The Impact of Admission Serum Creatinine Derived Estimated Glomerular Filtration Rate on Major Adverse Cardiac Events in ST-Segment Elevation Myocardial Infarction Patients Undergoing Primary Percutaneous Coronary Intervention

Background The impact of Cockroft-Gault (C-G) derived estimated glomerular filtration rate (eGFR) on mortality and major adverse cardiac events (MACEs) in patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PCI) was assessed. Methods A total of 884 patients were classified into four categories according to admission creatine derived eGFR: < 60, 60 - < 90, 90 - < 120, and ≥ 120 mL/min/1.73 m2. Results In-hospital and long-term MACEs were significantly higher in eGFR < 60 mL/min/1.73 m2 subgroup (P < 0.001 and P = 0.028). Multivariate analysis demonstrated 7.78-fold (95% CI: 0.91 - 66.8) higher mortality risk in eGFR < 60 mL/min/1.73 m2 subgroup. Conclusion As an easily applicable bedside method, C-G derived eGFR could be important for prediction of in-hospital and long-term mortality and MACE in STEMI patients undergoing primary PCI.


Introduction
Chronic kidney disease (CKD) is a significant and independent cause of cardiovascular adverse events and mortality in general population and patients with cardiovascular diseases [1][2][3][4]. Additionally, in many studies, it has shown that any degree of renal impairment assessed by estimated glomerular filtration rate (eGFR) is related to increased adverse cardiovascular outcomes and predicts in-hospital, short-and long-term mortality in acute coronary syndrome (ACS) and general population [5][6][7][8]. Decreased renal function assessed by eGFR predicts prognosis in both genders in ST-segment elevation myocardial infarction (STEMI) [9]. In STEMI patients undergoing primary percutaneous coronary intervention (PCI), poor myocardial perfusion and increased no-reflow phenomenon were detected in patients with decreased eGFR [10,11]. Two recent studies have shown relation between decreased eGFR and increased burden of atherosclerosis and decreased left ventricular ejection fraction (EF) in various forms of ACSs [12,13]. Renal dysfunction is associated with female gender, older age, hypertension (HT), and diabetes mellitus (DM) [7]. These factors are linked with worse clinical outcomes.
In the present study, we investigated the relation between admission serum creatine (Cr) derived eGFR and in-hospital and long-term adverse cardiac events in STEMI patients undergoing primary PCI.

Materials and Methods
This retrospective observational single-centered study was conducted between May 2009 and 2011. Patients presenting with STEMI undergoing primary PCI were enrolled. Patients with unstable angina, non-STEMI and STEMI that did not undergo primary PCI were excluded. The European Society Cardiology/American College of Cardiology Foundation/ American Heart Association (ESC/ACCF/AHA) committee advisement of ECG criteria was used for diagnosis of STEMI [14,15]. These were definitive new bundle branch block or > 0.1 mV new ST elevation in two contiguous leads. The proce- dures were conducted by a highly experienced staff (performing > 75 PCIs/year), in a high volume center (> 3,000 PCIs/ year). Following the diagnosis of STEMI, the blood samples were obtained from the patients at the emergency department. The eGFR was calculated from these initial blood samples.
Baseline demographic data were collected from the medical records. The left ventricular EF was measured just before discharge. The modified Simpson's method was used for measurement via a System V (Vingmed, GE, Horten, Norway) [17].
The written informed consent was obtained from all patients and hospital local ethic committee approved the study protocol.

Coronary angiography and primary PCI
All coronary angiographies and primary PCIs were performed via the femoral approach. The infarct related artery was graded by using thrombolysis in myocardial infarction (TIMI) classification [18]. The coronary artery stenosis was assumed as significant if stenosis was > 50% of diameter. The primary PCI was applied just to the infarct related artery. Following the diagnosis of STEMI, prior to primary PCI, all patients were loaded with 300 mg clopidogrel and 300 mg aspirin (unless contraindicated). Due to payment strategy of the Minister of Health, ticagrelor and prasugrel were not used. No statin loading was applied before primary PCI. After the revealing of the coronary anatomy, all patients were administered 100 U/kg heparin. Following the procedure, patients were transferred to the coronary care unit. Intravenous heparin (500 U/h) or subcutaneous low molecular heparin (enoxaparin) (1 mg/kg/day) was maintained in coronary care unit. Certain medications, e.g. aspirin 100 mg/day, clopidogrel 75 mg/day, rosuvastatin 20 mg/day or atorvastatin 40 mg/day, angiotensin converting enzyme inhibitor or angiotensin receptor blocker, and betablockers were administered according to the hemodynamic status of the patients. Glycoprotein IIb/IIIa receptor blocker (e.g. tirofiban) was used according to the burden of the thrombus on visual assessment.

Definitions
The Killip clinical examination was used for definition of patient clinical condition at presentation. HT was defined if a patient was on an antihypertensive treatment or requiring an antihypertensive drug in hospital. Any patient on an oral antidiabetic drug and/or insulin was assumed to be diabetic. Major adverse cardiac events (MACEs) consisted of cardiovascular mortality, re-infarction and target vessel revascularization (TVR). Re-infarction was defined by ST-segment elevation along with at least two-fold increase in serum creatine kinasemyocardial band (CK-MB) fraction activity.

Statistical analysis
All quantitative variables were expressed as mean ± standard deviation (SD), and qualitative variables were expressed as percent (%

Results
A total of 884 patients (113 female) undergoing primary PCI were enrolled to the study. Patients with renal dysfunction were often more female and older (P < 0.001 and P < 0.001) ( Table 1). Even though patients with eGFR < 60 mL/min/1.73 m 2 were more diabetic and hypertensive, this did not reach statistical significance (P = 0.137 and P = 0.072). The history of coronary artery disease and smoking did not differ between groups. With respect to laboratory findings, leucocyte count, CK-MB, admission glucose, and potassium levels were higher in the eGFR < 60 mL/min/1.73 m 2 group (P = 0.004, P < 0.001, P < 0.001, and P < 0.001, respectively). The left ventricular EF significantly differed between groups (P < 0.001). Patients with eGFR < 60 mL/min/1.73 m 2 group hospitalized longer (P = 0.004).
The unadjusted and adjusted models of logistic regression analysis for mortality according to eGFR are listed in Table 3. The mortality had the highest rates at eGFR < 60 mL/min/1.73 m 2 level and had 7.78 times higher mortality rates (95% CI: 0.91 -66.8) than eGFR > 120 mL/min/1.73 m 2 , which had the lowest rates as reference. After adjusted regression analysis, eGFR < 60 mL/min/1.73 m 2 still had a 2.92 times higher mortality rate (95% CI: 0.23 -36.8).

Discussion
The present study demonstrated that any degree of renal impairment is associated with increased in-hospital and longterm mortality and MACE in patients presenting with STEMI undergoing primary PCI. Also, patients with renal impairment had lower left ventricular EF and higher cardiac enzyme. On the other hand, no relation was found between stent thrombosis and renal impairment for short-and long-term period.
Serum Cr level solely did not predict the severity of renal function. Even serum Cr level could stay in normal range until moderate and severe renal impairment developed [2,7]. That is why eGFR was used for prediction of renal dysfunction. Generally, modification of diet in renal disease (MDRD) formula was used for prediction of eGFR in previous studies and in clinical practice [2,7]. In our previous study, we found that C-G derived eGFR predicts worse prognosis better than MDRD in STEMI [19]. This was a major difference between present and previous studies.
It was demonstrated that renal impairment is more frequently associated with HT and DM in addition to older age and female gender [7]. These clinical factors are associated with poor prognosis in acute MI [7]. Even though in the present study patients with renal impairment are mostly female, older and more frequently had HT and DM, after multivariate regression analysis, eGFR derived renal impairment is still prognostic in STEMI patients undergoing primary PCI.
The relation between CKD and cardiovascular mortality is well known [1][2][3]. Several different studies have shown that even mild degree of renal impairment is related with increased adverse outcomes in normal population and patients with cardiac disease. A similarly designed study with the present one shows that an eGFR < 30 mL/min/1.73 m 2 is associated with increased in-hospital and 1-year mortality [20]. In the present study, we show that long-term mortality and adverse events are sustained beyond 2 years. And also regression analysis showed that mild degree of renal impairment is related with adverse events. There are some potential explanations why renal impairment is associated with worse outcomes. Firstly, it has shown that as eGFR decreased, fibroblast and parathyroid hormone levels increased. As a result, vitamin D and phosphate metabolisms are deteriorated which leads to vascular dysfunction and coronary artery disease [6,21,22]. Secondly, potential explanation is that renal impairment is associated with enhanced pro-coagulant state, increased homocysteine levels, and endothelial dysfunction that eventually lead to atherosclerotic plaque formation and destabilization [2,23]. A recent study by Duran et al showed that mild and moderate renal impairment is associated with increased atherosclerosis burden [13]. Another important finding of our study was that as left ventricular EF decreased, eGFR also decreased. Also, CK-MB level was highest in the eGFR < 60 mL/min/1.73 m 2 group. In a recent study, Sonmez et al found that in patients with STEMI undergoing primary PCI, an eGFR < 60 mL/min/1.73 m 2 is associated with lower left ventricular EF and larger myocardial infarct size [12].
Some other findings of the present study were in accordance with the previous studies. The hospitalization duration was longer in lowest eGFR group as in Alfaleh et al's study, probably as a result of increased in-hospital adverse events [24]. Previous reports indicated that the risk of stroke increases with renal impairment [24,25]. Even though the number of stroke was low, in the present study, there was a significant inverse relation with eGFR and stroke.
In the present study, the rate of stent thrombosis did not differ significantly between groups. There are different results in previous studies in regard of different types of stents. In a study performed by Miao et al, CKD, in elective DES implantation patients, was associated with increased stent thrombosis [26]. Another study that excluded STEMI patients showed that in patients with eGFR < 60 mL/min/1.73 m 2 , DES usage is not associated with stent thrombosis but reduced myocardial infarction and rate of repeat revascularization [27]. Similar to Wang et al's study, Toutouzas et al showed that new generation DES did not increase the rate of stent thrombosis in CKD patients [28]. Another study performed with bare metal stents showed an inverse relation between renal dysfunction and neointima formation after BMS deployment. In that study the neointima intravascular ultrasound showed greater lipid volume, suggesting atherosclerosis development [29]. In the present study, BMS was deployed for stent implantation. For BMS, restenosis rather than thrombosis is a concern [30]. Probably this could be the explanation why no difference of stent thrombosis was found between groups.
The present study had several limitations. Firstly, this study was retrospective and single-centered. Secondly, just the effect of admission Cr derived eGFR was used for assessment. Thirdly, the impact of in-hospital medication on prognosis and eGFR was not assessed. Fourthly, the standard method of renal function assessment, 24 h Cr clearance, was not performed.

Conclusions
In conclusion, the present study demonstrated that any degree of renal impairment and especially eGFR < 60 mL/min/1.73 m 2 was associated with increased in-hospital and long-term mortality and MACE and decreased left ventricular EF. C-G derived eGFR is a simple, easily bedside applicable method of renal impairment for risk stratification in STEMI patients undergoing primary PCI.