or less after coronary

ure 15. Perioperative (PERIOP) mortality rate after aorticcular surgery as a function of preoperative (Preop) manage-t. CABG, Coronary artery bypass grafting. P values reflectparisons to no testing group.

ure 16. Perioperative (PERIOP) mortality rate after infrain-al vascular surgery as a function of preoperative (Preop)agement. P values reflect comparisons to no testing group.
G, Coronary artery bypass grafting; ns, not significant. P v
ting. Death or MI occurred in 7 patients, all in thepatients undergoing noncardiac within 6 weeks of

I (event rate � 4%). These studies suggest thatcardiac surgery should be postponed for at least 6ks after coronary stenting. Conversely, if noncardiac

gery is more urgent and a high risk of ischemicdiac events is identified, options other than PCI/ting should be considered, including intensivelockade, bypass surgery, or possibly non-stentingA.Impact of preoperative testing on out-e. The above discussion provides compelling evi-

ce that management strategies can reduce cardiac riskatients undergoing noncardiac surgery. Such strate-

s depend on identification of high-risk patients whomost likely to benefit from medical and revascular-

ure 17. One-year mortality rate after aortic vascular surgerya function of preoperative (Preop) management. CABG,onary artery bypass grafting. P values reflect comparisonso testing group.

ure 18. One-year mortality rate after infrainguinal vascularery as a function of preoperative (Preop) management.G, Coronary artery bypass grafting. Data for PTCA pa-

ts are not presented because of the very small sample size.
alues reflect comparisons to no testing group.

izatpredat

Meor1-yofimaperinsursur29%surPTCthetherevoperedimaurethePreing(Fighorhigfromwiteraparintetienup,patlowthebetimatestnotwitemlariculthaoveof simpoutpro

Sum

clintoideamrevativandpre

suppernonneestenIt wulaandbasneo

expthacatiactioutcloknomyto sprealre

Ac

inte

Ref

1.

2.

3.


ion interventions. Is there evidence that noninvasiveoperative testing can lead to a better outcome? Recenta suggest that it can.

Fleisher et al66 examined a 5% sample of thedicare population (6895 patients) undergoing aorticinfrainguinal vascular surgery. Perioperative andear mortality rates were then examined as a functionno preoperative testing versus preoperative stressging, bypass surgery, or PTCA. They found thatioperative and 1-year mortality rates were substantialboth groups (7% and 11%, respectively, for aorticgery; 6% and 16%, respectively, for infrainguinalgery). Preoperative stress imaging was performed in

of aortic surgery patients and 12% of infrainguinalgery patients. This was followed by bypass surgery or

A in 27% of patients undergoing stress imaging inaortic cohort and 29% undergoing stress imaging ininfrainguinal cohort. No patient underwent coronary

ascularization without prior stress imaging. The peri-rative mortality rate of aortic surgery patients wasuced by more than 50% when preoperative stressging was performed compared with no testing (Fig-15). Important differences were seen with regard toimpact of revascularization between the two cohorts.operative bypass surgery or PTCA after stress imag-improved outcome in the aortic surgery patientsure 15). However, in the infrainguinal surgery co-

t, coronary revascularization was associated with ah perioperative mortality rate that was not different

the no-testing group (Figure 16). In the subgrouph stress imaging but no revascularization, the periop-tive mortality rate was significantly reduced com-ed with the no-testing group, presumably reflectingnsification of perioperative medical therapy in pa-ts identified as being at high risk. At 1-year follow-the outcome in the aortic cohort remained better in

ients undergoing preoperative stress imaging and wasest in the revascularized subgroups (Figure 17). Ininfrainguinal cohort the 1-year outcome remained

ter in the subgroup undergoing preoperative stressging without revascularization compared with noing (Figure 18). However, the 1-year outcome wasdifferent in the revascularized patients compared

h the no-testing group. Importantly, these dataphasize the potential hazards of coronary revascu-zation in patients scheduled for infrainguinal vas-ar surgery in contrast to aortic surgery and suggestt the former cohort is a higher-risk populationrall. Overall, however, this study supports the usetress imaging before vascular surgery and providesortant evidence that perioperative and long-term

come can be improved by the information
vided.
mary

Preoperative stress MPI can be used in concert withical data to identify low-risk patients who can go onundergo planned noncardiac surgery. It can alsontify high-risk patients, providing an opportunity toeliorate this risk. Studies suggest that medical andascularization interventions can reduce the perioper-e risk in patients identified as high risk by stress MPIthat early and late outcome is improved when such

operative testing is used.

FUTURE DIRECTIONS

It will be important to further develop the dataporting the impact of preoperative stress MPI onioperative and long-term outcome, particularly invascular surgery. In addition, more information isded on the risks of preoperative coronary arteryting as it relates to the timing of noncardiac surgery.ill also be helpful to examine the impact of ventric-

r function on outcome and the interaction of functionischemia on prognosis, as the advent of technetium-

ed imaging agents has made acquisition of simulta-us function and perfusion data routine.There are also new developing technologies that are

loring the identification of the “vulnerable plaque”67

t may have implications for preoperative risk stratifi-on. It may be hypothesized that identification ofve, inflammatory plaques may strongly influencecome, especially in the perioperative state, wheretting activation is intensified. In combination withwn data on the prognostic value of jeopardized viableocardium, such new imaging techniques may be ableubstantially improve the positive predictive value ofoperative noninvasive testing and further advance theady well-established clinical value of this approach.

knowledgment

The author has indicated he has no financial conflicts ofrest.

erences

Massie BM, Mangano DT. Risk stratification for noncardiacsurgery. How (and why)? Circulation 1993;87:1752-5.Abraham SA, Coles NA, Coley CM, et al. Coronary risk ofnoncardiac surgery. Prog Cardiovasc Dis 1991;34:205-34.Filipovic M, Jeger R, Probst C, et al. Heart rate variability andcardiac troponin I are incremental and independent predictors ofone-year all-cause mortality after major noncardiac surgery inpatients at risk of coronary artery disease. J Am Coll Cardiol
2003;42:1767-76.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.


Yeager RA, Moneta GL, Edwards JM, et al. Reducing periopera-tive myocardial infarction following vascular surgery. Arch Surg1995;130:869-73.Hertzer NR, Beven EG, Young JR, et al. Coronary artery diseasein peripheral vascular patients: a classification of 1000 coronaryangiograms and results of surgical management. Ann Surg 1984;199:223-33.L’Italian GJ, Cambria RP, Cutler BS, et al. Comparative early andlate cardiac morbidity among patients requiring different vascularsurgery procedures. J Vasc Surg 1995;21:935-44.Norman PE, Semmens JB, Lawrence-Brown MM. Long-termrelative survival following surgery for abdominal aortic aneurysm:a review. Cardiovasc Surg 2001;9:219-24.Calvin JE, Kieser TM, Walley VM, et al. Cardiac mortality andmorbidity after vascular surgery. Can J Surg 1986;29:93-7.Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial indexof cardiac risk in noncardiac surgical procedures. N Engl J Med1977;297:845-50.Detsky AS, Abrams HB, McLaughlin JR. Predicting cardiaccomplications in patients undergoing noncardiac surgery. J GenIntern Med 1986;1:211-9.Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia insurgical mortality. JAMA 1961;178:261-6.Eagle KA, Singer DE, Brewster DC, et al. Dipyridamole-thalliumscanning in patients undergoing vascular surgery. JAMA 1987;257:2185-9.Yeager RA, Weigel RM, Murphy ES, et al. Application ofclinically valid cardiac risk factors to aortic aneurysm surgery.Arch Surg 1986;121:278-81.Eagle KA, Coley CM, Newell JB, et al. Combining clinical andthallium data optimizes preoperative assessment of cardiac riskbefore major vascular surgery. Ann Intern Med 1989;110:859-66.Jeffrey CC, Kunsman J, Cullen DJ, Brewster DC. A prospectiveevaluation of cardiac risk index. Anesthesiology 1983;58:462-4.Lette J, Waters D, Bernier H, et al. Preoperative and long-termcardiac risk assessment. Ann Surg 1992;216:192-204.Lee TH, Marcantonio ER, Mangione CM, et al. Derivation andprospective validation of a simple index for prediction of cardiacrisk of major noncardiac surgery. Circulation 1999;100:1043-9.Boucher CA, Brewster DC, Darling RC, et al. Determination ofcardiac risk by dipyridamole-thallium imaging before peripheralvascular surgery. N Engl J Med 1985;312:389-94.Cutler BS, Leppo JA. Dipyridamole thallium 201 scintigraphy todetect coronary artery disease before abdominal aortic surgery. JVasc Surg 1987;5:91-100.Fletcher JP, Antico VF, Gruenewald S, et al. Dipyridamole-thallium scan for screening of coronary artery disease prior tovascular surgery. J Cardiovasc Surg (Torino) 1988;29:666-9.Sachs RN, Tellier P, Larmignat P, et al. Assessment by dipyrid-amole-thallium-201 myocardial scintigraphy of coronary risk be-fore peripheral vascular surgery. Surgery 1988;103:584-7.Eagle KA, Coley CM, Newell JB, et al. Combining clinical andthallium data optimizes preoperative assessment of cardiac riskbefore major vascular surgery. Ann Intern Med 1989;110:859-66.Younis LT, Aguirre F, Byers S, et al. Perioperative and long-termprognostic value of intravenous dipyridamole thallium scintigra-phy in patients with peripheral vascular disease. Am Heart J1990;119:1287-92.Marshall ES, Raichlen JS, Forman S, et al. Adenosine radionuclideperfusion imaging in the preoperative evaluation of patientsundergoing peripheral vascular surgery. Am J Cardiol 1995;76:817-21.Mangano DT, London MJ, Tubau JF, Browner WS, Hollenberg M,
Krupski W, Layug EL, Massie B. Dipyridamole thallium-201
scintigraphy as a preoperative screening test: a reexamination of itspredictive potential. Circulation 1991;84:493-502.Lette J, Waters D, Champagne P, et al. Prognostic implications ofa negative dipyridamole-thallium scan: results in 360 patients.Am J Med 1992;92:615-20.Hendel RC, Whitfield SS, Villegas BJ, et al. Prediction of latecardiac events by dipyridamole thallium imaging in patientsundergoing elective vascular surgery. Am J Cardiol 1992;70:1243-9.Kresowik TF, Bower TR, Garner SA, et al. Dipyridamole thalliumimaging in patients being considered for vascular procedures. ArchSurg 1993;128:299-302.Baron J-F, Mundler O, Bertrand M, et al. Dipyridamole-thallium scintigraphy and gated radionuclide angiography toassess cardiac risk before abdominal aortic surgery. N EnglJ Med 1994;330:663-9.Bry JDL, Belkin M, O’Donnell TF Jr, et al. An assessment of thepositive predictive value and cost-effectiveness of dipyridamolemyocardial scintigraphy in patients undergoing vascular surgery. JVasc Surg 1994;19:112-21.Koutelou MG, Asimacopoulos PJ, Mahmarian JJ, et al. Preopera-tive risk stratification by adenosine thallium 201 single-photonemission computed tomography in patients undergoing vascularsurgery. J Nucl Med 1995;2:389-94.Massie MT, Rohrer MJ, Leppo JA, et al. Is coronary angiographynecessary for vascular surgery patients who have positive results ofdipyridamole thallium scans? J Vasc Surg 1997;25:975-82.Coley CM, Field TS, Abraham SA, et al. Usefulness of dipyri-damole-thallium scanning for preoperative evaluation of cardiacrisk for nonvascular surgery. Am J Cardiol 1992;69:1280-5.Shaw L, Miller DD, Kong BA, et al. Determination of periopera-tive cardiac risk by adenosine thallium-201 myocardial imaging.Am Heart J 1992;124:861-9.Brown KA, Rowen M. Extent of jeopardized viable myocardiumdetermined by myocardial perfusion imaging best predicts periop-erative cardiac events in patients undergoing noncardiac surgery.J Am Coll Cardiol 1993;21:325-30.Younis I, Stratmann H, Takase B, et al. Preoperative clinicalassessment and dipyridamole thallium-201 scintigraphy for predic-tion and prevention of cardiac events in patients having majornoncardiovascular surgery and known or suspected coronary arterydisease. Am J Cardiol 1994;74:311-7.Stratmann HG, Younis LT, Wittry MD, et al. Dipyridamoletechnetium 99m sestamibi myocardial tomography for preopera-tive cardiac risk stratification before major or minor nonvascularsurgery. Am Heart J 1996;132:536-41.Van Damme H, Pierard L, Gillain D, et al. Cardiac risk assessmentbefore vascular surgery: a prospective study comparing clinicalevaluation, dobutamine stress echocardiography, and dobutamineTc-99m sestamibi tomoscintigraphy. Cardiovasc Surg 1997;5:54-64.Brown KA. Prognostic value of thallium 201 myocardial perfusionimaging: a diagnostic tool comes of age. Circulation 1991;83:363-81.Brown KA. Prognostic value of myocardial perfusion imagingstate of the art and new developments. J Nucl Cardiol 1996;3(6 Pt1):516-37.Leppo J, Plaja J, Gionet M, et al. Noninvasive evaluation of cardiacrisk before elective vascular surgery. J Am Coll Cardiol 1987;9:269-76.Shaw LJ, Eagle KA, Gersh BJ, et al. Meta-analysis of intravenousdipyridamole-thallium-201 imaging (1985 to 1994) and dobut-amine echocardiography (1991 to 1994) for risk stratification
before vascular surgery. J Am Coll Cardiol 1996;27:787-98.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.


Lette J, Waters D, Cerino M, et al. Preoperative coronary arterydisease risk stratification based on dipyridamole imaging and asimple three-step, three-segment model for patients undergoingnoncardiac vascular surgery or major general surgery. Am JCardiol 1992;69:1553-8.L’Italien GJ, Paul SD, Hendel RC, et al. Development andvalidation of a Bayesian model for perioperative cardiac riskassessment in a cohort of 1,081 vascular surgical candidates. J AmColl Cardiol 1996;27:779-86.Hendel RC, Leppo JA. The value of perioperative clinical indexesand dipyridamole thallium scintigraphy for the prediction ofmyocardial infarction and cardiac death in patients undergoingvascular surgery. J Nucl Cardiol 1995;2:18-25.Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guidelineupdate for perioperative cardiovascular evaluation for noncardiacsurgery-executive summary: a report of the American College ofCardiology/American Heart Association Task Force on PracticeGuidelines (Committee to Update the 1996 Guidelines on Periop-erative Cardiovascular Evaluation for Noncardiac Surgery). J AmColl Cardiol 2002;39:542-53.Bartels C, Bechtel JFM, Hossmann V, et al. Cardiac risk stratifi-cation for high-risk vascular surgery. Circulation 1997;95:2473-5.Froehlich JB, Karavite D, Russman PL, et al. American College ofCardiology/American Heart Association preoperative assessmentguidelines reduce resource utilization before aortic surgery. J VascSurg 2002;36:758-63.Hendel RC, Whitfield SS, Billegas BJ, et al. Prediction of latecardiac events by dipyridamole thallium imaging in patientsundergoing elective vascular surgery. Am J Cardiol 1992;70:1243-9.Younis LT, Aguirre F, Byers S, et al. Perioperative and long-termprognostic value of intravenous dipyridamole thallium scintigra-phy in patients with peripheral vascular disease. Am Heart J1990;119:1287-92.Urbinati S, DiPasquale G, Andreoli A, et al. Frequency andprognostic significance of silent coronary artery disease in patientswith cerebral ischemia undergoing carotid endarterectomy. Am JCardiol 1992;69:1166-70.Pasternack PF, Imparato AM, Baumann FG, et al. The hemody-namics of beta-blockade in patients undergoing abdominal aorticaneurysm repair. Circulation 1987;76:III1-III7.Stone JG, Foex P, Sear JW, et al. Myocardial ischemia in untreatedhypertensive patients: effect of a single small oral dose of abeta-adrenergic blocking agent. Anesthesiology 1988;68:495-500.Mangano DT, Layug EL, Wallace A, Tateo I. Effect of atenolol onmortality and cardiovascular morbidity after noncardiac surgery.The Multicenter Study of Perioperative ischemia Research Group.N Engl J Med 1996;335:1713-20 [published erratum appears in
N Engl J Med 1997;336:1039].
Froehlich JB, L’Italien G, Paul S, Brown KA, Eagle KA. Improvedcardiovascular mortality with perioperative �-blockade in patientsundergoing major vascular surgery [abstract]. J Am Coll Cardiol1997;29:219A.Polderman D, Boersma E, Bax JJ, et al. The effect of bisoprolol onperioperative mortality and myocardial infarction in high-riskpatients undergoing vascular surgery. Dutch EchocardiographicCardiac Risk Evaluation Applying Stress Echocardiography StudyGroup. N Engl J Med 1999;341:1789-94.Golden MA, Whittemore AD, Donaldson MC, Mannick JA.Selective evaluation and management of coronary artery disease inpatients undergoing repair of abdominal aortic aneurysms. AmSurg 1990;212:415-20.Elliott BM, Robison JG, Zellner JL, Hendrix GH. Dobutamine-201T1 imaging assessing cardiac risks associated with vascularsurgery. Circulation 1991;84(Suppl III):III-54-III-60.Younis L, Stratmann H, Takas B, et al. Preoperative clinicalassessment and dipyridamole thallium 201 scintigraphy for predic-tion and prevention of cardiac events in patients having majornoncardiovascular surgery and know or suspected coronary arterydisease. Am J Cardiol 1994;74:311-7.Eagle KA, Rihal CS, Mickel MC, et al. Cardiac risk of noncardiacsurgery influence of coronary disease and type of surgery in 3368operations. CASS Investigators and University of Michigan HeartCare Program. Coronary Artery Surgery Study. Circulation 1997;96:1882-7.Landesberg G, Mosseri M, Wolf YG, et al. Preoperative thalliumscanning, selective coronary revascularization, and long-term sur-vival after major vascular surgery. Circulation 2003;108:177-83.Gottlieb A, Banoub M, Sprung J, et al. Perioperative cardiovascu-lar morbidity in patients with coronary artery disease undergoingvascular surgery after percutaneous transluminal coronary angio-plasty. J Cardiothorac Vasc Anesth 1998;12:501-6.Posner KL, Van Norman GA, Chan V. Adverse cardiac outcomesafter noncardiac surgery in patients with prior percutaneous trans-luminal coronary angioplasty. Anesth Analg 1999;89:553-60.Kaluza GL, Joseph J, Lee JR, et al. Catastrophic outcomes ofnoncardiac surgery soon after coronary stenting. J Am Coll Cardiol2000;35:1288-94.Wilson SH, Fasseas P, Orford JL, et al. Clinical outcome ofpatients undergoing non-cardiac surgery in the two months follow-ing coronary stenting. J Am Coll Cardiol 2003;42:234-40.Fleisher LA, Eagle KA, Shaffer T, et al. Perioperative andlong-term mortality rates after major vascular surgery: the rela-tionship to preoperative testing in the Medicare population. AnesthAnalg 1999;89:849-55.Lake Tahoe Invitation Meeting 2002 [published erratum appears in
J Nucl Cardiol 2003;10:342]. J Nucl Cardiol 2003;10:223–57.

Documents

Advances in nuclear cardiology: Preoperative risk ... · Advances in nuclear cardiology: Preoperative risk stratiﬁcation Kenneth A. Brown, MD, FACC The determination of cardiac