Preoperative echocardiographic parameters influencing quality of life five years after coronary artery bypass graft surgery

Background/Aim. Factors associated with mortality and morbidity following coronary artery bypass graft surgery have been well defined and the Parsonnet score is widely used in mortality prediction. The evaluation of quality of life has not been still implemented in everyday work and preoperative echocardiographic factors affecting the quality of life in patients undergoing coronary artery bypass graft surgery have been poorly documented. The aim of this study was to evaluate echocardiographic parameters influencing quality of life following coronary artery bypass graft surgery and its correlation with widely used Parsonnet score. Methods. A total of 449 consecutive patients with myocardial revascularization, operated during 1999 and 2000 were enrolled in this retrospective-prospective study. The patients with comorbidities were excluded as well as those with in complete myocardial revascularization. A group of 180 patients who accepted to participate in quality of life evaluation was followed for 60 months. The quality of life was evaluated using a questionnaire SF-36. Results. The mean patients’ age was 57.8 ± 7.8 years, 79.4% were males. A 5-year survival was 84.2%. The mean number of risk factors was 3.4 ± 1.0. Most of the patients were in New York Heart Associatiation (NYHA) II class (104 of them or 59.4%), 61 of them (34.9%) in NYHA III class and only 10 patients or 5.7% of them were in NYHA IV class. The mean End-Diastolic Diameter (EDD) was 55.3 ± 5.6 mm, mean EndSystolic Diameter (ESD) 38.7 ± 5.6 mm and mean ejection fraction (EF) 51.7 ± 9.6%. Left atrium dilatation (p < 0.001), as well as left ventricle dilatation (p < 0.001), low left ventricle ejection fraction (p < 0.001), multisegmental disorders of contractility (p < 0.001), and severe mitral regurgitation (p < 0.001) were in negative correlation with almost all dimensions of quality of life. ROC analysis showed that left ventricle EDD of 54.5 mm can be used as good cut-off value for prediction of optimal quality of life, with sensitivity of 57% and specificity of 70% (RR = 1.386), left ventricle ESD of 37.5 mm with sensitivity of 65% and specificity of 57% (RR = 0.855) and left ventricle EF of 50% with sensitivity of 61% and specificity of 70% (RR = 0.916). Conclusion. Echocardiographic parameters, that can easily be obtained preoperatively, have strong predictive value not only in postoperative survival, but also in determination of the quality of life of the patients five years after coronary artery bypass graft surgery.


Introduction
Coronary artery disease (CAD) is a major cause of mortality and morbidity in developed countries and coronary artery bypass graft (CABG) surgery is a primary treatment option when intervention cannot be done, usually in cases of multivessel coronary artery disease.Coronary artery disease is also a leading cause of mortality and morbidity in Serbia.The number of CABG procedures performed each year in the United States has risen from approximately 150 000 in 1979 to 598 000 in 1996, that is a 425% increase 1 .In our aging patient population improvement of quality of life (QoL) becomes more and more primary indication 2,3 .Several studies have been undertaken to evaluate patients assessment of their general health status following CABG, mostly reporting an improvement [2][3][4][5][6][7] .Although a great number of studies analyzed perioperative mortality, survival and prediction of patients recovery after CABG, only few of them concerned quality of life long term after CABG.Parsonnet score is a widely used mortality predictive model after coronary revascularization and echocardiographic parameters are only a small part of this model.However, the predictors of QoL after myocardial revascularization are not completely recognized and the echocardiographic parameters have not been discussed enough yet, although they can easily be obtained preoperatively.
The aim of this study was to assess the preoperative echocardiographic parameters influencing QoL five years after primary isolated CABG.

Methods
The study enrolled 449 consecutive patients with primary isolated CABG done in the years 1999 and 2000 at the Clinic for Cardiac Surgery (Second Surgery Clinic), at the Clinical Center of Serbia.Comorbidities that could influence QoL, reported in previous studies as statistical bias, (renal failure serum, diabetes mellitus, liver cirrhosis, previous cerebral insult, malignancy) were exclusive criteria.Only 258 patients fulfilled inclusion and exclusion criteria and thus enrolled in the further investigation.Written informed consent was provided for all patients included in the study and Ethical Committee approved study methodology.
A follow-up period was 60 months for all the patients.A total 180 of the patients accepted to participate in the study and gave back a completely fulfilled questionnaire with a written informed consent.The answer rate was 69.8%.

Quality of life and echocardiographic parameters assessment
Quality of life was assessed 60 months after CABG using a questionnaire SF-36, that was sent to the patients by mail.A sample size of approximately 175 patients was calculated to be sufficient to give 90% power to detect an increase of 10% in the mean score in any subscale of the SF-36 at the 5% significance level.
The questionnaire SF-36 (UK standard version -Serbian translation) was sent with a stamped, addressed envelope to the patients, to fill them in and send them back by post.The SF -36 is a validated questionnaire widely used in medical practice and research, especially with cardiac surgery patients.It consists of 36 items measuring seven domains of QoL: physical functioning, physical role, body pain, general health, vitality, social functioning, emotional role and mental health.Scores range from 0-100 points; higher scores indicate better health.Internal consistency (Cronbachs alpha) for this subscale has been reported at 0.84, that is regarded as a high value comparing to other questionnaires used in the similar type of studies.All the participants were asked to complete questionnaires within two weeks.
An optimal quality of life score was regarded to be an overall questionnaire score ≥ 70% of maximum SF-36 score.
The parsonnet score was also calculated for all patients using the original Parsonnet formula and correlation of SF-36 score and Parsonnet score was analyzed.
All the patients had (ABG) surgery using the standard cardiopulmonary bypass technique and standard cardioplegia.None of them was operated 'off-pump'.
Standard echocardiography was performed using commercially available equipment with 2.0-4.0MHz transducer.The M-mode echocardiographic study of the left ventricle was performed under 2-D control.Left ventricle end diastolic (EDD) and end systolic (ESD) diameters, left ventricular posterior wall thickness and interventricular septal thickness were measured.Left ventricle ejection fraction (EF) was estimated by visual method.Echocardiographic parameters were measured according to the American Society of Echocardiography.Values of each parameter were obtained by the same examiner.
Echocardiographic parameters EDD, ESD, EF segmental disorders of contractility, existence of dilatative ischemic cardiomyopathy and severity of mitral regurgitation were analyzed.
The data are expressed as mean values with standard deviations for continuous variables or as absolute numbers with percentages for categorical data.Correlation coefficients with related p values were also reported.Univariante and Multivariante Logistic Regression Analysis were used to assess the relationship between QoL and several preoperative variables.Sensitivity and specificity of these preoperative variables of patients with CABG for the prediction of optimal quality of life score > 70% were determined by the ROC (Receiver Operating Characteristics) curve method, and "cutoff" level was determined.Sensitivity was defined as a number of optimal quality of life score patients with Ejection Fraction (EF) above "cut-off" or End diastolic diameter/End Systolic Diameter (EDD/ESD) levels below "cut-off" (test positive) / number of all patients with optimal quality of life score.Specificity was defined as a number of patients with suboptimal quality of life score patients and EF levels below "cut-off" or EDD/EDS levels above "cut-off" (test negative) / number of all patients with suboptimal quality of life score.An area under the curve (AUC) was calculated representing a quantitative measure of predictive value of EF and EDD/EDS for optimal quality of life score in the patients with CABG.In all tests, p value < 0.05 was considered sta-Zdravković M, et al.Vojnosanit Pregl 2009; 66(9): 718-723.tistically significant and p value < 0.01 highly statistically significant.

Results
The mean age of the patients was 57.8 ± 7.8 years.Most of them (139 patients or 77.2%) were males.
A five-year survival after coronary artery by-pass surgery was 84.2 %.
Even 92 patients (51.1% patients) had prior myocardial infarction in history of disease, and nine patients (5.0%) had also prior myocardial reinfarction.
Most of the patients were in New York Hear Association (NYHA) II class (104 of them or 57.8%), 61 of them (33.9%) in NYHA III class and only 15 patients or 8.3% of them were in NYHA IV class (Table 1).
Only 81 of the patients (44.6%) received arterial graft.Left ventricle dilatation was present in 31 of the patients (17.2%).Left ventricle hypertrophy was present in 78 patients (43.3%), but with higher frequency in female (59.1% in women and 27.8 % in men, p < 0.05).
The mean EDD was 55.3 ± 5.6 mm, mean EDS 38.7 ± 5.6 mm and mean EF was 51.7 ± 9.6 %.More than fifth of the patients (21.6%) had left atrial dilatation (more than 40 mm transversal diameter in M mode).
A total of 133 patients (73.9%) achieved optimal quality of life (> 70% of total score in all dimensions) after surgical revascularization, while 47 of them (26.1%) did not and thus regarded to have suboptimal QoL.
Patients with bigger left ventricle EDD had worse QoL in all aspects (p < 0.001).Left ventricle ESD was also in strong negative correlation with all aspects of quality of life (p < 0.01).The correlation with emotional role of quality of life and left ventricle ESD had important statistical significance (p < 0.05), although correlation was not as strong as with other QoL aspects (p < 0.01).
Posterior and interventricular septum wall thicknes was in correlation with none of QoL dimensions five years after coronary artery bypass surgery (p > 0.05).
Left ventricle EF was very important prognostic factor for QoL five years after coronary artery bypass surgery.There was positive correlation between left ventricle EF and all QoL aspects.Higher preoperative left ventricle EF was a strong predictor of better 5-year postoperative QoL.
A level of left ventricle systolic dysfunction (mild, moderate and severe) was also in very strong correlation with all QoL aspects (p < 0.001).There was an exception in emotional role wich was not affected by the level of left ventricle systolic dysfunction.
Left atrium dilatation was in negative correlation with all QoL dimensions.
Severity of mitral regurgitation was in a very strong correlation with almost all QoL aspects, exceptemotional role, wich was again not affected by severity of mitral regurgitation.
The presence of segmental disorders of myocardial contractility, as well as multisegmental disorders in myocardial contractility were in strong correlation with all QoL aspects five years after coronary artery the bypass surgery (p < 0.001) (Table 2).ROC analysis showed that left ventricle EDD of 54.5 mm can be used as a good cut-off value for an optimal quality of life, with sensitivity of 57% and specificity of 70%, left ventricle ESD of 37.5 mm with sensitivity of 65% and specificity of 57% and left ventricle EF of 50% with sensitivity of 61% and specificity of 70%.(Table 3).These echocardiographic parameters can be also used with high sensitivity and specificity in Parsonnet score calculation.Multivariante analysis revealed that preoperative left ventricle EF had the strongest predictive value in distinguishing future optimal quality of life five years following CABG (Table 3).
Left ventricle echocardiographic characteristics (EDD, ESD and EF) also had high sensitivity and specificity in prediction of Parsonnet score.
The Parsonnet score was in positive correlation with left ventricle EDD (r = 0.538, p < 0.001) and ESD (r = 0.495, p < 0.001), and in negative correlation with left ventricle EF (r = 0.461, p < 0.001), with high level of statistical significance in all cases (Table 3).
Analyzing relationship between the Parsonnet and SF-36 QoL score, we revealed very strong negative correlation: patients with a high Parsonnet score had low QoL score (r = 0.351, p < 0.001).
Relative risk and confidence intervals for echocardiographic parameters and the Parsonnet score were analyzed in Table 4.

Discussion
The main goal of CABG is relief of symptoms of angina pectoris and prolongation of life expectancy 1 .An improvement of QoL is also another very important endpoint 2- 6 .In Serbia, considering economic and war crisis, during the last years of 1990, an average waiting time for coronarography and following CABG operation was 12 months and in the time of operation a great number of patients had progression of ischemic heart disease.This article deals with the problem of preoperative echocardiographic parameters influence on quality of life 5 years after CABG and proper identification of risk group of patients.
Since previous studies showed that comorbidities (diabetes mellitus, renal failure, liver failure and malignancies) are very important factors of poor postoperative quality of life, this study was designed to exclude these patients in order to avoid comorbidity bias and to evaluate in the best way narrow field of echocardiographic parameters and its relationship with postoperative QoL [4][5][6] .This on the other hand, resulted in a smaller, but more homogenous group of patients, that can be seen in some similar studies 4,5 .
The mean patients age in our study was 57.8 ± 7.8 years and most of them were males, which is similar to results of other investigators [3][4][5][6][7][8][9] .They had a great mean number of risk factors; most of them were smokers with hyperlipoproteinemia and hypertension, with positive family history for cardiovascular diseases.But only third of them had obesity, although it is well known risk factor for coronary artery disease.More than a half of the patients had prior myocardial infarction (that is a higher proportion than in other studies) and 5% of the patients had prior myocardial reinfarction.This information had to be regarded in analysis of primary and secondary coronary artery disease prevention and used as a good measure of a delay in myocardial revascularization 7,10 .
Most of the patients were in NYHA II class and only 5.7% of patients in NYHA IV, which could be considered as a good marker of mostly slightly impaired left ventricular function in the time of surgery.
The mean values of left ventricle echocardiographic parameters showed that in the time of surgery in most cases left ventricle systolic function was not significantly impaired.Small proportion of patients with moderate mitral regurgitation could also be used as a good marker of mostly preserved left ventricle function in the group of patients.The patients with heavy mitral regurgitation were elected not only for surgical myocardial revascularization but also for concomitant mitral valve surgery, which was not the objective of this study.
A the significantly smaller ratio of patients who received arterial graft compared to other world cardiac surgery centers could be explained by the fact that our study analyzed patients operated 10 years ago, in 1999 and 2000 when this technique was not so frequently implemented as nowadays in our country.
Several new studies addressed QoL 12 and 18 months after bypass surgery but none of them analyzed preoperative echocardiographic parameters influencing QoL after CABG 9, 10   .Evaluation of QoL is difficult because it is not independent of subjective perception and static measurements cannot define it 11 .Besides questionnaire SF-36, which was used in our study, there are several other questionnaires used for evaluation of QoL (Physical activity score, Nottingham Health Profile and Psychological General Well-being Index), but SF-36 is highly validated.In this study a very strong cor-relation was found between all SF-36 aspects and echocardiographic parameters, as in several other studies [11][12][13] .
Poor preoperative EF, as a measure of poor systolic function, had a very important negative influence at all aspects of QoL in our study.The results of other studies are not uniform.Herlitz and al. 15,16 have shown in a group of 1904 patients that preoperative left ventricle EF had a huge impact on 5-year mortality but no influence on physical activity, symptoms of chest pain, neither dyspnea nor any indices of QoL after CABG.Evaluation of the quality of life in this study was done by different questionnaires, not used in our study: Physical Activity Score, Nottingham Health Profile and Psychological General Well-Being Index and the difference is a product of using different tools for QoL measurement.
Our results are in correlation with results of Rumsfeld and al. 17 who used the same method for evaluation of QoL as we did: SF-36 questionnaire.Their results showed in a group of 1973 patients that preoperative left ventricle EF indeed is very important predictor of postoperative QoL dimensions.Analyzing a group of patients with advanced left ventricular dysfunction Trachiotis and al. 18 also suggested that there was a higher mortality in patients with sequentially decreased left ventricular function undergoing coronary artery bypass grafting, although more than 60% of patients with an EF less than 25% were alive and had good control of angina after five year.Our results are also in concordance with previously mentioned, since Parsonnet score was in a highly significant correlation with echocardiographic parameters.
Cut-off values of left ventricle EDD, ESD and EF can be used with good sensitivity and specificity for evaluation and prognosis of optimal postoperative quality of life, similarly as for Parsonnet score prediction.Specifity is even higher regarding quality of life analysis compared to widely conventionally used Parsonnet score analysis 13 .Multivariante analysis showed that although all tested echocardiographic parameters could be used with high statistical confidence, EF had the best balance of high sensitivity and specifity in prediction of both Parsonnet score and QoL score.
The patients with multisegmental disorders in contractility had poorer QoL in all dimensions.The presence and degree of mitral regurgitation were also very important factors for postoperative QoL.This could be explained by haemodynamical changes in left ventricle function, since functional mitral insuffitiency is a well-known parameter of elevated enddiastolic left ventricle pressure and poor systolic and diastolic function [18][19][20] .
A strong correlation between Parsonnet and QoL score indicates that the patients with high Parsonnet score (and high risk of mortality after myocardial revascularization) are in the same time with a low QoL after myocardial revascularization.This group of patients is a group with high risk and should be properly preoperatively identified and more often controlled.
High relative risks (RR) suggest the importance of preoperative echocardiographic parameters and their influence on quality of life (SF-36 score) as well as postoperative surviving (Parsonnet score).

Conclusion
Echocardiographic parameters are easily obtained preoperatively.They have a strong predictive value in prediction not only postoperative mortality, but also in prediction of quality of life after surgical myocardial revascularization.Postoperative quality of life can be improved by early echo-cardiographic screening in patients with coronary artery disease before CABG surgery.Considering the results of this study it could be concluded that in the patients with mild left ventricular dysfunction myocardial revascularization could be performed relatively safely with an optimal improvement in all aspects of the quality of life.
These predictors of health-related quality of life after coronary artery bypass surgery may be useful in preoperative risk assessment and counseling of patients with regard to the anticipated health status outcomes.