Aortic valve replacement with coronary bypass grafting

J

THoRAc CARDIOVASC SURG

85:705-711, 1983

Aortic valve replacement with coronary bypass grafting Significant determinants of ten-year survival Concomitant aortic valve replacement (AVR) and myocardial revascularization were performed on 197 patients between 1969 and 1981. Operative mortality during the period 1969 to 1975 was 15.6 % compared to 5.0% for the years 1976 to 1981 (p < 0.02). The incidence of perioperative myocardial infarction (PMI) declined over the same period from 14.2% to 2.0% (p < .01). Functional class and left ventricular end-diastolic pressure significantly influenced mortality, whereas age, sex, duration of symptoms, cardiac index, wall motion abnormality, type of valve lesion, and completeness of revascularization did not. Type of myocardial preservation did not significantly affect operative mortality, a/though a trend favoring either cardioplegia or continuous perfusion of both coronary ostia and grafts was observed. Life-table analysis shows a survival rate during the 10 year follow-up period equal to that of patients undergoing isolated AVR. Cornary bypass grafting (CABG) returns patients with combined aortic valve (AVD) and coronary artery disease (CAD) to a prognostic curve determined by their valvular disease alone.

David L. Nunley, M.D., Gary L. Grunkemeier, Ph.D., and Albert Starr, M.D., Port/and. Ore.

Coronary bypass grafting (CABG) improves the survival rate for certain subgroups of patients with coronary artery disease (CAD),u Patients with aortic valvular disease (AVO) have increased left ventricular work and greater myocardial oxygen demands, which may aggravate the effects of concomitant CAD. Recognition of the poor prognosis of AVD plus CAD3 has led to the practice of combining CABG and aortic valve replacement (AVR). Although early studies showed widely disparate operative mortalities ranging from 9% to 85% and implied that the late survival rate was worse than that seen with AVR alone.t" recent data suggest a comparable result in these two groups.' Follow-up is short in all reported series to date, however. This study From the Division of Cardiopulmonary Surgery, Oregon Health Sciences University, and St. Vincent Medical Center, Portland. Ore. Received for publication June 28, 1982. Accepted for publication July 19, 1982. Address for reprints: David L. Nunley, M.D., Oregon Health Sciences University, 3 I81 SW Sam Jackson Park Rd., Portland, Ore. 9720 I.

provides the longest follow-up with concurrent CABG and AVR over the last 12 years, shows the importance of time-frame in determining late morbidity and mortality, and demonstrates the efficacy of the combined operation in the current time-frame. Patients and methods

From 1969 to 1981, 197 patients underwent AVR and concomitant CABG at the Oregon Health Sciences University and St. Vincent Medical Center. All individuals had single valve replacement. Clinical information, angiographic findings, and hemodynamic parameters obtained at catheterization are summarized in Table I. Follow-up was achieved through office visits, questionnaires, and telephone contacts with patients and referring physicians. Special questionnaires were designed to assess current functional status for both angina and congestive heart failure according to the New York Heart Association (NYHA) classification. A mean follow-up of 4.3 years was obtained for 98.5% of the patients whose status was known as of Aug. 15, 1981. Although maximum follow-up was 11.3 years, only six patients were followed up past 10 years and there were 705

The Journal of

7 0 6 Nunley, Grunkemeier, Starr

Thoracic and Cardiovascular Surgery

Table I. Preoperative clinical and hemodynamic data No. of patients Male (%) Mean age (range) Median duration of symptoms (yr) NYHA Functional Class (%) Class I Class II Class III Class IV CAD risk factors (%) Family history Hypertension Previous infarction Diabetes Indications for CABG (%) Angina Chronic Progressive Unstable Left heart failure Prophylactic Valvular lesion (%) Stenosis Regurgitation Mixed Etiology of valvular disease (%) Calcific Rheumatic Other Extent of CAD (%) One vessel Two vessel Three vessel Left main Catheterization data: mean (range) Cardiac index (n = 136) (L/min/m') LVEDP (n = 156) (mm Hg) Ejection fraction (n = 64) (%) End-diastolic volume index (n = 52) (rnl/rn') Pulmonary wedge pressure (n = 150) (mm Hg)

197 74 64 (41-81) 3.0

3 43 39 15 31 26 17

9

59 16

5 3 17 52 12 36 53 35 12

Table Il, Operative mortality Factor

No. of patients

Overall Time-frame of operation 1969-1975 1976-1981

Functional Class I-II Ill-IV Age group (yr) 41-59 60-69 70-81

Sex Male Female Duration of symptoms 5 yr Myocardial protection Cardioplegia Continuous perfusion: ostia/grafts Intermittent perfusion: ostia only Continuous perfusion: ostia only

I Mortality (%)

197

10.2

96 101

16 5*

72 85

4 1St

61 85 51

13 7 12

146 51

14 9

60 65 57

8 8 15

27 65 71 34

7 8 10 18

*p < 0.02 tp < 0.05

Table

m. Causes of operative death

48

Cause

33 10 9

No. of patients

Myocardial infarction Low output syndrome Hemorrhage Infection Arrhythmia Renal failure

2.7 (1.2-5.0) 16.4 (2-45) 56 (16-86)

Total

6

5 4 3

I 1 20 (10.2%)

110.3 (27-283) 12.5 (3-40)

Legend: NYHA, New York Heart Association. CAD, Coronary artery disease. CABG, Coronary artery bypass grafting. LYEDP, Left ventricular end-diastolic pressure.

no deaths after this time; accordingly, survival curves are shown for 10 years. Five hundred ninety-five patients underwent AVR alone during the period 1970 to 1980. As it is our policy to obtain coronary angiograms on all adult patients with A VD, this group comprises individuals not having a significant (>50 %) degree of coronary artery obstruction. Patient groups were matched for time-frame of

operation and were adjusted by means of relative survival computations for age and sex differences. Thus 25% of our AVD patients had concomitant CAD.

Computer data analysis utilized a statistical interactive programming system. * Late survival probabilities were estimated by the life-table method"; actuarial curves were compared by means of the log-rank statistic." Relative survival curves were used for comparing survival between two groups with different age and sex distributions.'? Relative survival is the actual observed survival divided by the expected survival for an age- and sex-matched group from the normal Oregon population. Chi square statistics were used to compare proportions. As no children are included in this series, all calcified bicuspid valves were considered "calcific." Most (92%) valves used were of caged-ball design. Angiographic

*Medical

Data Research Center, Portland, Ore.

Volume 85 Number 5

70 7

Aortic valve replacement and coronary bypass

May. 1983

Table IV. Causes of late death Cause Cardiac-valve related Stroke Endocarditis Thrombosis Sudden unknown Cardiac-not valve related Noncardiac Pending Total

--._--!-.-

100

<,

No. oj patients 6

.............._.~

69%!.-4

3 2 I

52%:!:6

! :!: S.E.

6 16 II

I

6

2

descriptions of the extent of CAD were correlated with the number and distribution of grafted vessels to produce a "revascularization index," which is the ratio of number of diseased vessels grafted/number of diseased vessels identified on angiography, modified from the methodof Dash and associates. I I The mean revascularization index was 0.92 with an average of 1.6 grafts/patient. The occurrenceof perioperative myocardial infarction (PMI) was evaluated in all patients; the diagnosis was made on the basis of the following: (1) new Q waves :2:0.04 second duration and >3 mm in depth; (2) ST-segment depression > 1 mm with associated T-wave inversion in two or more adjacent electrocardiographic leads for 48 hours or more in conjunction with typical enzyme elevations, particularly a creatine kinase value greater than 500 IV with at least 10% myocardial specific isoenzyme; (3) autopsy evidence of recent infarction.

Results Overall operative mortality was 10.2%; however, marked diminution in risk was noted for the last 6 years of the study, and year of operation was the most important preoperative variable affecting early mortality (Table II). Sixteen patients had a PMI, for an overall PMI rate of 8.1 %. Of these infarctions, six were fatal, accounting for 30% of all operative deaths. Occurrence of PMI significantly influenced mortality (p < 0.001). The incidence of PMI declined dramatically during the latter half of the study period:from 1969to 1975the PMI rate was 14.2%, compared with only 2.0%from 1976to 1981 (p < 0.01). The incidence of PMI in relation to the four modes of myocardial protection was as follows: continuous perfusion of ostia and grafts, 3%; cardioplegia, 4%; continuous perfusion of ostia only, 9%; intermittent ostial perfusion, 14%. Although continuous perfusionof the ostia and graft produced a significantly lower rate of PMI than did either intermittent or continuous perfusion of the ostia alone (p < 0.05), cold cardioplegia did

4

5

6

7

8

9

10

YEARS POSTOP.

45*

*5.9/100 patient-years or follow-up.

3

Fig. 1. Actuarial survival for all patients undergoing aortic valve replacement with coronary bypass. Numbers in parentheses above horizontal axis show number of patients at risk. S.£., Standard error. 100

~2~_. 80% 80 % ·o-·_·~e-·--o

--.

~-._.e-._-o-._

__

73%

. - . AVR +CBS 0-'-0 AVR only 2

345

6

7

8

9

/0

YEARS POSTOP.

Fig. 2. Relative survival for 197 patients having aortic valve replacement (AVR) with coronary bypass surgery (CBS) and 595 patients undergoing AVR alone during 1970 to 1981.

Relative survival is the ratio of the observed survival of the group in question to that of an age- and sex-matched group from the normal population. The increase in AVR-CBS relative survival in the sixth and tenth years indicates a slightly better than normal survival.

not, since this latter technique was employed in a relatively small number of patients. Neither the length of cardiopulmonarybypassnor the myocardial ischemic time was related to occurrence of PMI. Of interest is that whereas only six patients displayed any hemodynamic instability, five of these also died of infarction. On postmortem examination, four patients were found to have had full-thickness infarcts. Two other patients probably had preoperative infarctions; profound hypotension developed after ventricular perforation during catheterization in one, and an unrecognized infarct occurred 3 weeks prior to elective operation in the other. Neither history of prior myocardial infarction nor functional status had an effect on the PMI rate. Five patients died of low output syndrome. Autopsy did not confirm myocardial injury in these individuals. No relationship could be found betweenfunctionalclass, left ventricular end-diastolic pressure, ejection fraction,

The Journal of Thoracic and Cardiovascular Surgery

7 0 8 Nunley, Grunkemeier, Starr

100 ' • -. 80 o;....

-...J ~

s:

0...

·-·0-._

S;:

~ 60 II)

'~40 ~ ~ 20 Cl:::

•

73°//0

_

'0--0--<>-._

~

0-.-2._._.0._._ ~~-.

.0-._.

63 %

1:

0 ...._.

0

3

47% -·-·0-·_·0

! !S.E.

~Iandn

..-. ill and nz: 4

5

6

7

8

9

10

2

3

YEARS POSTOP.

Fig. 3. Survival by time-frame of operation for 96 patients having aortic valve replacement with coronary bypass in 1969 to 1975 and 101 patients undergoing aortic valve replacement with coronary bypass in 1976 to 1981. This difference approaches significance at 5 years (p = 0.053). 100 -...J

§ 80 S;: 9::

,-0'e ·-·0-·_·0_._. i 4% <,

~60

~ 40 ~

~ 20

Cl:::

o Op.

----.

!

O-·_·~_·_o()_·_·o..._

56 %

~--e"

! S.E. 0-,-0 ~15 mm Hg . - > 15 mm Hg 2

3

4

·0-,-·

r

.-._1I

°

67 '10

6

7

8

9

4 5 6 YEARS POSTOP.

7

·0-·-'0-'-"'(

8

9

iO

Fig. 5. Survival for preoperative Functional Classes I-II and III-IV for patients undergoing aortic valve replacement with coronary bypass (p < 0.05).

l.oop(67-73)

Nunley (69-75)

N o80

N'96

,-·0

1

41% 2

5

68~

---."'-.32%1

61 %

-->---..~

:r!S.E. '-.1976-81 0-·-01969-75 2

75 %

·0--+-0-._.0......

10

YEARS POSTOP.

Fig. 4. Survival for patients undergoing aortic valve replacement with coronary bypass grouped according to left ventricular end-diastolic pressure (p < 0.05). cardiac index, method of myocardial protection, or duration of cardiac bypass and death in this subgroup. Of the seven patients dying of hemorrhage or infection, three with mediastinitis died of aortic suture line or aorta-coronary anastomosis disruptions; one developed endocarditis in the immediate postoperative period and died of a myocardial abscess; one had an intraoperative injury to the atrial-aortic junction and died of hemorrhage; one with generalized aneurysmal dilatation of the aorta underwent successful aneurysmorrhaphy but died on the fourth postoperative day of suture line disruption; and one man who required three gastric operations for stress bleeding after the cardiac operation died of intra-abdominal sepis 7 weeks after the initial operation. The causes of operative death are summarized in Table

III. Forty-five late deaths occurred during the follow-up period, for an actuarial survival rate (± standard error) of 82% ± 3% at I year, 69% ± 4% at 5 years, and 52% ± 6% at 10 years (Fig. 1). Although only six patients have survived beyond 10 years, the actuarial

3

4

5

6

7

8

9

10

YEARS POSTOP. Fig. 6. Comparison of survival curves for those studies in Table V using actuarial methods.

technique takes into account the varying amounts of experience contributed by all 197 patients in the series. Accordingly, the effective sample size" is calculated to be 69, which means that the precision obtained in the estimate of 10 year survival is the same as if 69 patients had been operated upon and followed up for 10 years. Late survival was most affected by cardiac related events, whereas valve related mortality accounted for only 13% of deaths (Table IV). Relative survival compared favorably to that of the control group undergoing AVR alone (Fig. 2). Long-term survival related to operative time-frame demonstrates improved survival rates in the more recently operated group, with 86% ± 4% alive at I year, 82 ± 4% at 3 years, and 73% ± 6% at 5 years (Fig. 3). This trend approaches statistical significance at 5 years. Functional class and left ventricular end-diastolic pressure were also significant determinants of long-term prognosis (Figs. 4 and 5), whereas age, sex, type of valve lesion, wall motion abnormality, cardiac index, duration of symptoms, and completeness of revascularization did not significantly affect overall survival. Currently, 86% of surviving patients are in NYHA

Volume 85

Aortic valve replacement and coronary bypass

Number 5 May, 1983

709

Table V. Comparison of studies reporting aortic valve replacement with coronary artery bypass grafting Authors (time-frame) Loop et al." (1967-1973) Berndt et al." (1970-1972) Richardson et al.' (1970-1977) Wisoff et al." (1970-1978) Lundell et al." (1973-1978) Jones" (1974-1980) Nunley et al." 1969-1975 1976-1981

No. of patients

Mean age (yr)

Male (%)

Operative mortality (%)

Mean follow-up (yr)

80 28 220 44 82 128

57 59 59 60 60 64

95 96 81 73t 88 83

9 14 5 7 20 6

2.9 1.5 1.9 3.5t 1.9

96 101

63 65

75 73

16 5

5.9 2.8

'Present series divided by time-frame of operation. tCombined aortic or mitral valve replacement with coronary bypass grafting.

Functional Classes I or II, and symptoms of heart failure rather than ischemic pain tend to limit activity, since 80% of patients are free of angina. Over 80% are improved from preoperative Classes III and IV. Discussion Arteriosclerotic heart disease is a leading cause of complications and death following valve replacement. Copeland and co-workers' emphasized that long-term mortality is doubled in those patients undergoing A VR in the presence of significant CAD. Early reports showed a greater operative mortality from combined valve replacement and CABG than from either procedure alone," but Anderson and colleagues" in 1973 noted an early mortality of 4% in 27 patients who had both AVR and CABG; subsequently, other investigators demonstrated a comparable operative mortality for AVR/CABG and AVR alone.i" Predictive factors of operative risk have not been conclusively elucidated. PMI, cardiac index, incomplete revascularization, and type of myocardial protection have all been shown to influence operative survival, although the relative importance of each variable is not well defined." 16, 17 In our series, time-frame of operation was the most significant preoperative determinant of early survival. It is difficult to identify specific factors responsible for this marked improvement in survival rate during the past 6 years. Although the incidence of PMI has also fallen during the same time-frame, this improvement does not entirely account for the greater than threefold reduction in mortality. While cardioplegia acts chiefly to reduce myocardial oxygen demands, continuous perfusion of both ostia and grafts maintains adequate oxygen supply to the fibrillating, cooled heart. Thus each method affects different sides of the supply-demand equation to preserve myocardium. Continuous perfusion of ostia and grafts has been shown to be an efficacious method of myocardial protection, with results comparable to those

obtained with cardioplegia." Although no single method of myocardial preservation was a significant determinant of survival, a trend favoring either cardioplegia or continuous perfusion of ostia and grafts is evident and may well reach significance with the addition of more cases. Better patient selection is not responsible for the improved results; on the average, the more recent patients were older, in a worse functional class, and had lower cardiac indices. This study reiterates the importance of time-frame of operation when considering results. This is in agreement with earlier work" and demonstrates the evolutionary nature of cardiac surgery, emphasizing the fallacy of attempting to compare data from different time periods. This continuum of progress is doubtlessly a consequence of many factors, including increasingly accurate preoperative diagnosis, improved anesthetic and postoperative care, advances in operative technique and methods of myocardial preservation, and an evolving base of experience with regard to the management of these patients. The severity of CAD in patients undergoing A VR is less than is typically seen in individuals having CABG alone. At our institution, multivessel or left main disease occurred in 52% of the A VD plus CAD group, with an average of 1.6 grafts placed; by comparison, multivessel disease or left main involvement was present in 78% of those with only CAD and necessitated 2.2 grafts/ patient. 20 Other series of combined CAD and A VD report percentages of one-, two-, and three-vessel CAD similar to ours.': 7, 21 This selection may reflect the increased hemodynamic burden placed on the myocardium by A VD, which causes relatively less severe degrees of CAD to become symptomatic. In our subgroup of patients with proved multivessel disease, the long-term survival rate remains comparable to that for A VR alone; thus it is not the single-graft patients who are responsible for the equivalent results. Although objections might be raised regarding the

7I0

Nunley, Grunkemeier, Starr

value of CABG for single-vessel disease, we believe this practice is indicated in view of the ventricular enlargement" and residual outflow gradient" seen in many of these patients after prosthetic valve implantation. These factors may contribute to postoperative myocardial dysfunction, which can be further exacerbated by CAD. Indeed Linhart and associates" found that in patients with impaired left ventricular function following valve replacement, the most frequent associated condition was CAD. Ischemia secondary to increased myocardial oxygen demand may be obviated by revascularization of a critically stenosed single vessel, especially if the left anterior descending artery (LAD) is involved, since a significantly greater area of muscle is jeopardized by disease in this artery than by lesions affecting the right or circumflex systems." The impressive survival for patients with isolated LAD lesions surgically treated at the Cleveland Clinic provides further support for the use of CABG in this setting." Fifty-one percent of single grafts in the present series were placed to the LAD for lesionsof greater than 70% luminal diameter narrowing. A less important, but relevant consideration is that of opportunity, i.e., performing CABG as an ancillary procedure since the incremental risk of operative death in negligible." Richardson and co-workers" showed an annual decrease in operative mortality but failed to find a significant difference between the early and late timeframes in their study. They reported an 11% incidence of definite PMI overall with a 9% rate in the last 3 years. The occurrence of PMI correlated strongly with the type of myocardial protection and also with early mortality; however, nearly 40% of their patients were thought to have had "definite or probable" infarctions. Such a high rate of infarction may exaggerate the effects of subsequent mitigating factors. Bonow and associates" recently concluded that routine revascularization is not indicated for patients with CAD and AVD who undergo AYR. This assertion is based on comparisons of mortality between AVR patients with and without CAD. The authors also state that CABG may increase the risk of PMI, although their own data do not support this point. An accompanying critical editorial emphasized the small number of patients involved in Bonow's study, making valid comparison groups unlikely, and rebutted the contention that concomitant CABG might increase hospital mortality after AVR. 27 Another criticism is the lack of adjustment for the nearly 10 year age difference between the two groups. Although a second survival analysis was performed excluding 28 patients less than 35 years of age, it is not stated to what degree this exclusion changed mean age, and the only published

The Journal of Thoracic and Cardiovascular Surgery

survival curve is for the entire group. Thirty-six percent of late deaths in the unoperated CAD group were due to complications of CAD, whereas no patient in the CAD-free group died of definite ischemic causes. Bonow's group also found no difference in survival rates for patients with one-, two-, or three-vesseldisease, a finding which stands in contradiction to other data2.28 and further implies that the population studied was not representative of patients with CAD. . Angina is a common symptom in AVD, particularly aortic stenosis. Of patients with AVD, 25% to 33% have significant CAD by angiography.s'A" Without myocardial revascularization, these patients have a significantly worse late survival rate than their counterparts with normal coronary vessels.' In the absence of a contemporary control group, retrospective comparisons of patients with AVD and unoperated CAD must suffice. However, patients with AVD alone may reasonably be expected to survive longer than the population with both AVD and CAD. Indeed, one group of investigators has concluded that in patients with AVD, "... severe coronary atherosclerosis may adversely affect longevity, despite bypass grafting.": More recently, other authors have presented data suggesting that the overall survival rate of patients undergoing AVR and CABG does not differ from the results seen with AVR alone.i" Our paper supports this contention. Table V presents a comparative analysis of several major studies with actuarial survival curves, where available, superimposed in Fig. 6. Our series is subdivided by timeframe. In the report of Copeland and colleagues,' the 3 year survival rate of patients with AVR and medically treated CAD was only 60%, compared to 85% for patients free of CAD who also underwent AYR. In our series, the 3 year relative survival rate for the combined procedure is 82%. Our data indicate that complete myocardial revascularization returns those patients with combined AVD and CAD to a prognostic class equivalent to individuals with AVD alone. This observation is consistent with findings from other studies analyzing survival after CABG which demonstrate life-table curves approaching the normal population for the surgically treated group," and it further supports the efficacy of CABG in mitigating the risk of death from CAD. We wish to thank Joan Livermore for advice and aid in graphics preparation, and Edie Stout for clerical assistance. REFERENCES Second Interim Report by the European Coronary Surgery Study Group: Prospective randomised study of coronary artery bypass surgery in stable angina pectoris. Lancet 2:491-495, 1980

Volume 85 Number 5 May, 1983

2 Murphy ML, Hultgren HN, Detre K, Thomsen J, Takaro T, et al: Treatment of chronic stable angina. A preliminary report of survival data of the randomized Veterans Administration cooperative study. N Engl J Med 297:621-627, 1977 3 Copeland JG, Griepp RB, Stinson EB, Shumway NE: Long-term follow-up after isolated aortic valve replacement. J THORAC CARDIOVASC SURG 74:875-889, 1977 4 Loop FD, Philips DF, Roy M, Taylor PC, Groves LK, Effler DB: Aortic valve replacement combined with myocardial revascularization. Late clinical results and survival of surgically-treated aortic valve patients with and without coronary artery disease. Circulation 55: 169-173, 1977 5 Flemma RJ, Johnson WD, Lepley D, Auer JE, Tector AJ, Blitz J: Simultaneous valve replacement and aorta-tocoronary saphenous vein bypass. Ann Thorac Surg 12:163-170,1971 6 Berndt TB, Hancock EW, Shumway NE, Harrison DC: Aortic valve replacement with and without coronary artery bypass surgery. Circulation 50:967-971, 1974 7 Richardson JV, Kouchoukos NT, Wright JO, Karp RB: Combined aortic valve replacement and myocardial revascularization. Results in 220 patients. Circulation 59:75-81, 1979 8 Grunkemeier GL, Starr A: Actuarial analysis of surgical results. Rationale and method. Ann Thorac Surg 24:404408, 1977 9 Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel N, McPherson K, Peto J, Smith PG: Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples. Br J Cancer 35:1-39, 1977 10 Ederer F, Axtell LM, Cutler SJ: The relative survival rate. A statistical methodology. National Cancer Institute Monograph No.6, Cancer: End Results and Mortality Trends, Bethesda, 1961, pp 101-121 11 Dash H, Johnson RA, Dinsmore RE, Harthorne JW: Cardiomyopathic syndrome due to coronary artery disease. I. Relation to angiographic extent of coronary disease and to remote myocardial infarction. Br Heart J 39:733739,1977 12 Culter SJ, Ederer F: Maximum utilization of the life table method in analyzing survival. J Chron Dis 8:699-712, 1958 13 Cooley DA, Dawson JT, Hallman GL, Sandiford FM, Wukasch DC, Garcia E, Hall RJ: Aortocoronary saphenous vein bypass. Results in 1,492 patients, with particular reference to patients with complicating features. Ann Thorac Surg 16:380-390, 1973 14 Anderson RP, Bonchek LI, Wood JA, Chapman RP, Starr A: The safety of combined aortic valve replacement and coronary bypass grafting. Ann Thorac Surg 15:249255, 1973 15 Wisoff BG, Fogle R, Weisz D, Garvey J, Hamby R: Combined valve and coronary artery surgery. Ann Thorac Surg 29:440-443, 1980 16 Merin G, Danielson GK, Wallace RB, Rutherford BD,

Aortic valve replacement and coronary bypass 7 1 1

Pluth JR: Combined one-stage coronary artery and valvular surgery. Circulation 48:Suppl 3:173-176,1973 17 Lundell DC, Laks H, Geha AS, Khachane VB, Hammond GL: The importance of myocardial protection in combined aortic valve replacement and myocardial revascularization. Ann Thorac Surg 28:501-508, 1978 18 Starr A: In discussion of Lundell et ali' 19 Macmanus Q, Grunkemeier GL, Lambert LE, Teply JF, Harlan BJ, Starr A: Year of operation as a risk factor in the late results of valve replacement. J THORAC CARDIOVASC SURG 80:834-841,1980 20 Rahimtoola SH, Grunkemeier GL, Teply JF, Lambert LE, Thomas DR, Suen Y, Starr A: Changes in coronary bypass surgery leading to improved survival. JAMA 246:1912-1916,1981 21 Bonow RO, Kent KM, Rosing DR, Lipson LC, Borer JS, McIntosh CL, Morrow AG, Epstein SE: Aortic valve replacement without myocardial revascularization in patients with combined aortic valvular and coronary artery disease. Circulation 63:243-251, 1981 22 Bristow JD, Kremkau EL: Hemodynamic changes after valve replacement with Starr-Edwards prostheses. Am J Cardiol 35:716-722, 1975 23 Pyle RB, Mayer JE, Lindsay WG, Jorgensen CR, Wang Y, Nicoloff DM: Hemodynamic evaluation of LilleheiKaster and Starr-Edwards prostheses. Ann Thorac Surg 26:336-343, 1978 24 Linhart JW, de la Torre A, Ramsey HW, Wheat MW: The significance of coronary artery disease in aortic valve replacement. J THORAC CARDIOVASC SURG 55:811-819, 1968 25 Iskandrian AS, Lichtenberg R, Segal BL, et al: Assessment of jeopardized myocardium in patients with onevessel disease. Circulation 65:242-247, 1982 26 Lytle BW, Loop FD, Thurer RL, Groves LK, Taylor PC, Cosgrove DM: Isolated left anterior descending coronary atherosclerosis. Long-term comparison of internal mammary artery and venous autografts. Circulation 61:869874, 1980 27 Kirklin JW, Kouchoukos NT: Aortic valve replacement with myocardial revascularization. Circulation 63:252253, 1981 28 Harris PJ, Phil D, Harrell FE, Lee KL, Behar VS, Rosati RA: Survival in medically treated coronary artery disease. Circulation 60:1259-1269, 1979 29 Hancock EW: Aortic stenosis, angina pectoris, and coronary artery disease. Am Heart J 93:382-393, 1977 30 Coleman EH, Soloff LA: Incidence of significant coronary artery disease in rheumatic valvular heart disease. Am J Cardiol 25:401-404, 1970 31 Jones EL: The conduct and results of combined operations: Coronary artery bypass and valve replacement; coronary artery bypass and brachiocephalic arterial operation, PostGraduate Course-Cardiac Surgery, Sixty-seventh Annual Clinical Congress, American College of Surgeons, San Francisco, Calif., 1981, pp 11-15