https://orcid.org/0000-0003-4800-2972
Abstract
Limited US outcome data exist among patients with myocardial injury and types 1 and 2 myocardial infarction (MI) evaluated with high-sensitivity cardiac troponin (hs-cTn).
This is an observational US cohort study of emergency department (ED) patients undergoing hs-cTnT measurement. Cases with ≥1 hs-cTnT increase >99th percentile were adjudicated following the Fourth Universal Definition of MI. Post-discharge major adverse cardiovascular events (MACE) included death, MI, heart failure (HF) hospitalization, stroke or transient ischaemic attack, and new-onset atrial fibrillation or flutter during 2 years follow-up. Among 2002 patients, 857 (43%) had ≥1 hs-cTnT >99th percentile. Among these, 702 (81.9%) had myocardial injury, 64 (7.5%) had type 1 MI, and 91 (10.6%) had type 2 MI. Compared with patients without myocardial injury, type 2 MI [8.4 vs. 50%; adjusted hazard ratio (HR) 2.31, 95% confidence interval (CI) 1.49–3.58] and myocardial injury (8.4 vs. 47%; adjusted HR 3.13, 95% CI 2.39–4.09) had a higher risk of MACE, in large part because of death and HF hospitalizations. Compared with patients with type 1 MI, type 2 MI (23 vs. 50%; adjusted HR 2.24; 95% CI 1.23–4.10) and myocardial injury (23 vs. 47%; adjusted HR 2.02; 95% CI 1.20–3.40) also have a higher risk of MACE.
Among unselected US ED patients undergoing hs-cTnT measurement, most increases are due to myocardial injury, and type 2 MI is more frequent than type 1 MI. Patients with myocardial injury and type 2 MI have morbid outcomes, in large part due to death and HF.
Long-term outcomes among patients with myocardial injury, types 1 and 2 myocardial infarction. The frequency of myocardial injury, types 1 and 2 myocardial infarction among patients with increased hs-cTnT above the 99th percentile and corresponding adjusted 2-year risk for MACE, death, and heart failure hospitalizations. HF, heart failure; MACE, major adverse cardiovascular events; MI, myocardial infarction; URL, upper reference limit.
Introduction
With the adoption of high-sensitivity cTn (hs-cTn) assays, there are often significant increases in the proportion of patients with cTn concentrations above the 99th percentile upper reference limit (URL).1–3 It is often difficult to determine whether these increases are isolated cTn increases due to acute or chronic non-ischaemic myocardial injury or due to either atherothrombotic type 1 myocardial infarction (MI) or non-atherothrombotic type 2 acute MI due to supply–demand imbalance.4–7 Distinguishing between these groups is important given these conditions have distinct clinical phenotypes and mechanisms, as well as different therapeutic and prognostic implications.5–7
Most studies indicate that hs-cTn assays are associated with increases in the frequency of non-ischaemic myocardial injury and type 2 MI.1–3 Multiple studies8–11 suggest that these patients have similar or worse outcomes than patients with type 1 MI. We previously have reported that with the transition from fourth to fifth Gen cTnT, there was a shift in the distribution of diagnoses,1 with more non-ischaemic myocardial injury and type 2 MI diagnoses than type 1 MI. However, there is a paucity of outcomes data from US centres.
The present analysis is a comprehensive evaluation of the real-life US experience using hs-cTnT of long-term clinical outcomes following systematic adjudication of cases with hs-cTnT above sex-specific 99th percentiles following the Fourth Universal Definition of Myocardial Infarction (UDMI).4 Our goals were two-fold: (i) to describe and compare the frequency, clinical features, and management of patients with non-ischaemic myocardial injury and type 1 and 2 MIs, and (ii) to examine short- and long-term post-discharge major adverse cardiovascular outcomes.
Methods
Study design and patient population
The MAyo Southwest WisConsin 5th Gen Troponin T ImplementatiON (ACTION) study is a retrospective, multicentre (n = 2), IRB approved, observational cohort study of adult patients presenting to the emergency department (ED) in whom at least one cTnT measurement was obtained for clinical purposes from the Southwest Wisconsin Mayo Clinic Health System hospitals at La Crosse and Sparta in Wisconsin, USA. The primary results and methods have been described.1 The present analyses address only unique patients based on their first presentation during the hs-cTnT study period from September 12th, 2018 through March 11th, 2019. Patients who did not present through the ED, those aged <18 years old, and those without 12-lead electrocardiograms, hs-cTnT measurements, or in whom hs-cTnT was not measured within 12 h of ED presentation, or those classified as having types 3–5 MI were excluded.
High-sensitivity cardiac troponin measurements
The hs-cTnT was measured using the Elecsys Troponin T Gen 5 STAT assay (Roche Diagnostics) on the Cobas e 601 (MCHS La Crosse) and Cobas e 411 (MCHS Sparta). Per US Food and Drug Administration (FDA) guidance,12 concentrations were reported down to the limit of quantitation of <6 ng/L. Sex-specific 99th percentile URLs of 10 ng/L for women and 15 ng/L for men were used.12,13 Concentrations >10 ng/L for women and >15 ng/L for men are considered indicative of myocardial injury. Results are reported as whole units (no decimals) in ng/L. The Mayo Clinic protocol and rationale for the rule-in and rule-out of MI using 5th Gen cTnT has been described.12,13 In brief, patients with suspected MI are evaluated using a 0 and 2 h hs-cTnT protocol that uses sex-specific 99th percentile URLs and an absolute delta (serial change) of >10 ng/L to identify patients with acute myocardial injury. Those with 0 and 2 h deltas of ≤3 ng/L are classified as having no significant change, and those with deltas of 4–9 ng/L as indeterminate. For the indeterminate group, a 6 h sample is automatically ordered by the laboratory and an empirical change ≥12 ng/L over 6 h is considered indicative of acute myocardial injury.
Myocardial injury and infarction adjudication
All encounters with at least 1 cTnT >99th percentile URL were adjudicated using the Fourth UDMI criteria4 by trained physicians (Supplemental Material). Cases with challenging adjudication were reviewed by the principal investigator (Y.S.), and if needed, by a member of the Task Force for the Fourth UDMI (A.S.J.). Patients with at least 1 cTnT concentration >99th percentile URL were classified as having either myocardial injury (i.e. isolated cTnT increases) or acute MI if there were clinical features of acute myocardial ischaemia, such as ischaemic symptoms, new ischaemic electrocardiogram changes, development of pathological Q waves, imaging evidence of new loss of viable myocardium or new regional wall motion abnormality in a pattern consistent with an ischaemic aetiology, and/or identification of a culprit lesion on coronary angiography. Those with clinical evidence of myocardial ischaemia were classified as having MI and further subclassified into types 1 or 2 MI subtypes. Type 1 MI is an atherothrombotic MI and type 2 MI is due to non-atherothrombotic supply–demand myocardial ischaemia.4,5 The term ‘myocardial injury’ is used to refer to patients adjudicated to have cTnT increases without overt clinical evidence of acute myocardial ischaemia.
Study endpoints
The primary endpoint was post-discharge major adverse cardiovascular events (MACE), which was a composite of all-cause mortality, acute MI, heart failure hospitalization, stroke or transient ischaemic attack (TIA), and new-onset atrial fibrillation or flutter. Post-discharge deaths were further classified based on chart review and available records as cardiovascular, non-cardiovascular, or unknown. Cardiovascular deaths were defined as those attributable to acute MI, sudden cardiac death, heart failure, stroke, cardiovascular procedures, cardiovascular haemorrhages, or other cardiovascular causes.14
Statistical analysis
Baseline characteristics based on the first ED presentation for unique patients are presented using means and standard deviation or medians and interquartile ranges and compared using Kruskal–Wallis test for continuous variables. Categorical variables are presented as numbers (percentages) and compared using the χ2 test. Follow-up endpoints were estimated using the Kaplan–Meier method. Potential risk factors for those endpoints were evaluated using Cox Proportional Hazard models, and hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) are presented. For the primary endpoint of post-discharge MACE, multivariable models were developed which included age, sex, heart failure, chronic kidney disease, and ischaemic heart disease. For each individual outcome of the MACE composite, similar multivariable models were used, but when limited events occurred during follow-up, multivariable models were adjusted to include age and sex. A P-value <0.05 was considered statistically significant. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA) and R version 4.0.2 (R Foundation, Vienna, Austria).
Results
The final study cohort included 2002 patients, among which 857 (43%) had ≥1 hs-cTnT above the sex-specific 99th percentile URL. Of these, 702 (81.9%) were adjudicated as myocardial injury, 64 (7.5%) as type 1 MI, and 91 (10.6%) as type 2 MI.
Baseline characteristics
Baseline characteristics are shown in Table 1. The mean age of the cohort was 62 years, 52% were women, and 50% had chest discomfort. There were no differences in the proportion of women among groups. Patients without myocardial injury were younger and less likely to have comorbidities. Patients with type 1 MI were more likely to have chest discomfort, ST elevation, and higher baseline and maximum hs-cTnT concentrations than patients with type 2 MI and myocardial injury. Patients with type 2 MI were older and more likely to have a history of atrial dysrhythmias, heart failure, and worse renal function than patients with type 1 MI. Similarly, besides being more likely to have dyspnoea, patients with myocardial injury also had more history of atrial dysrhythmias and heart failure than patients with type 1 MI. Patients with type 2 MI and myocardial injury were mostly similar with respect to demographics and comorbidities; however, those with type 2 MI had higher baseline and maximum hs-cTnT concentrations and were more likely to have ST-depression and/or T-wave inversion.
Baseline characteristics
| Overall N = 2002 | Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|---|
| Demographics | |||||||||
| Age, mean (SD) | 62 (18) | 53 (17) | 74 (14) | 71 (13) | 76 (15) | <0.0001 | 0.03 | 0.12 | 0.06 |
| Women, n (%) | 1036 (52) | 604 (53) | 356 (51) | 27 (42) | 49 (54) | 0.35 | 0.15 | 0.19 | 0.57 |
| Presenting symptoms | |||||||||
| Chest discomfort, n (%) | 1002 (50) | 703 (61) | 212 (30) | 56 (88) | 31 (34) | <0.0001 | <0.0001 | <0.0001 | 0.45 |
| Dyspnoea, n (%) | 832 (42) | 436 (38) | 311 (44) | 39 (61) | 46 (51) | 0.0001 | 0.20 | 0.01 | 0.26 |
| Medical history | |||||||||
| Hypertension, n (%) | 1157 (58) | 502 (44) | 542 (77) | 42 (66) | 69 (76) | <0.0001 | 0.17 | 0.04 | 0.77 |
| Obesity, n (%) | 855 (43) | 483 (42) | 320 (46) | 25 (39) | 27 (30) | 0.03 | 0.22 | 0.32 | 0.004 |
| Current or prior tobacco use, n (%) | 1120 (56) | 633 (55) | 404 (58) | 36 (56) | 47 (52) | 0.66 | 0.57 | 0.84 | 0.29 |
| Cerebrovascular disease, n (%) | 200 (10) | 62 (5.4) | 115 (16) | 7 (11) | 16 (18) | <0.0001 | 0.25 | 0.26 | 0.77 |
| History of atrial dysrhythmias, n (%) | 376 (19) | 112 (9.8) | 217 (31) | 7 (11) | 40 (44) | <0.0001 | <0.0001 | 0.0008 | 0.01 |
| Heart failure, n (%) | 379 (19) | 73 (6.4) | 254 (36) | 15 (23) | 37 (41) | <0.0001 | 0.03 | 0.04 | 0.40 |
| Diabetes mellitus, n (%) | 485 (24) | 166 (15) | 264 (38) | 23 (36) | 32 (35) | <0.0001 | 0.92 | 0.79 | 0.65 |
| Chronic kidney disease, n (%) | 403 (20) | 70 (6) | 281 (40) | 19 (30) | 33 (36) | <0.0001 | 0.39 | 0.10 | 0.49 |
| Family history of CAD, n (%) | 622 (31) | 320 (28) | 250 (36) | 25 (39) | 27 (30) | 0.0029 | 0.22 | 0.58 | 0.26 |
| Peripheral artery disease, n (%) | 208 (10) | 38 (3.3) | 129 (18) | 13 (20) | 28 (31) | <0.0001 | 0.15 | 0.70 | 0.0052 |
| Dyslipidaemia, n (%) | 1035 (52) | 468 (41) | 463 (66) | 43 (67) | 61 (67) | <0.0001 | 0.98 | 0.84 | 0.84 |
| Ischaemic heart disease, n (%) | 446 (22) | 151 (13) | 242 (35) | 26 (41) | 27 (30) | <0.0001 | 0.16 | 0.32 | 0.36 |
| Laboratory results | |||||||||
| Baseline hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 21) | <6 (<6, 8.0) | 23 (17, 38) | 89 (38, 220) | 35 (19, 73) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Maximum hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 23) | <6 (<6, 9.0) | 24 (18, 43) | 216 (95, 898) | 44 (23, 111) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Creatinine (mg/dL), median (IQR) (n = 1985) | 0.9 (0.8, 1.1) | 0.8 (0.7, 1.0) | 1.1 (0.8, 1.4) | 1.0 (0.8, 1.4) | 1.1 (0.9, 1.5) | <0.0001 | 0.20 | 0.54 | 0.27 |
| eGFR (mL/min/1.73 m2), median (IQR) (n = 1984) | 80 (59, 91) | 91 (77, 91) | 60 (41, 79) | 67 (47, 83) | 55 (37, 74) | <0.0001 | 0.01 | 0.13 | 0.06 |
| NT-proBNP (pg/mL), median (IQR) (n = 718) | 409 (94, 2147) | 95 (44, 294) | 1181 (270, 3268) | 1794 (363, 4608) | 2211 (727, 6537) | <0.0001 | 0.20 | 0.44 | 0.003 |
| Initial 12-lead electrocardiogram | |||||||||
| Sinus rhythm, n (%) | 1679 (84) | 1067 (93) | 508 (72) | 59 (92) | 45 (50) | <0.0001 | <0.0001 | 0.0005 | <0.0001 |
| ST elevation, n (%) | 23 (1.1) | 1 (0.1) | 3 (0.4) | 19 (30) | 0 (0.0) | <0.0001 | <0.0001 | <0.0001 | 0.53 |
| ST-depression/T-wave inversion, n (%) | 130 (6.5) | 39 (3.4) | 45 (6.4) | 17 (27) | 29 (32) | <0.0001 | 0.48 | <0.0001 | <0.0001 |
| Overall N = 2002 | Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|---|
| Demographics | |||||||||
| Age, mean (SD) | 62 (18) | 53 (17) | 74 (14) | 71 (13) | 76 (15) | <0.0001 | 0.03 | 0.12 | 0.06 |
| Women, n (%) | 1036 (52) | 604 (53) | 356 (51) | 27 (42) | 49 (54) | 0.35 | 0.15 | 0.19 | 0.57 |
| Presenting symptoms | |||||||||
| Chest discomfort, n (%) | 1002 (50) | 703 (61) | 212 (30) | 56 (88) | 31 (34) | <0.0001 | <0.0001 | <0.0001 | 0.45 |
| Dyspnoea, n (%) | 832 (42) | 436 (38) | 311 (44) | 39 (61) | 46 (51) | 0.0001 | 0.20 | 0.01 | 0.26 |
| Medical history | |||||||||
| Hypertension, n (%) | 1157 (58) | 502 (44) | 542 (77) | 42 (66) | 69 (76) | <0.0001 | 0.17 | 0.04 | 0.77 |
| Obesity, n (%) | 855 (43) | 483 (42) | 320 (46) | 25 (39) | 27 (30) | 0.03 | 0.22 | 0.32 | 0.004 |
| Current or prior tobacco use, n (%) | 1120 (56) | 633 (55) | 404 (58) | 36 (56) | 47 (52) | 0.66 | 0.57 | 0.84 | 0.29 |
| Cerebrovascular disease, n (%) | 200 (10) | 62 (5.4) | 115 (16) | 7 (11) | 16 (18) | <0.0001 | 0.25 | 0.26 | 0.77 |
| History of atrial dysrhythmias, n (%) | 376 (19) | 112 (9.8) | 217 (31) | 7 (11) | 40 (44) | <0.0001 | <0.0001 | 0.0008 | 0.01 |
| Heart failure, n (%) | 379 (19) | 73 (6.4) | 254 (36) | 15 (23) | 37 (41) | <0.0001 | 0.03 | 0.04 | 0.40 |
| Diabetes mellitus, n (%) | 485 (24) | 166 (15) | 264 (38) | 23 (36) | 32 (35) | <0.0001 | 0.92 | 0.79 | 0.65 |
| Chronic kidney disease, n (%) | 403 (20) | 70 (6) | 281 (40) | 19 (30) | 33 (36) | <0.0001 | 0.39 | 0.10 | 0.49 |
| Family history of CAD, n (%) | 622 (31) | 320 (28) | 250 (36) | 25 (39) | 27 (30) | 0.0029 | 0.22 | 0.58 | 0.26 |
| Peripheral artery disease, n (%) | 208 (10) | 38 (3.3) | 129 (18) | 13 (20) | 28 (31) | <0.0001 | 0.15 | 0.70 | 0.0052 |
| Dyslipidaemia, n (%) | 1035 (52) | 468 (41) | 463 (66) | 43 (67) | 61 (67) | <0.0001 | 0.98 | 0.84 | 0.84 |
| Ischaemic heart disease, n (%) | 446 (22) | 151 (13) | 242 (35) | 26 (41) | 27 (30) | <0.0001 | 0.16 | 0.32 | 0.36 |
| Laboratory results | |||||||||
| Baseline hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 21) | <6 (<6, 8.0) | 23 (17, 38) | 89 (38, 220) | 35 (19, 73) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Maximum hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 23) | <6 (<6, 9.0) | 24 (18, 43) | 216 (95, 898) | 44 (23, 111) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Creatinine (mg/dL), median (IQR) (n = 1985) | 0.9 (0.8, 1.1) | 0.8 (0.7, 1.0) | 1.1 (0.8, 1.4) | 1.0 (0.8, 1.4) | 1.1 (0.9, 1.5) | <0.0001 | 0.20 | 0.54 | 0.27 |
| eGFR (mL/min/1.73 m2), median (IQR) (n = 1984) | 80 (59, 91) | 91 (77, 91) | 60 (41, 79) | 67 (47, 83) | 55 (37, 74) | <0.0001 | 0.01 | 0.13 | 0.06 |
| NT-proBNP (pg/mL), median (IQR) (n = 718) | 409 (94, 2147) | 95 (44, 294) | 1181 (270, 3268) | 1794 (363, 4608) | 2211 (727, 6537) | <0.0001 | 0.20 | 0.44 | 0.003 |
| Initial 12-lead electrocardiogram | |||||||||
| Sinus rhythm, n (%) | 1679 (84) | 1067 (93) | 508 (72) | 59 (92) | 45 (50) | <0.0001 | <0.0001 | 0.0005 | <0.0001 |
| ST elevation, n (%) | 23 (1.1) | 1 (0.1) | 3 (0.4) | 19 (30) | 0 (0.0) | <0.0001 | <0.0001 | <0.0001 | 0.53 |
| ST-depression/T-wave inversion, n (%) | 130 (6.5) | 39 (3.4) | 45 (6.4) | 17 (27) | 29 (32) | <0.0001 | 0.48 | <0.0001 | <0.0001 |
Event rates are n (%) for secondary outcomes. P-values compared using Kruskal–Wallis or χ2 tests.
CAD, coronary artery disease; eGFR, estimated glomerular filtration rate; hs-cTnT, high-sensitivity cardiac troponin T; IQR, interquartile range; MI, myocardial infarction; NT-proBNP, N-terminal-prohormone brain natriuretic peptide; SD, standard deviation.
Baseline characteristics
| Overall N = 2002 | Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|---|
| Demographics | |||||||||
| Age, mean (SD) | 62 (18) | 53 (17) | 74 (14) | 71 (13) | 76 (15) | <0.0001 | 0.03 | 0.12 | 0.06 |
| Women, n (%) | 1036 (52) | 604 (53) | 356 (51) | 27 (42) | 49 (54) | 0.35 | 0.15 | 0.19 | 0.57 |
| Presenting symptoms | |||||||||
| Chest discomfort, n (%) | 1002 (50) | 703 (61) | 212 (30) | 56 (88) | 31 (34) | <0.0001 | <0.0001 | <0.0001 | 0.45 |
| Dyspnoea, n (%) | 832 (42) | 436 (38) | 311 (44) | 39 (61) | 46 (51) | 0.0001 | 0.20 | 0.01 | 0.26 |
| Medical history | |||||||||
| Hypertension, n (%) | 1157 (58) | 502 (44) | 542 (77) | 42 (66) | 69 (76) | <0.0001 | 0.17 | 0.04 | 0.77 |
| Obesity, n (%) | 855 (43) | 483 (42) | 320 (46) | 25 (39) | 27 (30) | 0.03 | 0.22 | 0.32 | 0.004 |
| Current or prior tobacco use, n (%) | 1120 (56) | 633 (55) | 404 (58) | 36 (56) | 47 (52) | 0.66 | 0.57 | 0.84 | 0.29 |
| Cerebrovascular disease, n (%) | 200 (10) | 62 (5.4) | 115 (16) | 7 (11) | 16 (18) | <0.0001 | 0.25 | 0.26 | 0.77 |
| History of atrial dysrhythmias, n (%) | 376 (19) | 112 (9.8) | 217 (31) | 7 (11) | 40 (44) | <0.0001 | <0.0001 | 0.0008 | 0.01 |
| Heart failure, n (%) | 379 (19) | 73 (6.4) | 254 (36) | 15 (23) | 37 (41) | <0.0001 | 0.03 | 0.04 | 0.40 |
| Diabetes mellitus, n (%) | 485 (24) | 166 (15) | 264 (38) | 23 (36) | 32 (35) | <0.0001 | 0.92 | 0.79 | 0.65 |
| Chronic kidney disease, n (%) | 403 (20) | 70 (6) | 281 (40) | 19 (30) | 33 (36) | <0.0001 | 0.39 | 0.10 | 0.49 |
| Family history of CAD, n (%) | 622 (31) | 320 (28) | 250 (36) | 25 (39) | 27 (30) | 0.0029 | 0.22 | 0.58 | 0.26 |
| Peripheral artery disease, n (%) | 208 (10) | 38 (3.3) | 129 (18) | 13 (20) | 28 (31) | <0.0001 | 0.15 | 0.70 | 0.0052 |
| Dyslipidaemia, n (%) | 1035 (52) | 468 (41) | 463 (66) | 43 (67) | 61 (67) | <0.0001 | 0.98 | 0.84 | 0.84 |
| Ischaemic heart disease, n (%) | 446 (22) | 151 (13) | 242 (35) | 26 (41) | 27 (30) | <0.0001 | 0.16 | 0.32 | 0.36 |
| Laboratory results | |||||||||
| Baseline hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 21) | <6 (<6, 8.0) | 23 (17, 38) | 89 (38, 220) | 35 (19, 73) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Maximum hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 23) | <6 (<6, 9.0) | 24 (18, 43) | 216 (95, 898) | 44 (23, 111) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Creatinine (mg/dL), median (IQR) (n = 1985) | 0.9 (0.8, 1.1) | 0.8 (0.7, 1.0) | 1.1 (0.8, 1.4) | 1.0 (0.8, 1.4) | 1.1 (0.9, 1.5) | <0.0001 | 0.20 | 0.54 | 0.27 |
| eGFR (mL/min/1.73 m2), median (IQR) (n = 1984) | 80 (59, 91) | 91 (77, 91) | 60 (41, 79) | 67 (47, 83) | 55 (37, 74) | <0.0001 | 0.01 | 0.13 | 0.06 |
| NT-proBNP (pg/mL), median (IQR) (n = 718) | 409 (94, 2147) | 95 (44, 294) | 1181 (270, 3268) | 1794 (363, 4608) | 2211 (727, 6537) | <0.0001 | 0.20 | 0.44 | 0.003 |
| Initial 12-lead electrocardiogram | |||||||||
| Sinus rhythm, n (%) | 1679 (84) | 1067 (93) | 508 (72) | 59 (92) | 45 (50) | <0.0001 | <0.0001 | 0.0005 | <0.0001 |
| ST elevation, n (%) | 23 (1.1) | 1 (0.1) | 3 (0.4) | 19 (30) | 0 (0.0) | <0.0001 | <0.0001 | <0.0001 | 0.53 |
| ST-depression/T-wave inversion, n (%) | 130 (6.5) | 39 (3.4) | 45 (6.4) | 17 (27) | 29 (32) | <0.0001 | 0.48 | <0.0001 | <0.0001 |
| Overall N = 2002 | Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|---|
| Demographics | |||||||||
| Age, mean (SD) | 62 (18) | 53 (17) | 74 (14) | 71 (13) | 76 (15) | <0.0001 | 0.03 | 0.12 | 0.06 |
| Women, n (%) | 1036 (52) | 604 (53) | 356 (51) | 27 (42) | 49 (54) | 0.35 | 0.15 | 0.19 | 0.57 |
| Presenting symptoms | |||||||||
| Chest discomfort, n (%) | 1002 (50) | 703 (61) | 212 (30) | 56 (88) | 31 (34) | <0.0001 | <0.0001 | <0.0001 | 0.45 |
| Dyspnoea, n (%) | 832 (42) | 436 (38) | 311 (44) | 39 (61) | 46 (51) | 0.0001 | 0.20 | 0.01 | 0.26 |
| Medical history | |||||||||
| Hypertension, n (%) | 1157 (58) | 502 (44) | 542 (77) | 42 (66) | 69 (76) | <0.0001 | 0.17 | 0.04 | 0.77 |
| Obesity, n (%) | 855 (43) | 483 (42) | 320 (46) | 25 (39) | 27 (30) | 0.03 | 0.22 | 0.32 | 0.004 |
| Current or prior tobacco use, n (%) | 1120 (56) | 633 (55) | 404 (58) | 36 (56) | 47 (52) | 0.66 | 0.57 | 0.84 | 0.29 |
| Cerebrovascular disease, n (%) | 200 (10) | 62 (5.4) | 115 (16) | 7 (11) | 16 (18) | <0.0001 | 0.25 | 0.26 | 0.77 |
| History of atrial dysrhythmias, n (%) | 376 (19) | 112 (9.8) | 217 (31) | 7 (11) | 40 (44) | <0.0001 | <0.0001 | 0.0008 | 0.01 |
| Heart failure, n (%) | 379 (19) | 73 (6.4) | 254 (36) | 15 (23) | 37 (41) | <0.0001 | 0.03 | 0.04 | 0.40 |
| Diabetes mellitus, n (%) | 485 (24) | 166 (15) | 264 (38) | 23 (36) | 32 (35) | <0.0001 | 0.92 | 0.79 | 0.65 |
| Chronic kidney disease, n (%) | 403 (20) | 70 (6) | 281 (40) | 19 (30) | 33 (36) | <0.0001 | 0.39 | 0.10 | 0.49 |
| Family history of CAD, n (%) | 622 (31) | 320 (28) | 250 (36) | 25 (39) | 27 (30) | 0.0029 | 0.22 | 0.58 | 0.26 |
| Peripheral artery disease, n (%) | 208 (10) | 38 (3.3) | 129 (18) | 13 (20) | 28 (31) | <0.0001 | 0.15 | 0.70 | 0.0052 |
| Dyslipidaemia, n (%) | 1035 (52) | 468 (41) | 463 (66) | 43 (67) | 61 (67) | <0.0001 | 0.98 | 0.84 | 0.84 |
| Ischaemic heart disease, n (%) | 446 (22) | 151 (13) | 242 (35) | 26 (41) | 27 (30) | <0.0001 | 0.16 | 0.32 | 0.36 |
| Laboratory results | |||||||||
| Baseline hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 21) | <6 (<6, 8.0) | 23 (17, 38) | 89 (38, 220) | 35 (19, 73) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Maximum hs-cTnT concentration (ng/L), median (IQR) | 10 (<6, 23) | <6 (<6, 9.0) | 24 (18, 43) | 216 (95, 898) | 44 (23, 111) | <0.0001 | <0.0001 | <0.0001 | <0.0001 |
| Creatinine (mg/dL), median (IQR) (n = 1985) | 0.9 (0.8, 1.1) | 0.8 (0.7, 1.0) | 1.1 (0.8, 1.4) | 1.0 (0.8, 1.4) | 1.1 (0.9, 1.5) | <0.0001 | 0.20 | 0.54 | 0.27 |
| eGFR (mL/min/1.73 m2), median (IQR) (n = 1984) | 80 (59, 91) | 91 (77, 91) | 60 (41, 79) | 67 (47, 83) | 55 (37, 74) | <0.0001 | 0.01 | 0.13 | 0.06 |
| NT-proBNP (pg/mL), median (IQR) (n = 718) | 409 (94, 2147) | 95 (44, 294) | 1181 (270, 3268) | 1794 (363, 4608) | 2211 (727, 6537) | <0.0001 | 0.20 | 0.44 | 0.003 |
| Initial 12-lead electrocardiogram | |||||||||
| Sinus rhythm, n (%) | 1679 (84) | 1067 (93) | 508 (72) | 59 (92) | 45 (50) | <0.0001 | <0.0001 | 0.0005 | <0.0001 |
| ST elevation, n (%) | 23 (1.1) | 1 (0.1) | 3 (0.4) | 19 (30) | 0 (0.0) | <0.0001 | <0.0001 | <0.0001 | 0.53 |
| ST-depression/T-wave inversion, n (%) | 130 (6.5) | 39 (3.4) | 45 (6.4) | 17 (27) | 29 (32) | <0.0001 | 0.48 | <0.0001 | <0.0001 |
Event rates are n (%) for secondary outcomes. P-values compared using Kruskal–Wallis or χ2 tests.
CAD, coronary artery disease; eGFR, estimated glomerular filtration rate; hs-cTnT, high-sensitivity cardiac troponin T; IQR, interquartile range; MI, myocardial infarction; NT-proBNP, N-terminal-prohormone brain natriuretic peptide; SD, standard deviation.
Aetiologies
Except for one patient with coronary spasm, all type 2 MI cases had primary non-coronary diagnoses on presentation. Most (72%) were identified as having a single predominant trigger/aetiology for their supply–demand mismatch. The most common aetiologies were tachyarrhythmias, other, and respiratory failure (see Supplementary material online, Table S1). Among those with myocardial injury, the most common systemic aetiologies were chronic kidney disease, sepsis, and respiratory failure, whereas the most common underlying cardiac conditions were heart failure, arrhythmias, and hypertension (see Supplementary material online, Table S2).
Cardiac evaluations
Cardiac evaluations are shown in Table 2. Serial hs-cTnT measurements were performed in 59% of the entire cohort, including in 88% of those with at least one hs-cTnT increase. Compared with patients without myocardial injury, patients with myocardial injury (6.8 vs. 26%, P < 0.0001), type 1 MI (6.8 vs. 81%, P < 0.0001), and type 2 MI (6.8 vs. 40%, P < 0.005) were more likely to undergo echocardiography during the index presentation. Patients with type 1 MI had the highest frequency of echocardiography, invasive angiography, and percutaneous coronary interventions performed. They also had the highest proportion of echocardiography, stress testing, invasive coronary angiography, and coronary revascularization procedures during the following 30 days post-discharge.
Cardiac evaluations during index hospitalization and 30 days post-discharge
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Index presentation | ||||||||
| Echocardiography performed, n (%) | 78 (6.8) | 179 (26) | 52 (81) | 36 (40) | <0.0001 | <0.0001 | <0.0001 | 0.0045 |
| Left ventricular dysfunction (<50%), n (%) | 11 (14) | 41 (23) | 25 (48) | 13 (36) | <0.0001 | 0.27 | 0.0004 | 0.096 |
| Regional wall motion abnormalities, n (%) | 10 (13) | 28 (16) | 37 (71) | 12 (33) | <0.0001 | 0.0004 | <0.0001 | 0.013 |
| Valvular heart disease, n (%) | 7 (9.0) | 57 (32) | 8 (15) | 16 (44) | <0.0001 | 0.0026 | 0.02 | 0.15 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 59 (5.2) | 38 (5.4) | 2 (3.1) | 6 (6.6) | 0.81 | 0.34 | 0.43 | 0.64 |
| Abnormal stress (new ischaemia), n (%) | 8 (0.7) | 2 (0.3) | 2 (3.1) | 2 (2.2) | 0.018 | 0.72 | 0.0025 | 0.015 |
| Invasive angiography, n (%) | 14 (1.2) | 6 (0.9) | 50 (78) | 5 (5.5) | <0.0001 | <0.0001 | <0.0001 | 0.0004 |
| Any coronary artery disease (>50%), n (%) | 6 (43) | 2 (33) | 46 (92) | 1 (20) | <0.0001 | <0.0001 | 0.0001 | 0.62 |
| Percutaneous coronary revascularization, n (%) | 3 (21) | 1 (17) | 33 (66) | 1 (20) | 0.0027 | 0.044 | 0.019 | 0.89 |
| Length of stay (h), mean (SD) | 16 (39) | 55 (94) | 62 (47) | 69 (81) | <0.0001 | 0.72 | 0.0006 | 0.0082 |
| Post-discharge evaluations within 30 days | ||||||||
| Echocardiography performed, n (%) | 210 (18) | 264 (38) | 32 (50) | 35 (39) | <0.0001 | 0.15 | 0.051 | 0.87 |
| Left ventricular dysfunction (<50%), n (%) | 30 (14) | 78 (30) | 15 (47) | 9 (26) | <0.0001 | 0.071 | 0.046 | 0.64 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 140 (12) | 92 (13) | 15 (23) | 10 (11) | 0.013 | 0.14 | 0.69 | 0.40 |
| Abnormal stress (new ischaemia), n (%) | 19 (14) | 17 (19) | 4 (27) | 3 (30) | 0.32 | 0.86 | 0.46 | 0.38 |
| Invasive angiography, n (%) | 38 (3.3) | 47 (6.7) | 7 (11) | 4 (4.4) | 0.0010 | 0.12 | 0.20 | 0.40 |
| Any coronary artery disease (>50%), n (%) | 24 (63) | 27 (57) | 5 (71) | 3 (75) | 0.81 | 0.90 | 0.48 | 0.49 |
| Coronary revascularization, n (%) | 12 (1.0) | 15 (2.1) | 5 (7.8) | 1 (1.1) | 0.0003 | 0.033 | 0.0064 | 0.51 |
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Index presentation | ||||||||
| Echocardiography performed, n (%) | 78 (6.8) | 179 (26) | 52 (81) | 36 (40) | <0.0001 | <0.0001 | <0.0001 | 0.0045 |
| Left ventricular dysfunction (<50%), n (%) | 11 (14) | 41 (23) | 25 (48) | 13 (36) | <0.0001 | 0.27 | 0.0004 | 0.096 |
| Regional wall motion abnormalities, n (%) | 10 (13) | 28 (16) | 37 (71) | 12 (33) | <0.0001 | 0.0004 | <0.0001 | 0.013 |
| Valvular heart disease, n (%) | 7 (9.0) | 57 (32) | 8 (15) | 16 (44) | <0.0001 | 0.0026 | 0.02 | 0.15 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 59 (5.2) | 38 (5.4) | 2 (3.1) | 6 (6.6) | 0.81 | 0.34 | 0.43 | 0.64 |
| Abnormal stress (new ischaemia), n (%) | 8 (0.7) | 2 (0.3) | 2 (3.1) | 2 (2.2) | 0.018 | 0.72 | 0.0025 | 0.015 |
| Invasive angiography, n (%) | 14 (1.2) | 6 (0.9) | 50 (78) | 5 (5.5) | <0.0001 | <0.0001 | <0.0001 | 0.0004 |
| Any coronary artery disease (>50%), n (%) | 6 (43) | 2 (33) | 46 (92) | 1 (20) | <0.0001 | <0.0001 | 0.0001 | 0.62 |
| Percutaneous coronary revascularization, n (%) | 3 (21) | 1 (17) | 33 (66) | 1 (20) | 0.0027 | 0.044 | 0.019 | 0.89 |
| Length of stay (h), mean (SD) | 16 (39) | 55 (94) | 62 (47) | 69 (81) | <0.0001 | 0.72 | 0.0006 | 0.0082 |
| Post-discharge evaluations within 30 days | ||||||||
| Echocardiography performed, n (%) | 210 (18) | 264 (38) | 32 (50) | 35 (39) | <0.0001 | 0.15 | 0.051 | 0.87 |
| Left ventricular dysfunction (<50%), n (%) | 30 (14) | 78 (30) | 15 (47) | 9 (26) | <0.0001 | 0.071 | 0.046 | 0.64 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 140 (12) | 92 (13) | 15 (23) | 10 (11) | 0.013 | 0.14 | 0.69 | 0.40 |
| Abnormal stress (new ischaemia), n (%) | 19 (14) | 17 (19) | 4 (27) | 3 (30) | 0.32 | 0.86 | 0.46 | 0.38 |
| Invasive angiography, n (%) | 38 (3.3) | 47 (6.7) | 7 (11) | 4 (4.4) | 0.0010 | 0.12 | 0.20 | 0.40 |
| Any coronary artery disease (>50%), n (%) | 24 (63) | 27 (57) | 5 (71) | 3 (75) | 0.81 | 0.90 | 0.48 | 0.49 |
| Coronary revascularization, n (%) | 12 (1.0) | 15 (2.1) | 5 (7.8) | 1 (1.1) | 0.0003 | 0.033 | 0.0064 | 0.51 |
Event rates are n (%) for secondary outcomes. P-values compared using Kruskal–Wallis or χ2 tests.
MI, myocardial infarction; SD, standard deviation.
Cardiac evaluations during index hospitalization and 30 days post-discharge
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Index presentation | ||||||||
| Echocardiography performed, n (%) | 78 (6.8) | 179 (26) | 52 (81) | 36 (40) | <0.0001 | <0.0001 | <0.0001 | 0.0045 |
| Left ventricular dysfunction (<50%), n (%) | 11 (14) | 41 (23) | 25 (48) | 13 (36) | <0.0001 | 0.27 | 0.0004 | 0.096 |
| Regional wall motion abnormalities, n (%) | 10 (13) | 28 (16) | 37 (71) | 12 (33) | <0.0001 | 0.0004 | <0.0001 | 0.013 |
| Valvular heart disease, n (%) | 7 (9.0) | 57 (32) | 8 (15) | 16 (44) | <0.0001 | 0.0026 | 0.02 | 0.15 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 59 (5.2) | 38 (5.4) | 2 (3.1) | 6 (6.6) | 0.81 | 0.34 | 0.43 | 0.64 |
| Abnormal stress (new ischaemia), n (%) | 8 (0.7) | 2 (0.3) | 2 (3.1) | 2 (2.2) | 0.018 | 0.72 | 0.0025 | 0.015 |
| Invasive angiography, n (%) | 14 (1.2) | 6 (0.9) | 50 (78) | 5 (5.5) | <0.0001 | <0.0001 | <0.0001 | 0.0004 |
| Any coronary artery disease (>50%), n (%) | 6 (43) | 2 (33) | 46 (92) | 1 (20) | <0.0001 | <0.0001 | 0.0001 | 0.62 |
| Percutaneous coronary revascularization, n (%) | 3 (21) | 1 (17) | 33 (66) | 1 (20) | 0.0027 | 0.044 | 0.019 | 0.89 |
| Length of stay (h), mean (SD) | 16 (39) | 55 (94) | 62 (47) | 69 (81) | <0.0001 | 0.72 | 0.0006 | 0.0082 |
| Post-discharge evaluations within 30 days | ||||||||
| Echocardiography performed, n (%) | 210 (18) | 264 (38) | 32 (50) | 35 (39) | <0.0001 | 0.15 | 0.051 | 0.87 |
| Left ventricular dysfunction (<50%), n (%) | 30 (14) | 78 (30) | 15 (47) | 9 (26) | <0.0001 | 0.071 | 0.046 | 0.64 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 140 (12) | 92 (13) | 15 (23) | 10 (11) | 0.013 | 0.14 | 0.69 | 0.40 |
| Abnormal stress (new ischaemia), n (%) | 19 (14) | 17 (19) | 4 (27) | 3 (30) | 0.32 | 0.86 | 0.46 | 0.38 |
| Invasive angiography, n (%) | 38 (3.3) | 47 (6.7) | 7 (11) | 4 (4.4) | 0.0010 | 0.12 | 0.20 | 0.40 |
| Any coronary artery disease (>50%), n (%) | 24 (63) | 27 (57) | 5 (71) | 3 (75) | 0.81 | 0.90 | 0.48 | 0.49 |
| Coronary revascularization, n (%) | 12 (1.0) | 15 (2.1) | 5 (7.8) | 1 (1.1) | 0.0003 | 0.033 | 0.0064 | 0.51 |
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 2 vs. type 1 MI | P-value injury vs. type 1 MI | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Index presentation | ||||||||
| Echocardiography performed, n (%) | 78 (6.8) | 179 (26) | 52 (81) | 36 (40) | <0.0001 | <0.0001 | <0.0001 | 0.0045 |
| Left ventricular dysfunction (<50%), n (%) | 11 (14) | 41 (23) | 25 (48) | 13 (36) | <0.0001 | 0.27 | 0.0004 | 0.096 |
| Regional wall motion abnormalities, n (%) | 10 (13) | 28 (16) | 37 (71) | 12 (33) | <0.0001 | 0.0004 | <0.0001 | 0.013 |
| Valvular heart disease, n (%) | 7 (9.0) | 57 (32) | 8 (15) | 16 (44) | <0.0001 | 0.0026 | 0.02 | 0.15 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 59 (5.2) | 38 (5.4) | 2 (3.1) | 6 (6.6) | 0.81 | 0.34 | 0.43 | 0.64 |
| Abnormal stress (new ischaemia), n (%) | 8 (0.7) | 2 (0.3) | 2 (3.1) | 2 (2.2) | 0.018 | 0.72 | 0.0025 | 0.015 |
| Invasive angiography, n (%) | 14 (1.2) | 6 (0.9) | 50 (78) | 5 (5.5) | <0.0001 | <0.0001 | <0.0001 | 0.0004 |
| Any coronary artery disease (>50%), n (%) | 6 (43) | 2 (33) | 46 (92) | 1 (20) | <0.0001 | <0.0001 | 0.0001 | 0.62 |
| Percutaneous coronary revascularization, n (%) | 3 (21) | 1 (17) | 33 (66) | 1 (20) | 0.0027 | 0.044 | 0.019 | 0.89 |
| Length of stay (h), mean (SD) | 16 (39) | 55 (94) | 62 (47) | 69 (81) | <0.0001 | 0.72 | 0.0006 | 0.0082 |
| Post-discharge evaluations within 30 days | ||||||||
| Echocardiography performed, n (%) | 210 (18) | 264 (38) | 32 (50) | 35 (39) | <0.0001 | 0.15 | 0.051 | 0.87 |
| Left ventricular dysfunction (<50%), n (%) | 30 (14) | 78 (30) | 15 (47) | 9 (26) | <0.0001 | 0.071 | 0.046 | 0.64 |
| Stress testing (exercise, nuclear, or echocardiography), n (%) | 140 (12) | 92 (13) | 15 (23) | 10 (11) | 0.013 | 0.14 | 0.69 | 0.40 |
| Abnormal stress (new ischaemia), n (%) | 19 (14) | 17 (19) | 4 (27) | 3 (30) | 0.32 | 0.86 | 0.46 | 0.38 |
| Invasive angiography, n (%) | 38 (3.3) | 47 (6.7) | 7 (11) | 4 (4.4) | 0.0010 | 0.12 | 0.20 | 0.40 |
| Any coronary artery disease (>50%), n (%) | 24 (63) | 27 (57) | 5 (71) | 3 (75) | 0.81 | 0.90 | 0.48 | 0.49 |
| Coronary revascularization, n (%) | 12 (1.0) | 15 (2.1) | 5 (7.8) | 1 (1.1) | 0.0003 | 0.033 | 0.0064 | 0.51 |
Event rates are n (%) for secondary outcomes. P-values compared using Kruskal–Wallis or χ2 tests.
MI, myocardial infarction; SD, standard deviation.
Compared with patients with type 1 MI, patients with type 2 MI were less likely to undergo echocardiography (81 vs. 40%, P < 0.0001) and invasive coronary angiography (78 vs. 5.5%, P < 0.0001) during the index presentation. Likewise, compared with patients with type 2 MI, patients with myocardial injury were less likely to undergo echocardiography (40 vs. 26%, P = 0.005) and invasive coronary angiography (5.5 vs. 0.9%, P = 0.0004).
Among patients undergoing echocardiography, patients with type 2 MI were more likely to have regional wall motion abnormalities (33 vs. 16%, P = 0.01) than those with myocardial injury. Valvular heart disease was more frequent in patients with type 2 MI and myocardial injury than in patients with type 1 MI. Stress testing was infrequent during the index presentation and at 30-day follow-up in patients with type 2 MI (6.6 and 11%) and myocardial injury (5.4 and 13%).
Major adverse cardiovascular events
During index hospitalization, death occurred in 0.2% of those without myocardial injury, when compared with 1.7% in those with myocardial injury, 1.6% in type 1 MI, and 7.7% in those with type 2 MI. Index hospitalization mortality was more likely to occur in patients with type 2 MI than patients with myocardial injury (7.7 vs. 1.7%, P = 0.0004) with a trend towards worse outcomes when compared with patients with type 1 MI (7.7 vs. 1.6%, P = 0.09).
At 2-year follow-up, mortality occurred in 4.7% of patients without myocardial injury, whereas, in contrast, mortality occurred in 20% of those with type 1 MI, 33% of those with myocardial injury, and 39% of those with type 2 MI. Modes of death are presented in Table 3. When compared with type 1 MI, mortality in type 2 MI and myocardial injury were most often non-cardiovascular. Cardiovascular mortality was the most common cause of death in patients with type 1 MI; however, no statistical differences were observed between groups.
Event rates of major adverse cardiovascular events at 2 years, by diagnosis
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 1 vs. type 2 MI | P-value type 1 MI vs. injury | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Post-discharge MACE, n (%) | 96 (8.4) | 328 (47) | 15 (23) | 45 (50) | <0.0001 | 0.0011 | 0.0003 | 0.62 |
| Death | 54 (4.7) | 229 (33) | 13 (20) | 35 (39) | <0.0001 | 0.016 | 0.043 | 0.27 |
| Cardiovascular | 7 (13) | 39 (17) | 6 (46) | 8 (23) | <0.0001 | 0.90 | 0.21 | 0.22 |
| Non-cardiovascular | 28 (52) | 103 (45) | 3 (23) | 15 (43) | <0.0001 | 0.024 | 0.027 | 0.65 |
| Unknown | 19 (35) | 87 (38) | 4 (31) | 12 (34) | <0.0001 | 0.16 | 0.15 | 0.83 |
| Acute myocardial infarction | 10 (0.9) | 26 (3.7) | 1 (1.6) | 4 (4.4) | 0.0002 | 0.33 | 0.37 | 0.75 |
| Heart failure hospitalization | 25 (2.2) | 122 (17) | 5 (7.8) | 18 (20) | <0.0001 | 0.039 | 0.049 | 0.57 |
| Stroke or transient ischaemic attack | 18 (1.6) | 44 (6.3) | 1 (1.6) | 6 (6.6) | <0.0001 | 0.14 | 0.13 | 0.90 |
| New-onset atrial fibrillation or flutter | 16 (1.4) | 23 (3.3) | 1 (1.6) | 2 (2.2) | 0.056 | 0.78 | 0.45 | 0.58 |
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 1 vs. type 2 MI | P-value type 1 MI vs. injury | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Post-discharge MACE, n (%) | 96 (8.4) | 328 (47) | 15 (23) | 45 (50) | <0.0001 | 0.0011 | 0.0003 | 0.62 |
| Death | 54 (4.7) | 229 (33) | 13 (20) | 35 (39) | <0.0001 | 0.016 | 0.043 | 0.27 |
| Cardiovascular | 7 (13) | 39 (17) | 6 (46) | 8 (23) | <0.0001 | 0.90 | 0.21 | 0.22 |
| Non-cardiovascular | 28 (52) | 103 (45) | 3 (23) | 15 (43) | <0.0001 | 0.024 | 0.027 | 0.65 |
| Unknown | 19 (35) | 87 (38) | 4 (31) | 12 (34) | <0.0001 | 0.16 | 0.15 | 0.83 |
| Acute myocardial infarction | 10 (0.9) | 26 (3.7) | 1 (1.6) | 4 (4.4) | 0.0002 | 0.33 | 0.37 | 0.75 |
| Heart failure hospitalization | 25 (2.2) | 122 (17) | 5 (7.8) | 18 (20) | <0.0001 | 0.039 | 0.049 | 0.57 |
| Stroke or transient ischaemic attack | 18 (1.6) | 44 (6.3) | 1 (1.6) | 6 (6.6) | <0.0001 | 0.14 | 0.13 | 0.90 |
| New-onset atrial fibrillation or flutter | 16 (1.4) | 23 (3.3) | 1 (1.6) | 2 (2.2) | 0.056 | 0.78 | 0.45 | 0.58 |
Event rates are n (%) for primary outcomes. P-values compared using Kruskal–Wallis test.
MACE, major adverse cardiovascular events; MI, myocardial infarction.
Event rates of major adverse cardiovascular events at 2 years, by diagnosis
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 1 vs. type 2 MI | P-value type 1 MI vs. injury | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Post-discharge MACE, n (%) | 96 (8.4) | 328 (47) | 15 (23) | 45 (50) | <0.0001 | 0.0011 | 0.0003 | 0.62 |
| Death | 54 (4.7) | 229 (33) | 13 (20) | 35 (39) | <0.0001 | 0.016 | 0.043 | 0.27 |
| Cardiovascular | 7 (13) | 39 (17) | 6 (46) | 8 (23) | <0.0001 | 0.90 | 0.21 | 0.22 |
| Non-cardiovascular | 28 (52) | 103 (45) | 3 (23) | 15 (43) | <0.0001 | 0.024 | 0.027 | 0.65 |
| Unknown | 19 (35) | 87 (38) | 4 (31) | 12 (34) | <0.0001 | 0.16 | 0.15 | 0.83 |
| Acute myocardial infarction | 10 (0.9) | 26 (3.7) | 1 (1.6) | 4 (4.4) | 0.0002 | 0.33 | 0.37 | 0.75 |
| Heart failure hospitalization | 25 (2.2) | 122 (17) | 5 (7.8) | 18 (20) | <0.0001 | 0.039 | 0.049 | 0.57 |
| Stroke or transient ischaemic attack | 18 (1.6) | 44 (6.3) | 1 (1.6) | 6 (6.6) | <0.0001 | 0.14 | 0.13 | 0.90 |
| New-onset atrial fibrillation or flutter | 16 (1.4) | 23 (3.3) | 1 (1.6) | 2 (2.2) | 0.056 | 0.78 | 0.45 | 0.58 |
| Without myocardial injury N = 1145 | Myocardial injury N = 702 | Type 1 MI N = 64 | Type 2 MI N = 91 | P-value overall | P-value type 1 vs. type 2 MI | P-value type 1 MI vs. injury | P-value type 2 MI vs. injury | |
|---|---|---|---|---|---|---|---|---|
| Post-discharge MACE, n (%) | 96 (8.4) | 328 (47) | 15 (23) | 45 (50) | <0.0001 | 0.0011 | 0.0003 | 0.62 |
| Death | 54 (4.7) | 229 (33) | 13 (20) | 35 (39) | <0.0001 | 0.016 | 0.043 | 0.27 |
| Cardiovascular | 7 (13) | 39 (17) | 6 (46) | 8 (23) | <0.0001 | 0.90 | 0.21 | 0.22 |
| Non-cardiovascular | 28 (52) | 103 (45) | 3 (23) | 15 (43) | <0.0001 | 0.024 | 0.027 | 0.65 |
| Unknown | 19 (35) | 87 (38) | 4 (31) | 12 (34) | <0.0001 | 0.16 | 0.15 | 0.83 |
| Acute myocardial infarction | 10 (0.9) | 26 (3.7) | 1 (1.6) | 4 (4.4) | 0.0002 | 0.33 | 0.37 | 0.75 |
| Heart failure hospitalization | 25 (2.2) | 122 (17) | 5 (7.8) | 18 (20) | <0.0001 | 0.039 | 0.049 | 0.57 |
| Stroke or transient ischaemic attack | 18 (1.6) | 44 (6.3) | 1 (1.6) | 6 (6.6) | <0.0001 | 0.14 | 0.13 | 0.90 |
| New-onset atrial fibrillation or flutter | 16 (1.4) | 23 (3.3) | 1 (1.6) | 2 (2.2) | 0.056 | 0.78 | 0.45 | 0.58 |
Event rates are n (%) for primary outcomes. P-values compared using Kruskal–Wallis test.
MACE, major adverse cardiovascular events; MI, myocardial infarction.
At 2-year follow-up, MACE occurred in 8.4% of patients without myocardial injury, compared with 23% in patients with type 1 MI, 47% in myocardial injury, and 50% in type 2 MI (Table 3). The unadjusted 2-year risk for the composite of MACE and its components are summarized in Figure 1 and Supplementary material online, Tables S3 and S4. Adjusted analyses demonstrated no statistical difference in the risk for MACE or its components between patients with type 1 MI and those without myocardial injury (see Supplementary material online, Table S3). In contrast, when compared with patients without myocardial injury, patients with type 2 MI had a higher risk for MACE (8.4 vs. 50%, P < 0.0001; adjusted HR 2.31, 95% CI 1.49–3.58), death (4.7 vs. 39%, P < 0.0001; adjusted HR 2.60, 95% CI 1.52–4.44), and heart failure hospitalizations (2.2 vs. 20%, P < 0.0001; adjusted HR 5.57, 95% CI 2.75–11.32; Figures 2–4, Graphical Abstract). Likewise, when compared with those without myocardial injury, patients with myocardial injury had a higher risk for MACE (8.4 vs. 47%, P < 0.0001; adjusted HR 3.13, 95% CI 2.39–4.09), death (4.7 vs. 33%, P < 0.0001; adjusted HR 3.46, 95% CI 2.45–4.88), acute MI (0.9 vs. 3.7%, P = 0.0002; adjusted HR 2.90, 95% CI 1.25–6.71), and heart failure hospitalizations (2.2 vs. 17%, P < 0.0001; adjusted HR 3.29, 95% CI 1.99–5.46; Table 3, Supplementary material online, Table S3, Graphical Abstract, and Figures 5 and 6).
Unadjusted 2-year risk for major adverse cardiovascular event and its components. Unadjusted hazard ratios and 95% confidence intervals for the composite outcome of major adverse cardiovascular event and its components among patients with myocardial injury and types 1 and 2 myocardial infarction, with patients without myocardial injury being the reference group. AF, atrial fibrillation; HF, heart failure; MACEs, major adverse cardiovascular events; MI, myocardial infarction; TIA, transient ischaemic attack.
Post-discharge major adverse cardiovascular events. Two-year Kaplan–Meier curve for major adverse cardiovascular events in patients without myocardial injury, myocardial injury, type 1 myocardial infarction, and type 2 myocardial infarction.
Post-discharge mortality. Two-year Kaplan–Meier curve for mortality in patients without myocardial injury, myocardial injury, type 1 myocardial infarction, and type 2 myocardial infarction.
Post-discharge heart failure hospitalization. Two-year Kaplan–Meier curve for heart failure hospitalizations in patients without myocardial injury, myocardial injury, type 1 myocardial infarction, and type 2 myocardial infarction.
Risk for post-discharge major adverse cardiovascular event and mortality. Unadjusted and adjusted risk of major adverse cardiovascular events and mortality in patients without myocardial injury, myocardial injury, type 1 myocardial infarction, and type 2 myocardial infarction at 2-year follow-up. CI, confidence interval; MACE, major adverse cardiovascular events; MI, myocardial infarction.
Risk for post-discharge heart failure hospitalization. Unadjusted and adjusted risk for heart failure hospitalization at 2-year follow-up in patients without myocardial injury, myocardial injury, type 1 myocardial infarction, and type 2 myocardial infarction at 2-year follow-up. CI, confidence interval; HF, heart failure; MI, myocardial infarction.
Compared with patients with type 1 MI, patients with type 2 MI had higher crude rates of MACE (50 vs. 23%, P = 0.001; unadjusted HR 2.69, 95% CI 1.50–4.83), because of higher mortality (39 vs. 20%, P = 0.02; unadjusted HR 2.19, 95% CI 1.16–4.15), and more heart failure hospitalizations (20 vs. 7.8%, P = 0.04, unadjusted HR 3.30, 95% CI 1.23–8.90; Table 3, Supplementary material online, Table S4, Figures 5 and 6). Following adjustment, those with type 2 MI were demonstrated to be at higher risk than those with type 1 MI for MACE (adjusted HR 2.24, 95% CI 1.23–4.10) and heart failure hospitalizations (adjusted HR 2.93, 95% CI 1.08–7.97) but not mortality (adjusted HR 1.76, 95% CI 0.91–3.41; see Supplementary material online, Table S4, Figures 5 and 6).
When compared with type 1 MI, patients with myocardial injury also had statistically significant higher crude rates of MACE (23 vs. 47%, P = 0.0003; unadjusted HR 2.24, 95% CI 1.33–3.75), because of mortality (20 vs. 33%, P = 0.04; unadjusted HR 1.68, 95% CI 0.96–2.94), and heart failure hospitalizations (7.8 vs. 17%, P = 0.05; unadjusted HR 2.46, 95% CI 1.01–6.01). Adjusted analyses demonstrated a higher risk for MACE (adjusted HR 2.02, 95% CI 1.20–3.40) but not for death or heart failure hospitalizations. There were no differences overall in long-term outcomes between type 2 MI and myocardial injury over time (Table 3, Supplementary material online, Table S4, Figures 2–6).
Discussion
The present study provides the first US data to comprehensively evaluate the clinical features and long-term major adverse cardiovascular outcomes of patients diagnosed with myocardial injury and types 1 and 2 MI based on a clinically used hs-cTnT assay and systematic adjudications according to the Fourth UDMI. Given hs-cTn assays were not cleared by the US FDA until 2017, there are no real-life US outcomes data. Anticipating an increasing adoption of hs-cTn assays across the US, our analyses are timely and important for US clinicians, particularly because cTn is used often more broadly in US clinical practice.
Our data provide several important findings. First, patients with type 2 MI and myocardial injury, which explain 93% of cases with hs-cTnT increases above sex-specific 99th percentile thresholds, have high rates of MACE and mortality at 2 years. Approximately, half experience MACE during follow-up and 33% of those with myocardial injury and 39% of patients with type 2 MI have died by 2 years. Second, patients with type 2 MI and myocardial injury have a substantially higher risk of MACE, including death and heart failure hospitalizations, compared with patients without myocardial injury. Third, compared with patients with type 1 MI for which there are robust clinical practice guidelines and evidence-based recommendations,15 patients with type 2 MI experience significantly higher rates of post-discharge MACE predominantly due to increased mortality and more heart failure hospitalizations. Fourth, while long-term outcomes are similar between patients with type 2 MI and myocardial injury, these patient groups exhibit distinct phenotypes with separate clinical and biomarker profile, with patients with type 2 MI having significantly higher index mortality than patients with myocardial injury.
Our study has multiple strengths. First, this study is the first to comprehensively compare clinical features and prognoses among patients with types 1 and 2 MI and myocardial injury based on a clinically used hs-cTn assay in a US regional healthcare system. Second, our analyses are among the first to examine US long-term outcomes based on systematic adjudication of cases according to the recommendations of the Fourth UDMI.4 Third, when compared with previous studies, our post-discharge analyses addressed not only mortality and MI but also heart failure, stroke, and atrial fibrillation. These analyses provided important insights about heart failure hospitalizations being a strong contributor to MACE.
Since the implementation of hs-cTn, we and others have reported substantial increases in the proportion of patients with cTn concentrations above the 99th percentile compared with contemporary cTn assays.1,4 In our study, most patients with ≥1 cTnT >99th sex-specific percentile had myocardial injury, followed by type 2 and type 1 MI. Similar distributions have been shown previously using hs-cTn assays.16,17 Previous studies demonstrate that patients with myocardial injury and type 2 MI experience significant adverse prognoses, often worse than those with type 1 MI.4,8–11,17–19 In our cohort, during 2-year follow-up, at least one-third of patients with type 2 MI and myocardial injury died. Like in previous studies,2,10,20 the excess deaths were largely attributable to non-cardiovascular death, findings which are likely explained by the older age and higher burden of comorbidities in this population.10 However, conjoint cardiovascular disease is often present and likely contributes to adverse outcomes, for which opportunities exist for cardiovascular risk reduction.4
Most importantly, during 2-year follow-up in our cohort, many of the patients with type 2 MI and myocardial injury had cardiovascular events that included a substantial proportion of heart failure hospitalizations at a rate higher than that seen in those with type 1 MI. Whether this reflects the relative paucity of evaluations and thus likely therapy in these groups compared with those with type 1 MI, which we have documented, is unclear but should call attention to the need for cardiovascular evaluation for these patients. In the Catheter Sampled Blood Archive in Cardiovascular Diseases (CASABLANCA),18 incident type 2 MI predicted all-cause and cardiovascular death, as well as the composite of all-cause death, non-fatal MI, heart failure, stroke, TIA, peripheral artery disease, and cardiac arrhythmias. However, those with type 2 MI were not compared with those with type 1 MI, myocardial injury, or those without myocardial injury.18 In contrast to our findings, Chapman et al.,10 despite fewer evaluations and treatments as in our study, found that patients with type 2 MI had a comparable unadjusted 5-year risk of composite MACE to those with type 1 MI, although a lower risk of MACE after adjustment for covariates. Our study demonstrates comparable risks of death and composite MACE between those with type 2 MI and myocardial injury, which are greater than those observed in patients with type 1 MI.
The increased risk for heart failure hospitalizations in patients with type 2 MI is an observation that has important clinical implications and underscores potential therapeutic opportunities. Similar results were seen in the CASABLANCA study.18 Cediel et al.11 also reported that those with type 2 MI had higher rates of heart failure readmissions compared with type 1 MI, although these results were not statistically significant following adjustment. This was also the most common cause of cardiovascular mortality in an analysis of patients with type 2 MI diagnoses in Olmsted County.9 This might reflect the fact that patients with type 2 MI have recurrent episodes over time that lead to heart failure. If so, therapeutic interventions early on if one could identify those at high-risk might improve outcomes. At the least, it suggests that those with type 2 MI might benefit from additional evaluations.
Patients with type 2 MI are often older and manifest more comorbidities, characteristics which are also seen in those with myocardial injury.21 The poor outcomes observed in these two populations are consistent with other experiences9–11,18,22 and are likely a reflection of the increased burden of comorbidities along with the lack of in-depth evaluations and a paucity of optimal guideline-directed therapies for conditions for which data exist that are often in these patients such as coronary artery disease and/or heart failure.10,23 Given the similar prognoses observed in these patients, some have suggested these patient subsets should be grouped together17; however, it is clear from our data that they have distinct clinical phenotypes. Patients with type 2 MI present with higher baseline and maximum troponin concentrations, a higher frequency of ischaemic ST-depression or T-wave inversions, and an increased frequency of regional wall motion abnormalities by echocardiography. Further, patients with type 2 MI had higher index hospitalization mortality compared with those with myocardial injury. These findings indicate these patients have larger ischaemic insults and are at higher short-term cardiovascular risk.
There are some limitations to this study. Our study is an observational study with a retrospective design, and residual confounding could exist. Second, as in all studies adjudicating diagnoses per the UDMI, misclassification between myocardial injury and MI subtypes is possible. Third, we did not differentiate between acute and chronic myocardial injuries, but rather categorized those with myocardial injury according to the UDMI aetiologies. Fourth, our study evaluates the use of the Roche’s hs-cTnT assay and studies using hs-cTnI assays are needed. Fifth, despite an institutional 0/2 h hs-cTnT protocol, serial measurements were not systematically performed in all patients in routine clinical practice; however, most patients (88%) with at least one hs-cTnT increase had serial measurements. Sixth, limited events during follow-up for individual components of MACE restricted multivariable models and larger studies are needed. Finally, while an important strength of our study is it evaluates the long-term prognoses of ED patients evaluated with hs-cTnT across two community hospitals, larger multicentre studies with larger numbers of patients and events are needed.
Conclusions
Using hs-cTnT, vast majority of increases observed in patients presenting to an ED are due to myocardial injury with a higher incidence of type 2 MI than type 1 MI. Most importantly, patients with type 2 MI and myocardial injury have high long-term mortality with one-third of patients dead at 2-year follow-up. Both populations are not often extensively evaluated. Perhaps in part for that reason, they have significantly higher risks of long-term MACE when compared with those with type 1 MI and those without myocardial injury.
Supplementary material
Supplementary material is available at European Heart Journal: Acute Cardiovascular Care.
Acknowledgement
This publication was made possible in part by the Mayo Clinic CTSA through grant number UL1TR002377 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health (NIH).
Funding
None declared.
Conflicts of interest: Y.S. has previously served on the Advisory Boards for Roche Diagnostics and Abbott Diagnostics without personal compensation. He has also been a speaker for Abbott Diagnostics without personal financial compensation. A.S.J. has consulted or presently consults for most of the major diagnostics companies, including Beckman-Coulter, Abbott, Siemens, ET Healthcare, Ortho Diagnostics, Roche, Radiometer, Sphingotec, RCE Technologies, and Amgen and Novartis. All other authors have nothing to disclose.
Data availability
The data underlying this article are available in the article and in its online Supplementary material.
Comments