Share this article on:

Postoperative High-Sensitivity Troponin and Its Association With 30-Day and 12-Month, All-Cause Mortality in Patients Undergoing On-Pump Cardiac Surgery

Mauermann, Eckhard MD, MSc*; Bolliger, Daniel MD*; Fassl, Jens MD*; Grapow, Martin MD; Seeberger, Esther E. DAS*; Seeberger, Manfred D. MD; Filipovic, Miodrag MD; Lurati Buse, Giovanna A. L. MD, MSc*

doi: 10.1213/ANE.0000000000002023
Cardiovascular Anesthesiology: Original Clinical Research Report

BACKGROUND: Troponin T is a predictor of cardiac morbidity and mortality after cardiac surgery with most data examining fourth generational troponin T assays. We hypothesize that postoperative high-sensitivity troponin T (hsTnT) measured in increments of the upper limit of the norm independently predicts 30-day all-cause mortality.

METHODS: We included consecutive patients undergoing on-pump cardiac surgery from February 2010 to March 2012 in a prospective cohort that measured hsTnT at 0600 of the first and second postoperative day. Our primary end point was 30-day, all-cause mortality. The secondary end point was 12-month, all-cause mortality in patients surviving the first 30 days. We divided hsTnT into 5 predetermined categorizes based on the upper limit of the norm (ULN). We used Cox regression to examine an association of hsTnT independent of the EuroSCORE II at both 30 days as well as at 12 months in patients surviving the first 30 days. We assessed the area under the receiver operating characteristics curve and the net reassignment improvement for examining the benefit of adding of hsTnT to the EuroSCORE II for prognostication and restratification of 30-day, all-cause mortality.

RESULTS: We included 1122 of 1155 eligible patients (75% male; mean age 66 ± 11 years). We observed 58 (5.2%) deaths at 30 days and another 35 (3.4%) deaths at 12 months in patients surviving 30 days. HsTnT categorized by ULN exhibited a graded response for the mortality. Furthermore, hsTnT remained an independent predictor of all-cause mortality at 30 days (adjusted hazard ratio 1.019 [1.014–1.024] per 10-fold increase in ULN) as well as at 12 months (adjusted hazard ratio 1.019 [1.007–1.032]) in patients surviving the first 30 days. The addition of hsTnT to the EuroSCORE II significantly increased the area under the receiver operating characteristics curve (area under curve: 0.816 [95% confidence interval, 0.754–0.878] versus area under curve: 0.870 [95% confidence interval, 0.822–0.917], respectively; P = .012). Finally, adding hsTnT to the EuroSCORE II improved restratification by the net reassignment improvement, primarily by improving rule-out of events.

CONCLUSIONS: This analysis suggests that, similar to previous assays, higher postoperative concentrations of hsTnT are independently associated with all-cause mortality in patients undergoing on-pump cardiac surgery.

Supplemental Digital Content is available in the text.Published ahead of print May 19, 2017.

From the *Department for Anesthesia, Surgical Intensive Care, Prehospital Emergency Medicine and Pain Therapy, Basel University Hospital, Basel, Switzerland; Division of Cardiac Surgery, Basel University Hospital, Basel, Switzerland; and Basel University Medical School, Basel, Switzerland.

Accepted for publication January 30, 2017.

Published ahead of print May 19, 2017.

Funding: Financial support was provided by the Swiss National Science Foundation, Berne, Switzerland, as well as by the Department for Anesthesia, Surgical Intensive Care, Prehospital Emergency Medicine and Pain Therapy, University Hospital Basel, Switzerland. No relationships with industry exist.

The authors declare no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Reprints will not be available from the authors.

Address correspondence to Eckhard Mauermann, MD, MSc, Department for Anesthesia, Surgical Intensive Care, Prehospital Emergency Medicine and Pain Therapy, Basel University Hospital, Spitalstrasse 21, 4031 Basel, Switzerland. Address e-mail to Eckhard.Mauermann@usb.ch.

Cardiac surgery remains both common and high risk with 30-day mortality rates around 4%, depending on a multitude of patient, cardiac, and operational factors.1 Medical management of patients at risk of death or major adverse cardiac events requires their identification throughout the course of their hospital stay. A number of preoperative risk scores such as the EuroSCORE II1 exist,2–4 but their predictive value inherently lacks prognostically relevant information available during or after surgery. No biomarker-based risk score for cardiac patients exists postoperatively, potentially due to a number of different substrates and assays.1

However, postoperative biomarkers concentrations—especially troponins—have been shown to independently, reliably, and substantially improve prognostication in patients undergoing on-pump cardiac surgery.5–12 These postoperative measurements may be particularly interesting as they may convey information about the patient’s course during and after surgery and after initial risk stratification. Troponin elevations occur after virtually every kind of cardiac surgery.13,14 This troponin elevation may be procedural (eg, incomplete myocardial protection, reperfusion injury, surgical trauma, defibrillation, etc)15–17 or may reflect a relevant new pathology (eg, new myocardial ischemia), both of which may be relevant for patient outcomes. A comparison between substrates (eg, troponin I or troponin T) remains difficult, although utilization of the upper limit of the norm may help to improve comparability.18

A high-sensitivity troponin T (hsTnT) assay is increasingly being implemented worldwide, in part due to earlier identification of myocardial infarction (MI).19–21 Although several large studies have examined the independent prognostic value of conventional troponin on mortality in patients undergoing various cardiac procedures,5–12 no studies powered to assess mortality in cardiac surgery have been conducted using hsTnT as of yet. The main benefit of hsTnT assay appears to be earlier detection of low levels of troponin elevation. Although one would expect higher troponin concentrations to be similar between the conventional (fourth generational) and hsTnT assays, hsTnT may be interesting for a number of reasons. First, hsTnT is much more sensitive (99th percentile/upper limit of the norm [ULN]) and more precise (10% coefficient of variation) than conventional assays.22 This greatly decreases the troponin concentration fulfilling the requirements to define elevation (99th percentile and ≤10% coefficient of variation).23,24 Consequently, as the biomarker component of the third universal definition of MI after cardiac surgery is based on 10× the ULN,24 the troponin limit for MI decreases from 300 ng/L to 140 ng/L. Second, as the bulk of literature has used troponin assays at a specific point in time (often the first postoperative day), it would be important to confirm this timeline for hsTnT. In summary, it is uncertain what independent prognostic information hsTnT—especially when presented in terms of the ULN—may give the clinician postoperatively.

We hypothesized that postoperative hsTnT concentrations presented in terms of the ULN are independent predictors of 30-day, all-cause mortality in patients undergoing on-pump cardiac surgery when adjusting for preoperative risk assessment (EuroSCORE II). We further hypothesize that postoperative hsTnT in terms of the ULN also independently predicts 12-month, all-cause mortality in patients surviving up to postoperative day 30. Finally, we hypothesize that hsTnT improves prognostication and restratification of 30-day all-cause mortality independent of the EuroSCORE II.

Back to Top | Article Outline

METHODS

Study Design and Participants

This study was approved by an IRB and preregistered on clinicaltrials.gov (NCT00468598, Lurati/Filipovic, May 2, 2007). In this single-center, prospective cohort study, we examined consecutive adult patients undergoing coronary artery bypass graft, valve, and combined on-pump cardiac surgery from February 2010 to March 2012 at a university hospital. Patients undergoing deep hypothermic circulatory arrest or off-pump surgery were excluded. Patients were included independent of preoperative risk stratification, performance criteria, or preoperative laboratory values. All patients provided written informed consent for their participation. This manuscript adheres to Strengthening the Reporting of Observational Studies in Epidemiology guidelines.25 All preoperative, intraoperative, and postoperative management decisions were at the discretion of treating physicians. The protocol did not interfere with clinical management by the attending physicians and hsTnT concentrations were visible for clinicians within electronic charts. Attending physicians were free in their options for interpretation of hsTnT concentrations and subsequent work-up (eg, to diagnose MI) and treatment. Data were prospectively collected by trained research personnel.

Back to Top | Article Outline

High-Sensitivity Troponin T Analysis

HsTnT (Roche Diagnostics, Rotkreuz, Switzerland) was measured at 0600 on both the first and second postoperative days by dedicated laboratory technicians blinded to patient data and medical course. The 99th percentile and 10% coefficient of variation are 14 and 13 ng/L, respectively, with a limit of detection of 3 ng/L.22 The higher hsTnT concentration on the first or second postoperative days was taken for analyses. In the event of only 1 available measurement, this concentration was used.

Back to Top | Article Outline

End Points and Adjudication Thereof

The primary end point was 30-day, all-cause mortality. The secondary end point was all-cause mortality from day 30 up to 12 months, that is, 12-month all-cause mortality in patients surviving the first 30 days. As a significant proportion of deaths occurred within the first 30 days, this end point definition examines a possible continued risk beyond the first 30 days. The number of major adverse cardiac events (MACE) at 30 days was also examined as an additional secondary end point. MACE was defined as acute coronary syndrome, cardiac arrest, congestive heart failure requiring rehospitalization, and coronary revascularization by either percutaneous coronary intervention or surgery.

Follow-up occurred after 12 months by a questionnaire sent by mail. Patients were asked to state whether or not a subsequent hospitalization had been required. Patients not responding to the questionnaire were contacted by telephone by a trained study nurse blinded to biomarker concentrations. In the event of death or a subsequent hospitalization, the hospital and/or general practitioner was contacted for documentation. On the basis of this documentation, outcomes were adjudicated by 2 blinded senior anesthesiologists.

Back to Top | Article Outline

Statistical Analysis

Categorical data are presented as absolute numbers (percent). Continuous data are presented as mean ± standard deviation or median (quartile 1, quartile 3). Differences in these data are calculated by Pearson χ2 test, Fisher exact test, Student t test, or Mann-Whitney U test, as appropriate.

We transformed hsTnT concentrations to the number of times exceeding the ULN by dividing hsTnT concentrations by 14 ng/L, allowing for a comparison with the third universal definition of MI and comparison with other assays and biomarkers. We also examined biomarkers by surgery type, both graphically as well as by the Mann-Whitney U test). For descriptive purposes and comparability, hsTnT concentrations in ULN were divided into categories as previously published (0 to <10 ULN, 10 to <20 ULN, 20 to <40 ULN, 40 to <100 ULN, and ≥100 ULN).18 The 30-day mortality rate of these ULN categories was calculated and differences assessed by global and post hoc log rank tests.

Cox regression was performed with hsTnT examined per 10 ULN, adjusting for the EuroSCORE II at 30 days and also at 12 months in patients surviving 30 days. Improved prognostication of 30-day mortality through the addition of hsTnT to the EuroSCORE II versus simply the EuroSCORE II was also assessed by the area under the receiver operating characteristics curve (DeLong test). Finally, we also used the net reassignment improvement (NRI)26 to further assess the incremental value of hsTnT for risk restratification for 30-day, all-cause mortality when adding hsTnT to the EuroSCORE II. We categorized the risk of 30-day, all-cause mortality by (a) predefined convenience mortality rates (≤5%, >5 to ≤10%, >10 to ≤15%, and >15%) as well as (b) on the expected mortality quartiles based on the EuroSCORE II (ie, the logistic model without hsTnT). Statistical significance was shown if the 2-tailed P value was ≤ .05. All statistical analyses were conducted with R 3.2.2 (The R Foundation for Scientific Computing, Vienna, Austria).

Sample size estimation was based on the rule of thumb that 10–12 events are required per variable to be examined. Based on a 30-day mortality rate of 3.7%, another 2.9% in the first 12 months in patients surviving 30 days,12 and an estimated 1200 patients, we expected approximately 44 all-cause deaths within the first 30 days and another 35 thereafter. This allowed for a robust multivariate model with up to 3–4 variables.27

This study is an a priori continuation of a previous, preregistered study examining conventional (fourth generational) TnT.12 The study design (patient inclusion/exclusion criteria, time point of hsTnT measurements, the definition of the higher of the 2 hsTnT concentrations, the adjustment of hsTnT for the EuroSCORE II, and outcomes) was set before accessing the data. The use of Cox regression and NRI were also determined before accessing the data. The following were constructed after accessing the data: splitting the initial 12-month mortality outcome into 30-day mortality and 12-month mortality in patients surviving the first 30 days. Similarly, the transformation of hsTnT concentrations from ng/L to the number of times exceeding the ULN was made after accessing the data.

Back to Top | Article Outline

RESULTS

Descriptive Analysis

A total of 1155 patients undergoing on-pump cardiac surgery were eligible (Figure 1). Of these, 33 patients were excluded: 3 (0.3%) patients due to duplicate entries, 5 patients (0.4%) died, 16 patients (1.4%) suffered MACE before first day sampling, and an additional 9 patients (0.8%) lacked troponin measurements. Of the 1122 patients in the main analysis, 26 (2.3%) were lost to follow-up.

Figure 1

Figure 1

Within the first 30 days, 58 (5.2%) patients died and 106 (9.4%) suffered all-cause death and/or MACE. An additional 35 (3.4%) patients died after 30 days but within the first 12 months. During the same period of time, 72 patients (6.9%) suffered all-cause death and/or MACE. Forty-two percent died without ever having left the intensive care unit and 62% suffered in-hospital deaths. Total median time to death was 15 days (4–58 days).

A total of 64 MACEs in 48 patients was observed within the first 30 days. These were comprised of 17 (1.5%) acute coronary syndromes, 14 (1.2%) nonfatal cardiac arrests, 18 (1.6%) congestive heart failures requiring rehospitalization, and 15 (1.3%) revascularizations either by percutaneous coronary intervention (9 [0.8%]) or by coronary artery bypass graft (6 [0.5%]).

Patient characteristics as well as procedural and postoperative variables stratified by 30-day survivorship status are shown in Table 1. Peak troponin occurred on the first postoperative day in 81.2% of patients. Detailed information relating to hsTnT concentrations by surgical category is shown in Supplemental Digital Content 1, Table 1, http://links.lww.com/AA/B704 as well as Supplemental Digital Content 2, Figure 1, http://links.lww.com/AA/B705. Preoperative hsTnT was not measured routinely, but rather by clinical preference. We registered 141 patients (12.6%) with an acute coronary syndrome within 7 days before surgery. These patients had a median preoperative hsTnT of 265 ng/L (78–929 ng/L). They also exhibited a higher median postoperative peak hsTnT than patients without acute coronary syndrome in the 7 days before surgery (1010 ng/L [388–2140 ng/L] vs 447 ng/L [265–828 ng/L]; P < .001).

Table 1

Table 1

Figure 2

Figure 2

The top panel of Figure 2 shows the distribution of peak postoperative hsTnT concentrations by the number of times the ULN (black) and in ng/L (gray). Dashed gray lines indicate the predetermined categorical groupings by the ULN (ie, 0 to <10× ULN, 10 to <20× ULN, 20 to <40× ULN, 40 to <100× ULN, and ≥100× ULN). Of the 544 patients undergoing isolated coronary artery bypass graft surgery in this cohort, 92.3% exhibited peak postoperative hsTnT ≥10× ULN (ie, 140 ng/L), that is, fulfilled the biomarker definition of MI after cardiac surgery (albeit with largely unknown baseline concentrations). Had identical troponin concentrations been measured with a conventional assay, only 58.6% of patients would have exceeded ≥10× ULN (ie, 300 ng/L). Of all patients, 81.2% exhibited a higher concentration on the first rather than on the second postoperative day. The bottom panel of Figure 2 shows the 30-day mortality rate and 95% confidence interval (CI) for these 5 categories. Log-rank P values indicate a graded response beginning with the category of 40 to <100× ULN.

Back to Top | Article Outline

Association Postoperative Concentrations and Events

Supplemental Digital Content 3, Figure 2, http://links.lww.com/AA/B706, shows the Kaplan-Meier survival curves for 30-day and total 12-month mortality by hsTnT categories based on the ULN. Table 2 shows the results of the Cox regression for 30-day mortality, 12-month mortality in patients surviving 30 days, and for 30-day MACE. HsTnT was significantly associated with 30-day mortality when adjusting for the EuroSCORE II (adjusted hazard ratio 1.019 [95% CI, 1.014–1.024], per 10-fold increase in ULN). Similarly, hsTnT continued to predict mortality within 12 months beyond 30 days and was also associated with 30-day MACE.

Table 2

Table 2

Figure 3 shows the receiver operating characteristic curves for the prediction of 30-day mortality. The addition of hsTnT to the EuroSCORE II significantly improved prognostication (area under curve: 0.870 [95% CI, 0.822–0.917] versus area under curve: 0.816 [95% CI, 0.754–0.878]; P = .012).

Figure 3

Figure 3

Table 3

Table 3

(b) Risk Categories by EuroSCORE II Quartiles

(b) Risk Categories by EuroSCORE II Quartiles

Table 3 shows that in terms of risk stratification, adding hsTnT to the EuroSCORE II led to overall improvement of restratification in both (a) categorization of convenience mortality rates (Table 3A) as well as by (b) categorization of mortality rates based on the expected mortality quartiles from the EuroSCORE II (Table 3B). In both cases, the main effect came from the rule-out of events (ie, NRInonevents).

Back to Top | Article Outline

DISCUSSION

This prospective cohort study showed an association between postoperative hsTnT and 30-day, all-cause mortality in patients undergoing on-pump cardiac surgery independent of the EuroSCORE II. Furthermore, increased postoperative hsTnT was independently associated with 12-month mortality in patients surviving the first 30 days. Finally, adding hsTnT to the EuroSCORE II significantly improved prognostication and risk restratification of 30-day all-cause mortality.

Back to Top | Article Outline

Comparison With Previous Studies

The association of postoperative troponin and mortality after on-pump surgery has been explored in a number of smaller and a few larger studies.5,6,8–12,18 A positive association has been found in the short,6,9,10,18 middle,5,6 and long term,6,11 and for patients undergoing coronary artery bypass grafting,5,9,11 valve surgery,5,10 and combined/unselected surgery.5,6 However, this study incorporates a number of novel approaches and findings.

First, we examined hsTnT normed by the ULN to confirm the prognostic significance of this assay at 24 hours. The previous evidence on hsTnT assays after cardiac surgery and the resultant decrease in the ULN consists of a few smaller studies. Wang et al28 evaluated the use of hsTnT measured a median 18 hours after surgery for the diagnosis of Type 5 MI in 560 (68% of their initial sample) patients undergoing isolated on-pump and off-pump coronary artery bypass grafting. Interestingly, and similar to our study, some 90% of patients exhibited postoperative hsTnT concentrations above the 140 ng/L threshold (10× the ULN), of whom, however, only 78 had MI as defined by both a hsTnT ≥140 ng/L as well as electrocardiographic or echocardiographic evidence (eg, third universal definition). Our study also suggests that, when applying a high-sensitivity assay, the definition of 10× ULN required by the third universal definition of Type 5 MI may be of limited value as it was reached in more than 90% of both all patients as well as isolated coronary artery bypass graft-only patients. This finding supports the literature advocating the avoidance of the adoption of cut-off/definitions previously determined for conventional assays and the generation of data with high-sensitivity assays within specific populations.29 Potentially, another absolute cut-off or the troponin trend (eg, the delta hsTnT) may be more useful for diagnosing MI.

Second, our study examined hsTnT and both short-term 30-day, all-cause mortality as well as for the remainder of the first 12 postoperative months, assuming a patient had survived the first 30 days. This is important because the proximity of hsTnT elevation and death may have implications for management decisions and the 30-day time period is an established for the short term. At the same time, however, the full scope of mid-term mortality is incrementally assessed, that is, independent of the large degree of overlap in effect on account of the 30-day mortality. In terms of the hsTnT assay, Wang et al28 also assessed the association with mortality. HsTnT was not an independent predictor of 30-day mortality (adjusted hazard ratio per 100 ng/L = 1.07 [95% CI, 0.93–1.23]), but it was a predictor of mid-term mortality at a mean of 1.8 years (adjusted hazard ratio per 100 ng/L = 1.04 [95% CI, 1.00–1.08]). This is most likely because of a limited number of deaths (n = 10 at 30 days and n = 16 for mid-term survival). Our study, however, was adequately powered to assess mortality in both the first 30 days as well as 12-month mortality in patients surviving the first 30 days. Furthermore, as the primary goal of our analysis was not to diagnose MI, but rather to the address the independent prognostic value of hsTnT after on-pump cardiac surgery, we did not exclude relevant patient groups. The association of hsTnT and 30-day mortality seems comparable with other studies examining troponin T and short-term mortality in previous assays.7 Finally, we also found a graded response of hsTnT categorized by the ULN and 30-day mortality. In both the group of all patients (Figure 2) as well as within surgical categories (Supplemental Digital Content 2, Figure 1, http://links.lww.com/AA/B705), 30-day mortality tended to rise beginning in the category 40× ULN to <100× ULN. This is the same category of ULN in which troponin I began to show a rise in 30-day mortality.18 Furthermore, Domanski et al18 were able to show that troponin I was independently associated with 6-month mortality in patients having survived the first 30 days (adjusted hazard ratio per 50× ULN 1.15 [95% CI, 1.10–1.21]). No analyses at 12 months were conducted.

Third, risk restratification was significantly improved through consideration of postoperative troponin T concentrations, as illustrated by both the receiver operating characteristic curve as well as the NRI. Interestingly, the main force driving the significant restratification benefit of adding hsTnT to the EuroSCORE II seems to be the rule-out of events (ie, the NRInonevents) rather than the identification of at-risk patients (NRIevents). This is in contrast to our previous publication in which the identification of at-risk patients improved risk restratification, while the rule-out of events did not.12 We attribute this difference to the use of a dichotomous cut-off (set at 0.8 µg/L) in the previous publication and the use of a continuous variable in this study. Additionally, the previous publication considered both TnT and B-type natriuretic peptide in the risk prediction model.

Back to Top | Article Outline

Strengths and Limitations

This study has a number of strengths. First, we examined a large study population undergoing unselected on-pump cardiac surgery. As illustrated in Supplemental Digital Content 2, Figure 1, http://links.lww.com/AA/B705, this approach appears to be justified and showed a graded response of hsTnT and mortality within surgical categories as defined by the EuroSCORE II. Second, we evaluated the newest preoperative risk stratification score (EuroSCORE II) as well as the newest troponin assay, both of which have yet to be examined in this setting. Of the larger studies, only Adabag et al5 as well as previous work by this group8,12 had adjusted for the EuroSCORE, whereas the full EuroSCORE II has not been examined in larger cohorts addressing cardiac biomarkers. We consider the adjustment using the EuroSCORE II a significant improvement because the original EuroSCORE was criticized for overestimating mortality30,31 and reclassification consideration might have been misleading when applying the EuroSCORE as a baseline risk estimator. Third, hsTnT concentrations were available in more than 99% of patients and follow-up completeness was almost 98%. Fourth, outcomes were adjudicated independently by 2 blinded senior anesthesiologists. Finally, we examined both short-term mortality risk (30 days) as well as continuing and incremental mid-term mortality (12 months in patients surviving 30 days).

This study also had some relevant limitations. First, our study was guided by pragmatic intent. As such, blood was drawn at 0600 on both the first and second postoperative days together with morning blood sampling and not at a time point relative to surgery. However, as we examined the higher of the 2 hsTnT concentrations on the first and second postoperative days, it is unlikely that pragmatic timing may have altered results significantly. Second, we did not consider preoperative biomarkers, which may influence postoperative biomarker concentrations (eg, age, renal function, etc). HsTnT has been shown to be elevated for some time after MI. However, our hypothesis was that postoperative biomarkers may contain additional prognostic information beyond recommended preoperative risk assessment (eg, the EuroSCORE II). A sensitivity analysis excluding the 141 patients (12.5%) with acute coronary syndrome within 7 preoperative days did not significantly alter our results. The cut-off of 7 days was chosen so as not to dilute a potential effect by including later times. Finally, our main analyses included patients undergoing various procedures according to our initial definition of the study population. As in a number of previous studies,5,6,12 we combined surgery types after finding no evidence of an interaction of between the odds ratios of surgery types and all-cause mortality. Furthermore, we also clearly illustrated differences in hsTnT and mortality by surgery type (Supplemental Digital Content 1, Table 1, http://links.lww.com/AA/B704, and Supplemental Digital Content 2, Figure 1, http://links.lww.com/AA/B705). Finally, the type of surgery is also considered in the EuroSCORE II.

In summary, in patients undergoing on-pump cardiac surgery, postoperative hsTnT is an early predictor of all-cause mortality and morbidity occurring within the first 30 days as well as for all-cause mortality for the remainder of the first 12 months. Furthermore, hsTnT may be added to the EuroSCORE II to restratify patients and improve the early rule-out of short-term mortality.

Back to Top | Article Outline

DISCLOSURES

Name: Eckhard Mauermann, MD, MSc.

Contribution: This author conceived and designed the study, analyzed and interpreted the data, and wrote the manuscript.

Name: Daniel Bolliger, MD.

Contribution: This author helped analyze and interpret the data and critically revised the manuscript.

Name: Jens Fassl, MD.

Contribution: This author helped analyze and interpret the data and critically revised the manuscript.

Name: Martin Grapow, MD.

Contribution: This author helped analyze and interpret the data and critically revised the manuscript.

Name: Esther E. Seeberger, DAS.

Contribution: This author acquired data, helped analyze and interpret the data, and critically revised the manuscript.

Name: Manfred D. Seeberger, MD.

Contribution: This author helped analyze and interpret the data and critically revised the manuscript.

Name: Miodrag Filipovic, MD.

Contribution: This author helped analyze and interpret the data and critically revised the manuscript.

Name: Giovanna A. L. Lurati Buse, MD, MSc.

Contribution: This author conceived and designed the study, acquired, analyzed and interpreted the data, and drafted and critically revised the manuscript.

All authors were also actively involved in the original research project with the exception of E.M.

This manuscript was handled by: W. Scott Beattie, PhD, MD, FRCPC.

Back to Top | Article Outline

REFERENCES

1. Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41:734–744.
2. Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg. 1999;16:9–13.
3. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J. 2003;24:881–882.
4. Nilsson J, Algotsson L, Höglund P, Lührs C, Brandt J. Comparison of 19 pre-operative risk stratification models in open-heart surgery. Eur Heart J. 2006;27:867–874.
5. Adabag AS, Rector T, Mithani S, et al. Prognostic significance of elevated cardiac troponin I after heart surgery. Ann Thorac Surg. 2007;83:1744–1750.
6. Croal BL, Hillis GS, Gibson PH, et al. Relationship between postoperative cardiac troponin I levels and outcome of cardiac surgery. Circulation. 2006;114:1468–1475.
7. Lurati Buse GA, Koller MT, Grapow M, Bolliger D, Seeberger M, Filipovic M. The prognostic value of troponin release after adult cardiac surgery—a meta-analysis. Eur J Cardiothorac Surg. 2010;37:399–406.
8. Lurati Buse GA, Koller MT, Grapow M, et al. 12-month outcome after cardiac surgery: prediction by troponin T in combination with the European system for cardiac operative risk evaluation. Ann Thorac Surg. 2009;88:1806–1812.
9. Nesher N, Alghamdi AA, Singh SK. Troponin after cardiac surgery: a predictor or a phenomenon? Ann Thorac Surg. 2008;85:1348–1354.
10. Paparella D, Guida P, Caparrotti S. Myocardial damage influences short- and mid-term survival after valve surgery: a prospective multicenter study. J Thorac Cardiovasc Surg. 2014;148:2373–2379.e1.
11. Søraas CL, Friis C, Engebretsen KV, Sandvik L, Kjeldsen SE, Tønnessen T. Troponin T is a better predictor than creatine kinase-MB of long-term mortality after coronary artery bypass graft surgery. Am Heart J. 2012;164:779–785.
12. Lurati Buse GA, Bolliger D, Seeberger E. Troponin T and B-type natriuretic peptide after on-pump cardiac surgery: prognostic impact on 12-month mortality and major cardiac events after adjustment for postoperative complications. Circulation. 2014;130:948–957.
13. Mair J, Wieser C, Seibt I. Troponin T to diagnose myocardial infarction in bypass surgery. Lancet. 1991;337:434–435.
14. Katus HA, Schoeppenthau M, Tanzeem A. Non-invasive assessment of perioperative myocardial cell damage by circulating cardiac troponin T. Br Heart J. 1991;65:259–264.
15. Pichon H, Chocron S, Alwan K. Crystalloid versus cold blood cardioplegia and cardiac troponin I release. Circulation. 1997;96:316–320.
16. Swaanenburg JC, Loef BG, Volmer M. Creatine kinase MB, troponin I, and troponin T release patterns after coronary artery bypass grafting with or without cardiopulmonary bypass and after aortic and mitral valve surgery. Clin Chem. 2001;47:584–587.
17. Lehrke S, Steen H, Sievers HH. Cardiac troponin T for prediction of short- and long-term morbidity and mortality after elective open heart surgery. Clin Chem. 2004;50:1560–1567.
18. Domanski MJ, Mahaffey K, Hasselblad V. Association of myocardial enzyme elevation and survival following coronary artery bypass graft surgery. JAMA. 2011;305:585–591.
19. Reichlin T, Schindler C, Drexler B. One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T. Arch Intern Med. 2012;172:1211–1218.
20. Giannitsis E, Becker M, Kurz K, Hess G, Zdunek D, Katus HA. High-sensitivity cardiac troponin T for early prediction of evolving non-ST-segment elevation myocardial infarction in patients with suspected acute coronary syndrome and negative troponin results on admission. Clin Chem. 2010;56:642–650.
21. Roffi M, Patrono C, Collet JP; Management of acute coronary syndromes in patients presenting without persistent ST-segmentelevation of the European Society of Cardiology. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST-Segment Elevation of the European Society of Cardiology (ESC). Eur Heart J. 2016;37:267–315.
22. Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA. Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem. 2010;56:254–261.
23. Jaffe AS, Apple FS, Morrow DA, Lindahl B, Katus HA. Being rational about (im)precision: a statement from the Biochemistry Subcommittee of the Joint European Society of Cardiology/American College of Cardiology Foundation/American Heart Association/World Heart Federation Task Force for the definition of myocardial infarction. Clin Chem. 2010;56:941–943.
24. Thygesen K, Alpert JS, Jaffe AS; Joint ESC/ACCF/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction. Third universal definition of myocardial infarction. Circulation. 2012;126:2020–2035.
25. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, 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.
26. Pencina MJ, D’Agostino RB Sr, D’Agostino RB Jr, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med. 2008;27:157–172.
27. Bagley SC, White H, Golomb BA. Logistic regression in the medical literature: standards for use and reporting, with particular attention to one medical domain. J Clin Epidemiol. 2001;54:979–985.
28. Wang TK, Stewart RA, Ramanathan T, Kang N, Gamble G, White HD. Diagnosis of MI after CABG with high-sensitivity troponin T and new ECG or echocardiogram changes: relationship with mortality and validation of the universal definition of MI. Eur Heart J Acute Cardiovasc Care. 2013;2:323–333.
29. Sherwood MW, Kristin Newby L. High-sensitivity troponin assays: evidence, indications, and reasonable use. J Am Heart Assoc. 2014;3:e000403.
30. Basraon J, Chandrashekhar YS, John R. Comparison of risk scores to estimate perioperative mortality in aortic valve replacement surgery. Ann Thorac Surg. 2011;92:535–540.
31. Parolari A, Pesce LL, Trezzi M. EuroSCORE performance in valve surgery: a meta-analysis. Ann Thorac Surg. 2010;89:787–793.

Supplemental Digital Content

Back to Top | Article Outline
© 2017 International Anesthesia Research Society