Early and long-term results of coronary artery bypass grafting with severely depressed left ventricular performance

Early and Long-Term Results of Coronary Artery Bypass Grafting with Severely Depressed Left Ventricular Performance ANTHONY P. FREEMAN, MB, FRACP, WARREN F. WALSH, MB, FRACP, ROBERT W. GILES, MD, DANIEL CHOY, MB, FRACP, DAVID C. NEWMAN, MB, FRACS, DAVID A. HORTON, MB, FRACS, JOHN S. WRIGHT, MB, FRACS, and I. PROVAN MURRAY, MD

The effects of coronary artery bypass grafting (CABG) on ventricular performance and long-term clinical status were studied in 18 consecutive patients with disabling angina pectoris and severely depressed left ventricular (LV) performance (ejection fraction [EF] 27 4- 9 % ). All patients survived CABG, although 1 patient had a perioperaUve myocardial infarction. There was no change in LVEF at rest, 29 -t- 12%, in the other 17 patients. However, LVEF during peak exercise increased from 22 -I- 7% to 27 -I- 14% (p <0.05). The 17 patients were separated into 2 groups: those who increased their peak exercise LVEF by at least 10% (group A, 8 patients) and those who increased it by less than 10 % (group B, 9 patients). Preoperatively, patients in group A had a higher LVEF at rest (p <0.001) and smaller end-systolic and end-diastolic volumes at rest (p <0.001) and during exercise (p <0.005).

Preoperatively, the LVEF in group A decreased with exercise, from 36 4- 4% to 27 4- 5% (p <0.01), but was unchanged in group B (19 -I- 3% vs 17 -I- 4%, difference not significant). After CABG, patients in group A had a smaller increase in end-systolic volume with exercise than those in group B (13 -I- 7 vs 34 4- 22 ml/m 2, p <0.05), but the changes in enddiastolic volume with exercise were not significantly different. At 27 -I- 5 months after CABG, 5 of 8 paUents in group A were asymptomatic. Of the 9 patients in group B, 2 had died and 5 had recurrence of angina or heart failure. In patients with depressed LV function who undergo CABG, improvement in LV performance and long-term clinical status is more likely to occur in patients with less severely impaired LV function at rest and evidence of exercise-induced ischemic dysfunction. (Am J Cardiol 1984;54:749-754)

Improvements in the anesthetic and surgical management of patients with left ventricular (LV) dysfunction have resulted in a steady decline in perioperative mortality and morbidity rates, permitting these patients to undergo coronary artery bypass grafting (CABG) with an acceptable riskJ '2 Most studies that have evaluated the effect of CABG on LV function and prognosis have been performed in patients with either normal or near-normal LV function at rest) -8 Consequently, little information is available about the effects of CABG in patients with severely depressed LV performance. 9,1°

Accordingly, a group of patients with marked depression of LV function who underwent CABG for relief of angina was studied prospectively.

Methods Patients: Eighteen consecutive patients were studied who were undergoing CABG for chronic stable angina (6 in New York Heart Association [NYHA] functional class III and 12 in class IV) and who had an angiographic LVEF of <40%. Patients who required additional LV aneurysm resection or cardiac valve surgery were excluded. The group included 15 men and 3 women, aged 58 + 6 years (range 50 to 69). All patients had at least I previous myocardial infarction, diagnosed by the triad of typical history, development of Q waves longer than 0.04 second on the electrocardiogram and elevated cardiac enzymes. The mean angiographic LVEF for the 18 patients was 28 + 8% (range 15 to 39%). When significant coronary arterial obstruction was defined as >50% diameter

From the Departments of Cardiovascular Medicine, Cardiothoracic Surgery and Nuclear Medicine, The Prince of Wales and Prince Henry Hospitals, Sydney,Australia. Manuscriptreceived December 30, 1983; revised manuscript received May 25, 1984, accepted June 5, 1984. Address for reprints: Anthony P. Freeman, MB, Department of Nuclear Medicine, The Prince of Wales Hospital, Randwick, Sydney, N.S.W. 2031, Australia. 749

750

CORONARY BYPASS GRAFTING WITH DEPRESSED VENTRICULAR FUNCTION

TABLE I

Radlonuclide Angiographic Data for the 17 Patients Preoperative

Pt CW JD JP AM LP MF BR LS BH JL ES MV NG FB PS GM KW Mean SD

R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex R Ex

Postoperative

LVEF (%)

ESVI (ml/m 2)

EDVI (ml/m 2)

40 30 37 20 35 25 32 32 17 17 38 24 29 24 21 16 18 18 23 22 40 30 35 34 22 15 22 23 14 10 17 13 21 22 27 ~9 22 • 7

85 140 37 88 69 93 98 113 150 164 83 153 52 61 199 258 160 186 96 119 59 91 56 75 155 183 159 180 238 322 171 237 178 196 120 ~60 155 ~69

141 202 59 110 107 124 144 167 181 197 133 202 73 80 251 306 195 226 125 153 98 131 86 114 199 216 203 207 276 357 205 274 226 251 159 ~64 195 ~74

LVEF (%) 42 43 41 33 35 43 43 44 19 11 41 34 34 34 14 11 21 21 21 21 44 43 44 47 17 10 24 22 14 10 14 11 21 14 29 ~12 27 ~14

ESVI (ml/m 2)

EDVI (ml/m 2)

70 65 32 48 57 68 27 42 165 209 56 89 61 81 154 156 151 181 272 272 35 40 21 33 146 184 123 166 176 204 147 150 103 142 106 ~69 125 ~71

121 114 51 72 87 119 48 75 204 235 95 133 92 122 179 175 192 230 344 343 62 71 37 62 176 204 161 213 205 227 171 169 131 165 139 ~78 160 ~75

EDVI = end-diastolic volume index; ESVI = end-systolic volume index; Ex = peak exercise; LVEF = left ventricular ejection fraction; R = rest; SD = standard deviation.

reduction, 12 patients had 3-vessel coronary artery disease (CAD) and 6 had 2-vessel CAD. Thirteen patients required long-acting nitrates such as isorbide dinitrate (45 =t=9 mg/day, range 30 to 120), 17 patients required ~-blocking drug therapy and 10 patients required furosemide (64 ~ 21 mg/day, range 40 to 160). Study protocol: Using standard techniques, gated equilibrium radionuclide angiography was performed at rest and during semisupine bicycleexercise in all patients 1 week before CABG and 3 months postoperatively (mean 99 ± 11 days). Using methods previously described, LVEF11,12 and enddiastolic and end-systolic volumes13at rest and peak exercise were measured. The left ventricle at rest was divided into 5 segments for qualitative regional wall motion analysis: anterolateral, apical, posterolateral, inferior and septal. Exercise wall motion was assessed in the segments seen in the left anterior oblique view only: septal, apicoinferior and posterolateral. Regional wall motion was assessed by 2 independent blinded observers without knowledge of the patient's clinical data. The patients were followed for 27 + 3 months (range 22 to 33). Details of their clinical course were obtained by personal examination, chart review and contact with the patient's own physician. Postoperative congestive heart failure was diagnosed by the presence of a third heart sound and the development of symptoms and signs of pulmonary congestion requiring diuretic therapy. All results are reported as mean + standard deviation. A 2-tailed paired or unpaired Student t test was used for analysis

of data. Results were considered significant if p <0.05. The study protocol was approved by the Prince Henry and Prince of Wales Hospital Ethics Committee. Results

Early clinical assessment: During CABG, a mean of 2.5 + 0.5 grafts per patient were performed and cold potassium cardioplegic arrest was used for myocardial protection in all cases. All 18 patients survived surgery. Postoperatively, 9 patients required parenteral inotropic therapy for at least 48 hours, and 4 of these 9 patients also received intraaortic balloon counterpulsation. One patient showed electrocardiographic evidence (new Q waves) of a perioperative myocardial infarction. Three months after CABG, all patients were free of angina. Thirteen patients (72%) were asymptomatic, 4 had dyspnea on moderate exertion and 1 patient became breathless during minimal activity. There was a marked reduction in the need for drug therapy, and no patient required either ~-blocking drugs or long-acting nitrates. The improved symptomatic status after 3 months was corroborated by measurement of exercise performance. Exercise capacity for the group improved from 40 ± 22 W preoperatively to 71 ± 30 W postoperatively (p

October 1, 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume54

781

with exercise before and after surgery in patients in both groups; however, these changes did not reach statistical significance for either group (Fig. 2). The preoperative status of groups A and B was examined to determine any differences. There were no significant differences between groups in anginal symptoms, drug therapy, severity of CAD or the number of vessels bypassed. The preoperative LV performance data for the groups are listed in Table II. Patients who improved their peak exercise LVEF by at least 10% (group A) had a higher LVEF at rest and during peak exercise (p <0.001), smaller end-systolic volumes both at rest (p <0.001) and during exercise (p <0.001), and smaller end-diastolic volumes at rest (p <0.001) and during exercise (p <0.005). The preoperative LVEF in patients in group A decreased with exercise by at least 10% EF units in 5 of the 8 patients and was unchanged

<0.001). Peak exercise double product also improved, from 176 + 52 × 1 0 2 to 255 4- 69 × 1 0 2 after CABG (p <0.001). All 18 patients were limited by angina in the preoperative exercise study, but at 3 months, 12 were limited by leg fatigue and 6 by dyspnea. Effect of s u r g e r y o n g l o b a l v e n t r i c u l a r f u n c t i o n : To assess more accurately the effect of CABG on LV performance, the patient who had a definite perioperative myocardial infarction was excluded from subsequent analysis of LV performance. This patient's EF at rest decreased from 40% preoperatively to 23% after CABG. In the interval between hospital discharge and the 3-month radionuclide study, no patient had a myocardial infarction or recurrence of angina that could have suggested early graft closure. The effect of CABG on LV performance for the remaining 17 patients is shown in Table I. The EF at rest for the patient group was 27 ± 9% preoperatively and 29 ± 12% postoperatively (difference not significant [NS]) (Fig. 1). There was no significant change in LV end-diastolic volume index at rest (159 + 64 vs 139 4- 78 ml/m 2) or end-systolic volume index (120 4- 60 vs 106 ± 69 ml/m 2) after CABG (Fig. 2, Table I). The postoperative LVEF at peak exercise was significantly higher than the preoperative value, 27 4- 14% vs 22 4- 7% (p <0.05), although LVEF decreased during exercise before and after surgery (Fig. 1). There was no significant change in LV end-diastolic volume index during peak exercise (195 + 75 vs 160 4- 75 ml/m 2) (Table I). Fourteen of the 17 patients (78%) showed a reduction in end-systolic volume index during peak exercise after CABG, although there was no significant change for the entire group (155 4- 69 vs 125 4- 71 ml/m2). The patients were separated into 2 groups for further analysis: those who improved their peak exercise LVEF after CABG by at least 10% EF units (group A, 8 patients) and those who increased it by less than 10% (group B, 9 patients) (Fig. 1). LV end-systolic volume index increased with exercise both before and after surgery in patients in group A, but the magnitude of increase in the end-systolic volume index was significantly smaller in the postoperative study (13 4- 7 vs 34 + 22 ml/m 2, p <0.05). In contrast, in patients in group B, the magnitude of the change in the end-systolic volume index with exercise was unchanged postoperatively (Fig. 2). The LV end-diastolic volume index increased

50-

. . . . Group A (n=8) • Group B (n=9) ,e

,';.1

oS s~

40LVEF 3O~o

2010-

0

I

I

I

Rest Preop

Peak Ex Peak Ex Postop Preop

FIGURE 1. Left venbicular ejection fraction (LVEF) in patients in groups A and B, measured at rest preoperatively (Preop) and at peak exercise (Peak Ex) both pre- and postoperatively (Postop). Before surgery, patients in group A had a greater exercise-induced decrease in LVEF than those in group B.

r - ' l Preop V/)J Postop 70. FIGURE 2. Changes in end-systolic volume index (ESVI) and end-diastolic volume index (EDVI) during exercise in groups A and B. The magnitude of change of end-systolic volume index for group A is significantly smaller after surgery. There was no difference in the exercise response of end-diastolic volume index for groups A and B pre- and postoperatively.

• p < 0.05

70. 60.

60' 50.

z~ E S V I 4o.

~, EDVI 40.

m l / m 2 30. 20"

m l / m 2 30" 20-

10" 0

1(> 0 A

B

A

B

752

CORONARYBYPASS C-~AFTING WITH DEPRESSED VENTRICULAR FUNCTION

TABLE II

PreoperaUve Left Ventrlcular Performance at Rest and During Exercise Group A

LVEF (%)

R Ex A R Ex %A R Ex %A

ESVI (ml/m 2) EDVI (ml/m 2)

36 27 --8 67 102 55 105 141 36

4- 4 -I- 5 -I- 6 4- 20 4- 32 -I- 39 4- 32 4- 45 -I- 25

Group B

p Value

19 4- 3 17 -1- 4 --2 -I- 3 167 -I- 38 205 4- 59 21 4- 11 207 4- 43 243 -t- 62 16 -4- 10

<0.001 <0.001 <0.05 <0.001 <0.001 <0.05 <0.001 <0.005 NS

% A = change; NS = not significant; other abbreviations as in Table I.

in 3 (36 4- 4% to 27 4- 5%, p <0.05). Patients in group B showed no significant change in LVEF with exercise preoperatively (19 4- 3% vs 17 4- 4%) (Fig. 1). Effects of s u r g e r y on regional v e n t r i c u l a r function: Twenty-five of 40 segments (62%) in patients in group A showed abnormal motion; 20 of these segments were hypokinetic and 5 were akinetic. In contrast, 41 of 45 segments (91%) in patients in group B had abnormal motion; 23 of these segments were hypokinetic, 16 were akinetic and 2 were dyskinetic. In the 17 patients analyzed, 12 of the 66 abnormal segments (18%) (7 hypokinetic and 5 akinetic) showed improved motion after CABG. Eight of the 12 segments (67%) that showed improved motion at rest were in patients in group A. Of the remaining 54 abnormal segments, regional wall motion was unchanged in 45. Nine segments showed worsening of wall motion; 6 (65%) were in patients in group B. Seven of the 9 segments that showed worsening of wall motion involved the septum. Preoperatively, worsening of wall motion during exercise occurred in 11 of 40 segments (28%) in group A and 2 of 45 (4%) in group B. After CABG, 10 of the 11 segments in group A and both segments in group B showed improvement in segmental wall motion with exercise. Long-term clinical follow-up: At long-term follow-up, 27 4- 3 months, there were 16 survivors. Two patients had died (LP and KW), 1 at 11 months and 1

NYHA

GROUPA

GROUPB

(n=8)

~=9)

3 H0. POSTOP

LON6 TERH F/U

FC

NYHA

3 HO, POSTOP

at 13 months after severe heart failure developed. Of the 16 survivors, 8 patients (including the patient who had a perioperative infarction) had clinical evidence of heart failure that required increased diuretic therapy (furosemide, 120 4- 146 mg/day, range 40 to 480). In 5 of these 8 patients, angina recurred and the patient required nitrate therapy. The long-term outcome for groups A and B is shown in Figure 3. Seven of the 8 patients in group A remained in NYHA functional class I or II, and only 1 of t h e s e patients required diuretic therapy. The eighth patient (MF) remained symptomatically unchanged after CABG, with functional class III dyspnea, although he remained free of angina. Patients who failed to increase their peak exercise LVEF by at least 10% after surgery (group B) had a less favorable long-term course than group A. Three of the 9 patients in group B had recurrent angina and increasing dyspnea. Four remained free of angina but showed clinical signs of increasing heart failure, including the 2 patients who died. Discussion

The patient group in this study had evidence preoperatively of severe CAD and marked LV dysfunction. Nevertheless, no patient in this series died during CABG, reflecting improvements in anesthetic techniques and the use of cold potassium cardioplegia for better myocardial protection. 1 All patients had relief from angina and showed substantial symptomatic benefit 3 months postoperatively. This excellent early clinical result is in agreement with other short-term studies of patients with angina who have depressed LV performance who undergo CABG. Effect on v e n t r i c u l a r performance: Most studies of patients with preoperative normal or mildly abnormal LV function have not shown improvement in LV performance at rest after CABG.3-8 In our study of patients with depressed LV performance, the LVEF at rest was similarly unchanged 3 months after surgery. Conflicting results have been reported in the few studies9,1°,14,15that have examined the effects of CABG on LV function at

LONG TERH F/U

FC

,

FI

,

FIGURE 3. Long-term follow-up (F/U) for groups A and B according to New York Heart Association (NYHA) functional class (FC). All patients showed early postoperative (POSTOP) clinical improvement, but only in group A was this sustained long term. Two patients in group B

died.

I,,

D

,v

[]

1

17]

,,I

[]

,v

2+

October 1, 1984 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 54

rest in patients with moderate or advanced LV dysfunction. This disagreement suggests that measures of global LV performance at rest may not be sufficiently sensitive to detect the effects of CABG because they do not allow for the assessment of cardiac performance reserve obtained with exercise. Despite lack of change in global LV function at rest after surgery, some improvement in regional myocardial function as assessed by segmental wall motion was demonstrated. Twelve of 66 segments (18%) with abnormal wall motion preoperatively showed an improvement in motion after CABG. This study has shown that LVEF at peak exercise tends to increase after CABG in patients with severely depressed LV performance. The mechanism of the improvement in peak exercise LVEF can be better understood when the changes in LV volumes in patients who increased their peak LVEF by at least 10% (group A) are examined. There was an increase in end-systolic and end-diastolic volumes with exercise, but in patients in group A the magnitude of the increase in the endsystolic volume was smaller after CABG. Because the changes in end-diastolic volume with exercise were not altered, postoperative peak exercise LVEF increased. Although groups A and B were not clinically different, their preoperative LV performance was significantly different. Patients in group A had smaller LV volumes and higher LVEFs at rest, indicating that cardiac function was less severely impaired in this subgroup. The marked reduction in LVEF with exercise in patients in group A suggests that these patients had more potentially ischemic myocardium and, hence, relatively less myocardial scarring than patients in group B. Patients in group A therefore showed a greater degree of exercise-induced ischemic dysfunction than would be possible in patients in group B, in whom there would likely be a larger amount of myocardial scar tissue. This explanation is consistent with the exercise data provided by Port et al, 16 who reported that the magnitude of decrease in EF with exercise in patients with CAD was greater in patients with only a moderate reduction in EF than in those with more severely depressed LV performance. Therefore, patients with depressed LV function who show a marked ischemic response to exercise, suggesting larger amounts of jeopardized myocardium, might be expected to improve their LV function during exercise after CABG to a much greater degree than patients who had substantial amounts of myocardial scar and, hence, less potential for improving LV wall motion. Postoperative radionuclide studies using either rubidium-8617 or thallium-20118,19 have documented an increase in myocardial perfusion to previously ischemic areas after CABG both at rest and during exercise. Late clinical results: Little information is available regarding the long-term follow-up of patients with marked ischemic LV dysfunction who undergo CABG. Earlier studies in which the survival of surgically treated patients was compared with that of medically treated patients were unduly influenced by the high operative mortality rates. 2°-2~ More recently, 3 long-term studies have examined survival in these groups of patients.

753

Faulkner et a124reported a low operative mortality rate and a 2-year survival rate of 83% in patients followed for a mean of 19 months. Coles et a125followed 59 surgically treated patients with an average preoperative LVEF of 28%. Their 5-year actuarial survival rate was 80 ~: 6%. These results are in agreement with our own findings: an 89% survival rate after a mean follow-up of 27 months. Alderman et al26 reported the results of CABG in patients with poor LV function identified from the registry of the Coronary Artery Surgery Study. In this study, long-term survival was compared in 420 medically treated and 231 surgically treated patients followed for as long as 7 years. The 5-year survival rate for the medically treated group was 54% and for the surgically treated group 68%. When these investigators examined the clinical results, they showed that most patients who had CABG for disabling angina had symptomatic benefit, whereas symptoms caused primarily by heart failure were usually not relieved. Despite the initial marked subjective clinical improvement in our patients, there was significant attrition with time. After more than 2 years of follow-up, 8 of the 18 patients had clinical evidence of heart failure that required diuretic therapy, and 5 of these 8 also had recurrence of angina. Most of the poor long-term clinical results, including the 2 late deaths, were in patients who failed to increase their peak exercise EF significantly after CABG (group B). As previously discussed, this group had more advanced LV dysfunction at rest preoperatively. Clinical implications: Patients with disabling angina who have depressed LV function may undergo CABG with a low risk and short-term symptomatic benefit. Patients who demonstrate substantial ischemic LV dysfunction with exercise preoperatively usually show a significant improvement in exercise LV performance postoperatively and have a better long-term prognosis. Our findings indicate that patients with the poorest LV function at rest (LVEF <25%), and in whom there is little change with exercise preoperatively long term, have both a high likelihood of having heart failure and a poorer prognosis despite relief of ischemic pain.

Acknowledgment: The authors gratefully acknowledge the secretarial assistance of Leah Crawleyand Linda Dixon and the technical assistance of Jenny Dixon. References 1. Rahlmtoola SH, Grunkemelr GL, Teply JF, Lambert LE, Thomas DR, Suen YF, Starr A. Changes in coronary bypass surgery leading to improved survival. JAMA 1981;246:1912-1916. 2. Kennedy JW Kaiser GC, Fisher LD, Fritz JK, Myers W, Mudd JG, Ryan TJ. Climcal and angiographic pred ctors of operative mortality from the Collaborative Study in Coronary Artery Surgery (CASS). Circulation 1981;63:793-802. 3. Barry WH, Pfelfer JF, Lipton MJ, "rllklen AG, Hultgren HN. Effects of coronary artery bypass grafting on resting and exercise hemodynamics in patients with stable angina pectoris: prospective, randomized study. Am J Cardiol 1976;37:823-830. 4. Kent KM, Borer JS, Green MV, Bacharach SL, Mclntosh CL, Conkle DM, Epstein SE. Effects of coronary-artery bypass on global and regional left ventricular function during exercise. N Engl J Med 1978;298:14341439. 5. Bussman WD, Mayer V, Kober G, Kaltenbach M. Ventricular function at rest, during leg raisingand physical exercise before and after aorto-coronary bypass surgery. Am J Cardiol 1979;43:488-493. 6. Freeman MR, Gray RJ, Berman DS, Maddahl J, Raymond MJ, Forrester JS, MaUoffJM. improvementin globaland segmentalleft ventricularfunction after coronary bypass surgery. Circulation 1981;64:supp111:11-34-11-39.

754

CORONARYBYPASS GRAFTING WITH DEPRESSED VENTRICULAR FUNCTION

7. Wolf NM, Krouh)n TH, Bore AA, McDonongh MT, Kessler KM, Strong M, Le Mole G, Spann JF. Left vantricular function following coronary bypass surgery. Circulation 1977;58:63-70. 8. Um YL.,Kalff V, Kelly MJ, Mason PJ, Cun'h) PJ, Harper RW, Anderson ST, Federman J, SUrllng GR, Plff A. Radlonuclide angiographlc assessment of global and segmental left ventricular function at rest and during exercise after coronary artery bypass graft surgery. Circulation 1982;66:972979. 9. Hellman CK, Schmldt DH, Kamath ML, Anholm J, Blau F, Johnson WD. Bypass graft surgery in severe left vantricular dysfunction. Circulation 1980;62:suppl 1:1-103-1-110. 10. Hung J, Kelly DT, Baird DK, Hendel PN, Leckle BD, Grant AF, Uron RF. Aorto-coronary bypass grafting in patients with severe left ventricular dysfunction. J Thorac Cardiovasc Surg 1980;79:718-723. 11, Freeman AP, Giles RW, Mcllveen BM, Murray IPC. Radionuclide ejection fraction: comparison of response to treadmill and bicycle exercise. Eur J Nucl Med 1982;7:393-396. 12. Freeman AP, GIh)s RW, Berdoukas VA, Walsh WF, Choy D, Murray IPC. Early left ventrlcular dysfunction and chelation therapy In thalassaemla major. Ann Intern IVied 1983;99:450-454. 13. Dehmer G, Lewis S, HIIIIs L, Twlng D, Folokoff M, Parkey R, WIIh)rson J. Non-geometric determination of left vantricular volumes from equilibrium blood pool scans. Am J Cardiol 1980;45:293-300. 14. ChatterJae K, Swan HJC, Parmley WW, Sustalta H, Marcus HS, Matloff J. Influence of direct myocardial revascuiarization on left ventricular asynargy and function in patients with coronary heart disease. Circulation 1973;47:276-286. 15. Hammermelnter KE, Kennedy JW, Hamilton GW, Stewart DK, Gould KL, Llpscomb K, Murray JA. Aortocoronary saphanous-vein bypass. Failure of successful grafting to improve resting left ventricular function in chronic angina. N Engl J Med 1974;290:186-192. 16. Pod S, McEwan P, Cobb FR, Jones RH. flnfluanca of resting left ventrlcular function on the left ventricularresponseto exercise in patientswith coronary artery disease. Circulation 1981;63:856-863. 17. Lurle AJ, Salel AF, Berman DS, De Nardo GL, Hurley EJ, Mason DT. De-

16.

19.

20.

21. 22. 23. 24. 25.

26.

termination of improved myocardial perfusion after aortocoronary bypass surgery by exercise rubidium-81 scintigraphy. Circulation 1976;54:suppl Iih111-20-111-23. RHchh) JL, Naharora KA, Trobangh GB, Williams DL, Hamilton GW. Thallium-201 myocardial imaging before and after coronary revascularizatlon; assessment of regional myocardial blood flow and graft patency. Circulation 1977;56:830-836. Plfnterer M, Emmenegger H, Schmlff HE, Muller-Brand J, H a m J, Gradel E, Laver MB, Burckhardt D, Burkart F. Accuracy of serial myocardial perfusion scintigraphywith thallium-201 for prediction of graft patency early and late after coronary artery bypass surgery. Circulation 1982;66: 1017-1024. Yatteau RF, Peter BH, Behar VS, Barrel AG, Rosotl RA, Kong Y. Ischemic cardiomyopathy. The myopathy of coronary artery disease. Natural history and results of medical versus surgical treatment. Am J Cardlol 1974;34: 520-525. Mlfchel BF, AIIvlzatos PA, Adam M, Gelsler GF, Thleh) JP, Lambert CJ. Myocardial revascularization in patients with poor ventricular function. J Thorac Cardiovasc Surg 1975;69:52-62. Manh)y JC, King JF, Zeft HJ, Jolmson W1D.The "bad" left venVich). Results of coronary surgery and effect on late survival. J Thorac Cardlovasc Surg 1976;72:841-848. Spencer FC, Green GE, "lice DA, Wallsh E, Mills NL, Glsssman E. Coronary artery bypass grafts for congestive heart failure. A report of experiences with 40 patients. J Thorac Cardlovasc Surg 1971;62:529-542. Faulkner L, Stoney WS, Alferd WC, Thomas CS, Bunus GR, Frlst RA, Page HL. Ischemic cardlomyopathy:medical versus surgical treatment. J Thorac Cardiovasc Surg 1977;74:77-82. Cotes JG, Del Campo C, Ahmed SN, Corpus R, MacDonald AC, Goldbach MM, Coles JC. Improved long term survival following myocardial revascularization in patients with severe left vantricular dysfunction. J Thorac Cardiovasc Surg 1981;81:846-850. Alderman EL, Fisher LD, LHwln P, Kalsor GL, Myers WO, Maynard C, Levlne F, Schloss M. Results of cort)nary artery surgery in patients with poor left vantricular function (CASS). Circulation 1983;68:785-795.