An epidemic of coronary heart disease (CHD) began

Transcription

1 AHA Scientific Statement Case Definitions for Acute Coronary Heart Disease in Epidemiology and Clinical Research Studies A Statement From the AHA Council on Epidemiology and Prevention; AHA Statistics Committee; World Heart Federation Council on Epidemiology and Prevention; the European Society of Cardiology Working Group on Epidemiology and Prevention; Centers for Disease Control and Prevention; and the National Heart, Lung, and Blood Institute Russell V. Luepker, MD, MS; Fred S. Apple, PhD; Robert H. Christenson, PhD; Richard S. Crow, MD; Stephen P. Fortmann, MD; David Goff, MD, PhD; Robert J. Goldberg, PhD; Mary M. Hand, MSPH, RN; Allan S. Jaffe, MD; Desmond G. Julian, MD; Daniel Levy, MD; Teri Manolio, MD; Shanthi Mendis, MD; George Mensah, MD; Andrzej Pajak, MD; Ronald J. Prineas, MD, PhD; K. Srinath Reddy, MBBS, MD, DM; Veronique L. Roger, MD; Wayne D. Rosamond, PhD; Eyal Shahar, MD; A. Richey Sharrett, MD, DrPH; Paul Sorlie, PhD; Hugh Tunstall-Pedoe, MD An epidemic of coronary heart disease (CHD) began during the 20th century in most industrialized countries, where CHD is a leading cause of mortality among adults. 1 Developing countries show the beginnings of the same epidemic. Reliable information on population incidence, prevalence, and case-fatality rates of CHD is essential to understanding, treating, and controlling the epidemic but is generally unavailable. Consistent and universal definitions of cases of CHD allow the determination of rates and comparisons within and between populations. These case definitions are essential to epidemiological studies and other research, such as clinical trials, quality assurance, and economic analysis of healthcare costs. The need for standardization is clear, and this statement recommends updated definitions. Definitions of cases for epidemiology studies and clinical trials in acute CHD rest on World Health Organization (WHO) (1959) and American Heart Association (1964) reports, followed by the WHO European AMI Registry criteria. 2,3 Myocardial infarction (MI) is based on cardiac symptoms, ECG changes, and/or elevation in biomarkers. This basic system has been widely used but variably interpreted, resulting in a lack of comparability among and within studies. Further specification and working definitions of CHD come from the Framingham Study. 4 The WHO criteria were revised in a joint report with the International Society and Federation of Cardiology in More recently, the WHO MONICA Study 6 and other surveillance and intervention studies, such as the Lipid Research Clinics in the United States, 7 have modified further the definition of CHD cases. These changes are usually based on a greater specification to the original WHO definition to allow for application in different settings. Advancing diagnostic technology, therapeutic interventions, and changing disease presentation in recent years forces a reevaluation of case definitions for acute CHD. New biomarkers, cardiac troponins, and creatine kinase (CK)-MB mass provide information that is more sensitive and/or specific in detecting even minor myocardial cell damage. 8 New imaging methods, such as MRI and radioisotope imaging, although not widely available today, will add to the diagnostic tools. These developments were recently reviewed in a Joint European Society of Cardiology/American College of Cardiology Workshop on the Redefinition of Myocardial Infarction. 9 That report, published in 2000, provided direction for clinicians faced with changing diagnostic testing and new information. Another recent report extends advice to clinical trials. 10 However, they fall short of providing direction for epidemiologists faced with evaluating and interpreting trends in event rates on the basis of retrospective surveillance. Patterns of CHD presentation are also changing. Whether because of changes in disease severity, improved diagnostic testing, increased professional awareness of the diagnosis, heightened public awareness of CHD symptoms, or insurance reimbursement to hospitals, cases of CHD in hospitalized The American Heart Association makes every effort to avoid any actual or potential conflicts of interest that may arise as a result of an outside relationship or a personal, professional, or business interest of a member of the writing panel. Specifically, all members of the writing group are required to complete and submit a Disclosure Questionnaire showing all such relationships that might be perceived as real or potential conflicts of interest. This statement was approved by the American Heart Association Science Advisory and Coordinating Committee on September 26, A single reprint is available by calling (US only) or writing the American Heart Association, Public Information, 7272 Greenville Ave, Dallas, TX Ask for reprint No To purchase additional reprints: up to 999 copies, call (US only) or fax ; 1000 or more copies, call , fax , or kgray@lww.com. To make photocopies for personal or educational use, call the Copyright Clearance Center, (Circulation. 2003;108: ) 2003 American Heart Association, Inc. Circulation is available at DOI: /01.CIR EA 2543

2 2544 Circulation November 18, 2003 TABLE 1. Classification of MI Biomarker Findings Cardiac Symptoms or Signs Present Cardiac Symptoms or Signs Absent ECG Findings Diagnostic Equivocal Missing Normal Diagnostic Equivocal Missing Normal Evolving diagnostic Definite Definite Definite Definite Definite Definite Definite Definite Positive Definite Probable Probable No Definite Probable Possible No Nonspecific Definite Possible No No Definite* Possible No No Normal or other ECG findings Definite Possible No No Definite* No No No Definite indicates definite MI; Probable, probable MI; Possible, possible MI; and No, no MI. Classification of case is at highest level allowed by combinations of 3 characteristics (cardiac signs and symptoms, ECG findings, biomarkers). *In absence of diagnostic troponin, downgrade to possible. patients are milder, and modern treatment methods reduce the extent of myocardial damage. These advances result in greater survival rates for patients hospitalized with MI and increased prevalence in the population of individuals with known disease who are susceptible to recurrence. Finally, the majority of CHD deaths occur outside of hospitals. This out-of-hospital death problem remains a diagnostic and therapeutic dilemma and a challenge for disease surveillance, because autopsy rates are relatively low. The combination of new diagnostic tests, changing disease presentation, increasing number of survivors, and the predicted incidence increases in developing countries all argue for better surveillance to establish valid rates and trends. 11 Such improvement depends on consistent, reliable, and valid case definitions. This Scientific Statement is based on a systematic review of evolving diagnostic strategies with the goal of developing standards for population studies of CHD. Clinical trials require consistent case definitions for end-point determination. They also typically collect data retrospectively, using clinic and hospital information recorded for clinical management well before the study review. Specifically, the Scientific Statement proposes case definition standards for CHD, from angina pectoris to MI to out-of-hospital death, on the basis of combinations of available data. These standards are also recommended for the collection of data in developing countries, where comprehensive information may not be available. Recommendations are made for the classification of sudden out-of-hospital death. Finally, recommendations are made for comparison of contemporary data that are based on new diagnostic methodologies with data collected in earlier periods. Such comparisons are necessary to accurately evaluate disease trends. A group of scientists from the American Heart Association Council on Epidemiology and Prevention; the American Heart Association Statistics Committee; the World Heart Federation Council on Epidemiology and Prevention; the European Society of Cardiology Working Group on Epidemiology and Prevention; the Centers for Disease Control and Prevention; the National Heart, Lung and Blood Institutes; and the WHO met at Airlie House in Warrenton, Va, in May 2002 to consider new standards and make recommendations. Support for this meeting was provided by the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md, and the Hawley Fund, Minneapolis, Minn. Definitions of Ischemic CHD The definition of a CHD case depends on symptoms, signs, biomarkers, and ECG and/or autopsy findings. These data may vary in quantity, quality, and timing. On the basis of the extent and diagnostic quality of data, definite, probable, and possible cases of fatal and nonfatal MI, procedure-related events, and angina pectoris are defined. The following outlines the definitions and recommendations, which are summarized in Table 1. The recommendations emphasize biomarkers in a setting in which signs, symptoms, and/or ECG findings suggest acute ischemia. Definition of Terms Cardiac Biomarkers Cardiac biomarkers are blood measures of myocardial necrosis, specifically CK, CK-MB, CK-MBm, or troponin (ctn). The order of diagnostic value is ctn CK-MBm CK-MB CK. A. Adequate set of biomarkers: At least 2 measurements of the same marker taken at least 6 hours apart B. Diagnostic biomarkers: At least 1 positive biomarker in an adequate set (see A above) of biomarkers showing a rising or falling pattern in the setting of clinical cardiac ischemia and the absence of noncardiac causes of biomarker elevation C. Equivocal biomarkers: Only 1 available measurement that is positive, or a rising or falling pattern not in the setting of clinical cardiac ischemia or in the presence of nonischemic causes of biomarker elevation D. Missing biomarkers: Biomarkers not measured E. Normal biomarkers: Measured biomarkers do not meet the criteria for a positive biomarker (see F below) F. Positive biomarkers: At least 1 value exceeding the 99th percentile of the distribution in healthy populations or the lowest level at which a 10% coefficient of variation can be demonstrated for that laboratory Cardiac Symptoms and Signs Cardiac symptoms and signs are findings from patient interview and examination. A. Cardiac symptoms: Presence of acute chest, epigastric, neck, jaw, or arm pain or discomfort or pressure without apparent noncardiac source. More general, atypical symptoms, such as fatigue, nausea, vomiting, diaphoresis, faintness, and back pain, should not be used as a diag-

3 Luepker et al Classification of Acute Coronary Heart Disease 2545 TABLE 2. Evolving Diagnostic ECG (Any of the Following: Q1 through Q4) Evolving Q1: No Q-code in previous study ECG or first ECG in event set of ECG(s) followed by a record with a diagnostic Q-code (Minnesota Code through plus 1.2-7) OR any code 1-3-X or in baseline ECG followed by a record with any code 1-1-X. Evolving Q2: An equivocal Q-code (Minnesota Code or any 1-3 code) and no major ST-segment depression in previous study ECG or first ECG in event set of ECG(s) followed by a record with a diagnostic Q-code PLUS a major ST-segment depression (Minnesota code 4-1-X or 4-2) and 100% increase in ST depression. Evolving Q3: An equivocal Q-code (Minnesota Code or any 1-3 code) and no major ST-segment depression in previous study ECG or first ECG in event set of ECG(s) followed by a record with a diagnostic Q-code PLUS a major T-wave inversion (Minnesota Code 5-1 or 5-2) and 100% increase in T-wave inversion. Evolving Q4: An equivocal Q-code and no ST-segment elevation in previous study ECG or first ECG in event set of ECG(s) followed by a record with a diagnostic Q-code PLUS ST-segment elevation (Minnesota code 9-2) and 100% increase in STE. Note: A significant Q-code change requires 50% increase in event Q/R ratio or 1-mm initial R-wave amplitude decrease in event ECG compared with corresponding lead(s) of baseline ECG. nostic criterion, although they are clinically useful in arriving at the correct diagnosis. B. Cardiac signs: Acute congestive heart failure or cardiogenic shock in the absence of non-chd causes. ECG Findings One or more ECG(s) may be collected in a possible cardiac event. These should be adjudicated or classified when possible. The evolution of ECG findings may be demonstrated (1) between the ECG(s) associated with the event or (2) between a previously recorded ECG and the event ECG(s). In cases in which only a single event ECG is available, an evolving diagnostic ECG pattern can be recorded only if a previous study ECG is available (eg, if there is no previous study ECG and only 1 event-related ECG, there can be no classification of evolving diagnostic ECG). Precise measurement guidelines for measurement of wave onset and offset to determine wave duration and voltage must TABLE 3. Positive ECG be followed. Most events likely to be MI occur in settings not controlled by epidemiology researchers, so that most ECG(s) will be hard copy, with varying levels of quality. The most extensively used measurement system for visual ECG findings is the Minnesota Code. 12 These measurement guidelines should be coupled with validated biologically acceptable degrees of change in ECG wave forms to code an evolution of change. These rules for proportional change are described elsewhere. 13 Another coding system that standardizes the measurement of ECG wave patterns is the Novacode (an extension of the Minnesota Code), which was designed for clinical trial ascertainment of MI. 14 More details on ECG coding are available on the Minnesota ECG Coding Center web site. 15 The categories are as follows: A. Evolving diagnostic ECG: See Table 2 B. Positive ECG: See Table 3 Includes: (a) Evolving ST elevation alone Evolving STE 1: No 9-2 in previous ECG or first ECG in event set of ECG(s) and 9-2 in at least 2 leads of a following event ECG with 100% increase in STE in both leads Evolving STE 2: 9-2 in previous ECG or first ECG in event set of ECG(s) with 100% increase in STE in at least 2 leads Evolving STE 3: 9-2 and no 5-1 or 5-2 in previous ECG in first ECG in event set of ECGs and appearance of 5-1 or 5-2 with 100% increase in T-wave inversions in at least 2 leads Evolving STE R1: Reversal of evolving STE 1 Evolving STE R2: Reversal of evolving STE 2 OR (b) Evolving equivocal Q-wave plus evolving ST-T depression/inversion Evolving Q5: No Q-code and neither 4-1-X nor 4-2 in previous study ECG or first ECG in event set of ECG(s) followed by a record with an equivocal Q-code PLUS 4-1-X or 4-2 and 100% increase in ST depression Evolving Q6: No Q-code and neither 5-1 nor 5-2 in previous study ECG or first ECG in event set of ECG(s) followed by a record with an equivocal Q-code PLUS a 5-1 or % increase in T-wave inversion Evolving Q7: No Q-code and no 9-2 in previous study ECG or first ECG in event set of ECG(s) followed by a record with an equivocal Q-code PLUS a 9-2 and 100% increase in STE OR (c) New left bundle-branch block (code 7-1-1, with the QRS duration increasing by at least 20 ms from less than 120 ms to 120 ms STE indicates ST elevation. In the case of a single ECG available only from a possible event hospital admission, a probable MI can be classified if compared with the previous study ECG, there is a new appearance of a diagnostic Q-wave ( MC through plus or any code 1-3-X in the previous ECG and the event ECG has any code 1-1-X or presence of 9-2 in at least 2 leads).

4 2546 Circulation November 18, 2003 TABLE 4. Nonspecific ECG: Evolution of Minor ST-T Depression/Inversion Alone or Minor Q-Wave Evolution Alone (a) Evolving non-ste non Q-wave pattern MI Evolving ST-T1: Either 4-0 (no 4-code), 4-4, or 4-3 in previous ECG or first ECG in event set of ECG(s) followed by a record with 4-2 or or and 100% increase in ST segment depression Evolving ST-T2: Either 4-2 or in previous ECG or first ECG in event set of ECG(s) followed by a record with and 100% increase in ST segment depression Evolving ST-T3: Either 5-0, 5-4, or 5-3 in previous ECG or first ECG in event set of ECG(s) followed by a record with 5-2 or 5-1 and 100% increase in T-wave inversion Evolving ST-T4: Code 5-2 in previous ECG or first ECG in event set of ECG(s) followed by a record with 5-1 and 100% increase in T-wave inversion Evolving ST-T5: Code in previous ECG or first ECG in event set of ECG(s) followed by a record with and 100% increase in ST depression Evolving ST-T6: Code 5-1 in previous ECG or first ECG in event set of ECG(s) followed by a record with 5-1 with 100% increase in T-wave inversion Evolving ST-T7: Code 5-2 in previous ECG or first ECG in event set of ECG(s) followed by a record with 5-2 with 100% increase in T-wave inversion Evolving ST-T R1 through ST-T R7 the reverse of ST-T1 to ST-T7, respectively OR (b) Evolving minor Q wave alone No Q code in previous study ECG or event ECG, followed by an event ECG with an equivocal Q-code (1-2-8 or any 1-3 code) C. Nonspecific ECG: See Table 4 D. ECG negative for ischemia: Normal ECGs or findings other than those described in Tables 2 through 4 Postmortem Consistent With Acute MI Postmortem findings consistent with acute MI are a cardiac pathology consistent with recent coronary occlusion or MI 28 days old. Case Classifications for CHD I. Nonfatal events A. Definite MI 1. Evolving diagnostic ECG, or 2. Diagnostic biomarkers B. Probable MI 1. Positive ECG findings plus cardiac symptoms or signs plus missing biomarkers, or 2. Positive ECG findings plus equivocal biomarkers C. Possible MI 1. Equivocal biomarkers plus nonspecific ECG findings, or 2. Equivocal biomarkers plus cardiac symptoms or signs, or 3. Missing biomarkers plus positive ECG D. Unrecognized MI 1. Appearance, in a nonacute setting, of a new diagnostic Q wave with or without ST-T wave depression, or ST elevation (Q1 through Q7 in Tables 2 and 3) E. Medical procedure-related event 1. Cardiac events after (up to 28 days) a medical procedure (eg, general surgery) with criteria for definite, probable, and possible MI identical to those described above (I.A C) 2. May be reported separately as procedure-related cardiac events or combined with overall event rates 3. If the medical procedure was performed for the treatment of acute ischemia (eg, angioplasty, coronary bypass surgery), an event should be classified as described above (I.A C) and not considered procedure-related F. Unstable angina pectoris 1. New cardiac symptoms and positive ECG findings with normal biomarkers 2. Changing symptom pattern and positive ECG findings with normal biomarkers G. Stable angina pectoris 1. Cardiac symptoms in a pattern that remains constant in presentation, frequency, character, and duration over time II. Fatal events (hospitalized patients) A. Definite fatal MI 1. Death within 28 days of hospital admission in MI cases defined in I.A 2. Postmortem findings consistent with MI within 28 days B. Probable fatal MI 1. Death within 28 days of hospital admission in cases defined in I.B 2. Death within 6 hours of hospital admission with cardiac symptoms and/or signs. Other confirmatory data (biomarkers, ECG) are absent or not diagnostic. C. Possible fatal coronary event 1. Death within 28 days of hospital admission in cases defined in I.C, I.F, and I.G 2. Postmortem findings show old infarct and/or 50% atherosclerotic narrowing of coronary arteries Ischemic CHD in Developing Countries The emerging epidemic of ischemic CHD and MI in developing countries requires increased awareness and surveillance in those areas. Definition of the scope, trends, magnitude, and outcomes associated with this epidemic is essential for prevention, diagnosis, and treatment. However, these are scarce resource settings in which diagnostic methods are costly, of varied reliability, or at times unavailable. There may be large differences between urban hospitals and rural facilities. Finally, advanced therapies for MI, such as emergency angioplasty or bypass surgery, although practiced in

5 Luepker et al Classification of Acute Coronary Heart Disease 2547 some settings, are infrequently available, making the newest diagnostic techniques less critical. Despite these limitations, it is essential to document the prevalence and trends in CHD in developing countries as the epidemic emerges and resource needs grow. The following definitions are recommended: Ischemic CHD I. Nonfatal events A. Criteria for definite, probable, and possible MI identical to those described above. This includes cardiac symptom/ sign, biomarker, and ECG findings. Although limitations of available data may result in fewer definite and probable cases, the inclusion of possible cases in the data will allow tracking of overall trends. The use of criteria similar to those in developed countries will also allow comparisons, even though diagnostic testing will be used with different frequency. Older biomarkers, such as total LDH, LDH isoenzymes, and AST (SGOT), are not recommended for criteria. They may be used, however, if more recent markers are unavailable. B. Unrecognized MI (as above) C. Procedure-related MI (as above) D. Unstable angina pectoris (as above) E. Stable angina pectoris (as above) II. Fatal events (in-hospital cases) Criteria are identical to those described above. Verbal autopsy, which involves interviews with witnesses and relatives by health workers, may be substituted for traditional postmortem. 16 III. Fatal events (out-of-hospital cases) Criteria as below. A verbal autopsy may be used as evidence except for category II.C.2. Out-of-Hospital Ischemic CHD Death Out-of-hospital CHD death is a major public health burden, accounting for 50% to 75% of all fatal cardiovascular disease events in countries in which it is documented. 17 It is often unexpected and affects all age, sex, and ethnic groups. Although the immediate mechanism of death is ventricular fibrillation or asystole, the underlying cause is commonly ischemic CHD. Other causes, including nonischemic forms of cardiac disease, are less common. Classification of out-of-hospital CHD death is usually deficient because of its sudden onset, lack of information from the victim, lack of witnesses, and low autopsy rates. These factors limit the accuracy and extent of classification. Prospective epidemiological studies can provide more preevent information and may collect relevant data from witnesses and other sources but may still face limited data on the circumstances surrounding the death. At the national level, the only source of information is the death certificate, which usually includes site of death. Much recent research focuses on arrhythmogenesis and the immediate control of malignant cardiac rhythms. Traditional emphasis on the suddenness of the out-of-hospital cardiac event has led to restricted definitions (eg, death within 1 hour of symptom onset) that may exclude most out-of-hospital events. The definitions below broadly consider out-ofhospital death as a general category, with subcategories based on site of death, presumptive cause, and timing. For data collection purposes, the following should be included: I. Site A. Before transport to medical facility (eg, home, work site, street) B. During transport (eg, ambulance, car) C. Pronounced dead in emergency department and not admitted to hospital II. Cause A. Death certification, based on WHO methods for assigning causation, is the primary source 1. Validated by postmortem examination, if available 2. Enhanced by other information, if available (See II.B and II.C below) B. Interviews with witnesses and family members C. Medical history from healthcare records and physicians III. Timing A. One hour and/or 24 hours since last seen or known to be alive B. Other time categories based on research needs Classification of Cause of Out-of-Hospital Death (Hierarchical) I. Definite fatal MI: Documented definite or probable MI in the previous 28 days and no evidence of a noncoronary cause of death, or autopsy evidence of recent coronary occlusion or MI 28 days old II. Definite fatal CHD: (1) A history of CHD and/or documented cardiac pain within 72 hours before death and (2) no evidence of a noncoronary cause of death or (3) autopsy evidence of chronic CHD, including coronary atherosclerosis and myocardial scarring III. Possible fatal CHD: An ICD code (underlying cause) for CHD death (ICD 9: 410 to 414, 427.5, 429.2, and 799; ICD 10: I20 to 25 and I46) and no evidence of a noncoronary cause of death IV. Cardiac death: When death certificates are the only source of information: ICD 9: 390 to 398, 402, 404 to 429; ICD 10: I00 to I09, I11, I13, I20 to I25, I27, I30 to I52 V. Non-CHD death: Evidence of a noncoronary cause of death VI. Unclassifiable: Insufficient information to determine whether the death was a CHD death (at any certainty level) or a noncardiac death Measuring Trends in Ischemic CHD Measurement of ischemic CHD trends in populations is important for several reasons. Among these are whether incidence and prevalence of disease are rising or falling, whether acute care is used appropriately, whether primary and secondary prevention guidelines are applied appropriately, to plan distribution of resources and the provision of medical services, and to detect the effect of changes in disease patterns and treatment. Mortality alone is less adequate for monitoring because of the lack of sensitivity to incidence and changing case fatality for CHD events.

6 2548 Circulation November 18, 2003 TABLE 5. Studies of CK-MB and Troponin Comparison Reference n Characteristics Positive CK-MB, n (%) Positive Troponin, n (%) % Troponin, Troponin/CK-MB Acute myocardial ischemia admissions 216 (27) 289 (36) 34% Possible myocardial ischemia in emergency ward 15 (5) 34 (12) 127% * MI discharge diagnosis plus biomarker, ECG, and pain algorithm 4157 (28) 4661 (32) 12% All ACS admissions 373 (22) 430 (25) 15% All ACS admissions except with diagnostic ECG of MI 23 (29) 32 (40) 39% ACS admits to cardiology service 189 (23) 228 (28) (21%) ACS indicates acute coronary syndrome. *Personal communication ( ), ARIC Executive Committee (August 27, 2003). Advances in diagnostic technology, in this case new biomarkers, along with changes in disease presentation make measuring and interpreting trends difficult. Only now is there a realization that changes in cardiac enzyme testing during the 1980s and 1990s presented obstacles to accurate trend estimation. A shift from transaminases and dehydrogenases to isoenzymes of the latter and then to CK, CK-MB, and CK-MB mass improved test precision, resulting in greater sensitivity and specificity in the diagnosis of MI over time and better case classification. Therefore, the trends in incidence and recurrence estimated during this period in areas in which surveillance existed are difficult to interpret without appropriate adjustment. The advent of even more sensitive and specific measures of myocardial cell damage, troponins, has significant implications for trend analysis (Table 5). Accumulating data suggest that the more sensitive troponin test results in greater rates of MI diagnosis than older markers. 8,18 20 Milder and smaller MI will be detected, cases that were earlier classified as unstable angina will be given a diagnosis of MI, and procedure-related troponin elevations will be labeled MI. The importance of small troponin increases is confirmed by their association with a poor prognosis on follow-up studies A diagnosis of MI or acute coronary syndrome, driven by more sensitive biomarkers, has the potential to create a spurious rise in ischemic CHD incidence if not recognized and reconciled. It could also further reduce recorded case-fatality rates erroneously as milder cases are detected and survive. It is also increasingly recognized that despite the sensitivity of troponins, they are known not to be 100% specific, and nonischemic causes of troponin elevation are documented in many settings. 28,29 Although elevated troponin levels predict poor outcomes in patients without clearly defined ischemic syndromes, 30,31 their use as a sole diagnostic criterion for MI in epidemiological studies is debated. The stated qualifiers requiring the setting of cardiac ischemia and excluding nonischemic causes should reduce false-positive results substantially, but the evidence base for a troponin-only definition is not strong. Current prospective epidemiological studies that include troponin levels along with chest pain and ECG criteria for acute MI address this issue for clinical trials and epidemiological studies. The present consensus is that the majority of troponin elevations fitting the workshop criteria do indicate MI. The use of troponin in the United States and elsewhere gradually increased from 1995 onward. However, older biomarkers are still used for MI diagnosis in many hospitals throughout the world. This gradual increase has led to a period in which a changing proportion of CHD is diagnosed by the new, more sensitive biomarkers. That proportion is largely unknown. In addition, there are several generations of troponin assays with different normal ranges and numerous manufacturers. Performance of the assays differs appreciably, and the field is gradually evolving, with improving quality of measurement. It will be some years before these tests are standardized worldwide or even within developed countries. Although the gradual spread of new diagnostic technology creates a period of transition, the need for accurate trend data remains. Administrative databases and retrospective studies are particularly vulnerable, because they must rely on available data. They are further compromised by changing insurance reimbursement rules based on diagnosis. 32 To more accurately interpret recent trends in CHD, the following are recommended: I. Comparison studies of these recommendations with previous studies and other MI criteria should be performed. This should include different observers who test, evaluate, and validate the current recommendations for observer variation, completeness, and consistency. II. Surveillance research documenting trends should use prospective methods when possible. These should include overlap methods in which old and new biomarkers are measured simultaneously to determine the effect of the new diagnostic tests. These data will allow the development of adjustment methods for trend analysis. They should be described in the literature to better understand the effects of the biomarker transition. III. Sentinel research centers should be established to monitor other indicators of acute CHD, including: A. ECG trends: 1. Q-wave infarction 2. ST-elevation and T-wave changes on admission ECG 3. Trends in infarct location B. CHD deaths out-of-hospital C. CHD deaths in-hospital D. Trends in infarct size on the basis of biomarkers or hemodynamic or clinical indicators E. Simultaneous use of older biomarkers

7 Luepker et al Classification of Acute Coronary Heart Disease 2549 IV. Administrative databases and retrospective studies seeking to determine incidence from prevalence and trends may consider adjustment factors before and after troponin-derived research studies. However, the use of adjustment factors must consider several important caveats, including: A. The variability in the use and spread of the new biomarkers in hospitals B. The evolution of the methodology and quality control for new biomarkers C. The relative paucity of overlap studies using new and old biomarkers simultaneously D. The sensitivity of available adjustment factors to selection characteristics for patient entry (Table 5) Acknowledgments Support for this meeting was provided by the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md, and the Hawley Fund, Minneapolis, Minn. The thoughtful comments of Kalevi Pyörälä, MD, PhD, Jaakko Tuomilehto, MD, PhD, and Yuling Hong, MD, PhD are appreciated, as are the administrative support of Kathy Ramel, Janelle Willard, and Jill Berkas. References 1. American Heart Association. Heart Disease and Stroke Statistics 2003 Update. Dallas, Tex: American Heart Association; World Health Organization Regional Office for Europe, Working Group on the Establishment of Ischaemic Heart Disease Registers. Report of the Fifth Working Group, Copenhagen, April Copenhagen, Denmark: Regional Office for Europe, World Health Organization; Report No. Eur 8201(5). 3. Weinstein BJ, Epstein FH; Working Subcommittee on Criteria and Methods, Committee on Epidemiological Studies, American Heart Association. Comparability of criteria and methods in the epidemiology of cardiovascular disease: report of a survey. Circulation. 1964;30: Section 8: Two-year incidence by exam interval by age and sex at exam 1. In: Kannel WB, Gordon T, eds. The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease. Bethesda, Md: US National Heart Institute; Nomenclature and criteria for diagnosis of ischemic heart disease. Report of the Joint International Society and Federation of Cardiology/World Health Organization task force on standardization of clinical nomenclature. Circulation. 1979;59: Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, et al, for the WHO MONICA Project. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project: registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation. 1994;90: Monograph: Coronary Primary Prevention Trial. Circulation. 1982; 66(suppl IV):IV-1 IV Falahati A, Sharkey SW, Christensen D, et al. Implementation of serum cardiac troponin I as marker for detection of acute myocardial infarction. Am Heart J. 1999;137: Myocardial infarction redefined a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. Eur Heart J. 2000; 21: Newby LK, Alpert JS, Ohman EM, et al. Changing the diagnosis of acute myocardial infarction: implications for practice and clinical investigations. Am Heart J. 2002;144: Yusuf S, Reddy S, Ôunpuu S, et al. Global burden of cardiovascular diseases: Part II: variations in cardiovascular disease by specific ethnic groups and geographic regions and prevention strategies. Circulation. 2001;104: Prineas RJ, Crow RS, Blackburn H. The Minnesota Code Manual of Electrocardiographic Findings: Standards and Procedures for Measurement and Classification. Boston, Mass: Wright-PSG, Crow RS, Prineas RJ, Hannan PJ, et al. Prognostic associations of Minnesota Code serial electrocardiographic change classification with coronary heart disease mortality in the Multiple Risk Factor Intervention Trial. Am J Cardiol. 1997;80: Rautaharju PM, Park LP, Chaitman BR, et al. The Novacode criteria for classification of ECG abnormalities and their clinically significant progression and regression. J Electrocardiol. 1998;31: University of Minnesota. Minnesota ECG Coding Center. Available at: Accessed October 7, Chandramohan D, Maude GH, Rodrigues LC, et al. Verbal autopsies for adult deaths: issues in their development and validation. Int J Epidemiol. 1994;23: Apple FS, Falahati A, Paulsen PR, et al. Improved detection of minor ischemic myocardial injury with measurement of serum cardiac troponin I. Clin Chem. 1997;43: Ohman EM, Armstrong PW, Christenson RH, et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. GUSTO IIA Investigators. N Engl J Med. 1996;335: Green GB, Beaudreau RW, Chan DW, et al. Use of troponin T and creatine kinase MB subunit levels for risk stratification of emergency department patients with possible myocardial ischemia. Ann Emerg Med. 1998;31: Antman EM, Grudzien C, Mitchell RN, et al. Detection of unsuspected myocardial necrosis by rapid bedside assay for cardiac troponin T. Am Heart J. 1997;133: Stubbs P, Collinson P, Moseley D, et al. Prospective study of the role of cardiac troponin T in patients admitted with unstable angina. BMJ. 1996; 313: Antman EM, Tanasijevic MJ, Thompson B, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med. 1996;335: Lindahl B, Venge P, Wallentin L. Relation between troponin T and the risk of subsequent cardiac events in unstable coronary artery disease. The FRISC study group. Circulation. 1996;93: Galvani M, Ottani F, Ferrini D, et al. Prognostic influence of elevated values of cardiac troponin I in patients with unstable angina. Circulation. 1997;95: Newby LK, Christenson RH, Ohman EM, et al. Value of serial troponin T measures for early and late risk stratification in patients with acute coronary syndromes. The GUSTO-IIa Investigators. Circulation. 1998; 98: Hamm CW, Ravkilde J, Gerhardt W, et al. The prognostic value of serum troponin T in unstable angina. N Engl J Med. 1992;327: Meier MA, Al-Badr WH, Cooper JV, et al. The new definition of myocardial infarction: diagnostic and prognostic implications in patients with acute coronary syndromes. Arch Intern Med. 2002;162: Hamm CW, Giannitsis E, Katus HA. Cardiac troponin elevations in patients without acute coronary syndrome. Circulation. 2002;106: Sobel BE, LeWinter MM. Ingenuous interpretation of elevated blood levels of macromolecular markers of myocardial injury: a recipe for confusion. J Am Coll Cardiol. 2000;35: Apple FS, Murakami MM, Pearce LA, et al. Predictive value of cardiac troponin I and T for subsequent death in end-stage renal disease. Circulation. 2002;106: Ammann P, Maggiorini M, Bertel O, et al. Troponin as a risk factor for mortality in critically ill patients without acute coronary syndromes. JAm Coll Cardiol. 2003;41: Assaf AR, Lapane KL, McKenney JL, et al. Possible influence of the prospective payment system on the assignment of discharge diagnoses for coronary heart disease. N Engl J Med. 1993;329: Lin JC, Apple FS, Murakami MM, et al. Rates of diagnosis of acute myocardial infarction utilizing cardiac troponin-i or creatine kinase MB. J Am Coll Cardiol. 2003;41:355A. 34. Ferguson JL, Beckett GJ, Stoddart M, et al. Myocardial infarction redefined: the new ACC/ESC definition, based on cardiac troponin, increases the apparent incidence of infarction. Heart. 2002;88: Apple FS, Johari V, Hoybook KJ, et al. Operationalizing cardiac troponin I testing along ESC/ACC consensus guidelines for defining myocardial infarction: increasing rate of detection. Clin Chim Acta. 2003;331: KEY WORDS: AHA Scientific Statements myocardial infarction coronary disease epidemiology diagnosis