Clinical and prognostic implications of C‐reactive protein levels in myocardial infarction with nonobstructive coronary arteries

Abstract Background Myocardial infarction with nonobstructive coronary arteries (MINOCA) is a heterogeneous condition. Recent studies suggest that MINOCA patients may have a proinflammatory disposition. The role of inflammation in MINOCA may thus be distinct to myocardial infarction with significant coronary artery disease (MI‐CAD). Hypothesis We hypothesized that inflammation reflected by C‐reactive protein (CRP) levels might carry unique clinical information in MINOCA. Methods This retrospective registry‐based cohort study (SWEDEHEART) included 9916 patients with MINOCA and 97 970 MI‐CAD patients, used for comparisons. Multivariable‐adjusted regressions were applied to investigate the associations of CRP levels with clinical variables, all‐cause mortality and major cardiovascular events (MACE) during a median follow‐up of up to 5.3 years. Results Median admission CRP levels in patients with MINOCA and MI‐CAD were 5.0 (interquartile range 2.0–9.0) mg/dl and 5.0 (interquartile range 2.1–10.0 mg/dl), respectively. CRP levels in MINOCA exhibited independent associations with various cardiovascular risk factors, comorbidities and estimates of myocardial damage. The association of CRP with peripheral artery disease tended to be stronger compared to MI‐CAD. The associations with female sex, renal dysfunction and myocardial damage were stronger in MI‐CAD. CRP independently predicted all‐cause mortality in MINOCA (hazard ratio 1.22 [95% confidence interval 1.17–1.26]), similar to MI‐CAD (p interaction = 0.904). CRP also predicted MACE (hazard ratio 1.08 [95% confidence interval 1.04–1.12]) but this association was weaker compared to MI‐CAD (p interaction<.001). Conclusions We found no evidence indicating the presence of a specific inflammatory pattern in acute MINOCA compared to MI‐CAD. However, CRP levels were independently, albeit moderately associated with adverse outcome.


| INTRODUCTION
Myocardial infarction with nonobstructive coronary arteries (MINOCA) is a condition that is gaining increasing interest. 1,2 The term MINOCA refers to a myocardial infarction (MI) fulfilling the criteria outlined in the Universal Definition 3 but without coronary stenosis ≥50%. Around 5%-10% of all MI are MINOCA. 1,4 The etiology of MINOCA is heterogenous with a multitude of causative or contributing mechanisms. 1 However, recent studies suggest that MINOCA patients as well may be characterized by a proinflammatory disposition in terms of more common inflammatory disorders 5,6 or as reflected by circulating biomarkers. 7 Inflammation also represents a key pathobiological mechanism in acute coronary syndromes. C-reactive protein (CRP), a sensitive downstream marker of the inflammation cascade, has been linked in various studies to adverse outcome in this condition. 8,9 Given the considerations above, we hypothesized that CRP levels might provide clinical and prognostic information that is unique in MINOCA and at difference to myocardial infarction with significant coronary artery disease (MI-CAD).
The aims of the present study were to closer investigate the associations of CRP levels in MINOCA with clinical findings and outcome overall, and in comparison to CRP levels in MI-CAD. For this purpose, a large cohort of Swedish MI patients with data available in the SWEDEHEART (Swedish Web-system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies) registry was assessed.

| Study population
This study is part of the TOTAL-AMI (Tailoring Of Treatment in All comers with Acute Myocardial Infarction) project. The primary aim of TOTAL-AMI is to study the mechanisms and implications of different MI subtypes 3

| Measurement of CRP levels
The CRP levels used in this analysis had been obtained upon admission as part of clinical routine practice in Sweden. Measurements were performed using standard assays at the central laboratories of the respective hospitals participating in SWEDEHEART.

| Statistical analysis
All continuous variables were skewed and are reported as medians with interquartile ranges (IQR). Differences were tested using the Cumulative probability curves were constructed using the Kaplan-Meier method, and the log-rank test was used to compare the occurrence of adverse outcome across CRP quartiles.
In all tests, a two-sided p value <.05 was considered significant.
The software package SPSS 27.0 (SPSS Inc., Chicago, IL) was used for the analyses.  Table 1. Increasing CRP quartiles exhibited significant associations across many cardiovascular risk factors and comorbidities, apart from hyperlipidemia and previous stroke. Increasing CRP quartiles were also associated with more adverse ECG findings, poorer left-ventricular ejection fraction and higher hs-cTnT levels. The clinical characteristics of MI-CAD patients are presented in the Table S1.
Upon multivariable adjustment, significant and consistent associations were noted between CRP (ln) and current smoking, diabetes mellitus, lower glomerular filtration rate, estimates of myocardial damage (i.e. lower left-ventricular ejection fraction and higher hs-cTnT levels), peripheral artery disease, chronic obstructive pulmonary disease, previous or present cancer and atrial fibrillation ( Table 2). As illustrated by the non-overlapping 95% confidence intervals of the regression coefficients in model 2, the association of CRP (ln) with peripheral artery disease tended to be stronger in MINOCA compared to MI-CAD, and there was a trend towards a stronger association with chronic obstructive pulmonary disease (Table S2B). Among patients with MI-CAD, CRP (ln) exhibited stronger associations with female sex, lower glomerular filtration rate and depressed left-ventricular ejection fraction (Table S2B)    <.001 -Abbreviations: BMI, body mass index; CCB, calcium channel blockers; CV, cardiovascular; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; RAAS, renin-angiotensinaldosterone system; hs-cTnT, high-sensitivity cardiac troponin T; LVEF, left-ventricular ejection fraction; MI, myocardial infarction; MACE, major cardiovascular adverse events; PAD, peripheral artery disease.
Note: Data given as numbers (with percentages) or medians (with interquartile ranges). study. However, the agreement between the information entered in the registry and the medical records is around 96%. 15 We lack detailed information on the assays used for CRP measurements. We would not regard this as a major shortcoming since guidelines recommend the use of uniform prognostic CRP thresholds despite interassay variations in detection ranges. 16 We are unable to comment on the potential associations of the severity of atheromatosis with CRP levels in MINOCA since the data available in SWEDEHEART do not allow for a subclassification of patients with coronary stenosis <50%.
Moreover, we lack information on intravascular imaging which might have identified high-risk plaque morphologies contributing to higher CRP levels in some MINOCA patients. We do not have data on the Our results in conclusion, demonstrate that CRP is independently, albeit moderately associated with adverse outcome in MINOCA. Contrasting to our primary hypothesis, we found no evidence indicating F I G U R E 1 Cumulative probability of (A) all-cause mortality and )B) major cardiovascular events in patients with myocardial infarction with nonobstructive coronary arteries (MINOCA) in relation to C-reactive protein (CRP) quartiles that acute MINOCA might be characterized by a specific inflammatory pattern, as reflected by CRP levels.

ACKNOWLEDGMENTS
The TOTAL-AMI project has received funding from the Swedish Foundation of Strategic Research. This organization had no role in the collection, analysis and interpretation of the data, in the writing of the report and in the discussion to submit this paper for publication.

CONFLICT OF INTEREST
None of the authors reported conflicts of interest relevant to this study.

DATA AVAILABILITY STATEMENT
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 and 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). We hereby confirm that other researchers are able to access the data 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.