Usefulness of Haemoglobin Level Combined with CAMI-STEMI Score for Predicting MACCE in Patients with Acute ST-Elevation Myocardial Infarction after PCI

Anaemia and high haemoglobin levels are common in ST elevation myocardial infarction (STEMI) patients, but the effect of the haemoglobin level on the prognosis of STEMI patients remains in dispute. This study aimed to evaluate the prognostic value of the haemoglobin level combined with the CAMI-STEMI score in STEMI patients after percutaneous coronary intervention (PCI). We included 360 STEMI patients who underwent PCI. The patients were divided into 3 groups according to the first haemoglobin value after PCI. Clinical characteristics and the incidence of major adverse cardiovascular and cerebral events (MACCE) during the follow-up period were recorded. The incidence of MACCE in the 3 groups increased with a decrease in the haemoglobin level. Multivariate regression analysis showed that the CAMI-STEMI score was an independent predictor of MACCE incidence at 30 days after PCI and that anaemia was an independent predictor of MACCE incidence at 6 months and 1 year after PCI. A high haemoglobin level was an independent predictor of MACCE incidence at 1 year after PCI. The area under receiver operating characteristic curves (AUCs) of the haemoglobin level, CAMI-STEMI score, and haemoglobin level combined with CAMI-STEMI score predicting the occurrence of MACCE in STEMI patients within 30 days after PCI were 0.604, 0.614, and 0.639, respectively. In conclusion, The CAMI-STEMI score was an independent predictor of MACCE incidence at 30 days after PCI. The haemoglobin level combined with the CAMI-STEMI score improved the predictive value of MACCE in STEMI patients within 30 days after PCI. Trial Registration. This trial was a prospective cohort study and registered with ChiCTR-ROC-17011542.


Background
Anaemia and high haemoglobin levels are common in ST elevation myocardial infarction (STEMI) patients, but the effect of the haemoglobin level on the prognosis of STEMI patients remains in dispute [1,2]. Numerous observational studies have suggested that anaemia is an independent predictor of short-term cardiovascular events during hospitalization and within 30 days in patients with acute coronary syndrome (ACS) [3][4][5]. However, there were different views on the role of anaemia in forecasting the long-term prognosis of patients with ACS [6,7]. The prognostic value of anaemia in patients with different types of myocardial infarction (STEMI and NSTEMI) was not the same [1]. Studies have shown that, in STEMI patients, anaemia should be considered as an additional risk factor at admission [5]. However, the correlation between elevated haemoglobin levels and cardiovascular risk remains unclear. Sabatine et al. reported that, in patients with ACS, both high and low haemoglobin levels were linked to increased rates of mortality [1]. In patients with ACS, the usefulness of risk score for predicting death and cardiovascular complications has been proved both in the short and long term [1,8]. Early risk stratification was strongly recommended to guide therapeutic management and to improve outcome [9]. Yang  Myocardial Infarction (CAMI-STEMI) score for the 2017 European Society of Cardiology (ESC). The CAMI-STEMI score can predict in-hospital mortality among Chinese STEMI patients, with similar performance to the wellestablished Thrombolysis in Myocardial Infarction (TIMI) score, while relying solely on simple and practical variables; that is, there is no need to ask the history or draw blood to check the biochemistry [10]. The purpose of this study was to explore the predictive value of the haemoglobin level combined with the CAMI-STEMI score on the prognosis of STEMI patients after PCI and to provide ideas for improving STEMI risk stratification and optimizing the treatment of myocardial infarction.

Methods
From June 2013 to April 2016, consecutive patients with STEMI admitted to Zhongda Hospital (Nanjing, People's Republic of China) and treated with PCI were enrolled. The inclusion criteria were as follows: patients with STEMI who were diagnosed based on the guidelines for the diagnosis and treatment of acute ST segment elevation myocardial infarction in 2010 (China) [11]. The exclusion criteria were as follows: a history of major surgery, trauma, and bleeding over the past three months; contraindications to anticoagulant and antiplatelet therapy; serious injury of liver and kidney; malignant tumour; and clinical data and coronary angiography being incomplete.
Anaemia was defined as a serum haemoglobin level of <120 g/L for men and <110 g/L for women [12]. The normal reference range of serum haemoglobin was defined as 120 g/L to 160 g/L for men and 110 g/L to 150 g/L for women [13]. The patients were divided into 3 groups according to the first haemoglobin value after PCI, group 1 (male: Hb<120 g/L, female: Hb<110 g/L; 42 cases), group 2 (male: 120 g/L ≤ Hb<160 g/L, female: 110 g/L≤Hb<150 g/L; 278 cases), and group 3 (male: Hb ≥160 g/L, female: Hb ≥150 g/L;40 cases). Haemoglobin was measured from venous blood 24 hours after PCI. Biochemical indices including blood sugar, blood lipid, and creatinine were measured.
All patients were given aspirin (300 mg), ticagrelor (180 mg), or clopidogrel (300 mg) before operation, and aspirin (100 mg, QD), ticagrelor (90 mg, BID), or clopidogrel (75 mg, QD) were administered after surgery. Heparin was utilized routinely during surgery and as appropriate after surgery. Statins, beta blockers, nitrates, and angiotensin-converting enzyme inhibitors were routinely used in all three groups without contraindications.
The endpoints were the major adverse cardiac and cerebrovascular events (MACCE) during the follow-up period (30 days, 6 months, and 1 year after PCI). The MACCE included all-cause death, reconstruction of target vascular lesions, myocardial infarction during follow-up, unstable angina pectoris requiring hospitalization, heart failure, stroke, or transient cerebral ischaemia.
Each patient's baseline clinical data (including demographic data, previous medical history, and vital signs on admission), biochemical and angiographic variables, and ultrasonic cardiogram results were recorded. All patients were followed up by a cardiovascular physician and by telephone. All patients included in the study volunteered to participate in this clinical study and signed an informed consent form.

Statistical Analysis.
Analyses were performed using SPSS software, version 16.0 (SPSS, Inc., Chicago, IL, USA). Continuous variables were expressed as the mean±SD or median (inter-quartile range). Categorical variables were expressed as frequencies with percentages. Survival was graphically represented using Kaplan-Meier curves. Differences in survival rates were compared using the log-rank test. Univariate and multivariate Cox proportional hazards models were used to identify study endpoint predictors. Variables with univariate P values <0.10 were selected for multivariate analysis and expressed as hazard ratio (HR) with 95% confidence intervals (CIs). Multivariate Cox regression analysis was performed using a forward stepwise method. The area under receiver operating characteristic curve (AUC) was used to indicate the predictive value of the target factor on individual prognosis. All tests were 2-tailed, and statistical significance was defined as P <0.05.

Results
This study selected 378 patients. Eighteen cases were lost to follow-up, for a lost-to-follow-up rate of 4.8%; thus, 360 patients were finally enrolled. Baseline clinical data are presented in Table 1. Biochemical and angiographic variables, and ultrasonic cardiogram results are shown in Table 2. There were statistically significant differences (P<0.05) among the three groups in terms of age, the proportion of female patients, body mass index (BMI), smoking history, the proportion of Killip class>1, blood potassium concentration, albumin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatine kinase (CK), creatinine, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and the proportion of thrombus aspiration during PCI.
The CAMI-STEMI scores and postoperative follow-up for each group is shown in Table 3. The Kaplan-Meier curve that depicts the follow-up without a MACCE (MACCEfree) survival curve of 3 groups is given in Figure 1. There was a statistically significant difference among the 3 groups at 30 days (0.021), 6 months (0.014) and 1 year (0.003) of follow-up. Multivariate regression analysis (Table 4) showed that the CAMI-STEMI score was an independent predictor of MACCE incidence at 30 days after PCI (HR:1.225; 95%CI:1.067-1.406; P=0.004). The Cox regression model at 30 days included the CAMI-STEMI score, haemoglobin level, albumin, thrombus aspiration during PCI, and left ventricular ejection fraction (LVEF). As is presented in Table 5, anaemia was an independent predictor of MACCE incidence at 6 months after PCI (HR:2.071; 95%CI:1.178-3.461;    Abbreviations. IQR = interquartile range. Data are presented as median (IQR) or n (%). * There was a statistically significant difference in CAMI-STEMI score between the 3 groups. * * There was a statistically significant difference in MACCE occurrence between the 3 groups at 30 days (0.021), 6 months (0.014), and 1 year (0.003) of followup. The incidence of MACCE (30 days, 6 months, and 1 year) in the 3 groups increased with the decrease in the haemoglobin level.

Discussion
Anaemia is a frequently encountered comorbidity among patients presenting with ACS [14]. Liu et al showed that the risks of short-term mortality, long-term mortality, heart failure, cardiogenic shock, and major bleeding were increased in patients with anaemia compared with patients without anaemia. Anaemia may become a promising risk stratification factor in ACS [15]. Furthermore, a low baseline haemoglobin level was found to be an independent predictor of the risks of in-hospital major bleeding and of death at 1 month in ACS patients [16]. STEMI poses a serious threat to human life. Previous studies of the effect of the haemoglobin    level on the prognosis of patients with coronary heart disease have mostly concentrated in ACS patients, but few studies have been conducted for STEMI patients in particular. Anaemia and high haemoglobin levels are common in STEMI patients, but the effect of the haemoglobin level on the prognosis of STEMI patients remains in dispute [1,2]. Wang et al. reported that a haemoglobin level<120 g/L at baseline was associated with adverse outcomes and an elevated incidence of major adverse cardiovascular events (MACE) during the follow-up period in STEMI patients undergoing primary PCI [17]. Previous studies have demonstrated that the relationship between haemoglobin level and prognosis in patients with coronary heart disease was J-shaped or U-shaped, which indicated that patients with a high haemoglobin level have poor prognosis [18]. This study showed that the levels of haemoglobin in male and smoking STEMI patients were higher and that the concentration of haemoglobin decreased with age. Albumin, ALT, AST, CK, TC, LDL-C, and the proportion of thrombus aspiration during PCI were lower in STEMI patients with low levels of haemoglobin; and creatinine and the proportion of Killip class greater than 1 were higher in STEMI patients with high levels of haemoglobin. The Kaplan-Meier curve ( Figure 1) showed that the incidence of MACCE in the 3 groups increased with the decrease of haemoglobin level within 30 days, 6 months, and 1 year after PCI. The CAMI-STEMI score is a newly published, simple, and practical risk stratification scoring system. It does not require blood tests and medical history. Predictive accuracy of hospitalized mortality in Chinese STEMI patients is similar to that obtained using TIMI score and the Global Registration of Acute Coronary Events (GRACE) score [10]. This study showed that the CAMI-STEMI score was an independent predictor of MACCE incidence at 30 days after PCI. Haemoglobin was the most widely used experimental datum in STEMI patients, and the result was stable and easy to observe. This study showed that anaemia was an independent predictor of MACCE incidence at 6 months and 1 year after PCI and that the risk of MACCE incidence at 6 months and 1 year after PCI was significantly higher than that of patients with normal haemoglobin levels. The probable reasons for the worse clinical outcome and mortality in anaemic patients with AMI are as follows. Anaemia leads to the reduction of the myocardial oxygen supply and the increased myocardial oxygen demand, which have a negative effect on the myocardium [19]. Anaemia also leads to volume expansion, thereby contributing to the development of chronic heart failure (CHF) [20]. High haemoglobin levels were shown to increase blood viscosity, which, in turn, caused increased coronary vascular resistance, decreased coronary blood flow, and a predisposition to thrombosis [21][22][23]. However, this study showed that a high haemoglobin level was an independent predictor of MACCE incidence at 1 year after PCI (HR:0.456; 95%CI:0.222-0.937; P=0.033) and was a protective factor. This may be linked to the fact that increases in haemoglobin could elevate systemic vascular resistance and increase blood pressure [24,25]. Gevaert et al. [26] showed that a systolic blood pressure less than 100 mmHg increased mortality by 3.5 times among STEMI patients. Meanwhile, high haemoglobin carried more oxygen and increased oxygen delivery to myocardium downstream of coronary stenoses. This study also indicated that the haemoglobin level and CAMI-STEMI score had a high predictive value for MACCE in STEMI patients within 30 days after PCI and that the haemoglobin level combined with the CAMI-STEMI score improved the predictive value of MACCE in STEMI patients within 30 days after PCI. By contrast, CAMI-STEMI score had a poor predictive value for MACCE in STEMI patients at 1 year, whereas the predictive value of the CAMI-STEMI score declined with time.

Study Limitations
The following limitations of the present study should be addressed. First, the number of patients included in this study was relatively limited. There was a selection bias, and whether these defects will influence the results also requires a prospective study of large samples. Second, in this study, we only included the hemoglobin 24 hours after PCI and the etiology of anaemia or low values of haemoglobin and the detail of procedural complications were not reported. Third, although we adjusted several known confounding variables in the multivariable Cox proportional hazards models, other unknown factors might have played roles in MACCE.

Conclusion
This study indicated that the incidence of MACCE increased with a decrease in the haemoglobin level in STEMI patients. Anaemia was an independent predictor of MACCE incidence at 6 months and 1 year after PCI. And a high haemoglobin level was an independent predictor of MACCE incidence at 1 year after PCI and was a protective factor. The CAMI-STEMI score was an independent predictor of MACCE incidence at 30 days after PCI. The haemoglobin level combined with the CAMI-STEMI score improved the predictive value of MACCE in STEMI patients within 30 days after PCI. The American Heart Association (ACC) and American College of Cardiology (AHA) guidelines recommend screening and correcting anaemia in ACS patients [27,28]. Owing to the lack of adequate clinical data support, the guidelines did not specify which haemoglobin level to target. Greater sample size, more haemoglobin subgroups, longer follow-up time, and multicentre trials are needed to further determine the ideal haemoglobin level for STEMI patients and to provide ideas for optimizing the treatment of myocardial infarction and improving the prognosis of patients.

Data Availability
The datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.