Clin Exp Nephrol DOI 10.1007/s10157-014-0955-4
ORIGINAL ARTICLE
Relationship of late arteriovenous fistula stenosis with soluble E-selectin and soluble EPCR in chronic hemodialysis patients with arteriovenous fistula Mukadder Ayse Bilgic • Hakki Yilmaz • Alper Bozkurt • Huseyin Tugrul Celik • Ismail Celal Bilgic • Ozgul Malcok Gurel Ismail Kirbas • Nuket Bavbek • Ali Akcay
•
Received: 16 January 2014 / Accepted: 25 February 2014 Ó Japanese Society of Nephrology 2014
Abstract Background/aims Vascular access dysfunction caused by stenosis is a major complication for hemodialysis (HD) patients. However, physiopathology of late arteriovenous fistula (AVF) stenosis is still under investigation. The aim of the present study was to evaluate the association between plasma soluble EPCR (sEPCR) with serum soluble E-selectin (sE-selectin) concentration and late AVF stenosis in HD patients. Methods Plasma sEPCR and serum sE-selectin concentrations were measured in 94 HD patients. Using these data, we studied the association of sEPCR and sE-selectin with the presence and degree of AVF stenosis using ultrasonography and fistulogram. Results Fifty-one patients have AVF stenosis, and the others (n = 43) have patent AVF. The degree of AVF
M. A. Bilgic H. Yilmaz (&) N. Bavbek A. Akcay Section of Nephrology, Department of Internal Medicine, School of Medicine, Turgut Ozal University, Alparslan Tu¨rkes Cad. No: 57, 06510 Ankara, Turkey e-mail: [email protected] A. Bozkurt I. Kirbas Department of Interventional Radiology, School of Medicine, Turgut Ozal University, Ankara, Turkey H. T. Celik Department of Biochemistry, School of Medicine, Turgut Ozal University, Ankara, Turkey I. C. Bilgic Department of General Surgery, School of Medicine, Turgut Ozal University, Ankara, Turkey O. M. Gurel Department of Cardiology, School of Medicine, Turgut Ozal University, Ankara, Turkey
stenosis was correlated with serum sE-selectin levels (r = 0.351, p = 0.01), but not sEPCR (r = 0.075, p = 0.702). The median level of sE-selectin was statistically higher in the group of AVF stenosis than in the group of patent AVF [463.2 pg/ml (275.4–671.4) vs. 162.5 pg/ml (96.7–285.3), p = 0.001]. Increased sE-selectin levels [OR (OR) = 6.356, p = 0.015] and high levels of LDL (OR = 4.321, p = 0.044) were independent predictors of late AVF stenosis in the multivariate model. Conclusions sE-selectin and the LDL were the most important predictors of late AVF stenosis. In addition, sEselectin correlated with the degree of AVF stenosis. We suggested that atherosclerosis might be contributing factor for development of late AVF stenosis. Keywords Arteriovenous fistula stenosis Biomarkers End-stage renal disease Angiography
Introduction A native arteriovenous fistula (AVF) is the preferred form of permanent vascular access in patients with end-stage renal disease (ESRD) undergoing hemodialysis (HD) due to fewer complications and decreased mortality [1]. Vascular access dysfunction (VAD) is the main cause of morbidity [3], and the problems associated with VAD represent roughly 20 % to 25 % of hospitalizations of patients on maintenance HD [3, 4]. Vascular access malfunction associated with some medical factors such as stasis (hypotension, hypoalbuminemia, compression), hypercoagulability [antiphospholipid antibodies, hyperhomocysteinemia, factor V leiden, lipoprotein (a)], endothelial cell injury (preexisting intimal hyperplasia, TNF-a, oxidative stress, calcium (Ca)
123
Clin Exp Nephrol
phosphate deposition, activated platelets), medications and red cell mass [1–5]. However, these risk factors include diabetes; location of the graft; age [65 years; time until use of graft after surgical procedure; smoking; and increased serum cholesterol and triglyceride (TG) levels [1–5]. The pathogenesis of arteriovenous (AV) access stenosis is not understood enough, although VAD caused by stenosis is a leading complication for HD patients [3–5]. The most common cause of this VAD is venous stenosis as a result of progressive venous neointimal hyperplasia and inadequate outward remodeling within the peri-anastomotic region (AV fistula) [3–5]. The primary cause of AVF stenosis resembles the histology of atherosclerosis [6]. E-selectin [endothelial leukocyte adhesion molecule-1 (ELAM-1), CD62E] is a cell surface glycoprotein that is expressed on inflamed endothelial cells in response to treatment with inflammatory cytokines [7, 8]. E-selectin mediates the interactions of leukocytes and platelets with endothelial cells [7, 8]. E-selectin is associated with severe atherosclerosis in ESRD [9, 10]. Some studies revealed that E-selectin contributes to the process of intimal hyperplasia [11] and decreasing E-selectin expression inhibits neointimal hyperplasia [12] in rat models. We hypothesized that E-selectin can be a major contributing factor for the development of AVF stenosis via neointimal hyperplasia and inflammation. The protein C pathway is a primary regulator of blood coagulation, and it plays key roles in the critical component of the host response to inflammatory stimuli [13]. The recent member of this pathway is the endothelial protein C receptor (EPCR) [14]. EPCR is a 46 kDa, type 1 transmembrane protein with homology to CD1d/MHC class I proteins, expressed primarily on the larger vessels and also microvascular endothelium [14, 15]. EPCR binds PC and facilitates formation of activated protein C (APC), a potent anticoagulant and anti-inflammatory agent [14–16]. EPCR can be cleaved to release a soluble form (sEPCR) in the circulation [17]. Soluble EPCR (sEPCR) binds PC and APC with similar affinity [16, 17]. sEPCR inhibits both PC activation and APC anticoagulant activity [15–17]. The increased soluble EPCR was associated with thromboembolic (e.g., deep venous thrombosis [18], stroke [19], STsegment elevation myocardial infarction [20]) and systemic inflammatory diseases (e.g., sepsis [21] and systemic lupus erythematosus [21]). In addition, sEPCR has been shown to modulate inflammation by binding to activated neutrophils [22]. We showed that higher level of sEPCR was related to the development of AVF thrombosis in HD patients [23]. No study has as yet investigated whether sEPCR influences on the development of AVF stenosis or not in HD patients. The aim of present study was to examine the relationship between E-selectin, sEPCR and the development of AVF stenosis in HD patients.
123
Materials and methods Over a period of 13 months, from January 2012 to February 2013, a total of 325 patients with AVF access dysfunction and AVF access routine control, who were referred to Department of Interventional Radiology, Turgut Ozal University, School of Medicine, Ankara. 181 patients were excluded, who have multiple repetitive endovascular treatments. Other causes of AVF dysfunction (such as thrombosis and aneurysm) except stenosis were detected in 45 patients. Patients were eligible for inclusion if they were established on HD and between 20 and 80 years of age. Patients were excluded if they had a history of hematological malignancy, a known bleeding diathesis or coagulopathy or concomitant warfarin therapy (n = 5). The single-center study included total 94 patients treated with chronic HD for more than 6 months. To identify the possible risk factors for AVF dysfunction, variables including age; gender; co-morbidities such as diabetes, hypertension, coronary arterial occlusive disease and peripheral arterial occlusive disease; age of AVF; side of the AVF; type of AVF; number of lesions; degree of stenosis; and residual stenosis were collected. Color Doppler ultrasonography (CDUS) studies were performed on a Doppler ultrasound device (Philips, HD15 PureWave Ultrasound System, Bothell, WA, USA) with a 5-MHz linear probe (C5-1 PureWave transducer) by two blinded radiologists experienced in the use of Doppler US. At CDUS analysis, AVF flow volume, peak systolic (PSV) and end-diastolic velocities, fistula diameter and subcutaneous depth were measured. Fistula anastomosis area and calibration, flow velocities and lumens of vascular structures in the arterial and venous tree were analyzed. Visual narrowing (diameter narrowing[50 %) as assessed on gray scale imaging and an increased ratio of the PSV at the stenosis as compared to the PSV measured 2 cm upstream from the stenosis are two major features to characterize AVF stenosis [24]. The diagnostic criteria vary in accordance with the location of the stenosis [24]. Cases with a PSV ratio in the anastomosis region of 3 and above were regarded as significant (50 % and above) anastomosis stenosis [24]. A draining vein and inflow artery stenosis are defined by visible narrowing and by a PSV ratio of C2:1 [24]. In addition, observation of focal color aliasing in the stenosis region at CDUS and peak systolic velocities rising above 400 cm/s were regarded as other assistant findings [24]. The site of stenotic lesions was classified into six categories based upon location [25]: (1) native artery; (2) AV anastomosis; (3) juxta-anastomotic vein (initial 2 cm of fistula); (4) venous outflow [2 cm from anastomosis (cannulation zone); (5) distal outflow (defined as above elbow joint for radial-cephalic fistulae and above mid-humerus for brachial fistulae); and (6) central venous system.
Clin Exp Nephrol
Fistulography with digital subtraction angiography (DSA) examinations was performed by two experienced interventional radiologists using a digital subtraction system (Philips Allura Xper FD 20 Release 2, Best, The Netherlands). Fistulogram with DSA was performed in these patients (n = 57) who have abnormal ultrasound findings. The degree of stenosis was classified into four groups: 50–74, 75–89, 90–99 and 100 % [26]. An accessory vein was defined as a branch coming off of the main venous channel that comprised the fistula. Blood was drawn from all study subjects under standardized conditions before fistulogram was performed. Samples were stored at -80 °C until analysis. Serum creatinine, urea, aspartate aminotransferase (AST), alanine aminotransferase (ALT), Ca, albumin, uric acid, CRP, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), TG, Ca and phosphorus (P) were measured by standard laboratory techniques using an autoanalyzer (Roche Diagnostics, COBAS INTEGRA 800, Indianapolis, Indiana, USA). Complete blood count (CBC) analysis was also performed with the use of a Coulter counter technique (Coulter Gen-S Hematology Analyzer, Beckman Coulter Corp, Hialeah, Florida). Plasma sEPCR levels were measured by solid-phase enzyme-linked immunosorbent assay (Catalog No: SK00507-01; Aviscera Bioscience Inc., Santa Clara, CA, USA) with intra- and inter-assay CVs of \6 and \ 12 %, respectively. This assay has high sensitivity (lower limit of detection 36 pg/mL). Human soluble E-selectin (sE-selectin) was analyzed in the serum with an enzyme-linked immunosorbent assay (Catalog No: EK0501; BOSTER BIOLOGICAL TECHNOLOGY Co., Ltd., Fremont, CA USA). This assay has high sensitivity (lower limit of detection \4 pg/mL) and exhibits no detectable cross-reactivity with other relevant proteins. All samples were assayed in duplicate. Calibration and standardization of the assays were performed according to the manufacturer’s protocol. Continuous variables were defined mean ± standard deviation or median (25–75 %); categorical variables were given as percentages. The Kolmogorov–Smirnov test was applied for a normal distribution. The independent sample unpaired Student’s t test or the Mann–Whitney U test was used for the continuous variables, and categorical data were compared with the chi-square test. We performed univariate logistic regression analyses to assess the association of clinical and laboratory markers with the binary presence of late AVF stenosis (yes/no). We reduced the model using stepwise multivariate logistic regression analyses. Odds ratios (OR) and 95 % confidence intervals were calculated. A p value \0.05 was considered statistically significant. All analyses were performed using SPSS software (version 17.0; SPSS Inc., Chicago, IL).
This study is approved by local Ethics Committee (IRB approval number: 99950669/1203).
Results Forty-eight (n = 48) patients were admitted for fistula control. 46 patients who had clinical findings of AVF dysfunction (difficulties in cannulation, painful arm edema, prolonged bleeding time after cannulation or after removal of the dialysis needles) were included in the study. All HD patients included in the current study (n = 94) were evaluated by using CDUS. 56 patients (59.6 %) had a luminal narrowing more than 50 %. 57 patients (60.6 %) had a peak systolic velocity ratio higher than 2:1. There was no abnormal ultrasonography finding in 37 patients; after all, all of the patients were admitted for control. Fistulography was performed for all patients with abnormal ultrasound findings (n = 57). Severe stenosis (C50 %) was determined in 51 out of 57 patients who underwent fistulography. Therefore, 51 patients were included in the stenosis group, while 43 patients were considered in the group without stenosis. Demographic features of the patients are presented in Table 1. The mean age was 57.2 ± 15.8 years (range 20–80), and 52.1 % of the patients (n = 49) were male. Renal failure had various causes, but the major cause was diabetic nephropathy (n = 43, 45.7 %). The patients were divided into two groups: patients with stenosis (Group 1; n = 51) and patients without stenosis (Group 2; n = 43). There was no difference in age, gender and body mass index between Group 1 and Group 2 (Table 2). The duration of fistula use was similar between Table 1 Baseline characteristics of study population Hemodialysis patients Numbers
94
Male/female (%)
49 (52.1)/45 (47.9)
Age (years)
57.2 ± 15.8
Body mass index (kg/m2)
27.6 ± 6.36
Hemodialysis duration (months)
36.23 ± 13.54
Smoking (%)
36.1
Hypertension (%)
89.3
Etiology of chronic renal failure Diabetic nephropathy
43
Hypertension related nephropathy
32
Chronic primary glomerulonephritis Interstitial nephritis
9 3
Polycystic kidney disease
2
Obstructive nephropathy
1
Unknown
4
123
Clin Exp Nephrol Table 2 Comparison among two groups Parameters
Group I (n = 51)
Group II (n = 43)
p
Male/female
26/25
23/20
0.786
Age (years)
57.8 ± 12.3
56.9 ± 14.5
0.812
BMI (kg/m2)
27.8 ± 6.32
26.9 ± 5.72
0.345
Smoker [n (%)]
19 (37.2)
15 (34.8)
0.098
Duration of AVF use (months)
42.8 ± 12.6
38.5 ± 8.7
0.245
URR
73.5 ± 2.8
74.1 ± 3.0
0.235
Albumin (g/dl)
3.87 ± 0.36
3.91 ± 0.42
0.189
Ca (corrected) (mg/dl)
8.40 ± 0.52
8.50 ± 0.65
0.231
P (mg/dl)
5.80 ± 1.90
5.75 ± 1.79
0.365
iPTH (pg/ml)
332.8 ± 160.5
319.5 ± 204.3
0.105
HDL-C (mg/dl) Triglycerides (mg/dl)
39.6 ± 10.3 279.5 ± 90.4
38.7 ± 11.2 288.9 ± 106.6
0.305 0.411
LDL-C (mg/dl)
154.5 ± 32.6
128.7 ± 28.6
\0.001
Hb (g/dl)
10.85 ± 1.15
10.90 ± 1.26
0.542 0.259
TSAT (%)
38.7 ± 10.2
37.8 ± 11.8
Ferritin (ng/ml)
422.4 ± 240.8
439.1 ± 230.7
0.346
CRP (mg/l)
18.77 ± 20.48
13.28 ± 12.47
0.271
sE-selectins (pg/ml)
463.2 (275.4–671.4)
162.5 (96.7–285.3)
0.001
sEPCR (pg/ml)
8,345.6 (1,456.7–12,347.9)
8,278.0 (1,265.8–11,782.4)
0.472
ACE or ARB [n (%)]
16 (31.3)
13 (30.2)
0.613
ß-Blockers [n (%)]
15 (29.4)
12 (27.9)
0.567
CCB [n (%)]
31 (60.7)
27 (62.7)
0.231
Alpha-blockers [n (%)]
10 (19.6)
9 (20.9)
0.102
Group 1 and Group 2 (42.8 ± 12.6 vs. 38.5 ± 8.7 months, respectively; p = 0. 245). There was no difference in medication use and incidence of cardiovascular diseases between the groups (Table 2). Biochemical tests showed that patients in Group 1 had higher LDL cholesterol (154.5 ± 32.6 vs. 128.7 ± 28.6 mg/dl; p \ 0.001) and median sE-selectin levels [463.2 pg/ml (275.4–671.4) vs. 162.5 pg/ml (96.7–285.3); p = 0.001] compared with patients in Group 2 (Table 2). There was no difference in the median sEPCR [8,345.6 pg/ml (1,456.7–12,347.9) vs. 8278 pg/ml (1,265.8–11,782.4); p = 0.435] level and other biochemical parameters between the groups (Table 2). We summarized CDUS and fistulographic findings of AVF stenosis in Table 3. Correlation analysis showed that serum sE-selectin levels increased with increasing degree of AVF stenosis determined by fistulography (r = 0.351, p = 0.01). However, there was no correlation between sEPCR and the degree of AVF stenosis (r = 0.075, p = 0.702). LDL-C levels were not associated with location and degree of AVF stenosis (r = 0.123, p = 0.345; r = 0.098, p = 0.689). Presence of DM was not related to location and degree of AVF stenosis (r = 0.097, p = 0.876; r = 0.126, p = 0.213).
123
Parameters that would cause AVF stenosis were determined by using univariate analysis, and then multivariate logistic regression analysis was performed. In the multivariate model (Table 4), sE-selectin was an independent predictor of AVF stenosis in chronic HD patients (OR = 6.356, p = 0.015) together with LDL (OR = 4.321, p = 0.044). Presence of DM and duration of AVF use did not predict for late AVF stenosis (Table 4).
Discussion In this study, sE-selectin level was determined as an independent and strong predictor of AVF stenosis, which is proved by using Doppler ultrasonography and fistulogram. In addition, (1) there was no difference in sEPCR levels between patients with AVF stenosis and patients without AVF stenosis; (2) While sE-selectin level was associated with the degree of AVF stenosis, there was no association between sEPCR level and the degree of AVF stenosis; (3) LDL level was higher in patients with AVF stenosis, and LDL was found to be an independent predictor of AVF stenosis. Arteriovenous fistulas represent the preferred vascular access route for HD patients [1–4]. Complications related
Clin Exp Nephrol
to vascular access are common in patients undergoing long-term HD. These complications are a major cause of morbidity, and they may lead to the loss of vascular access if the appropriate intervention is not performed on time. Table 3 CDUS and Fistulographic findings of AVF stenosis Category
n (%)
C50 % \50 %
56 (59.6) 38 (40.4)
Color Doppler ultrasonography Luminal diameter reduction A peak systolic velocity ratio Peak systolic flow velocity
C2:1
57 (60.6)
\2:1
37 (39.4)
[400 cm/s
49 (52.1)
\400 cm/s
45 (47.9)
Fistulography Degree of stenosis Total occlusion
100 %
6 (11.7)
Stenosis
90–99 %
20 (39.3)
75–89 %
15 (29.4)
50–74 %
10 (19.6)
1 site
40 (78.5)
2 sites
9 (17.6)
C3 sites
2 (3.9)
Native artery Arteriovenous anastomosis
3 (5.9) 16 (31.3)
Juxta-anastomotic vein
22 (43.1)
Number of lesions
Location of lesions
Table 4 Significant predictors of late AVF stenosis in univariable and multivariable logistic regression analyses
Venous outflow
18 (35.3)
Distal outflow
5 (9.8)
Central venous system
0
Especially in the long term, stenosis development is one of the major causes of AVF malfunction [1–5]. While certain factors that may lead to AVF thrombosis have been defined, factors that would lead to AVF stenosis are not sufficiently defined [3–5]. In this study, we showed that sE-selectin level was higher in HD patients with isolated AVF stenosis, and sEselectin level was an independent predictor of AVF stenosis. It is possible to explain the association between sEselectin and AVF stenosis as follows: The major cause of AVF stenosis is the development of neointimal hyperplasia [5], and E-selectin has been shown to play an important role in intima hyperplasia in rat models [11]. In another recent study, administration of cilostazol, which is a selective phosphodiesterase-3 inhibitor with antiatherogenic and antiproliferative features, to rats suppressed neointimal hyperplasia related to decreased E-selectin levels [12]. Therefore, it is possible to assume that E-selectin levels are associated with neointimal hyperplasia, and AVF stenosis develops with the progression of neointimal hyperplasia. The findings of this study also support this hypothesis. Previous studies have shown that E-selectin levels are associated with atherosclerosis in HD patients [9, 10]. Stenosis of the AVF is histologically similar to the atherosclerotic lesion [6, 27]. In activated endothelial cells, E-selectin expression increases to contribute to the formation of atherosclerosis, while it is also possible that increased E-selectin expression may lead to AVF stenosis. One of the major findings of this study is the higher LDL levels in patients with AVF stenosis. High serum LDL
Univariate OR (95 % CII)
p value
Age (year)
1.968 (1.645–2.143)
0.041
BMI (kg/m2)
0.973 (0.968–1.039)
0.727
Smoker
1.097 (1.065–1.159)
0.853
Duration of AVF use
1.185 (0.876–1.967)
0.345
DM (yes/no) Ca (mg/dl)
1.251 (0.945–1.423) 1.041 (1.006–1.198)
0.186 0.389
P (mg/dl)
0.970 (0.845–1.135)
0.818
Albumin (g/dl)
0.988 (0.968–1.035)
0.913
LDL (mg/l)
5.071 (3.987–7.256)
\0.001
HDL (mg/l)
1.018 (0.986–1.349)
0.658
iPTH (pg/ml)
1.174 (1.083–1.389)
0.592
Hb (g/dl)
0.815 (0.703–0.985)
0.439
CRP (mg/l)
1.236 (1.090–1.408)
0.246
TSAT (%)
1.072 (1.008–1.121)
0.857
Ferritin
1.211 (0.861–1.812)
0.312
sEPCR
1.023 (1.012–1.031)
0.428
sE-selectin
7.234 (5.891–9.213)
\0.001
Multivariate OR (95 % CI)
p value
1.245 (0.971–1.943)
0.206
4.321 (3.189–6.057)
0.044
6.356 (4.123–8.521)
0.015
123
Clin Exp Nephrol
level is known to be a major risk factor for atherosclerosis [28]. Therefore, high LDL levels in patients with AVF stenosis imply that AVF stenosis can develop through mechanisms similar to that of atherosclerosis. As a result of LDL cholesterol reduction with statins, neointimal formation has been shown to decrease after stent implantation [29]. Recently, Janardhanan et al. [30] have shown that simvastatin resulted in a significant reduction in venous neointimal hyperplasia and venous stenosis formation through decreasing VEGF-A/MMP-9 pathway activity in dialysis vascular access model. These studies and the findings of the current study imply that mechanisms leading to atherosclerosis play an important role in the development of AVF stenosis, increased LDL level is one of the factors leading to AVF stenosis, and intensive LDL management may prevent AVF stenosis. We previously showed the association between sEPCR and AVF thrombosis [23]. sEPCR has been shown to be associated with not only venous thromboemboli, but also stroke, ST-elevated MI and sepsis [19–21]. However, the association of sEPCR with neointimal hyperplasia remains obscure. The exact mechanisms of neointimal hyperplasia, the primary cause of AVF stenosis, are not currently understood, but are thought to be associated with the cumulative effects of inflammation, hemodynamic factors and coagulation [5, 31]. In addition, activation of coagulation cascade may indirectly play a role in mediating the extent of intimal hyperplasia and stenosis [32]. Tissue factor pathway inhibitor, a critical inhibitor of coagulation, markedly reduced intimal hyperplasia formation in experimental vein graft [33]. There is no study on isolated AVF stenosis without thrombosis—i.e., the association between neointimal hyperplasia and sEPCR—in the literature. In this study, we showed that there was no difference in sEPCR levels between patients with AVF stenosis and patients without AVF stenosis. This finding implies that sEPCR has no role in the pathogenesis of neointimal hyperplasia. We found no the relationship between presence of diabetes, degree localization of AVF stenosis. In addition, presence of diabetes was not a predictor of late AVF stenosis according to our results. Some studies revealed that presence of diabetes was a factor for patency of AVF [34, 35]. But some studies showed that presence of diabetes did not influence of AVF patency [36, 37]. The present study has some limitations. First, we are aware of the limitations of a cross-sectional study, which limits attributions about the direction of causality between variables. Second, the other limitation of the study was relative small sample of our hospital. Third, soluble E-selectin and EPCR were measured once, but serial measurements could be more appropriate. Nevertheless, our findings require further examination in experimental and prospective studies.
123
Conclusion HD patients with AVF stenosis have higher level of sEselectin and LDL-C, but not sEPCR. LDL-C and sEselectin are independent predictors of AVF stenosis in chronic HD patients. Angiographic stenosis degree is correlated with sE-selectin levels. Our results showed that AVF stenosis mimics some mechanisms of atherosclerosis. Acknowledgments This study was supported by the Scientific Research Fund of Turgut Ozal University. Conflict of interest
We have no conflict of interest.
References 1. Sousa CN, Apo´stolo JL, Figueiredo MH, Martins MM, Dias VF. Physical examination: how to examine the arm with arteriovenous fistula. Hemodial Int. 2013;17(2):300–6. 2. Vachharajani TJ. Diagnosis of arteriovenous fistula dysfunction. Semin Dial. 2012;25(4):445–50. 3. Stolic R. Most important chronic complications of arteriovenous fistulas for hemodialysis. Med Princ Pract. 2013;22(3):220–8. 4. Rose DA, Sonaike E, Hughes K. Hemodialysis access. Surg Clin N Am. 2013;93(4):997–1012. 5. Roy-Chaudhury P, Sukhatme VP, Cheung AK. Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint. J Am Soc Nephrol. 2006;17(4):1112–27. 6. Gagliardi GM, Rossi S, Condino F, Mancuso D, Greco F, Tenuta R, Savino O, Bonofiglio R, Domma F, Latorre G. Malnutrition, infection and arteriovenous fistula failure: is there a link? J Vasc Access. 2011;12(1):57–62. 7. Galkina E, Ley K. Vascular adhesion molecules in atherosclerosis. Arterioscler Thromb Vasc Biol. 2007;27(11):2292–301. 8. Roldan V, Marin F, Lip GY, Blann AD. Soluble E-selectin in cardiovascular disease and its risk factors. A review of the literature. Thromb Haemost. 2003;90:1007–20. 9. Testa A, Benedetto FA, Spoto B, Pisano A, Tripepi G, Mallamaci F, Malatino LS, Zoccali C. The E-selectin gene polymorphism and carotid atherosclerosis in end-stage renal disease. Nephrol Dial Transplant. 2006;21(7):1921–6. 10. Issac MS, Afif A, Gohar NA, Fayek NA, Zayed B, Sedrak H, El Din LAS. Association of E-selectin gene polymorphism and serum papp-a with carotid atherosclerosis in end-stage renal disease. Mol Diagn Ther. 2013;. doi:10.1007/s40291-013-0061-4. 11. Gotoh R, Suzuki J, Kosuge H, Kakuta T, Sakamoto S, Yoshida M, Isobe M. E-selectin blockade decreases adventitial inflammation and attenuates intimal hyperplasia in rat carotid arteries after balloon injury. Arterioscler Thromb Vasc Biol. 2004;24(11):2063–8. 12. Takigawa T, Tsurushima H, Suzuki K, Tsuruta W, Nakamura K, Matsumura A. Cilostazol suppression of arterial intimal hyperplasia is associated with decreased expression of sialyl Lewis X homing receptors on mononuclear cells and E-selectin in endothelial cells. J Vasc Surg. 2012;55(2):506–16. 13. Bouwens EA, Stavenuiter F, Mosnier LO. Mechanisms of anticoagulant and cytoprotective actions of the protein C pathway. J Thromb Haemost. 2013;11(Suppl 1):242–53. 14. Kurosawa S, Stearns-Kurosawa DJ, Hidari N, Esmon CT. Identification of functional endothelial protein C receptor in human plasma. J Clin Invest. 1997;100:411–8. 15. Montes R, Puy C, Molina E, Hermida J. Is EPCR a multi-ligand receptor? Pros and cons. Thromb Haemost. 2012;107(5):815–26.
Clin Exp Nephrol 16. Gleeson EM, O’Donnell JS, Preston RJ. The endothelial cell protein C receptor: cell surface conductor of cytoprotective coagulation factor signaling. Cell Mol Life Sci. 2012;69(5): 717–26. 17. Liaw PCY, Neuenschwander PF, Smirnov MD, Esmon CT. Mechanisms by which soluble endothelial cell protein C receptor modulates protein C and activated protein C function. J Biol Chem. 2000;275:5447–52. 18. Saposnik B, Reny JL, Gaussem P, Emmerich J, Aiach M, Gandrille S. A haplotype of the EPCR gene is associated with increased plasma levels of sEPCR and is a candidate risk factor for thrombosis. Blood. 2004;103(4):1311–8. 19. Yu¨ru¨rer D, Teber S, Deda G, Egin Y, Akar N. The relation between cytokines, soluble endothelial protein C receptor, and factor VIII levels in Turkish pediatric stroke patients. Clin Appl Thromb Hemost. 2009;15(5):545–51. 20. Tanalp AC, Oduncu V, Erkol A, Go¨zu¨bu¨yu¨k G, Ozveren O, Du¨ndar C, Canbay A, Kirma C. Soluble endothelial protein C receptor levels and protein C activity in patients with acute STsegment elevation myocardial infarction. Coron Artery Dis. 2013;24(3):209–16. 21. Kurosawa S, Stearns-Kurosawa DJ, Carson CW, et al. Plasma levels of endothelial cell protein C receptor are elevated in patients with sepsis and systemic lupus erythematosus: lack of correlation with thrombomodulin suggests involvement of different pathological processes. Blood. 1998;91:725–7. 22. Kurosawa S, Esmon CT, Stearns-Kurosawa DJ. The soluble endothelial protein C receptor binds to activated neutrophils: involvement of proteinase-3 and CD11b/CD18. J Immunol. 2000;165:4697–703. 23. Bilgic A, Ozdemir FN, Bayraktar N, Karakus S, Sasak G, Arat Z, Sezer S, Haberal M. Soluble endothelial protein C receptor: influence on arteriovenous fistula thrombosis development in hemodialysis patients. Am J Nephrol. 2007;27(4):366–72. 24. Hutchison SJ, Holmes KC. Arteriovenous fistulas. In: Hutchison SJ, Holmes KC, editors. Principles of Vascular and Intravascular Ultrasound. Philadelphia: Saunders; 2012. p. 82–96. 25. Clark TW, Hirsch DA, Jindal KJ, et al. Outcome and prognostic factors of restenosis after percutaneous treatment of native hemodialysis fistulas. J Vasc Interv Radiol. 2002;13:51–9. 26. Vazquez MA. Vascular access for dialysis: recent lessons and new insights. Curr Opin Nephrol Hypertens. 2009;18:116–21. 27. Marrone D, Pertosa G, Simone S, Loverre A, Capobianco C, Cifarelli M, Memoli B, Schena FP, Grandaliano G. Local
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
activation of interleukin 6 signaling is associated with arteriovenous fistula stenosis in hemodialysis patients. Am J Kidney Dis. 2007;49(5):664–73. Grundy SC. Role of low-density lipoproteins in development of coronary artery atherosclerosis. In: Kreisberg RA, Segrest JP, editors. Plasma lipoproteins and coronary artery disease. Cambridge: Blackwell Scientific; 1992. p. 93–124. Takamiya Y, Miura S, Kawamura A, Tanigawa H, Zhang B, Iwata A, Nishikawa H, Matsuo K, Shirai K, Saku K. Intensive LOwering of BlOod pressure and low-density lipoprotein ChOlesterol with statin theraPy (LOBOCOP) may improve neointimal formation after coronary stenting in patients with coronary artery disease. Coron Artery Dis. 2009;20(4):288–94. Janardhanan R, Yang B, Vohra P, Roy B, Withers S, Bhattacharya S, Mandrekar J, Kong H, Leof EB, Mukhopadhyay D, Misra S. Simvastatin reduces venous stenosis formation in a murine hemodialysis vascular access model. Kidney Int. 2013;84(2): 338–52. Lee T. Novel paradigms for dialysis vascular access: downstream vascular biology—is there a final common pathway? Clin J Am Soc Nephrol. 2013;. doi:10.2215/CJN.03490413. Hasenstab D, Lea H, Hart CE, Lok S, Clowes AW. Tissue factor overexpression in rat arterial neointima models thrombosis and progression of advanced atherosclerosis. Circulation. 2000;101: 2651–7. Huynh TT, Davies MG, Thompson MA, Ezekowitz MD, Hagen P, Annex BH. Local treatment with recombinant tissue factor pathway inhibitor reduces the development of intimal hyperplasia in experimental vein grafts. J Vasc Surg. 2001;33:400–7. Smith GE, Gohil R, Chetter IC. Factors affecting the patency of arteriovenous fistulas for dialysis access. J Vasc Surg. 2012;55(3):849–55. Huijbregts HJ, Bots ML, Wittens CH, Schrama YC, Moll FL, Blankestijn PJ. CIMINO study group. Hemodialysis arteriovenous fistula patency revisited: results of a prospective, multicenter initiative. Clin J Am Soc Nephrol. 2008;3(3):714–9. Mortaz SS, Davati A, Ahmadloo MK, Taheri HR, Golfam F, Tavakoli A, Khalaj AR. Evaluation of patency of arteriovenous fistula and its relative complications in diabetic patients. Urol J. 2013;10(2):894–7. Schild AF, Perez E, Gillaspie E, Seaver C, Livingstone J, Thibonnier A. Arteriovenous fistulae vs. arteriovenous grafts: a retrospective review of 1,700 consecutive vascular access cases. J Vasc Access. 2008;9(4):231–5.
123
ORIGINAL ARTICLE
Relationship of late arteriovenous fistula stenosis with soluble E-selectin and soluble EPCR in chronic hemodialysis patients with arteriovenous fistula Mukadder Ayse Bilgic • Hakki Yilmaz • Alper Bozkurt • Huseyin Tugrul Celik • Ismail Celal Bilgic • Ozgul Malcok Gurel Ismail Kirbas • Nuket Bavbek • Ali Akcay
•
Received: 16 January 2014 / Accepted: 25 February 2014 Ó Japanese Society of Nephrology 2014
Abstract Background/aims Vascular access dysfunction caused by stenosis is a major complication for hemodialysis (HD) patients. However, physiopathology of late arteriovenous fistula (AVF) stenosis is still under investigation. The aim of the present study was to evaluate the association between plasma soluble EPCR (sEPCR) with serum soluble E-selectin (sE-selectin) concentration and late AVF stenosis in HD patients. Methods Plasma sEPCR and serum sE-selectin concentrations were measured in 94 HD patients. Using these data, we studied the association of sEPCR and sE-selectin with the presence and degree of AVF stenosis using ultrasonography and fistulogram. Results Fifty-one patients have AVF stenosis, and the others (n = 43) have patent AVF. The degree of AVF
M. A. Bilgic H. Yilmaz (&) N. Bavbek A. Akcay Section of Nephrology, Department of Internal Medicine, School of Medicine, Turgut Ozal University, Alparslan Tu¨rkes Cad. No: 57, 06510 Ankara, Turkey e-mail: [email protected] A. Bozkurt I. Kirbas Department of Interventional Radiology, School of Medicine, Turgut Ozal University, Ankara, Turkey H. T. Celik Department of Biochemistry, School of Medicine, Turgut Ozal University, Ankara, Turkey I. C. Bilgic Department of General Surgery, School of Medicine, Turgut Ozal University, Ankara, Turkey O. M. Gurel Department of Cardiology, School of Medicine, Turgut Ozal University, Ankara, Turkey
stenosis was correlated with serum sE-selectin levels (r = 0.351, p = 0.01), but not sEPCR (r = 0.075, p = 0.702). The median level of sE-selectin was statistically higher in the group of AVF stenosis than in the group of patent AVF [463.2 pg/ml (275.4–671.4) vs. 162.5 pg/ml (96.7–285.3), p = 0.001]. Increased sE-selectin levels [OR (OR) = 6.356, p = 0.015] and high levels of LDL (OR = 4.321, p = 0.044) were independent predictors of late AVF stenosis in the multivariate model. Conclusions sE-selectin and the LDL were the most important predictors of late AVF stenosis. In addition, sEselectin correlated with the degree of AVF stenosis. We suggested that atherosclerosis might be contributing factor for development of late AVF stenosis. Keywords Arteriovenous fistula stenosis Biomarkers End-stage renal disease Angiography
Introduction A native arteriovenous fistula (AVF) is the preferred form of permanent vascular access in patients with end-stage renal disease (ESRD) undergoing hemodialysis (HD) due to fewer complications and decreased mortality [1]. Vascular access dysfunction (VAD) is the main cause of morbidity [3], and the problems associated with VAD represent roughly 20 % to 25 % of hospitalizations of patients on maintenance HD [3, 4]. Vascular access malfunction associated with some medical factors such as stasis (hypotension, hypoalbuminemia, compression), hypercoagulability [antiphospholipid antibodies, hyperhomocysteinemia, factor V leiden, lipoprotein (a)], endothelial cell injury (preexisting intimal hyperplasia, TNF-a, oxidative stress, calcium (Ca)
123
Clin Exp Nephrol
phosphate deposition, activated platelets), medications and red cell mass [1–5]. However, these risk factors include diabetes; location of the graft; age [65 years; time until use of graft after surgical procedure; smoking; and increased serum cholesterol and triglyceride (TG) levels [1–5]. The pathogenesis of arteriovenous (AV) access stenosis is not understood enough, although VAD caused by stenosis is a leading complication for HD patients [3–5]. The most common cause of this VAD is venous stenosis as a result of progressive venous neointimal hyperplasia and inadequate outward remodeling within the peri-anastomotic region (AV fistula) [3–5]. The primary cause of AVF stenosis resembles the histology of atherosclerosis [6]. E-selectin [endothelial leukocyte adhesion molecule-1 (ELAM-1), CD62E] is a cell surface glycoprotein that is expressed on inflamed endothelial cells in response to treatment with inflammatory cytokines [7, 8]. E-selectin mediates the interactions of leukocytes and platelets with endothelial cells [7, 8]. E-selectin is associated with severe atherosclerosis in ESRD [9, 10]. Some studies revealed that E-selectin contributes to the process of intimal hyperplasia [11] and decreasing E-selectin expression inhibits neointimal hyperplasia [12] in rat models. We hypothesized that E-selectin can be a major contributing factor for the development of AVF stenosis via neointimal hyperplasia and inflammation. The protein C pathway is a primary regulator of blood coagulation, and it plays key roles in the critical component of the host response to inflammatory stimuli [13]. The recent member of this pathway is the endothelial protein C receptor (EPCR) [14]. EPCR is a 46 kDa, type 1 transmembrane protein with homology to CD1d/MHC class I proteins, expressed primarily on the larger vessels and also microvascular endothelium [14, 15]. EPCR binds PC and facilitates formation of activated protein C (APC), a potent anticoagulant and anti-inflammatory agent [14–16]. EPCR can be cleaved to release a soluble form (sEPCR) in the circulation [17]. Soluble EPCR (sEPCR) binds PC and APC with similar affinity [16, 17]. sEPCR inhibits both PC activation and APC anticoagulant activity [15–17]. The increased soluble EPCR was associated with thromboembolic (e.g., deep venous thrombosis [18], stroke [19], STsegment elevation myocardial infarction [20]) and systemic inflammatory diseases (e.g., sepsis [21] and systemic lupus erythematosus [21]). In addition, sEPCR has been shown to modulate inflammation by binding to activated neutrophils [22]. We showed that higher level of sEPCR was related to the development of AVF thrombosis in HD patients [23]. No study has as yet investigated whether sEPCR influences on the development of AVF stenosis or not in HD patients. The aim of present study was to examine the relationship between E-selectin, sEPCR and the development of AVF stenosis in HD patients.
123
Materials and methods Over a period of 13 months, from January 2012 to February 2013, a total of 325 patients with AVF access dysfunction and AVF access routine control, who were referred to Department of Interventional Radiology, Turgut Ozal University, School of Medicine, Ankara. 181 patients were excluded, who have multiple repetitive endovascular treatments. Other causes of AVF dysfunction (such as thrombosis and aneurysm) except stenosis were detected in 45 patients. Patients were eligible for inclusion if they were established on HD and between 20 and 80 years of age. Patients were excluded if they had a history of hematological malignancy, a known bleeding diathesis or coagulopathy or concomitant warfarin therapy (n = 5). The single-center study included total 94 patients treated with chronic HD for more than 6 months. To identify the possible risk factors for AVF dysfunction, variables including age; gender; co-morbidities such as diabetes, hypertension, coronary arterial occlusive disease and peripheral arterial occlusive disease; age of AVF; side of the AVF; type of AVF; number of lesions; degree of stenosis; and residual stenosis were collected. Color Doppler ultrasonography (CDUS) studies were performed on a Doppler ultrasound device (Philips, HD15 PureWave Ultrasound System, Bothell, WA, USA) with a 5-MHz linear probe (C5-1 PureWave transducer) by two blinded radiologists experienced in the use of Doppler US. At CDUS analysis, AVF flow volume, peak systolic (PSV) and end-diastolic velocities, fistula diameter and subcutaneous depth were measured. Fistula anastomosis area and calibration, flow velocities and lumens of vascular structures in the arterial and venous tree were analyzed. Visual narrowing (diameter narrowing[50 %) as assessed on gray scale imaging and an increased ratio of the PSV at the stenosis as compared to the PSV measured 2 cm upstream from the stenosis are two major features to characterize AVF stenosis [24]. The diagnostic criteria vary in accordance with the location of the stenosis [24]. Cases with a PSV ratio in the anastomosis region of 3 and above were regarded as significant (50 % and above) anastomosis stenosis [24]. A draining vein and inflow artery stenosis are defined by visible narrowing and by a PSV ratio of C2:1 [24]. In addition, observation of focal color aliasing in the stenosis region at CDUS and peak systolic velocities rising above 400 cm/s were regarded as other assistant findings [24]. The site of stenotic lesions was classified into six categories based upon location [25]: (1) native artery; (2) AV anastomosis; (3) juxta-anastomotic vein (initial 2 cm of fistula); (4) venous outflow [2 cm from anastomosis (cannulation zone); (5) distal outflow (defined as above elbow joint for radial-cephalic fistulae and above mid-humerus for brachial fistulae); and (6) central venous system.
Clin Exp Nephrol
Fistulography with digital subtraction angiography (DSA) examinations was performed by two experienced interventional radiologists using a digital subtraction system (Philips Allura Xper FD 20 Release 2, Best, The Netherlands). Fistulogram with DSA was performed in these patients (n = 57) who have abnormal ultrasound findings. The degree of stenosis was classified into four groups: 50–74, 75–89, 90–99 and 100 % [26]. An accessory vein was defined as a branch coming off of the main venous channel that comprised the fistula. Blood was drawn from all study subjects under standardized conditions before fistulogram was performed. Samples were stored at -80 °C until analysis. Serum creatinine, urea, aspartate aminotransferase (AST), alanine aminotransferase (ALT), Ca, albumin, uric acid, CRP, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), TG, Ca and phosphorus (P) were measured by standard laboratory techniques using an autoanalyzer (Roche Diagnostics, COBAS INTEGRA 800, Indianapolis, Indiana, USA). Complete blood count (CBC) analysis was also performed with the use of a Coulter counter technique (Coulter Gen-S Hematology Analyzer, Beckman Coulter Corp, Hialeah, Florida). Plasma sEPCR levels were measured by solid-phase enzyme-linked immunosorbent assay (Catalog No: SK00507-01; Aviscera Bioscience Inc., Santa Clara, CA, USA) with intra- and inter-assay CVs of \6 and \ 12 %, respectively. This assay has high sensitivity (lower limit of detection 36 pg/mL). Human soluble E-selectin (sE-selectin) was analyzed in the serum with an enzyme-linked immunosorbent assay (Catalog No: EK0501; BOSTER BIOLOGICAL TECHNOLOGY Co., Ltd., Fremont, CA USA). This assay has high sensitivity (lower limit of detection \4 pg/mL) and exhibits no detectable cross-reactivity with other relevant proteins. All samples were assayed in duplicate. Calibration and standardization of the assays were performed according to the manufacturer’s protocol. Continuous variables were defined mean ± standard deviation or median (25–75 %); categorical variables were given as percentages. The Kolmogorov–Smirnov test was applied for a normal distribution. The independent sample unpaired Student’s t test or the Mann–Whitney U test was used for the continuous variables, and categorical data were compared with the chi-square test. We performed univariate logistic regression analyses to assess the association of clinical and laboratory markers with the binary presence of late AVF stenosis (yes/no). We reduced the model using stepwise multivariate logistic regression analyses. Odds ratios (OR) and 95 % confidence intervals were calculated. A p value \0.05 was considered statistically significant. All analyses were performed using SPSS software (version 17.0; SPSS Inc., Chicago, IL).
This study is approved by local Ethics Committee (IRB approval number: 99950669/1203).
Results Forty-eight (n = 48) patients were admitted for fistula control. 46 patients who had clinical findings of AVF dysfunction (difficulties in cannulation, painful arm edema, prolonged bleeding time after cannulation or after removal of the dialysis needles) were included in the study. All HD patients included in the current study (n = 94) were evaluated by using CDUS. 56 patients (59.6 %) had a luminal narrowing more than 50 %. 57 patients (60.6 %) had a peak systolic velocity ratio higher than 2:1. There was no abnormal ultrasonography finding in 37 patients; after all, all of the patients were admitted for control. Fistulography was performed for all patients with abnormal ultrasound findings (n = 57). Severe stenosis (C50 %) was determined in 51 out of 57 patients who underwent fistulography. Therefore, 51 patients were included in the stenosis group, while 43 patients were considered in the group without stenosis. Demographic features of the patients are presented in Table 1. The mean age was 57.2 ± 15.8 years (range 20–80), and 52.1 % of the patients (n = 49) were male. Renal failure had various causes, but the major cause was diabetic nephropathy (n = 43, 45.7 %). The patients were divided into two groups: patients with stenosis (Group 1; n = 51) and patients without stenosis (Group 2; n = 43). There was no difference in age, gender and body mass index between Group 1 and Group 2 (Table 2). The duration of fistula use was similar between Table 1 Baseline characteristics of study population Hemodialysis patients Numbers
94
Male/female (%)
49 (52.1)/45 (47.9)
Age (years)
57.2 ± 15.8
Body mass index (kg/m2)
27.6 ± 6.36
Hemodialysis duration (months)
36.23 ± 13.54
Smoking (%)
36.1
Hypertension (%)
89.3
Etiology of chronic renal failure Diabetic nephropathy
43
Hypertension related nephropathy
32
Chronic primary glomerulonephritis Interstitial nephritis
9 3
Polycystic kidney disease
2
Obstructive nephropathy
1
Unknown
4
123
Clin Exp Nephrol Table 2 Comparison among two groups Parameters
Group I (n = 51)
Group II (n = 43)
p
Male/female
26/25
23/20
0.786
Age (years)
57.8 ± 12.3
56.9 ± 14.5
0.812
BMI (kg/m2)
27.8 ± 6.32
26.9 ± 5.72
0.345
Smoker [n (%)]
19 (37.2)
15 (34.8)
0.098
Duration of AVF use (months)
42.8 ± 12.6
38.5 ± 8.7
0.245
URR
73.5 ± 2.8
74.1 ± 3.0
0.235
Albumin (g/dl)
3.87 ± 0.36
3.91 ± 0.42
0.189
Ca (corrected) (mg/dl)
8.40 ± 0.52
8.50 ± 0.65
0.231
P (mg/dl)
5.80 ± 1.90
5.75 ± 1.79
0.365
iPTH (pg/ml)
332.8 ± 160.5
319.5 ± 204.3
0.105
HDL-C (mg/dl) Triglycerides (mg/dl)
39.6 ± 10.3 279.5 ± 90.4
38.7 ± 11.2 288.9 ± 106.6
0.305 0.411
LDL-C (mg/dl)
154.5 ± 32.6
128.7 ± 28.6
\0.001
Hb (g/dl)
10.85 ± 1.15
10.90 ± 1.26
0.542 0.259
TSAT (%)
38.7 ± 10.2
37.8 ± 11.8
Ferritin (ng/ml)
422.4 ± 240.8
439.1 ± 230.7
0.346
CRP (mg/l)
18.77 ± 20.48
13.28 ± 12.47
0.271
sE-selectins (pg/ml)
463.2 (275.4–671.4)
162.5 (96.7–285.3)
0.001
sEPCR (pg/ml)
8,345.6 (1,456.7–12,347.9)
8,278.0 (1,265.8–11,782.4)
0.472
ACE or ARB [n (%)]
16 (31.3)
13 (30.2)
0.613
ß-Blockers [n (%)]
15 (29.4)
12 (27.9)
0.567
CCB [n (%)]
31 (60.7)
27 (62.7)
0.231
Alpha-blockers [n (%)]
10 (19.6)
9 (20.9)
0.102
Group 1 and Group 2 (42.8 ± 12.6 vs. 38.5 ± 8.7 months, respectively; p = 0. 245). There was no difference in medication use and incidence of cardiovascular diseases between the groups (Table 2). Biochemical tests showed that patients in Group 1 had higher LDL cholesterol (154.5 ± 32.6 vs. 128.7 ± 28.6 mg/dl; p \ 0.001) and median sE-selectin levels [463.2 pg/ml (275.4–671.4) vs. 162.5 pg/ml (96.7–285.3); p = 0.001] compared with patients in Group 2 (Table 2). There was no difference in the median sEPCR [8,345.6 pg/ml (1,456.7–12,347.9) vs. 8278 pg/ml (1,265.8–11,782.4); p = 0.435] level and other biochemical parameters between the groups (Table 2). We summarized CDUS and fistulographic findings of AVF stenosis in Table 3. Correlation analysis showed that serum sE-selectin levels increased with increasing degree of AVF stenosis determined by fistulography (r = 0.351, p = 0.01). However, there was no correlation between sEPCR and the degree of AVF stenosis (r = 0.075, p = 0.702). LDL-C levels were not associated with location and degree of AVF stenosis (r = 0.123, p = 0.345; r = 0.098, p = 0.689). Presence of DM was not related to location and degree of AVF stenosis (r = 0.097, p = 0.876; r = 0.126, p = 0.213).
123
Parameters that would cause AVF stenosis were determined by using univariate analysis, and then multivariate logistic regression analysis was performed. In the multivariate model (Table 4), sE-selectin was an independent predictor of AVF stenosis in chronic HD patients (OR = 6.356, p = 0.015) together with LDL (OR = 4.321, p = 0.044). Presence of DM and duration of AVF use did not predict for late AVF stenosis (Table 4).
Discussion In this study, sE-selectin level was determined as an independent and strong predictor of AVF stenosis, which is proved by using Doppler ultrasonography and fistulogram. In addition, (1) there was no difference in sEPCR levels between patients with AVF stenosis and patients without AVF stenosis; (2) While sE-selectin level was associated with the degree of AVF stenosis, there was no association between sEPCR level and the degree of AVF stenosis; (3) LDL level was higher in patients with AVF stenosis, and LDL was found to be an independent predictor of AVF stenosis. Arteriovenous fistulas represent the preferred vascular access route for HD patients [1–4]. Complications related
Clin Exp Nephrol
to vascular access are common in patients undergoing long-term HD. These complications are a major cause of morbidity, and they may lead to the loss of vascular access if the appropriate intervention is not performed on time. Table 3 CDUS and Fistulographic findings of AVF stenosis Category
n (%)
C50 % \50 %
56 (59.6) 38 (40.4)
Color Doppler ultrasonography Luminal diameter reduction A peak systolic velocity ratio Peak systolic flow velocity
C2:1
57 (60.6)
\2:1
37 (39.4)
[400 cm/s
49 (52.1)
\400 cm/s
45 (47.9)
Fistulography Degree of stenosis Total occlusion
100 %
6 (11.7)
Stenosis
90–99 %
20 (39.3)
75–89 %
15 (29.4)
50–74 %
10 (19.6)
1 site
40 (78.5)
2 sites
9 (17.6)
C3 sites
2 (3.9)
Native artery Arteriovenous anastomosis
3 (5.9) 16 (31.3)
Juxta-anastomotic vein
22 (43.1)
Number of lesions
Location of lesions
Table 4 Significant predictors of late AVF stenosis in univariable and multivariable logistic regression analyses
Venous outflow
18 (35.3)
Distal outflow
5 (9.8)
Central venous system
0
Especially in the long term, stenosis development is one of the major causes of AVF malfunction [1–5]. While certain factors that may lead to AVF thrombosis have been defined, factors that would lead to AVF stenosis are not sufficiently defined [3–5]. In this study, we showed that sE-selectin level was higher in HD patients with isolated AVF stenosis, and sEselectin level was an independent predictor of AVF stenosis. It is possible to explain the association between sEselectin and AVF stenosis as follows: The major cause of AVF stenosis is the development of neointimal hyperplasia [5], and E-selectin has been shown to play an important role in intima hyperplasia in rat models [11]. In another recent study, administration of cilostazol, which is a selective phosphodiesterase-3 inhibitor with antiatherogenic and antiproliferative features, to rats suppressed neointimal hyperplasia related to decreased E-selectin levels [12]. Therefore, it is possible to assume that E-selectin levels are associated with neointimal hyperplasia, and AVF stenosis develops with the progression of neointimal hyperplasia. The findings of this study also support this hypothesis. Previous studies have shown that E-selectin levels are associated with atherosclerosis in HD patients [9, 10]. Stenosis of the AVF is histologically similar to the atherosclerotic lesion [6, 27]. In activated endothelial cells, E-selectin expression increases to contribute to the formation of atherosclerosis, while it is also possible that increased E-selectin expression may lead to AVF stenosis. One of the major findings of this study is the higher LDL levels in patients with AVF stenosis. High serum LDL
Univariate OR (95 % CII)
p value
Age (year)
1.968 (1.645–2.143)
0.041
BMI (kg/m2)
0.973 (0.968–1.039)
0.727
Smoker
1.097 (1.065–1.159)
0.853
Duration of AVF use
1.185 (0.876–1.967)
0.345
DM (yes/no) Ca (mg/dl)
1.251 (0.945–1.423) 1.041 (1.006–1.198)
0.186 0.389
P (mg/dl)
0.970 (0.845–1.135)
0.818
Albumin (g/dl)
0.988 (0.968–1.035)
0.913
LDL (mg/l)
5.071 (3.987–7.256)
\0.001
HDL (mg/l)
1.018 (0.986–1.349)
0.658
iPTH (pg/ml)
1.174 (1.083–1.389)
0.592
Hb (g/dl)
0.815 (0.703–0.985)
0.439
CRP (mg/l)
1.236 (1.090–1.408)
0.246
TSAT (%)
1.072 (1.008–1.121)
0.857
Ferritin
1.211 (0.861–1.812)
0.312
sEPCR
1.023 (1.012–1.031)
0.428
sE-selectin
7.234 (5.891–9.213)
\0.001
Multivariate OR (95 % CI)
p value
1.245 (0.971–1.943)
0.206
4.321 (3.189–6.057)
0.044
6.356 (4.123–8.521)
0.015
123
Clin Exp Nephrol
level is known to be a major risk factor for atherosclerosis [28]. Therefore, high LDL levels in patients with AVF stenosis imply that AVF stenosis can develop through mechanisms similar to that of atherosclerosis. As a result of LDL cholesterol reduction with statins, neointimal formation has been shown to decrease after stent implantation [29]. Recently, Janardhanan et al. [30] have shown that simvastatin resulted in a significant reduction in venous neointimal hyperplasia and venous stenosis formation through decreasing VEGF-A/MMP-9 pathway activity in dialysis vascular access model. These studies and the findings of the current study imply that mechanisms leading to atherosclerosis play an important role in the development of AVF stenosis, increased LDL level is one of the factors leading to AVF stenosis, and intensive LDL management may prevent AVF stenosis. We previously showed the association between sEPCR and AVF thrombosis [23]. sEPCR has been shown to be associated with not only venous thromboemboli, but also stroke, ST-elevated MI and sepsis [19–21]. However, the association of sEPCR with neointimal hyperplasia remains obscure. The exact mechanisms of neointimal hyperplasia, the primary cause of AVF stenosis, are not currently understood, but are thought to be associated with the cumulative effects of inflammation, hemodynamic factors and coagulation [5, 31]. In addition, activation of coagulation cascade may indirectly play a role in mediating the extent of intimal hyperplasia and stenosis [32]. Tissue factor pathway inhibitor, a critical inhibitor of coagulation, markedly reduced intimal hyperplasia formation in experimental vein graft [33]. There is no study on isolated AVF stenosis without thrombosis—i.e., the association between neointimal hyperplasia and sEPCR—in the literature. In this study, we showed that there was no difference in sEPCR levels between patients with AVF stenosis and patients without AVF stenosis. This finding implies that sEPCR has no role in the pathogenesis of neointimal hyperplasia. We found no the relationship between presence of diabetes, degree localization of AVF stenosis. In addition, presence of diabetes was not a predictor of late AVF stenosis according to our results. Some studies revealed that presence of diabetes was a factor for patency of AVF [34, 35]. But some studies showed that presence of diabetes did not influence of AVF patency [36, 37]. The present study has some limitations. First, we are aware of the limitations of a cross-sectional study, which limits attributions about the direction of causality between variables. Second, the other limitation of the study was relative small sample of our hospital. Third, soluble E-selectin and EPCR were measured once, but serial measurements could be more appropriate. Nevertheless, our findings require further examination in experimental and prospective studies.
123
Conclusion HD patients with AVF stenosis have higher level of sEselectin and LDL-C, but not sEPCR. LDL-C and sEselectin are independent predictors of AVF stenosis in chronic HD patients. Angiographic stenosis degree is correlated with sE-selectin levels. Our results showed that AVF stenosis mimics some mechanisms of atherosclerosis. Acknowledgments This study was supported by the Scientific Research Fund of Turgut Ozal University. Conflict of interest
We have no conflict of interest.
References 1. Sousa CN, Apo´stolo JL, Figueiredo MH, Martins MM, Dias VF. Physical examination: how to examine the arm with arteriovenous fistula. Hemodial Int. 2013;17(2):300–6. 2. Vachharajani TJ. Diagnosis of arteriovenous fistula dysfunction. Semin Dial. 2012;25(4):445–50. 3. Stolic R. Most important chronic complications of arteriovenous fistulas for hemodialysis. Med Princ Pract. 2013;22(3):220–8. 4. Rose DA, Sonaike E, Hughes K. Hemodialysis access. Surg Clin N Am. 2013;93(4):997–1012. 5. Roy-Chaudhury P, Sukhatme VP, Cheung AK. Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint. J Am Soc Nephrol. 2006;17(4):1112–27. 6. Gagliardi GM, Rossi S, Condino F, Mancuso D, Greco F, Tenuta R, Savino O, Bonofiglio R, Domma F, Latorre G. Malnutrition, infection and arteriovenous fistula failure: is there a link? J Vasc Access. 2011;12(1):57–62. 7. Galkina E, Ley K. Vascular adhesion molecules in atherosclerosis. Arterioscler Thromb Vasc Biol. 2007;27(11):2292–301. 8. Roldan V, Marin F, Lip GY, Blann AD. Soluble E-selectin in cardiovascular disease and its risk factors. A review of the literature. Thromb Haemost. 2003;90:1007–20. 9. Testa A, Benedetto FA, Spoto B, Pisano A, Tripepi G, Mallamaci F, Malatino LS, Zoccali C. The E-selectin gene polymorphism and carotid atherosclerosis in end-stage renal disease. Nephrol Dial Transplant. 2006;21(7):1921–6. 10. Issac MS, Afif A, Gohar NA, Fayek NA, Zayed B, Sedrak H, El Din LAS. Association of E-selectin gene polymorphism and serum papp-a with carotid atherosclerosis in end-stage renal disease. Mol Diagn Ther. 2013;. doi:10.1007/s40291-013-0061-4. 11. Gotoh R, Suzuki J, Kosuge H, Kakuta T, Sakamoto S, Yoshida M, Isobe M. E-selectin blockade decreases adventitial inflammation and attenuates intimal hyperplasia in rat carotid arteries after balloon injury. Arterioscler Thromb Vasc Biol. 2004;24(11):2063–8. 12. Takigawa T, Tsurushima H, Suzuki K, Tsuruta W, Nakamura K, Matsumura A. Cilostazol suppression of arterial intimal hyperplasia is associated with decreased expression of sialyl Lewis X homing receptors on mononuclear cells and E-selectin in endothelial cells. J Vasc Surg. 2012;55(2):506–16. 13. Bouwens EA, Stavenuiter F, Mosnier LO. Mechanisms of anticoagulant and cytoprotective actions of the protein C pathway. J Thromb Haemost. 2013;11(Suppl 1):242–53. 14. Kurosawa S, Stearns-Kurosawa DJ, Hidari N, Esmon CT. Identification of functional endothelial protein C receptor in human plasma. J Clin Invest. 1997;100:411–8. 15. Montes R, Puy C, Molina E, Hermida J. Is EPCR a multi-ligand receptor? Pros and cons. Thromb Haemost. 2012;107(5):815–26.
Clin Exp Nephrol 16. Gleeson EM, O’Donnell JS, Preston RJ. The endothelial cell protein C receptor: cell surface conductor of cytoprotective coagulation factor signaling. Cell Mol Life Sci. 2012;69(5): 717–26. 17. Liaw PCY, Neuenschwander PF, Smirnov MD, Esmon CT. Mechanisms by which soluble endothelial cell protein C receptor modulates protein C and activated protein C function. J Biol Chem. 2000;275:5447–52. 18. Saposnik B, Reny JL, Gaussem P, Emmerich J, Aiach M, Gandrille S. A haplotype of the EPCR gene is associated with increased plasma levels of sEPCR and is a candidate risk factor for thrombosis. Blood. 2004;103(4):1311–8. 19. Yu¨ru¨rer D, Teber S, Deda G, Egin Y, Akar N. The relation between cytokines, soluble endothelial protein C receptor, and factor VIII levels in Turkish pediatric stroke patients. Clin Appl Thromb Hemost. 2009;15(5):545–51. 20. Tanalp AC, Oduncu V, Erkol A, Go¨zu¨bu¨yu¨k G, Ozveren O, Du¨ndar C, Canbay A, Kirma C. Soluble endothelial protein C receptor levels and protein C activity in patients with acute STsegment elevation myocardial infarction. Coron Artery Dis. 2013;24(3):209–16. 21. Kurosawa S, Stearns-Kurosawa DJ, Carson CW, et al. Plasma levels of endothelial cell protein C receptor are elevated in patients with sepsis and systemic lupus erythematosus: lack of correlation with thrombomodulin suggests involvement of different pathological processes. Blood. 1998;91:725–7. 22. Kurosawa S, Esmon CT, Stearns-Kurosawa DJ. The soluble endothelial protein C receptor binds to activated neutrophils: involvement of proteinase-3 and CD11b/CD18. J Immunol. 2000;165:4697–703. 23. Bilgic A, Ozdemir FN, Bayraktar N, Karakus S, Sasak G, Arat Z, Sezer S, Haberal M. Soluble endothelial protein C receptor: influence on arteriovenous fistula thrombosis development in hemodialysis patients. Am J Nephrol. 2007;27(4):366–72. 24. Hutchison SJ, Holmes KC. Arteriovenous fistulas. In: Hutchison SJ, Holmes KC, editors. Principles of Vascular and Intravascular Ultrasound. Philadelphia: Saunders; 2012. p. 82–96. 25. Clark TW, Hirsch DA, Jindal KJ, et al. Outcome and prognostic factors of restenosis after percutaneous treatment of native hemodialysis fistulas. J Vasc Interv Radiol. 2002;13:51–9. 26. Vazquez MA. Vascular access for dialysis: recent lessons and new insights. Curr Opin Nephrol Hypertens. 2009;18:116–21. 27. Marrone D, Pertosa G, Simone S, Loverre A, Capobianco C, Cifarelli M, Memoli B, Schena FP, Grandaliano G. Local
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
activation of interleukin 6 signaling is associated with arteriovenous fistula stenosis in hemodialysis patients. Am J Kidney Dis. 2007;49(5):664–73. Grundy SC. Role of low-density lipoproteins in development of coronary artery atherosclerosis. In: Kreisberg RA, Segrest JP, editors. Plasma lipoproteins and coronary artery disease. Cambridge: Blackwell Scientific; 1992. p. 93–124. Takamiya Y, Miura S, Kawamura A, Tanigawa H, Zhang B, Iwata A, Nishikawa H, Matsuo K, Shirai K, Saku K. Intensive LOwering of BlOod pressure and low-density lipoprotein ChOlesterol with statin theraPy (LOBOCOP) may improve neointimal formation after coronary stenting in patients with coronary artery disease. Coron Artery Dis. 2009;20(4):288–94. Janardhanan R, Yang B, Vohra P, Roy B, Withers S, Bhattacharya S, Mandrekar J, Kong H, Leof EB, Mukhopadhyay D, Misra S. Simvastatin reduces venous stenosis formation in a murine hemodialysis vascular access model. Kidney Int. 2013;84(2): 338–52. Lee T. Novel paradigms for dialysis vascular access: downstream vascular biology—is there a final common pathway? Clin J Am Soc Nephrol. 2013;. doi:10.2215/CJN.03490413. Hasenstab D, Lea H, Hart CE, Lok S, Clowes AW. Tissue factor overexpression in rat arterial neointima models thrombosis and progression of advanced atherosclerosis. Circulation. 2000;101: 2651–7. Huynh TT, Davies MG, Thompson MA, Ezekowitz MD, Hagen P, Annex BH. Local treatment with recombinant tissue factor pathway inhibitor reduces the development of intimal hyperplasia in experimental vein grafts. J Vasc Surg. 2001;33:400–7. Smith GE, Gohil R, Chetter IC. Factors affecting the patency of arteriovenous fistulas for dialysis access. J Vasc Surg. 2012;55(3):849–55. Huijbregts HJ, Bots ML, Wittens CH, Schrama YC, Moll FL, Blankestijn PJ. CIMINO study group. Hemodialysis arteriovenous fistula patency revisited: results of a prospective, multicenter initiative. Clin J Am Soc Nephrol. 2008;3(3):714–9. Mortaz SS, Davati A, Ahmadloo MK, Taheri HR, Golfam F, Tavakoli A, Khalaj AR. Evaluation of patency of arteriovenous fistula and its relative complications in diabetic patients. Urol J. 2013;10(2):894–7. Schild AF, Perez E, Gillaspie E, Seaver C, Livingstone J, Thibonnier A. Arteriovenous fistulae vs. arteriovenous grafts: a retrospective review of 1,700 consecutive vascular access cases. J Vasc Access. 2008;9(4):231–5.
123
