| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2005;27:861-869
© 2005 Elsevier Science NL
a Department of Thoracic and Cardiovascular Surgery, West-German Heart Center Essen, University Clinic of Essen, Hufelandstraße 55, 45122 Essen, Germany
b Institute for Medical Informatics, Biometry, and Epidemiology, University Clinic of Essen, Essen, Germany
Received 14 October 2004; received in revised form 19 January 2005; accepted 21 January 2005.
* Corresponding author. Tel.: +49 201 723 4928; fax: +49 201 723 5451. (E-mail: matthias.thielmann{at}uni-essen.de).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
Key Words: Coronary artery bypass grafting • Risk stratification • Patients outcome • Cardiac troponin I
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
To date, cardiac troponins have become the gold standard for the diagnosis of AMI and, in addition, are essential for the evaluation of ACS. However, cardiac troponins have not only improved AMI detection, but also enabled risk stratification in a variety of clinical settings. Numerous clinical studies revealed cardiac troponins as reliable and superior biomarkers for the prediction of short- and long-term prognosis and the probability of cardiac death in patients with acute coronary syndromes, such as unstable angina, non-ST-elevation or non-Q-wave MI [5]. An important prognostic role of the admission cTnT value in patients with ST-elevation MI has recently been demonstrated in large-scaled clinical studies [6,7]. Elevated cardiac troponins in the setting of percutaneous cardiac interventions (PCI) have been demonstrated to be strongly related to an increased probability of mortality [8,9]. In the setting of cardiac surgery, the relationship between postoperative release of cardiac troponins and patient outcome has been well described in several previous studies [10,11]. Whether preoperatively elevated cTnI serum levels similarly predict outcome in patients scheduled for elective CABG is currently unknown.
The presence of preoperative cardiac troponin I in the serum of patients undergoing elective CABG may be the result of unrecognized bygone, ongoing or even evolving small myocardial damage. Therefore, the present study focused on the clinical significance of a single preoperative cardiac troponin I serum level within 24h prior to surgery and its predictive value for postoperative morbidity, major adverse cardiac events and early mortality in patients undergoing elective coronary artery bypass surgery.
|
2. Patients and methods |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
2.2. Data management
All relevant preoperative, intraoperative, and postoperative data were prospectively recorded and documented with more than 1800 variables per case using a database tool according to the ‘Heidelberger Verein zur Multizentrischen Datenanalyse e.V.’ (HVMD) [12]. All patients gave their informed consent and the study was performed with institutional approval.
The primary study endpoints for comparison between the groups were in-hospital mortality, defined as all cause of death within 30 days after surgery or during the same time period of hospitalization as well as postoperative major adverse cardiac events (MACE) during the period of hospitalization including perioperative myocardial infarction (PMI) or low cardiac output syndrome (LCOS). Secondary study endpoints were the composite study endpoints (1) death or PMI and (2) death or PMI or LCOS as well as other postoperative complications like (3) stroke or minor adverse events like (4) new-onset ventricular arrythmia, (5) major bleeding, (6) neccessitiy for rethoracotomy or (7) postoperative renal failure requiring temporary hemodialysis.
The demographic characteristics studied as potential risk factors for in-hospital mortality and major adverse cardiac events and considered for inclusion in all the logistic regression analyses performed were: age, sex, and body mass index. The clinical and laboratory variables evaluated were left ventricular function, extend of coronary artery disease, left main stenosis, peripheral vascular disease, chronic obstructive pulmonary disease, smoking history, diabetes mellitus, arterial hypertension, hyperlipidemia, angina class (Canadian Cardiovascular Society), prior myocardial infarction, prior percutaneous coronary intervention (PCI), serum cardiac troponin I, serum creatine kinase activity and serum creatinine within 24h before surgery.
For risk analysis, patients were divided into three groups according to their preoperative cardiac troponin I serum levels, as previously described [13]. Group 1, negative and thus ‘normal’ cTnI serum levels within the institutions reference limit between 0.0 and 0.1ng/ml; Group 2, ‘low-level’ elevated and thus, ‘abnormal’ cTnI serum levels between 0.11 and 1.5ng/ml; Group 3, ‘conventional’ elevated cTnI serum levels above 1.5ng/ml.
Perioperative myocardial infarction was considered to have occurred, if one of the following diagnostic criteria were present. (1) A postoperaive cTnI serum level above 10.5ng/ml within the first 24h after CABG, as previously described [14]. (2) The appearance of ST-segment deviations at the J point in two or more contiguous leads with cut-off points 
0.2mV in leads V1, V2, or V3 and 0.1mV in other leads or T-wave abnormalities in two or more contiguous leads [4]. A postoperative LCOS was supposed to be present with a cardiac index below 2.0l/min/m2 or a systolic arterial pressure below 90mmHg, despite high-dose inotropic support (IV dopamine 8µg/kg/min or dobutamine 6µg/kg/min or epinephrine >0.1µg/kg/min or norepinephrine >0.1µg/kg/min) with or without the use of an IABP.
2.3. Surgical management
Internal thoracic artery and saphenous vein grafts were used as graft conduits in all patients. Proximal graft anastomoses to the aorta were performed with partial occlusion of the ascending aorta. Heparin was administered in order to achieve an activated coagulation time above 400s. Standard cardiopulmonary bypass (CPB) technique was used with ascending aortic and two-stage venous cannulation. During CPB, moderate hemodilution with a hematocrit level between 20 and 25% using mild systemic hypothermia (>32°C) was maintained. Myocardial protection was achieved using antegrade cold cristalloid (Bretschneider) cardioplegic arrest and additional topical cooling, and single aortic cross clamping for all distal anastomosis. Intraoperative Doppler graft flow measurement (Cardiomed, MediStim, Oslo, Norway) was routinely applied after CPB just before sternal closure under stable hemodynamic conditions for each graft.
2.4. Biochemical analysis
Venous blood samples were drawn from each patient preoperatively within 24h prior to surgery, and at 1, 6, 12, and 24h after aortic unclamping and were analyzed separately for cTnI. CTnI was measured using a specific two-side immunoassay (Dimension Flex®, Dade Behring Marburg GmbH, Marburg, Germany). The detection range for cTnI was 0.04–40ng/ml, requiring further dilutions if necessary. The assays reference interval was 0.00–0.05ng/ml. A cTnI value above 0.1ng/ml was considered as abnormal.
2.5. Statistical analysis
Data are reported as mean values±standard deviation (SD) and categorical variables by their number and summarized as percentage. For all categorical variables the odds ratios (ORs) and 95% confidence intervals were calculated. Comparisons of categorical variables between groups were performed by Pearson's Chi-square test or the Cochran-Armitage trend test (using equally spaced scores) [15], since expected frequencies <5 occurred all P-values were calculated exactly. Comparisons of continuous variables between groups were analyzed by Kruskal–Wallis test for one-way ANOVA or Jonckheere–Terpstra's trend test [16]. When a significant overall effect was detected, two group comparisons were performed with Fisher's exact test for categorical variables or the Mann–Whitney U test for continuous variables. Univariate and multivariate logistic regression analyses were performed to identify preoperative independent predictors for in-hospital mortality and postoperaive major adverse cardiac events like perioperative myocardial infarction and low cardiac output syndrome, respectively, as well as the cumulative risk of the composite study endpoints. Those variables identified by univariate regression analysis as a significant factor for at least one study endpoint were added to a multivariate logistic regression analysis model. A P value less than 0.05 was considered to indicate statistical significance. All statistical analyses between groups were performed using the StatXact 6.0 software (Cytel Software Corp., Cambridge, MA, USA). Logistic regression analyses were performed using the SAS System® (SAS Institute, Inc., Cary, USA), version 8.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
0.1ng/ml) was present in 1178 (83.8%), whereas 163 (11.6%) patients in group 2 could be observed with a low-level elevation of cTnI (cTnI 0.11–1.5ng/ml) and 64 (4.6%) patients in group 3 were found with a higher preoperative cTnI elevation (cTnI >1.5ng/ml). The demographics and baseline data of the groups according to the preoperative cTnI value are summarized in Table 1.
|
|
|
3.4. Patients outcomes and major adverse cardiac events
Among the 1405 patients of the groups with preoperative cardiac troponin I measurement, 29 (2.1%) died in the postoperative course within 30 days or within the same time of hospital stay. There was no patient, who died intraoperatively. There were 20 deaths among the 1178 patients with negative preoperative cTnI levels in group 1, 5 deaths among 163 patients with preoperative low-level cTnI elevation in group 2 and 4 deaths among 64 patients with a preoperative cTnI level above 1.5ng/ml in group 3. A one-sided Cochran-Armitage trend test showed a significant difference of in-hospital mortality between the groups (1.7 vs. 3.1 vs. 6.3%; P<0.02). The difference of in-hospital mortality was accompanied by a significant difference in the appearence of postoperative low cardiac output syndrome (LCOS; P<0.0001) and the incidence of perioperative myocardial infarction (PMI; P<0.001). Individuals with preoperatively increased cTnI level were at increased risk for in-hospital death and major adverse events, including a two- to threefold higher risk of suffering death, LCOS or PMI. Furthermore, the composite study endpoints of death or PMI (P<0.0001), and death, PMI or LCOS (P<0.0001) did further increase this highly significant difference between the groups (Fig. 2). According to the survival status, all adverse events and postoperative complications were significantly different.
|
3.6. Univariate and multivariate logistic regression analyses
Univariate logistic regression analyses identified some preoperative variables like age, angina class III–IV, preoperative cTnI serum level as a continuous variable and preoperative cTnI serum level above 1.5ng/ml stratified as group 3 as preoperative variables strongly associated with death, PMI, and LCOS, and with the composite study endpoints (Table 3). In a multivariate logistic regression analysis considering age, COPD, PVD, angina class III–IV, and preoperative cTnI levels stratified as groups 1–3, age, PVD, and cTnI >1.5ng/ml were siginficantly associated with death, whereas angina class III–IV and cTnI >1.5ng/ml were associated with PMI and only cTnI >1.5ng/ml was strongly associated with LCOS. By using the composite endpoints only angina class III–IV and cTnI >1.5ng/ml were associated variables. Considering preoperative cTnI levels in a multivariate logistic regression analysis as a continuous variable age, PVD, angina class III–IV and preoperative cTnI levels were associated with death, and angina class III–IV and cTnI with PMI and composite study endpoints, whereas preoperative cTnI level was found to be most strongly associated with PMI and LCOS and all composite study endpoints (Table 4).
|
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
0.1ng/ml), ‘low-level’ (group 2, 0.11–1.5ng/ml), and ‘conventional’ (group 3, 1.5ng/ml) elevated cTnI serum levels as well as preoperative cTnI levels as a continuous variable. Patients who had undergone an isolated elective CABG with preoperatively ‘low-level’ increased cTnI values were at an approximately twofold higher risk and patients with ‘conventional’ elevated cTnI values at an appromimately threefold higher risk of suffering death in hospital. Thus, a preoperatively elevated cTnI level prior to CABG could be identified as an independent predictor of surgical risk after adjustment for other confounding risk factors in a multivariate logistic regression analysis. However, not only the risk for in-hospital mortality, but also for postoperative major adverse cardiac events like PMI or LCOS was shown to be increased with preoperatively elevated cTnI serum levels. The introduction of highly sensitive and specific serological markers, such as cardiac troponins, fundamentally improved the clinical diagnostic criteria for the detection of acute myocardial infarction, thus leading to a redefinition of AMI [4]. Elevated cardiac troponins have been shown to be associated with adverse outcome and increased mortality rates in patients with acute STEMI [6], but also minor elevations of cardiac troponin among patients with non-ST-elevation acute coronary syndromes were demonstrated to have a higher risk of death and reinfarction [5]. It has also been demonstrated, that minor elevations of cardiac troponins and thus, a ‘minor myocardial damage’, is present in approximately 30% of patients with rest angina and negative CK/CK-MB values [18].
In the setting of acute coronary syndromes, elevations of cardiac troponins were found to be associated with multivessel disease, complex coronary lesions with unstable and ruptured plaques, distal coronary microembolization of platelet micro-aggregates and plaque debris [19] as well as abnormal microvascular myocardial perfusion [7]. However, cardiac troponins not only improved AMI diagnostics, but also enabled risk stratification and prediction of patients outcome in a broad variety of clinical scenarios, such as acute coronary syndromes [5,8], massive pulmonary embolism [20], early sepsis [21], and congestive heart failure [22]. Elevated levels of cardiac troponins on admission have been unequivocally demonstrated to predict the clinical outcome and a more complicated course after primary PCI for acute STEMI [6,7]. Patients with acute myocardial infarction and an elevated cTnI level at admission were found to be associated with a lower reperfusion rate on the one hand, and a lower success rate for primary PCI, a higher risk for in-hospital death, congestive heart failure, and a higher incidence of long-term cardiac mortality and of cardiac events.
Furthermore, cardiac troponins were also studied in several non-cardiac [23] and cardiac [10,11] surgical settings, in which most of the studies proposed a relationship between postoperative troponins elevations and short- and/or long-term outcome. In a recent clinical study, a single postoperative cTnT value was shown to enable the assessment of postoperative myocardial cell injury. This cTnT level was strongly correlated with long-term survival after cardiac surgery. Thus, cTnT was suggested as an additional independent variable to other established risk indicators [10].
In contrast to several studies, in which the postoperative release of cardiac troponin after cardiac surgery was associated with patient outcome, a previous study analyzed the possible relationship between preoperative myocardial cell injury, measured by cTnT, and the postoperative patient outcome. Carrier and co-workers could demonstrate that patients with preoperative positive (>0.02µg/l) serum levels of cTnT undergoing coronary artery bypass surgery were more likely to have postoperative complications and postoperative myocardial infarction in paticular [24]. However, in contrast to the present study, patients with preoperatively elevated cTnT levels were not at a higher risk for in-hospital mortality, although a significantly longer CPB time, more postoperative inotropic and mechanical support with a higher incidence of postoperative IABP support were reported. This is possibly explainable due to the smaller study size of patients in the study by Carrier et al. and, moreover, in the present study preoperative cTnI level was investigated as a continuous variable as well as a stratified variable in three different groups.
The present study did also show, that there is indeed a subgroup of elective CABG patients, in whom a ‘minor myocardial damage’ may rather be detected by an elevation of preoperative cTnI levels, than by the presence of angina or ECG abnormalities.
Numerous risk stratification systems have been developed over the recent years with variying complexity [25]. However, no variable has been incorporated so far inquiring a possible preoperative acute myocardial infarction or even the extent of acute myocardial cellular damage. The addition of a highly sensitive biomarker, such as cardiac troponin I, indicating myocardial cellular injury, that might be subclinical at the time of determination, may well enrich the exciting battery of preoperative risk stratification models for patients undergoing cardiac surgery. In addition, as shown by the results of the present study, the extent of preoperative cardiac troponin I elevation indicates the degree of myocardial cellular injury and thus, the patient surgical risk before CABG.
|
5. Clinical implications |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
Dr Thielmann: As I said, we derived the cutoff value of troponin I for the presence of a perioperative myocardial infarction from isolated CABG patients at our institution using the same cTnI test kit.
Last year in Vienna, I had the ability to present a paper where we could derive different cTnI cutoff levels by analyzing postoperatively CABG patients with evidence of PMI using acute repeat angiography. Those patients identified with an early graft failure had a cTnI cutoff level of about 32ng/ml after 24h compared to patients with a non-graft-related perioperative myocardial infarction with a very high sensitivity to predict graft failure. Patients with graft-related and non-graft-related PMI had a cut-off level of about 10.5ng/ml compared to a control group with patent grafts.
We evaluated the cutoff level of 10.5ng/ml within 24h after CABG by using ROC-analysis between groups with and without perioperative myocardial infarction.
The incidence of perioperative myocardial infarction of about 17.2% in group 3 is indeed significantly increased compared to groups 1 and 2, but clearly depends on the PMI diagnostic criteria. A cTnI cutoff level of 10.5ng/ml is a comparatively low cutoff point.
Mr D. Zindrou (London, UK): I wonder if troponin is a predictor of mortality or is it a marker?
And second, did you do interaction tests in your multiregression analysis to see whether Troponin is an independent predictor (i.e. not a cofounder)?
Dr Thielmann: I think with this presentation we could clearly show that, of course troponin I is a marker, but can predict mortality as well as PMI and low cardiac output syndrome. We performed univariate regression analyses and multivariate regression analysis, and we could show that cTnI is a predictor of these study end points adjusted for all other co-factors which are also predictive for death. So I think, yes, cardiac troponin I is not only a serum marker for AMI detection, but also an independent predictor for death, PMI and low cardiac output syndrome.
Dr T. Wahlers (Jena, Germany): Have you done any quality assessments of your levels whether your measurements were always the same?
Dr Thielmann: Yes, our biochemical laboratory routinely did every day, of course.
Dr Wahlers: And there is no variability in assessing the troponin I level?
Dr Thielmann: There is a variability of about 10% at 0.05ng/ml, and these variabilities were routinely controlled, but I can't tell you exactly how great these variabilities are. I am sorry.
Dr M. Antunes (Coimbra, Portugal): I am not sure that the majority of surgeons in this room would measure routinely the troponin I prior to surgery. What made you do that in your patients?
Secondly, you have to derive some consequences from these results. Are you prepared to delay surgery on patients in whom you predict a higher risk?
Thirdly, I couldn't understand what can be the relationship between preoperative troponin I levels and the individual graft failure. Can you please explain that?
Dr Thielmann: The last question, I didn't say that preoperative troponin I is a predictor for graft failure. It is a predictor for death LCOS and PMI, but not for graft failure. And I don't have the first question.
Dr Antunes: My question is, are you now going to do something on those patients you measure a high level of troponin I preoperatively? Are you delaying surgery, are you doing something to change the outcome?
Dr Thielmann: I think we have to change something. These studies are absolutely new. I don't have any data about that, but maybe we have to postpone for some days the CABG.
Dr T. Aberg (Umea, Sweden): Well, I think this is an exciting area, because obviously in the broad group of elective coronary artery surgery, there is a small subgroup which really should be into the unstable angina category, and we now have a good tool to detect that. I think the handling of this new knowledge will have to be new studies.
Dr Thielmann: As I said, the American College of Cardiology and the European Society of Cardiology redefined myocardial infarction because of those biomarkers like troponins, and if you measure preoperatively cardiac troponin, it is not just a biomarker, it is myocardial damage, you measure the myocardial damage preoperatively, and I think we have to be aware about those microinfarctlets, let's say.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Presented at the joint 18th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 12th Annual Meeting of the European Society of Thoracic Surgeons, Leipzig, Germany, September 12–15, 2004. |
References |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. Discussion5. Clinical implicationsAppendix A. Conference...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Gongora and T. M. Sundt III Myocardial Revascularization with Cardiopulmonary Bypass Card. Surg. Adult, January 1, 2008; 3(2008): 599 - 632. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
).