Curriculum in Cardiology

Appropriate Cardiac Cath Lab activation: Optimizingelectrocardiogram interpretation and clinical decision-making for acute ST-elevation myocardial infarctionIvanC. Rokos,MD,a William J. French,MD,b AmalMattu,MD,c GrahamNichol,MD,dMichael E. Farkouh,MD,MSc,e

James Reiffel, MD,f and Gregg W. Stone, MDf Los Angeles, CA; Baltimore, MD; Seattle, WA; Toronto, ON; andNew York, NY

During the last few decades, acute ST-elevation on an electrocardiogram (ECG) in the proper clinical context has been a reliablesurrogate marker of acute coronary occlusion requiring primary percutaneous coronary intervention (PPCI). In 2004, theAmerican College of Cardiology/American Heart Association ST-elevation myocardial infarction (STEMI) guidelines specifiedECG criteria that warrant immediate angiography in patients who are candidates for primary PPCI, but new findings haveemerged that suggest a reappraisal is warranted. Furthermore, as part of integrated and efficient STEMI systems, emergencydepartment and emergency medical services providers are now encouraged to routinely make the time-sensitive diagnosis ofSTEMI and promptly activate the cardiac catheterization laboratory (Cath Lab) team. Our primary objective is to provide apractical summary of updated ECG criteria for emergency coronary angiography with planned PPCI, thus allowing clinicians tomaximize the rate of appropriateCath Lab activation and minimize the rate of inappropriateCath Lab activation.We review theevidence for ECG interpretation strategies that either increase diagnostic specificity for “classic” STEMI and left bundle-branch block or improve diagnostic sensitivity in identifying 4 STEMI-equivalents: posterior MI, acute left main occlusion, deWinter ST/T-wave complex, and certain scenarios of resuscitated cardiac arrest. (Am Heart J 2010;160:995-1003.e8.)

The key trigger point for emergency cardiac catheter-ization laboratory (Cath Lab) activation is usually a singleelectrocardiogram (ECG) diagnostic of an acute ST--elevation myocardial infarction (STEMI), which instantlyreclassifies a patient with chest pain or other acutecardiac symptoms from “routine evaluation” status to“high-priority” STEMI procedure (Figure 1). The 2004American College of Cardiology/American Heart Associ-ation (ACC/AHA) guidelines1 specify ECG criteria thatwarrant immediate angiography in patients who arecandidates for primary percutaneous coronary interven-tion (PPCI). Although new findings related to the ECG

From the aUCLA-Olive View, Department of Emergency Medicine, Los Angeles, CA,bHarbor-UCLA, Division of Cardiology, Department of Medicine, Los Angeles, CA,cUniversity of Maryland, Department of Emergency Medicine, Baltimore, MD, dUniversityof Washington-Harborview Center for Pre-Hospital Emergency Care, Seattle, WA,eUniversity Health Network and Li Ka Shing Knowledge Institute, Toronto, ON, andfColumbia University Medical Center and the Cardiovascular, Research Foundation,New York, NY.Submitted March 29, 2010; accepted August 12, 2010.Reprint requests: Ivan C. Rokos, MD, FACEP, FAHA, (FACC), UCLA-Olive View MedicalCenter, Department of Emergency Medicine, North Annex, 14445 Olive View Drive,Sylmar, CA 91342-1495.E-mail: irokos@earthlink.net0002-8703/$ - see front matter© 2010, Mosby, Inc. All rights reserved.doi:10.1016/j.ahj.2010.08.011

diagnosis of STEMI have emerged since 2004, they havenot been reviewed by any of the more recent guide-lines.2-6 Hence, our primary objective is to provide apractical summary of updated criteria for emergencycoronary angiography with planned PPCI, thus allowingclinicians to maximize the rate of appropriate Cath Labactivation and minimize the rate of inappropriate CathLab activation.Three background concepts are important. First, ECG

analysis is a fundamental clinical skill that is usedroutinely by a broad range of clinicians, but high-riskECG findings consistent with acute ischemia continue tobe overlooked.7 Second, a common “efficiency chal-lenge” involves the need to quickly differentiate and treatthe small cohort of acute STEMI patients from the muchlarger group of undifferentiated “chest pain” patients(Figure 1). This must be balanced against the potential forpoor resource utilization, especially with the currentemphasis on early Cath Lab activation by either emer-gency department (ED) or emergency medical servicesproviders.8,9 Third, a coordinated systems-based ap-proach is currently emphasized by the ACC/AHA STEMIguidelines,2,3 the ACC Door-2-Balloon (D2B) Alliance,10

and the AHA Mission: Lifeline initiative.11 Proposedefforts within Mission:Lifeline12 to comprehensivelytrack all Cath Lab activations and improve overall STEMIsystem efficiency depend on the existence of clearly

Figure 1

STEMI Process #1 empowers frontline providers (ED triage nurses and paramedics) to promptly perform an ECG on all suspected cardiacpatients. The STEMI Diagnosis by 12-lead electrocardiography is the key trigger point for a series of events designed to rapidly restore blood flowin an acutely occluded coronary. The STEMI Process #2 includes all the treatments provided in the time interval from STEMI diagnosis to patientarrival in the Cath Lab. The Cath Lab STEMI Procedure begins with emergency coronary angiography, followed by PPCI and other invasiveprocedures as indicated. STEMI Process #3 represents active participation in a quality improvement registry for the entire system.

996 Rokos et alAmerican Heart Journal

December 2010

defined criteria that distinguish appropriate versusinappropriate Cath Lab activation.To maximize appropriate Cath Lab activations, we

critically review and summarize (Tables I and II) real-timeECG interpretation strategies that accurately identify bothSTEMI and STEMI-equivalents. Criteria defining the termSTEMI-equivalent have been gradually standardized byboth the National Cardiovascular Data Registry (NCDR)data dictionaries and international consensus,13 andbroadly include any ECG pattern potentially associatedwith an acute coronary occlusion but lacking “classic”ST-elevation. Efficient STEMI systems of care need tomaximize diagnostic specificity for both classic STEMIECG patterns and new (or presumed new) left bundle-branch block (ie, a STEMI-equivalent), whereas diagnos-tic sensitivity should be improved in identifying 4additional STEMI-equivalents: posterior MI, acute leftmain occlusion, de Winter ST/T-wave complex, andcertain scenarios of resuscitated cardiac arrest. Moreover,we propose that clinicians conceptually prioritize 3 ques-tions when analyzing any ECG: (1) Are the ST-segmentsconsistent with acute STEMI or STEMI-equivalent? (2) Haschronic ST-segment elevation that does not representacute myocardial infarction been excluded? (3) Is thepatient a reasonable candidate for reperfusion therapy(PPCI or fibrinolytics)?

Definitions and background conceptsAcute myocardial infarction can be defined from

various clinical perspectives: electrocardiography, bio-markers, angiography, imaging, and pathology.14 How-ever, the 2004 guidelines1 emphasize the central role ofthe ECG in decision-making during the acute phase anddefine “classic STEMI” as ≥1 mm ST-elevation in 2adjacent leads (class I-A recommendation). Acute andpersistent ST-elevation, meeting these criteria in eitherleads V1 through V4 or leads II/III/aVF, represents the

most common (≈80%) ECG finding that is customarilyreported, respectively, as an infarction of the anterior orinferior region of the heart in registries, clinical trials, andthe International Classification of Diseases, Ninth

Revision, billing codes.As specified in a recent expert consensus document,15

actual ECG leads should not be simply referred to as“anterior” or “inferior” because this de-emphasizes theimportant bioelectric principle that all ECG leads arebipolar. Thus, an acute coronary occlusion can manifeston any of the 12 individual leads of a standard ECG witheither ST-elevation at the positive pole or ST-depression atthe negative pole. For example, leads V1 through V3mightbe designated most accurately as “anteroseptal-postero-lateral” bipolar leads. This provides the rationale fordesignating certain types of ST-depression ECGpatterns astrue STEMI-equivalents (eg, posterior MI, acute left mainocclusion, de Winter ST/T-wave complex).Because current guidelines1 state that timely PPCI is the

preferred reperfusion modality, our review has principal-ly focused upon appropriate Cath Lab activation criteria.However, in many STEMI-care settings, prompt adminis-tration of fibrinolytics represents the only realisticreperfusion option for an acute coronary occlusion perthe guidelines.1 Fortunately, initial ECG interpretationstrategies for detecting acute STEMI or STEMI-equivalentare similar for treatment pathways involving either PPCIor fibrinolysis, but safe fibrinolytic administration alsorequires review of a standardized ContraindicationsChecklist.1

Maximizing diagnostic specificity foracute coronary occlusionClassic STEMIIn the proper clinical context, acute ST-elevation

≥1 mm in 2 contiguous ECG leads represents a reliablesurrogate marker for an acute coronary occlusion

Table I. Comparison of 2004 ACC/AHA guidelines and authors' proposed update for ECG criteria that enhance the rate of appropriate CathLab activation for acute MI

Indications forappropriate CathLab activation

Diagnostic criteria forpatients with

symptoms <12 h

2004 ACC/AHAguideline

recommendationProposed update vs.ACC/AHA guidelines Comment

Classic STEMIAnterior ST-elevation ≥1 mm in 2

contiguous leads V1-V4

Class I-A Agree ST-elevation ≥2 mm (menand ≥1.5 mm (women)improves diagnosticspecificity.15

Presence of reciprocalchanges (ST-depression inopposite leads) improvesdiagnostic specificity.

Inferior ST-elevation ≥1 mm in 2contiguous leads(II, III, or AVF)

Class I-A Agree Presence of reciprocalchanges improves diagnosticspecificity.

Lateral ST-elevation ≥1 mm in 2contiguous leads(I, AVL, V5, or V6)

Class I-A Agree As above.

STEMI-equivalentsNew or presumednew-onset LBBB

“Presumed new” LBBBassumed when priorECG unavailable”“New” LBBB when priorECG available

Class I-A Proposed demotion in futureACC/AHA guidelines

Unless clinically unstable,most LBBB should beevaluated with biomarkersand non-emergentangiography if indicated.An “old” ECG without LBBBdoes not necessarily confirmthat the “new LBBB” is acute.

Preexisting LBBB withSgarbossa concordance

Concordance noted betweenQRS complex and ST/T-wavecomplex, with ST elevation≥1 mm in ≥1 lead

None Proposed addition to futureACC/AHA guidelines

Use of these decision criteriaprovides N95% specificityand avoids the need to find aprior ECG for comparison.Discordant ST-elevation≥ 5 mm is also a Sgarbossacriteria, but some studiesfound it a weak predictor.

Posterior MI (isolated) ST-depression ≥0.5 mm inleads V1-V3

Associated T-waves are eitherupright or inverted.Appearance of tall R-wavesin V1-V2 may be delayed.

Fibrinolytics: class IIa-CPrimary PCI: class I-A implied

Proposed clarification infuture ACC/AHA guidelines

Recent data34 demonstratedthat most posterior MIs arecurrently evaluated withurgent (rather thanemergent) angiography, buthis delay is associated withworse clinical outcomes.

Left Main coronaryocclusion

ST-depression ≥ 1 mm in6 or more leadsLeadaVRwithST-elevation≥1mmST-elevation in lead aVR ≥V1

None Proposed addition to futureACC/AHA guidelines

Most relevant in any ECGwith diffuse ST-depression≥1 mm that does not meeclassic STEMI criteria, thusproviding a subtle clue thatemergency angiographymay be warranted

de Winter ST/T-wavecomplex

ST depression ≥1 mmup-sloping at the J-point inleads V1-V6

None Proposed addition to futureACC/AHA guidelines

Tall T waves and up-slopingST depression are persistentnot transient.

Precordial T waves are tall,upright, symmetricNormal QRS duration

Associated with proximalLAD occlusion

Hyper-acute T-waves Tall peaked T wavesimmediately followingsymptom onset mayrepresent acute ischemia,but clinical studiesare lacking.

None Potential addition to futureACC/AHA guidelines

Generally prudent to performserial ECGs, because trueHATW generally morphquickly into a classic STEMIpattern13

Hyperkalemia is anothercommon cause of tall T waves

Rokos et al 997American Heart JournalVolume 160, Number 6

)

t

t

,

Table II. Three scenarios for appropriate Cath Lab activation involving out-of-hospital cardiac arrest (OHCA)

First event Second event Action Supplemental action

Acute STEMI diagnosed byprehospital ECG

Cardiac arrest witnessed by EMSproviders, followed by promptdefibrillation and ROSC

Cath Lab team at STEMI receiving center shouldhave already been activated based on originaldiagnosis of STEMI or STEMI-equivalent

Initiate hypothermia protocolin postresuscitation patientswho remain unconscious

Bystander witnessed OHCA,early activation of 9-1-1,and early chestcompressions

Shockable rhythm and ROSCachieved, followed by ECGdiagnostic of acute STEMI orSTEMI-equivalent

Activate Cath Lab team at cardiacresuscitation center

Same as above

Same as above Shockable rhythm and ROSCachieved, but postresuscitationECG is not diagnostic of acuteSTEMI or STEMI-equivalent

Consider activation of Cath Lab team at cardiacresuscitation center in patients whose clinicalcircumstance suggests acute ischemia and whoare candidates for aggressive intervention

Same as above

EMS, Emergency medical services.

Table III. Classification of appropriate versus inappropriate CathLab activation

Appropriate Cath Lab Activation ⇒ Reasonable

Appropriate Cath Lab Activation ⇒ Ideal

Inappropriate Cath Lab Activation ⇒ Goal is <5% rate•No Angiography performed (Cath Lab activation cancelled by a physician)

•Angiography without a PPCI target-lesion identified and normal cardiac markers:–“Avoidable angiogram” based upon erroneous ECG interpretation

•Advanced co-morbidities: Patient is not a PPCI-candidate

•Angiography without PPCI performed:–Surgical revascularization indicated

–Coronary anatomy is not amenable to PPCI intervention (ie, Medical therapy)

–“Unavoidable angiogram” per index ECG and/or clinical scenario as documented by the real-time clinicians (eg, Tako-tsubo, myocarditis)

–No PPCI target-lesion identified but cardiac markers become elevated

•Before angiography, true STEMI per index ECG dies suddenly

•Angiography ± PPCI for ROSC following witnessed OHCA from a shockablerhythm. Some ROSC patients may deteriorate and die before angiography.

•Angiography and PPCI performed

Classification based on retrospective and multidisciplinary peer review of all indexclinical data. Green zone represents the ideal scenario, yellow zone events arereasonable, and red zone occurrences should be minimized (<5%).

998 Rokos et alAmerican Heart Journal

December 2010

requiring PPCI (Table I and online Appendix items I.a-d).However, ECG accuracy may be limited by multipleconditions: large individual variations in coronary anato-my, preexisting disease, collateral circulation, misplacedleads, and technical shortcomings in detecting electricalimpulses from certain regions of the heart.15,16

Inappropriate Cath Lab activations have a number ofnegative implications, including poor resource utiliza-tion, unnecessary costs, patient safety risk, increasedpatient anxiety, work-life-sleep issues that may adverselyaffect career longevity of Cath Lab staff (ie, “burn-out”),and potential community risk from entire Cath Lab teamsrushing (ie, driving fast) back to the hospital to treat anafter-hours STEMI. Initial reports of ED physicianperformance in diagnosing acute STEMI identified a 9%rate of false-positive Cath Lab activation (no culprit arteryand negative biomarkers).17 More recent surveillance of“unnecessary” Cath Lab activations (ie, cardiologist didnot perform emergency coronary angiography) by EDphysicians demonstrated a 5% rate from a single-centerexperience (n = 249 activations) in Virginia18 and a 6%rate across 14 hospitals (n = 2,213 activations) in a NorthCarolina STEMI system.19 In contrast, evaluation ofprehospital-ECG interpretation by paramedics (n = 646)from Los Angeles County, California, found a 20% rate ofinappropriate Cath Lab activation (ie, cardiologist didnot perform emergency coronary angiography), with anadditional 5% of suspected STEMI on PH-ECG that neverresulted in Cath Lab activation (although PH-ECGtransmission was rarely used).20 Similar analysis (n =529) from Orange County, California, found a 23% rateof inappropriate Cath Lab activation.21

As proposed previously,22 we believe a <5% rate ofinappropriate Cath Lab activation represents a reason-able STEMI systems goal. Table III provides our concep-tual framework summarizing appropriate versus

inappropriate Cath Lab activation scenarios, recognizingthat not all Cath Lab activations classified as appropriateactually result in PPCI. To help clinicians optimize

resource utilization in STEMI systems, 6 ECG-interpreta-tion strategies are reviewed below.First, the degree of ST-elevation is a continuous

variable that has been stratified (<1 or ≥1 mm) in theguidelines1 to yield a convenient binary diagnosis(STEMI absent vs. present, respectively). However,evidence exists for increased specificity when diagnosticcriteria involve ≥2 mm ST-elevation in certain precordialleads.6 In recognition of benign ST-elevation that isfrequently observed in healthy adults, the NCDR datadictionaries (Cath-PCI v4.3 and ACTION-GWTG v2.1)currently set the STEMI threshold in leads V2-V3 at 2 mmfor men and 1.5 mm for women. Similarly, a recentAHA/ACC statement15 in conjunction with the HeartRhythm Society listed the same criteria (except formen age <40, the threshold is N2.5 mm ST-elevation inleads V2-V3).

Table IV. Common STE-mimics

Narrow QRS ComplexSTE-mimics

Tall/Wide QRSComplex STE-mimics

Benign early repolarization (BER) Left ventricular hypertrophy (LVHPericarditis LBBBLeft ventricular aneurysm Paced rhythmNormal variant “male pattern” HyperkalemiaPreexcitation Brugada syndrome

A comprehensive list of all STE-mimics already exists.23

Rokos et al 999American Heart JournalVolume 160, Number 6

)

Second, clinician awareness regarding the existenceof ST-elevation mimics (STE-mimics) is essential andprovides the basis for distinguishing pathologic vs.nonpathologic ST-elevation.23 At the minimum, clinicianfamiliarity should exist broadly for 10 common condi-tions (Table IV and online Appendix items III.a-e) withchronic ST-elevation.Third, the presence of reciprocal changes on the ECG

(ie, ST-depression ≥0.5 mm in leads 180° opposite tothose meeting ST-elevation criteria) should exist for allECGs with acute STEMI based upon bioelectric princi-ples, but this reciprocal ST-depression might be of lessermagnitude or undetectable due to technical limitations ofthe standard 12-lead ECG.15,16 By definition, reciprocalST-depression should never occur in ECGs considerednormal or those with narrow-complex STE-mimics(except lead aVR with pericarditis). In a fibrinolytic erastudy,24 the presence of any reciprocal ST-depression in≥2 leads was associated with N95% diagnostic specificityand positive predictive value for acute MI. More recently,an AHA scientific statement15 recommended that com-puterized interpretation algorithms incorporate thelocation of ST-depression to predict the location of acoronary occlusion. Collectively, these data suggest thatclinicians should evaluate every ECG with suspectedSTEMI for reciprocal changes: if present, then appro-

priate Cath Lab activation is almost certain; if absent,the presence of a STE-mimic versus a true STEMIwithout reciprocal changes should be carefully butexpeditiously considered. Observation of reciprocalchanges may be also be helpful in resolving challenging“semi-STEMI” cases near the 1-mm threshold, where oneclinician may see 1.1 mm of ST-elevation but anothersees only 0.9 mm.Fourth, pathologic Q-waves may develop in leads

initially demonstrating ST-elevation, and hence theirpresence supports the diagnosis of true STEMI (buttheir absence does not exclude acute STEMI).1 Further-more, a recent APEX-AMI substudy25 analysis of 4,530patients with ≤6 hours of symptoms identified 2,514(56%) with a Q wave (defined as ≥40 milliseconds[1 mm] duration, excluding lead III) already present onthe baseline ECG. Even with contemporary PPCI-basedreperfusion, the cohort with Q waves at baseline had

worse outcomes (composite of death, heart failure, orshock at 90 days) after multivariate adjustment.Fifth, a commonly used ECG computer algorithm

demonstrated 90% specificity for identifying acuteSTEMI when evaluated in a clinical trial,26 thus providinga reasonable real-time diagnostic aide for cliniciansinvolved in time-pressured ECG interpretation. However,experience with the computer algorithm for prehospitalECGs as part of regional STEMI networks has demon-strated lower specificity and increased rates of inap-

propriate Cath Lab activation.9,20,21

Lastly, immediate Wireless transmission and PhysicianInterpretation (WiPI) of all prehospital-ECGs with sus-pected STEMI improves the rate of appropriate Cath Labactivation.27 By increasing the number of “critical eyes”that review an ECG in real-time (simultaneously withpatient transport), WiPI can provide an important“appropriateness filter” between a prehospital STEMIAlert and an in-hospital Code STEMI.9

Left bundle-branch blockLeft bundle-branch block (LBBB) represents a unique

challenge in the acute setting, because it can be cate-gorized as either a STE-mimic or a STEMI-equivalent. Asa STE-mimic, chronic LBBB patterns in asymptomaticpatients almost always have persistent ST-elevation≥1 mm in the anteroseptal and/or inferior leads.However, for decades, new (or presumed new) LBBBhas also been considered a STEMI-equivalent that requiresprompt therapy, with inclusion in guidelines, clinicaltrials, NCDR registries, and Joint Commission acute MIcore measures. Furthermore, both the ACC/AHA andEuropean guidelines1,5 strongly recommend (class I-A)immediate reperfusion, based predominately on old meta-analysis data showing reduced mortality in fibrinolytic-treated LBBB patients.28

Data have gradually accumulated suggesting thattraditional criteria using new (or presumed new) LBBBas a STEMI-equivalent results in low diagnostic specificityand high rates of inappropriate Cath Lab activation. Forexample, a fibrinolytic era study of 7,725 consecutivepatients with suspected cardiac ischemia presenting to asingle ED found only 182 (2.4%) patients with LBBB onECG, and only 24 (13%) of the 182 with LBBB had anacute myocardial infarction using positive biomarkersas the gold standard.29

In the modern era, angiographic assessment of “pre-sumed new LBBB” also raised concerns about treatingthese patients emergently as a true STEMI-equivalent.A recent single-center PPCI-focused study30 of 7,937“chest pain” patients in the ED identified 55 (0.7%)presumed new LBBB, 136 (1.7%) confirmed old LBBB,and 7,746 (97.6%) without LBBB. In the primary analysiscomparing the “new vs old vs no” LBBB cohorts, the rateof biomarker-confirmed acute MI was similarly low (7.3%,

1000 Rokos et alAmerican Heart Journal

December 2010

5.2%, 6.1%, respectively, P = .75). Thus, N90% of LBBBpatients evaluated in the ED do not have an acutecoronary occlusion nor require PPCI. Similarly, a recentanalysis from the PPCI-focused Minneapolis interhospitaltransfer network17 identified only 36 (2.6%) of 1335patients with “presumed new LBBB.”Moreover, 16 (44%)of 36 from this cohort did not have a culprit arteryvisualized during emergency coronary angiography, and13 (36%) of 36 had no elevation of serial biomarkers.The latest ESC guidelines5 acknowledge these more

recent findings by stating that biomarker results “maysometimes be helpful” in evaluating LBBB patients, thusimplying treatment of stable patients with an urgentrather than emergent PCI strategy. Importantly, clinicallyunstable LBBB patients or those with acutely abnormalanterior wall motion by echocardiography should stillreceive emergency angiography.Published ECG criteria (Table I and online Appendix

item II.e) exist that maximize diagnostic specificity fordetecting a true STEMI-equivalent in the setting ofchronic LBBB. A chronic LBBB normally has discor-

dance (the direction of the major component of theQRS complex is opposite that of ST segment/T wave inany lead), whereas QRS/ST concordance (with STelevation ≥1 mm in ≥1 lead) appears to be the mostrobust predictor of acute ischemia.6,15,31 A recent meta-analysis32 found only 20% sensitivity but 98% specificityfor the “concordance criteria,” but much of the data isfrom the fibrinolytic era when cardiac markers insteadof angiography were the gold standard.

Maximizing diagnostic sensitivity foracute coronary occlusionPosterior myocardial infarctionTrue posterior MI should be treated as a STEMI-

equivalent, in which case isolated ST-depression

≥0.5 mm in leads V1 through V3 represents the dominantfinding (Table I and online Appendix item II.a) on astandard 12-lead ECG.6,15 The use of additional posteriorchest wall leads (V7-V9 ≥0.5 mm) to detect ST elevationconsistent with posterior MI is often recommended,6,15

but their use remains uncommon in daily practice.Clinical trials commonly exclude posterior MI and theguidelines1 reflect this paucity of evidence with a classIIa-C recommendation for administering fibrinolytics inthis subset of patients. Paradoxically, the guidelines1

indirectly imply a class I-A recommendation for PPCI ofposterior MI.A recent AHA scientific statement15 critiqued the term

posterior MI (used in the 2004 ACC/AHA guidelines,clinical trials, NCDR registries, and International Classi-

fication of Diseases, Ninth Revision, codes) because theactual infarct occurs in the lateral wall of the left ventriclewhen assessed by advanced imaging techniques.

Although replacement terms like posterolateral MI wereconsidered, the expert statement ultimately recom-mended retaining the term posterior MI for ST-depression≥0.5mm in leads V2 through V3 (or ST-elevation≥0.5mmin leads V7-V9) until additional data become available.The TRITON-TIMI-38 study33 enrolled patients with

both STEMI and non-STEMI, and thus it provided a uniqueopportunity to evaluate contemporary posterior MItherapy. According to a recent substudy,34 isolatedST-depression in leads V1 through V3 on index ECG waspresent in 1,198 (8.8%) of the 13,608 enrolled patientsand retrospectively classified by the core lab as “posteriorMI” (26.2%), non-STEMI (53.5%), and unstable angina(20.3%) based upon cardiac markers and angiography.Despite guideline recommendations,1 the vast majority(1193 [99.6%]) of study patients with isolated ST-depressionin leads V1 through V3 were treated with non-emergent PCI(N6 hours after index ECG). However, the cohort of “slowlytreated” posterior MIs had an acutely occluded coronary(most commonly the left circumflex) and had a significantlyhigher rate of 30-day death or MI as compared with thoseclassified as non-STEMI.

Left main coronary occlusionAcute left main coronary occlusion (LMCO) generally

causes massive ischemia and is rapidly lethal, but a smallproportion of patients have enough perfusion from right-sided collaterals to arrive alive at the hospital. A recentCath-PCI Registry analysis35 spanning 2004 to 2008identified 434 (0.35% of all attempted PCIs in Cath-PCI)with acute STEMI from a left main culprit that was“unprotected” (ie, no prior coronary artery bypassgrafting to distal vessels). Interestingly, about one thirdof patients did not present in cardiogenic shock, eventhough angiography demonstrated that 85% of thesepatients had N90% stenosis and 55% had completeLMCO. Importantly, survival to hospital discharge was42% in this series.STEMI-equivalent ECG criteria (Table I and online

Appendix item II.b) exist for diagnosing the criticalcondition of acute LMCO, but these are not mentioned inany of the guidelines.1-3,5 In 2001, a small series reportedthat ≥0.5-mm ST-elevation in lead aVR is consistentwith acute LM occlusion, especially when the degree ofST-elevation in aVR is greater than lead V1 and inferiorST-depression is present.36 A combined analysis of 3studies with a total of 75 acute LM occlusions confirmedby angiography demonstrated N75% sensitivity andspecificity for these ECG findings.37

A 2009 AHA scientific statement15 recommended“when the resting ECG reveals ST-depression N1 mmin 8 or more surface leads coupled with ST-elevation inaVR and/or VI but is otherwise unremarkable, theautomated [computerized] interpretation should suggestischemia due to multi-vessel or left main coronary artery

Rokos et al 1001American Heart JournalVolume 160, Number 6

obstruction.” A 2010 international consensus document13

provided similar STEMI-equivalent criteria for acuteLMCO (or severe angiographic disease): a myocardialischemia vector causing ST-depression in 6 or more leads(maximal in V4-V6) and associated ST-elevation limitedto lead aVR and V1.

de Winter ST/T-wave complexIn 2008, de Winter et al38 described a novel STEMI-

equivalent ECG pattern that signifies acute occlusion ofthe proximal left anterior descending (LAD). Followingexamination of their Dutch registry, the “de Winter ST/T-wave” pattern was first recognized in 30 uniquepatients (all with normal baseline serum potassiumlevels) and represented 2% of all angiographicallyconfirmed anterior MIs. Diagnostic criteria on ECGinclude 1 to 3 mm of ST-depression that is up-slopingat the J-point in leads V1 through V6 and associatedwith persistently tall, upright, and symmetric precordialT-waves (Table I and online Appendix item II.c).The de Winter ST/T-wave complex is distinct from

transient hyper-acute T-waves (HATW) that may occurwithin minutes of an acute coronary occlusion (repre-senting another potential STEMI-equivalent ECG pattern)but generally morph quickly into a classic STEMIpattern.13,38 Transient HATW can be induced withexperimental occlusion of a coronary for a few minutes,but they have not been used as inclusion criteria forSTEMI clinical trials.In addition, de Winter ST/T-wave complex should not

be confused with Wellens syndrome,39 in which theECG demonstrates either biphasic or inverted T wavesin leads V2 through V3 (online Appendix item II.d).Wellens syndrome typically involves a chronic high-grade LAD stenosis that can usually be evaluated by non-emergent angiography.

Resuscitated cardiac arrestFor more than a decade, it has been known that

approximately 50% of patients with out-of-hospitalcardiac arrest (OHCA) and return of spontaneouscirculation (ROSC) have an acute coronary arteryocclusion when evaluated by immediate diagnosticangiography.40 Hence, in addition to the acute STEMIand STEMI-equivalent ECG criteria (Table I) reviewedthus far, certain scenarios of OHCA with ROSC appear tobe gaining acceptance as a trigger for appropriate CathLab activation (Table II). Many OHCA patients simplyrepresent a dramatic presentation of an acute coronaryocclusion, with sudden death as both the chief complaintand the clinical STEMI-equivalent. Neither the ACC/AHAguidelines nor the European guidelines1-3,5 comment onthe role of immediate diagnostic coronary angiographyfor OHCA patients, but both acknowledge that most of

the deaths in STEMI patients occur from lethal arrhyth-mias within 1 to 2 hours of symptom onset.Historically, the practice of taking resuscitated OHCA

patients for emergency coronary angiography andplanned PPCI was rare and usually restricted to a smallcohort of “ideal” patients: witnessed arrest, early chestcompressions, early defibrillation of a shockable rhythm(ventricular tachycardia or fibrillation), early ROSC, hemo-dynamically and electrically stable on arrival to thehospital, and clinically perceived on initial assessment tohave experienced only minimal neurologic injury fromischemia. Although randomized controlled trials of STEMIalmost uniformly exclude OHCA, observational registrydata has gradually accumulated and demonstrated goodoutcomes in resuscitated OHCA patients who traditionallyhave been considered “less than ideal” candidates for CathLab intervention. Thus, both the 2008 International Liai-son Committee on Resuscitation41 consensus statementand the 2009 state-of-the-art paper by Ewy and Kern42

recommended aggressive postresuscitation care involvingboth PPCI and invasive hemodynamic support for agreater proportion of patients resuscitated from OHCA.Despite the International Liaison Committee on Resus-

citation consensus statement, controversy still existsregarding the ability of a postresuscitation ECG to identifypatients with an acute coronary occlusion.43,44 However,the largest experience to date is the PROCAT45 registry,which provided compelling new evidence supporting aroutine Cath Lab based treatment pathway for OHCAregardless of ECG pattern after resuscitation. In thePROCAT registry spanning 2003 to 2008, all 435 resus-citated OHCA patients with no obvious extracardiaccause of arrest received emergency coronary angiogra-phy. Of the 134 patients with a post-ROSC ECGdemonstrating classic STEMI, 128 (96%) had at least onesignificant lesion (N50% reduction in luminal diameter)and 99 (74%) underwent successful PCI. In the remaining301 patients without classic STEMI on post-ROSC ECG,176 (58%) had a significant lesion and 78 (26%) receivedPCI. Multivariable analysis demonstrated successful PCIto be an independent predictor of survival (OR 2.1, 95%CI 1.2-3.7), irrespective of postresuscitation ECG pattern.As compared to a historically low rate of survival(<10%),41 overall survival to hospital discharge was 40%in the PROCAT registry. Moreover, there was a 94% rateof favorable neurologic outcomes in a cohort with a 68%rate of shockable first rhythm and 85% rate of in-hospitaltherapeutic hypothermia.A recent AHA policy statement proposed the develop-

ment of cardiac resuscitation centers46 in associationwith an existing network of STEMI receiving centers.11,22

The statement emphasized that optimal postresuscitationcare is complex, multi-disciplinary, and is probably mostefficiently provided in designated hospitals via a coordi-nated systems-based approach that is supported by arobust quality improvement infrastructure. Three different

1002 Rokos et alAmerican Heart Journal

December 2010

prehospital scenarios warrant consideration with regardto appropriate Cath Lab activation (Table II). Further-more, within each general scenario, real-world clinicaldecision-making likely involves assessment of the “downtime” before initiation of certain critical prehospitalinterventions (ie, time delay from patient collapse to9-1-1 call, to chest compressions, and to defibrillation)and each patient's clinical status upon hospital arrival.

Appropriate patient selectionThe final decision point before a Cath Lab activation

involves clinical correlation, and hence our deliberate useof the term appropriate Cath Lab activation. This broaderterm includes both traditional “false-positive” Cath Labactivation (ie, technical competency in ECG interpretationfor STEMI and STEMI-equivalents) as well as physicianjudgment regarding whether or not an individual patientis a “candidate” (class I-A recommendation1) for PPCI.Identification of STEMI and STEMI-equivalent patients forwhom Cath Lab activation is “inappropriate” requiresfrontline physicians to document a brief yet carefulassessment of the overall clinical scenario, age, comorbi-dities, functional status, do-not-resuscitate status, familyperspective, and patient preferences.New findings are also informing appropriate patient

selection strategies in patients resuscitated from suddencardiac arrest. Interest in early Cath Lab activation andPPCI-based myocardial reperfusion has traditionally beenovershadowed by significant concerns regarding mean-ingful neurologic recovery from the initial anoxic insult.However, data have gradually accumulated from bothclinical trials and observational registries supporting theability of therapeutic hypothermia to significantly improvethe rate of good cerebral recovery (number needed totreat = 6).41,46 Thus, baseline neurologic status immedi-ately after ROSC should not preclude early Cath Labactivation because current AHA expert statements41,46

emphasize deferral of cerebral recovery prognosticationfor at least 72 hours after collapse in patients treated withtherapeutic hypothermia. Furthermore, although the“best” hypothermia protocol has yet to be definitivelyproven, current consensus41,46 suggests initiating coolingas early as possible (ie, pre-arrival to the Cath Lab) usingsome approximation of a 4/24/8/@33 protocol (4-hourinduction [1°/h]) to a target temperature of 33°, 24-hourmaintenance at 33° with minimal temperature variation,and 8 hours of rewarming (0.5°/h) back to normothermia.

ConclusionIn summary, broad awareness should exist regarding

evidence-based triggers for appropriate Cath Lab activa-tion. A diverse group of frontline clinicians making thesetime-pressured decisions need a comprehensive list of pre-cise criteria, because not all “acute MIs” have classic ST-

elevation on ECG (eg, STEMI-equivalents and certainOCHAscenarios), not all ST-elevation patterns represent “trueSTEMI” (ie, STE-mimics), and some “true STEMI” patientsare not reasonable candidates for an aggressive treatmentstrategy involving PPCI. Optimal ECG interpretation pro-ficiency by all clinicians in identifying both classic STEMIand STEMI-equivalents constitutes a major cornerstone ofongoing efforts to maximize STEMI system efficiency.

DisclosuresNo external funding sources supported the writing of

this manuscript. Dr Stone reports research support fromBoston Scientific, Abbott Vascular, TherOx, and TheMedicines Company; consultant to St Jude Medical andRadiant; honoraria from Eli Lilly, Medtronic, and Glaxo-Smith-Kline. The remaining authors do not have financialor industry relationship disclosures relevant to this article.

References1. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines

for the management of patients with ST-elevation myocardialinfarction (committee to revise the 1999 guidelines). J Am CollCardiol 2004;44:671-719.

2. Antman EM, Hand M, Armstrong PW, et al. 2007 focused update ofthe ACC/AHA 2004 guidelines for the management of patients withST-elevation myocardial infarction. J Am Coll Cardiol 2008;51:210-47.

3. Kushner FG, Hand M, Smith SC, et al. Focused update: ACC/AHAguidelines for the management of patients with ST-elevationmyocardial infarction (updating 2004 guidelines and 2007 focusedupdate). JACC 2009;54:2205-41.

4. Patel MR, Dehmer GJ, Hirshfeld JW, et al. ACCF/SCAI/STS/AATS/AHA/ASNC 2009 appropriateness criteria for coronary revascu-larization. J Am Coll Cardiol 2009;53:530-53.

5. Van de Werf F, Bax J, Betriu A, et al. Management of acutemyocardial infarction in patients presenting with persistent ST-segment elevation: the Task Force on the Management of ST-SegmentElevation Acute Myocardial Infarction of the European Society ofCardiology. Eur Heart J 2008;29:2909-45.

6. Fesmire FM, Brady WJ, Hahn S, et al. ACEP clinical policy:indications for reperfusion therapy in emergency department patientswith suspected acute myocardial infarction. Ann Emerg Med 2006;48:358-83.

7. Masoudi FA, Magid DJ, Vinson DR, et al. Implications of the failure toidentify high-risk electrocardiogram findings for the quality of care ofpatients with acute myocardial infarction: results of the EmergencyDepartment Quality in Myocardial Infarction (EDQMI) study.Circulation 2006;114:1565-71.

8. Bradley EH, Herrin J, Wang Y, et al. Strategies for reducing the door-to-balloon time in acute myocardial infarction. N Engl J Med 2006;355:2308-20.

9. Rokos IC, French WJ, Koenig WJ, et al. Integration of pre-hospitalelectrocardiograms and ST-elevation myocardial infarction receivingcenter (SRC) networks: impact on Door-to-Balloon times across 10independent regions. JACC Cardiovasc Interv 2009;2:339-46.

10. Krumholz HM, Bradley EH, Nallamothu BK, et al. A campaign toimprove the timeliness of primary percutaneous coronary

Rokos et al 1003American Heart JournalVolume 160, Number 6

intervention: Door-to-Balloon: An Alliance for Quality. JACCCardiovasc Interv 2008;1:97-104.

11. Jacobs AK, Antman EM, Faxon DP, et al. Development of systems ofcare for ST-elevation myocardial infarction patients: executivesummary. Circulation 2007;116:217-30.

12. Peterson ED, Ohman EM, Brindis RG, et al. Development of systems ofcare for ST-elevation myocardial infarction patients: evaluation andoutcomes. Circulation 2007;116:e64-7.

13. Nikus K, Pahlm O, Wagner G, et al. Electrocardiographicclassification of acute coronary syndromes: a review by a committeeof the International Society for Holter and Non-Invasive Electro-cardiology. J Electrocardiol 2010;43:91-103.

14. Alpert JS, Thygesen K, Antman E, et al. Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition ofmyocardial infarction. J Am Coll Cardiol 2000;36:959-69.

15. Wagner GS, Macfarlane P, Wellens H, et al. AHA/ACCF/HRSscientific statement. Recommendations for the standardization andinterpretation of the electrocardiogram: part VI: acute ischemia/infarction. Circulation 2009;119:e262-70.

16. Zimetbaum PJ, Josephson ME. Use of the electrocardiogram in acutemyocardial infarction. N Engl J Med 2003;348:933-40.

17. Larson DM, Menssen KM. Sharkey SW, et al. “False-positive” cardiaccatheterization laboratory activation among patients with suspectedST-segment elevation myocardial infarction. JAMA 2007;298:2754-60.

18. Kontos MC, Kurz MC, Roberts CS, et al. An evaluation of theaccuracy of emergency physician activation of the cardiac catheter-ization laboratory for patients with suspected ST-segment elevationmyocardial infarction. Ann Emerg Med 2010;55:423-30.

19. Jollis JG, Garvey JL, Monk L, et al. False-positive cath lab activationrates in the RACE regional ST-elevation myocardial infarctionprogram (abstract 1101-286). JACC 2010;55:A109.

20. Rokos IC, Koenig WJ, Tadeo R, et al. Etiologies of inappropriatecardiac catheterization lab activations from pre-hospital electrocar-diograms in the Los Angeles County network (abstract 143).Circulation CVQO 2009;2(3 suppl):e33.

21. Swan PY, Nighswonger B, Boswell GL, et al. Factors associated withfalse-positive emergency medical services triage for percutaneouscoronary intervention. West J Emerg Med 2009;10:208-12.

22. Rokos IC, Larson DM, Henry TD, et al. Rationale for establishingregional ST-elevation myocardial infarction receiving center (SRC)networks. Am Heart J 2006;152:661-7.

23. Wang K, Asinger RW, Marriott HJ. ST-segment elevation inconditions other than acute myocardial infarction. N Engl J Med2003;349:2128-35.

24. Otto LA, Aufderheide TP. Evaluation of ST segment elevation criteriafor the prehospital electrocardiographic diagnosis of acute myocar-dial infarction. Ann Emerg Med 1994;23:17-24.

25. Armstrong PW, Fu Y, Westerhout CM, et al. Baseline Q-wave sur-passes time from symptom onset as a prognostic marker in ST-segmentelevation myocardial infarction patients treated with primary percu-taneous coronary intervention. J Am Coll Cardiol 2009;53:1503-9.

26. Kudenchuk PJ, Ho MT, Weaver WD, et al. Accuracy of computer-interpreted electrocardiography in selecting patients for thrombolytictherapy. MITI Project Investigators. J Am Coll Cardiol 1991;17:1486-91.

27. Ting HH, Krumholz HM, Bradley EH, et al. Implementation andintegration of prehospital ECGs into systems of care for acutecoronary syndrome: a scientific statement from the American HeartAssociation. Circulation 2008;118:1066-79.

28. Indications for fibrinolytic therapy in suspected acute myocardialinfarction: Fibrinolytic Therapy Trialists' (FTT) Group. Lancet 1994;343:311-22.

29. Kontos MC, McQueen RH, Jesse RL, et al. Can myocardial infarctionbe rapidly identified in emergency department patients who have leftbundle-branch block? Ann Emerg Med 2001;37:431-8.

30. Chang AM, Shofer FS, Tabas JA, et al. Lack of association betweenLBBB and AMI in symptomatic ED patients. Am J Emerg Med 2009;27:916-21.

31. Sgarbossa EB, Pinski SL, Barbagelata A, et al. Electrocardiographicdiagnosis of evolving acute MI in the presence of left bundle-branchblock. GUSTO-1 Investigators. N Engl J Med 1996;334:481-7.

32. Tabas JA, Rodriguez RM, Seligman HK, et al. Electrocardiographiccriteria for detecting acute myocardial infarction in patients with leftbundle branch block: a meta-analysis. Ann Emerg Med 2008;52:329-36.

33. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versusclopidogrel in patients with acute coronary syndromes. N Engl J Med2007;357:2001-15.

34. Pride YB, Tung P, Mohanavelu S, et al. Angiographic and clinicaloutcomes among patients with acute coronary syndromes presentingwith isolated anterior ST-segment depression: A TRITON-Timi 38sub-study. JACC Intv 2010;3:806-11.

35. Zoghbi GJ, Misra VK, Brott BC, et al. ST-elevation myocardialinfarction due to left main culprit lesions: PCI outcomes in cath-PCIregistry (abstract 2909-09). JACC 2010;55:A183.

36. Yamaji H, Iwasaki K, Kusachi S, et al. Prediction of acute left maincoronary artery obstruction by 12-lead electrocardiography. STsegment elevation in lead aVR with less ST segment elevation in lead V(1). J Am Coll Cardiol 2001;38:1348-54.

37. Rostoff P, Piwowarska W, Gackowski A, et al. Electrocardiographicprediction of acute left main coronary artery occlusion. Am J EmergMed 2007;25:852-5.

38. de Winter RJ, Verouden NJ, Wellens HJ, et al. A new ECG sign ofproximal LAD occlusion. N Engl J Med 2008;359:2071-3.

39. de Zwaan C, Bar FW, Wellens HJ. Characteristic electrocardio-graphic pattern indicating a critical stenosis high in left anteriordescending coronary artery. Am Heart J 1982;103:730-6.

40. Spaulding CM, Joly LM, Rosenberg A, et al. Immediate coronaryangiography in survivors of out-of-hospital cardiac arrest. N Engl JMed 1997;336:1629-33.

41. Neumar RW, Nolan JP, Adrie C, et al. Post-cardiac arrest syndrome:epidemiology, pathophysiology, treatment, and prognostication. Aconsensus statement from the AHA International Liaison Committeeon Resuscitation. Circulation 2008;118:2452-83.

42. Ewy GA, Kern KB. Recent advances in cardiopulmonary resuscitation:cardiocerebral resuscitation. J Am Coll Cardiol 2009;53:149-57.

43. Anyfantakis ZA, Baron G, Aubry P, et al. Acute coronaryangiographic findings in survivors of out-of-hospital cardiac arrest.Am Heart J 2009;157:312-8.

44. Muller D, Schnitzer L, Brandt J, et al. The accuracy of an out-of-hospital12-lead ECG for the detection of ST-elevation myocardial infarctionimmediately after resuscitation. Ann Emerg Med 2008;52:658-64.

45. Dumas F, Cariou A, Manzo-Silberman S, et al. Immediatepercutaneous coronary intervention is associated with better survivalafter out-of-hospital cardiac arrest: insights from the PROCAT(Parisian Region Out of hospital Cardiac ArresT) registry. CircCardiovasc Interv 2010;3:200-7.

46. Nichol G, Aufderheide TP, Eigel B, et al. Regional systems of care forout-of-hospital cardiac arrest: a policy statement from the AmericanHeart Association. Circulation 2010;121:709-29.

Rokos et al 1003.e1American Heart JournalVolume 160, Number 6

Appendix. Online Data Supplement

Title: Appropriate cardiac catheterization lab activation: optimizing electrocardiogram interpretation and clinicaldecision-making for acute ST-elevation myocardial infarctionAuthors: Rokos, French, Mattu, Nichol, Farkouh, Reiffel, and Stone.

14 Electrocardiogram (ECG) Examples

I. Classic STEMIa. Anterior-STEMI with reciprocal changes ⇒ proximal LAD occlusionb. Anterior-STEMI without reciprocal changes ⇒ middle LAD occlusionc. High-lateral STEMI ⇒ diagonal branch occlusiond. Lateral STEMI ⇒ left circumflex occlusion

II. STEMI-equivalentsa. Posterior MI (isolated)b. Left main coronary occlusion: lead aVR STE and infero-lateral STDc. de Winter ST/T-wave complexd. Wellens syndrome (distinguish from de Winter ST/T-wave complexe. LBBB with Sgarbossa criteria

III. ST-elevation mimics (STE-mimics)a. Benign early repolarization (BER)b. Acute pericarditisc. Left ventricular aneurysmd. Normal-variant chronic ST-elevatione. Brugada syndrome

1003.e2 Rokos et alAmerican Heart Journal

December 2010

I.a Anterior STEMI with reciprocal changes(IRA = proximal LAD)

I.b Anterior STEMI without reciprocal changes(IRA = middle LAD)

Rokos et al 1003.e3American Heart JournalVolume 160, Number 6

I.c “High” Lateral STEMI(IRA = Diagonal branch)

I.d Lateral STEMI(IRA = Left Circumflex)

1003.e4 Rokos et alAmerican Heart Journal

December 2010

II. STEMI-equivalents

II.a Posterior MI (Isolated)

II.b Left Main coronary occlusion (LMCO): Lead aVR with ST-elevation and associated infero-lateral ST-depression

Rokos et al 1003.e5American Heart JournalVolume 160, Number 6

II.c de Winter ST/T-wave complex

II.d Wellens syndrome

(This is not a STEMI-equivalent. Example provided to compare with leads V2-V3 of de Winter ST/T-wave complex)

1003.e6 Rokos et alAmerican Heart Journal

December 2010

II.e LBBB MI with Sgarbossa criteria concordance in V2 through V3

III. ST-elevation mimics (STE-mimics)

III.a Benign early repolarization (BER)

Rokos et al 1003.e7American Heart JournalVolume 160, Number 6

III.b Acute pericarditis

III.c Left Ventricular aneurysm

1003.e8 Rokos et alAmerican Heart Journal

December 2010

III.d Normal-variant chronic ST-elevation

III.e Brugada syndrome