Major Cardiac Events in Patients Admitted to Intensive Care After Vascular Noncardiac Surgery: A Retrospective Cohort

Abstract

Introduction. Patients proposed to vascular noncardiac surgery (VS) have several comorbidities associated with major adverse cardiac events (MACE). We evaluated incidence, predictors, and outcomes, and compared different scores to predict MACE after VS. Methods. We included all patients admitted from 2006 to 2013. Perioperative MACE included cardiac arrhythmias, myocardial infarction (MI), cardiogenic pulmonary edema (CPE), acute heart failure (AHF), and cardiac arrest (CA). Lee Revised Cardiac Risk Index (RCRI), Vascular Quality Initiative (VQI-CRI), Vascular Study Group of New England (VSG-CRI), and South African Vascular Surgical (SAVS-CRI) Cardiac Risk Indexes were calculated and analyzed. We performed multivariate logistic regression to assess independent predictors with calculation of odds ratio (OR) and 95% confidence interval (CI). To reduce overfitting, we used leave-one-out cross-validation approach. The Predictive ability of scores was tested using area under receiver operating characteristic curve (AUROC). Results. A total of 928 patients were included. We observed 81 MACE (28 MI, 22 arrhythmias, 10 CPE, 9 AHF, 12 CA) in 60 patients (6.5%): 3.3% in intermediate-risk surgery and 9.8% in high-risk surgery. Previous history of coronary artery disease (OR = 3.2, CI = 1.8-5.7), atrial fibrillation (OR = 5.1, CI = 2.4-11.0), insulin-treated diabetes mellitus (OR = 3.26, CI = 1.51-7.06), mechanical ventilation (OR = 2.75, CI = 1.41-4.63), and heart rate (OR = 1.02, CI = 1.01-1.03) at admission were considered independent risk factors in multivariate analysis. The AUROC of our model was 0.79, compared with RCRI (0.66), VSG-CRI (0.69), VQI-CRI (0.71), and SAVS-CRI (0.73). Conclusions. Observed MACE were within predicted range (1% to 5% after intermediate-risk surgery and >5% after high-risk surgery). SAVS-CRI and VQI-CRI had slightly better predictive capacity than VSG-CRI or RCRI.

Keywords

cardiac arrhythmia myocardial infarction pulmonary edema heart failure cardiac arrest intensive care unit vascular surgery

Introduction

Perioperative major adverse cardiac events (MACE) are common and increase length of hospital stay and mortality.¹ Lee et al² defined MACE as myocardial infarction, pulmonary edema (confirmed by chest radiograph in a plausible clinical setting), ventricular fibrillation or primary cardiac arrest (CA), and complete heart block. Patients submitted to vascular noncardiac surgery (VS) have several cardiovascular risk factors that are associated with MACE.³ The risk depends not only on patient or surgical factors but also on intra- or postoperative parameters.^1,3,4 Hemodynamic instability, blood loss, aortic cross clamping, reperfusion phenomena, and arterial embolism may increase the risk of complications after VS.⁴

Perioperative myocardial injury is often undetected because it does not exhibit typical symptoms of myocardial ischemia, such as chest pain, angina pectoris, or dyspnea.⁵ Acute myocardial infarction (MI) after major VS may range from 0.3% to 36%.⁶ This is why some authors recommend troponin screening after surgery.⁷ An accurate preoperative risk assessment is essential to guide patient management, allowing appropriate medical optimization, establish cardiac interventions, and early detect possible complications.⁸

Lee Revised Cardiac Risk Index (RCRI)² includes 6 independent predictors: high-risk type of surgery (suprainguinal vascular, intrathoracic, or intraperitoneal procedures); history of ischemic heart disease; history of congestive heart failure; history of cerebrovascular disease; preoperative treatment with insulin; and renal insufficiency (preoperative serum creatinine >2.0 mg/dL).² Although extensively used, RCRI may not be the best score to predict MACE after VS.⁹

More recently, Vascular Quality Initiative (VQI-CRI),¹⁰ Vascular Study Group of New England (VSG-CRI),¹¹ and South African Vascular Surgical (SAVS-CRI)¹² Cardiac Risk Indexes were derived to predict MACE after VS. The VQI-CRI has 5 variants depending on the surgery performed: carotid endarterectomy, endovascular aneurysm repair, open abdominal aortic aneurysm repair, suprainguinal bypass, and infrainguinal bypass. Age, type of surgery, history of coronary artery disease, diabetes, and creatinine concentration >1.8 mg/dL are included in all scores (Table 1). Only VQI-CRI uses critical limb ischemia, arterial hypertension, stress test status, and body mass index as predictors, whereas VSG-CRI and SAVS-CRI use chronic β-blockers as a risk factor and previous coronary surgery/percutaneous intervention as protective.

Table 1.

Comparison of the Different Cardiac Risk Scores^a.

RCRI	VQI-CRI^b	VSG-CRIs	SAVS-CRI
High-risk surgery (1)^c	Age	Age (2-4)	Age >65 years (2)
Ischemic heart disease (1)	Ischemic heart disease	Ischemic heart disease (2)	Ischemic heart disease (2)
Congestive heart failure (1)	Congestive heart failure	Congestive heart failure (2)	Suprainguinal surgery (7)
Cerebrovascular disease (1)	Chronic obstructive pulmonary disease	Chronic obstructive pulmonary disease (2)	Intermediate-risk surgery (3)
Insulin-treated diabetes mellitus (1)	Diabetes mellitus	Insulin-treated diabetes mellitus (1)	Diabetes mellitus (2)
Chronic kidney disease (creatinine >2.0 mg/dL) (1)	Chronic kidney disease (creatinine >1.8 mg/dL)	Chronic kidney disease (creatinine >1.8 mg/dL) (2)	β-blocker therapy (4)
	Critical limb ischemia	Current smoker (1)	CABG or PCI (-3)
	Arterial hypertension	β-blocker therapy (1)
	Abnormal cardiac stress test	CABG or PCI (−1)
	Body mass index

Abbreviations: RCRI, Revised Cardiac Risk Index; VQI, Vascular Quality Initiative; CRI, Cardiac Risk Index; VSG, Vascular Study Group of New England; SAVS, South African Vascular Surgery; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention.

Points in parenthesis.

Included variables and their relative weight is dependent on type of surgery.

Suprainguinal vascular, intrathoracic, or intraperitoneal procedures.

Our primary aim was to evaluate the incidence of MACE and its impact on the outcome of patients admitted to intensive care unit (ICU) after VS. Additionally, we wanted to identify and stratify the risk factors for MACE and compare the existing risk scores to predict MACE.

Methods

Study Design, Setting, and Participants

Retrospective cohort including all patients admitted to the surgical ICU after VS from January 2006 to July 2013 in a large academic hospital. We planned the analysis before looking at the data; exposures and outcomes were previously defined. The institutional ethics committee approved the protocol. This report complies with Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational cohort studies.¹³

Data Collection

We prospectively collected the following variables at ICU admission: age, gender, medical history, type of admission (elective or emergent), type of surgery, and ventilation. Surgeries were divided into intermediate-risk or high-risk surgery according to the joint guidelines of the European Society of Cardiology (ESC) and European Society of Anesthesiology (ESA).¹⁴ During ICU stay, prospective records included vital signs, laboratory results, and MACE such as cardiac arrhythmias, MI, cardiogenic pulmonary edema (CPE), acute heart failure (AHF), and CA. Acute MI was defined following the ESC/American College of Cardiology criteria.¹⁵ We also analyzed hospital or ICU mortality and length of stay (LOS). The ICU records at admission include the RCRI, whereas VSG-CRI, VQI-CRI, and SAVS-CRI were calculated in retrospective. We computed VQI-CRI according to the surgery performed.

Statistical Analysis

We used descriptive statistics to summarize data. We performed Kolmogorov-Smirnov test and histogram analysis to assess normality of data, and we selected parametric (independent samples t test) or nonparametric tests (Mann-Whitney U) accordingly. To compare proportions between groups in univariate analysis, we used the χ² test. We determined independent predictors of MACE using multivariate logistic regression with forward conditional method, calculating odds ratio (OR) and its 95% confidence interval (CI). We created a model using the adjusted OR of the independent variables as scoring points and analyzed the area under receiver operating characteristic (AUROC) of the different risk indexes to measure their predictive discrimination. We used the Hosmer-Lemeshow test to determine the goodness of fit of our model (calibration), P > .05 for no significant difference between predictive model and observed data. To reduce the potential of overfitting, we selected the leave-one-out cross-validation approach and the bootstrapping method. We performed Bonferroni correction for multiple comparisons. We used Stata 14 and SPSS 23 to analyze the data.

Results

During the study period, 928 patients were admitted to ICU after VS, and most of them were male. We included high-risk (open aortic surgery, lower limb revascularization or thromboembolectomy or amputation) and intermediate-risk surgeries (carotid endarterectomy, peripheral angioplasty, or endovascular aortic aneurysm repair). We observed 81 MACE (28 MI, 22 arrhythmias, 10 CPE, 9 AHF, and 12 CA) in 60 patients, representing an incidence of 6.5%: 3.3% in intermediate-risk surgery and 9.8% in high-risk surgery. Incidence of MI was 3.0% (28 out of 928 patients). Table 2 presents distribution of variables by MACE. Regarding type of admission, MACE incidence after elective surgery was 5.9% versus 12.3% after emergent surgery. Endovascular approach represented 10.8% of surgeries, and MACE incidence was no different from open surgery. Type of admission was not different in patients submitted to endovascular procedures; however, the endovascular group was older: 73 versus 67 years,P < .001. Type of ventilation (controlled vs spontaneous) but not surgery type increased ICU mortality and LOS. Carotid surgery (95% were submitted to cervical block regional anesthesia) had less MACE than lower limb or aortic surgery, 3.6% versus 8.1% versus 8.5%, respectively (P = .005).

Table 2.

Univariate Analysis of Major Adverse Cardiac Events (MACE).

Variables	No MACE Group (n = 868)	MACE Group (n = 60)	P
Age in years, median (IQR)	69 (61-76)	70 (60-77)	.458^a
Male gender, n (%)	697 (80.3)	47 (78.3)	.712^b
Prior medical history, n (%):
Arterial hypertension	438 (50.5)	31 (51.7)	.857^b
Diabetes mellitus	200 (23.0)	19 (31.7)	.128^b
Current smoker	182 (21.0)	13 (21.7)	.898^b
Peripheral arterial disease	250 (28.8)	22 (36.7)	.196^b
Coronary artery disease	278 (32.0)	35 (58.3)	<.001 ^b
Congestive heart failure	206 (23.7)	30 (50.0)	<.001^b
Atrial fibrillation	46 (5.3)	12 (20.0)	<.001 ^b
Cerebrovascular disease	350 (40.3)	19 (31.7)	.185^b
Chronic kidney disease	82 (9.4)	10 (16.7)	.070^b
Emergent surgery, n (%)	71 (8.2)	10 (16.7)	.025 ^b
High-risk surgery, n (%)	416 (48.3)	45 (75.0)	<.001 ^b
Endovascular surgery, n (%)	91 (10.5)	7 (11.7)	.773^b
At admission
Mechanical ventilation, n (%)	174 (20.0)	29 (48.3)	<.001 ^b
Vasopressors, n (%)	6 (0.7)	6 (10.0)	<.001 ^b
Body temperature, median (IQR)	35.8 (34.9-36.1)	35.8 (34.2-36.0)	.248^a
Systolic pressure, median (IQR)	132 (109-158)	119 (97-148)	.007 ^a
Mean arterial pressure, median (IQR)	89 (73-101)	82 (66-97)	.020 ^a
Heart rate, median (IQR)	78 (65-89)	88 (72-105)	<.001 ^a
Hematocrit, median (IQR)	33.0 (29.5-36.0)	30.6 (26.0-36.9)	.021 ^a
Serum urea, median (IQR)	33 (25-44)	40 (30-60)	.009 ^a
Serum creatinine, median (IQR)	0.9 (0.7-1.2)	1.1 (0.8-1.8)	.001 ^a
Serum potassium, median (IQR)	3.8 (3.5-4.1)	4.0 (3.5-4.3)	.155^a
Serum sodium, median (IQR)	139 (137-141)	139 (137-144)	.073^a
Leucocytes count, median (IQR)	10.2 (7.6-12.7)	12.4 (9.9-14.1)	.002 ^a
Length of stay (hours), median (IQR)	21 (16-43)	45 (21-90)	<.001 ^a
Hospital mortality, n (%)	28 (3.2)	16 (26.7)	<.001 ^b

Abbreviation: IQR, interquartile range (P25-P75).

Mann-Whitney test.

Chi-square test.

Table 3 displays the scores of the different cardiac risk indexes. All were significantly higher (P < .001) in the MACE group. Table 4 shows the multiple logistic regression used to assess the effect of variables on MACE at preadmission and at admission (the first OR column represents the univariate analysis OR while the adjusted OR column represents the forward selection). The surgery type was included in the multivariate analysis to control and adjust for confounding variables. Previous history of coronary artery disease, atrial fibrillation, insulin-treated diabetes mellitus, mechanical ventilation, and heart rate at admission were considered independent predictors. They increase the risk of MACE regardless of surgery type. The scoring system was based on the OR values for these factors, which were rounded to the nearest whole number. To include heart rate in our model, we used a categorical variable: <60; 60 to 80; 80 to 100; >100 beats per minute. The OR was 1.5 for every unit increase in that ordinal variable resulting in 0 to 4.5 model points (Table 4). Predictive ability remained the same after this transformation. Although hospital LOS and mortality differ between intermediate-risk and high-risk surgery, MACE remained an independent risk factor for mortality after adjusting for surgery type: OR 9.9 (5.0-19.9).

Table 3.

Cardiac Risk Indexes by Major Adverse Cardiac Events (MACE).

Risk Scores	No MACE Group (n = 884)	MACE Group (n = 44)	P
RCRI, median (IQR)	1 (1-2)	2 (1-3)	<.001 ^a
VQI-CRI, median (IQR)	7.3 (5.9-8.9)	9.3 (7.4-11.3)	<.001 ^a
VSG-CRI, median (IQR)	4 (2-6)	6 (4-9)	<.001 ^a
SAVS-CRI, median (IQR)	7 (5-9)	11 (8-13)	<.001 ^a

Abbreviations: RCRI, Revised Cardiac Risk Index; IQR, interquartile range (P25-P75); VQI, Vascular Quality Initiative; CRI, Cardiac Risk Index; VSG, Vascular Study Group of New England; SAVS, South African Vascular Surgery.

Mann-Whitney test.

Table 4.

Multivariate Analysis of Major Adverse Cardiac Events’ Predictors.

Variables	OR (95% CI)	P	Adjusted OR (95% CI)	P	Model Points
Coronary artery disease	2.97 (1.74-5.06)	<.001	3.18 (1.78-5.70)	<.001	3
Congestive heart failure	3.21 (1.89-5.46)	<.001	—	—
Atrial fibrillation	4.47 (2.22-8.99)	<.001	5.13 (2.39-11.00)	<.001	5
Diabetes mellitus ^a	3.75 (1.84-7.67)	<.001	3.26 (1.51-7.06)	.003	3
Emergent surgery	2.25 (1.09-4.62)	.028	—	—
High-risk surgery	3.22 (1.77-5.86)	<.001	—	—
Mechanical ventilation	3.73 (2.19-6.36)	<.001	2.75 (1.52-4.99)	.001	3
Systolic blood pressure	0.98 (0.97-0.99)	.012	—	—
Heart rate continuous	1.03 (1.01-1.04)	<.001	1.02 (1.01-1.03)	.002
Heart rate ordinal ^b	1.70 (1.27-2.28)	<.001	1.51 (1.11-2.06)	.006	1.5 increase per group ^b
Hematocrit	0.92 (0.88-0.97)	.002	—	—
Serum urea	1.01 (1.01-1.02)	.001	—	—
Serum creatinine	1.35 (1.11-1.63)	.002	—	—

Abbreviations: OR, odds ratio; CI, confidence interval.

Insulin-treated diabetes mellitus.

Heart rate points: 0 if <60; 1.5 if 60 to 80; 3 if 80 to 100; 4.5 if >100 beats per minute.

Figure 1 graphically displays the AUROC of the different scores. Our model had an AUROC of 0.79 with a Hosmer-Lemeshow test for the goodness of fit of 0.232 (good calibration). After leave-one-out cross-validation and bootstrapping, we obtained a similar AUROC of 0.77. The AUROC of RCRI (0.66) and VSG-CRI (0.69) were slightly worse than VQI-CRI (0.71) and SAVS-CRI (0.73). When comparing our model with the other scores, the P values were <.001 for RCRI, .002 for VSG-CRI, .0140 for VQI-CRI, and .0218 for SAV-CRI. After Bonferroni correction, only the P values for VSG-CRI and RCRI remained significant (P < .0125).

Discussion

Exact mechanisms of perioperative myocardial ischemia remain yet to be fully understood with multiple possible contributors. Postoperative period is potentially critical since sedation/analgesia may blunt any symptoms.^5,16 We believe that it is important to have an established protocol to systematically collect information to early detect MACE or MI. Definition of MACE or MI is different between studies and may explain some incidence variability (from 0.3% to 36%).⁶ Included surgeries and outcome measurements may also contribute to the disparity. Acute MI and MACE were prospectively collected in a cohort of 928 patients submitted to VS. Incidence of MACE was within the predicted range of ESC/ESA guidelines¹⁴ (1% to 5% after intermediate-risk surgery and >5% after high-risk surgery).

Medical history of atrial fibrillation, ischemic heart disease, or insulin-treated diabetes mellitus have been previously identified as risk factors for MACE.² The last 2 are included in RCRI and VSG, whereas VQI and SAVS include all diabetic patients. Bakker et al³ evaluated MACE in 1462 patients submitted to VS and reported that type 2 diabetes mellitus, independent of insulin use, is associated with an increased risk of cardiac complications after vascular surgery. Vanniyasingam et al¹⁷ also reported diabetes, together with brain natriuretic peptide, as risk factors for MACE after VS.

Both intra- and postoperative factors may contribute to MACE. van Lier et al¹ described tachycardia, anemia, hypoxemia, and hypotension as contributors to myocardial injury because of a supply and demand mismatch. We found mechanical ventilation and heart rate at admission to ICU as independent risk factors for MACE after VS. Scali et al¹⁸ reported less mortality when heart rate was less than 75 beats per minute, but the effect disappeared after controlling for β-blocker therapy. Unfortunately, it was impossible to determine β-blocker therapy in our sample.

A systematic review of 24 studies (792 740 patients) using the RCRI to predict MACE reported an AUROC of 0.75 for noncardiac surgery but less accurate (0.64) after VS.⁹ RCRI discriminates moderately well between patients at high versus low risk for MACE after noncardiac surgery; however, patients submitted to VS are at increased risk, especially because our sample includes intermediate-risk and high-risk surgery.⁹ Biomarkers such as brain natriuretic peptide, C-reactive protein, or copeptin may be used to predict outcomes after VS and improve the risk stratification capacity of the RCRI.^19-22 Another way of improving the RCRI is using age and history of arterial hipertension.²³ This was not true in our sample since age did not influence the risk of MACE. Despite the 5 variants of VQI depending on the surgery performed, its AUROC was similar to VSG or SAVS. The performance of the scores was better than previously reported in earlier studies.^24,25 All use serum creatinine to define chronic kidney disease, but they may be updated to use creatinine clearance in the near future.²⁶

Prediction is also influenced by the correct measurement of the outcomes. Some authors advocate troponin measurements systematically during the first 3 to 5 days after surgery.^27-29 High-sensitive troponin may be more sensitive but not as specific. These measurements are important regardless of symptoms because only 6% of patients have typical chest pain.⁵ In our study, acute MI and MACE were prospectively and systematically collected during ICU stay, but during that period, we used regular instead of high-sensitivity troponin. The exact values of troponin may influence short-term and long-term mortality until 5 years after surgery.³⁰

Some risk scores predict mortality/morbidity after VS but do not discriminate the site of complications. Others are specific for MACE but were developed including many types of noncardiac surgery and not specifically VS. Gupta et al reported that perioperative risk for myocardial infarction or cardiac arrest³¹ allows differentiating the type of surgery, but we did not have the functional status data to calculate it. The American College of Surgeons developed a surgical risk calculator, but authors did not make the equation available for clinical research to protect intellectual property and because they thought that external validation was not necessary.

Although all scores include age as a risk factor, we did not find it in our study. This may be because older patients were submitted to less invasive endovascular procedures, also explaining why endovascular had the same MACE incidence as open surgery. Despite a better prediction using our model, its usefulness in stratifying preoperative risk may be limited since some variables refer to parameters at admission to ICU. The sample size led authors to use the leave-one-out cross-validation approach and the bootstrapping method instead of the division into derivation and validation cohort. This is a limitation considering our model, but it does not interfere with the comparison of the other scores. Another limitation is that we have vital signs at ICU admission but not during the intraoperative period. In addition, we used only intermediate-risk or high-risk surgical patients admitted to the surgical ICU and recorded MACE only during ICU stay.

Conclusions

Observed MACE were within the predicted range (1% to 5% after intermediate-risk surgery and >5% after high-risk surgery). Previous history of coronary artery disease, atrial fibrillation, insulin-treated diabetes mellitus, mechanical ventilation, and heart rate at admission were considered independent risk factors. The SAVS-CRI and VQI-CRI had slightly better predictive capacity than the VSG-CRI or the RCRI.

Footnotes

Author Contributions

All authors were involved in data collection. Pedro Videira Reis was responsible for data analysis and manuscript writing. Fernando Abelha coordinated the project and revised the manuscript. All authors approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Pedro Videira Reis

References

van Lier

Wesdorp

Liem

VGB

, et al. Association between postoperative mean arterial blood pressure and myocardial injury after noncardiac surgery. Br J Anaesth. 2018;120:77-83.

Lee

Marcantonio

Mangione

, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043-1049.

Bakker

Valentijn

van de Luijtgaarden

, et al. Type 2 diabetes mellitus, independent of insulin use, is associated with an increased risk of cardiac complications after vascular surgery. Anaesth Intensive Care. 2013;41:584-590.

Biccard

Scott

DJA

Chan

MTV

, et al. Myocardial injury after noncardiac surgery (MINS) in vascular surgical patients: a prospective observational cohort study. Ann Surg. 2018;268:357-363.

Puelacher

Buse

Seeberger

, et al. Perioperative myocardial injury after noncardiac surgery: incidence, mortality, and characterization. Circulation. 2018;137:1221-1232.

Ollila

Vikatmaa

Virolainen

, et al. Perioperative myocardial infarction in non-cardiac surgery patients: a prospective observational study. Scand J Surg. 2017;106:180-186.

Liem

VGB

Hoeks

Grüne

, et al. Prognostic value of postoperative high-sensitivity troponin T in patients with different stages of kidney disease undergoing noncardiac surgery. Br J Anaesth. 2018;120:84-93.

Liu

Cohen

Sweitzer

BJ.

Defining the intrinsic cardiac risks of operations to improve preoperative cardiac risk assessment. Anesthesiology. 2018;128:283-292.

Ford

Beattle

Wijeysundera

DN.

Systematic review: prediction of perioperative cardiac complications and mortality by the Revised Cardiac Risk Index. Ann Intern Med. 2010;152:26-35.

10.

Bertges

Neal

Schanzer

, et al. The Vascular Quality Initiative Cardiac Risk Index for prediction of myocardial infarction after vascular surgery. J Vasc Surg. 2016;64:1411-1421.e4.

11.

Bertges

Goodney

Zhao

, et al. The Vascular Study Group of New England Cardiac Risk Index (VSG-CRI) predicts cardiac complications more accurately than the Revised Cardiac Risk Index in vascular surgery patients. J Vasc Surg. 2010;52:674-683,683.e1-683.e3.

12.

Moodley

Naidoo

Biccard

BM.

The South African Vascular Surgical Cardiac Risk Index (SAVS-CRI): a prospective observational study. S Afr Med J. 2013;103:746-750.

13.

von Elm

Altman

Egger

Pocock

Gøtzsche

Vandenbroucke

JP;

STROBE

Initiative.

The Strengthening the Reporting of observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370:1453-1457.

14.

Kristensen

Knuuti

Saraste

, et al; Authors/Task Force Members. 2014 ESC/ESA guidelines on non-cardiac surgery: cardiovascular assessment and management: the Joint Task Force on non-cardiac surgery: cardiovascular assessment and management of the European Society of Cardiology (ESC) and the European Society of Anaesthesiology (ESA). Eur J Anaesthesiol. 2014;31:517-573.

15.

Alpert

Thygesen

Antman

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;36:959-969.

16.

Biccard

BM.

Detection and management of perioperative myocardial ischemia. Curr Opin Anaesthesiol. 2014;27:336-343.

17.

Vanniyasingam

Rodseth

Buse

GAL

, et al. Predicting the occurrence of major adverse cardiac events within 30 days of a vascular surgery: an empirical comparison of the minimum p value method and ROC curve approach using individual patient data meta-analysis. Springerplus. 2016;5:304.

18.

Scali

Bertges

Neal

, et al. Heart rate variables in the Vascular Quality Initiative are not reliable predictors of adverse cardiac outcomes or mortality after major elective vascular surgery. J Vasc Surg. 2015;62:710-720.e9.

19.

Schrimpf

Gillmann

Sahlmann

, et al. Renal function interferes with copeptin in prediction of major adverse cardiac events in patients undergoing vascular surgery. PLoS One. 2015;10:e0123093.

20.

Jarai

Mahla

Perkmann

, et al. Usefulness of pre-operative copeptin concentrations to predict post-operative outcome after major vascular surgery. Am J Cardiol. 2011;108:1188-1195.

21.

Patelis

Kouvelos

Koutsoumpelis

Moris

Matsagkas

Arnaoutoglou

An update on predictive biomarkers for major adverse cardiovascular events in patients undergoing vascular surgery. J Clin Anesth. 2016;33:105-116.

22.

Singh

Morris

Smith

Moxon

Golledge

Systematic review and meta-analysis of the association between C-reactive protein and major cardiovascular events in patients with peripheral artery disease. Eur J Vasc Endovasc Surg. 2017;54:220-233.

23.

Welten

Schouten

van Domburg

, et al. The influence of aging on the prognostic value of the revised cardiac risk index for postoperative cardiac complications in vascular surgery patients. Eur J Vasc Endovasc Surg. 2007;34:632-638.

24.

Parmar

Torella

Prediction of major adverse cardiac events in vascular surgery: are cardiac risk scores of any practical value?

Vasc Endovascular Surg. 2010;44:14-19.

25.

Smeili

Lotufo

PA.

Incidence and predictors of cardiovascular complications and death after vascular surgery. Arq Bras Cardiol. 2015;105:510-518.

26.

Rochanov

Walsh

Deveraux

, et al. External validation of the Revised Cardiac Risk Index and update of its renal variable to predict 30-day risk of major cardiac complications after non-cardiac surgery: rationale and plan for analyses of the VISION study. BMJ Open. 2017;7:e013510.

27.

Gillmann

Meinders

Grohennig

, et al. Perioperative levels and changes of high-sensitivity troponin T are associated with cardiovascular events in vascular surgery patients. Crit Care Med. 2014;42:1498-1506.

28.

Sessler

Devereaux

PJ.

Perioperative troponin screening. Anesth Analg. 2016;123:359-360.

29.

Thomas

Cotter

Williams

van Rij

AM.

Diagnosis, incidence, and clinical implications of perioperative myocardial injury in vascular surgery. Vasc Endovasc Surg. 2016;50:247-255.

30.

Reed

Horr

Young

, et al. Associations between cardiac troponin, mechanism of myocardial injury, and long-term mortality after noncardiac vascular surgery. J Am Heart Assoc. 2017;6:e005672.

31.

Gupta

Sundaram

, et al. Development and validation of a risk calculator for prediction of cardiac risk after surgery. Circulation. 2011;124:381-387.