This was a retrospective clinical study to review the patients of acute STEMI underwent primary PCI (n = 664) in our hospital from Jul 2011 to Oct 2013. Baseline 12-lead electrocardiograms were performed at admission. A total of 553 patients were included in this study, while 111 patients were excluded according to follow exclusion criteria. The inclusion criteria were: (1) patients with 18 to 85 years of age; (2) diagnosed as acute STEMI; (3) underwent primary coronary intervention within 12 h after chest pain on-set. The exclusion criteria were as follows: (1) incomplete clinical history record (n = 23); (2) excluded the diagnosis of myocardial infarction by angiography, such as viral myocarditis, pericarditis or cardiomyopathy (n = 26); (3) not finish the detection of echocardiography in-hospitalization because of any reason (n = 17); (4) confirmed clinical heart failure before this admission, complicated with cardiomyopathy, congenital heart diseases and rheumatic heart disease (n = 19); (5) active chronic inflammation (n = 14); (6) dysfunction of hematological and immunological system (n = 4); (7) carcinoma or a condition treated with immunosuppressive agents (n = 8). This study and consent procedure were approved by our local ethics committee (Ethics Committee of Zhongshan Hospital affiliated to Fudan University), and were carried out in accordance with the principles of the Declaration of Helsinki. Consent for publication of these data was obtained from each patient when they were admitted in our hospital.

Several important clinical variables were record, such as age, gender, hypertension, diabetes, stable angina history, Time to hospital (from chest pain on-set to diagnosis) and D-to-B time (door to balloon).

Prior to primary PCI, all patients received adequate loading doses of aspirin (300 mg) and clopidogrel (300 mg) or ticagrelor (180 mg) immediately after diagnosed as STEMI. Procedure of PCI was performed immediately via the femoral or radial access route. The characteristics of coronary angiography were record, such as culprit artery, acute occlusive segment and multi-vessel disease (defined as having at least another vessel with 75 % or greater stenosis except the culprit occlusion artery, such as culprit vessel in LAD had LCX or RCA or LM stenosis, culprit vessel in LCX had RCA or LAD or LM stenosis, culprit vessel in RCA had LCX or LAD or LM stenosis). A lesion was considered proximal if it was located proximal to the first diagonal branch in the LAD, the first obtuse marginal branch in LCX, or the first acute marginal branch in RCA [12].

The usages of interventional techniques, thrombus aspiration and platelet glycoprotein IIb/IIIa receptor inhibitor were chosen at the operators’ discretion. The phenomenon of no reflow or slow flow post-stenting was record. Post-PCI blood pressure, including systolic blood pressure (SBP) and diastolic blood pressure (DBP), was detected by invasive blood pressure monitor from radial or femoral vascular sheaths.

Echocardiography was performed before discharge (4–6 days post-PCI) in all patients using a Philips IE33 instrument (Philips, Netherlands) with a 2–3.5 MHz transducer (X3-1), while left ventricular ejection fraction (LVEF) were detected by Simpson method. It was defined as reduced LVEF while LVEF less than 55 %. All exams were performed by one of three echocardiography operators. These three operators underwent standardized training before this study. Observers who detected LVEF were blinded to the results of coronary angiography and clinical record. The incidence of reduced LVEF after STEMI was compared between two different culprit artery systems, including LAD related STEMI (LM occlusion and LAD occlusion, n = 315) and non-LAD related STEMI (LCX occlusion, n = 63; RCA occlusion, n = 175).

All statistical analyses were performed with SPSS software 19.0. Data were presented as the percentage or mean ± standard deviation (SD). Chi-square analysis was used to compare the frequency for categorical variables, and Student’s t or correction t tests were used to compare means for continuous variables. Multivariable logistic analysis was performed to identify the independent risk factors for reduced LVEF (LVEF < 55 %). Stratification according to different risk subsets was also made by classification and regression tree (CART) analysis. All P-values were two-sided, and P < 0.05 was considered to indicate statistical significance.

A total of 553 STEMI patients enrolled with average age 64.0 ± 12.0 years. There were 447 men (80.8 %) and 106 women (19.2 %). The prevalence of hypertension and diabetes were 60.4 % (334 patients) and 48.8 % (270 patients), respectively. Culprit arteries of STEMI were 6 in LM, 309 in LAD, 63 in LCX and 175 in RCA, respectively. Baseline clinical and angiographic characteristics of patients with different culprit arteries were shown in Table 1. Compared with the non-LAD related STEMI (culprit arteries were RCA and LCX), patients of LAD related STEMI (culprit arteries were LM and LAD) had lower LVEF (52.4 ± 9.3 % vs. 57.1 ± 7.8 %, P < 0.01) and higher incidence of reduced LVEF (LVEF < 55 %: 53.7 and 26.9 %, P < 0.01).

In order to clarify the predictor difference for reduced LVEF in LAD system and non-LAD system groups, several clinical and angiographic predictors were analyzed by univariate analysis, shown in Table 1 and Fig. 1. It was demonstrated that elder (more than 65 years), time to hospital (from chest pain on-set to diagnosis), acute occlusion in proximal segment and post-PCI blood pressure significantly increased the risk of reduced in LAD system. However, in non-LAD related STEMI patients, beside the factors of age and time to hospital, multivessel stenosis significantly increased the risk of reduced LVEF. In order to further clarify the influence of post-PCI blood pressure, the occurrence of reduced LVEF was analyzed among four groups classified by the quartile of post-PCI blood pressure in patients with LAD system or non-LAD system STEMI, shown in Fig. 2. We found that lower SBP and DBP after primary PCI predicted the higher risk of reduced LVEF.

Multivariate logistic analysis was performed to demonstrate the independent effect of these predictors (confirmed statistic difference in univariate analysis) on the occurrence of reduced LVEF after STEMI, In this analysis, reduced LVEF (LVEF < 55 %) was employed as a dependent variable in both LAD and non-LAD system subgroups, while age > 65 years, multi-vessel stenosis, acute occlusion in proximal segment, time to hospital, post-PCI SBP <100 mmHg and post DBP <65 mmHg were set as independent variables, shown in Table 2. These results demonstrated that elder (OR = 1.984, 95 % CI = 1.205–3.266, P < 0.01), proximal occlusion (OR = 1.681, 95 % CI = 1.042–2.713, P = 0.033) and time to hospital (OR = 1.106, 95 % CI = 1.010–1.210, P = 0.029) were major independent predictors for reduced LVEF in LAD system, while time to hospital (OR = 1.246, 95 % CI = 1.097–1.414, P < 0.01), multi-vessel stenosis (OR = 2.394, 95 % CI = 1.185–4.836, P = 0.015) and post-PCI SBP < 100 mmHg (OR = 2.927, 95 % CI = 1.052121–7.643, P = 0.028) in non-LAD system.

In order to confirm the impact of predictors on reduced LVEF and simply the prediction process, CART analysis was also applied to assess the incidence of reduced LVEF after STEMI in multivariate subgroups. Reduced LVEF was employed as a dependent variable, while age > 65 years, male gender, stable angina history, diabetes, hypertension, culprit vessel (LAD system or non-LAD system), time to hospital >5 h, multi-vessel stenosis, occlusion in proximal segment, slow or no reflow, post-PCI SBP <100 mmHg and post DBP < 65 mmHg were set as independent variables. CART analysis results were shown in Fig. 3. We found that LAD system was the major determinant of reduced LVEF after STEMI. Beside culprit artery, elder, time to hospital > 5 h and proximal occlusion were most critical three steps in risk stratification for reduced LVEF in LAD system, while time to hospital and post-PCI diastolic blood pressure < 60 mmHg in non-LAD system.