Electrocardiographic PatternsMimicking ST SegmentElevation Myocardial Infarction

Peter Pollak, MDa, William Brady, MDb,*

KEYWORDS

� Electrocardiogram � ECG � ST segment elevation � STEMI � Mimic

KEY POINTS

� The 12-lead surface electrocardiogram (ECG) is inexpensive, portable, and transmittable; it remainsthe cornerstone of prompt diagnosis of and primary indication for the management of ST elevationmyocardial infarction (STEMI).

� Although the ECG is reasonably reliable, it remains an imperfect diagnostic tool. Some patients dopresent with classic symptoms and findings; however, approximately 60% to 80% of patients withST segment elevation on the presenting ECG are ultimately found to not be associated with STEMI.

� In certain difficult cases, a patient’s ECG can resemble STEMI yet manifest ST segment elevationfrom a non–acute coronary syndrome entity, the so-called STEMI mimics. In other situations, thepatient’s ECG makes it difficult or impossible to determine whether STEMI is present, the so-called STEMI confounders; these confounders to STEMI diagnosis are also mimickers of AMI.

� The ultimate goal with both the STEMI mimics and the confounders is to maximize rapid, accuratediagnosis while avoiding delays in treatment of alternative causes of ST segment elevation.

INTRODUCTION require the rapid mobilization of large teams of

m

Cardiovascular disease is the leading cause ofdeath worldwide. In fact, 7 million patients presentannually to emergency departments in the UnitedStates with symptoms concerning for myocardialischemia.1 Prompt reperfusion in the settingof ST segment elevation myocardial infarction(STEMI) can dramatically reduce the associatedmortality and morbidity. Unfortunately, the benefitof reperfusion therapies in acute myocardialinfarction (AMI) decays quickly over time. Improve-ments in care systems, such as the door-to-balloon initiative and the American HeartAssociation’s Mission Lifeline have altered thelandscape of STEMI management over the pastdecade, minimizing delays to reperfusion andsignificantly improving outcomes. These systems

a Division of Cardiovascular Medicine, Department of MeStreet, Charlottesville, VA 22908, USA; b Departments oVirginia School of Medicine, Lee Street, Charlottesville, V* Corresponding author.E-mail address: wb4z@virginia.edu

Cardiol Clin 30 (2012) 601–615http://dx.doi.org/10.1016/j.ccl.2012.07.0120733-8651/12/$ – see front matter � 2012 Elsevier Inc. All

practitioners and resources, involve not insignifi-cant risk to patients, and rely heavily on the clini-cian to quickly and accurately determine whetheran electrocardiographic finding represents acuteclosure of a coronary artery and related STEMI.The 12-lead surface electrocardiogram (ECG) isinexpensive, portable, and transmittable; itremains the cornerstone of the prompt diagnosisof and the primary indication for management ofSTEMI. Although reasonably reliable, the ECGremains an imperfect diagnostic tool. Somepatients do present with classic symptoms andfindings; however, approximately 60% to 80% ofpatients with ST segment elevation on the present-ing ECG ultimately are found to not be associatedwith STEMI.2,3 Refer to Fig. 1 for a depiction of the

dicine, University of Virginia School of Medicine, Leef Emergency Medicine and Medicine, University ofA 22908, USA

rights reserved. cardiology.th

eclinics.co

Fig. 1. Causes of ST segment elevation in adultpatients with chest pain. BER, benign early repolariza-tion; LBBB, left bundle branch block; LVH, left ventric-ular hypertrophy; MISC, miscellaneous; RBBB, rightbundle branch block; UNDEF, undefined.

Pollak & Brady602

causes of ST segment elevation in adult patientswith chest pain.2,3

In certain difficult cases, a patient’s ECG canresemble STEMI, yet manifest ST segment eleva-tion from a non–acute coronary syndrome (ACS)entity, the so-called STEMI mimics. In other situa-tions, the patient’s ECGmakes it difficult or impos-sible to determine whether STEMI is present, theso-called STEMI confounders; these confoundersto STEMI diagnosis are also mimickers of AMI.All suchcases can leave apractitionerwondering

whether to initiate reperfusion therapy, either viaadministration of a fibrinolytic agent or activationof STEMI alert process, in essence, whether toexpose patients to both the benefits and the risksof fibrinolysis or invasive coronary angiography. Insome cases, the astute clinician can detect analternative diagnosis masquerading as an STEMI.Failure to recognize these mimics can lead to inap-propriate use of resources, exposure of patients tounnecessary risk, and increased rather thandecreased morbidity and mortality. The ultimategoal with both the STEMI mimics and the con-founders is to maximize rapid, accurate diagnosiswhile avoiding delays in the treatment of alternativecauses of ST segment elevation. Because the riskof cerebral hemorrhage from fibrinolysis is notinsignificant, careful consideration of the ECG,looking for the STEMI mimics, is required inpatient-care situations in which primary percuta-neous coronary intervention (PCI) is not an option.More importantly, fibrinolysis given in the settingof certain STEMI mimics, such as acute myoperi-carditis, is associated with high mortality.Although each of the conditions discussed here

is unique, a common issue that must not be over-looked is the interpretation of the ECG within thecontext of the patient’s presentation; in otherwords, does the patient look like he or she is

experiencing a STEMI? The STEMI mimics andconfounders more often imitate the ECG findingsof AMI than the clinical syndrome, so the patientwith ST segment elevation but without a convinc-ingly clinical picture of STEMI should prompt theprovider to suspect a non-AMI presentation. Attimes, these diagnoses are very challenging,which will understandably impact the rapid appli-cation of reperfusion therapy and likely increasethe door-to-therapy time.

STEMI MIMICKING PATTERNSMyocarditis and Myopericarditis

Inflammation of thepericardiumandheartmuscle isa common cause of chest pain with ST segmentelevation. Seventy-three percent of patients diag-nosed with acute myopericarditis will have STsegment elevation on initial ECG.4 Additionally,44% of patients with chest pain and positivetroponin but who do not have obstructive coronarydisease by angiography demonstrate evidence ofmyocarditis by cardiac magnetic resonanceimaging (MRI) using late gadolinium enhancementto reveal areas ofmyocardial necrosis in a noncoro-nary distribution (ie, midwall or subepicardial ratherthan subendocardial).5

Myocarditis affects patients of all ages and hasa wide spectrum of clinical severity, ranging fromincidental chest discomfort to fulminant heart failurewith cardiogenic shock. The termmyocarditis refersto an inflammatory process of the heart muscle (asreflected by the presence of biomarkers and ECGchanges), whereas pericarditis refers to isolatedinflammation of the lining around theheart. Theperi-cardium is electrically silent, thus, when patientspresent with a clinical picture suggestive of pericar-ditis and demonstrate ST segment changes, themyocardium is also affected. Frequently, inflamma-tion involves both components, hence, the termmyopericarditis is used.The cause is not frequently elucidated, but a viral

cause is thought to be most common, witha minority of cases stemming from toxins or auto-immune processes. Patients may remembera prodrome of viral illness in the previous week ortwo. The pain is classically pleuritic in nature andchanges in severity with position (sitting forward orlying back).Biomarkers of myocardial necrosis, such as

troponin and creatinine kinase, are positive in thesetting of myocarditis, although they may be nega-tive in the setting of pure pericarditis. The absolutevalue of troponin elevation is associated with theextent of myocardial cell injury but only veryroughly correlates with clinical severity. A rub oncardiac auscultation is a highly specific but very

ECG Patterns Mimicking STEMI 603

insensitive physical examination finding and isfrequently absent when even a small effusion ispresent.6 Although a small pericardial effusion isnot uncommon, it is quite rare for pericarditis topresent with cardiac tamponade.

The electrocardiogram (Figs. 2 and 3) generallyevolves in patients with myocarditis/myopericardi-tis, yet this evolution occurs less rapidly than thatseen with STEMI; thus, serial ECGs, which demon-strate changes in the ST segment and T-wavemorphologies over minutes, are frequently usefulin making the distinction. ST segment elevationin multiple coronary distributions, especiallywhen seen in patients who are clinically stable,favors inflammation over STEMI in large partbecause patients who simultaneously occludemultiple coronary arteries typically present withshock or death.7 The presence of ST segmentdepression is less likely with myocarditis/myoper-icarditis (with the exception of lead AVR); whenpresent, ST segment depression likely representsreciprocal ST segment depression and suggestsSTEMI. The presence of PR segment depression

Fig. 2. Myopericarditis: Two examples of myopericarditis.(B) Myopericarditis with diffuse ST segment elevation and PV6 and elevation in lead aVR).

in leads with ST segment elevation as well as PRsegment elevation in lead AVR favors inflammationover STEMI. It is possible for myocarditis to be iso-lated anatomically in a single coronary artery distri-bution, leaving distinction from STEMI morechallenging. In cases involving a coexistent peri-cardial effusion, electrical alternans and/or dimin-ished electrical forces of the QRS complex canalso be seen.

Early Repolarization (Also Known as BenignEarly Repolarization)

Benign early repolarization (BER) is a normalvariant electrocardiographic pattern; it manifestsas ST segment elevation at the J point, predomi-nantly in the precordial leads. It is frequentlypresent in the general population, particularly inyounger patients and male gender. BER is associ-ated with young athletic men and is seen acrossall races; it has been linked to the black race,although some have disputed this association. Itis not itself a pathologic finding and has not been

(A) Myopericarditis with diffuse ST segment elevation.R segment changes (depression in leads II, III, aVF, and

Fig. 3. Myopericarditis versus STEMI. (A) ST segment elevation with PR segment depression of myopericarditis.(B) STEMI.

Pollak & Brady604

tied to a disease process; for unknown reasons,BER is seen frequently in patients with chest painwho have used cocaine.The electrocardiographic description (Figs. 4

and 5) of BER focuses on ST segment elevationwith the following associated features: STsegment elevation with an upward concavity ofits initial portion, notching or slurring of theterminal QRS complex (the J point), prominent Twaves that are symmetric and concordant withthe QRS complex (except in leads V1-V2), wide-spread distribution of ST segment elevation, andrelative temporal stability of the pattern. Thedegree of ST segment elevation, beginning at theJ-point elevation, is usually less than 3.0 mm,with a range of 0.5 to 5.0 mm. The ST segmentelevation morphologically seems as if the STsegment has been evenly lifted upwards from the

Fig. 4. Benign early repolarization with widespread, conc

isoelectric baseline at the J point with preservationof the normal concavity of the initial, up-slopingportion of the ST segment. The J point itself isfrequently notched or irregular in contour and isconsidered highly suggestive of BER. ProminentT waves are seen and are of large amplitude,slightly asymmetric morphology, and concordantwith the QRS complex; the height of the T wavesin BER ranges from approximately 6.5 mm in theprecordial distribution to 5.0 mm in the limb leads.The ST segment and T-wave abnormalities of BERare most often seen in leads V1 to V4. At times,coexistent changes are also seen in leads II, III,AVF, V5, and V6; importantly, BER-relatedchanges noted only in the limb leads are unusualand likely result from some other pathologicprocess, such as STEMI. Lastly, the chronic natureof the ST segment elevation is helpful in the

ave ST segment elevation.

Fig. 5. ST segment elevation in benign early repolarization (A) and STEMI (B).

ECG Patterns Mimicking STEMI 605

diagnosis of BER. The ST segment and T-waveabnormalities seen in BER will change very slowlyover time with a diminution of ST segment eleva-tion over decades; STEMI-related changes, onthe other hand, evolve over minutes to hours.

Previous Infarction with Ventricular Aneurysm

The presence of ST segment elevation on ECG inpatients with known prior myocardial infarctionpresents a special diagnostic challenge; thesepatients have known coronary disease and remainat an increased risk of recurrent STEMI. Patientswho have had a past myocardial infarction canpresent with the various mechanical complicationsand have lingering ST segment elevation that hasnot resolved. Other patients, after a past myocar-dial infarction with remodeling, can present withpersistent ST segment elevation attributable toaneurysmal dilatation of the infarcted segment ofmyocardium. In the prereperfusion era, rates of

Fig. 6. LV aneurysm. ST segment elevation in leads V1 to

left ventricular (LV) aneurysm approached 10%,but they have decreased dramatically since theintroduction of fibrinolysis and mechanical reper-fusion. Approximately one-quarter of patientswith persistent ST segment elevation followingSTEMI will have an LV aneurysm as determinedby echocardiography.8

When considered together, the patients’ historyof present illness, past medical history, and priorECG are the most useful tools to identify LV aneu-rysm and distinguish the ST segment elevationfrom that of STEMI. Patients should be able toclarify whether they had a prior heart attack andto what degree their current presentation is similaror dissimilar. In this case, the bedside echocardio-gram is generally unhelpful because the preexistingwall motion abnormality is difficult to distinguishfrom a new one.

The most frequent electrocardiographic mani-festation (Figs. 6 and 7) of ventricular aneurysm isST segment elevation, most often in the anterior

V5 along with Q waves and T-wave inversion.

Fig. 7. ST segment elevation in LV aneurysm (A) versus STEMI (B).

Pollak & Brady606

distribution; inferior and lateral aneurysms are alsoencountered. The 12-lead electrocardiogram de-monstrates ST segment elevation of varyingmorphologies and magnitudes, ranging fromobvious, convex ST segment elevation to minimal,concave elevations. Pathologic Q waves areusually observed in leads with ST segment eleva-tion. Inverted T waves of minimal magnitude arealso seen in these same leads. Of course,a comparisonwith apast ECGsince themyocardialinfarction, if available, can determine if any changehas occurred. Reciprocal ST segment depressionis usually not present and this is helpful in distin-guishing aneurysm from true STEMI. Lastly, theperformance of serial ECGs can be invaluable,with dynamic changes suggesting STEMI and theabsence of evolution suggesting aneurysm.

Coronary Vasospasm (Prinzmetal Angina)

Coronary vasospasm presents along a wide spec-trum of disease severity, ranging from intermittentchest pain (so-called variant angina) to clinicallysevere STEMI equivalents with concomitant heartfailure and ventricular fibrillation. It is exceptionallycommon among patients presenting to emergencydepartments with chest pain syndromes. TheClopidogrel and Acetyl Salicylic Acid in BypassSurgery for Peripheral ARterial Disease (CASPAR)trial found 49% of patients presenting with acutechest pain who did not have a culprit lesion byangiography did have coronary vasospasm byacetylcholine challenge.9 Coronary vasospasmcan occur in patients across a broad demographicwithout clear gender or race associations; more-over, it can occur along the length of any of thecoronary arteries. Coronary vasospasm is associ-ated with tobacco, cocaine use, and ergonovinederivatives and can occur in both atherosclerotic

as well as angiographically normal vessels. Thereis a form of coronary vasospasm caused byallergic histamine release, called Kounissyndrome.10 The final diagnosis of coronary vaso-spasm is made in the heart catheterization labora-tory. Prior use of provocation tests withacetylcholine and methylergonovines has largelyfallen out of favor.Because acute coronary vasospasm can lead to

near complete cessation of blood flow in theaffected coronary artery, its presentation and path-ophysiology parallels STEMI with the importantdistinction of absent vessel thrombosis. Rapidresponse to nitrates is a hallmark of coronary vaso-spasm. Treatment of normotensive patients withST segment elevation remains a useful strategy toidentify those with rapidly resolving syndromeswhen vasospasm is more likely. When consideringreperfusion treatment, it is important to considerthe possibility of coronary vasospasm.The association of coronary vasospasm with

cocaine (and other amphetamines) ingestion pres-ents another challenge. Cocaine is both vaso-spastic as well as thrombogenic, leaving users atan increased risk of both mechanisms of coronaryflow obstruction and resultant infarction. Inpatients with a clear amphetamine toxidrome (eg,hypertension, agitation, and so forth) as well asST segment elevation, it is prudent to treat the pa-tients with intravenous benzodiazepines and oralor topical nitrates with repeat performance of theECG to determine if persistent ST segment eleva-tion is present. It must be stressed that STEMIremains the working diagnosis and, if unable todemonstrate resolution of the ST segment eleva-tion within a short period of time, reperfusiontherapy should follow. Refer to Figs. 8 and 9 fora depiction of ST segment elevation as seen incoronary vasospasm.

Fig. 8. Coronary vasospasm: ST segment elevation in the inferior leads along with ST segment depression in leadsaVL, V2, and V3. This patient presented with chest pain and was ultimately diagnosed with coronary vasospasm(Prinzmetal angina). This form of ST segment elevation is indistinguishable from that seen in STEMI.

ECG Patterns Mimicking STEMI 607

Other Non-STEMI ST Segment ElevationSyndromes

Takotsubo cardiomyopathy (apical ballooningsyndrome)Apical ballooning syndrome (ABS) is a recentlyidentified disease in cardiology. In its short history,it has had different names, including Takostubocardiomyopathy and the broken heart syndrome,and it falls along the spectrum of acute stress-induced cardiomyopathies. Although the nameand classic presentation involve dyskinesis at theLV apex, multiple variations, including basal andmidcavitary dyskinesis, have also been reported.The clinical presentation of ABS is frequently indis-tinguishable from ACS and is well known to mimicSTEMI.11 Patients can present with the full spec-trum of chest pain equivalents from vague atypicalchest pain to severe chest pain with acute heartfailure and malignant dysrhythmia.

The spectrum of ECG findings is likewise varied,from minimal abnormality to profound anterior ST

Fig. 9. ST segment elevation in coronary vasospasm (A) ve

segment elevation (Fig. 10). In its most commonapical form, the ECG changes are limited to theanterior precordial leads, reflecting apical locationof injury. The inferior leads can also be involved,reflecting inferoapical injury. Because the historyand ECG and even echocardiogram can be indis-tinguishable from true AMI, the diagnosis of ABS ismade in the heart catheterization laboratoryfollowing diagnostic angiography.

Brugada syndrome and idiopathic ventricularfibrillationThe Brugada pattern includes right bundle branchblock (RBBB) (both incomplete and complete) withST segment elevation in leads V1 to V3; the naturalhistory of this syndrome is sudden cardiac death.Brugada syndrome was first described as a clinicalentity in 1992 to explain the observation thata particular cohort of patients prone to ventricularfibrillation despite structurally normal hearts hada distinct ECG pattern.12 Since its first description,

rsus STEMI (B).

Fig. 10. ST segment elevation in Takotsubocardiomyopathy.

Pollak & Brady608

a great deal has been learned about this relativelyrare inherited disease. It occurs in approximately 5out of 100 000 people across all nationalities, hasan autosomal-dominant inheritance pattern, anda male predominance (w70%).13 Three subtypeshave been described, with subtle variations in theshape of the ST segment. The underlying mecha-nism, resulting from a gene mutation, is a sodiumchannel loss of function.14

Clinically, the disease can present at any age, butthemeanageof suddendeath is 41 years. The char-acteristic ECG findings (Fig. 11) are dynamic andmay vary from one ECG to the next. Chest pain isnot part of the Brugada syndrome; however,syncope is a significant poor prognostic marker.The natural history includes cardiac arrest resultingfrom polymorphic ventricular tachycardia and/orventricular fibrillation. The disease andECG findingscan be exacerbated bymultiple medications aswellas conditions that alter myocardial membrane elec-trical stability (eg, fever, hypokalemia, ischemia,bradycardia, and increased autonomic tone).Vaughan Williams class Ic antiarrhythmic medica-tions (eg, flecainide and procainamide) are potentsodium channel blockers and are clinically used toelicit the Brugada pattern in suspected cases. Alarge number of medications in addition to sodiumchannel blockers have been reported to elicit a Bru-gadalike pattern on ECG. The significance of thesefindings to a particular patient remains unclear.

HyperkalemiaAs the serum potassium concentration increases,changes occur in ECG that can mimic ACS,

including STEMI.15 The elevation of the STsegment occurs following peaking of the T waveand widening of the QRS complex, suggestinghigh serum potassium concentrations. Althoughthe accompanying ST segment elevation may bemost impressive in a single region, the ECGchanges of hyperkalemia are present throughoutthe limb and precordial leads.

Postelectrical cardioversion/defibrillationFollowing transthoracic electrical cardioversion ordefibrillation, the ECG can manifest ST segmentelevation, among other abnormalities. ST segmentelevation has been noted following both cardiover-sion for atrial as well as ventricular tachyarrhyth-mias in up to 20% of patients. Additionally, STsegment deviation occurs following countershockby internal cardioverter-defibrillator in 25% ofpatients.16 The ST segment elevation is transientand resolves within minutes but can be profound,as much as 5 mm.17 Although impressive andalarming, this finding has not been associatedwith evidence of ongoing myocardial injury oradditional adverse sequelae.18

Hypothermia and Osborn wavesProminent J-point elevations, also known as Os-born waves, are a common finding in patientswho are hypothermic.19 When profound, the STsegment can be elevated, mimicking STEMI. Otherfindings include bradycardia and motion artifact.The mechanism and clinical implications of theOsborn wave remain unclear, but it is a transientfinding that resolves with normothermia. Osbornwaves can occur with either therapeutic hypo-thermia or accidental hypothermia.20

STEMI Confounding Patterns

The confounder electrocardiographic patterns arethose ECG entities that markedly reduce the elec-trocardiogram’s ability to detect changes relatedto ACS, not because the findings mimic STEMI butbecause the condition obscures the detection ofST segment elevation. The commonly confoundingpatterns include left bundle branch block (LBBB),ventricular paced rhythms, and LV hypertrophy.

LBBB

LBBB is the most classic example of a confounderand is exceptionally common in the evaluation ofpatients with chest pain. This pattern is linkedvery closely with ischemic heart disease in thatthe development of LBBB is associated withsignificant heart disease; it also identifies a popula-tion at an increased risk for STEMI, acute cardio-vascular complication, and poor outcome.21 TheLBBB pattern on the ECG includes an elevated

Fig. 11. Two forms of ST segment elevation as seen in the Brugada syndrome. (A) Convex form. (B) Concave(saddle-type) form.

ECG Patterns Mimicking STEMI 609

ST segment at baseline making it impossible touse the standard STEMI criteria for ST segmentelevation. The issue is further complicated by thefact that LBBB itself is a risk factor for cardiacdeath in that a new LBBB can be the presentingECG pattern in a proximal left anterior descendingcoronary artery (LAD) occlusion, thus a new LBBBpattern ECG in patients with a compatible clinicalscenario is considered a STEMI equivalent. Unfor-tunately, the rate of patients with AMI who presentwith new LBBB remains so small the mere pres-ence of new or presumed new LBBB does not it-self confer an increased risk of AMI.22 Moreover,although LBBB is associated with a poor prog-nosis for patients with chest pain, the diagnosticutility of its presence alone has been shown tobe weak with respect to the diagnosis of AMI.23

Two recurring scenarios illustrate the difficulty ofcorrectly identifying AMI in patientswith LBBB. Thefirst is patients with an intermediate risk story anda previously undiagnosed or unrecognized LBBB.The question, then, is this: should this new LBBBbe considered a STEMI? In considering the answerto this important question, it is instructive to recall

the mechanism through which AMI causes a newLBBB infarction of that portion of the conductionsystem related to LAD occlusion proximal to thefirst septal perforator branch. Given the proximallocation of the occlusion in the LAD, these AMIsinvolve a larger area of ischemic heart muscleand tend to have more severe presentations andhemodynamic compromise. It is likewise lesscommon for the new LBBB to be the ECG findingof AMI in patientswith an otherwise nonthreateningclinical presentation; in other words, these patientswith AMI tend to be rather ill on presentation, witha toxic appearance, pulmonary congestion, andcompromised perfusion.

The second clinical scenario occurs whenpatients with known LBBB present with acutechest pain, potentially suggestive of AMI; the clin-ical challenge is focusing on the possible presenceof AMI. This situation is themost appropriate appli-cation of the criteria Sgarbossa and colleagues24

derived from theGlobal Utilization of Streptokinaseand Tissue Plasminogen Activator for OccludedCoronary Arteries (GUSTO) trial. These criteria arebased on the finding that STEMI, when presenting

Pollak & Brady610

in the setting of LBBB, often seems different thanthe LBBB at baseline. Of course, an understandingof the ECG in the LBBB presentation is important.Such an understanding allows the clinician torecognize the expected findings of LBBB and,thus, the inappropriate findings that are associatedwith AMI, not only those described by Sgarbossaand colleagues but also other abnormalities. The12-lead ECG (Figs. 12 and 13) in patients withLBBB records the abnormal ventricular activationas it moves from right to left, producing a broad,mainly negative QS or rS complex in lead V1. Inlead V6, late intrinsicoid deflection is noted, result-ing in a positive, monophasic Rwave; similar struc-tures are frequently found in leads I and aVL. PoorR wave progression or QS complexes are noted inthe right to mid precordial leads, rarely extendingbeyond leads V4 or V5. QS complexes may alsobe encountered in leads III and aVF. The antici-pated or expected ST segment–T-wave configura-tions are discordant, directed opposite from theterminal portion of the QRS complex, and calledQRS complex–T-wave axes discordance. Assuch, leads with either QS or rS complexes mayhave markedly elevated ST segments, mimickingAMI. Leads with a large monophasic R wavedemonstrate ST segment depression. The Twave, especially in the right tomidprecordial leads,has a convex upward shape or a tall, vaultingappearance, similar to the hyperacute T wave ofearly myocardial infarction. The T waves in leadswith the monophasic R wave are frequently in-verted. Loss of this normal QRS complex–T-waveaxes discordance in patients with LBBBmay implyan acute process, such as AMI. An inspection ofthe ECG in patients with LBBBmust be performed,

Fig. 12. ST segment and T-wave changes seen in the norm

looking for a loss of this QRS complex–T-waveaxes discordance.Patients with known LBBB presenting with

a clinical presentation potentially suggestive ofAMI is the most appropriate application of thecriteria Sgarbossa and colleagues24 derivedfrom the GUSTO trial. These criteria are basedon the finding that STEMI, when presenting inthe setting of LBBB, often seems different thanthe LBBB at baseline. Unfortunately, althoughthe most accepted and widely used clinicalcriteria for diagnosis of AMI in LBBB, the Sgar-bossa criteria suffer from low sensitivity of 36%while retaining a relatively robust specificity of96% in the validation cohort. The most appro-priate course in these cases, then, is to proceedmost briskly in patients meeting the Sgarbossacriteria (with a score of 3 or more) and, in thosewho do not, relying on the pretest probabilitybased on history and clinical risk to determinethe overall risk of AMI. This rule states that 3specific ECG criteria (Fig. 14, Table 1) are inde-pendent predictors of AMI in patients withLBBB. The ECG criteria suggesting a diagnosisof AMI, ranked with a scoring system based onthe probability of such a diagnosis, include (1)ST segment elevation greater than 1 mm, whichwas concordant with the QRS complex (score of5); (2) ST segment depression greater than1 mm in leads V1, V2, or V3 (score of 3); and (3)ST segment elevation greater than 5 mm, whichis discordant with the QRS complex (score of 2).A total score of 3 or more suggests that patientsare likely experiencing an AMI based on theECG criteria. With a score less than 3, the electro-cardiographic diagnosis is less assured, requiring

al appearing left bundle branch block.

Fig. 13. ST segment elevation as seen in LBBB (A) versus STEMI (B).

ECG Patterns Mimicking STEMI 611

additional evaluation. This clinical predictioninstrument supports the contention that a detailedknowledge of the associated, or anticipated, STsegment–T-wave changes resulting from theabnormal ventricular conduction of the LBBB isa must. Such an understanding of the ECG inLBBB consequently allows the clinician to

Fig. 14. The Sgarbossa criteria. (A) Concordant ST segmelimited to leads V1, V2, and V3. (C) Excessive discordant ST sof the major, terminal portion of the QRS complex (thickarrow) is key determinant in the consideration of the Sga

recognize the unanticipated morphologies thatmay be suspicious for AMI.

LV Hypertrophy

LV hypertrophy (LVH) pattern is one of the mostcommon clinical pathologic findings on the

nt elevation. (B) Concordant ST segment depressionegment greater than 5 mm. Note that the relationshiparrow) and the initial portion of the ST segment (thinrbossa criteria.

Fig. 15. ST segment elevation in leads V1 to V3 as seenin the LVH with strain pattern.

Table 1The Sgarbossa criteria

Odds Rations Supporting AMI and PredictionScores for Independent Electrocardiographic

Criteria

CriterionOdds Ratio(95% CI) Score

ST segmentelevation �1 mmconcordant withQRS complex

25.2 (11.6–54.7) 5

ST segmentdepression �1 mmin lead V1, V2, or V3

6.0 (1.9–19.3) 3

ST segment elevation�5 mm & discordantwith QRS complex

4.3 (1.8–10.6) 2

Abbreviation: CI, confidence interval.

Pollak & Brady612

electrocardiogram. Although 50% of the adultpopulation diagnosed with hypertension will haveLVH by echocardiogram, only 20% of those willhave characteristic ECG findings of LVH. Indeed,the presence of LVH does carry a worse prognosisfor a given patient, and the presence of concomi-tant downsloping ST segment depression, thestrain appearance further degrades prognosis.25

There are many criteria for determining LVH byECG, all of which rely on the increase in voltage re-flected by the increased myocardial mass alongwith either leftward axis or left atrial abnormality.Unfortunately, all ECG criteria suffer from relativelypoor performance characteristics (sensitivityapproximately 50% and specificity approximately80%) when validated against LV mass calculatedby echocardiography or cardiac computed tomog-raphy (CT).26 Although the detection of underlying,anatomic LVH is important with respect to patientmanagement, in this particular instance, theauthors are only addressing the secondary repolar-ization abnormalities caused by LVH, in essence,the ST segment deviations (both elevation anddepression) as well as T-wave inversions.Approximately 80% of patients with the LVH by

voltage (Figs. 15 and 16) pattern demonstrate thestrain pattern; the strain pattern includes significantST segment changes (elevation and depression)and T-wave abnormalities (prominent T wavesand T-wave inversion). ST segment elevation inthe setting of LVH is almost exclusively seen inthe anterior distribution (leads V1 to V4). STsegment elevation is encountered in this distribu-tion alongwith prominent T waves; the ST segmentelevation can reach up to 5 mm in height in the

anterior leads. The lateral leads (leads I, aVL, V5,and V6) demonstrate large, prominent, positivelyoriented QRS complexes with marked ST segmentdepression and T-wave inversion. Although thestrain pattern can also be seen inferiorly, inferiorST segment elevation should not quickly beascribed to LVH. The LVH with pattern is associ-ated with poor R-wave progression, mostcommonly producing a QS pattern; thesecomplexes are located in leads V1, V2, and V3;furthermore, a leftward axis and left atrial abnor-mality add credence to ST segment elevationstemming from LVH. The ST segment elevationassociated with LVH is generally unchanging overtime, making a previous ECG for comparisonparticularly useful. Serial ECGs are similarly usefulbecause there should be no evidence of evolvingchanges unless there is active ischemia.

Ventricular Paced Pattern from ImplantedDevice

More than 2 million people in the United Stateshave an implanted pacemaker or internalcardioverter-defibrillator. Because a pacedventricular rhythm causes bundle-branch blockmorphology, it produces many of the same diag-nostic limitations as LBBB described earlier.Moreover, patients with implanted cardiac devices

Fig. 16. ST segment elevation in LVH with strain (A) versus STEMI (B).

ECG Patterns Mimicking STEMI 613

tend to be older and are more likely to have under-lying heart disease.

Ventricular pacing leads can be placed in theapex of the right ventricle causing a left bundlepattern or on the surface of the LV either in a coro-nary vein or in an epicardial location causing moreof a RBBB type of pattern. Patients with LV pacingleads frequently have such units placed for cardiacresynchronization therapy as part of the treatment

Fig. 17. ST segment elevation in the ventricular paced pat

of heart failure. Patients presenting with a pacedrhythm may or may not be dependent on theirpacemaker to maintain a perfusing ventricularrate. It can be tempting to pause pacing andanalyze the morphology of the underlying rhythm,but this is unreliable because of T-wave memorywherein the myocyte repolarization vector remainsdespite the change in depolarization vector.27

Sgarbossa and colleagues28 have investigated

tern.

Fig. 18. ST segment elevation in the ventricular paced pattern (A) versus STEMI (B).

Pollak & Brady614

the utility of applying their criteria to patients withan LBBB caused by paced rhythm and foundvery poor sensitivity and modest specificity; ina very basic sense, they found similar findings asnoted in the ECG diagnosis of AMI in the settingof LBBB. Refer to Figs. 17 and 18 for a depictionof the ST segment changes seen in the ventricularpaced pattern.The RBBB has long been recognized as a high-

risk feature in the presentation of AMI; in fact, ithas been noted that the presence of a newRBBB on presentation with AMI significantlyincreases 30-day mortality.29 Current guidelinesrequire ST segment elevation be present withRBBB for the diagnosis of STEMI; further, thelongstanding maxim has been the presence ofRBBB does not alter the detection of ST elevation.However, this notion has been challenged bythose who point out T-wave inversions in the ante-rior leads of patients with RBBB can impair thedetection of more subtle ST segment changes;also, severe proximal coronary occlusions can beassociated with new RBBB with hemiblock and,thus, do not demonstrate typical ST segmentelevation.30–35 Patients with a clinically worrisomepresentation and new bundle branch block (bothleft and right) should be carefully considered forAMI because the benefit of early reperfusiontherapy is quite high in these patients.

REFERENCES

1. Niska R, Bhulya F, Xu J. National Hospital Ambulatory

Medical Care Survey: 2007 emergency department

summary. Natl Health Stat Report 2010;26:1–32.

2. Brady WJ. Electrocardiographic ST segment ele-

vation: the diagnosis of AMI by morphologic anal-

ysis of the ST segment. Acad Emerg Med 2001;8:

961–7.

3. Brady WJ. Electrocardiographic ST segment eleva-

tion in emergency department chest pain centers:

etiology responsible for the STsegment abnormality.

Am J Emerg Med 2001;19:25–8.

4. Bruce M, Spodick D. Atypical electrocardiogram in

acute pericarditis: characteristics and prevalence.

J Electrocardiol 1980;13:61–6.

5. Codreanu A. Detection of myocarditis by contrast-

enhanced MRI in patients presenting with acute

coronary syndrome but no coronary stenosis.

J Magn Reson Imaging 2007;25:957–64.

6. Spodick D. Acute pericarditis. JAMA 2003;289:

1150–3.

7. Pollak P. Multiple culprit arteries in patients with ST

segment elevation myocardial infarction referred

for primary percutaneous coronary intervention.

Am J Cardiol 2009;104:619–23.

8. Galiuto L. Functional and structural correlates of

persistent STelevation after acute myocardial infarc-

tion successfully treated by percutaneous coronary

intervention. Heart 2007;93:1376–80.

9. Ong P. Coronary artery spasm as a frequent cause of

acute coronary syndrome. J AmColl Cardiol 2008;52:

523–7.

10. Kounis N. Kounis syndrome (allergic angina and

allergic myocardial infarction): a natural paradigm?

Int J Cardiol 2006;110:7–14.

11. Prasad A, Lerman A, Rihal C. Apical ballooning

syndrome (Tako-Tsubo or stress cardiomyopathy):

a mimic of acute myocardial infarction. Am Heart J

2008;155:408–17.

ECG Patterns Mimicking STEMI 615

12. Brugada P. Right bundle branch block, persistent

ST segment elevation and sudden cardiac death:

a distinct clinical and electrocardiographic syndrome.

J Am Coll Cardiol 1992;20:1391–6.

13. Benito B. Gender differences in clinical manifesta-

tions of Brugada syndrome. J Am Coll Cardiol

2008;52:1567–73.

14. Antzelvitch C. Brugada syndrome: report of the

second consensus conference. Circulation 2005;

111:659–70.

15. Simon B. Pseudomyocardial infarction and hyperka-

lemia: a case report and subject review. J Emerg

Med 1988;6:511–5.

16. Gurevitz O. ST-segment deviation following implant-

able cardioverter defibrillator shocks: incidence,

timing, and clinical significance. Pacing Clin Electro-

physiol 2002;25:1429–32.

17. Van Gelder IC. Incidence and clinical significance of

ST segment elevation after electrical cardioversion

of atrial fibrillation and atrial flutter. Am Heart J

1991;121:51–6.

18. Kok LC. Transient ST elevation after transthoracic

cardioversion in patients with hemodynamically

unstable ventricular tachyarrhythmia. Am J Cardiol

2000;85:878–81.

19. Antzelevitch C. J wave syndromes. Heart Rhythm

2010;7:549–58.

20. Noda T. Prominent J wave and ST segment elevation:

serial electrocardiographic changes in accidental hy-

pothermia. J Cardiovasc Electrophysiol 2003;14:223.

21. Schneider JF. Newly acquired left bundle-branch

block: the Framingham Study. Ann Intern Med 1979;

90(3):303–10.

22. Chang AM. Lack of association between left bundle-

branch block and acute myocardial infarction in

symptomatic ED patients. Am J Emerg Med 2009;

27(8):916–21.

23. Jain S. Utility of left bundle branch block as a diag-

nostic criterion for acute myocardial infarction. Am J

Cardiol 2011;107:1111–6.

24. Sgarbossa E. Electrocardiographic diagnosis of

evolving acute myocardial infarction in the presence

of left bundle branch block. N Engl J Med 1996;334:

481–7.

25. Sullivan J. Left ventricular hypertrophy: effect on

survival. J Am Coll Cardiol 1993;22:508–13.

26. Hancock EW. AHA/ACCF/HRS recommendation for

the standardization and interpretation of the electro-

cardiogram: part V: electrocardiogram changes

associated with cardiac chamber hypertrophy. J Am

Coll Cardiol 2009;53:992–1002.

27. Rosenbaum M. Electrotonic modulation of the T

wave and cardiac memory. Am J Cardiol 1982;50:

213–23.

28. Sgarbossa E. Early electrocardiographic diagnosis

of acute myocardial infarction in the presence of

ventricular paced rhythm. Am J Cardiol 1996;77:

423–4.

29. Scheinman M. Clinical and anatomic implications of

intraventricular conduction blocks in acute myocar-

dial infarction. Circulation 1972;46:753–60.

30. Widimsky P. Primary angioplasty in acute myocardial

infarction with right bundle branch block: should

new onset right bundle branch block be added to

future guidelines as an indication for reperfusion

therapy? Eur Heart J 2012;33(1):86–95.

31. Kamp T. Myocardial infarction, aortic dissection,

and thrombolytic therapy. Am Heart J 1994;128:

1234–7.

32. Luo JL. Type A aortic dissection manifesting as

acute myocardial infarction: still a lesson to learn.

Acta Cardiol 2009;64:499–504.

33. Ohlmann P. Diagnostic and prognostic value of

circulating d-dimers in patients with acute aortic

dissection. Crit Care Med 2006;34:1358–64.

34. Falterman T. Pulmonary embolism with ST segment

elevation in leads V1 to V4: case report and review

of the literature regarding electrocardiographic

changes in acute pulmonary embolism. J Emerg

Med 2001;21:255–61.

35. Fasullo S. An unusual presentation of massive

pulmonary embolism mimicking septal acute

myocardial infarction treated with tenecteplase.

J Thromb Thrombolysis 2009;27:215–9.