Admission high‐sensitivity C‐reactive protein levels improve the Grace risk score prediction on in‐hospital outcomes in acute myocardial infarction patients

Abstract Background Acute myocardial infarction (AMI) is the main cause of death and disability in cardiovascular and cerebrovascular diseases. Both the Global Registry of Acute Coronary Events (Grace) score and high‐sensitivity C‐reactive protein (hs‐CRP) were associated with prognosis in patients with AMI. However, whether the addition of the hs‐CRP to Grace risk score could improve the predictive power of Grace risk score on the prognosis of patients with AMI is unclear. Hypothesis: We hypothesized that the inclusion of hs‐CRP in the Grace risk score could improve the ability to correctly distinguish the occurrence of in‐hospital outcomes. Methods We retrospectively enrolled 1804 patients with AMI in the final analysis. Patients were divided into four groups by hs‐CRP quartiles. The relation between hs‐CRP and Grace risk score was analyzed by Spearman rank correlation. Logistic regression was used to identify independent risk factors. The predictive value of hs‐CRP add to Grace risk score was evaluated by C‐statistic, net reclassification improvement (NRI), integrated differentiation improvement (IDI), calibration plot, and decision curve analysis. Results The hs‐CRP and Grace risk score had a significantly positive correlation (r = .191, p < .001). hs‐CRP combined with Grace risk score could improve the ability of Grace risk score alone to correctly redistinguish the occurrence of in‐hospital outcome (C‐statistic = 0.819, p < .001; NRI = 0.05956, p = .007; IDI = 0.0757, p < .001). Conclusion Admission hs‐CRP level was a significant independent risk factor for in‐hospital outcomes in patients with AMI. The inclusion of hs‐CRP in the Grace risk score could improve the ability to correctly distinguish the occurrence of in‐hospital outcomes.


| INTRODUCTION
Acute myocardial infarction (AMI) is that the most serious form of coronary heart disease. About 550 000 individuals in the United States experience an AMI for the primary time every year. 1 By 2030, the number of AMI patients in China will reach 23 million. 2 In addition to the high incidence of AMI, it is the main cause of death and disability in cardiovascular and cerebrovascular diseases. Therefore, for patients with AMI, risk stratification is very important, especially identifying early adverse outcome risk.
To identify high-risk patients, Current guidelines recommend the Global Registry of Acute Coronary Events (Grace) score, which has been proved and widely used as a tool for risk stratification of AMI. 3 The Grace risk score system combines some clinical and biological variables and scores these variables to obtain an ultimate score, Patients were divided into low-risk, medium-risk, and high-risk groups based on the ultimate score to predict their risk of adverse outcomes.
This Grace risk score model only included two biological indicators: serum creatinine (SCr) and troponin, but some biological indicators closely related to myocardial infarction were not included. Therefore, whether the effect of Grace risk score on the prognosis of patients with AMI can be further improved by combining with these important biomarkers is unclear.
Inflammation is one of the mechanisms leading to AMI, 4 highsensitivity C-reactive protein (hs-CRP) as a biomarker of inflammation has been proven to be an important risk factor for cardiovascular disease, 5 It has also been shown to be associated with prognosis in patients with myocardial infarction. 6 The relationship between hs-CRP and the Grace risk score is not clear. The purpose of this study was to investigate the relationship between admission hs-CRP level and in-hospital outcome and whether the addition of the hs-CRP to Grace risk score could improve the predictive power of Grace risk score on the prognosis of patients with AMI.

| Study population
The present study is a single-center, retrospective observational cohort study. From January 2019 to December 2019, 1804 consecutive patients who were diagnosed with AMI performed percutaneous coronary intervention at Beijing Anzhen hospital were enrolled. The diagnosis of AMI must be based on at least two of the following three criteria: (1) a clinical history of ischemic chest pain,

| Demographic and clinical data
Patients' data of demographic, clinical, and angiographic characteristics were collected from Beijing An Zhen Hospital's medical information recording system.
The serum hs-CRP and the routine hematology, biochemical parameters were collected on admission and were measured by standard laboratory methods in the central lab of Beijing Anzhen Hospital.
Admission assessment indicators included in the Grace risk scoring model were obtained at hospital admission (age, heart rate, systolic blood pressure (BP), SCr level, Killip class, ST-segment deviation, elevated cardiac enzymes, and cardiac arrest). The Grace risk score was calculated according to the Grace risk calculator (https:// www.outcomes-umassmed.org/grace).

| Definitions
Malignant arrhythmia was defined as the arrhythmia that causes hemodynamic disturbance in a short time, leading to syncope or even sudden death, including ventricular fibrillation, ventricular tachycardia, thirddegree atrioventricular block, and so forth. The mechanical complication was defined as a complication of anatomical changes in the heart after myocardial infarction, including septal perforation, rupture of the papillary muscle of the mitral valve, and rupture of the heart. Cardiogenic shock was defined as a severe acute peripheral circulatory failure resulting from a significant decrease in cardiac output due to extreme cardiac dysfunction. Bleedings events were defined using the Bleeding Academic Research Consortium (BARC) classifications. 7

| In-hospital outcomes
The in-hospital outcomes were the composite of death, malignant arrhythmia, mechanical complication, congestive heart failure (HF), cardiogenic shock, thrombosis, bleeding, stroke.

| Statistical analysis
Continuous variables were presented as the mean ± standard deviation is consistent with a normal distribution, otherwise as to the median and interquartile range (IQR). Categorical variables were expressed as numbers and percentages. One-way analysis of variance or Mann-Whitney U test was used to analyze differences in continuous variables. The Pearson χ 2 test, Fisher's exact test, or the Cochran-Armitage Trend test was used to analyze categorical variables. The correlation between the Grace risk score and traditional hs-CRP was evaluated by Spearman correlation analysis. The admission values of hs-CRP were divided into four quartiles to stratify the incidence rates of in-hospital adverse events.
Univariate and multivariate logistic regression was used to estimate the in-hospital adverse events. The analysis results were presented by odds ratios and 95% confidence intervals (CIs).

| hs-CRP and the in-hospital outcome
The relationship between serum hs-CRP and in-hospital outcomes were shown in

| Association and combination of GRACE risk score with admission serum hs-CRP
The relation between hs-CRP and Grace risk score was estimated by  Table S3. Second, we evaluated the calibration degree of the model ( Figure S1). It  clinical benefit of the new model was significantly higher than the traditional Grace model (Figure 2).

| DISCUSSION
In this study, our main findings include: (1) a high level of admission hs-CRP is a significant independent risk factor for in-hospital outcome events in patients with AMI. (2) There was a significant positive correlation between hs-CRP and Grace risk score, and the inclusion of hs-CRP in the Grace risk score could improve the ability to correctly distinguish the occurrence of in-hospital outcomes.
Inflammation plays a key role in atherosclerosis, plaque rupture, and thrombosis, 10 which has been considered as one of the important risk factors of cardiovascular events. 4 hs-CRP is one of the biomarkers of inflammation, which has been proved to be associated with the risk of atherosclerotic cardiovascular disease. 5 Meanwhile, statins have also been shown to have anti-inflammatory effects in addition to lowering cholesterol. 11 The results of observation and randomized controlled trials showed that patients with cardiovascular disease benefit more significantly from the lower systemic inflammatory response. All these suggest that inflammation is involved I think it may be due to our short observation period and fewer cases.
The mechanism of hs-CRP in AMI patients is not only related to inflammation, but also related to different organ damage caused by oxidative stress, endothelial dysfunction, and mitochondrial abnormalities. [17][18][19] The Grace risk score has been widely recommended globally for risk stratification in patients with AMI to assess short-term (inhospital or 30 days) and medium-to long-term (≥1 year) outcomes. 3 The Grace score system includes two biological indicators, including SCr and myocardial enzymes, which indicates that these biological indicators can reflect the pathophysiological process of myocardial infarction. Previous studies have added other new biological The decision curve analysis (DCA) comparing the Grace risk score and its combination with admission serum hs-CRP. (The clinical benefit of the Grace + hs-CRP model was significantly higher than the traditional Grace model). Grace, Global Registry of Acute Coronary Events; hs-CRP, high-sensitivity C-reactive protein indicators to the Grace score system for predicting adverse clinical outcomes in patients with ACS, these novel biological indicators are NT-proBNP, 20 platelet reactivity, 21 serum calcium levels, 22 serum acid uric levels, 23 RDW/PDW, 24 and other factors. The results showed that these biomarkers in the Grace score may improve the identification of clinical outcomes in ACS patients, However, it is not clear whether hs-CRP as a nontraditional risk of AMI also enhances the ability of Grace risk stratification. In our study, the ACU area of hs-CRP combined with the Grace risk score model significantly increased from 0.785 of the traditional Grace model alone to 0.819, the difference was statistically significant. Moreover, we found hs-CRP combined Grace score model could improve the ability to correctly redistinguish the occurrence of hospitalization outcome events (the INR was 5.96%; p = .007). On the other hand, we found the new model of the prediction and actual events show good overall consistency. Finally, we found that the clinical benefit of the new model was significantly higher than the traditional Grace model. Therefore, our finding could support the assumption that the hs-CRP combined Grace risk score model enhances the ability of the Grace score system to predict the risk of in-hospital outcomes in patients with AMI.

| LIMITATION
This study had several limitations. First, this study was a singlecenter, retrospective study, which may have a selection bias or potential confounding factors. Second, we did not exclude patients with systemic inflammatory response, including acute infection, which may affect the concentration of hs-CRP. Third, the biochemical parameters were only measured at admission, and the dynamic changes of hs-CRP were not monitored during the in-hospital, which may result in measurement bias.

| CONCLUSION
Admission hs-CRP level was a significant independent risk factor for inhospital outcome events in patients with AMI. The inclusion of hs-CRP in the Grace risk score could improve the ability to correctly distinguish the occurrence of in-hospital outcomes. Further prospective, large, multicenter studies are needed to confirm our findings.