Improved long-term survival following myocardial revascularization in patients with severe left ventricular dysfunction

J

THoRAc CARDIOVASC SURG

81:846-850, 1981

Improved long-term survival following myocardial revascularization in patients with severe left ventricular dysfunction The natural history a/patients with coronary artery disease associated with poor left ventricular (LV) function is dismal. This report analyzes the efficacy of myocardial revascularization in this subset of patients with coronary artery disease manifesting severe LV dysfunction on the basis of LV angiography. LV ejection fraction (LVEF). and left ventricular end-diastolic pressure (LVEDP). For the 2 '12 year period ending November. 1977. 59 consecutive patients with coronary artery disease complicated by severe LV dysfunction underwent aorta-coronary bypass at the University of Western Ontario. All patients had angina refractory to medical therapy. Objective criteria for compromised LV function included the presence of three or more dysfunctional (hypokinetic of akinetic) segments on biplane LV angiography. Eighty-three percent (49/59) of patients had triple-vessel coronary artery disease. The mean LVEF for the series was 0.28 and the mean LVEDP was 18 mm Hg. The duration offollow-up was 24 to 60 months (mean 37 months). with follow-up survival data available on 100% of patients. The hospital mortality was 1.7% (1/59). and there were nine late deaths. The 5 year actuarial survival rate (±SEM) was 80% ± 6%. Of the 44 long-term survivors available for direct assessment. 98% (43/44) report improvement with respect to angina and 66% (29/44) are totally asymptomatic. Eighty percent (28/35) o] the long-term survivors under the age of 65 years are currently employed. These results indicate that myocardial revascularization can be performed in patients with severe ischemic LV dysfunction at very low risk and . further, that operation results in a dramatic improvement in survival expectations compared with optimal medical therapy.

1. G. Coles, M.D.,* C. Del Campo, M.D., F.R.C.S. (C),* S. N. Ahmed, M.D.,* R. Corpus, M.D., F.R.C.S. (C),* A. C. MacDonald, M.D., F.R.C.P. (C),** M. M. Goldbach, M.D., F.R.C.S. (C),*** and J. C. Coles, M.D., F.R.C.S. (C),**** London. Ontario. Canada

T

he role of myocardial revascularization in patients with symptomatic coronary artery disease associated with left ventricular (L V) dysfunction (ischemic cardiomyopathy) is not yet clearly defined. Although re-

From the Division of Cardiovascular and Thoracic Surgery. Victoria Hospital, University of Western Ontario, London, Ontario. Canada. Received for publication July 8. 1980. Accepted for publication Nov. 12, 1980. Address for reprints: Dr. John G. Coles. III Waterloo St., Suite 306, London. Ontario. Canada. N6B 2M4. *Resident. Cardiovascular and Thoracic Surgery. University of Western ontario. **Chairman. Department of Radiology, Victoria Hospital. ***Clinical Associate Professor of Surgery. University of Western Ontario. ****Clinical Professor of Surgery, University of Western Ontario; Chairman, Division of Cardiovascular and Thoracic Surgery.

846

cent reports have presented excellent results following revascularization in this group of patients;': 2 other series have reported lesss favorable":" and even contradictory results. 9 - 13 Much of this controversy stems from lack of uniform criteria for defining significant L V dysfunction. Moreover, selection of patients with true ischemic cardiomyopathy manifest by angina, rather than those with LV dysfunction resulting from irreversible myocardial fibrosis, will have a dramatic effect on both the surgical and late mortality.': 3. 5. 9 Patients whose L V dysfunction is the result of ischemic cardiomyopathy have a uniformly poor prognosis when managed by medical therapy exclusively. The most recent and comprehensive analysis of survival in medically treated coronary artery disease indicated a 5 year survival rate of 89% in patients with normal LV function; this figure dropped to 70% in patients with "moderately abnormal" L V contraction and to only

0022-5223/81/060846+05$00.50/0 © 1981 The C. V. Mosby Co.

Volume 61

Left ventricular dysfunction

Number 6 June. 1961

38% in patients with "diffusely abnormal" LV contraction.':' During the period of this study, the policy at our institution was to accept all patients for myocardial revascularization regardless of the status of LV function, provided that severe angina was the predominant symptom. This report analyzes the clinical course of 59 consecutive patients with severe LV dysfunction with angina pectoris who underwent coronary artery bypass grafting.

Patients and methods The clinical data of 59 consecutive patients undergoing myocardial revascularization for the 2 1/ 2 year period ending in November, 1977, were analyzed during a follow-up period of 24 to 60 months postoperatively (mean 37 months). There were 54 men and five women in the series with a mean age of 54 years at operation. All patients had severe angina that was not adequately controlled by conventional medical therapy, and none was operated upon for symptoms of congestive heart failure (CHF) alone. CHF, defined as the presence of major dyspnea on exertion, with or without orthopnea and/or pulmonary edema, was present in 20% (12159) of patients preoperatively. No patient was refused operation because of poor LV function. Patients with ventricular aneurysm and those in whom myocardial revascularization was combined with other cardiac procedures were excluded from the study. The entire series represented 13% (59/477) of all coronary bypass operations performed at our hospital during the period of study. Cardiac catheterization. Segmental wall motion was evaluated angiographically in all patients in both the right (RAO) and left anterior oblique projections (Fig. 1). The left ventriculograms were assessed by an independent cardiac radiologist as demonstrating severe LV dysfunction and were categorized according to the American Heart Association classification. 15 Fiftyone patients were designated as having Grade 4 (Grading I through 5) LV function, in which three or more of seven segments were hypokinetic or akinetic, and eight patients were designated as having Grade 5 function, in which five or more of seven segments were hypokinetic or akinetic. Ejection fraction (L VEF), calculated from the 30 degree RAO ventriculogram according to the method of Dodge and associates;" was available in 55 of the 59 patients. The mean LVEF (± 95 % CL *) for the series was 0.28% (±0.19), with 76% of patients having an *Confidence limits.

847

LAO

RAO

Fig. 1. Left ventricular angiogram in the right anterior oblique (RAO) and left anterior oblique (LAO) projections indicating segmental analysis used for designation of LV function. Grade 4: Three or more out of seven segments significantly hypokinetic or akinetic. Grade 5: Five or more out of sevensegments significantly hypokinetic or akinetic. 1. Basal (anterior). 2. Anterior. 3. Apical. 4. Inferior. 5. Basal (inferior). 6. Septal. 7. Posterobasal. LVEF of 0.35 or less. The mean left ventricular enddiastolic pressure (L VEDP) (±95% CL), determined prior to the injection of contrast medium, was 18 mm Hg (± 15 mm Hg). All patients had evidence of severe coronary obstructive disease, with 83% (49/59) having triple-vessel disease and all remaining patients having double-vessel disease. Operative technique. Propranolol therapy, although frequently tapered in dosage, was continued until less than 24 hours prior to the induction of anesthesia. In all patients, an emphasis on protective anesthesia involved careful control of heart rate and blood pressure by appropriate pharmacological means prior to the institution of cardiopulmonary bypass (CPB). Standard techniques of CPB were employed with hemodilution prime and moderate systemic hypothermia (27 to 31 C). Although intra-aortic balloon pump support was not used preoperatively, it was found to be essential to successfully wean four of the patients from CPB. Distal coronary anastomoses were performed first by means of a continuous suture technique with 6-0 polypropylene suture during periods of intermittent ischemic arrest. In the latter part of the study, cold (4 C) cardioplegic arrest, with a hyperkalemic (30 mEq/L) asanguineous solution introduced into the aortic root, provided myocardial protection during a single period of aortic cross-clamping. In all patients an attempt was made to achieve complete revascularization, as reflected in the average of 2.8 grafts per patient. Right coronary artery endarterectomy in conjunction with bypass was required in 19% (11/59) patients in order to revascularize a dominant right coronary artery. Operative mortality was defined as any death occur0

0

0

848

The Journal of Thoracic and Cardiovascular Surgery

Coles et al.

100

g

s.

80

>

60

.

40

.. II: .~

::l

III

.. :; ~

E

::l (.)

20

2

3

4

5

Years

Fig. 2. The 5 year actuarial survival rate including all deaths in the University of Western Ontariosurgical series was 80% ± 6%.

(± SEM)

ring during the immediate 30 day postoperative period. Actuarial analysis of survival from the time of operation was determined according to the method of Cutter and Ederer, I7 with inclusion of both cardiac and noncardiac deaths. More than 90% of patients were directly assessed in follow-up by the resident or consultant staff in the Cardiovascular Unit at Victoria Hospital, and the survival status of all of the remaining patients was ascertained by telephone contact with the patient or referring physician. No patient was lost to follow-up.

Results The operative mortality was 1.7% (1159), with the single operative death occurring 18 days postoperatively as a result of low cardiac output and sepsis. Of the nine late deaths occurring 3 to 37 months postoperatively, seven were cardiac in origin, occurring in association with myocardial infarction (five patients) or as a sudden unexplained event (two patients). The 5 year actuarial survival rate (± SEM) was 80% ± 6% (Fig. 2). This rate is contrasted with the findings of the recently published series from Duke University, which reported on survival of 1,214 medically treated patients with coronary artery disease, analyzed according to the angiographic status of L V function (Fig. 3).14 In this study, the 5 year cumulative survival rates (excluding noncardiac deaths) for patients with normal LV function, "moderately abnormal" LV function, and "diffusely abnormal" LV function were 89%, 70%, and 38%, respectively. Although the Duke series does not represent a strictly comparable group of patients, this analysis indicates the striking reduction in survival rate that occurs with increasing severity of LV dysfunction in medically treated patients. This adverse effect of LV dysfunction

on survival is clearly attenuated in the surgically treated patients with a comparable degree of LV dysfunction, whose survival curve approximates that for medically treated patients with normal LV function. Among the 44 long-term survivors available for direct assessment, 98% (43/44) report improvement in angina and 66% (29/44) are completely asymptomatic. Of the 12 patients significantly limited by exertional dyspnea preoperatively and designated as having CHF (New York Heart Association Class III or IV), nine are in significantly improved condition (more than one functional class), two are in slightly improved condition, and one patient experienced no improvement in symptoms. There were no operative or late deaths in this subgroup of patients. Eighty percent (28/35) of the long-term survivors under the age of 65 years are currently employed.

Discussion Angiographic analysis of LV segmental wall motion, as recommended by the American Heart Association, provides standard criteria for objectively designating the status of LV function. This method allows a more complete characterization of LV function than LVEF or L VEDP alone, which are indices reflecting global L V function. In this series, the presence of three dysfunctional segments or more out of seven segments on biplane LV angiography was considered indicative of severe LV dysfunction. The prognostic significance of ischemic LV dysfunction based on angiographic findings is well established.t'"?" Thus, in a retrospective analysis by Parker, 19 the 5 year survival rate for patients with coronary artery disease associated with LV asynergy was only 58%. Even the most favorable survival data available for patients with medically treated coronary artery disease':' (Fig. 3) are not substantially different from the widely quoted natural history study by Bruschke, Proudfit, and Sones.!" In the latter study, the 5 year cardiac survival rate for patients with localized impairment of L V contractility was 66%, and that for patients with diffuse impairment of LV contractility was only 29%. Nelson, Cohn, and Gorlin demonstrated that LVEF is an important predictor of short-term survival in patients with coronary artery disease treated medically, and in their study an LVEF of less than 0.3 was associated with a 42% mortality over a 14 month period. Moreover, in several surgical series,": II. 13. 23. 24 the L VEF was found to be an important determinant of both operative and late mortality. Hammermeister and Kennedy 13 concluded that myocardial revascularization

Volume 81 Number 6 June, 1981

Left ventricular dysfunction

849

100 UWO Surgical Series

~ 80 B

Duke Medical Series normal LV function moderate LV dysfunction severe LV dysfunction

•

II:

Ii 60 >

.

's

l>-----6

<>--0

:::s

en II

>

• :;

40

~

E :::s

u

20

1

2

3

4

5

Years

Fig. 3. The 5 year actuarial survival rate (excluding noncardiac deaths) comparing surgical results (UWO series) with medical results (Duke series), with medical survival analyzed according to the status of left ventricular (LV) function. The Duke data indicate a 5 year survival rate of 89o/c in patients with normal LV function, 70% in patients with' 'moderately abnormal" LV function, and 38% in patients with' 'diffusely abnormal" LV function. The 5 year cumulative survival rate in the UWO surgical series is 86%.

should not be offered to patients with an L VEF of 33% or less because of the high (33%) operative mortality. Likewise, L VEDP is an independent determinant of survival in medically treated coronary artery disease. 10. 14. 19-21 In the Duke study, there was a significant correlation between abnormal L VEDP (17 mm Hg or higher) and decreased survival rate, particularly in patients with triple-vessel disease. 14 A corresponding increase in early and late surgical mortality has been reported in patients with elevation of LVEDP.:l. 5.10. II. I:l. 25. 26 The reduction in L VEF (mean 0.28) and elevation in L VEDP (mean 18.1 mm Hg) observed at catheterization is a reflection of the ischemic L V dysfunction that characterized our patients. Although the L VEF was depressed in all patients, the L VEDP was only mildly elevated (12 to 18 mm Hg) in some patients, which is indicative of regional rather than global abnormalities in L V contraction. t , 27 The role of myocardial revascularization for ischemic L V dysfunction has been debated by various authors, who have reported a wide range of results with regard to both early and late survival after surgical treatment. 1-13 As suggested by previous authors, 1.3. 5. 9 we believe that the surgical survival is critically dependent upon the selection of patients in whom severe angina is the predominant symptom, indicating the presence of ischemic but potentially viable myocardium, and the exclusion of patients presenting solely with CHF secondary to irreversible myocardial fibrosis. With this criterion used for selection of patients, the 5

year survival rate in this surgical series (80%) approximates that for medically treated patients with normal L V function. Thus it appears that myocardial revascularization abrogates the adverse effect of LV dysfunction on survival observed in patients treated exclusi vely by medical therapy. The prohibitively high surgical risk previously reported 3 - 1:l was not observed in this series, for the operative mortality (1.7%) approaches that for bypass grafting in patients with normal LV function. It is probable that a better understanding of intraoperative factors influencing the myocardial oxygen supply/demand ratio-" as well as improvements in myocardial protection rendered by cardioplegic arrest account for the reduction in operative risk in this category of patients. Because of the encouraging long-term results in this retrospective analysis, we continue to offer an aggressive surgical approach to patients with angina pectoris complicated by severe L V dysfunction. We greatly appreciate the assistance of Mrs. T. Gerster in the preparation of the manuscript. REFERENCES Ellis LJ, Craver JM, Kaplan JA, Kine BK, Douglas JS, Morgan EA, Hatcher CR: Criteria for operability and reduction of surgical mortality in patients with severe left ventricular ischemia and dysfunction. Ann Thorac Surg 25:413, 1978 2 Faulkner SL, Stoney WS, Alford WC, Thomas CS, Burrus GR, Frist RA, Page HL: Ischemic cardiomyopathy. Medical versus surgical treatment. J THORAC CARDIOVASC SURG 74:77, 1977

The Journal of

850

Coles et al.

3 Isom OW, Spencer FC, Glassman E, et al: Long-term survival following coronary bypass surgery in patients with significant impairment of left ventricular function. Circulation 51,52:Suppl 1:141, 1975 4 Mundth ED, Harthorne JW, Buckley MJ, et al: Direct coronary arterial revascularization. Arch Surg 103:529, 1971 5 Manley JC, King JF, Zeft HJ, Johnson WD: The "bad" left ventricle. J THoRAc CARDIOVASC SURG 72:841, 1976 6 Kay JH, Redington JV, Mendez AM, et al: Coronary artery surgery for the patient with impaired left ventricular function. Circulation 45,46:Suppl 2:49, 1972 7 Manley JC, Johnson WD: Effects of surgery on angina (pre- and postinfarction) and myocardial function (failure). Circulation 46: 1208, 1972 8 Siderys H, Pittman IN, Herod G: Coronary artery surgery in patients with impaired left ventricular function. Chest 61:482, 1972 9 Spencer FC, Green GE, Tice DA, Wallsh E, Mills NL, Glassman E: Coronary artery bypass grafts for congestive heart failure. J THoRAc CARDIOVASC SURG 62:529, 1971 10 Yatteau RF, Peter RH, Behar VS, et al: Ischemic cardiomyopathy. The myopathy of coronary artery disease. Am J Cardiol 34:520, 1974 II Lea RE, Tector AJ, Flemma RJ. et al: Prognostic significance of a reduced left ventricular ejection fraction in coronary artery surgery. Circulation 45,46:Suppl 2:49, 1972 12 Bourassa MG: The bad left ventricle, Cleve C1in Q 45:179, 1978 13 Hammermeister KE, Kennedy JW: Predictors of surgical mortality in patients undergoing direct myocardial revascularization. Circulation 49,50:Suppl 2: 112, 1974 14 Harris PJ, Harrel FE, Lee KL, Behar VS. Rosati RA: Survival in medically treated coronary artery disease. Circulation 60: 1259, 1979 15 Austen WG, Edwards JE, Frye RL, Gersine GG, Gott VL, Griffith LSC, McGoon DC, Murphy ML, Roe BB: A reporting system on patients evaluated for cornary artery disease. Report of the ad hoc committee for grading of

Thoracic and Cardiovascular Surgery

16

17 18

19

20

21 22 23

24

25 26 27

28

coronary artery disease, Council on Cardiovascular Surgery, American Heart Association, Circulation 51:Suppl 4:30, 1975 Dodge HT, Sandler H, Ballew DW, Lord 10 Jr: Use of biplane angiocardiography for the measurement of left ventricular volume in man. Am Heart J 60:762, 1960 Cutter SJ, Ederer T: Maximum utilization of the life-table method of analyzing survival. J Chron Dis 8:699, 1958 Bruschke A V, Proudfit WL, Sones FM: Progress study of 590 consecutive nonsurgical cases of coronary disease followed 5-9 years. II. Ventriculographic and other correlations. Circulation 47: 1154, 1973 Parker JO: Prognosis in coronary artery disease. Arteriographic , ventriculographic and hemodynamic factors, The First Decade of Bypass Graft Surgery for Coronary Artery Disease, An International Symposium, Cleveland, Ohio, 1977 Proudfit WL, Bruschke A VG, Sones FM: Natural history of obstructive coronary artery disease. Ten year study of 601 non-surgical cases. Prog Cardiovasc Dis 21:53, 1978 Nelson GR, Cohn PF, Gorlin R: Prognosis in medicallytreated coronary artery disease. Circulation 52:408, 1975 Reeves TJ, Oberman A, Jones WB, Sheffield LT: Natural history of angina pectoris. Am J Cardiol 33:423, 1974 Cohn PF, Gorlin R, Cohn LH, et al: Left ventricular ejection fraction as a prognostic guide in surgical treatment of coronary and valvular heart disease. Am J Cardiol 34:136, 1974 Collins 11, Cohn LH, Sonnenblick EH, et al: Determinants of survival after coronary artery bypass surgery. Circulation 47,48:Suppl 3: 132, 1973 Oldham HN, Kong Y, Bartel AG, et al: Risk factors in coronary artery bypass surgery. Arch Surg 105:918, 1972 Hall RJ. Dawson JT. Cooley DA, et al: Coronary artery bypass. Circulation 47,48:Suppl 3:146, 1973 Cohn PF, Herman MV, Gorlin R: Ventricular dysfunction in coronary artery disease. Am J Cardiol 33:307, 1974 Brazier J, Cooper N, Buckberg G: The adequacy of subendocardial oxygen delivery. Circulation 49:968, 1974