Improved survival after coronary artery surgery in patients with extensive coronary artery disease

Improved survival after coronary artery surgery in patients with extensive coronary artery disease Survival in patients with ischemic heart disease is closely related to the extent of coronary artery obstruction as determined angiographically, One hundred forty-nine consecutive patients underwent coronary artery bypass surgery from November, 1971, to October, 1974. There were 2 late cardiac deaths, 1 late noncardiac death, and 1 hospital death, an operative mortality rate of 0.7 per cent and a total mortality rate of 2.7 per cent. Coronary angiograms were scored according to the method of Friesinger, Page, and Ross.' Fifty-four per cent (80/149) had scores of 10 or greater. Cumulative survival was analyzed according to life-table techniques; in the 80 surgically managed patients with scores of 10 or greater, survival at 3 years was 98 per cent. Friesinger's 47 nonoperated patients with similar angiographic scores had a 3 year cumulative survival of 68 per cent. Although this study compares different groups, the surgical series was composed of older patients (mean age 52 as compared to 44 years), includes 22 patients operated on urgently for preinfarction angina pectoris, and includes 18 patients with abnormal ventricular function. These data suggest that coronary artery bypass surgery can favorably influence prognosis in patients with severe coronary artery disease.

Daniel J. Ullyot, M.D.* (by invitation), Judith Wisneski, M.D. (by invitation), Robert W. Sullivan, M.D. (by invitation), and Edward W. Gertz, M.D. (by invitation), San Francisco, Calif. Sponsored by Benson B. Roe, M.D., San Francisco, Calif.

The single most powerful predictor of survival in patients with ischemic heart disease seems to be the extent of coronary artery obstruction determined angiographically. Numerous studies have shown a close correlation between the degree of coronary occlusive disease and late death in patients followed without surgical intervention after selective coronary angiography.v-" To examine the influence of coronary artery bypass on survival, we compared From the Departments of Surgery and Medicine, School of Medicine, University of California, the Veterans Administration Hospital, and the Public Health Service Hospital, San Francisco, Calif. Read at the Fifty-fifth Annual Meeting of The American Association for Thoracic Surgery, New York, N. Y., April 14, 15, and 16, 1975. "Address for reprints: Daniel J. Ullyot, M.D., Department of Surgery, University of California San Francisco, San Francisco, Calif. 94143.

a series of surgically treated patients with a published series- 2 of medically treated patients with similar extent of coronary artery disease demonstrated angiographically. Methods

One hundred forty-nine patients underwent coronary artery bypass between November, 1971, and October, 1974. Patients having bypass procedures combined with other cardiac procedures such as valve replacement, closure of ventricular septal defect, or ventricular aneurysmectomy during this period were excluded. All patients had selective coronary cineangiography in multiple projections and left ventriculography preoperatively. Angiographic scores according to the method of Friesinger, Page, and Ross' (Table I) were 405

The Journal of

406

Ullyot et ai.

Thoracic and Cardiovascular Surgery

Table I. Scoring system

for

coronary

cineangiograms" Score

Abnormality

o

None Trivial irregularities Single narrowing of 50-90% Multiple narrowings of 50-90% Narrowings greater than 90% Total obstruction of vessel

I 2 3 4 5

• A complete description of their system is provided by Friesinger, Page, and Ross.!

determined by two observers independently. Systolic ejection fractions by volume were determined in the 30 degree right anterior oblique projection according to the method of Greene and associates." In our laboratory, the normal value for systolic ejection fraction by volume is 66 per cent ± 6 per cent (S.D.). Clinical indications for surgery were stable, disabling angina pectoris in 42 per cent (63/149), crescendo angina in 29 per cent (43/149), and preinfarction angina . pectoris (defined as severe, prolonged chest pain at rest in patients admitted to the coronary care unit to rule out acute myocardial infarction) in 28 per cent (41/149). Two patients were operated upon for other manifestations of myocardial ischemia without angina pectoris. All patients were operated upon with the aid of total cardiopulmonary bypass with a disposable bubble oxygenator, hemodilution to a mixed venous hematocrit value of 25 per cent (± 3 per cent), moderate hypothermia (28° ± 2° C.), and electrically induced or spontaneous ventricular fibrillation. Individual saphenous vein grafts or direct internal mammary grafts were placed with the aid of aortic occlusion during the coronary anastomoses. An attempt was made to achieve complete revascularization, i.e., to bypass all obstructed major vessels that had patent distal segments demonstrated angiographically or found at operation. All patients were operated upon by or under the direct supervision of one surgeon. The perioperative infarction rate, defined as the appearance of new Q waves on serial

postoperative electrocardiograms, was determined. We followed all surviving patients by office visit or direct telephone contact with 100 per cent follow-up to February, 1975. The cause of death was determined by autopsy in all but 1 patient. The current symptomatic status of all surviving patients was determined. Graft patency was determined in 70 per cent of hospital survivors (104/148) studied on a random basis an average of 6 months postoperatively (11 days to 38 months). A total of 237 grafts in the 104 patients were examined. Patient survival 1, 2, and 3 years postoperatively was determined according to the life-table method of Cutler and Ederer. 12 Standard error of the survival rate at each interval was found according to the method developed by Greenwood" and also described by Merrell and Shulman.'! The differences in survival between our surgically treated group and the medically treated group of Friesinger and associates- 2 at each interval were tested for significance by performing a two-tailed Z test (see Appendix). The p values corresponding to each Z, i.e., the probability that the difference observed could occur by chance, were found by means of standard normal tables." Results

In the total series there were 3 late deaths and 1 hospital death for an operative mortality rate of 0.7 per cent and a total mortality rate of 2.7 per cent. The cumulative survival rate including all deaths was 97 per cent 39 months after surgery; the mean follow-up was 18 months (range 4 to 40 months). There were 13 perioperative myocardial infarctions, an incidence of 8.7 per cent (13/149). Graft patency (mean interval to restudy, 6 months) was 86 per cent (203/237). Ninety-five per cent (99/ 104) of those patients studied had one or more grafts patent. Seventy-two per cent (104 of 145 surviving patients) are currently free of angina. The angiographic scores of the 149 pa-

Volume 70

Coronary artery surgery

Number 3 September, 1975

407

25

4

5

6

7

8 9 10 ANGIOGRAPHIC SCORE

12

11

13

14

Fig. 1. Distribution of angiographic scores (see Table I) among 149 surgically treated patients. Eighty patients (54 per cent) had severe coronary artery disease (scores of 10 or higher).

tients are shown in Fig. 1. Fifty-four per cent (80/149) had scores of 10 or higher, corresponding to significant disease of all three major coronary branches. Twelve of these 80 patients (15 per cent) had significant (70 per cent or greater) obstruction of the left main-stem coronary artery. Two of the 4 deaths were in patients with scores less than 10 (Table II). One person died on the fifth postoperative day of pulmonary insufficiency. The other was a young man dying 18 months postoperatively while playing basketball. His death was presumably due to arrhythmia. In those with scores of 10 or greater there was 1 late noncardiac death caused by peritonitis associated with chronic lymphocytic leukemia 5 months postoperatively. Autopsy showed four patent grafts and no evidence of recent cardiac injury. The 1 cardiac death in the group with scores of 10 or greater occurred in a 55year-old diabetic man who died 6 months postoperatively after heat exposure. Interestingly, he had been restudied 1 month previously and his single left anterior descending graft was closed. The cumulative survival at 39 months in the surgically managed patients with scores of 10 or greater was 98 per cent (Fig. 2). At 1 year, comparison between the medically treated group of Friesinger and associates! who had scores greater than 10 and

Table II. All deaths in 149 surgically treated patients Patient

'~I-_In_te_r_va_I_IL.

A B C* D*

14 II

8 9

Cause

5 days 5 months 6 months 18 months

_

Pulmonary insufficiency Peritonitis Heat stroke Arrhythmia

'Cardiac death.

our surgically treated group with scores greater than 10 showed cumulative survival rates of 89 per cent and 98 per cent respectively; at 2 years 77 per cent and 98 per cent; and at 3 years 68 per cent and 98 per cent. Comparison of the two groups in life-table form shows the significance of the differences between the two groups 1, 2, and 3 years postoperatively (Table III). The differences were highly significant at each interval. The mean age in our group of 80 surgically managed patients with scores of 10 or greater was 52 years, with a range of 38 to 69 years. Seventy-five of the 80 patients had left ventriculograms suitable for analysis. The mean preoperative ejection fraction in this group was 68 per cent (range 20 to 90 per cent) (Fig. 3). Eighteen of the 75 patients (24 per cent) had abnormal ventricular function, defined as an ejection fraction of 50 per cent or less. Twenty-two of the 80

The Journal of

408

Ullyot et al.

Thoracic ond Cordiovascular Surgery

90

~s

80

z ~

70

...~

e,

~

,

Nonsurgically treated' ......

60

... ...

.•

50

o

12 24 36 MONTHS FOLLOWING ANGIOGRAM OR SURGERY

48

Fig. 2. Cumulative survival at 1, 2, and 3 years among surgically treated (our series) and medically treated ', 2 patients.

Table III. Cumulative survival among medically treated and surgically treated patients 3 years

Effective number exposed to risk of dying Cardiac deaths Cumulative survival (%) Standard error (%) Value for Z Probability

47 5 89 5

'From series of Friesinger, Humphries, and associates."

42 6 77 6

36 4 68 7

80 I 98 I 2.18 < 0.05

52 0 98 I 3.31 < 0.001

24 0 98 I 2.84 < 0.01

2

patients (28 per cent) underwent urgent surgery for preinfarction angina pectoris (Fig. 4). Twenty-six (33 per cent) had a progressive anginal pattern leading to study, but they did not require admission to a coronary care unit to rule out acute myocardial infarction. The remaining 32 patients (40 per cent) had stable disabling angina pectoris. A total of 193 individual grafts were placed in these 80 patients (an average of 2.4 grafts per patient) (Fig. 5). Nine patients (11 per cent) had single grafts, 34 (43 per cent) double grafts, 32 (40 per cent) triple grafts, and 5 (6 per cent) had

quadruple graft procedures. Eleven vessels required gas or mechanical endarterectomy in order to receive a bypass graft. The peri operative infarction rate was 10 per cent (8/80); three of these infarctions occurred in patients operated upon for preinfarction angina pectoris. Fifty-five patients had graft visualization at an average of 6 months postoperatively. Patency rate was 82 per cent (108/131 grafts). Ninety-five per cent (52/55) of patients restudied had one or more grafts open. The average number of patent grafts per patient studied with a score of 10 or greater was 1.96. Mean follow-up for these 80 patients was 18

Volume 70

Coronary artery surgery

Number 3 September, 1975

409

26 24 22 20

i'?

18

a;

16

~

14

0

12

>=

i z

10 8 6

O~..---

I

o

2

10

11-20

Fig. 3. Preoperative systolic ejection fractions by volume among patients with scores of 10 to 14. Of 80 patients with scores of 10 or above, 75 had left ventriculograms satisfactory for analysis.

I

40

20 30 NUMBER OF PATIENTS

so

Fig. 4. Clinical indications for surgery in 80 patients with scores of 10 or higher. 40

months (range 5 to 39). Of surviving patients, 73 per cent (57/78) report no angina pectoris on current follow-up. Discussion

The effect of coronary artery bypass on survival in patients with ischemic heart disease is still not certain. Varied methodologies, as well as a tendency to compare heterogeneous populations of medically and surgically treated patients with various degrees of coronary atherosclerosis, make interpretation difficult. For example, Cannom and associates" felt that their data at a mean follow-up of 9.9 months did not permit any definite statement regarding patient survival. Aronow and Stemmer" concluded that there was no difference in mortality rate at 1 year between their 20 surgically and 20 medically managed patients. Similarly, with the exception of one subgroup, McNeer and associates-" found no difference in 2 year survival between 379 surgically treated patients and 402 medically treated cohorts. Spencer and associates'" reported a 5 year survival rate of 81 per cent in 448 patients who had elective coronary artery bypass. These authors included operative deaths and late cardiac deaths and they compared their survival rate to that of Bruschke and associates," who reported 66 per cent 5 year survival rate in 590 medically treated patients with single-, double-, or triple-vessel disease. Anderson and co-

2

3

NUMBER OF GRAFTS PER PATIENT

Fig. 5. Number of grafts placed in 80 patients with scores of 10 or higher. Forty-six per cent of these patients had triple or quadruple graft procedures.

workers" reported an 84.4 per cent 4 year cumulative survival rate in 203 patients with three-vessel disease who were part of their series of 532 patients who had coronary bypass surgery. Their series excluded patients with unstable angina. These authors did not compare survival with that of medically treated patients with comparable disease. They did note that differences in intermediate survival among their surgically treated patients with one-, two-, or threevessel disease were small, unlike the large differences seen in medically treated patients. Sheldon and associates" showed that survival in their composite surgical group was better than that in medically managed patients" with double- or triple-vessel disease who would have been suitable candidates for surgery by criteria prevalent in 1970 to 1971. They did not specifically

4 10

Ullyot et al.

analyze survival in surgically managed patients with triple-vessel disease. Collins and associates" reported a composite 96 per cent survival in 93 patients managed surgically for angina pectoris and followed for an average of 9 months. There have been recent preliminary reports in abstract form~3-25 that address themselves to survival in patients with extensive coronary artery disease and that appear to corroborate our findings. The study by Friesinger and associates 1 was the first to show an association between extent of angiographically demonstrated coronary artery disease and survival in medically treated patients. A later report by Humphries and colleagues- provided a 12 year follow-up of this same group of 224 patients who had had coronary angiography between 1960 and 1967. Forty-seven of these patients had scores of 10 or greater, which are equivalent to significant involvement of all three major vessels. Cumulative survival among their 47 medically treated patients with extensive coronary disease was 53 per cent at 5 years; these results are similar to 5 year survival rates reported in other groups of medically treated patients with three-vessel disease." 10 We conclude that theirs was an important study and that the information provided is necessary to assess the influence of interventions, especially surgical interventions, on the natural history of the disease. 1 We have compared survival of 80 surgically treated patients who had scores of 10 or greater with that of Friesinger's 47 medically treated patients with similar scores. The statistically significant differences at 1, 2, and 3 year intervals suggest that surgical treatment exerts a favorable influence on survival in patients with extensive coronary artery disease. In addition, the difference in morbidity is striking: Only 2 of the 24 medically treated patients were pain free and without cardiovascular limitations at late follow-up"; 57 of 78 survivors in our surgically treated group were totally relieved of angina at current follow-up. There are several possible objections to

The Journal of Thoracic and Cardiovascular Surgery

this study. It is possible that the medical group represents patients with more severe coronary atherosclerosis. Some patients in the medically treated series were examined only in the left anterior oblique position, which could have resulted in systematic underscoring of some patients. Another objection might be that different observers scored each series. However, the scoring system is unambiguous and substantial differences are unlikely. Another objection might be that the medical group does not represent the course of severe ischemic heart disease that has been treated with long-term beta adrenergic blockade." There are no published data that show improved survival with these agents, however. A more serious criticism is that different groups are being compared. The medically treated patients, however, would seem to be a generally more favorable group as regards survival than the surgically managed patients. They were younger (mean age 44 years); all had achieved a stable level of ischemic symptoms (patients with unstable angina were excluded); left ventricular function was apparently good (patients with symptoms of congestive failure or cardiac enlargement on chest film were excluded); and none had severe hypertension (patients with diastolic blood pressure greater than 110 rom. Hg were excluded on retrospective analysis) .27 By contrast, the surgical group was older (mean age 52 years); 28 per cent had urgent surgery for preinfarction angina pectoris by strict clinical criteria; 33 per cent had progressive angina that often resulted in pain at rest but that did not require admission to a coronary care unit to rule out acute myocardial infarction; and 24 per cent had abnormal left ventricular function at rest with systolic ejection fractions of 50 per cent or less. Unfortunately, left ventriculography was not done routinely during the period the medical group was studied, and comparable analysis by left ventriculography is not possible. If one accepts the comparison between the two groups and the statistically significant difference in 3 year survival, one obvi-

Volume 70

Coronary artery surgery

Number 3 September, 1975

ous explanation for the difference is improved myocardial blood supply in the surgically treated group. Studies of graft patency disclosed an average of two patent grafts per patient studied. This may have had the effect of shifting patients with extensive disease into the more favorable prognostic category of patients with single-vessel disease. Two assumptions are implicit in this speculation: that grafts remained patent and that angiographic demonstration of graft patency is equivalent to adequate blood flow to the bypassed vessel. The group treated medically was not analyzed from the standpoint of suitability for surgery. It is possible that patency of at least one distal vessel, a necessary criterion for selection of patients for surgery, selects a subset of patients with more favorable chances of survival from among those with severe coronary atherosclerosis. Relevant to this point are the data of McNeer and associates," who found increased 2 year survival in their surgically treated cohort who had three-vessel disease, normal arteriovenous oxygen difference, and abnormal ventricular function. The authors noted that the difference in survival was specifically not due to the distribution of patients with good distal vessels in the two cohorts, 169 patients in all. The significance of our report is that we have compared medically and surgically managed patients with: (l) a similar degree of and (2) a severe extent of coronary artery disease determined angiographically. After achieving 100 per cent current followup of all patients, including those operated upon both electively and urgently, we have found significantly increased survival among the surgically treated group.

intervention in patients with extensive coronary artery disease. We thank Robert F. Mihalik and Beverly Hill for follow-up of patients, and Stanley Edlavitch, Ph.D., Assistant Chief, Community Medicine Program, U. S. Public Health Service Hospital, San Francisco, for statistical analysis. REFERENCES

2

3 4

5

6

7

8

9

Conclusions The data suggest that there is improved late survival in patients with extensive coronary atherosclerosis managed surgically compared to that in patients managed medically. We conclude that the possibility of improved survival, in addition to symptomatic improvement, should favor surgical

411

10 11

12

Friesinger, G. C., Page, E. E., and Ross, R. S.: Prognostic Significance of Coronary Arteriography, Trans. Assoc. Am. Phys. 83: 78, 1970. Humphries, J. 0., Kuller, L., Ross, R. S., Friesinger, G. c., and Page, E. E.: Natural History of Ischemic Heart Disease in Relation to Arteriographic Findings: A Twelve Year Study of 224 Patients, Circulation 49: 489, 1974. Lichtlen, P. R., and Moccetti, T.: Prognostic Aspects of Coronary Angiography, Circulation 46: 7, 1972 (Supp!. II). Slagle, R. C., Bartel, A. G., Behar, V. S., Peter, R. H., Rosati, R. A., and Kong, Y.: Natural History of Angiographically Documented Artery Disease, Circulation 46: 60, 1972 (Supp!. II). Moberg, C. H., Webster, J. S., and Sones, F. M., Jr.: Natural History of Severe Proximal Coronary Disease as Defined by Cineangiography (200 Patients, 7 Year Followup), Am. J. Cardio!. 29: 282, 1972. Bruschke, V. G., Proudfit, W. L., and Sones, F. M., Jr.: Progress Study of 590 Consecutive Nonsurgical Cases of Coronary Disease Followed 5-9 Years. I. Arteriographic Correlations, Circulation 47: 1147, 1973. Webster, J. S., Moberg, C., and Rincon, G.: Natural History of Severe Proximal Coronary Artery Disease as Documented by Coronary Cineangiography, Am. J. Cardio!. 33: 195, 1974. Burggraf, G. W., and Parker, J. 0.: Prognosis in Coronary Artery Disease: Angiographic, Hemodynamic, and Clinical Factors, Circulation 51: 146, 1975. Oberman, A., Jones, W. B., Riley, C. P., Reeves, T. J., and Sheffield, L. T.: Natural History of Coronary Artery Disease, Bul!. N. Y. Acad. Med. 48: 1109, 1972. Reeves, T. J., Oberman, A., Jones, W. B., and Sheffield, L. T.: Natural History of Angina Pectoris, Am. J. Cardio!. 33: 423, 1974. Greene, D. G., Carlisle, R., Grant, C.; and Bunnell, I. L.: Estimation of Left Ventricular Volume by One-Plane Cineangiography, Circulation 35: 61, 1967. Cutler, S. J., and Ederer, F.: Maximum Utili-

The Journal of

4 12

Ullyot et al.

Thoracic and Cardiovascular Surgery

13

14

15 16

17 18

. 19

20

21

22

23

24

zation of the Life Table Method in Analyzing Survival, J. Chron. Dis. 8: 699, 1958. Greenwood, M.: Appendix 1, The Errors of Sampling of the Survivorship Tables, in A Report on the Natural Duration of Cancer, Reports on Public Health and Medical Subjects, No. 33, London, 1926, His Majesty's Stationery Office. Merrell, M., and Shulman, L. E.: Determination of Prognosis in Chronic Disease, Illustrated by Systemic Lupus Erythematosus, J. Chron. Dis. 1: 12, 1955. Snedecor, G. W., and Cochran, W. G.: Statistical Methods, ed. 6, Ames, Iowa, 1967, Iowa State University Press. Cannom, D. S., Miller, D. C., Shumway, N. E., Fogarty, T. J., Daily, P.O., Hu, M., Brown, B., Jr., and Harrison, D. c.: The Long-Term Follow-up of Patients Undergoing Saphenous Vein Bypass Surgery, Circulation 49: 77, 1974. Aronow, W. S., and Stemmer, E. A.: Bypass Graft Surgery Versus Medical Therapy of Angina Pectoris, Am. J. Cardiol. 33: 415, 1974. McNeer, J. F., Starmer, C. F., Bartel, A. G., Behar, V. S., Kong, Y., Peter, R. H., and Rosati, R. A.: The Nature of Treatment Selection in Coronary Artery Disease: Experience With Medical and Surgical Treatment of a Chronic Disease, Circulation 49: 606, 1974. Spencer, F. c., Isom, O. W., Glassman, E., Boyd, A. D., Engelman, R. M., Reed, G. E., Pasternack, B. S., and Dembrow, J. M.: The Long-Term Influence of Coronary Bypass Grafts on Myocardial Infarction and Survival, Ann. Surg. 180: 439, 1974. Anderson, R. P., Rahmintoola, S. H., Bonchek, L. I., and Starr, A.: The Prognosis of Patients With Coronary Artery Disease After Coronary Bypass Operations: Time-Related Progress of 532 Patients With Disabling Angina Pectoris, Circulation 50: 274, 1974. Sheldon, W. C., Rincon, G., Effler, D. B., Proudfit, W. L., and Sones, F. M., Jr.: Vein Graft Surgery for Coronary Artery Disease: Survival and Angiographic Results in 1,000 Patients, Circulation 48: 184, 1973 (Supp!. III). Collins, J. J., Jr., Cohn, L. H., Sonnenblick, E. H., Herman, M. V., Cohn, P. F., and Gorlin, R.: Determinants of Survival After Coronary Artery Bypass Surgery, Circulation 48: 132, 1973 (Suppl, III). Cohn, L. H., and Collins, J. J., Jr.: Improved Long-Term Survival Following Coronary Artery Bypass, Circulation 50: 166, 1974 (Suppl. III). Wertheimer, M., and Liddle, H. V.: Results of Direct Coronary Artery Graft Reconstruction: A Five-Year Clinical and Arteriographic Appraisal, Program for the Eleventh Annual

Meeting of the Society of Thoracic Surgeons, Montreal, Canada, January, 1975, p. 84. 25 Manley, J., Friedberg, H. D., Auer, J., and Johnson, W. D.: Late Follow-up (to Five Years) after Coronary Surgery in Comparison to Reported Medical Series, Am. J. Cardio!. 35: 155, 1975. 26 Coronary-bypass Surgery [Editorial], Lancet 1: 137, 1973. 27 Humphries, J. 0.: Personal communication, March, 1975.

Appendix Hs: Pk (surgical)

= Pk (medical)

where H o is the null hypothesis, i.e., that there is no difference in cumulative survival p at the k th interval. Pk (surg) - Pk (med)

z

"'-\JIpQ

[

1 + 1 ] h' (surg) h' (med)

where h' (surg ) is the effective number exposed to the risk of dying during the k th interval in the surgical group and h' (med) is the effective number exposed to the risk of dying during the k th interval in the medical group. P = h' (surg) x Pk (surg) + h' (med) x Pk (medL h' (surg) + h' (med) and

Q = 1 - P. When the null hypothesis is true, the probability of finding Z of a given magnitude is found in a standard normal table.

Discussion DR. JOHN 1. COLLINS Boston, Mass.

It is a pleasure to see the results reported by Dr. Ullyot. Studies of this type are going to cause many to change their attitudes toward revascularization surgery. Death is an unarguable end point as opposed to recurrence of angina or myocardial infarction. Our results in this area are entirely consistent with those which have been reported by Dr. Ullyot. We have reviewed a series of 330 angina patients operated upon since 1970 and have found significantly improved longevity up to 4 years after surgery in those with two- and threevessel disease as compared to results in medically treated patients with a similar degree of angiographically proved disease. Longevity in patients with single-vessel disease did not appear to be improved, however.

Volume 70 Number 3

Coronary artery surgery

4 13

September, 1975

It is our policy now to recommend surgery for improvement of life expectancy in patients with greater than single-vessel obstruction and to reserve surgery in distal single vessel disease for those with truly intractable angina. DR. JEROME H. KAY Los Angeles, Calif.

For this discussion I will confine my remarks to those patients with an ejection fraction of 0.1 to 0.2 (normal 0.7). For the past 6 years, we have operated upon patients with an ejection fraction of 0.1 to 0.2. The patients were offered surgery, and a risk of death of 30 to 35 per cent was quoted with surgery. Between February, 1970, and August, 1974, 78 patients whom we considered operable declined the operation. Of these 78 patients, by actuarial studies, only 23 per cent were alive 5 years later. From August, 1969, to December, 1974, 171 patients with ejection fractions of 0.1 to 0.2 consented to myocardial revascularization despite the high surgical mortality rate quoted. There was a surgical mortality rate of 26 per cent in these patients. Actuarial studies revealed that 62 per cent were alive 5 years after the operation. Therefore, at the end of 5 years, 39 per cent more patients are alive with surgical treatment than with medical treatment. For the past 2 years our surgical mortality rate has decreased quite strikingly with these critically ill patients. This has been due primarily to the untiring efforts of Dr. Pablo Zubiate. We treated 59 patients in this group from Jan. 1, 1973 to December, 1974. There were only 4 deaths in these 59 consecutive patients (7 per cent mortality) who had an ejection fraction of 0.2 or

less and underwent myocardial revascularization. With this surgical mortality, 90 per cent are alive at 24 months by actuarial studies, compared to only 35 per cent alive after 2 years of medical treatment. We restudied a number of these patients with an ejection fraction of 0.1 to 0.2. The ejection fraction improved in 70 per cent of the patients. We feel strongly that surgery is indicated for the patient with a poorly functioning heart if the procedure can be performed with a low enough risk. DR. ULL YOT (Closing) I wish to thank the discussers for their remarks. Dr. Collins put it very articulately when he said that death is an unambiguous criterion of the efficacy of surgical versus medical therapy. There are at least two subgroups of patients with ischemic heart disease who are at high risk with nonoperative therapy. One is patients with significant stenosis of the left main-stem coronary artery. The other is patients with extensive coronary artery disease. When composite groups of medically and surgically managed patients are compared, the survival data of subgroups, such as those with triple-vessel disease, tend to be obscured. We have paid particular attention to comparing patients with (1) similar and (2) extensive degrees of coronary atherosclerosis in asserting the superiority of surgical management for survival. Dr. Kay has pointed out another subgroup at high risk with medical management. It is probable that there is some overlap between his patients with very poor left ventricular function and those with extensive coronary artery disease.