PbhocswUfh

Ctd

C2

0d

Comparison of the Prognostic Effect of Left Versus Right VersusNo Bundle Branch Block on Presenting Electrocardiogram in AcuteMyocardial Infarction Patients Treated With Primary Angioplasty

in the Primary Angioplasty in Myocardial Infarction Trials

Mayra Guerrero, MDa,*, Kishore Harjai, MDb, Gregg W. Stone, MDc, Bruce Brodie, MDd,David Cox, MDe, Judy Boura, MSa, Lorelei Grines, PhDa, William O’Neill, MDa, and

Cindy Grines, MDa

The presence of bundle branch block (BBB) has been associated with poor outcomes inpatients who have acute myocardial infarction. Whether this is true in the angioplasty erais not known. We sought to evaluate the outcome of patients with acute myocardialinfarction and BBB who were treated with primary angioplasty. We evaluated 3,053patients who underwent emergency catheterization in the PAMI trials. Patients who hadleft BBB (n � 48, 1.6%) on presenting electrocardiogram were compared with patientswho had right BBB (n � 95, 3.1%) or no BBB (n � 2,910, 95.3%). Patients who had BBBwere older and more frequently had diabetes mellitus, peripheral vascular disease, andprevious coronary artery bypass grafting. They had lower ejection fraction and moremultivessel disease. There were no significant differences in door-to-balloon time, finalThrombolysis In Myocardial Infarction flow grade or stent use. In-hospital major adversecardiac events (death, ischemic target vessel revascularization, and reinfarction) werehigher in patients who had BBB due primarily to increased in-hospital death (left BBB14.6%, right BBB 7.4%, no BBB 2.8%, p < 0.0001). In multivariate logistic regressionanalysis, left BBB was an independent predictor of in-hospital death (odds ratio 5.53, 95%confidence interval 1.89 to 16.1, p � 0.002). In conclusion, patients who have acutemyocardial infarction and BBB have increased co-morbidities and higher mortality ratesdespite treatment with primary angioplasty. Despite early identification of multivesseldisease with triage to angioplasty or coronary artery bypass grafting, if necessary, similartreatment times, and final Thrombolysis In Myocardial Infarction grade 3 flow, thepresence of left BBB on admission electrocardiogram in patients who have acute myo-cardial infarction is an independent predictor of in-hospital mortality. Because 85% ofdeaths in patients who have left BBB occur within the first week, these patients shouldbe recognized early and receive prompt and aggressive treatment. © 2005 Elsevier Inc.

All rights reserved. (Am J Cardiol 2005;96:482–488)

gctrWBwnlBm

M

TM

atients who have acute myocardial infarction (AMI) and leftundle branch block (BBB) on admission electrocardiogramave been found to have unfavorable short- and long-termutcomes. This observation has been reported for several de-ades1–9 and has not changed despite development of aggres-ive medical treatments.10–19 The poor prognosis is seen evenhen thrombolytic therapy is instituted.10,17 In the first Globaltilization of Streptokinase and Tissue Plasminogen Activator

or Occluded Coronary Arteries (GUSTO-I) trial, patients whoad AMI and any type of BBB on admission electrocardio-

The aWilliam Beaumont Hospital, Royal Oak, Michigan; the bGuthrielinic, Sayre, Pennsylvania; the cColumbia University Medical Center and

he Cardiovascular Research Foundation, New York, New York; theMoses Cone Hospital, Greensboro, North Carolina; and eMid Carolinaardiology, Charlotte, North Carolina. Manuscript received January 17,005; revised manuscript received and accepted April 8, 2005.

* Corresponding author. Tel. 248-898-5237; fax: 248-898-7239.

gE-mail address: mguerrero@beaumont.edu. (M. Guerrero).

002-9149/05/$ – see front matter © 2005 Elsevier Inc. All rights reserved.oi:10.1016/j.amjcard.2005.04.006

ram had higher 30-day mortality rates than did matchedontrol subjects (18% vs 11%), despite the use of thrombolyticherapy in all patients.10 Thrombolytic reperfusion efficacyates in patients who have left BBB have not been well defined.

hether the poor prognosis of patients who have AMI and leftBB improves in the current era of emergency catheterizationith triage to primary angioplasty, stenting, or bypass surgery hasot been systematically analyzed. This study evaluated short- andong-term outcomes of patients who had AMI and left or rightBB on admission electrocardiogram and were treated with pri-ary angioplasty.

ethods

his is a post hoc analysis of the Primary Angioplasty inyocardial Infarction (PAMI) database. We pooled demo-

raphic, clinical, angiographic, and outcomes data on 3,053

www.AJConline.org

pt�PLhip(t3PorstipPttrPPutli

idScotiiad1

ctIctf

gctStwp

uwtvwaboeoesawcl

R

Oec1habhPPl1t

orBwfwroI

trbff0tTlw

483Coronary Artery Disease / Please supply wording

atients with AMI who underwent emergency catheteriza-ion with planned primary angioplasty in the PAMI-II20,21 (n

1,099), Stent PAMI22 (n � 1,264), Air PAMI23 (n � 71),AMI with No Surgery On Site (No SOS)24 (n � 499), andocal PAMI25 (n � 120) trials. The protocols of these trialsave been published previously. In the PAMI studies, wencluded patients who presented within 12 hours of chestain onset and had electrocardiographic evidence of AMIST-segment elevation �1 mm in 2 contiguous leads, pos-erior injury demonstrated by ST-segment depression in 2 of

anterior precordial leads, or complete left BBB). In theAMI-II trial, high-risk patients were randomized to receiver not receive an intra-aortic balloon pump,20 and the low-isk patients were randomized to coronary care unit admis-ion or early discharge on day 3 without stress testing.21 Inhe Stent-PAMI trial, patients who had AMI were random-zed to receive percutaneous transluminal coronary angio-lasty (PTCA) alone or with stent placement.22 In the Air-AMI trial, patients who had AMI were randomized to

hrombolytics or early transfer to an institution that was ableo perform primary PTCA (we excluded patients who wereandomized to thrombolytics in this trial).23 In theAMI-No SOS trial, all patients were treated with primaryTCA in centers without elective PTCA or surgical back-p.24 The Local PAMI trial was a pilot study that evaluatedhe effect of intramural infusion of heparin into the infarctesion after primary PTCA for AMI.25 All patients includedn these trials underwent emergency coronary angiography.

Study groups and end points: Patients were classifiednto 3 groups: complete left BBB on admission electrocar-iogram (n � 48), complete right BBB (n � 95), andT-segment elevation without BBB (n � 2,910). We ex-luded patients who had no information about the presencef BBB (4.9%). Baseline clinical and angiographic charac-eristics were analyzed. The following end points were stud-ed: death, reinfarction, ischemic target vessel revascular-zation, and the composite of these outcomes as majordverse cardiac events. These end points were evaluateduring index hospitalization and at 1 month, 6 months, andyear after randomization.Research nurses collected the data prospectively and

ompleted case report forms. Independent data monitorsraveled to the participating sites to verify hospital records.ndependent reviewers analyzed cineangiograms to assessoronary anatomy and Thrombolysis In Myocardial Infarc-ion flow grade. Complete 1-year follow-up was availableor 2,838 patients (93%).

Statistical analysis: Demographics, clinical and angio-raphic characteristics, treatment received, and clinical out-omes of patients who had left BBB were compared withhose of patients who had right BBB and those who hadT-segment elevation without BBB. Wilcoxon’s rank-sum

est was used for continuous variables and chi-square testas used for categorical variables where appropriate (ex-

ected frequency �5); otherwise, Fisher’s exact test was T

sed. Stepdown multivariate logistic regression analysesere performed to identify independent predictors of mor-

ality. We included in the first step the variables with uni-ariate p values �0.10 and variables that were associatedith increased in-hospital death. These variables included

ge �70 years, gender, heart rate �100 beats/min, systoliclood pressure �100 mm Hg, Killip’s class �1, left BBBn admission electrocardiogram, right BBB on admissionlectrocardiogram, diabetes mellitus, hypertension, previ-us myocardial infarction, previous coronary bypass graft,jection fraction �40%, multivessel disease, final percenttenosis, final Thrombolysis In Myocardial Infarction flow,nd right coronary infarct-related artery. A p value �0.05as considered statistically significant. All analyses were

ompleted with SAS 8.0 (SAS Institute, Cary, North Caro-ina).

esults

f 3,053 patients, 48 (1.6%) had left BBB on admissionlectrocardiogram and 95 (3.1%) had right BBB. Baselinelinical and angiographic characteristics are listed in Table. Patients who had BBB were older and more frequentlyad diabetes mellitus, previous coronary artery bypass graft,nd peripheral vascular disease. Patients who had left BBB,ut not those who had right BBB, also tended to haveypertension, previous myocardial infarction, previousTCA, and Killip’s class �1 on admission more frequently.atients who had RBBB had higher creatine phosphokinase

evels (median 1,443 mU/ml in patients with left BBB,,935 mU/ml in those with right BBB, 1,445 mU/ml inhose with no BBB, p � 0.03).

Procedural data for patients who underwent percutane-us coronary intervention are listed in Table 2. The infarct-elated artery was identified in 94% of patients who hadBB and 97% of patients who did not have BBB. Patientsho had BBB had more diseased vessels, lower ejection

raction, and fewer right coronary infarct-related arteriesith a trend toward more left anterior descending infarct-

elated arteries. There was no significant difference in initialr final percent stenosis or Thrombolysis In Myocardialnfarction flow grade.

In-hospital treatment: The treatment received by pa-ients who had BBB did not differ significantly from thateceived by patients who did not have BBB. Aspirin and �lockers in the emergency room were used with the samerequency (aspirin use 88% for those with left BBB, 92%or those with right BBB, 89% for those with no BBB, p �.65; �-blocker use 27% for those with left BBB, 30% forhose with right BBB, 35% for those with BBB, p � 0.4).here was a trend for less heparin use in patients who had

eft or right BBB (69% for those with BBB, 77% for thoseith right BBB, 84% for those with no BBB, p � 0.054).

he median time from emergency room to balloon expan-

sf0aBwPc

dTtg

Ti

TB

V

AWHSKDHPCCCPPPP

TP

V

CSEEIFIFMN

I

484 The American Journal of Cardiology (www.AJConline.org)

ion was similar (105 minutes for those with left BBB, 115or those with right BBB, 111 for those with no BBB, p �.65). All patients underwent emergency catheterization,nd PTCA was performed in 88% of patients who had leftBB, 86% of those who had right BBB, and 91% of thoseho had no BBB. The remaining patients did not undergoTCA due to unidentified infarct-related artery or need fororonary artery bypass grafting (Table 2). There was no

able 1aseline clinical and angiographic characteristics

ariable BBB

Left(n � 48)

ge (yrs) 69 � 10omen 10 (21%)eart rate (beats/min) 92 � 21ystolic blood pressure (mm Hg) 111 � 24illip’s class �I 11 (23%)iabetes mellitus 13 (27%)ypertension 29 (60%)eripheral vascular disease 6 (13%)erebrovascular disease 2 (4.4%)ongestive heart failure 2 (4.4%)urrent smoker 16 (34%)revious angina pectoris 6/29 (21%)revious myocardial infarction 10 (21%)revious coronary angioplasty 8 (17%)revious coronary bypass 4 (8.5%)

Values are numbers of patients (percentages) or means � SD.

able 2rocedural data in patients undergoing percutaneous coronary intervention

ariable BBB

Left(n � 48)

oronary angioplasty 42 (88%)tent used 18 (38%)mergency room to balloon time (min) 163 � 165 (105)jection fraction (%) 43.9 � 14

nitial stenosis (%) 98 � 4 (100)inal stenosis (%) 13 � 18 (10)nitial TIMI grade 3 flow 8 (17%)inal TIMI grade 3 flow 37 (90%)ultivessel coronary disease 30 (63%)o. of coronary arteries narrowed1 18 (37.5%)2 17 (35.4%)3 12 (25%)4 1 (2.1%)

nfarct-related arteryLeft anterior descending 21 (43.8%)Right coronary 15 (31.2%)Left circumflex 7 (14.6%)Left main 1 (2.1%)Vein graft or internal mammary 1 (2.1%)Unidentified 3 (6.2%)

Values are numbers of patients (percentages) or means � SD (medians

TIMI � Thrombolysis In Myocardial Infarction.

ifference in the frequency of use of stents, incidence ofhrombolysis In Myocardial Infarction grade 3 flow after

he procedure, or periprocedural complications acrossroups.

Outcomes: In-hospital outcomes are listed in Table 3.he composite of in-hospital major adverse cardiac events,

ncluding death, ischemic target vessel revascularization,

p Value

ight� 95)

None(n � 2,910)

66 � 13 61 � 12 �0.000123 (24%) 794 (27%) 0.50

90 � 19 89 � 21 0.2717 � 26 109 � 26 0.029

14 (15%) 408 (14%) 0.2523 (24%) 481 (17%) 0.02349 (53%) 1,320 (46%) 0.07612 (13%) 163 (5.8%) 0.0038 (8.5%) 145 (5.1%) 0.295 (5.3%) 70 (2.5%) 0.098

29 (31%) 1,177 (41%) 0.117/88 (19%) 363/2,187 (17%) 0.68

15 (16%) 423 (15%) 0.429 (9%) 282 (10%) 0.259 (10%) 111 (4%) 0.0094

p Value

Right(n � 95)

None(n � 2,910)

82 (86%) 2,646 (91%) 0.2326 (29%) 914 (32%) 0.51133 � 76 (115) 139 � 159 (111) 0.65

44.6 � 13 48.8 � 12 0.001698 � 5 (100) 97 � 7 (100) 0.4317 � 19 (15) 16 � 17 (10) 0.30

10 (11%) 318 (11%) 0.4772 (88%) 2,453 (93%) 0.1848 (51%) 1,353 (47%) 0.069

47 (49.5%) 1,557 (53.5%) 0.0319 (20%) 762 (26.2%)29 (30.5%) 575 (19.8%)0 (0%) 16 (0.5%)

46 (48.4%) 1,178 (40.5%) 0.1327 (28.4%) 1,195 (41%) 0.0414 (14.7%) 385 (13.2%) 0.790 (0%) 31 (1.1%)1 (1.1%) 17 (0.6%)7 (7.4%) 104 (3.6%)

R(n

1

1

).

ap(h0mw0pBEBgP1t�r

(odsn

nhrs

dgss(

dw(a

hwBopwrt

TI

V

DIRMHP

485Coronary Artery Disease / Please supply wording

nd reinfarction, was higher in patients who had BBB, duerimarily to a significant increase in in-hospital death14.6% in patients who had left BBB, 7.4% in those whoad right BBB, 2.8% in those who had no BBB, p �.0013). The difference in mortality was significant at 1onth (15% in patients who had left BBB, 7.5% in thoseho had right BBB, 3.1% in those who had no BBB, p �.0001; Figure 1) and remained significant at 1 year (16% inatients who had left BBB, 15% in those who had rightBB, 6% in those who had no BBB, p � 0.001; Figure 2).ighty-six percent of deaths among patients who had leftBB occurred within the first 7 days, whereas the otherroups demonstrated additional deaths over the next year.atients who had right BBB had a higher reinfarction rate atmonth (6.7% in patients who had right BBB, 2.3% in

hose who had left BBB, 2.2% in those who had no BBB, p0.03) and tended to have more ischemic target vessel

evascularizations at 1 year (Table 4).

Cause of in-hospital death: Seven of 48 patients14.6%) who had left BBB died during hospitalization. Sixf 7 deaths (86%) were of cardiovascular origin. One patientied of stroke, and 5 deaths were cardiac (i.e., cardiogenichock in 3 patients and arrhythmic cardiac arrests in 2). Theoncardiac death was due to multiorgan failure after coro-

able 3n-hospital outcomes of all patients

ariable BBB

Left(n � 48)

eath 7 (14.6%)schemic target vessel revascularization 1 (2.1%)einfarction 0 (0%)ajor adverse cardiac events* 8 (16.7%)ospital Stay (d) 7.1 � 8.6 (4.5)eak creatine phosphokinase (mU/ml) 2,050 � 2,026 (1,443)

Values are numbers of patients (percentages) or means � SD (medians* Death, ischemic target vessel revascularization, and reinfarction.

Figure 1. Thirty-day Kaplan-Meier survival curves.

ary artery bypass grafting. Six of the 7 subjects (86%) whoad left BBB and died had multivessel disease; the infarct-elated artery was the left anterior descending artery in 5ubjects (71%).

Seven of 95 subjects (7.4%) who had right BBB dieduring hospitalization. All deaths were cardiac, i.e., cardio-enic shock in 4 patients and lethal arrhythmia in 3. Theseubjects had multivessel disease, and the left anterior de-cending artery was the infarct-related artery in 4 subjects57%).

Eighty-two of 2,910 patients (2.8%) who had no BBBied during hospitalization. Sixty-seven of these deathsere of cardiac origin (82%), 14 (17%) were noncardiac

cerebrovascular accident, bleeding, or medical reasons),nd cause of death was not known in 1 patient.

Cause of death at 1 year: More deaths in subjects whoad BBB had a cardiac origin at 1 year of follow-up. Thereere 164 deaths among 2,584 patients (6%) who had noBB at 1 year; a cardiovascular cause was identified in 102f these deaths (62%). There were 7 deaths among the 45atients (16%) who had left BBB, and 6 (86%) of theseere cardiovascular. Twelve of 79 patients (15%) who had

ight BBB died by 1 year of follow-up, and 10 (83%) ofhese had a cardiac origin.

p Value

Right(n � 95)

None(n � 2,910)

7 (7.4%) 82 (2.8%) �0.00014 (4.2%) 77 (2.7%) 0.464 (4.2%) 53 (1.8%) 0.21

11 (11.6%) 176 (6.1%) 0.00137.1 � 5 (6) 6.5 � 5.7 (5) 0.11

3,226 � 3,449 (1,935) 2,036 � 2,022 (1,445) 0.03

).

Figure 2. One-year Kaplan-Meier survival curves.

hydtwr0dp

D

ItlaIPhdtapAIpc

amdMwBomf

pctpuitrgmtBfrahBpna

rfnPfltpb

(

TO

V

DIRM

TIr

V

RHWKLAEF

486 The American Journal of Cardiology (www.AJConline.org)

Multivariate analysis: Independent predictors of in-ospital death found in multivariate logistic regression anal-sis are listed in Table 5. Right BBB was not an indepen-ent predictor of death. After adjusting for other variables,he presence of left BBB on admission electrocardiogramas an independent predictor of in-hospital death (odds

atio 5.53, 95% confidence interval 1.89 to 16.1, p �.0018) and death at 1 month (odds ratio 5.18, 95% confi-ence interval 1.79 to 14.9, p � 0.0024) but was not aredictor of death at 1 year.

iscussion

n our study, all patients underwent emergency catheteriza-ion, which provided information about coronary anatomy,eft ventricular ejection fraction, and infarct-related arterynd produced higher rates of Thrombolysis In Myocardialnfarction grade 3 flow as a result of primary angioplasty.revious studies that analyzed the prognosis of patients whoad AMI and BBB lacked this information. Therefore, ourata allow a better understanding of the coronary pathologyhat might influence outcome. Further, it provides insightsbout the effect of primary angioplasty on outcome in theseatients. Whether the poor prognosis of patients who haveMI and BBB is neutralized by establishing Thrombolysis

n Myocardial Infarction grade 3 flow with primary angio-lasty had not been systematically analyzed in large multi-enter trials.

able 4ne-year outcomes

ariable BBB

Left(n � 48)

eath 7/45 (16%)schemic target vessel revascularization 3/39 (7.7%)einfarction 0/42 (0%)ajor adverse cardiac events* 11/46 (24%)

Values are numbers of patients (percentages).* Death, ischemic target vessel revascularization, and reinfarction.

able 5ndependent predictors of in-hospital mortality in multivariate logisticegression analysis (n � 2,479)

ariable p Value OddsRatio

95% CI

ight BBB 0.26 1.92 0.62–5.98eart rate �100 beats/min 0.031 1.86 1.06–3.25omen 0.0077 2.12 1.22–3.67illip’s class �I 0.0053 2.3 1.28–4.13eft BBB 0.0018 5.53 1.89–16.1ge �70 yrs 0.0005 2.71 1.55–4.73jection fraction �40% �0.0001 3.69 2.12–6.40inal TIMI flow grade �3 �0.0001 5.89 3.29–10.5

tCI � confidence interval; other abbreviation as in Table 2.

We found that the prognosis of patients who have AMInd right or left BBB on admission electrocardiogram re-ains poor compared with patients who do not have BBB,

espite emergency catheterization and primary angioplasty.oreover, in our database, patients who had left BBB hadorse prognosis than did patients who had right BBB. LeftBB was predictive of early mortality despite adjusting forther known predictors of death, including ejection fraction,ultivessel disease, and Thrombolysis In Myocardial In-

arction flow.Previous reports have suggested that the poor outcome in

atients who have left BBB may be due to a higher risklinical profile and less aggressive treatment offered tohem. In the National Registry of Myocardial Infarction,atients who had left BBB (6.7% of the population) had annadjusted risk of 22.6% for in-hospital death versus 13.1%n patients who did not have left BBB (p � 0.001).17,18 Inhis database, patients who had left BBB were less likely toeceive aspirin, � blockers, heparin, and intravenous nitro-lycerin. Patients who did not have left BBB were 4.2-foldore likely to receive thrombolytic therapy than were pa-

ients who had left BBB. In addition, patients who had leftBB and received thrombolytics had a longer median time

rom hospital arrival to initiation of treatment. Any initialeperfusion treatment (thrombolytics, PTCA, or coronaryrtery bypass grafting) was used less often in patients whoad left BBB (23.6%) than in patients who did not have leftBB (84.8%, p � 0.001). These data suggest that the poorrognosis of patients who had AMI and left BBB was dueot only to a more severe clinical profile but also to lessggressive treatment received by these patients.

In our database, patients who had left or right BBBeceived aspirin, heparin, and � blockers with the samerequency as did patients who did not have BBB. We foundo significant difference in door-to-balloon time, use ofTCA or stents, and Thrombolysis In Myocardial Infarctionow after the procedure. Nevertheless, the outcome of pa-

ients who had AMI and BBB remained poor. The poorrognosis of patients who had left BBB was independent ofaseline clinical characteristics or treatment received.

In the GUSTO-I database, patients who had left BBB1.6% of the population) had higher 30-day mortality rates

p Value

Right(n � 95)

None(n � 2,910)

12/79 (15%) 164/2,584 (6%) �0.00117/78 (22%) 388/2,552 (15%) 0.126/75 (8%) 116/2,480 (5%) 0.15

29/85 (34%) 597/2,707 (22%) 0.031

han did matched control subjects (18% vs 11%, odds ratio

1wmBmliiw

iofGoiRspBt06ld

tpwapgilBtBottmr(rG

Bpascp5tm

dh

1

1

1

1

1

1

1

487Coronary Artery Disease / Please supply wording

.8, p � 0.003) despite the use of thrombolytics.10 Patientsho had reversion of left BBB after thrombolysis had lowerortality rates than did patients who had persistent leftBB.10,26 This suggests that prompt reperfusion may restoreyocardial contractile and electrical function and, hence,

ead to better outcomes. We can speculate that similar find-ngs would be seen with primary angioplasty. However,nformation about persistence versus reversal of left BBBas not available in the PAMI database.The presence of right BBB or even intermediate QRS

nterval prolongation has also been associated with poorutcomes in patients who have AMI.27,28 In previous reportsrom the National Registry of Myocardial Infarction andUSTO-I databases, patients who had AMI and right BBBn admission electrocardiogram had a poor prognosis sim-lar to that in patients who had left BBB.10 In the Nationalegistry of Myocardial Infarction, right BBB was even a

tronger predictor of in-hospital death than left BBB inatients with AMI.17 In our study, patients who had rightBB had greater than twofold increase in in-hospital death

han did patients who had no BBB (7.4% vs 2.8%, p �.0001). This difference remained statistically significant atmonths and 1 year after randomization. However, unlike

eft BBB, right BBB was not an independent predictor ofeath on multivariate analysis.

Limited data exist with regard to the prognosis of pa-ients who have AMI and left BBB and are treated withrimary angioplasty. Moreno et al29 analyzed 945 patientsho had AMI and were treated with primary angioplasty atsingle institution in Spain from 1991 to 1999. Only 17

atients (1.8%) had left BBB on admission electrocardio-ram, and this was associated with a 3.7-fold increase inn-hospital mortality compared with patients who had noeft BBB (41% vs 11%, p � 0.002). Patients who had leftBB had a higher risk clinical profile that probably con-

ributed to the poor outcome, because the presence of leftBB was not found to be an independent predictor of deathn multivariate analysis.29 In addition, these patients werereated at a single center, so the patient population andreatment strategies might not represent the experience inost institutions worldwide. Their in-hospital mortality

ates for patients who had left BBB and those who did not41% and 11%) were much higher than the mortality rateeported in the National Registry of Myocardial Infarction,USTO, or PAMI databases.Our study is the first to report that the presence of left

BB on admission electrocardiogram is an independentredictor of in-hospital mortality in patients who have AMInd undergo primary angioplasty. Previous reports haveuggested that the cause of death is pump failure and not theonduction defect per se,5 and that the use of prophylacticacemakers has no added benefit.5,30 In our study, 3 of thecardiac deaths among patients who had left BBB were due

o cardiogenic shock and only 2 were due to lethal arrhyth-

ia. However, conclusions about the exact mechanism of

eath cannot be drawn because the sample of patients whoad LBBB was small, with only 7 deaths among them.16,19

1. Fazzini PF, Marchi F, Pucci P. Prognostic value of intraventricularconduction blocks in acute myocardial infarction. Acta Cardiol 1976;31:455–465.

2. Otterstad JE, Gundersen S, Anderssen N. Left anterior hemiblock inacute myocardial infarction. Incidence and clinical significance inrelation to the presence of bundle branch block and to the absence ofintraventricular conduction defects. Acta Med Scand 1978;203:529–534.

3. Woo KS, Norris RM. Bundle branch block after myocardial infarc-tion—a long term follow-up. Aust N Z J Med 1979;9:411–416.

4. Michalopoulos A, Athanasopoulos C, Lekakis J. The fate of patientswith intraventricular conduction disturbances complicating acute myo-cardial infarction. Acta Cardiol 1979;34:189–208.

5. Gaspard P, Delahaye JP, Touboul P, Milon H, Michelon G, Moussa M.Long-term course of patients with myocardial infarction and bundlebranch block. Arch Mal Coeur Vaiss 1980;73:1287–1297.

6. Dubois C, Pierard LA, Smeets JP, Foidart G, Legrand V, KulbertusHE. Short- and long-term prognostic importance of complete bundle-branch block complicating acute myocardial infarction. Clin Cardiol1988;11:292–296.

7. Hassi M, Kunstmann S, Corbalan R, Arriagada D, Casanegra P.Intraventricular conduction disorders in acute myocardial infarction:early and late clinical significance. Rev Med Chil 1989;117:1381–1386.

8. Hod H, Goldbourt U, Behar S. Bundle branch block in acute Q waveinferior wall myocardial infarction. A high risk subgroup of inferiormyocardial infarction patients. The SPRINT Study Group. SecondaryPrevention Reinfarction Israeli Nifedipine Trial. Eur Heart J 1995;16:471–477.

9. Laji K, Wilkinson P, Ranjadayalan K, Timmis AD. Prognosis in acutemyocardial infarction: comparison of patients with diagnostic andnondiagnostic electrocardiograms. Am Heart J 1995;130:705–710.

0. Sgarbossa EB, Pinski SL, Topol EJ, Califf RM, Barbagelata A, Good-man SG, Gates KB, Granger CB, Miller DP, Underwood DA, WagnerGS. Acute myocardial infarction and complete bundle branch block athospital admission: clinical characteristics and outcome in the throm-bolytic era. GUSTO-I Investigators. Global Utilization of Streptoki-nase and t-PA (tissue-type plasminogen activator) for Occluded Cor-onary Arteries. J Am Coll Cardiol 1998;31:105–110.

1. Hathaway WR, Peterson ED, Wagner GS, Granger CB, Zabel KM,Pieper KS, Clark KA, Woodlief LH, Califf RM. Prognostic signifi-cance of the initial electrocardiogram in patients with acute myocardialinfarction. GUSTO-I Investigators. Global Utilization of Streptokinaseand t-PA for Occluded Coronary Arteries. JAMA 1998;279:387–391.

2. Melgarejo Moreno A, Galcera Tomas J, Garcia Alberola A, GonzalezSanchez A, Jimenez Pagan F, Vignote Mingorance G, Galan Ayuso J,Rodriguez Garcia P. The incidence, clinical characteristics and prog-nostic significance of a left bundle-branch block associated with anacute myocardial infarct. Rev Esp Cardiol 1999;52:245–252.

3. Gunnarsson G, Eriksson P, Dellborg M. Bundle branch block andacute myocardial infarction. Treatment and outcome. Scand Cardio-vasc J 2000;34:575–579.

4. Melgarejo-Moreno A, Galcera-Tomas J, Garcia-Alberola A. Prognos-tic significance of bundle-branch block in acute myocardial infarction:the importance of location and time of appearance. Clin Cardiol2001;24:371–376.

5. Miller WL, Sgura FA, Kopecky SL, Asirvatham SJ, Williams BA,Wright RS, Reeder GS. Characteristics of presenting electrocardio-grams of acute myocardial infarction from a community-based popu-lation predict short- and long-term mortality. Am J Cardiol 2001;87:1045–1050.

6. Sgarbossa EB, Pinski SL, Barbagelata A, Underwood DA, Gates KB,

Topol EJ, Califf RM, Wagner GS. Electrocardiographic diagnosis of

1

1

1

2

2

2

2

2

2

2

2

2

2

3

488 The American Journal of Cardiology (www.AJConline.org)

evolving acute myocardial infarction in the presence of leftbundle-branch block. GUSTO-1 (Global Utilization of Streptokinaseand Tissue Plasminogen Activator for Occluded Coronary Arteries)Investigators. N Engl J Med 1996;334:481–487.

7. Go AS, Barron HV, Rundle AC, Ornato JP, Avins AL. Bundle-branchblock and in-hospital mortality in acute myocardial infarction. Na-tional Registry of Myocardial Infarction 2 Investigators. Ann InternMed 1998;129:690–697.

8. Barron HV, Bowlby LJ, Breen T, Rogers WJ, Canto JG, Zhang Y,Tiefenbrunn AJ, Weaver WD. Use of reperfusion therapy for acutemyocardial infarction in the United States: data from the NationalRegistry of Myocardial Infarction 2. Circulation 1998;97:1150–1156.

9. Shlipak MG, Go AS, Frederick PD, Malmgren J, Barron HV, CantoJG. Treatment and outcomes of left bundle-branch block patients withmyocardial infarction who present without chest pain. National Reg-istry of Myocardial Infarction 2 Investigators. J Am Coll Cardiol2000;36:706–712.

0. Stone GW, Marsalese D, Brodie BR, Griffin JJ, Donohue B, CostantiniC, Balestrini C, Wharton T, Esente P, Spain M, et al. A prospective,randomized evaluation of prophylactic intraaortic balloon counterpul-sation in high risk patients with acute myocardial infarction treatedwith primary angioplasty. Second Primary Angioplasty in MyocardialInfarction (PAMI-II) Trial Investigators. J Am Coll Cardiol 1997;29:1459–1467.

1. Grines CL, Marsalese DL, Brodie B, Griffin J, Donohue B, CostantiniCR, Balestrini C, Stone G, Wharton T, Esente P, et al. Safety andcost-effectiveness of early discharge after primary angioplasty in lowrisk patients with acute myocardial infarction. PAMI-II Investigators.Primary Angioplasty in Myocardial Infarction. J Am Coll Cardiol1998;31:967–972.

2. Grines CL, Cox DA, Stone GW, Garcia E, Mattos LA, Giambar-tolomei A, Brodie BR, Madonna O, Eijgelshoven M, Lansky AJ, et al.Coronary angioplasty with or without stent implantation for acutemyocardial infarction. Stent Primary Angioplasty in Myocardial In-

farction Study Group. N Engl J Med 1999;341:1949–1956.

3. Grines CL, Westerhausen DR Jr, Grines LL, Hanlon JT, LogemannTL, Niemela M, Weaver WD, Graham M, Boura J, O’Neill WW,Balestrini C. A randomized trial of transfer for primary angioplastyversus on-site thrombolysis in patients with high-risk myocardial in-farction: the Air Primary Angioplasty in Myocardial Infarction study.J Am Coll Cardiol 2002;39:1713–1719.

4. Wharton TP Jr, Grines LL, Turco MA, Johnston JD, Souther J, LewDC, Shaikh AZ, Bilnoski W, Singhi SK, Atay AE, et al. Primaryangioplasty in acute myocardial infarction at hospitals with no surgeryon-site (the PAMI-No SOS study) versus transfer to surgical centersfor primary angioplasty. J Am Coll Cardiol 2004;43:1943–1950.

5. Esente P, Kaplan AV, Ford JK, Martin JL, Ayres M, Kosinski EJ,Lasorda DM, Graham M, Gallant P, Grines LL, Grines CL. Localintramural heparin delivery during primary angioplasty for acute myo-cardial infarction: results of the Local PAMI Pilot Study. CatheterCardiovasc Interv 1999;47:237–242.

6. Sgarbossa EB, Pinski SL, Gates KB, Wagner GS. Predictors of in-hospital bundle branch block reversion after presenting with acutemyocardial infarction and bundle branch block. GUSTO-I Investiga-tors. Global Utilization of Streptokinase and t-PA for Occluded Cor-onary Arteries. Am J Cardiol 1998;82:373–374.

7. Ricou F, Nicod P, Gilpin E, Henning H, Ross J Jr. Influence of rightbundle branch block on short- and long-term survival after acuteanterior myocardial infarction. J Am Coll Cardiol 1991;17:858 –863.

8. Pudil R, Feinberg MS, Hod H, Boyko V, Mandelzweig L, Behar S.The prognostic significance of intermediate QRS prolongation in acutemyocardial infarction. Int J Cardiol 2001;78:233–239.

9. Moreno R, Garcia E, Lopez de Sa E, Abeytua M, Soriano J, Ortega A,Rubio R, Lopez-Sendon JL. Implications of left bundle branch blockin acute myocardial infarction treated with primary angioplasty. Am JCardiol 2002;90:401–403.

0. Alpman A, Guldal M, Erol C, Akgun G, Kervancioglu C, Sonel A,Akyol T. The role of arrhythmia and left ventricular dysfunction inpatients with acute myocardial infarction and bundle branch block. Jpn

Heart J 1993;34:145–157.