LGDMAaD

tlgc

ttdswAl

Dtibf

TgAhvofpCppclpam

i

A

AtE

©P

CA

RD

IOV

ASC

ULA

R

ong-Term Results of Coronary Artery Bypassrafting in Patients With Left Ventricularysfunction

ohamed A. Soliman Hamad, MD, M. Erwin S. H. Tan, MD, PhD,lbert H. M. van Straten, MD, André A. J. van Zundert, MD, PhD,nd Jacques P. A. M. Schönberger, MD, PhD

epartments of Cardiothoracic Surgery and Anaesthesiology, Catharina Hospital, Eindhoven, the Netherlands

pTsfatpcf0

ttlnm

Background. In this prospective study, we investigatedhe determinants of long-term outcome, symptoms, andeft ventricular function after coronary artery bypassrafting in patients with a moderate to severely de-reased left ventricular ejection fraction.

Methods. Between 1997 and 1998, 75 consecutive pa-ients with moderate to severe left ventricular dysfunc-ion underwent coronary artery bypass grafting proce-ures. The operative mortality rate was 4.0%, and the 72urvivors were monitored for 8 years. The end pointsere mortality, symptomatic status (New York Heartssociation [NYHA] functional class), and left ventricu-

ar function.Results. The total survival rate after 8 years was 89.3%.uring follow-up, 8 patients died. Death was attributed

o a cardiac cause in 5 patients and to a noncardiac causen 3. There was no statistically significant differenceetween preoperative and late postoperative NYHA

unctional class, despite a statistically significant im-

raeigpmdIotrc

P

BwCwoRb

horacic Surgery, Catharina Hospital, Michelangelolaan 2, Postbus 1350,indhoven, 5602 ZA, the Netherlands; e-mail: aasmsn@cze.nl.

2008 by The Society of Thoracic Surgeonsublished by Elsevier Inc

rovement that persisted for up to 4 years after CABG.he results of echocardiography showed a statisticallyignificant improvement in the left ventricular ejectionraction (from 0.322 � 0.06 preoperatively to 0.463 � 0.02t follow-up, p < 0.001). Multivariate analysis revealedhat the left ventricular end-systolic volume index, theresence of angina pectoris, and absence of symptoms ofongestive heart failure were preoperative indicators ofreedom from heart failure after coronary operations (p <.05).Conclusions. Coronary artery bypass grafting for pa-

ients with moderate-to-severe left ventricular dysfunc-ion is associated with acceptable long-term results. Theeft ventricular end-systolic volume index is a simpleoninvasive method to aid in the preoperative decisionaking in such patients.

(Ann Thorac Surg 2008;85:488 –93)

© 2008 by The Society of Thoracic Surgeons

he number of patients with advanced left ventriculardysfunction who undergo coronary artery bypass

rafting (CABG) has increased in the past few years [1].lthough the recovery of impaired myocardial functionas been shown to occur after CABG, contractile leftentricular dysfunction remains a negative determinantf postoperative outcome [2]. Impaired left ventricularunction, which has been shown to be an independentredictor of operative mortality in patients who undergoABG [3], can also lead to low cardiac output and a highostoperative mortality rate [4]. Many patients with im-aired left ventricular function require inotropic or me-hanical support for hours to days after surgery [4]. In theast decade, an improvement in surgical techniques anderioperative management has led to a better postoper-tive course and decreased postoperative mortality andorbidity rates in this high-risk group of patients [5, 6].Although surgical revascularization may be successful

n the short term, little is known about the long-term

ccepted for publication Sept 10, 2007.

ddress correspondence to Dr Soliman Hamad, Department of Cardio-

esults of successful CABG in patients with coronaryrtery disease and left ventricular dysfunction [7]. How-ver, certain high-risk subgroups of patients remain atncreased risk for complications and death despite pro-ressive advances in perioperative care and myocardialrotection [8]. We prospectively studied 75 patients withoderate to severe left ventricular dysfunction who un-

erwent primary elective CABG between 1997 and 1999.mmediate and short-term results have shown the safetyf CABG in this group of patients [9]. Our study presentshe long-term outcome in this group of patients withegard to clinical status and predictors of freedom fromongestive heart failure (CHF).

atients and Methods

etween July 1997 and March 1999, 75 consecutive patientsith impaired left ventricular function who underwentABG in Catharina Hospital, Eindhoven, the Netherlands,ere enrolled in this prospective study of long-termutcome. This study was approved by the Medico-Ethicaleview Committee, and an informed consent was signed

y every patient before enrollment. The inclusion criteria

0003-4975/08/$34.00doi:10.1016/j.athoracsur.2007.09.010

wfcos

tpps

PAthTipdHhse

SAmfvvcmwpma

F(

T

P

AMN

P

P

P

L

LRDHC

CtLd

TS

C

I

8

T

E

CVIRP

489Ann Thorac Surg SOLIMAN HAMAD ET AL2008;85:488–93 CABG AND LEFT VENTRICULAR DYSFUNCTION

CA

RD

IOV

ASC

ULA

R

ere the presence of left ventricular dysfunction (ejectionraction [EF] � 0.40), isolated CABG, and no acute myo-ardial infarction during the month that preceded theperation. Exclusion criteria were prior cardiogenichock, emergency surgery, off-pump CABG, implanta-

able 1. Preoperative Characteristics of Patients

atients (n) 75

ge, y (mean � SD) 64 � 13ale/female ratio 22/3YHA functional class, No.I 15II 13III 19IV 28

reoperative mitral regurgitation, No.No 46Trivial 22Mild 7

reoperative Q-wave MI (�30 days),No.

46

rimary indication for operation, No.Heart failure 18Angina 34Both 10Ventricular arrhythmia, No. 2Critical anatomy, No. 11

VESVI (mL/m2)Mean � SD (range) 86.09 � 18.9 (61.6–134.2)Median 78.4

VEF, mean � SD 0.322 � 0.06epeat CABG, No. 8iabetes, No. 19emodialysis, No. 4OPD, No. 6

ABG � coronary artery bypass grafting; COPD � chronic obstruc-ive pulmonary disease; LVEF � left ventricular ejection fraction;VESVI � left ventricular end-systolic volume index; MI � myocar-ial infarction; NYHA � New York Heart Association.

able 2. Causes of Postoperative Death in the Studyubjects

ause of Death Patients, No.

mmediate mortality rate (hospitalized patients)Low cardiac output 1Myocardial infarction 1Mediastinitis 1

-year mortality rateCardiac cause

Sudden death 2Massive pulmonary embolism 1

Noncardiac causeCerebral hemorrhage 1Cancer 1

Total 8 Ci

ion of an intraaortic balloon pump, combined valverocedures, or a left ventricular aneurysmectomy. Thereoperative characteristics of the patients studied areummarized in Table 1.

reoperative Assessmentll patients with ventricular dysfunction (EF � 0.40 on

ransthoracic-transesophageal echocardiography) andibernating myocardium were studied preoperatively.he presence of this preserved myocyte functional activ-

ty was estimated by a transesophageal echocardiogra-hy stress test using dobutamine infusion according toosages and infusion rates as previously described [10].ibernating myocardium was diagnosed if the patientad 5 or more reversible segments of the 16 myocardialegments [8], and 61 patients (81.3%) had preoperativevidence of hibernating myocardium.

urgical Techniquesll patients underwent isolated CABG using cardiopul-onary bypass. A similar operative technique was per-

ormed throughout the study. After sternotomy and har-esting of the left internal mammary artery, saphenousein, or in selected cases, the radial artery, or both,ardiopulmonary bypass was initiated with normother-ic perfusion. Cold St. Thomas crystalloid cardioplegiaas intermittently given through the aortic root. Com-lete revascularization was achieved in all patients. Theean cardiopulmonary bypass time was 74 � 26 minutes,

nd the mean ischemia time was 55 � 18 minutes.

ig 1. Total survival (dotted line) and freedom from heart failuresolid line) at the 4- and 8-year follow-up. (HF � heart failure.)

able 3. Cardiac Events at Follow-Up

vent Patients, No. Percentage

HF symptoms 24 32entricular arrhythmia 2 2.7

nfarction 5 6.7eoperation 0 0CI 3 4

HF � congestive heart failure; PCI � percutaneous coronaryntervention.

PTsdp

he

STdwv

uvadfettacs1

R

Oddrpl

Lm

Fllm

Fe0

Ft(0

Fo

490 SOLIMAN HAMAD ET AL Ann Thorac SurgCABG AND LEFT VENTRICULAR DYSFUNCTION 2008;85:488–93

CA

RD

IOV

ASC

ULA

R

ostoperative Follow-Uphe mean follow-up period was 8.1 � 0.6 years. Allurvivors were clinically monitored in our outpatientepartment and underwent transthoracic echocardiogra-hy at 4 and 8 years postoperatively.The clinical diagnosis of CHF was made if the patient

ad the following criteria (the first two criteria aressential):

● History of exertional dyspnea or orthopnea, orboth.

● Clinical examination findings of bilateral basalcrepitation on auscultation, bilateral pitting edemaof the legs, and congested neck veins, with orwithout hepatomegaly.

● Clinical response to medical therapy.

tatistical Analysishe values are presented as the mean � the standardeviation. Linear mixed models for repeated measuresere performed to predict the postoperative trend of left

entricular function. The two-sample Student t test was

ig 2. Freedom from heart failure in patients with a preoperativeeft ventricular end-systolic volume index (LVESVI) of 100 mL/m2 oress (dashed line) and in those with an LVESVI exceeding 100L/m2 (solid line). (*p � 0.015; **p � 0.001; ***p � 0.001.)

ig 3. Long-term freedom from heart failure in patients with preop-rative angina (dashed line) and those without (solid line). (*p �

.01; **p � 0.001; ***p � 0.001.) a

sed to test the impact of the individual preoperativeariables on long-term CHF-free survival. Multivariatenalysis was applied to determine the independent pre-ictors of CHF-free survival. We have considered the

ollowing variables in the multivariate analysis: the pres-nce of hibernating myocardium, presence of preopera-ive angina, the left ventricular ejection fraction (LVEF),he left ventricular end-systolic volume index (LVESVI),nd the New York Heart Association (NYHA) functionallass. Values of p � 0.05 were considered statisticallyignificant. Statistical analysis was performed with SPSS5.0 software (SPSS Inc, Chicago, IL).

esults

f the 75 patients enrolled in the study, 3 patients (4%)ied within 30 days of the operation. Five patients (6.7%)ied during follow-up, and the long-term total survivalate was 89.3%. The causes of postoperative deaths areresented in Table 2. Only 1 patient, who emigrated, was

ost to follow-up.There was a statistically significant improvement in the

VEF during the short-term follow-up, defined as 6onths after the date of surgery, from 0.32.2% � 6.0%

ig 4. Long-term freedom from heart failure in patients at preopera-ive New York Heart Association (NYHA) functional class III/IVdashed line) and those at NYHA I/II (solid line). (*p � 0.18; **p �.001; ***p � 0.012.)

ig 5. Long-term freedom from heart failure in patients with an pre-perative ejection fraction (EF) exceeding 0.30 (solid line) and those

t 0.30 or less (dashed line) (*p � 0.17; **p � 0.12; ***p � 0.34).

tldd0oep

ppwific(ac0C

otcsaendiodo

sLo(cami

C

Ibcnhspppe

ASPpvceqp[wcrnme

aouhrup

vtatotc“tpet

PDifpEs7

TS

V

NALEP

a

Cide

491Ann Thorac Surg SOLIMAN HAMAD ET AL2008;85:488–93 CABG AND LEFT VENTRICULAR DYSFUNCTION

CA

RD

IOV

ASC

ULA

R

o 44.0% � 4.0% (p � 0.0001). However, the results ofong-term follow-up showed that the postoperative LVEFid not differ with statistical significance from the imme-iate postoperative value (mean EF, 0.463 � 0.02, p �.12). There was no echocardiographic evidence of post-perative development of mitral regurgitation or wors-ning of preexisting mild mitral regurgitation in anyatient.Freedom from CHF was found in 74.7% � 6% of the

atients 4 years after surgery and in 61.3% � 8% of theatients 8 years after surgery (Fig 1). When comparedith the preoperative value, no statistically significant

mprovement in NYHA class was noted during long-termollow-up. Four years after CABG, the number of patientsn NYHA class IV had decreased, with statistical signifi-ance, from a preoperative rate of 37.7% (n � 28) to 24%n � 18) of the total population (p � 0.001). Eight yearsfter the procedure, the number of patients in NYHAlass IV significantly increased to 24 patients (32%; p �.02). All of these patients had preoperative symptoms ofHF.Nine patients (12%) who exhibited an immediate post-

perative arrhythmia responded to medical therapy, withhe exception of 1 patient who received an implantableardiac defibrillator within 2 weeks postoperatively. Aecond patient required the implantation of an implant-ble cardiac defibrillator 3 years later because of anpisode of ventricular tachycardia. Postoperative coro-ary angiography was performed in 14 patients (18.7%)ue to recurrence of symptoms. Percutaneous coronary

ntervention was performed in 3 patients because of anccluded venous graft. No repeat CABG was performeduring the follow-up period. The cardiac events thatccurred during long-term follow-up are in Table 3.Multivariate regression analysis revealed a statistically

ignificant correlation between CHF-free survival and anVESVI of 100 mL/m2 or less (p � 0.001; Fig 2), symptomsf angina (p � 0.001; Fig 3), and preoperative NYHA classp � 0.01; Fig 4). There was no statistically significantorrelation between CHF-free survival and the preoper-tive value of the EF (Fig 5) or the presence of hibernatingyocardium. The results of the multivariate analysis are

able 4. Multivariate Predictors of Heart Failure-Freeurvival at 8 Years

ariable Coefficient SE OR 95% CI p Value

YHA class 0.78 0.36 1.2 2.12–6.67 0.03a

ngina 0.64 0.19 2.3 1.03–1.42 0.03a

VESVI 0.32 0.15 0.998 0.997–0.999 0.02a

F 0.004 0.003 0.996 0.991–1.011 0.096resence ofHM

0.794 0.590 0.995 0.990–0.132 0.077

Statistically significant.

I � confidence interval; EF � ejection fraction; HM � hibernat-ng myocardium; LVESVI � left ventricular end-systolic volume in-ex; NYHA � New York Heart Association; SE � standardrror; OR � odds ratio.

n Table 4. t

omment

n this study, we showed that CABG in patients affectedy depressed left ventricular function (LVEF � 0.40)onfers a favorable but still not optimal long-term prog-osis. We observed that surgical revascularization en-ances left ventricular recovery and ensures long-termurvival. Recurrent symptoms of CHF were not com-letely prevented, however; they developed primarily inatients with a limited functional capacity, those withoutreoperative symptoms of angina, and those with anxtremely dilated ventricle (eg, LVESVI � 100 mL/m2).

ssessment of Hibernating Myocardium and Patientelectionredicting the likelihood of myocardial recovery and theostoperative prognosis in patients with depressed leftentricular function who underwent CABG remains ahallenge [11]. Noninvasive tests that establish the pres-nce and extent of hibernating myocardium are a re-uirement for accurate patient selection and a favorableostoperative result, but those tests are still being refined

12]. In our study, dobutamine stress echocardiographyas performed preoperatively in all patients by the same

ardiologist, and thallium-201 rest-redistribution scintig-aphy was performed in 43 (57%) patients. Both tech-iques are valuable in the identification of hibernatingyocardium and in the prediction of a favorable postop-

rative outcome [13].We confirmed in earlier research [9] that the preoper-

tive presence of hibernating myocardium is a predictorf immediate postoperative improvement of left ventric-lar function in these patients. However, we found thatibernating myocardium had no predictive value withespect to long-term outcome. Freedom from heart fail-re was not statistically significantly correlated with thereoperative presence of hibernating myocardium.The predictive role of the preoperative assessment of a

iable myocardium has been confirmed by several au-hors [8, 14, 15]. Kron, commenting on the study by Channd colleagues [8], noted that the problem of postopera-ive deterioration lies in patients who have no evidencef ischemia and hence do not benefit from revasculariza-ion. Elefteriades and colleagues [16] suggested that myo-ardial viability should be considered as a continuumfrom full thickness viability without any scar to fullhickness scarring without viable cells.” To confirm this,reoperative viability studies (dobutamine stress-chocardiography or thallium) should be correlated withhe histologic results of biopsied cardiac tissue.

atient Survivalata on the long-term results of CABG in patients with

mpaired left ventricular function are few [13, 17]; there-ore, we conducted this prospective study of selectedatients with compromised left ventricular function.ight years after CABG, 89.3% of the patients in oureries were alive. CHF-free survival after CABG was4.7% at 4 years and 61.3% at 8 years (an outcome better

han that of previous series) [7]. However, those studies

icfiwshobml

RoOvwttcmanlpaw

mtttansasetdat

PICptsstssm

taqmi

ic

bsmsTtivTpsilssoEstht

fbprWcaTp

R

492 SOLIMAN HAMAD ET AL Ann Thorac SurgCABG AND LEFT VENTRICULAR DYSFUNCTION 2008;85:488–93

CA

RD

IOV

ASC

ULA

R

ncluded patients without established hibernating myo-ardium at the preoperative assessment [7]. A surprisingnding was that the presence of hibernating myocardiumas not a statistically significant predictor of long-term

urvival in our patients. Hibernating myocardium mayave more greatly affected the immediate postoperativeutcome than the long-term outcome, a theory supportedy the fact that an immediate postoperative improve-ent in left ventricular function indicators was not fol-

owed by a further improvement in late follow-up.

ecovery of Left Ventricular Function and Recurrencef Heart Failureur data showed a substantial improvement in left

entricular function after revascularization. However,hen we compared the recovery of left ventricular func-

ion in long-term follow-up with immediate postopera-ive improvement, we noted no statistically significanthange. Shivalkar and colleagues [18] showed that theyocardial structure of the hypocontractile segments hascritical role in functional recovery. The extent of hiber-ating myocardium appears to be a positive predictor of

eft ventricular functional improvement after CABG. Theresence of a well-developed coronary collateral net haslso been shown to predict the resumption of segmentalall motion [19].Revascularization reduces postoperative mortality andorbidity in these patients but does not protect against

he progression of ischemic cardiomyopathy in some ofhem. The time-limited beneficial effects of revasculariza-ion in patients with left ventricular dysfunction havelso been shown by Luciani and colleagues [20], whooted that 47% of the patients studied were free of theymptoms of heart failure 5 years after CABG, despite anbsolute survival rate of 75% for that time period. In theireries, only 48% of the patients had undergone a preop-rative assessment of myocardial viability; this renderedhe need for CABG in patients with irreversible myocar-ial damage likely. Further homogenous clinical seriesre needed to further elucidate the optimal patient selec-ion for the treatment of ischemic cardiomyopathy.

redictors of Late Outcomen our series, preoperative symptoms of ischemia andHF appeared to be the most statistically significantredictors of long-term outcome. Our finding supports

he observation of Lansman and colleagues [21], whotated that the presence of angina predicted a goodurvival rate and a consistent improvement in left ven-ricular function. However, other investigators [13] havehown that the presence of angina had no statisticallyignificant predictive value with regard to long-termortality and the recurrence of heart failure.Dreyfus and colleagues [22] emphasized the impor-

ance of using viability tests instead of clinical evaluations a crucial determinant of patient selection and subse-uent postoperative outcome. In our series, hibernatingyocardium was a statistically significant predictor of

mmediate postoperative recovery, but the preoperative

schemia class and NYHA functional class were statisti-ally significant predictors of long-term outcome.

The LVESVI proved to be an important predictor ofoth short- and long-term functional improvement in oureries. In patients who had a preoperative LVESVI of 100L/m2 or less, we observed a favorable outcome for

urvival and the recovery of left ventricular function.hat finding supports the conclusion of other investiga-

ors [23, 24] who noted that the LVESVI was the onlyndependent measurement that was predictive of sur-ival and the improvement of left ventricular function.he LVESVI is a simple and noninvasive preoperativeredictor of postoperative functional recovery. It can beolely used in the decision making in patients withschemic cardiomyopathy; however, further studies witharger groups of patients are needed to confirm thistatement. On the other hand, the EF value was not atatistically significant predictor of long-term outcome inur series. Like other authors [25], we observed that theF was not a strong predictor of outcome in patients withevere left ventricular dysfunction. That theory rests onhe fact that a low EF did not differentiate betweenibernation and infarction as the cause of poor contrac-

ile function.In conclusion, we found that revascularization con-

erred a low postoperative mortality rate and a potentialenefit to left ventricular function in patients with im-aired left ventricular function. The long-term survivalate in our series was better than that in previous series.

e found that long-term freedom from heart failureould be predicted by the preoperative presence ofngina and a preoperative LVESVI of 100 mL/m2 or less.he presence of preoperative NYHA class III or IV was aredictor of a less favorable long-term outcome.

eferences

1. Ascione R, Narayan P, Rogers CA, Lim KH, Capoun R,Angelini GD. Early and midterm clinical outcome in patientswith severe left ventricular dysfunction undergoing coro-nary artery surgery. Ann Thorac Surg 2003;76:793–9.

2. Milano CA, White WD, Smith LR, et al. Coronary arterybypass in patients with severely depressed ventricular func-tion. Ann Thorac Surg 1993;56:487–93.

3. Ferguson TB Jr, Hammill BG, Peterson ED, DeLong ER,Grover FL; STS National Database Committee. A decade ofchange–risk profiles and outcomes for isolated coronaryartery bypass grafting procedures, 1990–1999: a report fromthe STS National Database Committee and the Duke Clini-cal Research Institute. Society of Thoracic Surgeons. AnnThorac Surg 2002;73:480–489; discussion 489–90.

4. Yau TM, Fedak PW, Weisel RD, Teng C, Ivanov J. Predictorsof operative risk for coronary bypass operations in patientswith left ventricular dysfunction. J Thorac Cardiovasc Surg1999;118:1006–13.

5. Davierwala PM, Maganti M, Yau TM. Decreasing signifi-cance of left ventricular dysfunction and reoperative surgeryin predicting coronary artery bypass grafting-associatedmortality: a twelve-year study. J Thorac Cardiovasc Surg2003;126:1335–44.

6. Cimochowski GE, Harostock MD, Foldes PJ. Minimal oper-ative mortality in patients undergoing coronary artery by-

pass with significant left ventricular dysfunction by maximi-

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

493Ann Thorac Surg SOLIMAN HAMAD ET AL2008;85:488–93 CABG AND LEFT VENTRICULAR DYSFUNCTION

CA

RD

IOV

ASC

ULA

R

zation of metabolic and mechanical support. J ThoracCardiovasc Surg 1997;113:655–64; discussion 664–6.

7. Appoo J, Norris C, Merali S, et al. Long-term outcome ofisolated coronary artery bypass surgery in patients withsevere left ventricular dysfunction. Circulation 2004;110:II13–II27.

8. Chan RK, Raman J, Lee KJ, et al. Prediction of outcome afterrevascularization in patients with poor left ventricular func-tion. Ann Thorac Surg 1996;61:1428–34.

9. Soliman Hamad MA, Peels K, Van Straten A, Van Zundert A,Schonberger J. Coronary artery bypass surgery in patientswith impaired left ventricular function. Predictors of hospitaloutcome. Acta Anaesthesiol Belg 2007;58:37–44.

0. La Canna G, Alfieri O, Giubbini R, Gargano M, Ferrari R,Visioli O. Echocardiography during infusion of dobutaminefor identification of reversibly dysfunction in patients withchronic coronary artery disease. J Am Coll Cardiol 1994;23:617–26.

1. Dilsizian V, Bonow RO. Current diagnostic techniques ofassessing myocardial viability in patients with hibernatingand stunned myocardium. Circulation 1993;87:1–20; erra-tum, Circulation 1993;87:2070.

2. La Canna G, Rahimtoola SH, Visioli O, et al. Sensitivity,specificity, and predictive accuracies of non-invasive tests,singly and in combination, for diagnosis of hibernatingmyocardium. Eur Heart J 2000;21:1358–67.

3. Lorusso R, La Canna G, Ceconi C, et al. Long-term results ofcoronary artery bypass grafting procedure in the presence ofleft ventricular dysfunction and hibernating myocardium.Eur J Cardiothorac Surg 2001;20:937–48.

4. Elefteriades JA, Tolis G Jr, Levi E, Mills LK, Zaret BL.Coronary artery bypass grafting in severe left ventriculardysfunction: excellent survival with improved ejection frac-tion and functional state. J Am Coll Cardiol 1993;22:1411–7.

5. Tan J, Kejriwal N, Vasudevan A, Maria PL, Alvarez JM.Coronary bypass surgery for patients with chronic poorpreoperative left ventricular function (EF � 30%): 5-yearfollow-up. Heart Lung Circ 2006;15:130–6.

6. Samady H, Elefteriades JA, Abbott BG, Mattera JA, McPher-

son CA, Wackers FJ. Failure to improve left ventricular

function after coronary revascularization for ischemiccardiomyopathy is not associated with worse outcome.Circulation 1999;100:1298–304.

7. Gunning MG, Chua TP, Harrington D, et al. Hibernatingmyocardium: clinical and functional response to revasculari-sation. Eur J Cardiothorac Surg 1997;11:1105–12.

8. Shivalkar B, Maes A, Borgers M, et al. Only hibernatingmyocardium invariably shows early recovery after coronaryrevascularization. Circulation 1996;94:308–15.

9. Kozman H, Cook JR, Wiseman AH, Dann RH, EngelmanRM. Presence of angiographic coronary collaterals predictsmyocardial recovery after coronary bypass surgery in pa-tients with severe left ventricular dysfunction. Circulation1998;98:II57–61.

0. Luciani GB, Montalbano G, Casali G, Mazzucco A. Predict-ing long-term functional results after myocardial revascular-ization in ischemic cardiomyopathy. J Thorac CardiovascSurg 2000;120:478–89.

1. Lansman SL, Cohen M, Galla JD, et al. Coronary bypass withejection fraction of 0.20 or less using centigrade cardioplegia:long-term follow-up. Ann Thorac Surg 1993;56:480–485; dis-cussion 485–6.

2. Dreyfus GD, Duboc D, Blasco A, et al. Myocardial viabilityassessment in ischemic cardiomyopathy: benefits of coro-nary revascularization. Ann Thorac Surg 1994;57:1402–7; dis-cussion 1407–8.

3. Hamer AW, Takayama M, Abraham KA, et al. End-systolicvolume and long-term survival after coronary artery bypassgraft surgery in patients with impaired left ventricular func-tion. Circulation 1994;90:2899–904.

4. Yamaguchi A, Ino T, Adachi H, Mizuhara A, Murata S,Kamio H. Left ventricular end-systolic volume index inpatients with ischemic cardiomyopathy predicts postopera-tive ventricular function. Ann Thorac Surg 1995;60:1059–62.

5. Langenburg SE, Buchanan SA, Blackbourne LH, et al. Pre-dicting survival after coronary revascularization for ischemiccardiomyopathy. Ann Thorac Surg 1995;60:1193–6; discus-

sion 1196–7.