Original research
289
Relation of red cell distribution width to contrast-induced acute kidney injury in patients undergoing a primary percutaneous coronary intervention Fatih Akina, Omer Celikb, Ibrahim Altuna, Burak Aycac, Derya Ozturkb, Seckin Satilmisd, Ahmet Ayazb and Omer Tasbulakb Background and aim We investigated the utility of the preprocedural red cell distribution width (RDW) for predicting contrast-induced acute kidney injury (CI-AKI) in patients with ST-segment-elevation myocardial infarction (STEMI) who underwent a primary percutaneous coronary intervention. Materials and methods A total of 630 consecutive patients who were routinely referred to coronary angiography for STEMI were included in the present study. Results CI-AKI was observed in 79 patients (12.5%). The RDW, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and the mean platelet volume were significantly higher in the CI-AKI group than in the non-CI-AKI group (P < 0.001, P = 032, P = 0.025, and P = 0.039, respectively). Serum total bilirubin and direct bilirubin levels were not different among the study groups. Using multivariate logistic regression analysis, we found that left ventricular ejection fraction [odds ratio (OR) = 0.972, 95% confidence interval (CI) 0.945–0.998, P = 0.033], estimated glomerular filtration rate (OR = 0.970, 95% CI 0.959–0.981, P < 0.001), contrast volume (OR = 1.007, 95% CI 1.002–1.012, P = 0.009), and
Introduction Contrast-induced acute kidney injury (CI-AKI) is a common cause of acute renal failure, especially after a primary percutaneous coronary intervention (PPCI) [1]. The incidence of CI-AKI ranges from 5 to 20% among patients who undergo coronary angiography [2–4]. The patients who have undergone PPCI are at a high risk for CI-AKI and the incidence of CI-AKI is increased up to 20% after PPCI [4]. CI-AKI after a PPCI is related to worse clinical outcomes such as myocardial infarction, repeat revascularization, and end-stage renal failure [3,5, 6]. Moreover, CI-AKI is related to increased in-hospital morbidity and mortality rates [3,4]. Early risk estimation for the development of CI-AKI is crucial in preventing its development. The red cell distribution width (RDW) is a measure of variation in the size of circulating red blood cells, and is reported routinely as part of an automated full blood count [7]. Recent studies have shown a strong independent association between increased RDW and poor prognoses in patients with stable coronary disease [8], 0954-6928 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
RDW (OR = 1.406, 95% CI 1.120–1.792, P = 0.005) were independent predictors of CI-AKI. Conclusion Red blood cell distribution width, an inexpensive and easily measurable laboratory variable, is associated independently with the development of CI-AKI. Our data suggest that RDW may be a useful marker in CI-AKI risk stratification. Coron Artery Dis 26:289–295 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. Coronary Artery Disease 2015, 26:289–295 Keywords: contrast-induced acute kidney injury, in-hospital mortality, primary percutaneous coronary intervention, red cell distribution width a Department of Cardiology, Muğla Sıtkı Kocman University School of Medicine, Muğla, bDepartment of Cardiology, Mehmet Akif Ersoy Chest and Cardiovascular Surgery Education and Research Hospital, cDepartment of Cardiology, Bağcılar Education and Research Hospital and dDepartment of Cardiology, Acıbadem University School of Medicine, Istanbul, Turkey
Correspondence to Fatih Akin, MD, Department of Cardiology, Muğla Sıtkı Kocman University School of Medicine, Muğla 48000, Turkey Tel: + 90 506 627 2903; fax: + 90 212 471 9494; e-mail: [email protected] Received 19 October 2014 Revised 25 December 2014 Accepted 4 January 2015
heart failure [9], and ST-segment-elevation myocardial infarction (STEMI) treated with PPCI [10]. RDW has also been found to be associated with kidney function [11]. In a recent study, RDW has been linked to CI-AKI in patients with acute coronary syndrome [12]. Although the association between RDW and several cardiovascular diseases is well known, its relation with CI-AKI in patients with STEMI undergoing PPCI remains unclear. The neutrophil-to-lymphocyte (N/L) ratio provides a simple but promising method to evaluate systemic inflammation and it is used widely as a prominent marker for cardiovascular diseases [13]. Recently, the N/L ratio has been linked to CI-AKI in patients undergoing PPCI [14]. Platelet function disorder has also been implicated in the pathogenesis of STEMI. Platelet indices such as the mean platelet volume (MPV) and platelet-to-lymphocyte (P/L) ratio are associated with atherosclerosis and inflammation [15,16]. MPV and P/L ratio levels are higher in patients with STEMI and have been linked to increased mortality in that setting [15,17]. DOI: 10.1097/MCA.0000000000000223
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
290 Coronary Artery Disease 2015, Vol 26 No 4
Recently, many strategies have been studied to prevent CI-AKI development, but, unfortunately, no medication or strategy has been proven to effectively prevent the development of CI-AKI. Careful patient assessment is the most effective strategy to reduce the incidence of CIAKI. Therefore, in this study, we aimed to determine the association of RDW with CI-AKI in patients undergoing PPCI. Furthermore, we hypothesized that RDW and CIAKI would also be associated with in-hospital mortality in these patients.
Methods Study population
A total of 630 consecutive patients who were routinely referred to coronary angiography for acute myocardial infarction (AMI) were included in the present crosssectional study after the following exclusions: patients with cardiogenic shock, hypotension (< 90/60 mmHg), and those who need inotropic therapy, receiving longterm peritoneal, or hemodialysis treatment, undergoing cardiac surgery for emergency coronary revascularization or STEMI with mechanical complications, and contrast medium exposure within 10 days. Inclusion criteria of the study were patients who presented within 12 h of the onset of symptoms with typical, prolonged chest pain at rest (>30 min), not responsive to nitrates, with ECG STsegment elevation of at least 0.2 mV in two or more contiguous leads, or left bundle-branch block. Patients were recruited between January 2011 and August 2013. An absolute 0.3 mg/dl or more increase in serum creatinine (SCr) levels compared with baseline levels within 48 h after administration of contrast medium was considered as CI-AKI. Patients were divided into two groups. Patients without CI-AKI (group-1) were compared with patients developing CI-AKI (group-2). Because of the urgent setting of the patients with baseline renal impairment [estimated glomerular filtration rate (eGFR) < 60 ml/min], preprocedural hydration therapy could not be applied, but postprocedural hydration therapy was applied until 12 h after a PCI. The study protocol was approved by the local ethics committee.
chewable 300 mg aspirin and clopidogrel 600 mg loading dose before and at the time of coronary angiography. A nonionic low osmolarity contrast medium was used in all patients. The artery that was presumed to be nonobstructed was injected first. All patients received heparin (100 IU/kg) when the coronary anatomy was defined. Angiographic evaluation was performed by visual assessment. At enrollment, only the infarct-related artery (IRA) was revascularized in patients with multivessel disease. After primary angioplasty and/or stenting, a successful procedure was defined as a regression under 30% of obstruction of the IRA with TIMI-3 flow. After the procedure, all patients were admitted to the coronary care unit. The use of glycoprotein IIb/IIIa inhibitors was left to the discretion of the operators. Follow-up and endpoints
In all cases, blood samples were drawn at admission before catheterization procedures. A 12-lead ECG was recorded for each patient at hospital presentation, and the localization of myocardial infarction and IRA were determined from the ECGs. Echocardiographic analyses were carried out within 24–48 h of revascularization. Biplane Simpson’s method was used for the measurement of left ventricular ejection fraction (LVEF) [18]. eGFRs were determined using eGFRs that were calculated using the modification of diet in renal disease [19]. AKI was defined according to the Kidney Disease Improving Global Outcomes criteria [20]. The N/L ratio was obtained by dividing the total count of neutrophils by the lymphocytes count. The P/L ratio was obtained by dividing the total count of platelets by the lymphocyte count. SCr levels were measured at admission before angiography and daily after the procedure until discharge. An absolute 0.3 mg/dl or more increase in SCr levels compared with baseline levels within 48 h after administration of contrast medium was considered as CI-AKI. Data of in-hospital mortality rates of the patients were obtained from the medical records of hospitals.
Definitions
Statistical analysis
Hypertension was defined as a blood pressure of 140/90 mmHg or greater or having a history of antihypertensive drug use. Diabetes mellitus (DM) was defined as a fasting blood glucose of at least 126 mg/dl on two occasions or as being on treatment. Present smokers were defined as having a history of smoking within the past year. Admission anemia was defined as a baseline hemoglobin (Hb) level less than 12 mg/dl in women and less than 13 mg/dl in men. Chronic kidney disease was defined as an eGFR less than 60 ml/min/1.7 m2.
All analyses were carried out using SPSS 20.0 (released 2011, IBM statistics for Windows, version 20; IBM Corp., Armonk, New York, USA), MedCalc version 11.3 (MedCalc Software, Mariakerke, Belgium), and Matlap (7.5.0 r2007b; The Mathworks, Natick, Massachusetts, USA). All data are presented as mean ± SD unless otherwise stated. Comparison of parametric values between the two groups was performed using an independent-samples t-test. Comparisons of nonparametric values between the two groups were performed using the Mann–Whitney U-test. Categorical variables were compared using the χ2-test. Logistic regression analysis was used to assess predictors of CI-AKI and inhospital mortality. Those variables with P less than 0.1 by
Coronary angiography, angioplasty, and stenting
Emergency coronary angiography was performed using the standard Judkins technique. All patients were given a
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Red cell distribution width, nephropathy Akin et al. 291
univariate analysis were included in the multivariate logistic regression analysis model and the respective odds ratios (OR) with 95% confidence intervals (CI) were calculated. The usefulness of the RDW value in predicting the presence of CI-AKI was analyzed using receiver operating characteristics (ROC) curve analyses. When a significant cut-off value was observed, the sensitivity and specificity values were presented. The Delong method was used to compare area under the curve (AUC). In addition, the contribution of RDW was analyzed using the net reclassification improvement calculator. All statistical tests were two-sided, and statistical significance was determined at a P value less than 0.05.
Results The mean age of the patients studied was 56.7 ± 12 years and 19% of the patients were women. A total of 79 of 630 (12.5%) patients experienced CI-AKI. Demographic and clinical patient characteristics in the CI-AKI and non-AKI groups are listed in Table 1. The patients in CI-AKI group were older than the patients in the non-AKI group, the incidence of hypertension was higher in the CI-AKI group than in the non-AKI group, creatinine levels were significantly higher, the levels of eGFR were lower, and Baseline clinical and laboratory characteristics of CI-AKI and non-CI-AKI groups
Table 1
Age (years) Male sex [n (%)] Hypertension [n (%)] Diabetes [n (%)] Smoking [n (%)] Chronic kidney disease [n (%)] Previous history of MI [n (%)] Previous history of CABG [n (%)] Hemoglobin (g/dl) Mean platelet volume (fl) RDW (%) Neutrophil count (109/l) N/L ratio P/L ratio Fasting glucose (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglyceride (mg/dl) Creatinine (mg/dl) eGFR (ml/min/1.73 m2) Total bilirubin (mg/dl) Direct bilirubin (mg/dl) Left ventricular EF Multivessel coronary disease [n (%)] Contrast volume Infract-related artery [n (%)] Left anterior descending Left circumflex Right coronary
No CI-AKI (n = 551)
CI-AKI (n = 79)
P value
55.7 ± 11.7 448 (81.3) 187 (33.9) 135 (24.5) 200 (36.3) 41 (7.4)
63.6 ± 12.3 62 (78.5) 45 (57) 27 (34.2) 30 (37.5) 32 (40.5)
< 0.001 0.542 < 0.001 0.074 0.589 < 0.001
101 (18.3) 17 (3.1)
18 (22.8) 5 (6.3)
13.9 ± 1.7 8.3 ± 1 13.1 ± 1 10.2 ± 12.8 6.2 ± 5.7 158 ± 108.4 133.3 ± 56.8 126.3 ± 37.6 41 ± 22.5 130 ± 95.6 0.9 ± 0.4 98.2 ± 28.5 0.38 ± 0.52 0.18 ± 0.43 49.7 ± 8.3 251 (45.6)
13.4 ± 1.6 8.6 ± 1.1 13.7 ± 1.3 10.3 ± 3.7 6.7 ± 4.9 187 ± 114.5 149.7 ± 75.3 119.7 ± 38.3 36.2 ± 13.4 125.8 ± 120.7 1.5 ± 1.4 67.5 ± 30.1 0.36 ± 0.35 0.13 ± 0.14 45 ± 11.1 48 (60.8)
141.8 ± 49
153.9 ± 49
295 (53.5) 113 (20.5) 143 (26)
0.357 0.179 0.031 0.039 < 0.001 0.969 0.032 0.025 0.087 0.133 0.070 0.714 < 0.001 < 0.001 0.791 0.632 0.001 0.016 0.042 0.541
38 (48.1) 16 (20.3) 25 (31.6)
CABG, coronary artery bypass grafting; CI-AKI, contrast-induced acute kidney injury; EF, ejection fraction; eGFR, estimated glomerular filtration rate; MI, myocardial infarction; N/L ratio, neutrophil/lymphocyte ratio; P/L ratio, platelet/lymphocyte ratio; RDW, red cell distribution width.
LVEF at the admission was also lower in the CI-AKI group compared with the non-CI-AKI group (Table 1). The RDW, N/L ratio, P/L ratio, and MPV were significantly higher in the CI-AKI group than in the non-CIAKI group (Table 1). The volume of contrast administered during the procedure was also higher in the CI-AKI group. In the angiographic parameters, there were no differences in IRA between two groups, but in the AKI group, there was a higher incidence of multivessel disease than that in the non-AKI group. There was no difference in sex, presence of DM, lipid profiles – including triglyerides to HDL cholesterol – total bilirubin, and direct bilirubin between the CI-AKI and the non-CI-AKI groups. In all patients, RDW was correlated positively with the N/L ratio (r = 0.099, P = 0.013) and the P/L ratio (r = 0.237, P = 0.001). There was also a mildly significant inverse association between RDW level and eGFR (r = − 0.118, P = 0.003) (Fig. 1). Significant univariate predictors of CI-AKI were age, hypertension, LVEF, Hb, eGFR, MPV, RDW, N/L ratio, and multivessel disease. Using multivariate logistic regression analysis, we found that LVEF (OR = 0.972, 95% CI 0.945–0.998, P = 0.033), eGFR (OR = 0.970, 95% CI 0.959–0.981, P < 0.001), contrast volume (OR = 1.007, 95% CI 1.002–1.012, P = 0.009), and RDW (OR = 1.406, 95% CI 1.120–1.792, P = 0.005) were independent predictors of CI-AKI. The results of multivariate logistic regression analysis are presented in Table 2. For in-hospital mortality, age, hypertension, RDW, eGFR, CI-AKI, LVEF, N/L ratio, and multivessel coronary disease were included in the multivariate logistic regression model. In the multiple logistic regression analysis, only LVEF (OR = 0.891, 95% CI 0.852–0.932, P < 0.001) and CI-AKI (OR = 2.949, 95% CI 1.139–7.636, P = 0.026) were found to be independent correlates of inhospital mortality (Table 3). Using ROC comparison analysis, eGFR showed superior predictive value than RDW, ejection fraction, and contrast volume for the prediction of CI-AKI. We also observed that RDW had a significantly larger AUC than that of contrast volume (0.663, 95% CI 0.601–0.725 vs. 0.582, 95% CI 0.540–0.624; P < 0.01) for the prediction of CI-AKI (Table 4). The ROC analysis yielded a cut-off value of 13.25 for RDW to predict CI-AKI with 62% sensitivity and 64% specificity (AUC = 0.663, 95% CI = 0.601–0.725, P < 0.001) (Fig. 2). Higher RDW nonsignificantly improved the discriminatory ability (AUC from 0.786 ± 0.031 to 0.799 ± 0.030, P = 0.345) of the baseline clinical model, but did not improve the NRI index (1.251, P = 0.426).
Discussion In this study, we found a significant association between serum RDW levels and the development of CI-AKI in patients with STEMI undergoing primary PPCI. In the
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
292 Coronary Artery Disease 2015, Vol 26 No 4
Fig. 1
RDW (%)
(a)
(b) 22.00
22.00
20.00
20.00
18.00
18.00
16.00
16.00
14.00
14.00
12.00
12.00 R Linear = 0.01
R2 Linear = 0.056
2
10.00
10.00 0.00
20.00
40.00 N/L ratio
60.00
0.00
200.00
400.00 600.00 P/L ratio
800.00
1000.00
(c) 22.00
20.00
RDW (%)
18.00
16.00
14.00
12.00 R2 Linear = 0.014 10.00 0.00
50.00
100.00
150.00
200.00
250.00
eGFR (ml/min/1.73 m2) Correlation between RDW and the N/L ratio (a). Correlation between RDW and the P/L ratio (b). Correlation between RDW and eGFR (c). eGFR, estimated glomerular filtration rate; N/L ratio, neutrophil-to-lymphocyte ratio; P/L ratio, platelet-to-lymphocyte ratio; RDW, red cell distribution width.
present study, the incidence of CI-AKI was 13%, which was in agreement with previous reports. In addition, CIAKI was an independent predictor of in-hospital mortality in multivariate models. Recent studies have reported an independent association between RDW and intrahospital and long-term outcomes in patients with AMI [10,21]. Ihan et al. [21] enrolled 763
patients with AMI undergoing a PPCI and documented these patients’ intrahospital mortality from hospital registries. In their analysis, RDW was associated with intrahospital cardiovascular mortality. Uyarel et al. [22] investigated the association between RDW and inhospital and long-term cardiovascular mortality in 2506 STEMI patients. Their study suggests that RDW could predict mortality among PPCI patients. In our study, we
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Red cell distribution width, nephropathy Akin et al. 293
Variables Age Hypertension Left ventricular EF Hemoglobin eGFR Mean platelet volume RDW N/L ratio Multivessel coronary disease Contrast volume
Odds ratio
95% confidence interval
P
1.013 1.331 0.972 1.096 0.970 1.267 1.406 1.014 1.359 1.007
0.988–1.041 0.743–2.246 0.945–0.998 0.942–1.275 0.959–0.981 0.980–1.640 1.120–1.792 0.971–1.060 0.781–2.372 1.002–1.012
0.320 0.337 0.033 0.236 < 0.001 0.070 0.005 0.525 0.285 0.009
EF, ejection fraction; eGFR, estimated glomerular filtration rate; N/L ratio, neutrophil/lymphocyte ratio; RDW, red cell distribution width.
Age Hypertension Left ventricular EF CI-AKI eGFR RDW N/L ratio Multivessel coronary disease
1.0
0.8
0.6
0.4 AUC : 0.663 95% CI : 0.601–0.725
Independent predictors of in-hospital mortality in multivariate logistic regression analysis
Table 3
Variables
Fig. 2
Sensitivity
Independent predictors of contrast-induced acute kidney injury in multivariate logistic regression analysis
Table 2
0.2
Odds ratio
95% confidence interval
P
1.020 1.645 0.891 2.949 0.987 0.967 1.033 2.380
0.981–1.060 0.649–4.167 0.852–0.932 1.139–7.636 0.971–1.003 0.704–1.328 0.975–1.094 0.915–6.195
0.319 0.294 < 0.001 0.026 0.116 0.835 0.273 0.076
0.0 0.0
0.2
0.4 0.6 1 − specificity
0.8
1.0
Receiver-operating characteristic curve analysis of RDW for the prediction of contrast-induced acute kidney injury. AUC, area under the curve; CI, confidence interval; RDW, red cell distribution width.
CI-AKI, contrast-induced acute kidney injury; EF, ejection fraction; eGFR, estimated glomerular filtration rate; N/L ratio, neutrophil/lymphocyte ratio; RDW, red cell distribution width.
investigated the association between RDW and CI-AKI and in-hospital mortality. In our cohort, RDW was associated with CI-AKI, but not with in-hospital mortality. It was also reported that CI-AKI is associated with a higher in-hospital complication rate and mortality in patients undergoing PPCI [23]. Consistently, we found a significant association between CI-AKI and in-hospital mortality. Our study suggests that the association between RDW and in-hospital mortality among PPCI patients can be explained by CI-AKI. To the best of our knowledge, the study by Kurtul et al. [12] is the only study in the literature determining the association between RDW and CI-AKI. They investigated the association between RDW and CI-AKI in heterogeneous cardiac disorders including STEMI, nonSTEMI, and unstable angina pectoris. Our study included patients with STEMI undergoing PPCI – a high-risk group for the development of CI-AKI – we confirm and Table 4
extend the previous findings. Their study included 673 patients with acute coronary syndrome; 424 of these patients had STEMI. Our sample size of STEMI patients was larger and thus more reflective of CI-AKI risk in patients undergoing PPCI. We also investigated the utility of the preprocedural RDW for in-hospital mortality in patients STEMI who underwent PPCI. Mehran et al. [24] developed a simple risk score of CIAKI after PCI. They suggested a CI-AKI risk stratification score on the basis of eight variables including age, chronic kidney disease, hypotension, DM, chronic congestive heart failure, anemia, and contrast volume. In our study, RDW was an independent predictor of CI-AKI and it showed superior predictive value than contrast volume. Our study suggests that RDW may advance the prediction of CI-AKI. The underlying pathophysiological mechanisms of CIAKI are not known. Implicated mechanisms for the
Univariate predictors of contrast-induced acute kidney injury
Parameters
AUC
95% CI
P
Sensitivity (%)
Specificity (%)
Youden index J
eGFR LVEF RDW CV
0.763 0.611 0.663 0.582
0.728–0.796 0.571–0.648 0.601–0.725 0.540–0.624
< 0.001 0.003 < 0.001 0.009
68.4 53.2 62.8 67.4
75.5 68.2 64.7 52.3
0.438 0.209 0.278 0.208
P(LVEF − RDW) = 0.199, P(LVEF − eGFR) = 0.001, P(RDW − eGFR) = 0.027, P(RDW − CV) = 0.047, P(eGFR − CV)
289
Relation of red cell distribution width to contrast-induced acute kidney injury in patients undergoing a primary percutaneous coronary intervention Fatih Akina, Omer Celikb, Ibrahim Altuna, Burak Aycac, Derya Ozturkb, Seckin Satilmisd, Ahmet Ayazb and Omer Tasbulakb Background and aim We investigated the utility of the preprocedural red cell distribution width (RDW) for predicting contrast-induced acute kidney injury (CI-AKI) in patients with ST-segment-elevation myocardial infarction (STEMI) who underwent a primary percutaneous coronary intervention. Materials and methods A total of 630 consecutive patients who were routinely referred to coronary angiography for STEMI were included in the present study. Results CI-AKI was observed in 79 patients (12.5%). The RDW, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and the mean platelet volume were significantly higher in the CI-AKI group than in the non-CI-AKI group (P < 0.001, P = 032, P = 0.025, and P = 0.039, respectively). Serum total bilirubin and direct bilirubin levels were not different among the study groups. Using multivariate logistic regression analysis, we found that left ventricular ejection fraction [odds ratio (OR) = 0.972, 95% confidence interval (CI) 0.945–0.998, P = 0.033], estimated glomerular filtration rate (OR = 0.970, 95% CI 0.959–0.981, P < 0.001), contrast volume (OR = 1.007, 95% CI 1.002–1.012, P = 0.009), and
Introduction Contrast-induced acute kidney injury (CI-AKI) is a common cause of acute renal failure, especially after a primary percutaneous coronary intervention (PPCI) [1]. The incidence of CI-AKI ranges from 5 to 20% among patients who undergo coronary angiography [2–4]. The patients who have undergone PPCI are at a high risk for CI-AKI and the incidence of CI-AKI is increased up to 20% after PPCI [4]. CI-AKI after a PPCI is related to worse clinical outcomes such as myocardial infarction, repeat revascularization, and end-stage renal failure [3,5, 6]. Moreover, CI-AKI is related to increased in-hospital morbidity and mortality rates [3,4]. Early risk estimation for the development of CI-AKI is crucial in preventing its development. The red cell distribution width (RDW) is a measure of variation in the size of circulating red blood cells, and is reported routinely as part of an automated full blood count [7]. Recent studies have shown a strong independent association between increased RDW and poor prognoses in patients with stable coronary disease [8], 0954-6928 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
RDW (OR = 1.406, 95% CI 1.120–1.792, P = 0.005) were independent predictors of CI-AKI. Conclusion Red blood cell distribution width, an inexpensive and easily measurable laboratory variable, is associated independently with the development of CI-AKI. Our data suggest that RDW may be a useful marker in CI-AKI risk stratification. Coron Artery Dis 26:289–295 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. Coronary Artery Disease 2015, 26:289–295 Keywords: contrast-induced acute kidney injury, in-hospital mortality, primary percutaneous coronary intervention, red cell distribution width a Department of Cardiology, Muğla Sıtkı Kocman University School of Medicine, Muğla, bDepartment of Cardiology, Mehmet Akif Ersoy Chest and Cardiovascular Surgery Education and Research Hospital, cDepartment of Cardiology, Bağcılar Education and Research Hospital and dDepartment of Cardiology, Acıbadem University School of Medicine, Istanbul, Turkey
Correspondence to Fatih Akin, MD, Department of Cardiology, Muğla Sıtkı Kocman University School of Medicine, Muğla 48000, Turkey Tel: + 90 506 627 2903; fax: + 90 212 471 9494; e-mail: [email protected] Received 19 October 2014 Revised 25 December 2014 Accepted 4 January 2015
heart failure [9], and ST-segment-elevation myocardial infarction (STEMI) treated with PPCI [10]. RDW has also been found to be associated with kidney function [11]. In a recent study, RDW has been linked to CI-AKI in patients with acute coronary syndrome [12]. Although the association between RDW and several cardiovascular diseases is well known, its relation with CI-AKI in patients with STEMI undergoing PPCI remains unclear. The neutrophil-to-lymphocyte (N/L) ratio provides a simple but promising method to evaluate systemic inflammation and it is used widely as a prominent marker for cardiovascular diseases [13]. Recently, the N/L ratio has been linked to CI-AKI in patients undergoing PPCI [14]. Platelet function disorder has also been implicated in the pathogenesis of STEMI. Platelet indices such as the mean platelet volume (MPV) and platelet-to-lymphocyte (P/L) ratio are associated with atherosclerosis and inflammation [15,16]. MPV and P/L ratio levels are higher in patients with STEMI and have been linked to increased mortality in that setting [15,17]. DOI: 10.1097/MCA.0000000000000223
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
290 Coronary Artery Disease 2015, Vol 26 No 4
Recently, many strategies have been studied to prevent CI-AKI development, but, unfortunately, no medication or strategy has been proven to effectively prevent the development of CI-AKI. Careful patient assessment is the most effective strategy to reduce the incidence of CIAKI. Therefore, in this study, we aimed to determine the association of RDW with CI-AKI in patients undergoing PPCI. Furthermore, we hypothesized that RDW and CIAKI would also be associated with in-hospital mortality in these patients.
Methods Study population
A total of 630 consecutive patients who were routinely referred to coronary angiography for acute myocardial infarction (AMI) were included in the present crosssectional study after the following exclusions: patients with cardiogenic shock, hypotension (< 90/60 mmHg), and those who need inotropic therapy, receiving longterm peritoneal, or hemodialysis treatment, undergoing cardiac surgery for emergency coronary revascularization or STEMI with mechanical complications, and contrast medium exposure within 10 days. Inclusion criteria of the study were patients who presented within 12 h of the onset of symptoms with typical, prolonged chest pain at rest (>30 min), not responsive to nitrates, with ECG STsegment elevation of at least 0.2 mV in two or more contiguous leads, or left bundle-branch block. Patients were recruited between January 2011 and August 2013. An absolute 0.3 mg/dl or more increase in serum creatinine (SCr) levels compared with baseline levels within 48 h after administration of contrast medium was considered as CI-AKI. Patients were divided into two groups. Patients without CI-AKI (group-1) were compared with patients developing CI-AKI (group-2). Because of the urgent setting of the patients with baseline renal impairment [estimated glomerular filtration rate (eGFR) < 60 ml/min], preprocedural hydration therapy could not be applied, but postprocedural hydration therapy was applied until 12 h after a PCI. The study protocol was approved by the local ethics committee.
chewable 300 mg aspirin and clopidogrel 600 mg loading dose before and at the time of coronary angiography. A nonionic low osmolarity contrast medium was used in all patients. The artery that was presumed to be nonobstructed was injected first. All patients received heparin (100 IU/kg) when the coronary anatomy was defined. Angiographic evaluation was performed by visual assessment. At enrollment, only the infarct-related artery (IRA) was revascularized in patients with multivessel disease. After primary angioplasty and/or stenting, a successful procedure was defined as a regression under 30% of obstruction of the IRA with TIMI-3 flow. After the procedure, all patients were admitted to the coronary care unit. The use of glycoprotein IIb/IIIa inhibitors was left to the discretion of the operators. Follow-up and endpoints
In all cases, blood samples were drawn at admission before catheterization procedures. A 12-lead ECG was recorded for each patient at hospital presentation, and the localization of myocardial infarction and IRA were determined from the ECGs. Echocardiographic analyses were carried out within 24–48 h of revascularization. Biplane Simpson’s method was used for the measurement of left ventricular ejection fraction (LVEF) [18]. eGFRs were determined using eGFRs that were calculated using the modification of diet in renal disease [19]. AKI was defined according to the Kidney Disease Improving Global Outcomes criteria [20]. The N/L ratio was obtained by dividing the total count of neutrophils by the lymphocytes count. The P/L ratio was obtained by dividing the total count of platelets by the lymphocyte count. SCr levels were measured at admission before angiography and daily after the procedure until discharge. An absolute 0.3 mg/dl or more increase in SCr levels compared with baseline levels within 48 h after administration of contrast medium was considered as CI-AKI. Data of in-hospital mortality rates of the patients were obtained from the medical records of hospitals.
Definitions
Statistical analysis
Hypertension was defined as a blood pressure of 140/90 mmHg or greater or having a history of antihypertensive drug use. Diabetes mellitus (DM) was defined as a fasting blood glucose of at least 126 mg/dl on two occasions or as being on treatment. Present smokers were defined as having a history of smoking within the past year. Admission anemia was defined as a baseline hemoglobin (Hb) level less than 12 mg/dl in women and less than 13 mg/dl in men. Chronic kidney disease was defined as an eGFR less than 60 ml/min/1.7 m2.
All analyses were carried out using SPSS 20.0 (released 2011, IBM statistics for Windows, version 20; IBM Corp., Armonk, New York, USA), MedCalc version 11.3 (MedCalc Software, Mariakerke, Belgium), and Matlap (7.5.0 r2007b; The Mathworks, Natick, Massachusetts, USA). All data are presented as mean ± SD unless otherwise stated. Comparison of parametric values between the two groups was performed using an independent-samples t-test. Comparisons of nonparametric values between the two groups were performed using the Mann–Whitney U-test. Categorical variables were compared using the χ2-test. Logistic regression analysis was used to assess predictors of CI-AKI and inhospital mortality. Those variables with P less than 0.1 by
Coronary angiography, angioplasty, and stenting
Emergency coronary angiography was performed using the standard Judkins technique. All patients were given a
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Red cell distribution width, nephropathy Akin et al. 291
univariate analysis were included in the multivariate logistic regression analysis model and the respective odds ratios (OR) with 95% confidence intervals (CI) were calculated. The usefulness of the RDW value in predicting the presence of CI-AKI was analyzed using receiver operating characteristics (ROC) curve analyses. When a significant cut-off value was observed, the sensitivity and specificity values were presented. The Delong method was used to compare area under the curve (AUC). In addition, the contribution of RDW was analyzed using the net reclassification improvement calculator. All statistical tests were two-sided, and statistical significance was determined at a P value less than 0.05.
Results The mean age of the patients studied was 56.7 ± 12 years and 19% of the patients were women. A total of 79 of 630 (12.5%) patients experienced CI-AKI. Demographic and clinical patient characteristics in the CI-AKI and non-AKI groups are listed in Table 1. The patients in CI-AKI group were older than the patients in the non-AKI group, the incidence of hypertension was higher in the CI-AKI group than in the non-AKI group, creatinine levels were significantly higher, the levels of eGFR were lower, and Baseline clinical and laboratory characteristics of CI-AKI and non-CI-AKI groups
Table 1
Age (years) Male sex [n (%)] Hypertension [n (%)] Diabetes [n (%)] Smoking [n (%)] Chronic kidney disease [n (%)] Previous history of MI [n (%)] Previous history of CABG [n (%)] Hemoglobin (g/dl) Mean platelet volume (fl) RDW (%) Neutrophil count (109/l) N/L ratio P/L ratio Fasting glucose (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglyceride (mg/dl) Creatinine (mg/dl) eGFR (ml/min/1.73 m2) Total bilirubin (mg/dl) Direct bilirubin (mg/dl) Left ventricular EF Multivessel coronary disease [n (%)] Contrast volume Infract-related artery [n (%)] Left anterior descending Left circumflex Right coronary
No CI-AKI (n = 551)
CI-AKI (n = 79)
P value
55.7 ± 11.7 448 (81.3) 187 (33.9) 135 (24.5) 200 (36.3) 41 (7.4)
63.6 ± 12.3 62 (78.5) 45 (57) 27 (34.2) 30 (37.5) 32 (40.5)
< 0.001 0.542 < 0.001 0.074 0.589 < 0.001
101 (18.3) 17 (3.1)
18 (22.8) 5 (6.3)
13.9 ± 1.7 8.3 ± 1 13.1 ± 1 10.2 ± 12.8 6.2 ± 5.7 158 ± 108.4 133.3 ± 56.8 126.3 ± 37.6 41 ± 22.5 130 ± 95.6 0.9 ± 0.4 98.2 ± 28.5 0.38 ± 0.52 0.18 ± 0.43 49.7 ± 8.3 251 (45.6)
13.4 ± 1.6 8.6 ± 1.1 13.7 ± 1.3 10.3 ± 3.7 6.7 ± 4.9 187 ± 114.5 149.7 ± 75.3 119.7 ± 38.3 36.2 ± 13.4 125.8 ± 120.7 1.5 ± 1.4 67.5 ± 30.1 0.36 ± 0.35 0.13 ± 0.14 45 ± 11.1 48 (60.8)
141.8 ± 49
153.9 ± 49
295 (53.5) 113 (20.5) 143 (26)
0.357 0.179 0.031 0.039 < 0.001 0.969 0.032 0.025 0.087 0.133 0.070 0.714 < 0.001 < 0.001 0.791 0.632 0.001 0.016 0.042 0.541
38 (48.1) 16 (20.3) 25 (31.6)
CABG, coronary artery bypass grafting; CI-AKI, contrast-induced acute kidney injury; EF, ejection fraction; eGFR, estimated glomerular filtration rate; MI, myocardial infarction; N/L ratio, neutrophil/lymphocyte ratio; P/L ratio, platelet/lymphocyte ratio; RDW, red cell distribution width.
LVEF at the admission was also lower in the CI-AKI group compared with the non-CI-AKI group (Table 1). The RDW, N/L ratio, P/L ratio, and MPV were significantly higher in the CI-AKI group than in the non-CIAKI group (Table 1). The volume of contrast administered during the procedure was also higher in the CI-AKI group. In the angiographic parameters, there were no differences in IRA between two groups, but in the AKI group, there was a higher incidence of multivessel disease than that in the non-AKI group. There was no difference in sex, presence of DM, lipid profiles – including triglyerides to HDL cholesterol – total bilirubin, and direct bilirubin between the CI-AKI and the non-CI-AKI groups. In all patients, RDW was correlated positively with the N/L ratio (r = 0.099, P = 0.013) and the P/L ratio (r = 0.237, P = 0.001). There was also a mildly significant inverse association between RDW level and eGFR (r = − 0.118, P = 0.003) (Fig. 1). Significant univariate predictors of CI-AKI were age, hypertension, LVEF, Hb, eGFR, MPV, RDW, N/L ratio, and multivessel disease. Using multivariate logistic regression analysis, we found that LVEF (OR = 0.972, 95% CI 0.945–0.998, P = 0.033), eGFR (OR = 0.970, 95% CI 0.959–0.981, P < 0.001), contrast volume (OR = 1.007, 95% CI 1.002–1.012, P = 0.009), and RDW (OR = 1.406, 95% CI 1.120–1.792, P = 0.005) were independent predictors of CI-AKI. The results of multivariate logistic regression analysis are presented in Table 2. For in-hospital mortality, age, hypertension, RDW, eGFR, CI-AKI, LVEF, N/L ratio, and multivessel coronary disease were included in the multivariate logistic regression model. In the multiple logistic regression analysis, only LVEF (OR = 0.891, 95% CI 0.852–0.932, P < 0.001) and CI-AKI (OR = 2.949, 95% CI 1.139–7.636, P = 0.026) were found to be independent correlates of inhospital mortality (Table 3). Using ROC comparison analysis, eGFR showed superior predictive value than RDW, ejection fraction, and contrast volume for the prediction of CI-AKI. We also observed that RDW had a significantly larger AUC than that of contrast volume (0.663, 95% CI 0.601–0.725 vs. 0.582, 95% CI 0.540–0.624; P < 0.01) for the prediction of CI-AKI (Table 4). The ROC analysis yielded a cut-off value of 13.25 for RDW to predict CI-AKI with 62% sensitivity and 64% specificity (AUC = 0.663, 95% CI = 0.601–0.725, P < 0.001) (Fig. 2). Higher RDW nonsignificantly improved the discriminatory ability (AUC from 0.786 ± 0.031 to 0.799 ± 0.030, P = 0.345) of the baseline clinical model, but did not improve the NRI index (1.251, P = 0.426).
Discussion In this study, we found a significant association between serum RDW levels and the development of CI-AKI in patients with STEMI undergoing primary PPCI. In the
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
292 Coronary Artery Disease 2015, Vol 26 No 4
Fig. 1
RDW (%)
(a)
(b) 22.00
22.00
20.00
20.00
18.00
18.00
16.00
16.00
14.00
14.00
12.00
12.00 R Linear = 0.01
R2 Linear = 0.056
2
10.00
10.00 0.00
20.00
40.00 N/L ratio
60.00
0.00
200.00
400.00 600.00 P/L ratio
800.00
1000.00
(c) 22.00
20.00
RDW (%)
18.00
16.00
14.00
12.00 R2 Linear = 0.014 10.00 0.00
50.00
100.00
150.00
200.00
250.00
eGFR (ml/min/1.73 m2) Correlation between RDW and the N/L ratio (a). Correlation between RDW and the P/L ratio (b). Correlation between RDW and eGFR (c). eGFR, estimated glomerular filtration rate; N/L ratio, neutrophil-to-lymphocyte ratio; P/L ratio, platelet-to-lymphocyte ratio; RDW, red cell distribution width.
present study, the incidence of CI-AKI was 13%, which was in agreement with previous reports. In addition, CIAKI was an independent predictor of in-hospital mortality in multivariate models. Recent studies have reported an independent association between RDW and intrahospital and long-term outcomes in patients with AMI [10,21]. Ihan et al. [21] enrolled 763
patients with AMI undergoing a PPCI and documented these patients’ intrahospital mortality from hospital registries. In their analysis, RDW was associated with intrahospital cardiovascular mortality. Uyarel et al. [22] investigated the association between RDW and inhospital and long-term cardiovascular mortality in 2506 STEMI patients. Their study suggests that RDW could predict mortality among PPCI patients. In our study, we
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Red cell distribution width, nephropathy Akin et al. 293
Variables Age Hypertension Left ventricular EF Hemoglobin eGFR Mean platelet volume RDW N/L ratio Multivessel coronary disease Contrast volume
Odds ratio
95% confidence interval
P
1.013 1.331 0.972 1.096 0.970 1.267 1.406 1.014 1.359 1.007
0.988–1.041 0.743–2.246 0.945–0.998 0.942–1.275 0.959–0.981 0.980–1.640 1.120–1.792 0.971–1.060 0.781–2.372 1.002–1.012
0.320 0.337 0.033 0.236 < 0.001 0.070 0.005 0.525 0.285 0.009
EF, ejection fraction; eGFR, estimated glomerular filtration rate; N/L ratio, neutrophil/lymphocyte ratio; RDW, red cell distribution width.
Age Hypertension Left ventricular EF CI-AKI eGFR RDW N/L ratio Multivessel coronary disease
1.0
0.8
0.6
0.4 AUC : 0.663 95% CI : 0.601–0.725
Independent predictors of in-hospital mortality in multivariate logistic regression analysis
Table 3
Variables
Fig. 2
Sensitivity
Independent predictors of contrast-induced acute kidney injury in multivariate logistic regression analysis
Table 2
0.2
Odds ratio
95% confidence interval
P
1.020 1.645 0.891 2.949 0.987 0.967 1.033 2.380
0.981–1.060 0.649–4.167 0.852–0.932 1.139–7.636 0.971–1.003 0.704–1.328 0.975–1.094 0.915–6.195
0.319 0.294 < 0.001 0.026 0.116 0.835 0.273 0.076
0.0 0.0
0.2
0.4 0.6 1 − specificity
0.8
1.0
Receiver-operating characteristic curve analysis of RDW for the prediction of contrast-induced acute kidney injury. AUC, area under the curve; CI, confidence interval; RDW, red cell distribution width.
CI-AKI, contrast-induced acute kidney injury; EF, ejection fraction; eGFR, estimated glomerular filtration rate; N/L ratio, neutrophil/lymphocyte ratio; RDW, red cell distribution width.
investigated the association between RDW and CI-AKI and in-hospital mortality. In our cohort, RDW was associated with CI-AKI, but not with in-hospital mortality. It was also reported that CI-AKI is associated with a higher in-hospital complication rate and mortality in patients undergoing PPCI [23]. Consistently, we found a significant association between CI-AKI and in-hospital mortality. Our study suggests that the association between RDW and in-hospital mortality among PPCI patients can be explained by CI-AKI. To the best of our knowledge, the study by Kurtul et al. [12] is the only study in the literature determining the association between RDW and CI-AKI. They investigated the association between RDW and CI-AKI in heterogeneous cardiac disorders including STEMI, nonSTEMI, and unstable angina pectoris. Our study included patients with STEMI undergoing PPCI – a high-risk group for the development of CI-AKI – we confirm and Table 4
extend the previous findings. Their study included 673 patients with acute coronary syndrome; 424 of these patients had STEMI. Our sample size of STEMI patients was larger and thus more reflective of CI-AKI risk in patients undergoing PPCI. We also investigated the utility of the preprocedural RDW for in-hospital mortality in patients STEMI who underwent PPCI. Mehran et al. [24] developed a simple risk score of CIAKI after PCI. They suggested a CI-AKI risk stratification score on the basis of eight variables including age, chronic kidney disease, hypotension, DM, chronic congestive heart failure, anemia, and contrast volume. In our study, RDW was an independent predictor of CI-AKI and it showed superior predictive value than contrast volume. Our study suggests that RDW may advance the prediction of CI-AKI. The underlying pathophysiological mechanisms of CIAKI are not known. Implicated mechanisms for the
Univariate predictors of contrast-induced acute kidney injury
Parameters
AUC
95% CI
P
Sensitivity (%)
Specificity (%)
Youden index J
eGFR LVEF RDW CV
0.763 0.611 0.663 0.582
0.728–0.796 0.571–0.648 0.601–0.725 0.540–0.624
< 0.001 0.003 < 0.001 0.009
68.4 53.2 62.8 67.4
75.5 68.2 64.7 52.3
0.438 0.209 0.278 0.208
P(LVEF − RDW) = 0.199, P(LVEF − eGFR) = 0.001, P(RDW − eGFR) = 0.027, P(RDW − CV) = 0.047, P(eGFR − CV)
