Management and outcome trends in type 2 myocardial infarction: an investigation from the SWEDEHEART registry

Despite poor prognosis, patients with type 2 myocardial infarction (MI) tend to be underdiagnosed and undertreated compared to those with type 1 MI. Whether this discrepancy has improved over time is uncertain. We conducted a registry-based cohort study investigating type 2 MI patients managed at Swedish coronary care units (n = 14,833) during 2010–2022. Multivariable-adjusted changes (first three vs last three calendar years of the observation period) were assessed regarding diagnostic examinations (echocardiography, coronary assessment), provision of cardioprotective medications (betablockers, renin–angiotensin–aldosterone-system inhibitors, statins) and 1-year all-cause mortality. Compared to type 1 MI patients (n = 184,329), those with type 2 MI less often had diagnostic examinations and cardioprotective medications. Increases in the use of echocardiography (OR 1.08 [95% confidence interval 1.06–1.09]) and coronary assessment (OR 1.06 [95% confidence interval 1.04–1.08]) were smaller compared to type 1 MI (pinteraction < 0.001). The provision of medications did not increase in type 2 MI. All-cause mortality rate in type 2 MI was 25.4% without temporal change (OR 1.03 [95% confidence interval 0.98–1.07]). Taken together, the provision of medications and all-cause mortality did ot improve in type 2 MI despite modest increases in diagnostic procedures. This emphasizes the need of defining optimal care pathways in these patients.

Type 2 myocardial infarction (MI) is defined as acute ischemic myocardial injury due to oxygen supply/demand mismatch caused by a condition other than coronary plaque disruption or coronary intervention 1 . Compared to patients with type 1 MI, those with type 2 MI tend to be older and more frequently suffer from comorbidities, both cardiovascular and non-cardiovascular [2][3][4][5][6][7] . Outcomes after type 2 MI are poor; up to one third of the population will die within one year from the event 5,6 . Paradoxically however, type 2 MI patients tend to be underdiagnosed and undertreated [2][3][4][5][6][7] . This likely reflects clinical challenges related to the heterogeneity of underlying etiologies, the distinction of non-ischemic myocardial injury in patients with complex conditions, the presence of increasing age and co-morbidites, and the lack of evidence from prospective randomised controlled trials to guide treatment.
The concept of type 2 MI was introduced in 2007 together with the publication of the Universal Definition of MI 1 . It is conceivable that an increased awareness of type 2 MI as a high-risk condition may have influenced patient management through increases in the use of diagnostic examinations or of cardioprotective medications. This could plausibly have led to changes in outcomes for patients with type 2 MI. To investigate this issue, longitudinal information obtained over a sufficiently long period is needed. We here present registry-based data exploring changes in management patterns and all-cause mortality for a large cohort of type 2 MI patients who had been admitted during a 11-year period to Swedish coronary care units (CCU).

Material and methods
Study population. This study is part of the TOTAL-AMI (Tailoring Of Treatment in All comers with Acute Myocardial Infarction) project 8 . The primary aim of TOTAL-AMI is to study the mechanisms and implications of different MI subtypes 1 and comorbidities (e.g. chronic obstructive pulmonary disease, atrial fibrillation, renal dysfunction) in MI. TOTAL-AMI uses data from SWEDEHEART (Swedish Web-system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies) which www.nature.com/scientificreports/ is a Swedish nationwide registry prospectively collecting data from patients admitted to CCUs or other specialized facilities because of suspected acute coronary syndrome. SWEDEHEART provides information on patient demographics, medical history, symptoms, physical and ECG findings upon admission, blood test results, inhospital management, and discharge medications. Upon hospital admission, patients receive information about the registry, have the right to deny participation and get their data erased upon request. Written informed consent is not required according to Swedish law. The cohort of interest for the present study included all type 2 MI patients admitted between January 1, 2010 and February 13, 2022. For comparative purposes, we also considered type 1 MI patients who had been hospitalized during the same period. All patients had been discharged with I21 or I22 (International Classification of Diseases, 10th revision, Clinical Modification) as primary diagnostic code.
All data had been made pseudonymized before the statistical analyses. The study was conducted according to the principles of the 1975 Declaration of Helsinki and had been approved by the Regional Ethical Review Board in Stockholm (2012/60-31/2). Diagnostic classification. The diagnoses recorded in SWEDEHEART are set by the treating physicians at each respective hospital. Since 2010, the use of the Universal Definition including its subclassifications 1 is recommended within the SWEDEHEART framework 9 . Completeness of this information has gradually increased in the registry from 33.4% in 2010 to almost 100% from 2012 onwards (Supplemental Fig. 1).

Investigated medical interventions.
We investigated the annual rates of various medical interventions considered as quality measures for the management of type 2 MI: in-hospital echocardiography, in-hospital coronary assessment (invasive coronary angiography, treadmill test, myocardial perfusion scintigraphy, stress echocardiography), and provision of cardioprotective medications upon discharge: betablockers, renin-angiotensin-aldosterone-system (RAAS)-inhibitors or statins. For discharge medications, we focused on patients with coronary artery disease (CAD) since such treatments are strongly advised in these patients [10][11][12] . CAD was defined as previous MI, previous coronary revascularization, and coronary stenosis ≥ 50% upon invasive angiography or a pathologic non-invasive test performed during the hospitalization. We did not consider antiplatelet or anticoagulatory medications since decisions on treatment and treatment duration tend to be highly individualized in type 2 MI, depending on preexisting comorbidities, bleeding risk and the presence or absence of specific triggering conditions.
Since management decisions in type 2 MI might have been affected by specific comorbidities or contraindications, the following intervention-specific exclusion criteria were applied: • Total cohort: dementia; • Coronary assessment: ST-elevation upon admission, hemoglobin < 80 g/L; • Discharge medication with betablockers: heart rate < 50/min upon admission; • Discharge medication with RAAS-inhibitors: estimated glomerular filtration rate (eGFR; CKD-EPI equation) < 20 mL/min/1.73m 2 , left-ventricular ejection fraction > 0.50 in patients without concomitant diabetes, hypertension or known heart failure.
Prognostic evaluation. The prognostic outcome was 1-year all-cause mortality. Information on mortality until end of May 5, 2022 was obtained from the Swedish Population Registry.
Statistical analysis. All continuous variables were skewed and are reported as medians with 25th and 75th percentiles. Categorical variables are expressed as frequencies and percentages. Changes in management were assessed using multivariable logistic regressions using the first three vs last three calendar years of the observation period (i.e. 2010-2012 vs. 2020-2022) as explanatory covariates. Adjustment was made for hospital, sex, age, current smoking, diabetes, previous MI, previous coronary revascularization, previous heart failure, previous stroke, atrial fibrillation upon admission, chronic obstructive pulmonary disease, previous or present cancer and peripheral artery disease. Subanalyses were conducted in men and women, respectively. To ensure the validity of our results, two sensitivity analyses were applied: • Restricting the analyses to patients aged < 80 years since management decisions tend to be highly individualized in patients at higher ages; • Restricting the analyses to patients with a first-time admission since specific management decisions could have been based on information obtained during a previous hospitalization.
Interaction terms were added to the fully adjusted models for comparisons of management patterns in type 2 MI patients with and without CAD, female and male type 2 MI patients and between patients with type 2 MI and type 1 MI.
Changes in all-cause mortality were investigated using the same approach with additional adjustment for in-hospital coronary revascularization. Kaplan-Meier curves were plotted to illustrate the cumulative incidence of all-cause mortality per admission year with differences calculated using the Log-rank test.
No imputation was performed in case of missing data. In all tests, a two-sided p-value < 0.05 was considered significant, apart from the interaction analyses where significance was reached at a p value < 0.10. The software package SPSS 27.0 (SPSS Inc., Chicago, IL) was used for the analyses. In patients with type 2 MI, significant temporal changes in the rates of echocardiography and coronary assessment by + 8% and + 6%, respectively were noted following multivariable adjustment ( Table 2; Supplementary  Fig. S2). These increases were similar in men and women, as indicated by non-significant interaction terms (Supplementary Table S1). The sensitivity analyses did not provide different results (Supplementary Table S2). However, the temporal increases in examination rates were smaller as noted for type 1 MI (echocardiography: odds ratio 1.13 [95% confidence interval 1.13-1.14]; p interaction < 0.001; coronary assessment: odds ratio 1.12 [95% confidence interval 1.11-1.13]; p interaction < 0.001).
The provision of cardioprotective medications decreased in type 2 MI by 1-6% in adjusted models ( Table 2, Supplementary Fig. S3). The sensitivity analyses yielded similar results (Supplementary Table S2) apart for patients with a first-time admission in whom neutral temporal trends were seen. Prescription rates were generally higher in type 2 MI patients with CAD compared to those without. However, the temporal trends were similar in both cohorts (Supplementary Table S3) as also noted when men and women were assessed separately (Supplementary Table S1). Treatment patterns differed to type 1 MI where an increase in statin prescription was noted (odds ratio 1.05 [95% confidence interval 1.05-1.06; p interaction < 0.001]), a stronger decrease in betablocker prescription (odds ratio 0.88 [95% confidence interval 0.88-0.89; p interaction < 0.001]) and no change in the prescription of RAAS-inhibitors (odds ratio 1.00 [95% confidence interval 0.99-1.00; p interaction = 0.012]).
Information on all-cause mortality was available in 11,608 unique type 2 MI patients. A total of 2954 (25.4%) patients died within one year from admission. Mortality rates for patients admitted 2010-2012 were 26.0% (n = 748/2882), compared to 25.8% (n = 597/2313) for those admitted 2019-2021. The Kaplan-Meier curves yielded no clear pattern indicating an unidirectional mortality change in outcome over time ( Fig. 2A and B).

Discussion
The concept of type 2 MI was introduced in 2007 in the context of the publication of the Universal Definition 1 . Since then, type 2 MI has been recognized as a condition associated with a considerable burden of comorbidities and poor prognosis [2][3][4][5][6][7] . Paradoxically however, our data from Swedish type 2 MI patients admitted over a 11-year period (over 14 years from the publication of the Universal Definition) demonstrate lower use of diagnostic tests and lower medication provision without improvement over time. Moreover, we observed no major change in 1-year all-cause mortality. Our findings are intriguing, in particular since patients were managed by cardiology staff with a presumably high awareness on the clinical implications of type 2 MI. www.nature.com/scientificreports/ Overall, we noted minor increases in the rates of echocardiography and coronary assessment. These changes however, were smaller compared to type 1 MI. Accordingly, they appear to represent rather a management change in the general MI population than improved routines in type 2 MI. The provision of RAAS-inhibitors and statins did not increase in the overall type 2 MI population including those with CAD, known to have the poorest prognosis 2,4 . This was in contrast to type 1 MI patients for whom interaction analyses indicated more favorable prescription trends. Prescriptions of betablockers decreased significantly in type 2 MI patients but remained at overall high levels, e.g. 82.5% in 2020-2022 in patients with CAD.
The lack of increases in prescription rates of cardioprotective medications in type 2 MI is intriguing. We can only speculate on possible reasons why physicians were reluctant to initiate such treatments. For some patients, the management of the triggering condition may have overshadowed the need of cardioprotective medications. Polypharmacy may have been an issue in other patients. Limited perceived life expectancy could have mattered as well since type 2 MI patients in general represent an aged population. However, the results from the sensitivity analyses restricted to patients < 80 years yielded at least similar decreases in prescription rates as noted for the entire type 2 MI cohort.
Sex-disparities in the management of type 2 MI may exist but evidence is still limited 14,15 . In a previous investigation from the SWEDEHEART registry, no indication of sex bias in the selection of type 2 MI patients for cardiac investigations was found 16 . We here extend these findings by demonstrating that temporal trends in medical interventions did not differ between women and men. However, we acknowledge that selection bias may be present since type 2 MI patients warded outside cardiology departments more often tend to be female 7 .
Our data emphasize the need of evidence-based care pathways in type 2 MI. Notably, these patients share many risk factors with those having type 1 MI 17 . It is thus, conceivable that aggressive risk factor management may achieve similar prognostic benefits in both MI types. Supporting evidence however, is scarce. Medication with Table 1. Clinical characteristics in patients with type 2 MI. *Data from hospital survivors (all type 2 MI: n = 14,116; type 2 MI with CAD: n = 7391; type 1 MI: n = 174,866). Data are given as numbers with percentages, and as medians with 25th and 75th percentiles. MI Myocardial infarction; eGFR Estimated glomerular filtration rate; PCI Percutaneous coronary intervention; CABG Coronary artery bypass grafting; adm: admission; COPD Chronic obstructive pulmonary disease; RAAS Renin-angiotensin-aldosterone-system; CAD coronary artery disease. www.nature.com/scientificreports/ statins has been shown to improve outcomes in MI patients without obstructive coronary arteries (MINOCA) 18 among whom many may have had type 2 MI. Lipidlowering and antidiabetic treatments moreover, reduce the risk of future type 2 MI 19,20 . On a similar note, medications with betablockers or RAAS-inhibitors exhibit prognostic benefit in patients with type 2 MI 3 and MINOCA 18 . While there still are knowledge gaps on how to manage type 2 MI optimally, we believe that more intense diagnostic assessment and medication might improve outcomes in this vulnerable population. Recently published data suggest that the T2-risk score might facilitate the identification of high-risk type 2 MI patients with particular need for further investigation 21 . The Targeting Investigation and Treatment in Patients With Type 2 Myocardial Infarction (TARGET-Type 2) pilot study (NCT05419583, clinicaltrials.org) and the Appropriateness of Coronary Investigation in Myocardial Injury and Type 2 Myocardial Infarction (ACT-2) study 22 will provide clarifying evidence in this regard. However, an agreement on optimal timing and the setting of cardiac assessment and follow-up in type 2 MI may be difficult since these patients represent a diverse cohort, often with complex health issues. We suggest that patients aged < 80 years should undergo cardiac assessment, either during the initial hospitalization or in the outpatient setting. Identified risk factors and cardiac conditions should be treated according to current guidelines. The underlying triggering condition should be managed by the respective  www.nature.com/scientificreports/ specialty. We suggest individualized management approaches in patients aged ≥ 80 years depending on biological age and patient preference. We acknowledge some study limitations that need to be considered. Our investigation is limited to a single healthcare system. Although all hospitals participating in SWEDEHEART are annually monitored, the data cannot be of the same quality as in a prospective study. However, the accuracy of the data and the registry has been found to be high 23 . The diagnosis of MI was set locally by the treating physicians. While the SWEDEHEART framework recommends the use of criteria outlined in the Universal Definition 9 , there was no independent adjudication. We can thus, not exclude erroneous diagnosis or subtyping of MI in some cases 24 . There may have been unmeasured confounders not documented in SWEDEHEART that could have influenced management decisions, e.g. patient refusal, comorbidities, frailty or short life expectancy. We lack information on the dosage of medications, examination results or treatment modifications performed after hospital discharge. Finally, our analysis is restricted to patients admitted to CCUs why selection bias may be present. Patients with type 2 MI are often given ward in other facilities with even lesser use of diagnostic procedures and medications according to recent Swedish data 7 . This implies the possibility of even greater management disparities compared to type 1 MI. Extrapolation of our findings to these patients should thus, be done with caution.
In conclusion, our data demonstrate no favorable trend in the use of diagnostic tests and provision of cardioprotective medications in patients with type 2 MI. This represents a missed opportunity to improve outcome. Accordingly, all-cause mortality in our study cohort remained largely unchanged during the observation period. Our findings thus, emphasize the need of defining optimal care pathways for type 2 MI patients.

Data availability
The data used in this study originates from the SWEDEHEART registry and contains sensitive patient information. The dataset analyzed in this study is not publicly available due to Swedish patient privacy, secrecy laws regulating access to SWEDEHEART, and due to ethical restrictions regarding the current analysis from the TOTAL-AMI project (Regional Ethical Review Board in Stockholm; reference number 2012/60-31/2). Data access can be made available at Uppsala Clinical Research Center upon reasonable request and under the provision that the data is accessed onsite and does not leave Uppsala University. This request can be sent to info@ucr.uu.se.