Author + information
- Received October 19, 2017
- Revision received February 7, 2018
- Accepted February 8, 2018
- Published online May 2, 2018.
- Pierluigi Tricoci, MD, MHS, PhDa,∗ (, )
- L. Kristin Newby, MD, MHSa,
- Robert M. Clare, MSa,
- Sergio Leonardi, MD, MHSb,
- C. Michael Gibson, MD, MS, MAc,
- Robert P. Giugliano, MD, MScd,
- Paul W. Armstrong, MDe,
- Frans Van de Werf, MDf,
- Gilles Montalescot, MD, PhDg,
- David J. Moliterno, MDh,
- Claes Held, MD, PhDi,
- Philip E. Aylward, MDj,
- Lars Wallentin, MD, PhDi,
- Robert A. Harrington, MDk,
- Eugene Braunwald, MDd,
- Kenneth W. Mahaffey, MDk and
- Harvey D. White, DScl
- aDuke Clinical Research Institute, Durham, North Carolina
- bFondazione IRCCS Policlinico San Matteo, Pavia, Italy
- cBeth Israel Deaconess Medical Center, Boston, Massachusetts
- dTIMI Study Group, Cardiovascular Division, Brigham and Women’s Hospital, Boston, Massachusetts
- eDivision of Cardiology, University of Alberta, Edmonton, Alberta, Canada
- fDepartment of Cardiology, University of Leuven, Leuven, Belgium
- gSorbonne Université Paris 06, ACTION Study Group, Centre Hospitalier Universitaire Pitié-Salpêtrière (AP-HP), Paris, France
- hGill Heart Institute and Division of Cardiovascular Medicine, University of Kentucky, Lexington, Kentucky
- iDepartment of Medical Sciences, Cardiology, Uppsala University, Uppsala Clinical Research Center, Uppsala, Sweden
- jSouth Australian Health and Medical Research Institute, Flinders University and Medical Centre, Adelaide, Australia
- kDepartment of Medicine, Stanford University, Stanford, California
- lGreen Lane Cardiovascular Service, Auckland City Hospital, Auckland, New Zealand
- ↵∗Address for correspondence:
Dr. Pierluigi Tricoci, Duke Clinical Research Institute, Box 3850, 2400 Pratt Street, Durham, North Carolina 27705.
Objectives In 13,038 patients with non–ST-segment elevation acute coronary syndrome undergoing index percutaneous coronary intervention (PCI) in the EARLY ACS (Early Glycoprotein IIb/IIIa Inhibition in Non–ST-Segment Elevation Acute Coronary Syndrome) and TRACER (Thrombin Receptor Antagonist for Clinical Event Reduction in Acute Coronary Syndrome) trials, the relationship between PCI-related myocardial infarction (MI) and 1-year mortality was assessed.
Background The definition of PCI-related MI is controversial. The third universal definition of PCI-related MI requires cardiac troponin >5 times the 99th percentile of the normal reference limit from a stable or falling baseline and PCI-related clinical or angiographic complications. The definition from the Society for Cardiovascular Angiography and Interventions (SCAI) requires creatine kinase–MB elevation >10 times the upper limit of normal (or 5 times if new electrocardiographic Q waves are present). Implications of these definitions on prognosis, prevalence, and implementation are not established.
Methods In our cohort of patients undergoing PCI, PCI-related MIs were classified using the third universal type 4a MI definition and SCAI criteria. In the subgroup of patients included in the angiographic core laboratory (ACL) substudy of EARLY ACS (n = 1,401) local investigator– versus ACL-reported angiographic complications were compared.
Results Altogether, 2.0% of patients met third universal definition of PCI-related MI criteria, and 1.2% met SCAI criteria. One-year mortality was 3.3% with the third universal definition (hazard ratio: 1.96; 95% confidence interval: 1.24 to 3.10) and 5.3% with SCAI criteria (hazard ratio: 2.79; 95% confidence interval: 1.69 to 4.58; p < 0.001). Agreement between ACL and local investigators in detecting angiographic complications during PCI was overall moderate (κ = 0.53).
Conclusions The third universal definition of MI and the SCAI definition were both associated with significant risk for mortality at 1 year. Suboptimal concordance was observed between ACL and local investigators in identifying patients with PCI complications detected on angiography. (Trial to Assess the Effects of Vorapaxar [SCH 530348; MK-5348] in Preventing Heart Attack and Stroke in Participants With Acute Coronary Syndrome [TRA·CER] [Study P04736]; NCT00527943; EARLY ACS: Early Glycoprotein IIb/IIIa Inhibition in Patients With Non–ST-Segment Elevation Acute Coronary Syndrome [Study P03684AM2]; NCT00089895)
Recurrent myocardial infarction (MI) is a common complication among patients with coronary heart disease, and the risk is highest following acute coronary syndromes (ACS) (1,2). Given their frequency and prognostic implications, MIs are a key efficacy endpoint in cardiovascular clinical trials and an increasingly important safety endpoint in trials assessing noncardiovascular therapies (3). MI associated with percutaneous coronary intervention (PCI) is a clinical entity whose diagnostic criteria have been subject to an ongoing controversy (4). The lack of a common pathophysiological mechanism, the relatively high incidence of elevated markers of cardiac cell necrosis during PCI, and the uncertainty over their prognostic significance have resulted in a lack of universally accepted criteria. Several diagnostic criteria and revisions have been proposed over the years, and definitions used in clinical trials have been inconsistent (5).
The third universal definition of MI document has proposed a new definition of PCI-related MI (referred to as type 4a), which requires, in addition to increased cardiac biomarkers, the presence of clinical or electrocardiographic signs of ischemia or the documentation of procedural complications detected by coronary angiography (6). The third universal definition also raised the diagnostic cardiac troponin (cTn) threshold to >5 times the 99th percentile of a normal reference range (the upper reference limit) from a normal or decreasing pre-PCI value (6). This new definition aims to increase diagnostic specificity and prognostic relevance, while maintaining enough sensitivity in using the MI diagnosis for significant biomarker elevations in the appropriate clinical context. The definition proposed by the Society for Cardiovascular Angiography and Interventions (SCAI) relies on a >10 times increase of creatine kinase (CK)–MB (5 times in the presence of new Q waves on electrocardiography) and aims to capture “clinically relevant” MIs, without specific focus on sensitivity (7). The prognostic implications of these definitions have not been well established. Finally, as the third universal definition of PCI-related MI heavily relies on angiographic findings, it is important to define the accuracy of reporting angiographic complications by local physicians and investigators.
We used a combined dataset from 2 large non–ST-segment elevation (NSTE) ACS randomized clinical trials—TRACER (Thrombin Receptor Antagonist for Clinical Event Reduction in Acute Coronary Syndrome) and EARLY ACS (Early Glycoprotein IIb/IIIa Inhibition in Non–ST-Segment Elevation Acute Coronary Syndrome)—to assess the prognostic implication of the third universal definition of type 4a MI and SCAI definitions (2,8). We assessed the prognostic impact of the additional clinical and angiographic requirements on mortality in reference to isolated biomarker elevation only. Finally, using data from the EARLY ACS angiographic core laboratory (ACL) cohort, we measured the concordance of angiographic complications between ACL readings and site investigator reports on the case report form (CRF). The ultimate goal was to assess the applicability of the new definition in real practice and especially in clinical trials.
We used the databases of the EARLY ACS and TRACER trials, which are available at the Duke Clinical Research Institute (2,8). Briefly, the 2 trials enrolled patients with NSTE ACS during the acute phase of presentation. One-year follow-up was available in both trials. Clinical events, including MIs, were reviewed and adjudicated by a central clinical event committee (CEC) using pre-specified definitions. Local CK-MB and cTn values were collected as part of the protocols every 8 h after randomization prior to PCI; at 8, 16, and 24 h following PCI; and daily until discharge. Any additional cardiac biomarkers collected as part of standard of care from hospital presentation to discharge were also collected on the CRF.
Information on complications during PCI was collected on the CRF, including signs or symptoms of ischemia or clinical instability during PCI, as well as angiographically detected complications. Major angiographic complications were collected on the CRF, and these included major dissection, side branch closure, distal embolization, no reflow, abrupt closure, and TIMI (Thrombolysis In Myocardial Infarction) flow decrease. In the EARLY ACS trial, a subset of patients undergoing angiography (n = 2,070), including 1,401 undergoing PCI, were included in an angiographic substudy in which angiograms from the index NSTE ACS hospitalization were collected and read centrally by the ACL (PERFUSE, Boston, Massachusetts). All angiograms were assessed by trained reviewers who were blinded to treatment assignment as well as clinical and procedural data reported on the CRF. Epicardial coronary flow was assessed using the TIMI flow grade and the corrected TIMI frame count. Angiographic complications included major coronary dissection, decrease in TIMI flow grade, no reflow, abrupt closure, distal embolization, and transient or sustained side branch closure.
Third universal definition of PCI-related MI (type 4a) and the SCAI definition
The third universal definition of MI defines a PCI-related MI (type 4a) as any elevation of cTn >5 times the 99th percentile of the normal reference range occurring within 48 h of the procedure, plus: 1) evidence of prolonged ischemia (≥20 min) as demonstrated by prolonged chest pain; 2) ischemic ST-segment changes or new pathological Q waves; 3) angiographic evidence of a flow-limiting complication, such as loss of patency of a side branch, persistent slow flow or no reflow, or embolization; or 4) imaging evidence of new loss of viable myocardium or new regional wall motion abnormality. If the baseline cTn values are elevated and are stable or falling, then a rise of >20% is required for the diagnosis of a type 4a MI, as with re-infarction.
The main SCAI criteria to define PCI-related MI are an elevation of CK-MB >10 times the upper limit of normal (ULN) post-PCI in patients with normal pre-PCI biomarkers or in patients with abnormal but stable or falling peri-PCI biomarkers. If new Q waves are present after PCI, then CK-MB >5 times the ULN is required for the diagnosis.
Classification of PCI-related MI in the dataset
In the EARLY ACS and TRACER trials, PCI-related MIs were adjudicated by the CEC using a similar definition: an increase in CK-MB >3 times the local ULN post-PCI with normal values prior to PCI. The third universal definition type 4a MI definition requires a cTn elevation of >5 times the ULN with a stable or falling pattern prior to PCI. Therefore, we re-reviewed the cardiac biomarker trends of 5,785 patients (2,331 in EARLY ACS, 3,454 in TRACER) who had cTn >5 times the ULN and CK-MB <3 times the ULN, because those cases would not have been captured as MIs according to the trial definition. For each patient, values of cTn over time were plotted according to a method we have previously published (9–11) and were re-reviewed to identify patients meeting the following criteria: 1) stable or decreasing cTn levels prior to PCI; and 2) >20% relative increase of cTn from PCI at baseline to >5 times the ULN following PCI. Patients meeting these criteria were reclassified as meeting biomarker criteria for MI and were included in the analysis. This review was performed by an experienced CEC reviewer (P.T.). The goal of the review was to assess whether numeric values met the aforementioned biomarker criteria, and no subjective or clinical assessment was used. Following this review, 1,503 patients (26%) met the MI biomarker criteria of new cTn elevation >5 times the ULN post-PCI with stable or decreasing pre-PCI values.
Patients were then grouped according to: 1) the original CEC-adjudicated PCI MI definition (CK-MB >3 times the ULN); 2) third universal definition criteria (cTn >5 times the ULN with new evidence of ischemia, electrocardiographic changes, or angiographically detected complication); or 3) cTn >10 times the ULN with new evidence of ischemia, electrocardiographic changes, or angiographically detected complications. In each of those groups, mortality rates were calculated using biomarker criteria only and the complete criteria with biomarker and clinical or angiographic criteria. Finally, the SCAI definition (elevation of CK-MB >10 times the ULN post-PCI or CK-MB >5 times the ULN with new Q waves) was used as a fourth group. All patients included in the analysis had normal or stable or falling pre-PCI biomarkers.
The distribution of all categorical variables was examined by displaying the frequencies. Medians (25th and 75th percentiles) were displayed for continuous variables. We calculated the prevalence of PCI-related MI according to the pre-defined categories. We assessed the relationship between MI and subsequent risk for all-cause death at 1 year. Kaplan-Meier cumulative incidences of all-cause death at 1 year according to the PCI-related MI categories were calculated. We also calculated 1-year death rates among patients who had a spontaneous MI within 30 days of the index hospitalization and the death rates of those patients who did not have an MI within 30 days of admission, to provide clinical context to observed death rates after PCI-related MI.
Using a proportional hazards regression model, we made pairwise comparisons of unadjusted and adjusted mortality risks associated with each definition of PCI-related MI versus patients who did not have an MI. We also compared the hazard of death associated with PCI-related MI definitions versus spontaneous MI as a time-dependent covariate in unadjusted and adjusted proportional hazards models (1 comparison per model). In this analysis, PCI-related MIs were defined as “baseline” events, with time to death measured from randomization, and spontaneous MIs were treated as a time-dependent covariate. Linearity was tested, and splines were fit as appropriate. Covariates used for adjustment included age, sex, electrocardiographic ST-segment changes, multivessel coronary disease, race, weight, history of hypertension, history of diabetes mellitus, prior stroke, prior MI, peripheral arterial disease, history of heart failure, prior PCI, prior coronary artery bypass graft, smoking, systolic blood pressure, heart rate, creatinine clearance, and index MI troponin peak.
Statistical analysis was performed at the Duke Clinical Research Institute using SAS (SAS Institute, Cary, North Carolina) using original datasets.
Concordance between investigators and ACL
In the subgroup of patients from the EARLY ACS trial who were included in the angiographic substudy and underwent PCI, we compared the reporting of angiographic complications during the index PCI by site investigators with the presence of angiographic complications according to the reading provided by the ACL. A descriptive concordance table with rates of angiographic complications as reported by each of the 2 sources was created. Cohen’s kappa was calculated as a general measure of agreement between ACL and local investigators in detecting patients with angiographic complication during PCI.
The analysis included a total of 13,038 patients (5,559 from EARLY ACS and 7,479 from TRACER) with NSTE ACS who underwent PCI during the index hospitalization. Demographic and baseline characteristics by trial are provided in Online Table 1. The prevalence of PCI-related MI according to the different criteria of interest is shown in Table 1. Of patients who had a new increase of cTn >5 times the ULN post-PCI (13.7%), 2.0% (n = 258) had additional clinical and angiographic criteria reported by local investigators, thus meeting the third universal definition of type 4a MI criteria. Among patients who met the third universal MI definition, 221 (85.7%) had angiographic complications reported, of whom 178 (69.0%) had angiographic complications only and 43 (16.7%) had both angiographic complication and clinical signs of ischemia. The remaining 37 patients (14.3%) had clinical signs of ischemia only. The incidence of PCI-related MI with the SCAI criteria was 1.2%.
PCI-related MI and 1-year mortality
The mortality rate at 1 year among patients with the original PCI-related MI definition used in the 2 trials (CK-MB >3 times the ULN) was 3.7% (hazard ratio [HR] 1.66 vs. no MI; 95% confidence interval [CI]: 1.19 to 2.33; p = 0.003) (Tables 2 and 3). The mortality rate using biomarker-only criteria with cTn >5 times the ULN was 2.4% (HR: 0.90 vs. no MI; 95% CI: 0.69 to 1.17; p = 0.42). In comparison, 1-year mortality following type 4a MI according to the third universal definition was 3.3% (HR: 1.82 vs. no MI; 95% CI: 1.13 to 2.92; p = 0.014). Using the SCAI criteria, the observed mortality rate among patients with PCI-related MI was 5.3% (HR: 2.93 vs. no MI; 95% CI: 1.72 to 5.00; p < 0.001). Irrespective of the definition used, the hazard of death was nearly 4-fold higher with spontaneous MI than with PCI-related MI (Table 4). Finally, we assessed whether higher cTn thresholds could be used in place of the addition of clinical and angiographic complications. We did not observe an incremental mortality rate using a higher isolated cTn cutoff (Table 5).
Concordance between local investigators and ACL for angiographic complications
Among the 1,401 patients who had core laboratory assessment of the index PCI, angiographic complications were reported in 89 patients (6.4%) by local investigators and 184 patients (13.1%) by the core laboratory (Table 6). Of the 184 ACL-identified procedural complications on angiograms, 46 (25%) were also reported by the investigator. Of the 89 patients with procedural complications reported by the investigator, 46 (51.7%) were confirmed by the core laboratory. Overall, the agreement between ACL and local investigators in the detection of angiographic complication was moderate (κ = 0.53).
In this analysis, we assessed the association between PCI-related MI and mortality using various MI definitions, with special focus on the third universal definition of MI type 4a and the SCAI criteria (6,7). There is a lack of consensus on the definition of PCI-related MI because of its uncertain prognostic significance and absence of a common pathophysiologic causal mechanism (4). A criterion used in several randomized clinical trials has been an elevation of CK-MB >3 times the ULN (5). This arbitrary cutoff was derived from prior studies assessing the relationship between CK-MB elevation following PCI and subsequent mortality (12). The introduction of cTn as the preferred marker to assess myocardial cell necrosis after PCI has increased the capability to detect small amounts of myocardial cell necrosis, but with more sensitivity in detecting small myocardial injuries, questions have arisen on the clinical relevance of cTn elevations post-PCI. In fact, abnormal values of cTn are frequent even after an otherwise successful PCI (13–15). Despite these uncertainties, an association between cTn increase post-PCI and mortality has been shown (9,16,17), especially after properly accounting for stable or falling pre-PCI cTn values (9). The third universal definition of MI raised the cTn threshold to 5 times the 99th percentile of the assay normal reference range and required clinical symptoms or electrocardiographic changes or angiographic complications during the PCI (6). The new criteria represent an attempt to associate the cTn increase with identifiable clinical or angiographic evidence of ischemia, avoiding “calling” MI solely on the basis of isolated cTn increase. The SCAI definition is based on a different approach. First, it does not pursue diagnostic sensitivity, but rather it only aims to identify biomarker increases presumed to be “clinically relevant” (7). Second, it uses CK-MB as the biomarker of choice, rather than cTn, based on the fact that more studies were able to show the prognostic significance of CK-MB. Third, it does not incorporate angiographically detected complications in the diagnosis.
In this analysis, we show that only approximately 15% to 30% of patients with significant increases in biomarkers post-PCI present with clinical or angiographic evidence of periprocedural ischemia. These numbers support the concept that the additional requirements indicated by the third universal definition identify only those events with documentation of ischemia, thus increasing specificity in the diagnosis. Importantly, although isolated increases in cTn above the threshold of 5 times or even 10 times were not associated with mortality, events that had cTn elevations with documented complication were associated with a statistically significant 2-fold increase in the hazard of death. The American College of Cardiology/American Heart Association Task Force on Clinical Data Standards, in a joint paper with the U.S. Food and Drug Administration, has proposed use of the third universal definition of MI as a standardized endpoint definition for MI in all cardiovascular clinical trials (18). Our data validate the use of the third universal definition criteria for PCI-related MI, thus supporting their application in randomized clinical trials.
PCI-related MI defined according to the SCAI criteria was also significantly associated with 1-year mortality, with an HR of 2.9, which was higher than that of the third universal definition, although 95% CIs are largely overlapping. The higher observed rate of mortality observed with SCAI criteria is possibly due to the fact that the SCAI definition identifies larger MIs but likely at the price of losing diagnostic sensitivity (i.e., patients with biomarker elevation due to intraprocedural coronary occlusion would not be captured with the SCAI definition unless CK-MB values are very high). About 40% of patients with significant cTn increase post-PCI and clinical or angiographic complications are not identified using the SCAI criteria. In contrast, one may argue that the SCAI criteria may identify “more severe” PCI-related MI with stronger prognostic implications. Yet our data suggest that the estimated hazard of death falls within a similar range for both definitions, and there is no substantial change in the hazard relative to that of spontaneous MI, which remains 4-fold higher irrespective of the definition used. The latter finding is consistent with prior studies (19). With application of the clinical and angiographic criteria, raising the PCI-related MI cTn diagnostic threshold to 10 times the ULN has a modest effect on the associated risk for death and proportion of patients meeting PCI-related criteria, suggesting that the presence of clinical and angiographic criteria for periprocedural ischemia already identifies patients who are at higher risk. Furthermore, using a high threshold of isolated (i.e., without corresponding evidence of ischemia or abnormal coronary flow) cTn increase does not appear to be associated with further increase of 1-year mortality.
A major issue in the application of the third universal definition is the systematic ascertainment of angiographic criteria. In this analysis, we have directly compared the report of PCI complications documented on the CRF according to the local investigators with a rigorous and systematic assessment performed by a central ACL according to established angiographic definitions aimed at minimizing interobserver variability. We show only a moderate agreement between local and ACL investigators in the report of angiographic complications. In particular, only 25% of the procedural complications detected by the core laboratory were reported on the CRF by the local investigators. Our results are consistent with other studies demonstrating moderate agreement between ACL and local investigators in quantitative and semiquantitative angiographic measures, such as TIMI flow, and even lower agreement for qualitative measures, such as the presence or absence of coronary thrombosis (20–22). This analysis, to our knowledge, is the first investigation of the concordance in angiographic measures used in the application of the PCI-related MI definition. We are unable to investigate further as to what led to suboptimal agreement between local and ACL investigators and whether CRF-collected data accurately reflected the PCI reports. However, in a recent report published by our group where low agreement between an ACL and CEC was observed, we found that central event adjudication was consistent with information provided in the cardiac catheterization reports, thus suggesting that those documents may have low sensitivity in the reporting of angiographic measures (22).
Our results overall highlight that a rigorous and systematic application of the PCI-related MI criteria on the basis of the presence of angiographically detected complications likely requires the use of a central ACL. However, the implementation of core laboratory review on a large scale is costly and presents operational challenges. Nonetheless, a definition based on higher cTn thresholds used in place of evidence of coronary ischemia not only may result in lower sensitivity in the detection of PCI-related MI, but our data suggest that this may result in capturing events that are less prognostically relevant than those with lower cTn thresholds but evidence of coronary ischemia.
First, it was a retrospective study in which MI definitions were derived using existing data. We believe that the datasets allowed an adequate application of the MI definitions studied; however, because of the retrospective nature, some simplification to the original definitions were used. For example, the trials did not routinely collect imaging to assess for new loss of viable myocardium, which is 1 of the criteria for the third universal definition. However, because it is not standard of care to routinely perform pre- and post-PCI myocardial imaging, we do not think this limitation affects the applicability of our results (6). The third universal definition is based on elevation of cTn times the 99th percentile of the upper reference limit, while we used the ULN provided by participating centers. For the SCAI definition, we did not use the criteria allowing the definition in presence of not stable or falling values (ST-segment elevation or depression, new worsening heart failure or hypotension with CK-MB >10 times the ULN), because our study focused on patients with stable or falling pre-PCI biomarker values in order to be able to apply the definitions of interest (7). We also believe that it is not possible to determine a new PCI-related MI unless pre-PCI biomarkers are stable or falling. When the trials were conducted, the use of high-sensitivity cTn was low; therefore, generalization of our results to high-sensitivity cTn must take this into consideration. With these limitations in mind, we believe that our analysis captures the key element of the PCI-related MI definitions and allows an adequate assessment of their prognostic relevance.
PCI-related MI defined according to the third universal definition of MI type 4a criteria is associated with a 2-fold increase in the risk for 1-year death. The inclusion of additional clinical and angiographic criteria significantly improves the strength in the association with death compared with isolated biomarker increase. The SCAI definition is associated with a higher mortality rate, at the price of reduced diagnostic sensitivity. There is suboptimal agreement between ACL and local investigators in detecting angiographic complications, with most core laboratory–identified complications not reported by the local investigators, which suggests ACL assessment may be required to ensure a systematic application of the third universal definition of MI type 4a criteria.
WHAT IS KNOWN? PCI-related MI is a frequently reported endpoint in randomized clinical trials, but the appropriate definition is controversial, as the prognostic implications of an increase in myocardial necrosis markers with PCI are unclear. The third universal definition of PCI-related MI is based on the presence of objective evidence of peri-procedural ischemia, such as ischemic symptoms or electrocardiographic changes or the presence of an angiographically documented procedural complication in addition to troponin elevation. The SCAI definition is based on a very high threshold of the less sensitive biomarker CK-MB. Those definitions have not been validated.
WHAT IS NEW? In our analysis, we show that the third universal definition detects more PCI-related MIs than the SCAI definition. PCI-related MIs defined with both the third universal and SCAI definitions are associated with a significant increase in the risk of 1-year mortality, with the patients with SCAI-defined MIs having a higher risk for death. The presence of ischemic or angiographic complications significantly adds to the prognostic relevance of the MI in addition to troponin elevation only. One of the key aspects of implementing the third universal definition is accurate reporting of PCI-related complications. In our study, we show only a moderate agreement between site investigators and an ACL in the detection of angiographically detected complications.
WHAT IS NEXT? Our data validate the prognostic value of the third universal definition and the SCAI definition in the detection of PCI-related MI, thus supporting their use in practice and in randomized clinical trials, with the third universal definition appearing more sensitive, but with a lower event rate, than the SCAI definition. More studies are needed to understand the clinical and prognostic implications of the underreporting of angiographic complications without the use of an ACL.
The TRACER trial was supported by Merck. The EARLY ACS trial was supported by Schering-Plough and Millennium Pharmaceuticals. Dr. Tricoci has a consultant agreement and research grant from Merck and CSL and a research grant from Sanofi. Dr. Newby has received research support from Bristol-Myers Squibb, GlaxoSmithKline, Google Life Sciences (Verily), NHLBI, MURDOCK Study, and NIH; and consulting or other services (including CME) for American Heart Association, AstraZeneca HCF, CAMC Health Education & Research Institute, Center for Human Genetics, North Carolina State University, Medscape, LLC/TheHeart.org, Metanomics, NIH, Philips, Roche Diagnostic Corp., Society of CV Patient Care, and University of Alberta. Dr. Gibson reports serving as the Chief Executive Officer at the Baim Institute for Clinical Research; present research/grant funding from Angel Medical Corporation, Bayer Corp, CSL Behring, Janssen Pharmaceuticals, Johnson & Johnson Corporation, Portola Pharmaceuticals; consulting and peer-to-peer communication for Amarin Pharama, Amgen, Arena Pharmaceuticals, Bayer Corporation, Boehringer Ingelheim, Boston Clinical Research Institute, Cardiovascular Research Foundation, Chiesi, CSL Behring, Eli Lilly, Gilead Sciences, Inc., Janssen Pharmaceuticals, Johnson & Johnson Corporation, The Medicines Company, Merck, Novo Nordisk, Pfizer, Pharma Mar, Portola Pharmaceuticals, Sanofi, Somahlution, St. Francis Hospital, Vereson Corportation, WebMD; and royalties as a contributor for UpToDate in Cardiovascular Medicine. Dr. Giugliano has received honoraria from Merck-Schering Plough and from Daiichi Sankyo as a consultant and for continuing medical education lectures; research grant funding for EARLY ACS through the TIMI Study Group from Merck-Schering Plough, and from Daiichi Sankyo for a trial with a novel anticoagulant. Dr. Armstrong has received research support from Merck, Bayer, Sanofi, Recherche & Développement, and CSL Limited; consulting or other services for AstraZeneca, Merck, Bayer, Novartis, Mast Therapeutics, Cardiome Pharma Corp., and CSL Limited. Dr. Van de Werf has received a research grant and honoraria for lectures and advisory board membership from Merck. Dr. Montalescot has received research funds for his institution or fees from Abbott, Amgen, Actelion, AstraZeneca, Bayer, Boehringer Ingelheim, Boston Scientific, Bristol-Myers Squibb, Beth Israel Deaconess Medical, Brigham Women's Hospital, Cardiovascular Research Foundation, Daiichi Sankyo, Idorsia, Lilly, Europa, Elsevier, Fédération Française de Cardiologie, ICAN, Medtronic, Journal of the American College of Cardiology, Lead-Up, Menarini, Merck Sharpe & Dohme, Novo-Nordisk, Pfizer, Sanofi, Servier, The Mount Sinai School, TIMI Study Group, and WebMD. Dr. Held has received institutional research grants from AstraZeneca, GlaxoSmithKline, Pfizer/Bristol-Myers Squibb, Roche, and Schering-Plough (now Merck); and is a consultant for AstraZeneca. Dr. Aylward has received research grants from Merck, AstraZeneca, Sanofi, and GlaxoSmithKline; and has received honoraria for Speakers Bureau and advisory board membership from AstraZeneca, Eli Lilly, Boehringer Ingelheim, Bayer Johnson & Johnson, Servier, and Bristol-Myers Squibb. Dr. Wallentin has received research grants from AstraZeneca, Merck, Boehringer Ingelheim, Bristol-Myers Squibb/Pfizer, and GlaxoSmithKline; Speakers Bureau and lecture fees from AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb/Pfizer, GlaxoSmithKline, and Merck; honoraria from Boehringer Ingelheim, AstraZeneca, Bristol-Myers Squibb/Pfizer, GlaxoSmithKline, and Merck; is a consultant or advisory board member for Merck, Regado Biosciences, Evolva, Portola, CSL Behring, Athera Biotechnologies, Boehringer Ingelheim, AstraZeneca, GlaxoSmithKline, and Bristol-Myers Squibb/Pfizer; and has received travel support from Bristol-Myers Squibb/Pfizer. Dr. Harrington has received consulting fees and honoraria from Adverse Events, Amgen, Daiichi-Lilly, Gilead Sciences, Janssen Research & Development, Medtronic, Merck, Novartis Corporation, The Medicines Company, Vida Health, Vox Media, and WebMD; has received research grants from AstraZeneca, Bristol-Myers Squibb, CSL Behring, GlaxoSmithKline, Merck, Portola, Sanofi, and The Medicines Company; has ownership interest and is a partner or principal at Element Science and MyoKardia; is an officer, a director, a trustee, or holds another fiduciary role with Evidint and Scanadu; is a member of the Data Safety Monitoring Board for Regado; and has is an officer, director, trustee, or holds another fiduciary role with the American Heart Association. Dr. Braunwald has received consulting fees, lecture fees, and grant support from Eli Lilly, Johnson & Johnson, Merck, and Bayer. Dr. Mahaffey reports research support from Afferent, Amgen, Apple, Inc., AstraZeneca, Cardiva Medical, Inc., Daiichi Sankyo, Ferring, Google (Verily), Johnson & Johnson, Luitpold, Medtronic, Merck, Novartis, Sanofi, St. Jude Medical, Tenax; consulting or other services (including CME) forAblynx, AstraZeneca, Baim Institute, Boehringer Ingelheim, Bristol-Myers Squibb, Cardiometabolic Health Congress, Elsevier, GlaxoSmithKline, Johnson & Johnson, Medergy, Medscape, Merck, Mitsubishi, Myokardia, Novartis, Oculeve, Portola, Radiometer, Springer Publishing, Theravance, UCSF, WebMD; and has equity in BioPrint Fitness. Dr. White has received research grants from Sanofi, Eli Lilly, the National Institutes of Health, Merck Sharpe & Dohme, AstraZeneca, GlaxoSmithKline, and Daiichi Sankyo Pharma Development; and has received consulting fees from AstraZeneca. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- angiographic core laboratory
- acute coronary syndrome(s)
- clinical event committee
- creatine kinase
- case report form
- cardiac troponin
- myocardial infarction
- non–ST-segment elevation
- percutaneous coronary intervention
- Society for Cardiovascular Angiography and Interventions
- upper limit of normal
- Received October 19, 2017.
- Revision received February 7, 2018.
- Accepted February 8, 2018.
- 2018 American College of Cardiology Foundation
- Tricoci P.
- Thygesen K.,
- Alpert J.S.,
- Jaffe A.S.,
- Simoons M.L.,
- Chaitman B.R.,
- White H.D.
- Moussa I.D.,
- Klein L.W.,
- Shah B.,
- et al.
- Tricoci P.,
- Leonardi S.,
- White J.,
- et al.
- Leonardi S.,
- Truffa A.A.,
- Neely M.L.,
- et al.
- Mahaffey K.W.,
- Held C.,
- Wojdyla D.M.,
- et al.
- Prasad A.,
- Gersh B.J.,
- Bertrand M.E.,
- et al.
- Cantor W.J.,
- Newby L.K.,
- Christenson R.H.,
- et al.
- Hicks K.A.,
- Tcheng J.E.,
- Bozkurt B.,
- et al.
- Leonardi S.,
- Thomas L.,
- Neely M.L.,
- et al.
- Popma J.J.,
- Lansky A.J.,
- Yeh W.,
- et al.
- Popma C.J.,
- Sheng S.,
- Korjian S.,
- et al.