This was a prospective, multicenter study conducted at 16 hospitals in the Liaoning Province of northeast China from May 2009 to August 2011. The 16 hospitals were: First Affiliated Hospital, China Medical University; First Affiliated Hospital, Dalian Medical University; Changtu xian People’s Hospital; Fuxin Mongolian Autonomous County People’s Hospital; Yixian People’s Hospital; Benxi Steel Company Hospital; Fushun Coal Mining Administration Hospital; Chaoyang Center Hospital; Fuxin Center Hospital; Zhuanghe Center Hospital; Wafangdian Center Hospital; Pulandian Center Hospital; Donggang Center Hospital; Dashiqiao Center Hospital; Zhangwu xian People’s Hospital; Kuandian xian Center Hospital.

This study was conducted in accordance with the declaration of Helsinki, and was conducted with approval from the Ethics Committee of China Medical University. Written informed consent was obtained from all participants.

We enrolled 479 consecutive rural female STEMI patients from May 2009 to August 2011 from all of the centers. The inclusion criteria were: (1) STEMI was diagnosed according to European Society of Cardiology (ESC) criteria [3]; (2) it was the first time STEMI was diagnosed; (3) all patients were given primary PCI treatment within 12 h or thrombolytic therapy within 6 h after symptom onset. The exclusion criteria were: (1) acute myocardial infarction patients with acute kidney injury (AKI); (2) patients undergoing dialysis treatment. (3) patients whose non-infarct-related artery was treated during primary PCI; (4) PCI was undertaken after thrombolytic therapy.

Acute renal failure was diagnosed by an increase in serum creatinine levels of 50 % or an absolute increase of ≥26.5 μmol/l in 48 h [17].

Demographic and basic clinical data were obtained from all patients, which included age, gender, body mass index and cardiovascular risk factors. Additional clinical data including clinical laboratory tests, coronary imaging data, therapeutic strategies and adverse cardiac events were collected by trained personnel.

Kidney function was measured by estimated glomerular filtration rate (eGFR), calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) Modification of Diet in Renal Disease (MDRD) equation [18] [CKD-EPI formula If female and if serum creatinine (Scr) ≤0.7 mg/dL:

If female and if Scr > 0.7 mg/dL:

Renal insufficiency (RI) was defined as eGFR <60 mL/min/1.73m²according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines [19] eGFR were calculated by the Scr examination results immediately after admission.

The patients were divided into 3 groups based on their eGFR values; the Normal group (high eGFR group, eGFR ≥ 90 mL/min/1.73 m², n = 233), the Moderate group (middle eGFR group, 60 mL/min · m² < eGFR <89 mL/min/1.73 m², n = 108) who had moderate RI, and the RI group (low eGFR group, eGFR <60 mL/min/1.73 m², n = 95).

PCI was considered to be successful if the infarct-related artery residual stenosis was <10 %, and the Thrombolysis in Myocardial Infarction (TIMI) flow level reached 3 [20]. All patients were estimated immediately after PCI.

1.5 million Units of urokinase were administrated to patients by intravenous infusion within 30 min for thrombolytic therapy.

Thrombolytic therapy was considered successful if more than 2 of 4 criteria were met [9]: ST segment resolved by ≥ 50 % by electrocardiogram (ECG) within 2 h, chest pain disappeared within 2 h; reperfusion arrhythmia occurred; serum myocardial enzyme peaks were detected in advance.

All patients received the recommended standard management for STEMI, including 300 mg loading dose of aspirin and clopidogrel after admission, aspirin, clopidogrel, low-molecular-weight heparin, beta-blockers, statins and angiotensin-converting enzyme (ACE) inhibitors/ angiotensin receptor antagonist (ARB) as appropriate. The details of the medications are included in Appendix 1.

Abnormal body mass index (BMI) was defined as BMI ≥ 25 kg/m [2, 21]. Diabetes was diagnosed by previous medical history or fasting glucose ≥ 7.0 mmol/L and/or 2 h plasma glucose level ≥11.1 mmol/l (measured after 75 g oral glucose load). Hypertension was diagnosed by previous medical history or systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg after admission. Hypercholesterolemia was diagnosed by previous medical history or low-density lipoprotein (LDL) cholesterol ≥ 2.6 mmol/L and/or non high-density lipoprotein (HDL) cholesterol ≥3.3 mmol/L. Definition of contrast-induced nephropathy (CIN) was an increase in serum creatinine ≥ 0.5 mg/dL, occurring 48 h after exposure to contrast media [22]. Current smoker was defined as smoking >300 cigarettes/year.

Major adverse cardiac events (MACE) included: death, recurrent myocardial infarction, target vessel revascularization and stroke.

Key bleeding end points were analyzed on the basis of global use of strategies to open occluded coronary arteries (GUSTO) criteria [23].

CIN, contrast-induced nephropathy was defined as either a greater than 25 % increase of serum creatinine or an absolute increase in serum creatinine of 0.5 mg/dL [24].

Patients underwent physical examination, ECG examination and fasting blood biochemical examination after admission including Scr, serum creatine kinase MB (CKMB), and cardiac troponin I (cTNI). Patients were examined for myocardial necrosis markers and by ECG once every 8 h in 72 h, then every 24 h they were examined again. Blood leukocyte counts were examined 24 h after admission and echocardiography 48 h after admission.

To ensure the standard of data collection, laboratory procedures, data management and coordination between the multiples centers involved in the study was up to our quality control all the clinicians involved in this study received uniform training before the research began.

The patients were followed up by telephone for 2 years; once per year; by the same doctor.

Categorical data were expressed with absolute numbers and percentages and analyzed using the χ ² test. Continuous data with normal distribution were described with mean and standard deviation (SD) and median and interquartile range (IQR; 25th to 75th). Data among groups were compared using ANOVA. Further Student–Newman–Keuls (SNK) analyses were performed between the three groups. Non-normally distributed data among groups were compared using rank test.

Corresponding Kaplan-Meier curves with the log-rank test were constructed. Univariate analysis of eGFR, Scr, Microalbuminuria, age, gender, diabetes, hypertension, hyperlipidemia, Killip class, heart rate (HR), ejection fraction (EF), white blood cell (WBC) counts, and medication treatment were performed to determine the predictors for mortality. Multiple Cox proportional hazard model was used to estimate associations between significant factors identified in the univariate analysis. Hazard ratios (HR) and 95 % confidence intervals (CI) were calculated and p value <0.05 was considered statistically significant. All analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA).

The mean follow-up period was 741 ± 118 days. Among the 436 individuals in the final cohort, the mean age was 67.52 years, and 35.78 % had diabetes (Table 1). Elderly patients and those with hypertension and diabetes accounted for a high proportion of patients in the RI group. Median symptom to door time, door to balloon time and door to needle time were 183, 134 and 56 min, respectively. In the three groups the mean ages were 61.11 ± 8.42 years in the Normal group, 64.08 ± 6.91 years in the Moderate group and 75.57 ± 7.53 years in the RI group with a significant difference between all of the groups (p < 0.05). The diabetes rates were 29.61 % in the Normal group, 37.96 % in the Moderate group and 48.42 % in the RI group with a significant difference between the Normal and RI groups (p < 0.05). The hypertension rates were 32.19 % in the Normal group 44.44 % in the Moderate group and 52.63 % in the RI group with a significant difference between the Normal and RI group (p < 0.05). There were higher proportions in the RI group of Killip ≥ 2 at 21.05 % compared to 9.26 % (p < 0.05) in the Moderate group and 9.01 %, (p < 0.05) in the Normal group, longer hospital stay at 12.05 ± 5.74 days compared to 8.36 ± 5.11 days (p < 0.05) in the Moderate group and 8.53 ± 4.78 days (p < 0.05) in the Normal group, low EF values at 44.38 ± 13.05 % compared to 48.92 ± 14.02 % (p < 0.05) in the Moderate group and 50.11 ± 13.60 % (p < 0.05) in the Normal group, high Scr values at 116.67 ± 59.01 mmol/l compared to 85.47 ± 44.90 mmol/l (p < 0.05) in the Moderate group and 78.12 ± 31.13 mmol/l (p < 0.05) in the Normal group, and cTnI peak at 45.33 ± 15.26 ng/ml compared to 30.19 ± 18.73 ng/ml (p < 0.05) for the Moderate group and 31.06 ± 16.28 ng/ml for the Normal group (p < 0.05) (Table 1). There were 298 patients who underwent primary PCI and 138 thrombolytic therapy.

The mortality rates of the patients in the Normal group, Moderate group and RI group were 1.72 %, 3.7 %, and 7.37 %, respectively, during hospitalization. Patients in the RI group had a significantly higher mortality rate compared with Normal group during hospitalization (p = 0.045). There were no significant differences in RMI, TVR, stroke and bleeding between the three groups. In-hospital MACE developed more frequently in the patients in the RI group compared with Normal group (p = 0.022). But there was no significant difference in the incidence of CIN (Table 2).

A similar trend was observed during 1 year of follow up after hospital discharge. Patients in the RI group had a significantly higher mortality rate compared with those in the Normal group (16.84 % v.s.4.29 %, p < 0.001), and a significantly higher MACE rate compared with the other two groups (p < 0.001) (Table 2).

During 2 years of follow up, patients in the RI group had a significantly higher mortality rate compared with the other two groups (24.21 % v.s. 6.87 % and 10.19 %, p < 0.001). There were no significant differences in recurrent myocardial infarction (RMI), target vessel revascularization (TVR), stroke and bleeding between the three groups. Compared with the RI group respectively, the Moderate group and the Normal group had higher MACE rates (36.11 % v.s. 22.32 %, 52.63 % v.s. 22.32 %, p < 0.001) (Table 2).

During 2 years of follow-up, there were 1 case of moderate bleeding and 4 cases of minor bleeding in the RI group. There were 3 cases of minor bleeding in the other two groups, respectively.

Variables were analyzed by univariate analysis for significant factors for 2 year mortality and are presented in Table 3, this suggested that eGFR of less than 90 ml/min/1.73 m², may be a predictor of 2 year mortality as both 60–90 ml/mim/ m² eGFR and eGFR <60 ml/min/1.73 m² were significant (both p > 0.001). Age ≥75 years was another significant factor (p = 0.049) as well as hypertension (p = 0.013), Killip class ≥2 (p = 0.023), and EF <40 (p = 0.025). The Kaplan–Meier survival curves are depicted in Fig. 2. The survival rate of the RI group was significantly lower than in the other two groups (log-rank test, p < 0.001).

Multivariate analysis then identified eGFR <90 ml/min/1.73 m² and age ≥75 years as independent predictors of mortality at 2 years. In detail these were eGFR 60–90 ml/min/1.73 m² with HR 2.081, 95%CI 1.250–2.842, p < 0.001; eGFR <60 ml/min/1.73 m² with HR 3.872, 95%CI 2.004–6.131, p < 0.001; age ≥75 with HR 1.461, 95%CI 1.011–1.952, p = 0.024.

All variables were also included into univariate analysis to investigate the independent predictors for in hospital mortality. The independent predictors of mortality in hospital were eGFR <60 ml/min/1.73 m² and Killip ≥ 2 (Table 4). Further analysis of the significant factors by multiple cox proportional hazard modeling identified independent predictors of mortality. This showed that the independent predictors of hospital mortality were eGFR <60 ml/min/1.73 m² [HR 1.832, 95 % CI 1.017–3.091, p = 0.033] and Killip ≥ 2 [HR 1.340, 95 % CI 1.012–1.647, p = 0.018].