Original Paper Received: February 21, 2013 Accepted: August 23, 2013 Published online: October 25, 2013
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
Arteriovenous Fistula Aneurysm in Patients on Regular Hemodialysis: Prevalence and Risk Factors Aleksandar Jankovic a Branislav Donfrid b Jelena Adam b Marjan Ilic b Zivka Djuric a Tatjana Damjanovic a Jovan Popovic a Gordana Popovic a Zoran Radojicic c Nada Dimkovic a, d
a Clinical Department for Renal Diseases, b Clinic for Surgery, Zvezdara University Medical Center, c Faculty of Organizational Sciences, Institute for Statistics, d Medical Faculty, University of Belgrade, Belgrade, Serbia
Abstract Background/Aims: Compared to all other complications, literature data about vascular access aneurysm (VAA) are the scarcest. The aim of this cross-sectional study was to evaluate the prevalence of arteriovenous fistula (AVF) aneurysms and to confirm the risk factors for their appearance. Methods: The presence, number and morphological characteristics of AVF aneurysms were confirmed, and according to the score of AVF aneurysm (the sum of the length and width in cm), patients were classified into group 1 (score ≤12) and group 2 (score >12). Analysis included the last data from the medical records including vascular calcifications score. Results: Out of 181 patients, 150 with native fistula were included in this study. Aneurysmatic changes were detected in 90 (60%) patients, and the majority had two or more aneurysms. VAA were more frequent in patients with adult polycystic kidney disease (ADPKD) than in other diagnostic categories. By using forward stepwise logistic regression, we confirmed that patients on high-flux hemodialysis (HD) had 5.3-fold higher risk, and patients with diabetes mellitus had
© 2013 S. Karger AG, Basel 1660–2110/13/1242–0094$38.00/0 E-Mail [email protected] www.karger.com/nec
5.8-fold less risk for developing AVF aneurysm. While vascular calcification score did not influence the incidence of VAA, higher PWV had significant negative influence on formation of AVF aneurysm (OR 1.25, 95% CI 1.003–1.56, p = 0.047). By ROC curve analysis, it was determined that patients who were longer than 5.7 years on HD had greater risk for developing VAA (area = 0.741, p = 0.000). Conclusion: This single-center study confirmed the very high prevalence of VAA (60%). Aneurysms were more frequent in patients with ADPKD and in those who had longer dialysis vintage on high-flux membranes with higher blood flow rate. © 2013 S. Karger AG, Basel
Introduction
A potent vascular access (VA) is a major prerequisite for adequate hemodialysis (HD), and the type of VA is highly connected to patient mortality: it is significantly higher in patients with vascular catheter (15.2%) as compared to patients with vascular graft (9.1%) or native fistula (7.3%) [1]. Similar results have been observed by the CHOICE study: mortality was 16.1% for patients who were dialyzed by vascular catheters, 14.2% for patients dialyzed by vascular graft and 11.7% for patients dialyzed Aleksandar Jankovic, MD, MrSci Dimitrija Tucovica 161 SRB–11000 Belgrade (Serbia) E-Mail sashajan22 @ yahoo.com
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
Key Words Blood flow rate · Calcification score · Dialysis membrane · Pulse wave velocity · Vascular access aneurysm
Patients and Methods Patients This single-center cross-sectional study included 181 patients treated with chronic HD for more than 6 months at Clinical Department for Renal Diseases, Zvezdara University Medical Center, Belgrade. Patients were dialyzed 4–5 h three times per week by using low-flux or high-flux polysulfone membranes. All AVF were within the forearm region (93 on the left and 57 on the right hand). According to the clinical assessment, all aneurysmal changes of at least 1 cm in width were classified as VAAs. Measurement was always done by the same physician by using a ruler. According to the score of AVF aneurysm (the sum of the length and width in cm), patients were classified into group 1 (score ≤12) and group 2 (score >12). Apart from general data, analysis included the last data from the medical records at the time of analysis: type of VA, HD vintage, characteristics of dialysis membrane, blood flow during HD, mineral metabolism indices, the use (and dose) of vitamin D analogues and previous surgical intervention on VA. Out of 181 patients, 150 had native fistula, 26 had vascular graft and 5 had permanent vascular catheter, consecutively. Patients with grafts and vascular catheters were excluded from further analysis. Out of 150 patients (men/women 86/64) with AVF, blood flow rate during HD was 250 ml/min in 72 (48%), consecutively. Low-flux membranes were used in 42 (28%) patients and high-flux in 108 (72%): among them, 22 (15% out of overall) were treated by hemodiafiltration. Cannulation was performed using the rope-ladder technique. Skin quality was examined routinely in all patients before puncturing. In the case of skin change and threatening rupture, patients were referred for aneurysmectomy (14 patients).
surements, and data were expressed as a mean value of two measurements as previously described [6]. Statistical calculations were performed using the SPSS 16.0 software program. Data were expressed as percentages for discrete factors, and mean values for continuous variables. Medians and interquartile range were used for continuous variables without normal distribution. Statistical analyses included descriptive statistics, exploratory analyses and cross-table statistical tests. For discrete variables, the χ2 test and likelihood ratio were performed to estimate differences between groups. Differences between groups without normal distribution were analyzed by the Mann-Whitney test. To divide all patients with AVF aneurysm into two different groups, we used cluster analysis. Forward stepwise logistic regression was used to determine odds ratios for parameters that have an influence on AVF aneurysm formation. ROC curve analysis was used to determine cutoff values for significant variables. In all comparisons, a p value 12 (RR 2.007, OR 2.78, p = 0.035), indicating that calcitriol use was associated with lower aneurysmatic score. Table 1 shows the risk factors for AVF aneurysm. There was a significant positive association between aneurysm appearance and blood access flow rate and a negative association between AVF aneurysm and PWV and dose of calcitriol. The effect of vascular calcifications on AVF aneurysm prevalence was not significant. By ROC curve analysis, it was determined that patients who were treated with HD for more than 5.7 years had greater risk for developing VAA (area = 0.741, p = 0.000; fig. 1).
Prevalence and Risk Factors of AVF Aneurysms
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
95
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
by native fistula [2]. According to data from USRDS, VA complications are the most frequent reason for hospitalization of HD patients [3]. These complications include maturation failure, VA stenosis, thrombosis, ischemia, infection and VA-associated congestive heart failure. Compared to all other complications, literature data about VA aneurysm (VAA) are the scarcest. Therefore, the aim of this study was to evaluate the prevalence of arteriovenous fistula (AVF) aneurysms and to confirm the risk factors for their appearance.
Table 1. Risk factors and AVF aneurysm
AVF aneurysm
Aneurysm score
yes
no
p
Age, years Males/females Blood flow rate, ml/min Total calcification score3 Calcification of AVF PWV iPTH, pg/ml Calcitriol dose4
60.0±15.5 51/39 270±50 5±7 1±2 8.56±4.69 662±565 1.50±1.75
61.0±21.5 35/25 250±38 4±3 0±3 10.10±5.45 613±606 1.75±0.63
0.283 0.8402 0.006 0.055 0.094 0.008 0.268 0.043
1
group 1
group 2
p1
61±17 32/26 260±35 5±7 1±2 8.69±5.04 662±861 1.5±1.88
60±16.75 19/13 300±49 3±7 1±2 7.82±3.12 665±541 1.625±2.06
0.980 0.7002 0.270 0.246 0.271 0.329 0.680 0.712
iPTH = Intact parathyroid hormone. 1 Mann-Whitney test; 2 χ2; 3 Adragao score + vascular access calcification score; 4 number of tablets (25 μg) per week. Values are expressed as median±interquartile range unless otherwise indicated.
Table 2. Influence of dialysis modality, underlying renal disease, use of calcitriol, diabetes and ACEi usage on the appearance of AVF aneurysm and on the aneurysm score
AVF aneurysm
Aneurysm score
yes
no
p
group 1
group 2
p
Dialysis modality LFHD HFHD HDF
18 (20.0%) 58 (64.4%) 14 (15.6%)
24 (40.0%) 27 (45.0%) 9 (15.0%)
0.023
14 (23.7%) 35 (59.3%) 10 (16.9%)
4 (12.9%) 23 (74.2%) 4 (12.9%)
0.350
Underlying renal disease HTA DM ADPKD Pyelonephritis GN Unknown
43 (47.8%) 3 (3.3%) 15 (16.7%) 14 (15.6%) 12 (13.3%) 3 (3.3%)
29 (48.3%) 13 (21.7%) 3 (5.0%) 9 (15.0%) 4 (6.7%) 2 (3.3%)
0.005
31 (52.5%) 3 (5.1%) 9 (15.3%) 9 (15.3%) 5 (8.5%) 2 (3.4%)
12 (38.7%) 0 (0.0%) 6 (19.4%) 5 (16.1%) 7 (22.6%) 1 (3.2%)
0.266
Calcitriol use Yes No
37 (41.1%) 53 (58.9%)
20 (33.3%) 40 (66.7%)
0.336
29 (49.2%) 30 (50.8%)
8 (25.8%) 23 (74.2%)
0.032
DM Yes No
15 (25.0%) 45 (75.0%)
0.000
4 (4.4%) 86 (95.6%)
4 (6.8%) 55 (93.2%)
0 (0.0%) 31 (100.0%)
0.062
ACEi usage Yes No
41 (45.6%) 49 (54.4%)
32 (53.3%) 28 (46.7%)
0.350
27 (45.7%) 32 (54.2%)
14 (45.2%) 17 (54.8%)
0.798
LFHD = Low-flux hemodialysis; HFHD = high-flux hemodialysis; HDF = hemodiafiltration; HTA = hypertension; GN = glomerulonephritis; ACEi = angiotensin-converting enzyme inhibitor.
96
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
Jankovic et al.
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
Prevalence and Risk Factors of AVF Aneurysms
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
1.0
Sensitivity
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1.0
1 – Specificity
Fig. 1. ROC curve – determination of cutoff values for HD vintage. Area = 0.741.
Table 3. Forward stepwise logistic regression of parameters that influenced formation of AVF aneurysm
Parameter
p
OR
95% CI
PWV HFHD DM
0.047 0.012 0.022
1.25 5.284 5.81
1.003–1.56 1.444–19.329 1.28–26.32
By using forward stepwise logistic regression analysis, we confirmed that patients on high-flux HD had 5.3-fold higher risk for developing AVF aneurysm (table 3). Patients with DM had 5.8-fold lower risk for developing AVF aneurysm. Higher PWV had significant negative influence on formation of AVF aneurysm (OR 1.25, 95% CI 1.003–1.56, p = 0.047).
Discussion
97
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
Color version available online
Although potentially serious and life-threatening complications, VAAs are rarely the topic of the current literature. For aneurysmal changes, we used a subjective method since we did not find any other specific method-
ology regarding HD patients in the current literature. Most of the studies are case reports or reports on small numbers of patients, and this study contributes to the topic since we analyzed 150 HD patients who will be followed as a cohort. Our data confirmed that aneurysmatic changes are frequent: they were confirmed in 60% of our dialysis population. This is in agreement with data presented by Derakhshanfar et al. [7], who found that 51% of their dialysis population had aneurysmatic changes. In their 5-year examination of VA survival/complications, Brems et al. [8] reported VAA in only 6% of patients, while Raju [9] observed aneurysms in 16% of patients. The possible explanation for the difference between our and their reports is that our patients had native fistulas instead of vascular grafts. The role of method of cannulation in VAA incidence has to be confirmed. The most frequent method of cannulation uses change of cannulation place on every dialysis (rope-ladder technique), which is difficult to achieve in patients with ‘short fistula’ [10]. Two independent groups from Netherlands [10, 11] reported lower incidence of VAA in patients who were cannulated at the same place on every HD (buttonhole technique). In our group of patients, the buttonhole technique was not applied. Our study confirmed the role of underlying renal disease in aneurysmatic changes: aneurysms were more frequent in patients with APCKD. At the same time, patients with diabetes had AVF aneurysm less frequently than patients without diabetes. These results are difficult to interpret since the consequences of both diabetes and hypertension on blood vessels are well known. It is possible that these patients may have stiff, calcified blood vessels that could ‘protect’ them from the aneurysmatic changes. The present study confirmed that dialysis vintage significantly influenced VAA formation. Patients that used the same VA for more than 5.7 years were at a higher risk of developing VAA. We also confirmed that dialysis membrane and blood flow rate may have an influence on the formation of aneurysmatic changes: dialysis on highflux polysulfone membrane and with higher blood flow rate carried a higher risk for VAA. Similar data in the literature are missing, but we may speculate that higher blood flow rate may contribute to vascular stress in the long term. Mineral metabolism disturbances are a well-known complication of chronic renal failure and may lead to vascular and soft tissue calcifications. There are no literature data about the influence of impaired mineral metabolism on VAA formation. In our study, patients without VAA used a significantly higher dose of calcitriol and had high-
er PWV, but calcification score was not found to be a significant risk factor for VAA formation. Of the proposed risk factors, only dialysis vintage had an influence on the score of aneurysms. These results are difficult to interpret since similar results in the literature are missing. Angiotensin II and transforming growth factor-βmediated degradation of elastin, collagen and the other constituents of the vessel wall may be of importance in VAA formation as well as matrix metalloproteinase (MMP). These proteolytic enzymes are secreted by vascular smooth muscle cells and by inflammatory cells [12–14]. In patients with aneurysm of abdominal aorta, authors found a high concentration of MMP-9 [15], while the mice genetically deficient in MMP-2 and MMP-9 are resistant to developing aneurysm [12]. Similar data in humans are missing, but in case such connection exists, the ACE inhibitors could be an important preventive strategy for the VAA in dialysis patients. In our group of patients, angiotensin-converting enzyme inhibitor use was equal in patients with and without VAA. In addition, we did not find that ACE and MMP-3 gene polymorphism influence aneurysmatic changes (data not shown). This finding has to be confirmed in a large population of patients with aneurysm. The present study has several weak points. Aneurysmatic changes have not been visualized by ultrasonogram or angiogram where more precise details could be available including high- or down-flow stenosis. Unfortunate-
ly, we have no data about the consequences these aneurysms may have on the cardiovascular system. Apart from the present risk factors, there could be other potential risk factors that were not included in this analysis. Finally, clinical applicability of previous findings has to be confirmed. Higher blood flow rate during HD is beneficial for dialysis adequacy but it carries a risk for VAA formation, and longer, slow dialysis could be a solution. Also, attention should be directed toward careful puncturing of AVF and more frequent monitoring of complications such as rupture and cardiovascular consequences once aneurysmatic changes become apparent. Taken together, results presented in this study should be taken with caution and need to be confirmed by further investigations.
Conclusion
The prevalence of VAA in our dialysis population was very high (60%), and in more than half (71%) they were multiple. Aneurysms were more frequent in patients with ADPKD than in other diagnostic categories, and in those who had longer dialysis vintage on high-flux membranes with higher blood flow rate during HD. To establish an adequate preventive and therapeutic strategy, we need more prospective studies and detailed analysis of additional risk factors for VAA.
1 Pastan S, Soucie JM, McClellan WM: Vascular access and increased risk of death among hemodialysis patients. Kidney Int 2002; 62: 620–626. 2 Astor BC, Eustace JA, Powe NR, Klag MJ, Fink NE, Coresh J, CHOICE Study: Type of vascular access and survival among incident hemodialysis patients: the Choices for Healthy Outcomes in Caring for ESRD (CHOICE) Study. J Am Soc Nephrol 2005;16:1449–1455. 3 US Renal Data System (2007). USRDS annual data report: atlas of chronic kidney disease in the United States. http://www.usrds.org/atlas. htm (accessed 6 August 2008). 4 Adragao T, Pires A, Lucas C, Birne R, Magalhaes L, Gonçalves M, et al: A simple vascular calcification score predicts cardiovascular risk in haemodialysis patients. Nephrol Dial Transplant 2004;19:1480–1488. 5 Schlieper G, Krüger T, Djuric Z, Damjanovic T, Markovic N, Schurgers LJ, et al: Vascular access calcification predicts mortality in hemodialysis patients. Kidney Int 2008; 74: 1582–1587.
98
6 Schlieper G, Westenfeld R, Krüger T, Cranenburg EC, Magdeleyns EJ, Brandenburg VM, Djuric Z, Damjanovic T, Ketteler M, Vermeer C, Dimkovic N, Floege J, Schurgers LJ: Circulating nonphosphorylated carboxylated matrix gla protein predicts survival in ESRD. J Am Soc Nephrol 2011;22:387–395. 7 Derakhshanfar A, Gholyaf M, Niayesh A, Bahiraii S: Assessment of frequency of complications of arteriovenous fistula in patients on dialysis: a two-year single center study from Iran. Saudi J Kidney Dis Transpl 2009; 20: 872–875. 8 Brems J, Castaneda M, Garvin PJ: A five-year experience with the bovine heterograft for vascular access. Arch Surg 1986;121:941–944. 9 Raju S: PTFE grafts for hemodialysis access. Techniques for insertion and management of complications. Ann Surg 1987; 206: 666– 673. 10 Verhallen AM, Kooistra MP, van Jaarsveld BC: Cannulating in haemodialysis: rope-ladder or buttonhole technique? Nephrol Dial Transplant 2007;22:2601–2604.
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
11 Van Loon MM, Goovaerts T, Kessels AG, van der Sande FM, Tordoir JH: Buttonhole needling of haemodialysis arteriovenous fistulae results in less complications and interventions compared to the rope-ladder technique. Nephrol Dial Transplant 2010;25:225–230. 12 Longo GM, Xiong W, Greiner TC, Zhao Y, Fiotti N, Baxter BT: Matrix metalloproteinases 2 and 9 work in concert to produce aortic aneurysms. J Clin Invest 2002;110:625–632. 13 Wilson WR, Anderton M, Schwalbe EC, Jones JL, Furness PN, Bell PR, et al: Matrix metalloproteinase-8 and -9 are increased at the site of abdominal aortic aneurysm rupture. Circulation 2006;113:438–445. 14 Longo GM, Buda SJ, Fiotta N, Xiong W, Griener T, Shapiro S, Baxter BT: MMP-12 has a role in abdominal aortic aneurysms in mice. Surgery 2005;137:457–462. 15 Wilson WR, Anderton M, Choke EC, Dawson J, Loftus IM, Thompson MM: Elevated plasma MMP1 and MMP9 are associated with abdominal aortic aneurysm rupture. Eur J Vasc Endovasc Surg 2008;35:580–584.
Jankovic et al.
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
References
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
Arteriovenous Fistula Aneurysm in Patients on Regular Hemodialysis: Prevalence and Risk Factors Aleksandar Jankovic a Branislav Donfrid b Jelena Adam b Marjan Ilic b Zivka Djuric a Tatjana Damjanovic a Jovan Popovic a Gordana Popovic a Zoran Radojicic c Nada Dimkovic a, d
a Clinical Department for Renal Diseases, b Clinic for Surgery, Zvezdara University Medical Center, c Faculty of Organizational Sciences, Institute for Statistics, d Medical Faculty, University of Belgrade, Belgrade, Serbia
Abstract Background/Aims: Compared to all other complications, literature data about vascular access aneurysm (VAA) are the scarcest. The aim of this cross-sectional study was to evaluate the prevalence of arteriovenous fistula (AVF) aneurysms and to confirm the risk factors for their appearance. Methods: The presence, number and morphological characteristics of AVF aneurysms were confirmed, and according to the score of AVF aneurysm (the sum of the length and width in cm), patients were classified into group 1 (score ≤12) and group 2 (score >12). Analysis included the last data from the medical records including vascular calcifications score. Results: Out of 181 patients, 150 with native fistula were included in this study. Aneurysmatic changes were detected in 90 (60%) patients, and the majority had two or more aneurysms. VAA were more frequent in patients with adult polycystic kidney disease (ADPKD) than in other diagnostic categories. By using forward stepwise logistic regression, we confirmed that patients on high-flux hemodialysis (HD) had 5.3-fold higher risk, and patients with diabetes mellitus had
© 2013 S. Karger AG, Basel 1660–2110/13/1242–0094$38.00/0 E-Mail [email protected] www.karger.com/nec
5.8-fold less risk for developing AVF aneurysm. While vascular calcification score did not influence the incidence of VAA, higher PWV had significant negative influence on formation of AVF aneurysm (OR 1.25, 95% CI 1.003–1.56, p = 0.047). By ROC curve analysis, it was determined that patients who were longer than 5.7 years on HD had greater risk for developing VAA (area = 0.741, p = 0.000). Conclusion: This single-center study confirmed the very high prevalence of VAA (60%). Aneurysms were more frequent in patients with ADPKD and in those who had longer dialysis vintage on high-flux membranes with higher blood flow rate. © 2013 S. Karger AG, Basel
Introduction
A potent vascular access (VA) is a major prerequisite for adequate hemodialysis (HD), and the type of VA is highly connected to patient mortality: it is significantly higher in patients with vascular catheter (15.2%) as compared to patients with vascular graft (9.1%) or native fistula (7.3%) [1]. Similar results have been observed by the CHOICE study: mortality was 16.1% for patients who were dialyzed by vascular catheters, 14.2% for patients dialyzed by vascular graft and 11.7% for patients dialyzed Aleksandar Jankovic, MD, MrSci Dimitrija Tucovica 161 SRB–11000 Belgrade (Serbia) E-Mail sashajan22 @ yahoo.com
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
Key Words Blood flow rate · Calcification score · Dialysis membrane · Pulse wave velocity · Vascular access aneurysm
Patients and Methods Patients This single-center cross-sectional study included 181 patients treated with chronic HD for more than 6 months at Clinical Department for Renal Diseases, Zvezdara University Medical Center, Belgrade. Patients were dialyzed 4–5 h three times per week by using low-flux or high-flux polysulfone membranes. All AVF were within the forearm region (93 on the left and 57 on the right hand). According to the clinical assessment, all aneurysmal changes of at least 1 cm in width were classified as VAAs. Measurement was always done by the same physician by using a ruler. According to the score of AVF aneurysm (the sum of the length and width in cm), patients were classified into group 1 (score ≤12) and group 2 (score >12). Apart from general data, analysis included the last data from the medical records at the time of analysis: type of VA, HD vintage, characteristics of dialysis membrane, blood flow during HD, mineral metabolism indices, the use (and dose) of vitamin D analogues and previous surgical intervention on VA. Out of 181 patients, 150 had native fistula, 26 had vascular graft and 5 had permanent vascular catheter, consecutively. Patients with grafts and vascular catheters were excluded from further analysis. Out of 150 patients (men/women 86/64) with AVF, blood flow rate during HD was 250 ml/min in 72 (48%), consecutively. Low-flux membranes were used in 42 (28%) patients and high-flux in 108 (72%): among them, 22 (15% out of overall) were treated by hemodiafiltration. Cannulation was performed using the rope-ladder technique. Skin quality was examined routinely in all patients before puncturing. In the case of skin change and threatening rupture, patients were referred for aneurysmectomy (14 patients).
surements, and data were expressed as a mean value of two measurements as previously described [6]. Statistical calculations were performed using the SPSS 16.0 software program. Data were expressed as percentages for discrete factors, and mean values for continuous variables. Medians and interquartile range were used for continuous variables without normal distribution. Statistical analyses included descriptive statistics, exploratory analyses and cross-table statistical tests. For discrete variables, the χ2 test and likelihood ratio were performed to estimate differences between groups. Differences between groups without normal distribution were analyzed by the Mann-Whitney test. To divide all patients with AVF aneurysm into two different groups, we used cluster analysis. Forward stepwise logistic regression was used to determine odds ratios for parameters that have an influence on AVF aneurysm formation. ROC curve analysis was used to determine cutoff values for significant variables. In all comparisons, a p value 12 (RR 2.007, OR 2.78, p = 0.035), indicating that calcitriol use was associated with lower aneurysmatic score. Table 1 shows the risk factors for AVF aneurysm. There was a significant positive association between aneurysm appearance and blood access flow rate and a negative association between AVF aneurysm and PWV and dose of calcitriol. The effect of vascular calcifications on AVF aneurysm prevalence was not significant. By ROC curve analysis, it was determined that patients who were treated with HD for more than 5.7 years had greater risk for developing VAA (area = 0.741, p = 0.000; fig. 1).
Prevalence and Risk Factors of AVF Aneurysms
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
95
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
by native fistula [2]. According to data from USRDS, VA complications are the most frequent reason for hospitalization of HD patients [3]. These complications include maturation failure, VA stenosis, thrombosis, ischemia, infection and VA-associated congestive heart failure. Compared to all other complications, literature data about VA aneurysm (VAA) are the scarcest. Therefore, the aim of this study was to evaluate the prevalence of arteriovenous fistula (AVF) aneurysms and to confirm the risk factors for their appearance.
Table 1. Risk factors and AVF aneurysm
AVF aneurysm
Aneurysm score
yes
no
p
Age, years Males/females Blood flow rate, ml/min Total calcification score3 Calcification of AVF PWV iPTH, pg/ml Calcitriol dose4
60.0±15.5 51/39 270±50 5±7 1±2 8.56±4.69 662±565 1.50±1.75
61.0±21.5 35/25 250±38 4±3 0±3 10.10±5.45 613±606 1.75±0.63
0.283 0.8402 0.006 0.055 0.094 0.008 0.268 0.043
1
group 1
group 2
p1
61±17 32/26 260±35 5±7 1±2 8.69±5.04 662±861 1.5±1.88
60±16.75 19/13 300±49 3±7 1±2 7.82±3.12 665±541 1.625±2.06
0.980 0.7002 0.270 0.246 0.271 0.329 0.680 0.712
iPTH = Intact parathyroid hormone. 1 Mann-Whitney test; 2 χ2; 3 Adragao score + vascular access calcification score; 4 number of tablets (25 μg) per week. Values are expressed as median±interquartile range unless otherwise indicated.
Table 2. Influence of dialysis modality, underlying renal disease, use of calcitriol, diabetes and ACEi usage on the appearance of AVF aneurysm and on the aneurysm score
AVF aneurysm
Aneurysm score
yes
no
p
group 1
group 2
p
Dialysis modality LFHD HFHD HDF
18 (20.0%) 58 (64.4%) 14 (15.6%)
24 (40.0%) 27 (45.0%) 9 (15.0%)
0.023
14 (23.7%) 35 (59.3%) 10 (16.9%)
4 (12.9%) 23 (74.2%) 4 (12.9%)
0.350
Underlying renal disease HTA DM ADPKD Pyelonephritis GN Unknown
43 (47.8%) 3 (3.3%) 15 (16.7%) 14 (15.6%) 12 (13.3%) 3 (3.3%)
29 (48.3%) 13 (21.7%) 3 (5.0%) 9 (15.0%) 4 (6.7%) 2 (3.3%)
0.005
31 (52.5%) 3 (5.1%) 9 (15.3%) 9 (15.3%) 5 (8.5%) 2 (3.4%)
12 (38.7%) 0 (0.0%) 6 (19.4%) 5 (16.1%) 7 (22.6%) 1 (3.2%)
0.266
Calcitriol use Yes No
37 (41.1%) 53 (58.9%)
20 (33.3%) 40 (66.7%)
0.336
29 (49.2%) 30 (50.8%)
8 (25.8%) 23 (74.2%)
0.032
DM Yes No
15 (25.0%) 45 (75.0%)
0.000
4 (4.4%) 86 (95.6%)
4 (6.8%) 55 (93.2%)
0 (0.0%) 31 (100.0%)
0.062
ACEi usage Yes No
41 (45.6%) 49 (54.4%)
32 (53.3%) 28 (46.7%)
0.350
27 (45.7%) 32 (54.2%)
14 (45.2%) 17 (54.8%)
0.798
LFHD = Low-flux hemodialysis; HFHD = high-flux hemodialysis; HDF = hemodiafiltration; HTA = hypertension; GN = glomerulonephritis; ACEi = angiotensin-converting enzyme inhibitor.
96
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
Jankovic et al.
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
Prevalence and Risk Factors of AVF Aneurysms
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
1.0
Sensitivity
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
1.0
1 – Specificity
Fig. 1. ROC curve – determination of cutoff values for HD vintage. Area = 0.741.
Table 3. Forward stepwise logistic regression of parameters that influenced formation of AVF aneurysm
Parameter
p
OR
95% CI
PWV HFHD DM
0.047 0.012 0.022
1.25 5.284 5.81
1.003–1.56 1.444–19.329 1.28–26.32
By using forward stepwise logistic regression analysis, we confirmed that patients on high-flux HD had 5.3-fold higher risk for developing AVF aneurysm (table 3). Patients with DM had 5.8-fold lower risk for developing AVF aneurysm. Higher PWV had significant negative influence on formation of AVF aneurysm (OR 1.25, 95% CI 1.003–1.56, p = 0.047).
Discussion
97
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
Color version available online
Although potentially serious and life-threatening complications, VAAs are rarely the topic of the current literature. For aneurysmal changes, we used a subjective method since we did not find any other specific method-
ology regarding HD patients in the current literature. Most of the studies are case reports or reports on small numbers of patients, and this study contributes to the topic since we analyzed 150 HD patients who will be followed as a cohort. Our data confirmed that aneurysmatic changes are frequent: they were confirmed in 60% of our dialysis population. This is in agreement with data presented by Derakhshanfar et al. [7], who found that 51% of their dialysis population had aneurysmatic changes. In their 5-year examination of VA survival/complications, Brems et al. [8] reported VAA in only 6% of patients, while Raju [9] observed aneurysms in 16% of patients. The possible explanation for the difference between our and their reports is that our patients had native fistulas instead of vascular grafts. The role of method of cannulation in VAA incidence has to be confirmed. The most frequent method of cannulation uses change of cannulation place on every dialysis (rope-ladder technique), which is difficult to achieve in patients with ‘short fistula’ [10]. Two independent groups from Netherlands [10, 11] reported lower incidence of VAA in patients who were cannulated at the same place on every HD (buttonhole technique). In our group of patients, the buttonhole technique was not applied. Our study confirmed the role of underlying renal disease in aneurysmatic changes: aneurysms were more frequent in patients with APCKD. At the same time, patients with diabetes had AVF aneurysm less frequently than patients without diabetes. These results are difficult to interpret since the consequences of both diabetes and hypertension on blood vessels are well known. It is possible that these patients may have stiff, calcified blood vessels that could ‘protect’ them from the aneurysmatic changes. The present study confirmed that dialysis vintage significantly influenced VAA formation. Patients that used the same VA for more than 5.7 years were at a higher risk of developing VAA. We also confirmed that dialysis membrane and blood flow rate may have an influence on the formation of aneurysmatic changes: dialysis on highflux polysulfone membrane and with higher blood flow rate carried a higher risk for VAA. Similar data in the literature are missing, but we may speculate that higher blood flow rate may contribute to vascular stress in the long term. Mineral metabolism disturbances are a well-known complication of chronic renal failure and may lead to vascular and soft tissue calcifications. There are no literature data about the influence of impaired mineral metabolism on VAA formation. In our study, patients without VAA used a significantly higher dose of calcitriol and had high-
er PWV, but calcification score was not found to be a significant risk factor for VAA formation. Of the proposed risk factors, only dialysis vintage had an influence on the score of aneurysms. These results are difficult to interpret since similar results in the literature are missing. Angiotensin II and transforming growth factor-βmediated degradation of elastin, collagen and the other constituents of the vessel wall may be of importance in VAA formation as well as matrix metalloproteinase (MMP). These proteolytic enzymes are secreted by vascular smooth muscle cells and by inflammatory cells [12–14]. In patients with aneurysm of abdominal aorta, authors found a high concentration of MMP-9 [15], while the mice genetically deficient in MMP-2 and MMP-9 are resistant to developing aneurysm [12]. Similar data in humans are missing, but in case such connection exists, the ACE inhibitors could be an important preventive strategy for the VAA in dialysis patients. In our group of patients, angiotensin-converting enzyme inhibitor use was equal in patients with and without VAA. In addition, we did not find that ACE and MMP-3 gene polymorphism influence aneurysmatic changes (data not shown). This finding has to be confirmed in a large population of patients with aneurysm. The present study has several weak points. Aneurysmatic changes have not been visualized by ultrasonogram or angiogram where more precise details could be available including high- or down-flow stenosis. Unfortunate-
ly, we have no data about the consequences these aneurysms may have on the cardiovascular system. Apart from the present risk factors, there could be other potential risk factors that were not included in this analysis. Finally, clinical applicability of previous findings has to be confirmed. Higher blood flow rate during HD is beneficial for dialysis adequacy but it carries a risk for VAA formation, and longer, slow dialysis could be a solution. Also, attention should be directed toward careful puncturing of AVF and more frequent monitoring of complications such as rupture and cardiovascular consequences once aneurysmatic changes become apparent. Taken together, results presented in this study should be taken with caution and need to be confirmed by further investigations.
Conclusion
The prevalence of VAA in our dialysis population was very high (60%), and in more than half (71%) they were multiple. Aneurysms were more frequent in patients with ADPKD than in other diagnostic categories, and in those who had longer dialysis vintage on high-flux membranes with higher blood flow rate during HD. To establish an adequate preventive and therapeutic strategy, we need more prospective studies and detailed analysis of additional risk factors for VAA.
1 Pastan S, Soucie JM, McClellan WM: Vascular access and increased risk of death among hemodialysis patients. Kidney Int 2002; 62: 620–626. 2 Astor BC, Eustace JA, Powe NR, Klag MJ, Fink NE, Coresh J, CHOICE Study: Type of vascular access and survival among incident hemodialysis patients: the Choices for Healthy Outcomes in Caring for ESRD (CHOICE) Study. J Am Soc Nephrol 2005;16:1449–1455. 3 US Renal Data System (2007). USRDS annual data report: atlas of chronic kidney disease in the United States. http://www.usrds.org/atlas. htm (accessed 6 August 2008). 4 Adragao T, Pires A, Lucas C, Birne R, Magalhaes L, Gonçalves M, et al: A simple vascular calcification score predicts cardiovascular risk in haemodialysis patients. Nephrol Dial Transplant 2004;19:1480–1488. 5 Schlieper G, Krüger T, Djuric Z, Damjanovic T, Markovic N, Schurgers LJ, et al: Vascular access calcification predicts mortality in hemodialysis patients. Kidney Int 2008; 74: 1582–1587.
98
6 Schlieper G, Westenfeld R, Krüger T, Cranenburg EC, Magdeleyns EJ, Brandenburg VM, Djuric Z, Damjanovic T, Ketteler M, Vermeer C, Dimkovic N, Floege J, Schurgers LJ: Circulating nonphosphorylated carboxylated matrix gla protein predicts survival in ESRD. J Am Soc Nephrol 2011;22:387–395. 7 Derakhshanfar A, Gholyaf M, Niayesh A, Bahiraii S: Assessment of frequency of complications of arteriovenous fistula in patients on dialysis: a two-year single center study from Iran. Saudi J Kidney Dis Transpl 2009; 20: 872–875. 8 Brems J, Castaneda M, Garvin PJ: A five-year experience with the bovine heterograft for vascular access. Arch Surg 1986;121:941–944. 9 Raju S: PTFE grafts for hemodialysis access. Techniques for insertion and management of complications. Ann Surg 1987; 206: 666– 673. 10 Verhallen AM, Kooistra MP, van Jaarsveld BC: Cannulating in haemodialysis: rope-ladder or buttonhole technique? Nephrol Dial Transplant 2007;22:2601–2604.
Nephron Clin Pract 2013;124:94–98 DOI: 10.1159/000355548
11 Van Loon MM, Goovaerts T, Kessels AG, van der Sande FM, Tordoir JH: Buttonhole needling of haemodialysis arteriovenous fistulae results in less complications and interventions compared to the rope-ladder technique. Nephrol Dial Transplant 2010;25:225–230. 12 Longo GM, Xiong W, Greiner TC, Zhao Y, Fiotti N, Baxter BT: Matrix metalloproteinases 2 and 9 work in concert to produce aortic aneurysms. J Clin Invest 2002;110:625–632. 13 Wilson WR, Anderton M, Schwalbe EC, Jones JL, Furness PN, Bell PR, et al: Matrix metalloproteinase-8 and -9 are increased at the site of abdominal aortic aneurysm rupture. Circulation 2006;113:438–445. 14 Longo GM, Buda SJ, Fiotta N, Xiong W, Griener T, Shapiro S, Baxter BT: MMP-12 has a role in abdominal aortic aneurysms in mice. Surgery 2005;137:457–462. 15 Wilson WR, Anderton M, Choke EC, Dawson J, Loftus IM, Thompson MM: Elevated plasma MMP1 and MMP9 are associated with abdominal aortic aneurysm rupture. Eur J Vasc Endovasc Surg 2008;35:580–584.
Jankovic et al.
Downloaded by: Claude Moore Health Sciences Lib - Univ of Virginia 128.143.23.241 - 11/7/2013 11:21:47 AM
References
![[Echocardiographic abnormalities in chronic hemodialysis patients: prevalence and risk factors].](/assets/img/hold.png)
