|
|
||||||||
Ann Thorac Surg 2002;74:102-108
© 2002 The Society of Thoracic Surgeons
a Department of Cardiothoracic, Friedrich-Schiller-University Hospital Jena, Jena, Germany
b Department of Vascular Surgery, Friedrich-Schiller-University Hospital Jena, Jena, Germany
c Department of Medical Statistics, Informatics, and Documentation, Friedrich-Schiller-University Hospital Jena, Jena, Germany
Accepted for publication March 19, 2002.
* Address reprint requests to Dr Albes, Department of Cardiothoracic and Vascular Surgery, University Hospital Jena, Jena 07740, Germany
e-mail: johannes.albes{at}med.uni-jena.de
| Abstract |
|---|
|
|
|---|
Methods. Five hundred eighteen patients undergoing isolated coronary artery bypass grafting were investigated. Thirty-nine patients underwent acute revascularization because of enzyme-proven or electrocardiogram-proven MI accompanied by crescendo angina, hemodynamic compromise, or both. They were compared with 33 emergent, 63 urgent, and 383 elective patients without MI. Preoperative risk factors for early mortality and necessity of continuous venovenous hemofiltration were analyzed by means of logistical regression analysis. Perioperative data were compared.
Results. Early mortality of the MI cohort was 15.4%, in contrast to 15.2% in emergent, none in urgent, and 2.1% in elective patients. Left internal thoracic artery was used in 87% of MI, 97% of emergent, 94% of urgent, and 97% of elective patients. Intraaortic balloon pump was necessary in 50% of MI patients, 27% of emergent, 6.3% of urgent, and 3.1% of elective cases. Continuous venovenous hemofiltration was performed in 29% of MI patients, 15% of emergent, 4.9% of urgent, and 3.4% of elective patients. Hemodynamic instability significantly increased the odds ratio for early mortality and continuous venovenous hemofiltration.
Conclusions. Patients undergoing acute revascularization carried an elevated risk to die early notwithstanding the presence or absence of acute MI. Liberal use of left internal thoracic artery grafts was not detrimental in acute patients whereas liberal use of intraaortic balloon pump was beneficial. In almost 30% of MI patients, continuous venovenous hemofiltration was not necessary, implying a severely impaired perioperative hemodynamic condition. Immediate revascularization in the presence of acute MI is therefore indicated although it may be addressed as a separate high-risk group.
| Introduction |
|---|
|
|
|---|
| Patients and methods |
|---|
|
|
|---|
Myocardial infarction and emergent revascularization were defined when patients were directly transferred from the catheter laboratory to the operating theater. Owing to specific regional characteristics of our hospital, which is located in a rural area with a variety of small external referring catheter laboratories, the transferral time frame occasionally exceeded 6 hours. Urgent revascularization was defined when the operation was performed within 48 hours after invasive diagnosis whereas all other patients were defined as elective cases. Acute MI was proven by means of respective electrocardiogram alterations (pathologic Q-wave, S-T elevation, T-wave alteration) or a significant (minimum 10-fold) increase of initial postoperative (<20 minutes) troponin I [9, 10], as well as significant creatine kinase isoenzyme (CK-MB) serum levels. Preoperative risk factors were obtained. Preoperative hemodynamic stability status was assessed at anesthesia, and patients were stratified into three groups: stableconscious, no hypotension, no inotropic support; unstablehypotension, low cardiac output, necessity of inotropic support; resuscitationnecessity of resuscitation in the operating theater. Preoperative intraaortic balloon pump (IABP) was not used in the acute patients. Complete revascularization as well as the use of the left internal thoracic artery (LITA) was the operative aim. Therefore, LITA was occasionally prepared after emergent institution of extracorporeal circulation. Indication for continuous venovenous hemofiltration (CVVH) was made clinically with respect to the expected course of renal function: Severe decrease of urine production to less than 0.5 mL/kg per day (30 to 50 mL/h) despite high-dose furosemide support (>1 g/d) was accompanied by massive fluid overload (central venous pressure of >18 mm Hg) or subsequent serum potassium levels of more than 6.5 mmol/L. An IABP was implanted intraoperatively when cardiac output appeared to be inadequate. An 8F balloon catheter was percutaneously advanced by means of the femoral artery without a sheath. Indication for extracorporeal membrane oxygenation was made on a highly individual basis: persistent low cardiac output accompanied by lactate increase under appropriate catecholamine support when an IABP was already in place. Type of cardioplegia, number and type of grafts, necessity of IABP, extracorporeal membrane oxygenation, necessity of temporary or permanent CVVH, hospitalization, EM, Cleveland-Score, and EuroSCORE were assessed and statistically compared among all groups. Cleveland-Score and EuroSCORE were used for operative risk assessment. Both scores are validated risk scores [11, 12] for cardiac procedures comprising comorbidities, demographic data, cardiac function, and previous cardiac procedures.
Statistical analysis was performed using SPSS statistical package (SPSS Inc, Chicago, IL) for Windows (Microsoft Inc, Seattle, WA). Numeric variables with normal distribution were analyzed by means of analysis of variance and post hoc comparisons with Tukey-HSD (honestly significant difference) adjustment. Nonnormally distributed variables were analyzed by means of Kruskal-Wallis test and subsequent Mann-Whitney U test. Categorical variables were analyzed by means of Pearson
2 test with
adjustment. Logistical regression analysis of preoperative risk factors was performed to assess the odds ratio for EM and necessity of CVVH. Data are shown as mean percentages or means ± standard deviation (SD). Odds ratios are presented with 95% confidence intervals. Significance was assumed if p was <0.05. Analysis of variance, Kruskal-Wallis, or Pearson
2 test p values or Students t test or Mann-Whitney U test p values between specific groups are shown.
| Results |
|---|
|
|
|---|
|
Troponin and cardiac enzymes
The patients suffering from acute MI showed the highest initial peak serum troponin I and CK-MB levels (troponin I, 32.7 ± 37.6 ng/mL [normal range, 0.1 ng/mL]; CK-MB, 1.58 ± 1.28 µmol/L x s [normal range, 0.16 µmol/L x s]. In the emergent cohort, troponin I levels and CK-MB levels were lower than in the acute group but higher than in the urgent cohort (troponin I, 2.02 ± 3.9 ng/mL; CK-MB, 0.29 ± 0.24 µmol/L x s). In the urgent group, troponin I levels were slightly elevated (troponin I, 0,91 ± 1.03 ng/mL; CK-MB, 0.21 ± 0.13 µmol/L x s). The differences of troponin I and CK-MB among the MI and both the emergent and urgent groups were statistically significant (p < 0.001), whereas the observed differences between emergent and urgent patients did not exhibit statistical significance.
Type of cardioplegia
All but one of the MI and the emergent patients, respectively, received cold blood cardioplegia (96%; 95%) in contrast to 61% of the urgent patients and 47% of the elective patients. The remaining patients received crystalloid cardioplegia.
Number and type of grafts
The majority of all patients received a LITA graft to the left anterior descending artery. Although in the MI group the percentage was lower than in the other groups, the difference was only significant when compared with the elective patients. Ninety-two (24%) of the elective patients received more than one artery (LITA, right internal thoracic artery, or radial artery); in 62 (16%) of the elective cases the revascularization was entirely arterial. Because of the higher number of total arterial revascularizations, the average number of venous anastomoses in the elective group was significantly lower than in all other groups (Table 2).
|
Extracorporeal membrane oxygenation
Extracorporeal membrane oxygenation was instituted in one instance in all groups. The elective as well as the MI patient died, whereas the 2 patients of the urgent and the emergent groups survived (Table 2).
Continuous venovenous hemofiltration
Use of CVVH was necessary in 28.2% of MI patients, 15.2% of emergent cases, 4.8% of urgent patients, and 3.2% of elective patients. Three of the 11 patients in the MI group died, 4 of the 5 patients in the emergent groups died, whereas none of the urgent and 5 of the 12 patients in the elective cohort died. The differences between the total number of CVVH of the MI and the emergent patients were statistically significant when compared with the total number of urgent as well as elective cases, whereas no statistically significant differences were seen between MI and emergent as well as between urgent and elective patients. Hemofiltration became permanent in 10% of the MI group, 9% of emergent patients, 5% of the urgent cohort, and 1.6% of elective cases (Table 2). Female sex, age, urgency status, and stability significantly affected the odds ratio for the necessity of postoperative CVVH whereas preoperative serum creatinine or compensated renal function did not show a significant increase (Table 3).
|
|
|
| Comment |
|---|
|
|
|---|
For many years studies using the huge numbers of coronary artery bypass grafting could demonstrate distinct differences in the outcome of patients stratified into typical risk groups with regard to the urgency of surgical intervention. It was demonstrated that urgent, emergent, or elective revascularization significantly affects EM and morbidity, as well as late mortality [16]. Reinecke and colleagues [4] compared historic data with recent results of acute coronary artery bypass grafting and revealed a risk progression of emergency coronary artery bypass grafting as a consequence of increasing comorbidities. Several authors have already analyzed emergent revascularization after failed PTCA and reported that it consistently showed an EM between 10% and 15%. Unfortunately, the presence of acute MI has not been assessed differentially in these studies [4, 13, 17].
In our study, electrocardiogram alterations served as one of the two indicators for acute MI. These were mainly S-T elevations or pathologic Q waves. Because of the rapidly changing electrocardiogram pattern during the evolving phase of acute MI, a differential analysis on the basis of electrocardiogram findings was not performed. Troponin I as well as CK-MB levels were used for further verification of acute MI. The observed differences were therefore expected. Surprisingly, however, emergent patients also exhibited higher values than the urgent group. This may point to some acute but minor myocardial damage in the emergent cohort. Our findings support the theory that ischemia cannot necessarily be separated sharply from cellular damage [18, 19]. A substantial number of acute MI patients were transferred to our unit without interventional or therapeutic measures. In our rural area we cooperate with several cardiologists working in small units, who are not as aggressive as the colleagues of a large university center.
We used blood cardioplegia in nearly all emergent situations and also in the majority of the urgent cases. Owing to financial reasons, however, crystalloid cardioplegia was used in half of the elective cases. Cold-blood cardioplegia has been unequivocally recommended as a sufficient myocardial protective agent [20]. Because blood cardioplegia was mainly used, both immediate groups received the most homogeneous and optimal myocardial protection measures of all four groups of our study. Therefore, inadequate myocardial protection did not play a detrimental role regarding the results of the immediate groups.
According to our general policy, almost 90% of the acute patients received an LITA graft to the left anterior descending artery. Although it has been suggested by other authors that in the acute situation arterial grafts may be avoided because of spastic reactions, we did not encounter hemodynamic problems caused by a compromised initial blood flow to the left anterior descending artery in acute patients [21]. Because of the absence of problems and considering the superior long-term patency rate of LITA grafts, we believe that they should indeed be used particularly in the acute setting. The number of venous grafts was lower in the elective group than in the emergent cohorts. This is caused by our regimen of performing total arterial revascularizations in the elective patients whenever the coronary morphology and availability of adequate arterial conduits allow it.
Half of the acute patients received an IABP. This policy is very much in concordance with other studies [8]. As we did not encounter any adverse effects of counterpulsation with regard to compromised blood flow of the intestinal or femoral arteries, we advocate a liberal use of IABP during acute revascularization. The necessity of CVVH reflected the severity of the hemodynamic situation of our immediately revascularized patients. However, prerenal failure occurred most often in MI patients, although a statistical difference from the emergent group was not observed. Fortunately, in only a minority of these patients, renal failure became permanent. As expected, the logistical regression analysis demonstrated that unstable preoperative conditions increased the odds for renal failure. However, female sex and age also had an influence on postoperative renal function, thereby contributing to the well-known facts that female and elderly patients represent a separate risk group in coronary surgical procedures [2].
The EM of the patients who were revascularized during ongoing MI was similar to the emergent patients without acute infarction. Acute MI therefore does not appear to be an independent predictor for early death [3]. In the literature, EM has often been found to be in the range of 10% to 20% in acutely revascularized cohorts independent from the presence of acute MI [4, 15]. In contrast, the prognosis of patients suffering from acute MI and hemodynamic compromise is poor when treated conservatively. Early death has been reported to range from 25% to more than 50% [3, 6]. The rationale for a surgical intervention is therefore to save the life of the patient, although the risks are high. In contrast, isolated elective coronary artery bypass grafting exhibits an EM of 2% to 3% in many large studies, which was confirmed by our results [21]. The urgent group revealed a surprisingly good outcome in this study. None of these patients died early, and few patients required perioperative counterpulsation although all remaining factors were similar to those obtained in the elective group. Urgent patients obviously represent a subgroup with an exceptionally low risk. It can be hypothesized that cardiologists refer those patients, who are not in a critical state of myocardial ischemia but exhibit severe stenoses, for an urgent procedure. The surgeon on the other hand may be more alert than in a routine procedure, and finally, less waiting time under unclear condition at home may also play a favorable role. The hypothesis that the odds for early death significantly increase when a patient is hemodynamically unstable or is suffering from impaired renal function could be proven by our study. Also female sex and age increased the risks for an adverse outcome, which has been shown earlier [16].
Risk scores are a valuable tool to estimate the particular risks of patients undergoing cardiac surgical procedures [12, 2224]. Both the Cleveland-Score and the EuroSCORE are easy to obtain as they comprise only 15 to 16 single paragraphs. Myocardial infarction adds to the risk score in the EuroSCORE although a differentiation between acute and subacute MI does not exist. In contrast, the Cleveland-Score does not itemize MI, but assigns a substantial amount of points for an emergency. Both scores did indeed disclose the particularly elevated risks of acute operation. However, the Cleveland-Score showed a trend to reflect the differences between both immediate groups statistically whereas the EuroSCORE failed to distinguish adequately.
Although both analyzed immediate groups represent inhomogeneous cohorts, our results show that distinct differences exist and can be used to aim at specific factors in subsequent studies on a prospective basis. All patients in our study were operated on using extracorporeal circulation. Recent papers suggested that acute revascularization be performed off-pump, although the presently published numbers of patients are small [25]. However, in the future, this approach may become feasible in certain subgroups of stable patients who require immediate revascularization and exhibit contraindications for extracorporeal circulation [26]. For these stable patients a preoperative IABP can be an additional option, if ventricular function is substantially reduced [26]. Those patients can then be transferred into the urgency status, which showed a favorable outcome in our study. Although we did not yet find criteria that are in favor of such an approach, identification of a respective subgroup may be an interesting topic for further studies.
We conclude that revascularization during acute MI carries a considerable but acceptable risk to die early, which is not substantially higher than the risk of patients undergoing an emergent procedure without acute MI. Revascularization can be successfully accomplished using blood cardioplegia and intensified hemodynamic support. The use of the LITA is recommended considering the absence of acute detriments and the improved long-term results. However, an increased perioperative morbidity must be expected.
Our current experience obtained under identical procedural characteristics in a relatively short period indicates that the current risk groupselective, urgent, and emergent revascularizationfind proper reflection in established score systems. However, an adjustment regarding the presence of acute MI may even improve the stratification properties.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Thielmann, M. Neuhauser, A. Marr, U. Herold, M. Kamler, P. Massoudy, and H. Jakob Predictors and Outcomes of Coronary Artery Bypass Grafting in ST Elevation Myocardial Infarction Ann. Thorac. Surg., July 1, 2007; 84(1): 17 - 24. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Voisine, P. Mathieu, D. Doyle, J. Perron, R. Baillot, G. Raymond, J. Metras, and F. Dagenais Influence of time elapsed between myocardial infarction and coronary artery bypass grafting surgery on operative mortality Eur. J. Cardiothorac. Surg., March 1, 2006; 29(3): 319 - 323. [Abstract] [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 |