High-sensitivity Troponin I Predicts Major Cardiovascular Events after Non-Cardiac Surgery: A Vascular Events in Non-Cardiac Surgery Patients Cohort Evaluation (VISION) Substudy Free

Abstract

Background

Myocardial injury after non-cardiac surgery (MINS), based on measurement of troponin T, is associated with perioperative major adverse cardiovascular events (MACE). We therefore determined the high-sensitivity troponin I (hsTnI) thresholds associated with 30 day MACE after non-cardiac surgery.

Methods

We performed a nested biobank cohort study of 4553 patients from the Vascular Events in Non-Cardiac Surgery Patients Cohort Evaluation (VISION) Study. We measured hsTnI (ADVIA Centaur® hsTnI assay) on postoperative days 1 to 3 in patients ≥45 years undergoing non-cardiac surgery. An iterative Cox proportional hazard model determined peak postoperative hsTnI thresholds independently associated with MACE (i.e., death, myocardial infarction occurring on postoperative day 4 or after, non-fatal cardiac arrest, or congestive heart failure) within 30 days after surgery.

Results

MACE occurred in 89/4545 (2.0%) patients. Peak hsTnI values of <75 ng/L, 75 ng/L to <1000 ng/L, and ≥1000 ng/L were associated with 1.2% (95% CI, 0.9–1.6), 7.1% (95% CI, 4.8–10.5), and 25.9% (95% CI, 16.3–38.4) MACE, respectively. Compared to peak hsTnI <75 ng/L, values 75 ng/L to <1000 ng/L and ≥1000 ng/L were associated with adjusted hazard ratios (aHR) of 4.53 (95% CI, 2.75–7.48) and 16.17 (95% CI, 8.70–30.07), respectively. MACE was observed in 9% of patients with peak hsTnI ≥75 ng/L vs 1% in patients with peak hsTnI <75 ng/L (aHR 5.76; 95% CI, 3.64–9.11). A peak hsTnI ≥75 ng/L was associated with MACE in the presence (aHR 9.35; 95% CI, 5.28–16.55) or absence (aHR 3.99; 95% CI, 2.19–7.25) of ischemic features of myocardial injury.

Conclusion

A peak postoperative hsTnI ≥75 ng/L was associated with >5-fold increase in the risk of 30 days MACE compared to levels <75 ng/L. This threshold could be used for MINS diagnosis when the ADVIA Centaur hsTnI assay is used.

Clinicaltrials.gov Registration Number: NCT00512109.

Introduction

Myocardial injury after non-cardiac surgery (MINS) is a prognostically relevant troponin elevation thought to be due to ischemia (i.e., excluding non-cardiac causes for troponin elevation such as sepsis or pulmonary embolism) within 30 days after non-cardiac surgery. MINS includes patients with diagnosis of myocardial infarction (MI) and those that do not fulfil the definition of MI (i.e., patients with acute troponin elevation due to ischemic etiology without clinical ischemic features such as chest pain or electrocardiographic ischemic changes) (1). MINS occurs in up to 1 in 6 patients, depending on patient and surgery-related characteristics (2). MINS is associated with serious short- and long-term cardiovascular events including mortality (3–5).

The VISION Study (Vascular Events in Non-Cardiac Surgery Patients Cohort Evaluation; ClinicalTrials.gov NCT00512109) was a large (n = 40 004) international prospective cohort that evaluated major complications in patients undergoing non-cardiac surgery (6). It has been established that elevated values of the fourth-generation troponin T (TnT) assay (1) and the fifth-generation high-sensitivity TnT assay (hsTnT) (4) are strong independent predictors of 30-day mortality after non-cardiac surgery. For that reason, the diagnostic criteria for MINS were established with these assays, and the troponin thresholds for the diagnostic criteria are specific to them. It is not known what thresholds should be used for high-sensitivity troponin I (hsTnI) as the diagnostic criteria for MINS which would optimally predict 30-day major adverse cardiovascular events (MACE). While there is only one hsTnT assay, there are several hsTnI assays, with different immunoassays and varying reference ranges. Thus, in the absence of hsTnI assay standardization each hsTnI assay may require validation when establishing new thresholds for MINS.

This study aims to determine the thresholds for the ADVIA Centaur® High-Sensitivity Troponin I Assay. We evaluated patients recruited into the VISION Biobank to determine: (a) the association between peak hsTnI levels during the first 3 postoperative days and their relationship to the risk of MACE at 30 days after surgery; (b) the optimal hsTnI thresholds to predict the risk of MACE 30 days after surgery; (c) the diagnostic criteria for MINS based on this assay.

Materials and Methods

VISION eligibility criteria have been described elsewhere (4). This nested cohort within the VISION Study population included all patients (≥45 years who underwent non-cardiac surgery, under regional or general anesthesia, that required overnight hospital admission) enrolled in the VISION Biobank who had at least one blood sample drawn postoperatively. All VISION Study sites were offered the opportunity to participate in the Biobank. Some sites were not able to facilitate the collection, processing, storage, and shipment requirements. Participating sites tried to include all patients in the Biobank after the Biobank was initiated in their centre. Blood samples were obtained preoperatively, and on days 1 to 3 after surgery (4).

This study was conducted in compliance with the Declaration of Helsinki. All participating sites had institutional/ethics review board approval before starting patient enrollment. Patients in the VISION Biobank provided consent to have blood samples collected for future analysis. Samples were shipped and stored below −70°C in the Clinical Research Laboratory and Biobank in Hamilton, Canada. Samples were then thawed, centrifuged, and hsTnI was measured with the ADVIA Centaur High-Sensitivity Troponin I (TNIH) Assay (manufacturer overall serum 99th percentile = 46.5 ng/L) using the ADVIA Centaur XP system from Siemens Healthcare Diagnostics.

Patients with a troponin elevation obtained during the main VISION Study were assessed for ischemic symptoms and electrocardiographic ischemic findings (4). Research personnel followed patients while in hospital and performed a 30 day follow-up call. Variables and outcomes data were collected through patients, next of kin interview, and chart review. Study personnel obtained source documents related to outcome events from the primary care physician or hospital records. Case report forms and supporting documentation were stored at the coordination centre (Population Health Research Institute, Hamilton, Canada) in a secure data management system. Experts adjudicated all troponin elevations above the 99th percentile to determine if it was believed to be due to an ischemic etiology and if they met the third universal definition of MI (7). Adjudicators were blinded to whether the sites reported the event as an MI or not. The primary outcome was the composite of MACE (i.e., all-cause death, MI occurring on postoperative day 4 or after, non-fatal cardiac arrest, or congestive heart failure [CHF] within 30 days after surgery). MI was defined by the universal definition (7), diagnosed with any troponin assay, and occurring between postoperative days 4 and 30. As we monitored troponins in the first 3 days after surgery, and MINS diagnosis includes MI diagnosis, we excluded MIs happening during this period to properly address the prognostic impact of troponin elevation to predict future MIs. All other outcomes were assessed between surgery and 30 days.

Sample size was based upon the primary objective, which was to determine the relationship between the peak postoperative hsTnI measurement and the risk of MACE during the first 30 days after surgery. A sample size of 4500 ensured a stable model with >10 events per variable in the 7-covariate model based on a 2.0% event rate of the primary outcome (online Supplemental Table 1) (8, 9).

Statistical Analysis

We considered the peak troponin level the highest value obtained in the first 3 days after surgery. For patients who experienced the primary outcome within the first 3 postoperative days (i.e., MI not included), the hsTnI value on the day of the event or earlier was considered the peak value, and subsequent hsTnI values were not used. We excluded patients who had a primary outcome on the day of surgery (n = 8) from the analysis of hsTnI as a predictor of the primary outcome. We censored patients at the time of last assessment if they did not complete the 30-day follow-up. We analysed outcomes according to the Adjudication Committee.

We undertook a Cox proportional hazards model analysis. The dependent variable was MACE at 30 days after non-cardiac surgery. Independent variables were independent predictors of 30-day mortality previously demonstrated in VISION analyses (i.e., age, peripheral vascular disease, chronic obstructive pulmonary disease, urgent or emergent surgery, active cancer; general surgery vs other surgery, and peak postoperative hsTnI) (4).

We used a modification of the Mazumdar approach (i.e., an iterative process to explore potential hsTnI thresholds) to determine if there were prognostically important postoperative hsTnI thresholds (i.e., a hazard ratio [HR] ≥3 and a 30 day risk of MACE ≥3%, with a P-value <0.01) to predict MACE at 30 days (10). If more than one hsTnI value fulfilled these criteria, the lowest value fulfilling the criteria would represent the optimal threshold. We pre-specified exploration of the following hsTnI thresholds: from 5 to 100 ng/L by increments of 5 ng/L (except that 46 was used instead of 45 ng/L because 46 ng/L represents the 99th percentile for this assay), 100 to 200 ng/L by increments of 10 ng/L, and 200 to 1000 ng/L by increments of 100 ng/L. We used a conservative P-value <0.01 to account for multiple testing.

We investigated if an absolute change between preoperative and peak postoperative hsTnI was associated with an important increase of MACE using the strategy outlined above. Additional post-hoc analyses were performed assessing sensitivity, specificity, and likelihood ratio for the 50, 75, and 100 ng/L hsTnI cutoffs.

To establish MINS diagnostic criteria, we excluded 24 patients (6 events) with troponin elevation due to a clear non-ischemic etiology (online Supplemental Table 2); these patients were kept in the other analyses and considered not to have suffered a MINS. Patients with an elevated hsTnI value before surgery that was higher than or equal to the peak postoperative hsTnI value were considered as no elevation. We undertook a Cox proportional hazards model where the dependent variable was MACE within 30 days after surgery and the same independent variables described above. This model included only patients with a hsTnI value equal or greater than the lowest absolute threshold (i.e., the lowest troponin value associated with a HR ≥3 and a 30-day risk of MACE ≥3%) with an ischemic feature, and separately an elevated postoperative hsTnI measurement without an ischemic feature as independent variables. We established a priori that if an elevated hsTnI with and without ischemic features independently predicted MACE at 30 days, the MINS diagnostic criteria would only require an elevated hsTnI, without the need for the presence of an ischemic feature. We would establish MINS diagnostic criteria as a peak hsTnI greater than or equal to the important threshold.

For regression analyses, we report the adjusted HR (aHR), 95% confidence intervals (CI), and P-value. We assessed model discrimination using the C-statistic. All tests were two-sided and a P < 0.05 was considered statistically significant, unless specified otherwise. Analyses were performed using SAS version 9.4 (SAS Institute Inc) and R version 3.3.2 (R Project). This manuscript adheres to the applicable Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Results

We included 4553 patients from 9 sites across Canada, the United Kingdom, Hong Kong, and the USA (online Supplemental Table 3). Preoperative and postoperative day 1 samples were available for 4141 and 4176 patients, respectively (online Supplemental Table 4). Overall, mean age was 65.1 (standard deviation 10.6) years, 49.2% were female, 56.9% had hypertension, 19.0% had diabetes, and 16% a history of coronary artery disease. Table 1 and online Supplemental Table 5 presents patients’ baseline characteristics. We obtained 30 day follow-up data on 4544/4553 (99.8%) patients, and 9 patients were censored at hospital discharge.

The composite of MACE occurred in 89/4545 (2.0%) patients at 30 days. The distribution of each component of the primary outcome was the following: 35 patients died, 15 patients had a MI after postoperative day 3, 4 patients had a non-fatal cardiac arrest, and 44 patients had a new acute CHF within 30 days after surgery. Peak postoperative hsTnI was an independent predictor of the primary outcome (aHR 3.17; 95% CI, 1.61–6.23); P < 0.001; online Supplemental Table 6). Table 2 presents the optimal thresholds of peak postoperative hsTnI to predict the primary outcome. Peak hsTnI values of <75 ng/L, 75 ng/L to <1000 ng/L, and ≥1000 ng/L were associated with an incidence of MACE of 1.2% (95% CI, 0.9–1.6), 7.1% (95% CI, 4.8–10.5), and 25.9% (95% CI, 16.3–38.4), respectively. Compared to peak hsTnI <75 ng/L (reference), hsTnI values 75 ng/L to <1000 ng/L and ≥1000 ng/L were associated with an aHR of 4.53 (95% CI, 2.75–7.48) and 16.17 (95% CI, 8.70–30.07), respectively. These peak hsTnI thresholds were associated with higher incidence of all-cause mortality (online Supplemental Table 7). The likelihood ratio test did not improve the model by including the pre-specified absolute changes from 5 to 75 ng/L (n = 3781 patients, 41 events) by increments of 5 ng/L compared to the model that did not include any change (P = 0.051). The best cutpoint was a change of 40 ng/L; however, this yielded a corresponding HR <1. The post-hoc analyses assessing sensitivity, specificity, and likelihood ratio for the 50, 75, and 100 ng/L hsTnI cutoffs are reported in online Supplemental Table 8.

Incidence of MACE was 31/343 (9%) in patients with postoperative peak hsTnI ≥75 ng/L vs 52/4178 (1%) in patients with postoperative peak hsTnI <75 ng/L (aHR 5.76; 95% CI, 3.64–9.11). A postoperative peak hsTnI ≥75 ng/L was associated with increased risk of MACE either in the presence (aHR 9.35; 95% CI, 5.28–16.55) or absence (aHR 3.99; 95% CI, 2.19–7.25) of ischemic features around the time of the peak troponin elevation (Table 3).

Considering those findings and our pre-specified criteria, we determined our diagnostic criteria for MINS as a peak postoperative hsTnI ≥75 ng/L using the ADVIA Centaur High-Sensitivity Troponin I Assay in patients with a presumed ischemic etiology for the troponin elevation, regardless of the presence of an ischemic feature. Based on these criteria, the incidence of MINS in this cohort was 7.6% (348/4553 patients). Patients with MINS had a higher incidence of death (12/343 [3.5%] vs 21/1478 [0.5%]; odds ratio [OR] 7.18; 95% CI, 3.50–14.41) and CHF (18/343 [5.2%] vs 22/4178 [0.5%]; OR 10.46; 95% CI, 5.56–19.71) compared with patients without MINS (online Supplemental Table 9). Most patients with a MINS diagnosis (85.9%) were asymptomatic (online Supplemental Table 10).

Table 4 reports the Cox model of variables associated with MACE 30 days after surgery. MINS, based on the criteria determined above, was the strongest predictor of the primary outcome (aHR 5.76; 95% CI, 3.64–9.11). Cox models demonstrated no interaction between the 3 established important thresholds (i.e., <75 ng/L, 75 to <1000 ng/L, and ≥1000 ng/L) and sex (interaction P value 0.823; online Supplemental Table 11). An additional post-hoc analysis including preoperative N-terminal pro brain natriuretic peptide (NTproBNP) in the Cox model demonstrated similar results (online Supplemental Table 12).

Discussion

Our study demonstrates that postoperative hsTnI levels measured by the ADVIA Centaur High-Sensitivity Troponin I Assay independently predict 30 day MACE in patients undergoing non-cardiac surgery. Patients with a peak hsTnI ≥75 ng/L in the first 3 days after surgery, due to an ischemic etiology, are at higher risk of suffering MACE 30 days after surgery compared to patients with lower hsTnI levels (aHR 5.76; 95% CI, 3.64–9.11). Patients with asymptomatic peak postoperative hsTnI ≥75 ng/L had a 4-fold increase in the risk of MACE at 30 days as compared to peak postoperative hsTnI <75 ng/L (aHR 3.99; 95% CI, 2.19–7.25). Therefore, for a diagnosis of MINS, physicians should consider using the proposed threshold of 75 ng/L or greater for the ADVIA Centaur High-Sensitivity Troponin I Assay.

Previous studies have demonstrated that MINS is associated with adverse prognosis. Smilowitz and colleagues performed a systematic review (169 studies, >530 000 participants) demonstrating that patients with MINS had higher 30-day mortality (relative risk [RR] 5.6; 95% CI, 4.1–7.7) and 1-year mortality after non-cardiac surgery (RR 4.1; 95% CI, 3.0–5.6) compared to patients without MINS (2). However, few studies aimed to identify the troponin threshold predicting adverse clinical outcomes and defining MINS in patients with postoperative troponin elevation. Although the risk is continuous, establishing prognostic thresholds is relevant in clinical practice to guide physicians on assessment of patients, to implement strategies to better monitor or risk stratify, and to consider therapies to mitigate further events in high-risk patients. Prognostic thresholds are also helpful to identify low-risk patients that may benefit from early discharge or a shorter period of cardiovascular monitoring when the negative predictive value is high, as in this case (98%).

With the advent of the high-sensitivity assays, interpretation of positive troponin results has become even more critical due to the ability to detect troponin elevations at lower levels (11, 12). Clinicians cannot assume that troponin values from different assays are comparable (13, 14). However, the ADVIA Centaur and the Atellica IM assays are closely correlated with a slope around 1.0, a very low intercept, and a high correlation factor (15). Therefore, although this study has not used the newer Atellica IM assay, a threshold of 75 ng/L or greater for the ADVIA Centaur High-Sensitivity Troponin I Assay could probably also be applied for the newer Atellica assay or to other assays which demonstrate similar comparability of results. More importantly, clinicians need to know the thresholds providing prognostic information after non-cardiac surgery for the assay used by their local laboratory, and use that information according to the clinical context of each patient individually.

Other VISION Study analyses have also identified that, for some of the high-sensitivity troponin assays, the prognostically important thresholds to predict perioperative cardiovascular outcomes in non-cardiac surgery are higher than the 99th percentile proposed by the manufacturer (16). In the hsTnT VISION cohort (21 842 patients), compared with the reference group (peak hsTnT <5 ng/L), participants with peak postoperative hsTnT levels of ≥20 to <65 ng/L, 65 to <1000 ng/L, and >1000 ng/L were independently associated with 30 day mortality rates of 3.0%, 9.1%, and 29.6%, respectively (4). An absolute hsTnT change of ≥5 ng/L was associated with an increased risk of 30 day mortality (aHR, 4.69; 95% CI, 3.52–6.25). An elevated postoperative hsTnT from 20 to <65 ng/L with an absolute change ≥5 ng/L or an hsTnT ≥65 ng/L were the thresholds established for MINS with the Roche specific assay, while levels of ≥14 ng/L are proposed by the manufacturer as the 99th percentile for an overall cutoff.

In this study, we have used the same statistical approach to identify thresholds that were utilized in previous VISION cohorts. There is no gold standard analytic approach to determining the optimal threshold for a biomarker. We used a modified Mazumdar approach because it is a well-established approach and, in our experience, physicians can understand this approach. In contrast, our experience suggests physicians find other approaches (e.g., spline or fractional polynomial functions) more difficult to understand.

Limitations

Due to challenges of getting Biobank blood draws and consent before urgent/emergent surgeries, fewer patients undergoing urgent/emergent surgery participated in the Biobank than the overall study. Thus, the VISION Biobank had a lower representation of high-risk patients. Therefore, although using a high-sensitivity troponin, MINS incidence was lower in this population compared with the hsTnT VISION Study (4). The ADVIA Centaur High-Sensitivity Troponin I Assay has an area under the curve of 0.978 (95% CI, 0.937–0.996) for non-perioperative MI, and long-term storage and freeze-thaw stability (17). Hence, it is unlikely that the difference between MINS incidence is related to the assay. Due to a smaller sample size, we may not have identified lower prognostic troponin thresholds or a significant change value in our model. However, the threshold levels identified were associated with an incidence of MACE >3% and HR ≥3, pre-specified meaningful thresholds (4). Additionally, we are not aware of any other research group with a similar cohort of patients that has also measured the ADVIA Centaur High-Sensitivity Troponin I Assay in this context, and our results should be validated in other studies. Despite this study using the third definition of MI, we do not believe this has impacted the diagnosis of MI.

Conclusion

Within the first 3 days after non-cardiac surgery, hsTnI elevation independently predicts MACE at 30 days. Postoperative peak hsTnI ≥75 ng/L was associated with a >5-fold increase in the risk of MACE at 30 days as compared to peak postoperative hsTnI <75 ng/L. This threshold was prognostically relevant regardless of the presence of ischemic features and can be used to diagnose MINS with the ADVIA Centaur High-Sensitivity Troponin I Assay combined with clinical reasoning.

Supplemental Material

Supplemental material is available at Clinical Chemistry online.

Nonstandard Abbreviations

MINS, myocardial injury after non-cardiac surgery; MACE, major adverse cardiovascular events; hsTnI, high-sensitivity troponin I; VISION, Vascular Events in Non-cardiac Surgery Patients Cohort Evaluation; aHR, adjusted hazard ratios; MI, myocardial infarction; hsTnT, high-sensitivity TnT assay.

Author Contributions

The corresponding author takes full responsibility that all authors on this publication have met the following required criteria of eligibility for authorship: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Nobody who qualifies for authorship has been omitted from the list.

F. Borges (Conceptualization-Lead, Data curation-Supporting, Formal analysis-Supporting, Funding acquisition-Lead, Investigation-Lead, Methodology-Lead, Project administration-Lead, Resources-Lead, Supervision-Lead, Validation-Supporting, Writing—original draft-Lead, Writing—review & editing-Lead), E. Duceppe (Conceptualization-Lead, Formal analysis-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Lead, Project administration-Lead, Resources-Supporting, Supervision-Supporting, Writing—original draft-Supporting, Writing—review & editing-Supporting), D. Heels-Ansdell (Conceptualization-Supporting, Data curation-Lead, Formal analysis-Lead, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Lead, Project administration-Lead, Resources-Supporting, Software-Lead, Supervision-Supporting, Validation-Lead, Writing—original draft-Supporting, Writing—review & editing-Supporting), A. Patel (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Project administration-Supporting, Resources-Supporting, Writing—review & editing-Supporting), D. Sessler (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), V. Tandon (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting) M. Chan (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), R. Pearse (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), S. Srinathan (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), A. Garg (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), R. Sapsford (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), S. Ofori (Conceptualization-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Validation-Supporting, Writing—review & editing-Supporting), M. Marcucci (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), P. Kavsak (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), S. Pettit (Conceptualization-Supporting, Data curation-Lead, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Project administration-Lead, Resources-Lead, Supervision-Lead, Validation-Supporting, Writing—review & editing-Supporting), J. Spence (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Validation-Supporting, Writing—review & editing-Supporting), E. Belley-Cote (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Validation-Supporting, Writing—review & editing-Supporting), M. McGillion (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), R. Whitlock (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), A. Lamy (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting) D. Conen (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), S. Thomas (Conceptualization-Supporting, Funding acquisition-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), C. Mueller (Conceptualization-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), A. Jaffe (Conceptualization-Supporting, Investigation-Supporting, Methodology-Supporting, Resources-Supporting, Writing—review & editing-Supporting), P. Devereaux (Conceptualization-Lead, Data curation-Supporting, Formal analysis-Supporting, Funding acquisition-Lead, Investigation-Lead, Methodology-Lead, Project administration-Lead, Resources-Lead, Supervision-Lead, Writing—original draft-Supporting, Writing—review & editing-Supporting).

Authors’ Disclosures or Potential Conflicts of Interest

Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest are listed in the sections below:

Employment or Leadership

A.S. Jaffe, leadership role in the Convalescent Plasma trial.

Consultant or Advisory Role

E. Duceppe, advisory board member for Roche Diagnostics. P.A. Kavsak, consulting fees were received from Abbott Point of Care, Beckman Coulter, Quidel, Roche Diagnostics, and Siemens Healthcare Diagnostics. Payments were made to the institution. D. Conen, consulting fees were received from Roche Diagnostics. Payments were made to the institution. C. Mueller, consulting fees received from Roche and Osler & Acon. Payments were made to the institution. A.S. Jaffe, consulting fees received from Abbott Laboratories, Beckman-Coulter, Siemens, Roche Diagnostics, Radiometer, Ortho Diagnostics, ET Healthcare, Sphingotec, Astellas, RCE Technologies, and Amgen. Payments for all used biomarkers were made to the institution. A.S. Jaffe, participation in an advisory board for Conve. P.J. Devereaux, consulting fees received from Trimedic Canada. Payments made to the institution. P.J. Devereaux, member of the Quidel Corporation advisory board, November 2020, and Bayer Canada advisory board, October 2020.

Stock Ownership

None declared.

Honoraria

E. Duceppe, Roche Diagnostics. D. Conen, Servier and BMS/Pfizer. P.A. Kavsak, Beckman Coulter, Roche Diagnostics, Siemens Healthcare Diagnostics, and Thermo Fisher Scientific. C. Mueller, honoraria for speaking engagements received from several diagnostic companies, including Roche Diagnostics. Payments were made to the institution. P.J. Devereaux, speaker’s honorarium received from Roche Diagnostics for the proCardia 2020 conference.

Research Funding

F.K. Borges, Siemens Healthcare Diagnostics provided the high-sensitivity troponin I assays and some additional funding to conduct this analysis, grant and reagents from Roche Diagnostics not related to the scope of the study. All payments were made to the institution. F.K. Borges is the recipient of a Research Early Career Award from Hamilton Health Sciences Foundation. E. Duceppe, Roche Diagnostics research grant, Abbott Laboratories research grant. All grant payments were made to the institution. M. Chan, Public Policy Research Fund (grant CUHK-4002-PPR-3), Research Grant Council, Hong Kong SAR, China, General Research Fund (grant 461412), Research Grant Council, Hong Kong SAR, China. All payments were made to the Chinese University of Hong Kong. S. Srinathan, support for the present manuscript from the Health Sciences Centre Foundation, the Manitoba Medical Society Foundation, Manitoba Health Research Council, and Faculty of Dentistry, University of Manitoba. Payments were made to the Department of Surgery and the Department of Anesthesiology, University of Manitoba. R. Sapsford, National Institute of Health Research—funding received for research nurses. P.A. Kavsak, grants received from Abbott Laboratories, Beckman Coulter, Ortho Clinical Diagnostics, Randox Laboratories, Roche Diagnostics, and Siemens Healthcare Diagnostics. Payments were made to the institution. S. Pettit, Siemens Healthcare Limited provided support for the present manuscript. Payments were made to the institution. D. Conen, grant received from the Canadian Institutes of Health Research. Payments were made to the institution. C. Mueller, grants from several diagnostic companies, including Roche diagnostics. Payments were made to the institution. P.J. Devereaux, grants received from Abbott Diagnostics, Roche Diagnostics, and Siemens Canada. All payments were made to the institution. VISION Study had more than 70 grants (all grants are listed on page 3 of the supplemental material). P.J. Devereaux, recipient of monitoring devices from CloudDx and Philips Healthcare.

Expert Testimony

None declared.

Patents

McMaster University has filed patents with P.A. Kavsak listed as an inventor on “Quality Control Materials for Cardiac Troponin Testing and Identifying Pregnant Women at Increased Risk for Hypertension and Future Cardiovascular Disease.” McMaster University has filed a patent with P.A. Kavsak listed as an inventor in the acute cardiovascular biomarker field (EP 3 341 723 A1), in particular, a patent has been awarded in Europe (EP 3 341 723 B1) on a method of determining risk of an adverse cardiac event.

Other Remuneration

S. Srinathan, compensation received as a visiting professor, Society of Cardiothoracic Surgeons of the UK and Ireland. P.A. Kavsak, support for attending meetings and/or travel from Randox Laboratories. A.S. Jaffe, financial support for attending meetings/travel from Roche Diagnostics. Payments were made to the institution.

Role of Sponsor

The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, preparation of manuscript, or final approval of manuscript.

Acknowledgments

We would like to acknowledge the exceptional contributions from Dr. Yannick LeManach, PhD, Department of Anesthesia, McMaster University, Hamilton, ON, Canada (in memoriam) in the VISION study and specifically in this substudy.

References

Author notes