|
|
||||||||
Department of Thoracic and Cardiovascular Surgery, University of Göttingen, Göttingen, Germany
Accepted for publication June 19, 2009.
* Address correspondence to Dr Danner, Department of Thoracic and Cardiovascular Surgery, University of Göttingen, Robert-Koch-Str 40, Göttingen, 37075, Germany (Email: drdanner{at}arcor.de).
| Abstract |
|---|
|
|
|---|
Methods: Over a 3-year period, patient data including acute myocardial infarction and emergent CABG were retrospectively reviewed. Univariate and multivariate analysis for in-hospital mortality was performed. The EuroSCORE analysis and follow-up was investigated.
Results: Overall in-hospital mortality was 18.3%. Preoperative cardiac related predictors for in-hospital mortality were cardiogenic shock (p < 0.001), very poor left ventricular function (p = 0.001), and ST-segment elevation (p = 0.012). In multivariate regression analysis, age, cardiogenic shock, and pulmonary hypertension were independent preoperative risk factors. According to the EuroSCORE, we could define three statistically different groups: intermediate-risk, high-risk, and very high risk, with an observed mortality of 3.3%, 20.0%, and 63.2%, respectively. The EuroSCORE correlates with but overestimates the mortality risk. In subgroup analysis, the creatine kinase-myocardial band/hour ratio for the intermediate-risk group and ST-segment elevation for the high-risk group were additional cardiac risk factors.
Conclusions: Patients with an acute myocardial infarction and emergency aortocoronary CABG have an elevated operative risk. Logistic EuroSCORE overestimates the mortality rate. Three different risk groups can be defined, in which creatine kinase-MB/h-ratio and ST-segment elevation can more accurately predict operative risk.
| Introduction |
|---|
|
|
|---|
The European System for Cardiac Operative Risk Evaluation (EuroSCORE) as a standardized risk score, not only in Europe, can predict the mortality risk [9, 10], but for patients with acute myocardial infarction, these predictors seem not to be accurate enough. We focused on these high-risk patients to find further preoperative risk factors that can be essential for decision making in the catheterization laboratory. Thus, the current study was undertaken to analyze factors for in-hospital mortality and for impact on the follow-up of patients undergoing operation on an emergent basis due to a myocardial infarction, either NSTEMI or STEMI.
| Material and Methods |
|---|
|
|
|---|
Hospital records of patients were reviewed to determine demographic, clinical, operative, and perioperative data. The maximum preoperative value of the myocardial biomarkers CK-MB (U · L–1) and troponin T (µg · L–1) were analyzed. Electrocardiogram abnormalities were categorized as STEMI or NSTEMI. Onset time of angina was given by the emergency records and was calculated as time span from onset of the symptoms to the beginning of the surgical intervention. For estimation of myocardial cell necrosis, the ratio of CK-MB and troponin T to this timespan was calculated, and is given as U/L–1 · h–1 or µg/L–1 · h–1, respectively. In-hospital mortality was defined as death during the periprocedural hospital stay; in our cohort, after the period of 100 days, either death occurred at the hospital or patients were discharged. Data were transformed to additive and logistic EuroSCORE values. The calculation was performed in accordance to the EuroSCORE definitions (available at: www.euroscore.org) [9]. Follow-up after discharge was given by medical records. The study was approved by the local Ethics Committee; patient consent was not necessary owing to the retrospective character of the study. No conflict of interest is declared.
Operative Procedure
All patients were operated on using standard cardiopulmonary bypass with ascending aortic and right atrial cannulation after median sternotomy. Mild hypothermia was applied, and cardiac arrest was induced either by cold crystalloid cardioplegia (Bretschneider) or antegrade and retrograde blood cardioplegia (Buckberg).
Statistical Analysis
Data are expressed for categorical variables as proportions (%) or number (n), and for continuous variables as mean ± SD. Differences in categorical variable were analyzed by Fisher's exact test, and for calculating the estimated odds ratio (OR), by the Mantel-Haenzel test. Differences in continuous variables were analyzed by independent Student t tests, controlled by the Levene test for equal variances. Results were considered significant if p values were less than 0.05; then, mean difference and confidence interval (CI) were given. Multivariate regression (likelihood function) analysis of preoperative factors was performed to identify independent factors for in-hospital mortality if p values were less than 0.05 in univariate analysis.
Kappa-means cluster analysis was performed to define groups, considering the predicted logistic EuroSCORE and observed mortality. After statistical definition of these groups, univariate risk factor analysis was performed to detect risk factors for in-hospital mortality in addition to EuroSCORE variables. Midterm survival (cardiac- and noncardiac-related deaths) was determined by the Kaplan-Meier method, and differences between groups were tested by the generalized Wilcoxon (Breslow) test.
Statistical modeling and testing were performed with SPSS statistical software (version 15.0.0; SPSS Inc, Chicago, IL).
| Results |
|---|
|
|
|---|
|
|
|
EuroSCORE Analysis
We could identify three statistically different groups related to the logistic EuroSCORE (additive EuroSCORE). In all three groups, the observed in-hospital mortality was lower than the predicted logistic EuroSCORE risk: intermediate group, 3.3% versus 17.5 ± 8.7% (additive 9.7 ± 2.0); high risk group, 20.0% versus 44.2 ± 11.2% (additive 13.9 ± 1.5); and very high risk group, 63.2% versus 74.5 ± 9.4% (additive 18.1 ± 1.8), respectively. Univariate analysis for preoperative risk factors beside the EuroSCORE variables for the three groups revealed additional predictors for mortality.
In the intermediate-risk group, the ratio of CK-MB per hour discriminates risk more accurately. Creatine kinase-MB was significantly lower (6.6 ± 5.5 versus 15.2 ± 1.9 U/L–1 · h–1, p = 0.036) in survivors versus nonsurvivors. The presence of STEMI, poor left ventricular function or cardiogenic shock, and other factors reached no statistical significance in this subgroup. In the high-risk group, the presence of STEMI was the only additional significant risk factor for in-hospital mortality (p = 0.005); in this group, no patient died in NSTEMI conditions. In the very high risk group, no further cardiac risk factor for in-hospital mortality was detected. All further preoperative factors (listed in Table 1) were not statistically different in this subgroup analysis for mortality. In all three groups, the EuroSCORE was not different between survivors and nonsurvivors. Disease characteristics for appropriate EuroSCORE calculations are listed in Table 4.
|
|
|
| Comment |
|---|
|
|
|---|
The EuroSCORE adequately determines mortality rate for medium- and low-risk groups, but for the high-risk group, no further subgroups are established [9, 10]. In our opinion, preoperative variables such as type of myocardial infarction, time point of intervention, or critical preoperative status are not well determined by the EuroSCORE. In acute ongoing myocardial infarction with failure of percutaneous intervention, further factors are desired for the prediction of the mortality risk for these patients, and particularly, there were no differences of EuroSCORE levels in the subgroups. The EuroSCORE correlates with prediction of mortality risk even for this distinguished group of patients with ongoing acute myocardial infarction. As have we, other groups also revealed an overestimation of the mortality risk with the EuroSCORE [12]. With acute myocardial infarction, nearly all patients have a higher EuroSCORE than 6 points. In this retrospective analysis, the mortality risk increases with higher EuroSCORE, but as high as a 10-point (intermediate risk) mortality rate seems rather low for these patients. It is influenced additionally by the mass of myocardial necrosis, which is indicated by the CK-MB ratio [13]. This factor could indicate patients who should be refused an emergent CABG, and that needs to be evaluated in a larger series. A STEMI was an additional risk factor in the high-risk group (additive EuroSCORE 11 to 17) and raises the mortality risk 6.7-fold compared with that for NSTEMI patients. A STEMI was more often correlated with cardiogenic shock or poor left ventricular function, which is also considered in EuroSCORE, but was an additional factor in this group.
A greater portion of myocardial necrosis is indicated by a higher CK-MB per hour ratio in STEMI patients, but it reached no statistical significance. That could have an influence on intervention time, but the analysis of time dependency (data not shown) did not reveal a significant difference between time groups, for patients with STEMI or NSTEMI. Time interval differences were reported for STEMI patients operated on at 3 to 8 days versus the first 24 hours. In many studies, mortality associated with STEMI decreased the more the surgical intervention was delayed, but when focusing on patients operated on immediately, similar mortality rates as high as 24% were seen [7, 8, 14]. Appearance of cardiogenic shock raises the mortality rate in large studies as high as 46% to 51%, and with early percutaneous intervention and CABG is as high as 40% to 59%, respectively [16, 17]. In our patient cohort, the mortality rate of patients in cardiogenic shock was 55%. If sufficient coronary blood flow is not reached, outcome is much worse, with a very high mortality rate of 78% [16]. These patients can profit from aggressive surgical management including complete revascularization.
There are several studies in which off-pump or beating-heart technique revealed decreased mortality rate, with significantly less bypass grafting, especially for the circumflex and right coronary artery, resulting in incomplete revascularization [15, 18]. Complete revascularization is an essential factor for long-term survival, and we reached a complete revascularization in 98% of patients [19, 20]. Our standard surgical approach is operation with cardiac arrest with cardioplegia. In most cases, we used antegrade and retrograde blood cardioplegia—which is superior to crystalloid cardioplegia [21–23]—but we found no statistically significant difference for in-hospital mortality rate.
Nonuse of the left internal thoracic artery was mostly in elderly patients and in those without left anterior descending artery stenosis, but in all these patients, revascularization was complete. Because the nonuse of left internal thoracic artery interfered with age, it was not an independent factor in multivariate analysis. The use of left internal thoracic artery is an important factor for late survival, and we used it in 90% of patients younger than 75 years [20].
After discharge from hospital, the follow-up was without relevant major cardiac adverse events, and the functional capacity of the patients compares mostly with NYHA class II or even less. That indicates high surgical performance and is probably based on complete revascularization and the consequent use of the left internal artery even in emergent circumstances. In our renewed high-risk groups (intermediate, high, and very high risk), we observed statistically different midterm survival rates, which were based on the in-hospital mortality rate. The slope of survival in the intermediate group is caused by the incidence of malignant neoplasm.
Limitations of the study were the retrospective setting of the study and the single-center database, but CABG emergently performed is in most studies a small subgroup [7, 19, 24], so a multicenter prospective protocol should be initiated. The three-group analysis is based on a rather low number of patients, and has to be validated in a prospective investigation.
In conclusion, patients with ongoing acute myocardial infarction and emergent need for CABG had a significantly higher in-hospital mortality risk that is correlated with but overestimated by the EuroSCORE. We could define three groups of intermediate risk, high risk, and very high risk, which had different additional cofactors for mortality prediction.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Siregar, R. H. H. Groenwold, F. de Heer, M. L. Bots, Y. van der Graaf, and L. A. van Herwerden Performance of the original EuroSCORE Eur J Cardiothorac Surg, January 26, 2012; (2012) ezr285v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. D. Hillis, P. K. Smith, J. L. Anderson, J. A. Bittl, C. R. Bridges, J. G. Byrne, J. E. Cigarroa, V. J. DiSesa, L. F. Hiratzka, A. M. Hutter Jr, et al. 2011 ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery: A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines Developed in Collaboration With the American Association for Thoracic Surgery, Society of Cardiovascular Anesthesiologists, and Society of Thoracic Surgeons J. Am. Coll. Cardiol., December 6, 2011; 58(24): e123 - e210. [Full Text] [PDF] |
||||
![]() |
Writing Committee Members, L. D. Hillis, P. K. Smith, J. L. Anderson, J. A. Bittl, C. R. Bridges, J. G. Byrne, J. E. Cigarroa, V. J. DiSesa, L. F. Hiratzka, et al. 2011 ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery: A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines Circulation, December 6, 2011; 124(23): e652 - e735. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |