Prognostic Analysis for Cardiogenic Shock in Patients with Acute Myocardial Infarction Receiving Percutaneous Coronary Intervention

Cardiogenic shock (CS) is uncommon in patients suffering from acute myocardial infarction (AMI). Long-term outcome and adverse predictors for outcomes in AMI patients with CS receiving percutaneous coronary interventions (PCI) are unclear. A total of 482 AMI patients who received PCI were collected, including 53 CS and 429 non-CS. Predictors for AMI patients with CS including recurrent MI, cardiovascular (CV) mortality, all-cause mortality, and repeated-PCI were analyzed. The CS group had a lower central systolic pressure and central diastolic pressure (both P < 0.001). AMI patients with hypertension history were less prone to develop CS (P < 0.001). Calcium channel blockers and statins were less frequently used by the CS group than the non-CS group (both P < 0.05) after discharge. Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) score, CV mortality, and all-cause mortality were higher in the CS group than the non-CS group (all P < 0.005). For patients with CS, stroke history was a predictor of recurrent MI (P = 0.036). CS, age, SYNTAX score, and diabetes were predictors of CV mortality (all P < 0.05). CS, age, SYNTAX score, and stroke history were predictors for all-cause mortality (all P < 0.05). CS, age, and current smoking were predictors for repeated-PCI (all P < 0.05).


Introduction
Cardiogenic shock (CS) is uncommon in patients with acute myocardial infarction (AMI). However, an AMI complicated by CS is a complex syndrome which may induce low cardiac output and hypotension resulting in multiorgan dysfunction and mortality. Even with the introduction of modern intensive care units (ICUs), advanced medical treatment, and invasive devices, the short-term mortality and morbidity of AMI complicated by CS remain high [1][2][3][4]. The mortality rate for AMI complicated by CS after early revascularization, including percutaneous coronary intervention (PCI), is approximately 40% to 60%. In addition, as for age and gender, patients with AMI complicated by CS who are older than 75 years of age may have a higher one-year mortality than their younger counterparts [5,6]. In addition, compared with men, women suffering from STEMI more often have concurrent CS, according to some studies [7].
In terms of an invasive strategy for CS, there was no difference in 30-day survival rate between PCI and coronary artery bypass graft (CABG) group [8]. On the other hand, the outcome data comparing multivessel with culprit lesion PCI is controversial. Thus, the best revascularization strategy for CS patients remains obscure [9][10][11][12].
Long-term prognosis of patients with AMI complicated by CS is still unclear and analysis of predictors for adverse clinical outcomes has not been well studied. Therefore, this study aimed to survey the clinical features and outcomes of patients with AMI complicated by CS compared to those without CS. Moreover, the risk factors for recurrent myocardial infarction (MI), cardiovascular (CV) mortality, all-cause mortality, and repeated-PCI in patients with AMI complicated by CS were analyzed.

Study Population.
A retrospective survey of a prospective database was conducted via medical record review over a period from 2007 through 2014. AMI patients between 20 and 90 years of age were consecutively recruited from the inpatient clinic at Taichung Tzu Chi Hospital, Taiwan. They were divided into two groups: patients with AMI complicated by CS (the CS group) and AMI patients without CS (non-CS group). Patients suffering from out of hospital death (OHCA) and patients with malignancy were excluded from the analysis. All patients were followed up regularly via the outpatient department (OPD). A survey focusing on MI, repeated-PCI, CV mortality, and all-cause mortality was completed for each patient at the end of the study.

Definition, Data
Collection, and Measurement. CS was defined as systemic blood pressure (BP) less than 80/50 mmHg or less than 90/60 mmHg after vasopressor therapy during admission to the emergency department. Diabetes was defined as a fasting plasma glucose level of more than 126 mg/dL, a casual plasma glucose level greater than 200 mg/dL, or a hemoglobin A1c (HbA1c) level of more than 6.5%. Hypercholesterolemia was defined as a serum cholesterol level of more than 200 mg/dL or an LDL-C level of more than 100 mg/dL. Chronic kidney disease (CKD) was defined as an estimated glomerular filtration rate (eGFR) of less than 60 mL/min/1.73 m 2 , which was equal to or more than stage III chronic kidney disease (CKD). Previous MI history was defined as a history of MI prior to admission, accompanied by a threefold elevation of cardiac enzymes from the baseline value.
Measurements of body parameters included body height, body weight, and body mass index (BMI). Baseline biochemical data collected on admission included fasting plasma glucose, serum creatinine, total cholesterol, high-density lipoprotein-cholesterol (HDL-C), low-density lipoproteincholesterol (LDL-C), and serum triglyceride level. As for the hemodynamic data, central aorta systolic pressure and central aorta diastolic pressure during cardiac catheterization were also collected. The central aortic pressure (CAP) was measured via a pigtail catheter while performing a coronary angiography. The angiographic findings included number and distribution of diseased vessels, number of treated vessels, and number of lesions. Left ventricular systolic function was usually calculated via two-dimensional echocardiography. Lesion severity and complexity were evaluated using the Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) score. Related clinical parameters including baseline characteristics, related risk factors, hemodynamic data, angiographic findings, and treatment strategies such as drug medications after discharge or invasive procedures (balloon angioplasty, bare-metal stent deployment, or drug-eluting stent deployment) were compared between patients with CS and those without CS. In addition, this study attempted to identify the significant predictor of AMI patients developing cardiogenic shock and to analyze the adverse predictors of recurrent MI, CV mortality, all-cause mortality, and repeated-PCI procedures in patients with CS.

Statistical Analysis.
The patients were divided into two groups: patients with AMI complicated by CS (the CS group) and AMI patients without CS (non-CS group). The independent -test, chi-squared test, Fisher's exact test, and multivariate logistic regression analysis were used to compare the differences between the two groups. The log rank test and Kaplan-Meier curves were used for the survival analysis.
The Cox proportional hazards model was used to test the effects of independent variables on hazards. values less than 0.05 were considered significant. All analyses were performed using the statistical package SPSS for Windows (Version 23.0, SPSS Inc., Chicago, IL).

Results
A total of 482 patients who suffered from AMI were collected during the study period. Among them, there were 53 patients with AMI complicated by CS on admission, while 429 patients had no CS during admission. The mean age in the CS group compared with the non-CS group was 62.5 ± 12.1 years versus 63.6 ± 13.1 years, respectively ( = 0.573). The mean follow-up period was 94.4±97.9 weeks for the CS group and 152.2 ± 108.4 weeks for the non-CS group.
Baseline clinical characteristics are shown in Table 1. Concerning the hemodynamic parameters, after inotropic agents and intra-aortic balloon pumping (IABP) usage, the CS group had a lower central systolic pressure (CSP) (110.8 ±  As for the baseline biochemistry, there was no significant difference between the two groups. The demographic data of the study population are shown in Table 2. ST-elevation myocardial infarction (STEMI) was more prevalent in the CS group compared with the non-CS group, and non-ST-elevation myocardial infarction was less prevalent in the CS group than in the non-CS group ( = 0.001). On the other hand, a history of hypertension was less prevalent in the CS group compared with the non-CS group ( < 0.001). In addition, patients in the CS group used P2Y12 receptor inhibitor of platelet (P2Y12 inhibitors) more frequently than those in the non-CS group ( = 0.04). By contrast, CCB and statins were less frequently used by the CS group compared with the non-CS group ( = 0.012 and = 0.04, resp.). The angiographic findings and clinical outcomes are shown in Table 3. The distributions of diseased vessels in the CS group compared with the non-CS group were single vessel disease, 34.0% versus 39.2%; dual vessel disease, 30.2% versus 35.2%; and triple vessel disease, 35.8% versus 25.6% ( = 0.285). The SYNTAX score was higher in the CS group than in the non-CS group (17.3±10.4 versus 13.1±8.0, < 0.001). Figure 1 shows the cumulative rate of freedom from recurrent MI, CV mortality, all-cause mortality, and repeated-PCI between the two groups. Freedom from CV mortality, allcause mortality, and repeated-PCI was lower in the CS group compared with the non-CS group (all < 0.001, resp.), but there was no significant difference between the two groups for recurrent MI ( = 0.305).
Medical factors predicting AMI complicated by CS are shown in Table 4. Based on the results of multivariate logistic regression analysis, SYNTAX score was the only predictor of AMI complicated by CS ( = 0.002). Furthermore, adverse predictors associated with clinical outcome in patients with AMI complicated by CS are listed in Table 5. For the CS group, history of stroke was a predictor of recurrent MI ( = 0.036), and CS, age, SYNTAX score, and diabetes were associated with CV mortality ( < 0.001, < 0.001, = 0.008, and = 0.047, resp.). On the other hand, use of beta-blockers (BBs) and angiotensin-converting enzyme inhibitor (ACEIs) could reduce CV mortality. Moreover, CS, age, SYNTAX score, and history of stroke were associated with all-cause mortality ( < 0.001, < 0.001, = 0.002, and = 0.003, resp.), whereas use of BBs and statins could reduce all-cause mortality. For repeated-PCI, CS, age, and current smoking were related risk factors ( < 0.001, = 0.018, and = 0.027, resp.), whereas usage of ACEIs could reduce the rate of repeated-PCI.

Discussion
AMI complicated by CS is one of the leading causes of death in patients hospitalized with AMI. Despite relevant progress even after invasive strategy, prognosis in this population remains poor and risk of future cardiac events is high. In this study, long-term CV mortality, all-cause mortality, and rate of repeated-PCI were higher in the CS group compared with the non-CS group. However, there was no difference between groups regarding recurrent MI. In addition, we found that Clinical factor used in analysis included sex, baseline biochemical data, angiographic findings on cardiac catheterization, exposed risk factors, and medications during admission. STEMI: ST-segment elevation myocardial infarction; Non-STEMI: non-ST-segment elevation myocardial infarction; BB: beta-blockers; ACEI: angiotensin-converting enzyme inhibitors. * Significant.
SYNTAX score was an important risk factor for patients with AMI complicated by CS. In our study, we also found that both CSP and CDP were lower in the CS group compared with the non-CS group in spite of use of inotropic agents or/and intra-aortic balloon pump (IABP). This finding was compatible with the reduced use of potent hypotensive agents such as CCB and angiotensin receptor blockers (ARB) in the CS group. Moreover, we found that hypertension was more prevalent in the non-CS group compared with the CS group.
The role of hypertension in AMI patients developing CS remains controversial. In a large observational study, the presence of hypertension in AMI patients may protect against developing CS [13], but, according to Menon and colleagues in the Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries (GUSTO) III Trial, hypertension was a predictor for developing CS in AMI patients [14]. There was no difference in terms of number or distribution of disease vessels from angiographic findings, but the left anterior descending artery (LAD) was the most common location of lesions in patients with AMI complicated by CS.
As has been reported, STEMI occurred more frequently in the CS group, but our CS group had higher SYNTAX scores than our non-CS group, which indicated that they had more complex and more severe lesion anatomy. Simple infarct-related artery (IRA) intervention in the CS group due to STEMI may have led to inadequate revascularization in these cases which could have affected long-term mortality and repeat-PCI rate. Immediate multivessel revascularization may be helpful in patients with CS due to STEMI [9] but whether this applies to all patients with AMI complicated by CS is still controversial [10].
The 30-day predictors for clinical outcomes of CS such as CS itself, DM, hypertension, previous MI, and old age have been previously studied [15]. From the results of logistic regression analysis, SYNTAX score was the only factor strongly related to developing CS in the AMI patients in our study. On the other hand, based on the results of our Cox proportional hazards model, CS, age, and SYNTAX score were risk factors associated with both long-term CV mortality and all-cause mortality. In patients with acute coronary syndrome (ACS), SYNTAX score was a significant predictor of both short-term and long-term outcomes. For STEMI patients receiving primary PCI, SYNTAX score was an independent predictor of short-term [16,17] and long-term cardiac mortality [18,19]. For NSTEMI patients receiving primary PCI, SYNTAX score was also an independent predictor of 1year major adverse cardiac events (MACE) including death, cardiac death, MI, and target vessel revascularization (TVR) [20]. Clinically, SYNTAX score should be carefully evaluated during the index catheterization. AMI patients with high SYNTAX score deserve more attention, and more aggressive revascularization should be considered in these patients.

Conclusion
Patients with AMI complicated by CS may have higher longterm mortality rates and higher repeated-PCI rates than those without CS, but we found no significant difference in the occurrence of recurrent MI between the two groups. SYNTAX Score strongly correlated with development of CS in AMI patients during initial admission and may also predict long-term mortality in AMI patients with CS.