HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Shigeki Morita Pablo Zubiate Gregory Louis Kay Jerome Harold Kay Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Morita, S. Articles by Kay, J. H. Search for Related Content PubMed PubMed Citation Articles by Morita, S. Articles by Kay, J. H.

J Thorac Cardiovasc Surg 1995;110:944-951
© 1995 Mosby, Inc.

SURGERY FOR ACQUIRED HEART DISEASE

PREDICTION OF LONG-TERM SURVIVAL BY PREOPERATIVE EXERCISE TESTING IN PATIENTS WITH DEPRESSED EJECTION FRACTION UNDERGOING MYOCARDIAL REVASCULARIZATION

Los Angeles, Calif.

Supported in part by Los Angeles Thoracic and Cardiovascular Foundation.

Received for publication April 28, 1994. Accepted for publication March 10, 1995. Address for reprints: Gregory Louis Kay, MD, 123 South Alvarado St., Los Angeles, CA 90057.

Abstract

Ejection fraction is a major determinant of morbidity and mortality for patients with ischemic heartdisease. Patients with an ejection fraction of 0.40 or less are generally recognized as having apoorer prognosis than those patients with an ejection fraction of 0.50 or better and remain a heterogeneous group. It would be useful if patients with a favorable surgical prognosis could be identified preoperatively. Fifty-five patients who underwent coronary artery bypass grafting and had an ejection fraction less than 0.40 (mean of 0.23 ± 0.07 standard deviation) were studied by catheter measurement of pulmonary arterial pressure and radionuclide left ventriculography. Heartrate, systemic blood pressure, pulmonary artery pressures, cardiac output, and ejection fraction were measured, at rest, after nitroglycerin was given intravenously and with supine bicycle exercise.Forty-seven patients who had follow-up longer than 4 years were divided into two groups according to their life status (alive or dead) 4 years after operation. Measured variables of exercise stress tests and clinical characteristics were entered into factor analysis to obtain a cardiac function factor score for predicting the life status after 4 years. The cardiac function factor score was highly loaded by stroke index (rest, nitroglycerin), cardiac index (exercise), systemic vascular resistance index(exercise), and history of congestive heart failure. The cardiac function factor provided a predictive value superior to that of any individual variable. By dividing the patients into two groups by cardiac function factor score, the actuarial 5-year survival was 72% versus 17% for good and poor prognosis groups, respectively (p < 0.0001). Preoperative exercise stress testing data integrated by factor analysis provide a predictive tool for patients with a low ejection fraction. (J THORAC CARDIOVASC SURG 1995;110:944-51)

Left ventricular function is an important determinant of survival and functional capacity in patients with ischemic heart disease. ¹ Our own experience ^2,3 and the experience of others demonstrate a substantial survival advantage for patients who undergo surgical revascularization compared with patients treated medically. ^4-11 Nonetheless, patients with poor ventricular function do not, as a group, do as well as patients witha normal ejection fraction (EF) who undergo revascularization. ^8-13 Since 1968 our group has undertaken myocardial revascularization for patients with poor ventricular function. The hospital mortality rate since 1973 is 7%. In 1984 we reported our experience inpatients with an EF of 0.20 or less. ³ We found that 5-year survival was influenced by the presence of congestive heart failure (CHF): the 5-year survival was 24% for patients with CHF at the time of operation, 40% for patients successfully treated for CHF before operation, and 60% for those patients who had never had CHF.Besides the presence or absence of heart failure, it would be of great use if we could identify other significant predictors of survival in these patients with high-risk conditions.

A number of methods have been proposed to predict the prognosis after myocardial revascularization for patients with ischemic heart disease. ^4,6,9,14 However, no study, to the best of our knowledge, has been undertaken relating preoperative exercise testing combined with radionuclide ventriculography and catheterization of the right side ofthe heart to prognosis in a cohort of patients with poor ventricular function who then undergo myocardial revascularization. To improve predictability, several variables were integrated by the use of a factor analysis technique to calculate a factor score, which we shall call thecardiac function factor (CFF).The usefulness of this CFF score was tested in terms of long-term actuarial survival.

METHODS

From February 11, 1980, through June 6, 1987, a total of 6166 coronary artery bypass operationswere done at our institution. Twenty-four percent (1455) of the patients had a left ventricular EF lessthan 0.40. Of those patients who were primarily referred to one surgeon (P.Z.), 73 patientsunderwent exercise stress testing as part of the preoperative evaluation. The mean EF in thesepatients was 0.23 ± 0.07 standard deviation by cineventriculography. There were 61 menand 12 women whose ages ranged from 37 to 79 years (mean 60 ± 9 years). All patientsenrolled in the test were believed to be questionable candidates for operation as judged by thesurgeon. After informed consent was obtained, the patient underwent catheterization of the rightside of the heart and gated radionuclide left ventriculography.

Of those 73 patients who underwent exercise testing, 17 patients (14 men and 3 women) wereexcluded from surgical treatment. Three patients did not have operable disease and six patientsrefused operation. Five patients had pulmonary hypertension that was not reduced by nitroglycerininfusion. In three patients, the EF was significantly decreased by nitroglycerin injection or byexercise. At the time, we thought that these results obtained from the nitroglycerin challenge andexercise stress test were contraindications to operation, although we no longer believe this isnecessarily the case. The average age of the medically treated patients was 62 ± 6 yearsand the New York Heart Association (NYHA) functional class was 3.31 ± 0.61. The mean EF determined by cineangiography was 0.22 ± 0.08. The actuarial 5-year survival ofmedically treated patients was 19%.

Fifty-six patients underwent operation. One patient who also required left ventricular aneurysmectomy was eliminated from the study group. The remaining 55 patients (mean age 60± 10 years, 46 men, 9 women) underwent myocardial revascularization.

Measurements.
With the patient placed in the supine position, electrocardiographic leads were placed to monitor lead II activity. A pulmonary artery pressure line capable of measuring cardiac output was inserted via an internal jugular vein and the radial artery was cannulated. Heart rate; systolic, diastolic, and mean arterial pressure; systolic and diastolic pulmonary artery pressure; pulmonary artery wedge pressure; and cardiac output were determined. Cardiac output was normalized by body surface area to give a cardiac index (CI). Gated radionuclide blood pool imaging studies were acquired in the left anterior oblique position and EF was measured. From these measurements, stroke index (SI =1000 x CI/heart rate), left ventricular end-diastolic volume index (EDVI = SI/EF), left ventricular end-systolic volume index (ESVI = EDVI - SI), and total systemic vascular resistance index (TSVRI = 80 x mean arterial pressure/CI) were calculated.

Protocols.
Three sets of measurements were obtained. First, the control measurement was done with thepatient in the supine position. Next, measurements were made after nitroglycerin (0.2 to 0.4 mg)was given intravenously to produce a reduction in systolic blood pressure of 10 to 20 mm Hg. After the systolic pressure returned to control levels, supine bicycle exercise was initiated.Hemodynamic variables were measured at the peak level of exercise. Exercise was discontinued if there were any ischemic electrocardiographic changes or if the patient reported exhaustion orangina. No measurement of peak work was made.

Before the test, the patient was given an intravenous injection of cold pyrophosphate (15 mg). After 30 minutes, 25 mCi of technetium 99m pertechnetate was injected. Gated blood pool imaging studies were done in the left anterior oblique position.

Surgical technique.
The operative technique used has been described previously by Zubiate, Kay, and Mendez. ² In brief, before cardiopulmonary bypass was begun, the proximal anastomoses were done by excluding a portion of the ascending aorta. The heart-lung machine was filled with lactated Ringer'ssolution. After completion of the proximal anastomoses, cardiopulmonary bypass was instituted andthe patient was cooled to a core temperature of 25° to 28° C. Potassium cardiac arrest was accomplished and the distal anastomoses completed. The left ventricle was not vented. A leftatrial pressure line was usually inserted via the right superior pulmonary vein. Care was taken toavoid ventricular distention. Isoproterenol or epinephrine was used judiciously for a cardiotonic effect. Nitroprusside was used as needed for afterload reduction. The intraaortic balloon assist device was used if adequate pressure was not maintained after cardiopulmonary bypass.

Statistics.
Four-year survival after operation was taken to be the end point. Of the 55 patients considered, 47 had such a complete 4-year history. These 47 patients were divided into two groups:nonsurvivors,which includes 17 patients who died within the 4-year period, and survivors, which includes 30 who survived beyond 4 years. For each hemodynamic variable, differences between groups were examined by means of t tests. For attribute clinical variables, ² or Fisher's exact tests were done. After the aforementioned univariate screening of variables associated with outcome was done, we performed a factor analysis on those hemodynamic variables significant atp = 0.10 or less. In this way, the variables were screened to see which appeared individually to discriminate between the two groups. To enter attribute variables into factor analysis, arbitraryscores were given, such as sex (1 = men, 2 = women), angina (0 = none, 1 = stable, 2 = unstable),and CHF (0 = no history, 1 = positive history but compensated, 2 = uncontrolled and symptomatic).

Factor analysis is a technique used to create a new variable, which in this case we call the CFF,that conceptually averages correlated variables. With factor analysis, we can create principal components that provide an adequate summary of the data or a set of variables being considered ¹⁵ (see Appendix A). A CFF was computed for each patient. A t test was done to see whether the CFF could discriminate between the groups. In addition, the relation between the CFF and the survival was determined by logistic analysis (see Appendix B).

Means and standard deviations were computed for all the continuous variables. Actuarial survival analysis was done by the Kaplan-Meier method. The difference between the cumulative survival between the two groups was determined by the Cox-Mantel test.

RESULTS

General information is summarized inTable I. Mean NYHA functional class was 3.3 ± 0.6. Thirty-eight patients (69%) had either compensated or symptomatic CHF. Eighteen patients (33%) had noangina. The average EF determined by cineangiography was 0.23 ± 0.07. Most of the patients (94%) received from two to four bypass grafts (mean 2.8 ± 0.8). No patient had a myocardial infarction within 30 days before the operation.

The CFF was significantly different between the two groups: -0.366 ± 0.826 in the survivor group versus 0.616 ± 1.142 in the nonsurvivor group (p = 0.0045). It had greater discriminant ability in predicting 4-year survival than any individual variable.We determined a CFF of 0.45 as a cutoff point to yield the highest ² value; 82.4% of the patients with a CFF less than 0.45 survived after 4 years, whereas 15.4% of those with a CFF higher than 0.45 died within 4 years (² = 18.3,p < 0.001). The discriminant ability of CFF was also tested in terms of actuarial life-table analysis with a follow-up period up to 7 years. All 55 patients were divided into two groups by CFF of 0.45.The two curves were different with a p value of less than 0.0001 (Fig. 1).

View larger version (19K): [in this window] [in a new window]   Fig. 1. Comparison of probability of survival after operation between patients whose CFF was greater than 0.45 and those with values less than 0.45.

View larger version (19K): [in this window] [in a new window]   Fig. 2. Comparison of probability of survival after operation between patients with exercise CI (CI Ex) greater than 3.5 L/min · m2 and those with values less than 3.5 L/min · m2.

View larger version (22K): [in this window] [in a new window]   Fig. 3. Values of 47 patients whom we were able to follow up to 4 years after operation were plotted toCFF.Reversed triangles represent values from patients who were alive 4 years after operation.Triangles show values from patients who did not survive 4 years after operation. Note that only two patient swere alive when the CFF was greater than 0.45. Regression analysis was done for patients' values plotted above to fit logistic model; probability (alive beyond 4 years) = 1/(1 + exp[-BO - B1 xCFF]). Best fit was obtained when parameters B0 and B1 were set to B0 = 0.6764 and B1 = -1.1853.

DISCUSSION

Our group has undertaken myocardial revascularization for patients with poor ventricular function for more than 20 years. We know from experience that some patients with low EF do extremely well,whereas others do not benefit from operation. In patients with poor ventricular function, EF is an incomplete and inadequate predictor of outcome. Postextrasystolic potentiation certainly implies viability of the myocardium. ⁴ A history of CHF is a particularly ominous sign. ³ The presence of angina is reassuring, but how bad a sign is the absence of angina?

Although individual clinical and hemodynamic factors may exhibit predictive ability, a combination of relevant clinical characteristics and hemodynamic variables by factor analysis produces a superior predictor, the CFF. Our findings have practical implication. Proper selection of candidates for revascularization from this population of patients can, at times, be as much an art as a science.

The patients subjected to study by our protocol were all patients to whom the surgeons "felt uncomfortable" about offering surgical treatment. Patients not treated surgically had a 5-year survivorship of 19%. The 5-year survival of patients in our series with a CFF greater than 0.45 was17%. This does not differ from the survival reported in the literature for patients treated medically. ¹⁶ It may not be reasonable to proceed with surgical therapy in such a patient.

For those patients with CHF at the time of evaluation for revascularization, a prohibitive CFF would provide the basis for delaying operation to allow for adequate medical treatment. If the heart failure can be controlled, one might expect enough improvement in the CFF to warrant operation at thattime. Further, for those patients with a favorable CFF, it is encouraging to the patient, cardiologist,and surgeon to know that, despite poor ventricular function, greater than 70% 5-year survival can be expected. This is important information to be able to convey to the patient and cardiologist when therapeutic options are being discussed.

In our series, CI during exercise was a significant predictor of the outcome. Franciosa ¹⁷ and Fuster ¹⁸ and their colleagues showed CI at rest was a predictor of survivorship in patients with heart failure caused by either coronary artery disease or idiopathic dilated cardiomyopathy. Observations have previously been limited to medically treated patients. To the best of our knowledge, no data have been available in the surgical literature that relate survival with preoperative CI, either at rest or during exercise, in patients with normal or depressed ventricular function. Further study is required to confirm our positive correlation between exercise CI and surgical survival.

The predictive ability of EF at rest was negligible. In the three protocols, only EF after nitroglycerin injection helped distinguish between the survivors and nonsurvivors (Table III). It is noteworthy also that there was no change in EF with exercise. Although the response of EFto exercise may be useful in predicting prognosis in some settings, it seems applicable only topatients with better ventricular function. ^19,20 The poor predictive ability of EF at rest is consistent with recent findings that showed a lack of correlation between EF and CHF ^21,22 or exercise capacity. ^23,24 The interpretation of EF is difficult, because EF is afterload dependent, ²⁵ and both contractility and vascular resistance may vary during exercise. In addition, diastolic function has been recognized as the cause for some forms of heart failure. ²² Our preliminary study shows that diastolic function is impaired in patients who had a poor response in CI to exercise. ²⁶ Further study is warranted to reveal the complicated relations among exercise, contractility,diastolic function, vascular resistance, and EF.

It would benefit the patients if we could obtain the CFF noninvasively. Of those 14 variables used to calculate CFF, those most highly loaded were related to cardiac output. Because cardiac output is closely related to oxygen consumption, ²⁷ oxygen consumption during exercise may be used as a substitute for cardiac output. Indeed, it has been reported that oxygen consumption at the peak level of exercise is related to the survival of patients with heart failure. ²⁴ Pulmonary artery pressure and EF can be estimated by echocardiography. Thus there is the possibility of reconstructing the CFF score noninvasively.

This study is retrospective. It suffers from the usual problems associated with retrospective studies.For the predictability of CFF score to be tested strictly, a prospective study should be done. The peak exercise level obtained for each patient was not noted. We can thus make no recommendations as to the minimum level of exercise needed to test the patient "fairly." This point may be moot, however, inasmuch as these patients have severelycompromised conditions and can only "do their best." Despite these limitations, we believe our method of predicting long-term survival for patients with ischemic heart disease with poor ventricular function is utilitarian and has benefit in the practical process of identifying which patient with a high-risk condition will benefit from surgical revascularization.

Appendix: APPENDIXES

Appendix A.
Factor analysis is a method of creating principal components that provide an adequate summary of the data or a set of variables being considered. When factor analysis is done, multiple factors are created. Because there were 14 variables in this study, 14 factors were created. Each factor is a linear equation of the 14 variables. In some cases, more than one factor may be necessary to adequately describe the dependence and independence among the set of variables being considered. One method to screen the factors is to calculate "eigenvalue." In our study, there were four factors that showed eigenvalue greater than one. The first factor, which wecalled CFF, showed highest eigenvalue of 5.017. The eigenvalues of the remaining three factorswere 2.703, 1.252, and 1.170. CFF explained 35.9% of the standard variance, whereas thepercentages explained by the remaining factors were 19.3%, 8.9%, and 8.4%, respectively. Wetherefore used only the first factor, CFF, which turned out to be a simple and effective way to predict long-term survival.

Appendix B.
The relation between CFF and 4-year survival was estimated by means of the logistic model;probability (alive beyond 4 years) = 1/(1 +exp [-B0 - B1 x CFF]), where B0 and B1 are the coefficients obtained from regression analysis. Shown inFig. 3 is the relationship between CFF and the estimated 4-year survival as expressed by the logistic model with the coefficients of B0 = 0.6764 and B1 = -1.1853. For example, there is an 86.5%probability of survival to 4 years after operations for this type of patient with a CFF = -1.0. There is a37.5% probability of survival to 4 years after operations for this type of patient with a CFF = 1.0.

Footnotes

*Current address: Department of Cardiovascular Surgery, Faculty of Medicine, Kyushu University,Fukuoka, Japan.

**Dr. Sun is responsible for the statistical work.

References

1. Massie BM, Conway M. Survival of patients with congestive heart failure: past, present, and future.Circulation 1987;75(Suppl):IV11-9.
   
2. Zubiate P, Kay JH, Mendez M. Myocardial revascularization for the patient with drastic impairment of function of the left ventricle. J THORAC CARDIOVASC SURG 1977;73:84-6.[Abstract]
   
3. Zubiate P, Kay JH, Dunne EF. Myocardial revascularization for patients with an ejection fraction of 0.2 or less. West J Med 1985;140:745-9.[Medline]
   
4. Cohn LH, Collins JJ Jr, Cohn PF. Use of the augmented ejection fraction to select patients with left ventricular dysfunction for coronary revascularization. J THORAC CARDIOVASC SURG1976;72:835-40.
   
5. Faulkner SL, Stoney WS, Alford WC, et al. Ischemic cardiomyopathy: medical versus surgical treatment. J THORAC CARDIOVASC SURG 1977;74:77-82.[Abstract]
   
6. Jones RH, Floyd RD, Austin EH, Sabiston DC Jr. The role of radionuclide angiocardiography in the preoperative prediction of pain relief and prolonged survival following coronary artery bypass grafting.Ann Surg 1983;197:743-54.[Medline]
   
7. Passamani E, Davis KB, Gillespie MJ, Killip T, CASS principal investigators and their associates. A randomized trial of coronary artery bypass surgery: survival of patients with a low ejection fraction. NEngl J Med 1985;312:1665-71.[Abstract]
   
8. Alderman EL, Fisher LD, Litwin P, et al. Results of coronary artery surgery in patients with poor left ventricular function (CASS). Circulation 1983;68:785-95.[Abstract/Free Full Text]
   
9. Balu V, Szmedra L, Dean D, Bhayana J. Long-term survival of patients with low ejection fraction.Tex Heart Inst J 1988;15:44-8.
   
10. Manley JC, King FH, Zeft HJ, Johnson WD. The "bad"ventricle: results of coronarysurgery and effect on late survival. J THORAC CARDIOVASC SURG 1976;72:841-8.[Abstract]
    
11. Vlietstra RE, Assad-Morel JL, Frye RL, et al. Survival predictors in coronary artery disease: medical and surgical comparisons. Mayo Clin Proc 1977;5:85-90.
    
12. Loop FD, Cosgrove DM, Lytle BW, et al. An 11 year evolution of coronary arterial surgery (1967-1978). Ann Surg 1979;190:444-55.[Medline]
    
13. Lawrie GM, Morris GC Jr, Calhoon JH, et al. Clinical results of coronary bypass in 500 patients at least 10 years after operation. Circulation 1982;66(Suppl):I1-5.
    
14. David YB, Shefer A, Weiss AT, et al. Early postoperative assessment of coronary artery bypass surgery using nuclear left ventriculography and atrial pacing. J THORAC CARDIOVASC SURG1983;31:377-81.
    
15. Cureton EE, D'Agostino RB. Factor analysis: an applied approach. Hillsdale, New Jersey: Lawrence Erlbaum Associates, 1983.
    
16. Califf RM, Harrell FE Jr, Lee KL, et al. The evolution of medical and surgical therapy for coronaryartery disease: a 15-year perspective. JAMA 1989;261:2077-86.[Abstract]
    
17. Franciosa JA, Wilen M, Ziesche S, Cohn JN. Survival in men with severe chronic left ventricular failure due to either coronary heart disease or idiopathic dilated cardiomyopathy. Am J Cardiol1983;51:831-6.
    
18. Fuster V, Gerish ER, Tajik AJ, Brandenburg RO, Frye RL. The natural history of dilated cardiomyopathy. Am J Cardiol 1981;47:525-31.[Medline]
    
19. Bonow RO, Kent KM, Rosing DR, et al. Exercise-induced ischemia in mildly symptomatic patients with coronary-artery disease and preserved left ventricular function: identification of subgroups atrisk of death during medical therapy. N Engl J Med 1984;311:1339-45.[Abstract]
    
20. Gibbons RJ, Lee KL, Cobb FR, Coleman RE, Jones RH. Ejection fraction response to exercise in patients with chest pain, coronary artery disease and normal resting ventricular function. Circulation1982;66:643-8.
    
21. Dodek A, Kassebaum DG, Bristow JD. Pulmonary edema in coronary artery disease without cardiomegaly: paradox of the stiff heart. N Engl J Med 1972;286:1347-50.
    
22. Dougherty AH, Naccarelli GV, Gray EL, Hicks CH, Goldstein RA. Congestive heart failure with normal systolic function. Am J Cardiol 1984;54:778-82.[Medline]
    
23. Franciosa JA, Park M, Levine TB. Lack of correlation between exercise capacity and indexes of resting left ventricular performance in heart failure. Am J Cardiol 1981;47:33-9.[Medline]
    
24. Cohn JN, Johnson GR, Shabetai R, et al. Ejection fraction, peak exercise oxygen consumption,cardiothoracic ratio, ventricular arrhythmias, and plasma norepinephrine as determinants of prognosis in heart failure. Circulation 1993;87(Suppl):VI5-16.
    
25. Kass DA, Maughan WL, Guo ZM, Kono A, Sunagawa K, Sagawa K. Comparative influence of load versus inotropic states on indexes of ventricular contractility: experimental and theoretical analysis based on pressure-volume relationships. Circulation 1987;76:1422-36.[Abstract/Free Full Text]
    
26. Morita S, Kay GL, Zubiate P, Mendez M, Kay JH. Diastolic function determines the cardiac output in patients with low ejection fraction secondary to ischemic heart disease [Abstract]. Circulation1989;80(Suppl):II275.
    
27. Koike A, Hiroe M, Adachi H, et al. Cardiac output–O₂uptake relation during incremental exercise in patients with previous myocardial infarction.Circulation 1992;85:1713-9.[Abstract/Free Full Text]
    

This article has been cited by other articles:

				S. Morita, G.-W. Sun, and G. L. Kay Predicting survivorship after coronary artery bypass grafting in patients with low ejection fraction J. Thorac. Cardiovasc. Surg., March 1, 1997; 113(3): 614 - 614. [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Shigeki Morita Pablo Zubiate Gregory Louis Kay Jerome Harold Kay Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Morita, S. Articles by Kay, J. H. Search for Related Content PubMed PubMed Citation Articles by Morita, S. Articles by Kay, J. H.