Abstract
Objectives
We aimed to determine whether high-sensitivity cardiac troponin I (hs-cTnI) thresholds associated with increased 30-day mortality isolated coronary artery bypass grafting (CABG) differed between those undergoing off-pump (OPCAB) and on-pump (ONCAB) CABG.
Methods
We conducted a subanalysis of patients who underwent isolated CABG in the Vascular Events in Surgery Patients Cohort Evaluation (VISION) Cardiac Surgery Study. Cox regression was used to determine the hazard ratios (HRs) for mortality based on postoperative day 1 log-transformed hs-cTnI adjusted by EuroSCORE II, with OPCAB versus ONCAB as an interaction term. HRs were modelled as a function of hs-cTnI, and the lowest troponin associated with HR ≥ 1.00 was identified for each group.
Results
Of the original VISION cohort, 6505 patients underwent isolated CABG (OPCAB = 1141, ONCAB = 5364). Median hs-cTnI after CABG was 2446 ng/L (interquartile range [IQR] 1164-5654), and lower after OPCAB (640 ng/L [264-1689]) than ONCAB (2972 ng/L [1536-6448], P < .001). There were no differences in 30-day mortality between OPCAB and ONCAB (1.7% vs 1.4%, P = .5). Increased log-peak hs-cTnI was associated with greater mortality after CABG (adjusted HR = 1.7 [95% CI, 1.4-2.1]). The hs-cTnI threshold associated with HR ≥ 1.00 for isolated CABG was 6549 ng/L (95% CI, 3609-8381). OPCAB versus ONCAB had a significant interaction effect on the association between hs-cTnI and mortality (interaction P = .002). The hs-cTnI threshold associated with mortality after OPCAB was ≥4708 ng/L (95% CI, 581-7177), compared to ≥6806 ng/L (95% CI, 4001-13 993) after ONCAB.
Conclusions
The clinically significant hs-cTnI threshold after CABG associated with an increased risk of 30-day mortality above the baseline is substantially higher than defined by current guidelines, but lower in patients undergoing OPCAB compared to ONCAB.
Keywords: troponin, CABG, on-pump CABG, off-pump CABG, myocardial infarction
Coronary artery bypass grafting (CABG) remains the gold standard for treating severe multivessel coronary artery disease.
Graphical abstract
INTRODUCTION
Coronary artery bypass grafting (CABG) remains the gold standard for treating severe multivessel coronary artery disease.1 However, it is an invasive procedure that often involves manipulating the heart and disrupting physiological coronary blood flow. One concerning complication of CABG is perioperative myocardial infarction (MI). A transient rise in cardiac enzymes, such as creatine-kinase myocardial band, troponin T or I, is observed after surgery and typically poses no major threat on long-term outcomes.2,3 However, high levels of cardiac biomarkers are concerning for postoperative MI due to graft failure, poor myocardial protection, incomplete revascularization, or surgical trauma.4
High-sensitivity cardiac troponin I (hs-cTnI) has become the gold standard biomarker used to diagnose myocardial injury.5 The Vascular Events in Surgery Patients Cohort Evaluation (VISION) Cardiac Surgery Study (2022) examined routine hs-cTnI levels measured on postoperative day (POD) 1 and POD 2-3 associated with clinically important myocardial injury after cardiac surgery.6 The authors found that the hs-cTnI threshold associated with an increase in adverse outcomes was meaningfully higher than defined by existing guidelines and varied according to the procedure performed.
The original VISION Cardiac Surgery study combined CABG and aortic valve replacement but did not explore whether the hs-cTnI threshold for prognostically relevant myocardial injury after CABG differed for on-pump (ONCAB) and off-pump (OPCAB) CABG.6 OPCAB is thought to be less invasive due to avoidance of cardiopulmonary bypass and aortic manipulation and is associated with less blood transfusion and, in some studies, lower risk of early stroke.7,8 However, numerous studies have found either comparable or slightly inferior long-term survival and graft patency in OPCAB compared to ONCAB.8,9 There is currently no clear consensus on the clinically significant levels of hs-cTnI after CABG, and specific to ONCAB and OPCAB, that should represent prognostically important myocardial injury.
We aimed to determine if the hs-cTnI thresholds associated with increased 30-day mortality and major vascular complications differed for patients undergoing OPCAB and ONCAB.
METHODS
Ethical statement
This study was a subanalysis of the VISION Cardiac Surgery Study, for which ethical approval was obtained by the Population Health Research Institute.6 Further ethical approval was waived as de-identified data was used for this analysis.
Study population and data acquisition
The VISION Cardiac Surgery study was a multicentre prospective cohort study conducted across 12 countries from 2013 to 2019. Its design and protocol have been previously described.6 To summarize, patients ≥18 undergoing any cardiac surgical procedure, except for isolated pericardial window, pericardiectomy, pacemaker or defibrillator implantation, cardiac transplant, isolated aortic surgery, or mechanical assist device insertion, were recruited. Baseline variables were collected and consisted of their demographics, cardiovascular risk factors, variables used to calculate EuroSCORE II, and medications. Blood samples 3-12 hours after surgery or on postoperative day (POD) 1 were measured for hs-cTnI via a standardized assay (ARCHITECT STAT, Abbott Laboratories) which had an upper reference limit (URL) of 26 ng/L. The highest hs-cTnI value measured until the end of POD1 was denoted as “peak POD1 hs-cTnI.”6
We conducted a secondary analysis of patients who underwent isolated CABG in the VISION Cardiac Surgery cohort. In addition to the original study criteria, we excluded patients who: underwent any other concomitant cardiac surgery; had preoperative MI <24 hours; had hs-cTnI ≥300 ng/L < 12 hours preoperatively; lacked a hs-cTnI measurement POD1-3; underwent “salvage” cardiac surgery (cardiopulmonary resuscitation en route to the operating room or before induction); or had no follow-up mortality data. The statistical plan for this subanalysis was developed in partnership with the VISION Cardiac Surgery study leadership.
Outcomes
The primary outcome of interest was 30-day all-cause mortality. The secondary outcome of interest was 30-day major vascular complications, a composite of vascular mortality, postoperative MI between days 4-30 after surgery, and insertion of a mechanical assist device up to the end of the patient’s index stay in the intensive care unit after cardiac surgery. Postoperative MI between days 4-30 after surgery was defined based on the fourth Universal Definition of Myocardial Infarction (UDMI). It required both (1) increased troponin >99xURL and (2) clinical symptoms, ECG changes, wall motion abnormalities, imaging evidence of new loss of viable myocardium, or identification of a coronary thrombus by either angiography or autopsy.2,6,10 Outcomes were determined for all patients undergoing isolated CABG, then compared between patients undergoing ONCAB and OPCAB.
Statistical analysis
All statistical analyses were conducted in R 4.1.0 (R Foundation for Statistical Computing, Vienna), with OPCAB as the exposure and ONCAB as the control. Baseline characteristics and 30-day outcomes were compared using Student’s t-test or Kruskal-Wallis test for continuous outcomes, and Chi-square and Fisher’s exact test for categorical outcomes. Patients without recorded POD1 hs-cTnI or hs-cTnI <100 ng/L or >51 000 ng/L were excluded from Cox regression analyses to improve model fit.6
Multivariable Cox regression was used to determine hazard ratios (HRs) for 30-day all-cause mortality and major vascular complications as a function of log-transformed POD1 peak hs-cTnI. Models were adjusted for EuroSCORE II to account for baseline mortality risk while avoiding overfitting, with OPCAB versus ONCAB as an interaction term. We then derived fitted models that predicted HRs as a function of log-transformed hs-cTnI via natural cubic spline regression with 3 degrees of freedom at equal tertiles. From these fitted models, the lowest hs-cTnI threshold with HR ≥ 1.00 was identified for All CABG and OPCAB versus ONCAB, with 95% confidence intervals (95% CI) determined via bootstrapping.
RESULTS
Baseline characteristics
The VISION Cardiac Surgery Study included 13 862 patients, of which 6505 (OPCAB = 1141, ONCAB = 5364) underwent isolated CABG (Figure 1). Median age was 65.5 (58.6-71.8) years, and 80.7% (5247/6505) were men. OPCAB patients were less likely to have previous MI (OPCAB = 40.3% vs ONCAB = 51.3%, P < .001) and MI <90 days (OPCAB = 18.1% vs ONCAB = 29.1%, P < .001). OPCAB patients had lower median EuroSCORE II (1.1% [interquartile range (IQR) 0.7-1.8] vs 1.2% [0.8-1.9]) (P = .002), New York Heart Association (NYHA) status (P < .001) and Canadian Cardiovascular Society (CCS) class (P < .001) (Table 1). OPCAB patients presented with less medication use (Table S1) and had lower rates of complete revascularization (26.3% vs 11.2%, P < .001).
Figure 1.
CONSORT Flow Diagram of Included Patients. Abbreviations: CABG, coronary artery bypass grafting; hs-cTnI, high-sensitivity cardiac troponin I; ONCAB, on-pump CABG; OPCAB, off-pump CABG; POD1, postoperative day 1.
Table 1.
Baseline Characteristics
| Overall (n = 6505) | OPCAB (n = 1141) | ONCAB (n = 5364) | P-value | |
|---|---|---|---|---|
| Age (median [IQR]) | 65.5 [58.6, 71.8] | 65.8 [58.2, 72.0] | 65.5 [58.7, 71.8] | .7 |
| Female sex (%) | 1258 (19.3) | 197 (17.3) | 1061 (19.8) | .06 |
| Baseline BMI (median [IQR]) | 28.0 [25.2, 31.2] | 27.3 [24.8, 30.7] | 28.1 [25.2, 31.4] | <.001 |
| Hypertension (%) | 5090 (78.2) | 856 (75.0) | 4234 (78.9) | .004 |
| Diabetes (%) | 2633 (40.5) | 413 (36.2) | 2220 (41.4) | .001 |
| COPD (%) | 513 (7.9) | 87 (7.6) | 426 (7.9) | .8 |
| Tobacco use (%) | 3870 (59.5) | 646 (56.6) | 3224 (60.1) | .03 |
| Previous MI (%) | 3211 (49.4) | 460 (40.3) | 2751 (51.3) | <.001 |
| MI within 90 days (%) | 1769 (27.2) | 207 (18.1) | 1562 (29.1) | <.001 |
| Heart failure (%) | 794 (12.2) | 121 (10.6) | 673 (12.5) | .08 |
| Previous stroke (%) | 339 (5.2) | 50 (4.4) | 289 (5.4) | .2 |
| Atrial fibrillation (%) | 438 (6.7) | 62 (5.4) | 376 (7.0) | .07 |
| Peripheral arterial disease (%) | 600 (9.2) | 123 (10.8) | 477 (8.9) | .05 |
| CCS class (%) | <.001 | |||
| 0 | 789 (12.1) | 169 (14.8) | 620 (11.6) | |
| 1 | 890 (13.7) | 235 (20.6) | 655 (12.2) | |
| 2 | 2018 (31.0) | 356 (31.2) | 1662 (31.0) | |
| 3 | 1969 (30.3) | 267 (23.4) | 1702 (31.7) | |
| 4 | 839 (12.9) | 114 (10.0) | 725 (13.5) | |
| NYHA class (%) | <.001 | |||
| Not applicable | 1108 (17.0) | 289 (25.3) | 819 (15.3) | |
| 1 | 1210 (18.6) | 208 (18.2) | 1002 (18.7) | |
| 2 | 2246 (34.5) | 405 (35.5) | 1841 (34.3) | |
| 3 | 1497 (23.0) | 205 (18.0) | 1292 (24.1) | |
| 4 | 444 (6.8) | 34 (3.0) | 410 (7.6) | |
| Urgency (%) | ns | |||
| Urgent | 1881 (28.9) | 252 (22.1) | 1629 (30.4) | |
| Emergent | 97 (1.5) | 16 (1.4) | 81 (1.5) | |
| Elective | 4527 (69.6) | 873 (76.5) | 3654 (68.1) | |
| Poor mobility (%) | 248 (3.8) | 48 (4.2) | 200 (3.7) | .5 |
| Baseline creatinine (µmol/L) (mean [SD]) | 84.2 (78.8) | 76.6 (49.6) | 85.9 (83.7) | <.001 |
| EuroSCORE II (median [IQR]) | 1.2 [0.8, 1.9] | 1.1 [0.7, 1.8] | 1.2 [0.8, 1.9] | .002 |
Bold values indicates significant p-values. Abbreviations: BMI, body mass index; CCS, Canadian Cardiovascular Society; COPD, chronic obstructive pulmonary disease; IQR, interquartile range; MI, myocardial infarction; ns, non-significant; NYHA, New York Heart Association; ONCAB, on-pump CABG; OPCAB, off-pump CABG.
Postoperative hs-cTnI and 30-day outcomes
Of the 6505 isolated CABG patients included in this subanalysis, 6382 had POD1 hs-cTnI (median = 2446 ng/L [1164-5654], 94xURL). Three patients died before POD1 hs-cTnI could be measured. At 30 days, 1.4% (92/6505) patients died and 2.3% (150/6505) suffered a major vascular complication (Table 1).
Median peak hs-cTnI was lower at 640 ng/L ([264-1689], 25xURL) after OPCAB than 2972 ng/L ([1536-6448], 114xURL) after ONCAB (P < .001) (Figure 2). There was no difference between OPCAB and ONCAB in unadjusted 30-day mortality (1.7% [19/1141] vs 1.4% [73/5364], P = .5), major vascular complications (2.1% [24/1141] vs 2.3% [125/5364], P = .7), or its components (Table 2).
Figure 2.
hs-cTnI After All Isolated CABG, OPCAB, and ONCAB by Mortality Status. Abbreviations: CABG, coronary artery bypass grafting; hs-cTnI, high-sensitivity cardiac troponin I; ONCAB, on-pump CABG; OPCAB, off-pump CABG; POD1, postoperative day 1.
Table 2.
30-Day Outcomes
| 30-Day outcome | Overall (n = 6505) | OPCAB (n = 1141) | ONCAB (n = 5364) | P-value |
|---|---|---|---|---|
| Peak hs-cTnI day 1 (median [IQR]) | 2445.8 [1164.2-5653.5] | 640 [264-1688.5] | 2972 [1536.5-6448.8] | <.001 |
| Complete revascularization (%) | 5594 (86.1) | 839 (73.7) | 4755 (88.8) | <.001 |
| All-cause mortality (%) | 92 (1.4) | 19 (1.7) | 73 (1.4) | .5 |
| Vascular death (%) | 72 (1.1) | 15 (1.3) | 57 (1.1) | .6 |
| Mechanical assist device insertion (%) | 75 (1.2) | 8 (0.7) | 67 (1.2) | .2 |
| Major vascular complication (%) | 150 (2.3) | 24 (2.1) | 126 (2.3) | .7 |
| MI (%) | 23 (0.4) | 6 (0.5) | 17 (0.3) | .4 |
Bold values indicates significant p-values. Abbreviations: IQR, interquartile range; MI, myocardial infarction; ONCAB, on-pump CABG; OPCAB, off-pump CABG.
hs-cTnI thresholds and 30-day mortality
After excluding extreme hs-cTnI values, 6213 patients were eligible for Cox regression (OPCAB = 1017, ONCAB = 5196). Figure 3 shows the relationship between each unit increase in log-transformed peak hs-cTnI and 30-day mortality (HR = 1.7 [95% CI, 1.4-2.1], P < .001) after adjusting for EuroSCORE II. There was a significant interaction for ONCAB versus OPCAB regarding the association between hs-cTnI and 30-day mortality (P-interaction = .002).
Figure 3.
Predicted 30-Day Mortality as a Function of POD1 hs-cTnI for All Isolated CABG. (A) Predicted 30-day mortality and POD1 hs-cTnI for all CABG. (B) Predicted 30-day mortality and POD1 hs-cTnI for OPCAB. (C) Predicted 30-day mortality and POD1 hs-cTnI for ONCAB. Abbreviations: CABG, coronary artery bypass grafting; hs-cTnI, high-sensitivity cardiac troponin I; ONCAB, on-pump CABG; OPCAB, off-pump CABG; POD1, postoperative day 1.
For isolated CABG, the lowest hs-cTnI threshold at which 30-day mortality risk increases above the baseline was 6549 ng/L (95% CI, 3609-8381, 252xURL). The hs-cTnI threshold associated with increased risk of 30-day mortality after OPCAB was ≥4708 ng/L (95% CI, 581-7177, 181xURL), versus ≥6806 ng/L (95% CI, 4001-13 993, 262xURL) after ONCAB (Figure 4). Discrimination testing via a receiver operating characteristic (ROC) curve showed that hs-cTnI had an area under the curve of 0.73 (Figure S1).
Figure 4.
Predicted Hazard Ratio as a Function of POD1 hs-cTnI for All Isolated CABG. (A) Hazard ratio and POD1 hs-cTnI for all isolated CABG. (B) Predicted 30-day mortality and POD1 hs-cTnI for OPCAB. (C) Predicted 30-day mortality and POD1 hs-cTnI for ONCAB. Abbreviations: CABG, coronary artery bypass grafting; hs-cTnI, high-sensitivity cardiac troponin I; ONCAB, on-pump CABG; OPCAB, off-pump CABG; POD1, postoperative day 1 .
hs-cTnI thresholds and 30-day major vascular complications
Increased hs-cTnI was associated with higher likelihood of major vascular complications, adjusting for EuroSCORE II (HR = 1.7 [95% CI, 1.5-1.9], P < .001). There was a significant interaction between ONCAB versus OPCAB regarding the association between hs-cTnI and major vascular complications (P-interaction = .01) (Figure S2).
The hs-cTnI value at which major vascular complications were significantly more likely to occur (HR > 1.00) in isolated CABG was 4781 ng/L (95% CI, 1470-7557, 184xURL). The threshold was lower for patients who underwent OPCAB (4062 ng/L, 156xURL) compared to ONCAB (6912 ng/L, 266xURL) (Figure S3). The distribution of hs-cTnI values was too broad to perform bootstrapping for the isolated OPCAB and ONCAB cohorts.
HRs for mortality and major vascular complications corresponding to incremental hs-cTnI thresholds are shown in Table S2.
DISCUSSION
We conducted a secondary analysis of the VISION Cardiac Surgery study examining clinically-relevant hs-cTnI levels in isolated CABG and demonstrated a significant interaction between OPCAB and ONCAB.
The hs-cTnI predictive of increased 30-day mortality and major vascular complications is higher than defined by existing guidelines
Isolated biomarker rise after CABG is a strong prognosticator of adverse outcomes and could indicate procedure-related myocardial injury or treatment failure in the form of graft thrombosis, technical errors, or distal embolization.3,11 Troponin levels have become more sensitive with new biomarker assays. Numerous attempts to define postoperative MI after CABG have been made, between >10xURL (with ancillary findings, fourth UDMI) to >70xURL (without ancillary findings, Society for Cardiovascular Angiography and Interventions [SCAI] 2).10,12–14 These definitions have not been adapted to high-sensitivity troponin assays, nor rooted in clear evidence.11,15 Routine hs-cTnI collection after CABG varies among hospitals as the threshold for periprocedural MI remains contested. Although ancillary criteria can improve diagnostic accuracy, standard echocardiograms, MRI, and CT are not usually performed until several days post-surgery, and patients may have angiographically significant MI without structural or functional abnormalities.
Our subanalysis provides a definition of clinically-significant myocardial injury based on isolated biomarker levels post-CABG that (i) exceeds current guideline thresholds for perioperative MI and (ii) is a prognosticator to prompt definitive testing in patients at risk of missed MI.
Hs-cTnI has a continuous risk relationship to both 30-day mortality and major vascular complications. However, biomarker thresholds can be useful in establishing a framework at which the risk of adverse events becomes higher than established baseline. We found the POD1 hs-cTnI threshold associated with increased mortality after isolated CABG was 6549 ng/L (252xURL), which aligns with previous results for patients who underwent isolated CABG or AVR. Omran et al16 defined >13 000 ng/L (500xURL) as predictive for 48-hour repeat revascularization (n = 4684), suggesting a higher and non-equivocal hs-cTnI for early graft failure resulting in urgent reintervention. Wang et al17 examined 48-hour hs-cTnI and long-term mortality (n = 7813), which may account for their lower threshold of 903 ng/L (35xURL).17 Pölzl et al18 (n = 8292) reported a hs-TnT cutoff defining mortality-associated MI of 2385 ng/L (170xURL). In 2024, the European Association of Cardio-Thoracic Surgery (EACTS) proposed a higher hs-cTnI-only threshold for post-CABG MI of 500xURL based on the VISION findings.2 Hs-cTnI elevation from 35-500xURL was reclassified as “perioperative myocardial injury” to indicate a zone of ambiguity.2 In these scenarios, elevated hs-cTnI level may serve as a warning sign that prompts closer clinical monitoring or additional workup in the form of echo imaging or serial ECGs. However, if clinical signs or positive test findings are present, more invasive testing such as coronary angiography may be required. Overall, these variable thresholds across studies emphasize the importance of establishing nuanced criteria for clinically-significant myocardial injury. However, the consensus remains that study-validated thresholds exceed levels defined by current guidelines.19
Having a data-driven hs-cTnI threshold also allows for fairer outcome assessments in trials comparing PCI to CABG. SCAI-2 diagnoses perioperative myocardial injury in the absence of supportive ancillary findings for hs-cTnI levels >70xURL and does not differentiate between CABG and PCI.12 Our median POD1 hs-cTnI level was 94xURL with 61% of patients exceeding the SCAI-2 definition of >70xURL on POD1, implying this would significantly over diagnose postprocedural MI in patients who received CABG.11 However, SCAI-2 has been used in guideline-influencing trials such EXCEL,20 thus generating controversy regarding potentially biased outcomes favouring PCI for left main disease.21,22 Similarly, a secondary analysis of the ISCHEMIA trial found a significantly higher number of Type 5 MIs when periprocedural MI was defined using identical thresholds for CABG and PCI. Unlike Type 1 MI, which was lower in the CABG group and associated with mortality, Type 5 MIs thus defined had no reliable impact on mortality.19 Similarly, applying the SCAI-2 and EXCEL definitions of postprocedural MI to the SYNTAX trial resulted in a higher incidence of MIs in the CABG cohort compared to analyses performed using the UDMI or original trial definitions.23 These scenarios illustrate controversies regarding the overdiagnosis of adverse events in the CABG cohorts resulting in bias that favoured PCI in major clinical trials where non-evidence based outcome definitions of postoperative MI were used.
Average hs-cTnI distributions and levels predictive of adverse outcomes after OPCAB are lower than that for ONCAB
Our procedure-specific diagnostic criteria for prognostically important postoperative troponin elevation between ONCAB and OPCAB supports using different thresholds for these 2 procedures. We found that the hs-cTnI threshold associated with increased mortality was higher in ONCAB (6806 ng/L, 261xURL) than OPCAB (4708 ng/L, 181xURL), with similar findings for major vascular complications. This trend aligns with results reported by Wang et al17 where hs-cTnI thresholds indicative of long-term mortality were 1446 ng/L after ONCAB and 564 ng/L after OPCAB. The median POD1 hs-cTnI in ONCAB was also higher than the OPCAB cohort.
Proponents of OPCAB appreciate the avoidance of cardiopulmonary bypass and its associated cannulation incisions, cellular oedema, reperfusion injury, and aortic cross clamping.24,25 However, OPCAB is liable to unique risk factors for myocardial ischaemia including brief periods of coronary hypoperfusion without myocardial protection and a higher likelihood of incomplete revascularization, which has been demonstrated in previous studies and our outcomes (Table 2). Historical studies before high-sensitivity assays have reported similar phenomena. Khan et al found that while average cTnI was lower after OPCAB, this had no impact on long-term mortality and 3-month graft patency was better after ONCAB.26,27 Our study likewise supports the notion that hs-cTnI elevations relating to cardiopulmonary bypass may incur minimal long-term consequences. Similarly, Wang et al17 found a lower hs-cTnI threshold to be associated with risk of long-term mortality in OPCAB compared to ONCAB. Their lower thresholds overall could be attributed to their outcome measure of long-term (2-year) mortality; we hypothesize that mortality <30 days requires greater myocardial damage as indicated by higher immediate hs-cTnI levels. Nevertheless, these findings may signal to clinicians that hs-cTnI thresholds used for prognostication should be procedure-specific to either ONCAB or OPCAB.
Limitations
Our study results should be interpreted in the context of several limitations. As per VISION Cardiac Surgery study, we used a short-term outcome of 30-day all-cause mortality and major vascular complications as a marker of clinically significant myocardial injury but did not examine long-term outcomes. Although in the VISION Cardiac Surgery study hs-cTnI was measured on POD 2-3, there were insufficient events in the CABG cohort to trend troponin levels in this subanalysis.6 Furthermore, we were limited by low event rates in the OPCAB group (ie, 15 deaths and 24 major vascular complications), which led to wider confidence intervals. The general distributions of hs-cTnI in the isolated CABG, OPCAB, and ONCAB cohorts were also wide, despite median hs-cTnI values being significantly different between OPCAB and ONCAB. We aimed to control for this by removing extreme hs-cTnI values before Cox regression and spline analysis. The threshold of removal for very high hs-cTnI values was based on the upper dilution protocol of the ARCHITECT STAT assay. The lower limit of exclusion for hs-cTnI values was validated by sensitivity analyses (hs-cTnI <100 ng/L vs <300 ng/L), with no difference in study results.
Importantly, this was an observational post-hoc study that did not employ randomization of patients to OPCAB versus ONCAB. Although model results were adjusted by EuroSCORE II, this does not account for surgical risk factors that may allocate patients towards on-pump or off-pump procedures such as porcelain aorta or frailty. As such, there may be unexplored contributors to the lower hs-cTnI threshold seen in OPCAB and ONCAB due to patient selection. Finally, there is evidence suggesting differences in hs-cTnI between men and women.28 To avoid model overfitting, our analysis did not stratify for this but rather controlled for EuroSCORE II which contains sex as a predictor.
CONCLUSION
We conducted a subanalysis of isolated CABG procedures in the VISION Cardiac Surgery Study. We found that the POD1 hs-cTnI thresholds after CABG associated with 30-day mortality and major vascular complications are substantially lower than defined by current guidelines. Furthermore, the hs-cTnI value was lower in patients undergoing OPCAB compared to ONCAB. Our findings represent a data-driven and procedure-specific diagnostic criteria of clinically important hs-cTnI thresholds after CABG.
Supplementary Material
Contributor Information
Grace S Lee, Division of Cardiac Surgery, University of Toronto, Toronto, ON M5G 2C4, Canada.
Derrick Y Tam, Division of Cardiac Surgery, University of Toronto, Toronto, ON M5G 2C4, Canada; Schulich Heart Program, Sunnybrook Health Sciences Centre, Toronto, ON M4N 3M5, Canada; Institute of Health Policy, Management and Evaluation, Toronto, ON M5T 3M6, Canada.
Dominique Vervoort, Institute of Health Policy, Management and Evaluation, Toronto, ON M5T 3M6, Canada.
Shun Fu Lee, Population Health Research Institute, Hamilton, ON L8L 2X2, Canada; Health Research Methods, Evidence, and Impact, Hamilton, McMaster University, ON L8S 4L8, Canada.
Katheryn Brady, Population Health Research Institute, Hamilton, ON L8L 2X2, Canada.
Emilie Belley-Cote, Population Health Research Institute, Hamilton, ON L8L 2X2, Canada; Division of Cardiology, McMaster University, Hamilton, ON L8S 4L8, Canada; Division of Critical Care, McMaster University, Hamilton, ON L8S 4L8, Canada.
P J Devereaux, Health Research Methods, Evidence, and Impact, Hamilton, McMaster University, ON L8S 4L8, Canada; Division of Cardiology, McMaster University, Hamilton, ON L8S 4L8, Canada.
Andre Lamy, Population Health Research Institute, Hamilton, ON L8L 2X2, Canada; Department of Surgery, McMaster University, Hamilton, ON L8S 4L8, Canada.
Richard Whitlock, Population Health Research Institute, Hamilton, ON L8L 2X2, Canada; Department of Surgery, McMaster University, Hamilton, ON L8S 4L8, Canada.
Ryan Louie, Population Health Research Institute, Hamilton, ON L8L 2X2, Canada.
Stephen E Fremes, Division of Cardiac Surgery, University of Toronto, Toronto, ON M5G 2C4, Canada; Schulich Heart Program, Sunnybrook Health Sciences Centre, Toronto, ON M4N 3M5, Canada; Institute of Health Policy, Management and Evaluation, Toronto, ON M5T 3M6, Canada.
AUTHOR CONTRIBUTIONS
All authors contributed meaningfully to the manuscript.
SUPPLEMENTARY MATERIAL
Supplementary material is available at ICVTS online.
FUNDING
The VISION Cardiac Surgery Study (upon which the current study is based) was supported by: a Canadian Institute of Health Research (CIHR) operating grant, the CIHR Foundation, the CIHR Strategy for Patient Oriented Research (Ontario SPOR Support Unit and Ministry of Health and Long-Term Care [MOHLTC]), Abbott Laboratories; in Canada, grants from the Hamilton Health Sciences New Investigator Fund and Research Strategic Initiative, the Canadian Network and Centre for Trials Internationally, McMaster University Surgical Associates, the Academic Health Science Centres Alternative Funding Plan Innovation Fund Ontario, and the Population Health Research Institute (PHRI) Internal Funding Program; in China, General Research Fund grant 14101414, the Research Grant Council, Hong Kong Special Administrative Region; in Malaysia, University of Malaya research grant (RG302-14AFR); in Australia, the National Heart Foundation of Australia Vanguard Grant (101034) and the National Health and Medical Research Program; in the UK, the British Heart Foundation Personal Chair grant (CH/F/21/90010) and Research Excellence grant (RE/18/5/34216); and in Spain, an Instituto de Salud Carlos III grant (PI13/00502).
CONFLICTS OF INTEREST
None declared.
DATA AVAILABILITY
The data underlying this article are available in the article and in its online supplementary material.
REFERENCES
- 1. Lawton JS, Tamis-Holland JE, Bangalore S, et al. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization. Circulation. 2022;145:e18-e114. 10.1161/CIR.0000000000001038 [DOI] [PubMed] [Google Scholar]
- 2. Gaudino M, Flather M, Capodanno D, et al. European Association of Cardio-Thoracic Surgery (EACTS) expert consensus statement on perioperative myocardial infarction after cardiac surgery. Eur J Cardio-Thorac Surg. 2024;65:1–6. 10.1093/ejcts/ezad415 [DOI] [PubMed] [Google Scholar]
- 3. Heuts S, Gollmann-Tepeköylü C, Denessen EJS, et al. Cardiac troponin release following coronary artery bypass grafting: mechanisms and clinical implications. Eur Heart J. 2023;44:100-112. 10.1093/eurheartj/ehac604 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Gaudino M, Dangas GD, Angiolillo DJ, et al. ; American Heart Association Council on Cardiovascular Surgery and Anesthesia; Council on Clinical Cardiology; Council on Cardiovascular and Stroke Nursing; and Stroke Council. Considerations on the management of acute postoperative ischemia after cardiac surgery: a scientific statement from the American Heart Association. Circulation. 2023;148:442-454. 10.1161/CIR.0000000000001154 [DOI] [PubMed] [Google Scholar]
- 5. Januzzi JL, Mahler SA, Christenson RH, et al. Recommendations for institutions transitioning to high-sensitivity troponin testing. J Am Coll Cardiol. 2019;73:1059-1077. 10.1016/j.jacc.2018.12.046 [DOI] [PubMed] [Google Scholar]
- 6. Devereaux PJ, Lamy A, Chan MTV, et al. ; VISION Cardiac Surgery Investigators. High-sensitivity troponin I after cardiac surgery and 30-day mortality. N Engl J Med. 2022;386:827-836. 10.1056/NEJMoa2000803 [DOI] [PubMed] [Google Scholar]
- 7. Lamy A, Devereaux PJ, Prabhakaran D, et al. ; CORONARY Investigators. Five-year outcomes after off-pump or on-pump coronary-artery bypass grafting. N Engl J Med. 2016;375:2359-2368. 10.1056/NEJMoa1601564 [DOI] [PubMed] [Google Scholar]
- 8. Deppe AC, Arbash W, Kuhn EW, et al. Current evidence of coronary artery bypass grafting off-pump versus on-pump: a systematic review with meta-analysis of over 16 900 patients investigated in randomized controlled trials. Eur J Cardiothorac Surg. 2016;49:1031-1041; discussion 1041. 10.1093/ejcts/ezv268 [DOI] [PubMed] [Google Scholar]
- 9. Afilalo J, Rasti M, Ohayon SM, Shimony A, Eisenberg MJ. Off-pump vs. on-pump coronary artery bypass surgery: an updated meta-analysis and meta-regression of randomized trials. Eur Heart J. 2012;33:1257-1267. 10.1093/eurheartj/ehr307 [DOI] [PubMed] [Google Scholar]
- 10. Thygesen K, Alpert JS, Jaffe AS, et al. ; Executive Group on behalf of the Joint European Society of Cardiology (ESC)/American College of Cardiology (ACC)/American Heart Association (AHA)/World Heart Federation (WHF) Task Force for the Universal Definition of Myocardial Infarction. Fourth universal definition of myocardial infarction (2018). Circulation. 2018;138:e618-e651. 10.1161/CIR.0000000000000617 [DOI] [PubMed] [Google Scholar]
- 11. Spagnolo M, Occhipinti G, Laudani C, Greco A, Capodanno D. Periprocedural myocardial infarction and injury. Eur Heart J Acute Cardiovasc Care. 2024;13:433-445. 10.1093/ehjacc/zuae014 [DOI] [PubMed] [Google Scholar]
- 12. Moussa ID, Klein LW, Shah B, et al. Consideration of a new definition of clinically relevant myocardial infarction after coronary revascularization. J Am Coll Cardiol. 2013;62:1563-1570. 10.1016/j.jacc.2013.08.720 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Garcia-Garcia HM, McFadden EP, Farb A, et al. ; Academic Research Consortium. Standardized end point definitions for coronary intervention trials: the academic research consortium-2 consensus document. Circulation. 2018;137:2635-2650. 10.1161/CIRCULATIONAHA.117.029289 [DOI] [PubMed] [Google Scholar]
- 14. Thielmann M, Sharma V, Al-Attar N, et al. ESC joint working groups on cardiovascular surgery and the cellular biology of the heart position paper: peri-operative myocardial injury and infarction in patients undergoing coronary artery bypass graft surgery. Eur Heart J. 2017;38:2392-2407. 10.1093/eurheartj/ehx383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Heuts S, Denessen EJS, Daemen JHT, et al. Meta-analysis evaluating high-sensitivity cardiac troponin T kinetics after coronary artery bypass grafting in relation to the current definitions of myocardial infarction. Am J Cardiol. 2022;163:25-31. 10.1016/j.amjcard.2021.09.049 [DOI] [PubMed] [Google Scholar]
- 16. Omran H, Deutsch MA, Groezinger E, et al. High-sensitivity cardiac troponin I after coronary artery bypass grafting for post-operative decision-making. Eur Heart J. 2022;43:2388-2403. 10.1093/eurheartj/ehab918 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wang J, Wang P, Liu H, et al. Impact of high-sensitivity cardiac troponin I elevation after on- and off-pump coronary artery bypass grafting on long-term prognosis. Can J Cardiol. 2025;41:294-305. 10.1016/j.cjca.2024.10.018 [DOI] [PubMed] [Google Scholar]
- 18. Pölzl L, Engler C, Sterzinger P, et al. Association of high-sensitivity cardiac troponin T with 30-day and 5-year mortality after cardiac surgery. J Am Coll Cardiol. 2023;82:1301-1312. 10.1016/j.jacc.2023.07.011 [DOI] [PubMed] [Google Scholar]
- 19. Chaitman BR, Alexander KP, Cyr DD, et al. ; ISCHEMIA Research Group. Myocardial infarction in the ISCHEMIA trial. Circulation. 2021;143:790-804. 10.1161/CIRCULATIONAHA.120.047987 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Stone GW, Sabik JF, Serruys PW, et al. ; EXCEL Trial Investigators. Everolimus-eluting stents or bypass surgery for left main coronary artery disease. N Engl J Med. 2016;375:2223-2235. 10.1056/NEJMoa1610227 [DOI] [PubMed] [Google Scholar]
- 21. Ben-Yehuda O, Chen S, Redfors B, et al. Impact of large periprocedural myocardial infarction on mortality after percutaneous coronary intervention and coronary artery bypass grafting for left main disease: an analysis from the EXCEL trial. Eur Heart J. 2019;40:1930-1941. 10.1093/eurheartj/ehz113 [DOI] [PubMed] [Google Scholar]
- 22. Gregson J, Stone GW, Ben-Yehuda O, et al. Implications of alternative definitions of peri-procedural myocardial infarction after coronary revascularization. J Am Coll Cardiol. 2020;76:1609-1621. 10.1016/j.jacc.2020.08.016 [DOI] [PubMed] [Google Scholar]
- 23. Hara H, Serruys PW, Takahashi K, et al. ; SYNTAX Extended Survival Investigators. Impact of peri-procedural myocardial infarction on outcomes after revascularization. J Am Coll Cardiol. 2020;76:1622-1639. 10.1016/j.jacc.2020.08.009 [DOI] [PubMed] [Google Scholar]
- 24. Gaudino M, Angelini GD, Antoniades C, et al. ; Arterial Grafting International Consortium (ATLANTIC) Alliance. Off‐pump coronary artery bypass grafting: 30 years of debate. J Am Heart Assoc. 2018;7:1–11. 10.1161/JAHA.118.009934 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. van Dijk D, Nierich AP, Jansen EWL, et al. ; Octopus Study Group. Early outcome after off-pump versus on-pump coronary bypass surgery. Circulation. 2001;104:1761-1766. 10.1161/hc4001.097036 [DOI] [PubMed] [Google Scholar]
- 26. Khan NE, De Souza A, Mister R, et al. A randomized comparison of off-pump and on-pump multivessel coronary-artery bypass surgery. N Engl J Med. 2004;350:21-28. 10.1056/NEJMoa031282 [DOI] [PubMed] [Google Scholar]
- 27. Mohammed AA, Agnihotri AK, van Kimmenade RRJ, et al. Prospective, comprehensive assessment of cardiac troponin T testing after coronary artery bypass graft surgery. Circulation. 2009;120:843-850. 10.1161/CIRCULATIONAHA.108.837278 [DOI] [PubMed] [Google Scholar]
- 28. Kavsak PA, Belley-Cote EP, Whitlock RP, Lamy A. Cardiac troponin testing in cardiac surgery. Expert Rev Cardiovasc Ther. 2023;21:729-731. 10.1080/14779072.2023.2283123 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article are available in the article and in its online supplementary material.