Abstract
OBJECTIVES
Using data from the CORONARY trial (n = 4752), we evaluated the incidence and prognostic significance of myocardial infarction (MI) applying different definitions based on peak postoperative creatine kinase-MB isoenzyme and cardiac troponin levels. We then aimed to identify the peak cardiac troponin during the first 3 postoperative days that was independently associated with a 2-fold increase in 30-day mortality.
METHODS
To combine different assays, we analysed cardiac troponins in multiples of their respective upper limit of normal (ULN). We identified the lowest threshold with a hazard ratio (HR) >2 for 30-day mortality independent of EuroSCORE and on- versus off-pump surgery.
RESULTS
Depending on the definition used based on creatine kinase-MB, the incidence of MI after coronary artery bypass grafting (CABG) ranged from 0.6% to 19% and the associated HRs for 30-day mortality ranged from 2.7 to 6.9. Using cardiac troponin (1528 patients), the incidence of MI ranged from 1.7% to 13% depending on the definition used with HRs for 30-day mortality ranging from 5.1 to 7.2. The first cardiac troponin threshold we evaluated, 180xULN, was associated with an adjusted HR for 30-day mortality of 7.6 [95% confidence interval (CI) 3.4–17.1] when compared to <130xULN. The next independent threshold was 130xULN with an adjusted HR for 30-day mortality of 7.8 (95% CI 2.3–26.1). The next cardiac troponin tested threshold (70xULN) did not meet criteria for significance.
CONCLUSIONS
Our results illustrate that the incidence and prognosis of a post-CABG MI varies based on the definition used. Validated post-CABG MI diagnostic criteria formulated from their independent association with important clinical outcomes are needed.
Keywords: Cardiac surgery, Creatine kinase-MB, Myocardial infarction, Troponin
INTRODUCTION
The universal definitions for myocardial infarction (MI) after coronary artery bypass grafting (CABG) surgery are based on a cardiac biomarker elevation >10 times the 99th percentile concentration (designated as the upper limit of normal; ULN) from a healthy population [1, 2]. These post-CABG MI diagnostic criteria are based on an arbitrary biomarker elevation threshold in association with signs of cardiac necrosis [1]. In 2012, the Task Force elected to increase the biomarker elevation threshold for the Third Universal definition because the initial >5xULN threshold was considered too sensitive [1, 3]. In 2013, the Society for Cardiovascular Angiography and Interventions suggested another definition using different biomarker elevation thresholds depending on whether new pathological Q waves or left bundle branch block (LBBB) was present [4]. These different MI definitions have never undergone rigorous validation in the context of CABG to assess their association with clinically important events [1, 2, 4]. They were also established when creatine kinase (CK)-MB was the biomarker of choice to diagnose post-CABG MI. In many centres, cardiac troponins have replaced CK-MB [5]. Given the ubiquitous release of cardiac troponin during CABG surgery, the prognostic value of using a 10-fold ULN threshold has been questioned by clinicians when diagnosing post-CABG MI [6]. Therefore, we aimed to identify a prognostically relevant cardiac troponin threshold post-CABG MI.
The CABG Off- or On-Pump Revascularization Study (CORONARY) is a large (n = 4752) randomized controlled trial that compares CABG with and without cardiopulmonary bypass [7–9]. Using data from that trial, we aimed to evaluate the incidence of MI and the prognostic implication of post-CABG MI as determined using different diagnostic criteria utilized by clinicians and in clinical studies of cardiac surgery.
METHODS
Details of the CORONARY Trial (NCT 00463294) methods have been previously published and are briefly described below [8].
Patients
Patients were recruited between 2006 and 2011 from 79 centres in 19 countries. Patients were eligible if they were scheduled to undergo CABG and had at least one of the following risk factors: age ≥70 years, peripheral artery disease, cerebrovascular disease, carotid stenosis ≥70% or renal insufficiency. Patients aged 60–69 years were enrolled if they had diabetes, required urgent revascularization, had a left ventricular ejection fraction ≤ 35% or a recent history of smoking. After the recruitment of 1700 patients, a protocol amendment allowed patients 55–59 years to be enrolled if they had one of the risk factors listed for patients aged 60–69 years. Patients were excluded if they required valve surgery, were not suitable for one of the 2 CABG techniques, had a life expectancy <2 years, required emergent revascularization or repeat CABG surgery or were previously enrolled in CORONARY. Written informed consent was obtained for each participant. Research ethics board approval was obtained at each participating centre.
Follow-up
In all patients, CK-MB measurements and electrocardiogram (ECG) readings were mandated at 24 and 48 h after surgery. Centres could report postoperative cardiac troponins at 24 and 48 h on the standardized case report forms, but troponin measurement was not mandated. When reporting biomarker values, centres also provided the 99th percentile of the ULN for the assay used.
Patients were seen in the clinic or at the hospital 30 days after their surgery. At 1 year, patients were assessed for death, non-fatal stroke, non-fatal MI or non-fatal renal failure requiring dialysis at 30 days after surgery. All deaths within 30 days of surgery were considered cardiovascular. Follow-up was complete for 4752 patients (100%) at 30 days and for 4690 patients (98.7%) at 1 year [7].
Statistical analyses
Creatine kinase-MB
Using data from participants in the CORONARY Trial as a cohort, we evaluated the incidence of post-CABG MI according to 5 different definitions: the Second Universal MI definition [3], the Third Universal MI definition [1], the definition proposed by Moussa et al. [4], the definition used in the CORONARY Trial [8] and the definition used in the Steroids In cardiac Surgery (SIRS) Trial [10]. The detailed definitions are presented in Table 1. To evaluate the clinical relevance of the definitions, we calculated the adjusted hazard ratio (aHR) for 30-day mortality adjusted for standard additive EuroSCORE associated with each of them [11]. To determine if there was a significant interaction (i.e. P < 0.05) between on-pump surgery and off-pump surgery with the MI definitions, we added an interaction term.
Table 1:
Criteria used in each definition of post-CABG myocardial infarction
| Definition | Rationale for biomarker threshold | Biomarker and ECG criteria |
|---|---|---|
| CK-MB | ||
| Universal definition 2007 | Arbitrary | CK-MB >5xULN with ECG abnormalities or angiographic evidence of new graft or native coronary artery occlusion or imaging evidence of new loss of viable myocardium |
| Universal definition 2012 | Arbitrary | CK-MB >10xULN with ECG abnormalities or angiographic evidence of new graft or native coronary artery occlusion or imaging evidence of new loss of viable myocardium |
| Moussa definition 2013 | Arbitrary | CK-MB ≥10xULN or ≥5xULN with ECG abnormalities |
| CORONARY definition 2012 | Arbitrary | CK-MB ≥5xULN or angiographic evidence of new graft or native coronary artery occlusion or imaging evidence of new loss of viable myocardium |
| SIRS study definition 2015 | Event-driven | CK-MB mass ≥6xULN or CK-MB activity ≥40 |
| Cardiac troponin | ||
| Universal definition 2012 | Arbitrary | Cardiac troponin >10xULN with ECG abnormalities or angiographic evidence of new graft or native coronary artery occlusion or imaging evidence of new loss of viable myocardium |
| Universal definition 2018 | Arbitrary | Cardiac troponin values >10xULN with pathological Q waves or angiographic evidence of new graft or native coronary artery occlusion or imaging evidence of new loss of viable myocardium |
| Moussa 2013 | Arbitrary | Cardiac troponin values ≥70xULN or ≥35xULN with ECG abnormalities |
ECG abnormalities: new pathological Q waves or new left bundle branch block.
CABG: coronary artery bypass grafting; CK-MB: creatine kinase-MB isoenzyme; ECG: electrocardiogram; SIRS: Steroids In cardiac Surgery; ULN: 99th percentile upper limit of normal.
In the SIRS trial, a 7507-patient trial evaluating prophylactic steroid versus placebo in on-pump cardiac surgery, we observed that CK-MB mass and activity assays performed differently based on thresholds set by their ULN [10]. For this reason, a separate threshold for each assay type was established. We aimed to confirm this finding using CORONARY data by evaluating the incidence and aHR for 30-day mortality associated with MI as defined in CORONARY separately for patients in whom CK-MB mass and CK-MB activity were measured. Furthermore, we determined the aHRs for 30-day mortality using the SIRS definitions for CK-MB mass and CK-MB activity using the CORONARY trial data.
Cardiac troponin
In patients for whom cardiac troponin levels were available, we evaluated the incidence of post-CABG MI according to 3 different definitions: the Third [1] and Fourth [2] Universal MI definitions, and the definition proposed by Moussa et al. [4] (Table 1). To combine different cardiac troponin assays, we analysed cardiac troponin in multiples of their respective ULN, as there is no standardization across assays [12]. For CORONARY trial participants with at least 1 cardiac troponin measurement in the first 48 h after surgery, we calculated the proportion that had a peak cardiac troponin concentration >10xULN (the currently suggested threshold), the associated aHR for 30-day and 1-year mortality adjusting for EuroSCORE.
To identify a prognostically relevant cardiac troponin concentration threshold after CABG surgery, we used a proportional hazards Cox model with 30-day mortality as the dependent variable. We identified the lowest cardiac troponin threshold that had a statistically significant independent aHR >2 by exploring peak cardiac troponin levels in multiples of the ULN, while adjusting for EuroSCORE and on-pump versus off-pump surgery. A priori, we specified that the first threshold we would test would be the peak postoperative troponin threshold close to the 95th percentile of the overall cohort. Additional cut-offs were chosen a posteriori based on the distribution of data. We conducted sensitivity analyses that repeated these thresholds identifying analyses but used 1-year mortality as the dependent variable.
RESULTS
In the CORONARY trial, 119 (2.5%) patients died during the first 30 days and 241 (5%) patients died within 1 year. The baseline characteristics of the CORONARY participants have been previously reported and are summarized in Table 2 [9].
Table 2:
Participant characteristics and additional outcomes for all participants in the CORONARY trial
| Total number of participants | 4752 |
|---|---|
| Age (years), mean (SD) | 68 (7) |
| Male gender, n (%) | 3843 (80.9) |
| Body mass index (kg/m2), mean (SD) | 26.7 (4.4) |
| Clinical history, n (%) | |
| Prior myocardial infarction | 1641 (34.5) |
| Prior percutaneous coronary intervention | 463 (9.7) |
| Cerebrovascular disease | 456 (9.8) |
| Peripheral arterial disease | 385 (8.1) |
| History of smoking | 2565 (54.0) |
| Any diabetes | 2228 (46.9) |
| Chronic kidney failure (dialysis) | 66 (1.4) |
| Congestive heart failure | 296 (6.2) |
| Hypertension | 3604 (75.8) |
| Chronic atrial fibrillation | 128 (2.7) |
| Left ventricular function, n (%) | |
| Grade 1 (≥50%) | 3294 (70.7) |
| Grade 2 (35–49%) | 1103 (23.7) |
| Grade 3 (20–34%) | 244 (5.2) |
| Grade 4 (<20%) | 11 (0.2) |
| EuroSCORE grade, n (%) | |
| 0–2 | 1339 (28.2) |
| 3–5 | 1932 (40.7) |
| >5 | 1412 (29.7) |
| EuroSCORE, mean (SD) | 3.8 (2.3) |
| Urgent surgery, n (%) | 1842 (38.8) |
| Any antiplatelet agent (preoperative), n (%) | 3620 (76.2) |
| Numbers of vessels diseased, n (%) | |
| Left main | 1001 (21.5) |
| Triple vessel disease | 2711 (58.2) |
| Double vessel disease | 817 (17.5) |
| Single vessel disease | 119 (2.6) |
| Cardioplegia technique, n (%) | 2377 (50.0) |
| Type | |
| Blood | 1987 (83.6) |
| Crystalloid | 175 (7.4) |
| Unknown | 215 (9.0) |
| Temperature | |
| Warm | 562 (23.6) |
| Cold | 1600 (67.3) |
| Unknown | 215 (9.0) |
| Delivery | |
| Antegrade | 1886 (79.3) |
| Retrograde | 17 (0.7) |
| Combined | 259 (10.9) |
| Unknown | 215 (9.0) |
| Additional outcomes, median (IQR) | |
| Intensive care unit length of stay | 2 (1–3) |
| Hospital length of stay | 5 (4–7) |
The reported EuroSCORE is a standard additive.
IQR: interquartile range; SD: standard deviation.
Three patients (0.06%) developed a new LBBB and 55 developed new Q waves (1.16%) after surgery.
Creatine kinase-MB
For each of the 5 definitions of interest for CK-MB, the incidence of MI and aHRs for 30-day and 1-year mortality are presented in Table 3. Depending on the diagnostic criteria used, the incidence of MI after CABG surgery ranged from 0.6% to 19% and the aHRs for 30-day mortality ranged from 2.7 to 6.9. The Third Universal definition was associated with the lowest post-CABG MI incidence and the SIRS trial definition with the highest. The SIRS trial definition resulted in the lowest aHR [2.7; 95% confidence interval (CI) 1.9–4.0] for 30-day mortality and the Moussa definition had the highest aHR (6.9; 95% CI 4.2–11.5). There was no significant interaction between on-pump surgery and off-pump surgery with the MI definitions (i.e. all interaction P-values were ≥0.2).
Table 3:
Myocardial infarction incidence and associated mortality according to diagnostic criteria used
| Definition | Incidence, N (%) | Unadjusted mortality 1 year, N (%) | 30-Day aHR for mortality | 1-Year aHR for mortality | |
|---|---|---|---|---|---|
| (95% CI)a | (95% CI)a | ||||
| CK-MB | |||||
| Universal definition 2007 | MI | 50 (1.1) | 8 (16.0) | 5.1 (2.2–11.4) | 2.8 (1.4–6.0) |
| No MI | 4702 (98.9) | 233 (5.0) | |||
| Universal definition 2012 | MI | 29 (0.6) | 5 (17.2) | 5.3 (2.0–14.2) | 2.5 (1.0–6.5) |
| No MI | 4723 (99.4) | 236 (5.0) | |||
| Moussa definition 2013 | MI | 127 (2.7) | 22 (17.3) | 6.9 (4.2–11.5) | 3.9 (2.5–6.0) |
| No MI | 4625 (97.3) | 219 (4.7) | |||
| CORONARY definition 2012 | MI | 328 (6.9) | 41 (12.5) | 4.0 (2.6–6.2) | 2.9 (2.1–4.1) |
| No MI | 4424 (93.1) | 200 (4.5) | |||
| SIRS study definition 2015 | MI | 902 (19.0) | 73 (8.1) | 2.7 (1.9–4.0) | 1.9 (1.4–2.5) |
| No MI | 3850 (81.0) | 168 (4.4) | |||
| Cardiac troponin | |||||
| Universal definition 2012 | MI | 27 (1.8) | 5 (18.5) | 7.2 (2.4–21.3) | 3.7 (1.5–9.3) |
| No MI | 1473 (98.2) | 79 (4.4) | |||
| Universal definition 2018 | MI | 26 (1.7) | 4 (15.4) | 5.1 (1.5–17.6) | 2.9 (1.1–8.1) |
| No MI | 1474 (98.3) | 80 (5.4) | |||
| Moussa 2013 | MI | 196 (13.1) | 23 (11.7) | 5.6 (2.8–11.0) | 3.0 (1.8–4.8) |
| No MI | 1304 (86.9) | 61 (4.7) | |||
Adjusted for EuroSCORE.
aHR: adjusted hazard ratio; CI: confidence interval; CK-MB: creatine kinase-MB isoenzyme; MI: myocardial infarction; N: number; SIRS: Steroids In cardiac Surgery.
The aHRs for the CK-MB mass and activity assays using the assay-specific SIRS definitions and the CORONARY definition are presented in Table 4. Using the CORONARY definition, the incidence of MI in patients for whom CK-MB mass was reported was 4%, whereas it was 14% when CK-MB activity was reported. The aHRs for 30-day mortality were 2.5 (95% CI 1.3–4.6) and 2.6 (95% CI 0.9–7.5), respectively.
Table 4:
Myocardial infarction associated mortality according to the SIRS and CORONARY definition for CK-MB mass and activity assays
| CK-MB assay | Incidence, N (%) | aHR for 30-day mortality (95% CI)a | aHR for 1-year mortality (95% CI)a |
|---|---|---|---|
| CORONARY definition | |||
| Mass | 103 (3.5) | 2.5 (1.3–4.6) | 2.4 (1.5–3.9) |
| Activity | 215 (14.2) | 2.6 (0.9–7.5) | 1.2 (0.6–2.4) |
| SIRS definition | |||
| Mass | 713 (24.5) | 3.8 (1.6–8.9) | 2.2 (1.1–4.2) |
| Activity | 175 (11.6) | 3.3 (2.1–5.2) | 2.0 (1.5–2.8) |
Adjusted for EuroSCORE.
aHR: adjusted hazard ratio; CI: confidence interval; CK-MB: creatine kinase-MB isoenzyme; SIRS: Steroids In cardiac Surgery.
Cardiac troponin
Peak cardiac troponin results were available for 1528 patients who underwent on-pump (n = 760) or off-pump CABG (n = 768) in the CORONARY trial. For 1085 of these patients (71%), cardiac troponin I was reported; for the other patients, cardiac troponin T was reported. The characteristics of patients with and without cardiac troponins reported are compared in Supplementary Material 1, Table S1. Patients who had cardiac troponin reported were systematically different from those who did not. Patients who had a troponin measurement were significantly older and more likely to have a history of hypertension, diabetes, percutaneous coronary intervention, renal failure requiring dialysis, peripheral and cerebrovascular disease. Patients for whom cardiac troponin was reported had a significantly higher body mass index and were significantly less likely to have had urgent surgery.
Peak cardiac troponins were >10xULN in 46% (705/1538) of patients. The positively skewed distribution of peak postoperative cardiac troponin concentrations is shown in Fig. 1. The mean for the peak cardiac troponin results was 53.0xULN (standard deviation 328.8). The median for peak cardiac troponin results was 8.7xULN (interquartile range 2.3–30.1). After adjustment for EuroSCORE and on- and off-pump surgery, the aHR for mortality associated with a peak cardiac troponin >10xULN was 4.0 (95% CI 0.8–19.3) at 30 days. There was no significant interaction between on-pump surgery and off-pump surgery with the MI definitions (i.e. the interaction P-value was 1).
Figure 1:
Distribution of cardiac troponins in multiples of the 99th percentile of the upper limit of normal. ULN: upper limit of normal.
For each of the 3 definitions of interest for cardiac troponins, the incidence of MI and aHR for 30-day and 1-year mortality are presented in Table 3. Depending on the diagnostic criteria used, the incidence of MI ranged from 1.7% to 13% and the aHRs for 30-day mortality ranged from 5.1 to 7.2. The Fourth Universal definition was associated with the lowest post-CABG MI incidence, whereas the Moussa definition resulted in the highest incidence. The Fourth Universal definition resulted in the lowest aHR for 30-day mortality 5.1 (95% CI 1.5–17.6), whereas the Third Universal definition resulted in the highest aHR, 7.2 (95% CI 2.4–21.3). For each of the definitions, we present the associated survival curve in Fig. 2 and receiver operating characteristic curve in the Supplementary Material 2.
Figure 2:
Adjusted survival probability at 30 days associated with each of the definitions. CK-MB: creatine kinase-MB isoenzyme; SIRS: Steroids In cardiac Surgery.
To identify a prognostically relevant threshold, the first threshold we evaluated was 180xULN because 177xULN corresponded to the 95th percentile. This threshold was associated with an aHR for 30-day mortality of 7.6 (95% CI 3.4–17.1) when compared to <130xULN. The next independent threshold that we evaluated was 130xULN to 180xULN with an aHR for 30-day mortality of 7.8 (95% CI 2.3–26.1) when compared to <130xULN. The next threshold (>70xULN) did not meet the criteria for significance with an aHR of 2.5 (95% CI 0.7–8.5). The HRs for 1-year mortality are presented in Table 5.
Table 5:
Association of cardiac troponins with mortality
| Covariate | aHR (95% CI)a | P-value | |
|---|---|---|---|
| 30-Day mortality | 130≤ cTn <180xULN | 7.8 (2.3–26.1) | 0.0009 |
| cTn ≥180xULN | 7.6 (3.4–17.1) | <0.0001 | |
| Off-pump | 1.3 (0.7–2.5) | 0.5 | |
| EuroSCORE | 1.1 (1.0–1.2) | 0.0002 | |
| 1-Year mortality | 130≤ cTn <180xULN | 3.7 (1.4–10.3) | 0.01 |
| cTn ≥180xULN | 4.2 (2.3–7.8) | <0.0001 | |
| Off-pump | 1.0 (0.7–1.6) | 1.0 | |
| EuroSCORE | 1.1 (1.1–1.1) | <0.0001 |
Reference is cardiac troponin <130xULN.
aHR: adjusted hazard ratio; CI: confidence interval; cTn: troponin; ULN: 99th percentile upper limit of normal.
DISCUSSION
Key results
We examined the incidence and associated mortality of different post-CABG MI definitions within the CORONARY trial data set. The incidence and associated mortality varied substantially based on which definition of MI after CABG surgery was applied. Based on the CORONARY trial data set, definitions based on CK-MB measures were associated with a 2.7- to 6.9-fold increase in 30-day mortality. However, some definitions were found to result in a high incidence of MI (SIRS definition, incidence of 19% which included CK-MB levels determined by enzyme activity and mass), whereas others were much less (Third Universal Definition, incidence of 0.6% which does not endorse CK-MB activity assays).
Cardiac troponin elevations >10xULN after CABG surgery occurred in 46% patients in the CORONARY trial and were not significantly associated with 30-day mortality when adjusting for EuroSCORE and on- versus off-pump CABG. When adding new Q waves or new LBBB, as per the Third and Fourth Universal definitions, the incidence of MI decreased to <2% and the HR for 30-day mortality was more than 5. These results suggest that a clinically relevant cardiac troponin threshold for post-CABG MI should be higher than >10xULN, the value advocated in the Universal Definition of MI. Including new Q waves or LBBB to post-CABG MI criteria leads to a higher HR for 30-day mortality; however, with a 5-fold increase in 30-day mortality, mandating these ECG changes in the criteria likely misses a stratum of the population who have suffered an MI and are at substantial risk for short-term death.
Existing studies
An association between cardiac biomarker elevations and mortality after CABG was demonstrated in a systematic review of 7 studies (18 908 patients) by Domanski et al. [13]. Their adjusted and unadjusted pooled results suggested that postoperative CK-MB and cardiac troponin increases were associated with increased short and long-term mortality. With increasing biomarker elevations, mortality increased: for CK-MB, a doubling in risk occurred with CK-MB ≥5xULN, whereas for cardiac troponin the risk doubled with elevations of 20–40xULN [13]. Another systematic review of 23 studies (29 483 patients) by Petaja et al. [14] also demonstrated worse outcomes with postoperative biomarker elevations.
Diagnosing an MI after a cardiac surgery is complicated for multiple reasons. MI after CABG surgery can be caused by early graft failure, distal embolization of plaque material and inadequate myocardial protection [15]. Based on different mechanisms, these events can have different clinical presentations. The surgical ‘trauma’ to the myocardium in itself causes some biomarker release. Cardiac and pericardial manipulation can result in ECG changes. The assessment for typical symptoms is confounded by normal postoperative pain, delirium and analgesia. Accordingly, biomarker elevations are important as they can be the main reliable manifestation of significant myocardial injury. In the non-cardiac perioperative setting, where assessment for symptoms is also unreliable, isolated cardiac biomarker elevations are associated with substantial mortality [16].
The SIRS definition is the only prognostic-based definition, but also the only definition that relies solely on biomarkers [10, 17]. The SIRS biomarker threshold was derived with the explicit goal of identifying patients at risk of a poor outcome. Using blinded data from the first 7000 participants, the SIRS investigators undertook a planned analysis and used a modified Mazumdar approach to identify the lowest peak postoperative CK-MB threshold independently associated with a HR >2 for 30-day mortality when adjusting for EuroSCORE [18]. Separate analyses of CK-MBs measured with the mass and the activity assays are another strength of the SIRS MI definition. Other MI definitions do not differentiate between these assays even though they produce significantly different results, and the CK-MB mass assay being recommended over activity assays [10, 19].
Our analysis using the CORONARY trial data validates the SIRS definition when using the CK-MB mass assay by demonstrating that the thresholds identified in SIRS were also associated with a more than 2-fold increase in 30-day mortality in the CORONARY cohort. In the absence of ECG, imaging or angiographic criteria, it could be argued that the SIRS definition identifies patients with myocardial injury rather than MI. However, any definition that requires new Q waves or a new LBBB is likely to demonstrate an MI incidence of <2%, no matter what is the required increase in the biomarker, and these data suggest such a definition is going to miss prognostically important events. As for the imaging and angiographic criteria, they cannot be assessed routinely in all patients and are therefore likely under detect MI.
Myocardial injury after cardiac surgery may actually be a better designation for the pathophysiological mechanisms that lead to biomarker release after cardiac surgery. A minority of cardiac surgery patients have classic MI with acute coronary or graft occlusion with resulting necrosis in a specific myocardial territory [20, 21]. Most cardiac injury after cardiac surgery is related to the extent of the procedure as demonstrated with higher biomarkers after valvular surgery or combined surgery compared to isolated CABG [22]. The quality of myocardial protection may also play a role; inadequate administration of cardioplegia may result in greater myocardial injury. Clinicians should recognize that troponin elevations above established thresholds after cardiac surgery provide important prognostic information. To what extent this risk is modifiable is still unknown and warrants further research. The ongoing VISION (Vascular events In Surgery patIents cOhort evaluatioN) Cardiac Surgery Study, a 15 000-patient cohort will allow derivation of a more precise estimate for the optimal event-driven MI definition after cardiac surgery using high-sensitivity cardiac troponins.
Strengths and limitations
Our work represents the first comparison of the incidence and prognostic value of different post-CABG MI definitions. Using patients from a large cardiac surgery randomized trial as a cohort provided us with a large sample size and high-quality data. The excellent follow-up (100% at 30 days and 98.7% at 1 year) allowed evaluation of the association of the MI definitions with important clinical outcomes.
Important limitations of this study include the absence of information on cardiac surgery procedures other than CABG. The subgroup of patients with cardiac troponin measurements was smaller and the overall CORONARY cohort, and this sample size limited our ability to explore potential thresholds. In addition, the CORONARY trial was conducted before the widespread use of high-sensitivity cardiac troponins and we cannot comment on reported differences in ULN between men and women and those older than 70 years than younger when using high-sensitivity cardiac troponin assays [23]. Though these data of non-high-sensitivity cardiac troponins provide insights, further research is needed using high-sensitivity cardiac troponins.
CONCLUSIONS
The various definitions for MI after CABG result in substantial variations in the incidence of MI and in the associated aHRs for mortality. These variations have important implications for clinical practice; depending on diagnostic criteria used, an absolute difference of 18% of patients will be labelled as having a perioperative MI. Our current study validates the prognostic impact of the SIRS diagnostic criteria, which are the only criteria established based on clinical data. Further work is needed to clarify the underlying pathophysiology of these events and help differentiate across the various pathophysiological mechanisms. Clinically relevant post-CABG MI diagnostic criteria should be independently associated with mortality. Our results illustrate the need for validated post-CABG MI diagnostic criteria formulated from their independent association with important clinical outcomes.
Funding
The CORONARY Trial was supported by the Canadian Institutes of Health Research. E.P.B.-C. received a Frederick Banting and Charles Best Graduate Scholarship—Doctoral Award from the Canadian Institutes of Health Research to support the conduct of this study. P.J.D. and D.J.C. are supported by Canada Research Chairs.
Conflict of interest: Peter Kavsak has received grants/reagents/consultant/advisor/honoria from Abbott Laboratories, Abbott Point of Care, Abbott Diagnostics Division Canada, Beckman Coulter, Ortho Clinical Diagnostics, Randox Laboratories, Roche Diagnostics and Siemens Healthcare Diagnostics. McMaster University has filed patents with Peter Kavsak listed as an inventor in the acute cardiovascular biomarker field. P.J. Devereaux is a member of a research group with a policy of not accepting honorariums or other payments from the industry for their own personal financial gain. They do accept honorariums/payments from the industry to support research endeavours and costs to participate in meetings. On the basis of study questions he has originated and grants he has written, he has received grants from Abbott Diagnostics, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers-Squibb, Coviden, Octapharma, Philips Healthcare, Roche Diagnostics and Stryker. He has also participated in an advisory board meeting for GlaxoSmithKline and an expert panel meeting with AstraZeneca and Boehringer Ingelheim. All other authors declared no conflict of interest.
Supplementary Material
REFERENCES
- 1. Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD. et al. Third universal definition of myocardial infarction. Eur Heart J 2012;33:2551–67. [DOI] [PubMed] [Google Scholar]
- 2. Thygesen K, Alpert JS, Jaffe AS, Chaitman BR, Bax JJ, Morrow DA. et al. Fourth universal definition of myocardial infarction. Eur Heart J 2019;40:237–69. [DOI] [PubMed] [Google Scholar]
- 3. Thygesen K, Alpert JS, White HD; Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction. Universal definition of myocardial infarction. Eur Heart J 2007;28:2525–38. [DOI] [PubMed] [Google Scholar]
- 4. Moussa ID, Klein LW, Shah B, Mehran R, Mack MJ, Brilakis ES. et al. Consideration of a new definition of clinically relevant myocardial infarction after coronary revascularization: an expert consensus document from the Society for Cardiovascular Angiography and Interventions (SCAI). J Am Coll Cardiol 2013;62:1563–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Prochaska MT, Hohmann SF, Modes M, Arora VM.. Trends in troponin-only testing for AMI in academic teaching hospitals and the impact of choosing wisely(R). J Hosp Med 2017;12:957–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. van Beek DE, van Zaane B, Buijsrogge MP, van Klei WA.. Implementation of the third universal definition of myocardial infarction after coronary artery bypass grafting: a survey study in Western Europe. J Am Heart Assoc 2015;4:e001401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Lamy A, Devereaux PJ, Prabhakaran D, Taggart DP, Hu S, Paolasso E. et al. Effects of off-pump and on-pump coronary-artery bypass grafting at 1 year. N Engl J Med 2013;368:1179–88. [DOI] [PubMed] [Google Scholar]
- 8. Lamy A, Devereaux PJ, Prabhakaran D, Hu S, Piegas LS, Straka Z. et al. Rationale and design of the coronary artery bypass grafting surgery off or on pump revascularization study: a large international randomized trial in cardiac surgery. Am Heart J 2012;163:1–6. [DOI] [PubMed] [Google Scholar]
- 9. Lamy A, Devereaux PJ, Prabhakaran D, Taggart DP, Hu S, Paolasso E. et al. Off-pump or on-pump coronary-artery bypass grafting at 30 days. N Engl J Med 2012;366:1489–97. [DOI] [PubMed] [Google Scholar]
- 10. Whitlock RP, Devereaux PJ, Teoh KH, Lamy A, Vincent J, Pogue J. et al. Methylprednisolone in patients undergoing cardiopulmonary bypass (SIRS): a randomised, double-blind, placebo-controlled trial. Lancet 2015;386:1243–53. [DOI] [PubMed] [Google Scholar]
- 11. Roques F, Nashef SA, Michel P, Gauducheau E, de VC, Baudet E. et al. Risk factors and outcome in European cardiac surgery: analysis of the EuroSCORE multinational database of 19030 patients. Eur J Cardiothorac Surg 1999;15:816–22; discussion 22–3. [DOI] [PubMed] [Google Scholar]
- 12. Estis J, Wu AHB, Todd J, Bishop J, Sandlund J, Kavsak PA.. Comprehensive age and sex 99th percentiles for a high-sensitivity cardiac troponin I assay. Clin Chem 2018;64:398–9. [DOI] [PubMed] [Google Scholar]
- 13. Domanski MJ, Mahaffey K, Hasselblad V, Brener SJ, Smith PK, Hillis G. et al. Association of myocardial enzyme elevation and survival following coronary artery bypass graft surgery. JAMA 2011;305:585–91. [DOI] [PubMed] [Google Scholar]
- 14. Petaja L, Salmenpera M, Pulkki K, Pettila V.. Biochemical injury markers and mortality after coronary artery bypass grafting: a systematic review. Ann Thorac Surg 2009;87:1981–92. [DOI] [PubMed] [Google Scholar]
- 15. Brener SJ, Lytle BW, Schneider JP, Ellis SG, Topol EJ.. Association between CK-MB elevation after percutaneous or surgical revascularization and three-year mortality. J Am Coll Cardiol 2002;40:1961–7. [DOI] [PubMed] [Google Scholar]
- 16. Botto F, Alonso-Coello P, Chan MT, Villar JC, Xavier D, Srinathan S. et al. Myocardial injury after noncardiac surgery: a large, international, prospective cohort study establishing diagnostic criteria, characteristics, predictors, and 30-day outcomes. Anesthesiology 2014;120:564–78. [DOI] [PubMed] [Google Scholar]
- 17.Writing Committee for the VSI, Devereaux PJ, Biccard BM, Sigamani A, Xavier D, Chan MTV. et al. Association of postoperative high-sensitivity troponin levels with myocardial injury and 30-day mortality among patients undergoing noncardiac surgery. JAMA 2017;317:1642–51. [DOI] [PubMed] [Google Scholar]
- 18. Whitlock R, Teoh K, Vincent J, Devereaux PJ, Lamy A, Paparella D. et al. Rationale and design of the steroids in cardiac surgery trial. Am Heart J 2014;167:660–5. [DOI] [PubMed] [Google Scholar]
- 19. Alpert JS, Thygesen K, Antman E, Bassand JP.. Myocardial infarction redefined—a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959–69. [DOI] [PubMed] [Google Scholar]
- 20. Rasmussen C, Thiis JJ, Clemmensen P, Efsen F, Arendrup HC, Saunamaki K. et al. Significance and management of early graft failure after coronary artery bypass grafting: feasibility and results of acute angiography and re-re-vascularization. Eur J Cardiothorac Surg 1997;12:847–52. [DOI] [PubMed] [Google Scholar]
- 21. Fabricius AM, Gerber W, Hanke M, Garbade J, Autschbach R, Mohr FW.. Early angiographic control of perioperative ischemia after coronary artery bypass grafting. Eur J Cardiothorac Surg 2001;19:853–8. [DOI] [PubMed] [Google Scholar]
- 22. Mastro F, Guida P, Scrascia G, Rotunno C, Amorese L, Carrozzo A. et al. Cardiac troponin I and creatine kinase-MB release after different cardiac surgeries. J Cardiovasc Med 2015;16:456–64. [DOI] [PubMed] [Google Scholar]
- 23. Welsh P, Preiss D, Shah ASV, McAllister D, Briggs A, Boachie C. et al. Comparison between high-sensitivity cardiac troponin T and cardiac troponin I in a large general population cohort. Clin Chem 2018;64:1607–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.