In this study, we evaluated the effects of various medications on the patency of vascular access (VA) for hemodialysis.
We analyzed data from the Longitudinal Health Insurance Database of Taiwan. We adopted a case–control study design within a cohort of patients who had received regular hemodialysis between 2002 and 2012; 34,354 patients with first VA failure were identified, and the duration from VA creation date to the first VA failure date was calculated. We then classified these patients into two groups, namely arteriovenous fistula (AVF, n = 25,933) and arteriovenous graft (AVG, n = 8,421). Each group was further divided into two subgroups, namely short-term (<1 year) and long-term (≥1 year) patency.
The risk factors for early VA failure were age ≥65 years, diabetes mellitus, hyperlipidemia, cerebral vascular disease, congestive heart failure, peripheral artery disease, and sepsis. Male sex, hypertension, cancer, and peptic ulcer were associated with early AVF failure. Antiplatelet therapy increased the AVF and AVG patency times with adjusted odds ratios of 0.748 (95% confidence interval [CI]: 0.703–0.796, p < 0.0001) and 0.810 (95% CI: 0.728–0.901, p = 0.0001), respectively. A significant decrease in the VA failure risk was observed with an increase in the cumulative defined daily dose of antiplatelet agents.
This nationwide study demonstrated that some risk factors were associated with early VA failure and that the use of antiplatelet agents prevented the loss of VA patency in a dose–response manner. Thus, antiplatelet drugs should be routinely administered to high-risk patients receiving dialysis.
Citation: Hsu Y-H, Yen Y-C, Lin Y-C, Sung L-C (2018) Antiplatelet agents maintain arteriovenous fistula and graft function in patients receiving hemodialysis: A nationwide case–control study. PLoS ONE 13(10): e0206011. https://doi.org/10.1371/journal.pone.0206011
Editor: Tatsuo Shimosawa, International University of Health and Welfare, School of Medicine, JAPAN
Received: July 14, 2018; Accepted: October 4, 2018; Published: October 18, 2018
Copyright: © 2018 Hsu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data used in this study were sourced from the National Health Insurance Research Database (NHIRD) in Taiwan. The NHIRD data that support the findings of this study were available from The National Health Research Institutes (NHRI) before December 31, 2013 upon request from researchers for the purpose of research. The NHRI stopped providing this service after December 31, 2013. The data underlying this study have been transferred to the Health and Welfare Data Science Center (HWDC). Owing to the legal restrictions imposed by the Government of Taiwan related to the Personal Information Protection Act, the database cannot be made publicly available. Interested researchers can obtain the data through a formal application to the Health and Welfare Data Science Center, Department of Statistics, Ministry of Health and Welfare, Taiwan (https://dep.mohw.gov.tw/DOS/np-2497-113.html). Researchers are not allowed to carry raw data outside the HWDC. Interested researchers can access these data in the same manner as the authors. The authors did not have any special access privileges that other researchers would not have.
Funding: This work was supported by Taipei Medical University (105-6102-001-102). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Functional vascular access (VA) is essential for hemodialysis (HD), and VA can be achieved using an arteriovenous fistula (AVF), an arteriovenous graft (AVG), or a central venous catheter. VA failure is a major cause of morbidity among patients receiving HD, and it accounts for the hospitalization of approximately 20% of these patients [1–4]. An AVF is the preferred type of VA for HD because the complication rates and patient survival rates associated with AVFs are lower and higher, respectively, than those of associated with AVGs or central venous catheters [5, 6]. The 1-year patency rates of native AVFs are extremely variable, ranging from 40% to 80% . The most crucial cause of AVF failure is progressive neointimal hyperplasia at the venous anastomosis, which results in stenosis and subsequent thrombosis [8–10]. An AVG is generally cannulated more easily and can be used for HD sooner after surgery than an AVF. However, in AVGs, stenosis frequently develops at the graft–vein anastomosis, thus leading to access thrombosis . Central venous catheters are used for rapid access in immediate dialysis and are associated with relatively high rates of infection and complications.
The mechanisms of VA failure are not completely understood; however, studies have shown that inflammation may play a critical role [7, 12–14]. Several clinical studies have reported that antiplatelet drugs improve VA patency rates [10, 11, 15, 16]. A recent Cochrane review reported that antiplatelet treatment can improve the 1-month patency rates of AVFs and AVGs . Two meta-analyses have revealed that aspirin prevents thrombosis of AVFs but not AVGs [17, 18]. Some in vitro and in vivo studies have demonstrated that statins reduce neointimal proliferation, vascular inflammation, and VA failure [4, 16, 19, 20]. However, one retrospective analysis that used data from the United States Renal Data System (USRDS) reported that statins are not associated with significant reduction in VA failure risk and that antiplatelet agents are associated with a significantly increased risk of AVF dysfunction . These inconsistent findings might have resulted partially from differences in methodology, clinical factors, and populations among the studies. Percutaneous transluminal angioplasty (PTA) and surgical thrombectomy are the two principal treatments for patients with progressive AVF or AVG failure [7, 8], which may affect VA survival time. Some trials have not considered the effects of PTA or surgical history on VA survival time.
The long-term efficacy of drug therapy in preventing the first AVF or AVG dysfunction in Taiwanese patients with chronic HD has not been reported thus far. We conducted a nationwide population-based case–control study by using the claims data of patients between 2001 and 2012 from the National Health Insurance Research Database (NHIRD) of Taiwan. We investigated the associations between various drug regimens and patency rates after VA creation.
The National Health Insurance (NHI) program has been operational in Taiwan since 1995 and provides comprehensive health insurance coverage for all residents of Taiwan. Currently, the NHI program includes 97% of all hospitals and clinics in Taiwan. In addition, 98% of the >23 million potential enrollees are covered under the NHI program . We studied the entire population of Taiwan between 2001 and 2012, and the follow-up period extended to the end of 2012.
In this study, we obtained medical insurance claims data from the Health and Welfare Data Science Center (HWDC), Ministry of Health and Welfare of Taiwan. These data constitute the medical information of each insured individual in Taiwan. Data files from the NHI claims database contain registration numbers and original claims data for reimbursement. All the data are anonymous and are deidentified by scrambling the identification codes of both the patients and medical facilities, thus rendering the NHI reimbursement data suitable for academic research . The International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) codes are used for reporting diagnoses. Thus, the diagnoses in the NHI claims database of Taiwan exhibit high accuracy and validity. Taiwanese studies have demonstrated the high accuracy and validity of ICD-9-CM diagnoses in the Taiwan’s NHI claims database [22, 23]. This study was approved by the Joint Institutional Review Board of Taipei Medical University (TMU-JIRB N201609023; see S1 File) and supervised by the HWDC.
This study used a case–control design. A flowchart of the patient selection procedure is presented in Fig 1. From the Registry for Catastrophic Illness Patients Database, a subdatabase of the NHI claims database, we identified all patients with end-stage renal disease (ESRD) (ICD-9-CM code: 585) who had received HD between January 1, 2001, and December 31, 2012 (n = 112,194). We included all patients with ESRD and first VA dysfunction between 2002 and 2012 (n = 47,818). VA dysfunction was defined as VA failure that occurs >3 months after the initial HD; this criterion excluded patients with very early VA failure (n = 5,088). The causes of very early VA failure differ from those associated with failure >3 months; therefore, patients with very early failure were excluded . We retrospectively searched for the first VA-related procedure (surgical thrombectomy or PTA) codes, which could be considered surrogates for first VA failure because stable patients on HD would receive these procedures only after VA failure. Patients who received renal transplantation (n = 56) or peritoneal dialysis (n = 25), either before or after HD, were excluded. In addition, we excluded patients if they had unclear sex data (n = 44), were younger than 20 years (n = 65), or had received HD for <90 days (n = 8,186). The final study cohort comprised 34,354 patients. We divided the patients into two groups, namely AVF and AVG. Each group was further divided into two subgroups according to the duration between VA creation date and first VA failure date, namely short-term (S) (<1 year) and long-term (L) (≥1 year) patency.
The time of occurrence of VA dysfunction was used to define two subgroups, namely short-term (S) (<1 year) and long-term (L) (≥1 year) VA dysfunction. AVF, arteriovenous fistula; AVG, arteriovenous graft; ESRD, end-stage renal disease.
Definition of comorbidities
The comorbidities of the patients included in this study were hypertension (HTN), diabetes mellitus (DM), coronary artery disease (CAD), congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), cerebral vascular disease (CVD), hyperlipidemia, peripheral artery disease (PAD), peptic ulcer disease (PUD), sepsis or systemic inflammatory response syndrome (SIRS), and cancer (Tables 1 and 2). The comorbidities were identified using ICD-9-CM codes . The existing comorbidities of the patients were defined using the following criteria: ICD-9-CM codes appearing in the inpatient database at least once within 1 year before initial HD, diagnoses of the comorbidity in the outpatient database at least twice, and timestamps of first and last diagnoses at least 30 days apart within 1 year before HD initiation.
Medication use and risk of VA failure
Some medications, including statins, antiplatelet agents, anticoagulants, and dipyridamole, that have been reported to affect VA outcome were examined [4, 10, 11, 15, 16, 19, 20, 26–29]. All prescription information was extracted from the NHIRD. The prescription claims of candidate medications were tracked for up to 1 year before the first VA dysfunction date by using NHIRD drug codes (see S2 File). The defined daily dose (DDD) recommended by the World Health Organization is a unit for measuring a prescribed amount of drug . It is the assumed average maintenance dose per day of a drug consumed for its main indication in adults. In this study, cumulative DDD (cDDD) was defined as a time-independent variable in which the daily supply of each prescription dispensed was summed over time within a 90-day period before the first VA failure date. To examine the drug–response relationship, we further categorized the medications of each candidate into two groups within the AVF and AVG groups (<10 and ≥10 cDDD in the anticoagulants for their rapid onset of action, or <30 and ≥30 cDDD in other drugs). The patients who used drugs for <30 or <10 cDDD were defined as drug nonusers (Table 3).
Baseline demographic and comorbidity characteristics are presented as proportions as appropriate and were compared using chi-squared statistics. We used a logistic model to estimate the odds ratios (ORs) and 95% confidence intervals (CIs) for the association between drug use and risk of VA dysfunction. The adjusted OR (aOR) was estimated using multiple logistic regression; the ORs were adjusted for sex, age group, comorbidities, and medication. We also performed a stratified analysis to identify the dose–response effect in both the VA groups (Tables 4 and 5). Furthermore, we divided the cDDDs of antiplatelet agents into four group variables, namely cDDD = 0, 1 < cDDD < 30, 31 < cDDD < 90, and cDDD > 90, to compare the VA failure risk in the different cDDD groups by using multiple logistic regression estimation. The data were analyzed using SAS statistical software (version 9.3 for Windows; SAS Institute Inc., Cary, NC, USA). The results were considered statistically significant if two-tailed p < 0.05.
Baseline characteristics of the AVF and AVG groups
The baseline characteristics of the AVF and AVG groups in the study cohort of 34,354 patients are listed in Tables 1 and 2. Among the 25,933 AVF patients, we identified 13,614 and 12,319 patients who did (AVF-S) and did not (AVF-L), respectively, experience AVF failure (and require intervention) within 1 year after the index date of VA creation. Among the 8,421 AVG patients, we identified 5,044 and 3,377 patients who did (AVG-S) and did not (AVG-L), respectively, experience AVG failure within 1 year after the index date of VA creation (Fig 1). In the patients with VA dysfunction, the proportion of men was higher among the patients with AVF than among those with AVG (53.3% vs 38.6%). The short-term patency groups, AVF-S and AVG-S, exhibited higher mean ages and significantly higher prevalence rates of comorbidities than the long-term patency groups, AVF-L and AVG-L. Furthermore, more patients in the AVF-S group used dipyridamole and statins than did in the AVF-L group. However, no differences in prescribed medications were observed between the AVG-S and AVG-L groups (Tables 1 and 2).
Risk factors for early VA dysfunction
After adjustment for potential confounders through multiple logistic regression, early VA dysfunction remained associated with advanced age, DM, hyperlipidemia, CVD, CHF, PAD, sepsis or SIRS (all aOR > 1 and p < 0.05). In addition, male sex and the comorbidities of HTN, cancer, and PUD were associated with early AVF failure but not early AVG failure (Table 3).
Outcomes in medication users and nonusers
Among the four types of medications, only antiplatelet agent users exhibited significantly longer patency durations than drug nonusers after adjustment for related covariates (Table 3). Antiplatelet agent use resulted in longer durations of AVF and AVG patency, with aORs of 0.748 (95% CI: 0.703–0.796, p < 0.0001) and 0.810 (95% CI: 0.728–0.901, p = 0.0001), respectively (Table 3).
An analysis of antiplatelet agent users based on cDDD within the 90-day period before the first VA failure date revealed a dose–response effect (Tables 4 and 5). When the AVF-group patients were stratified according to the cDDD of the medications, the aORs for AVF failure were 0.985 (95% CI: 0.917–1.057, p = 0.6689), 0.756 (95% CI: 0.700–0.815, p < 0.0001), and 0.728 (95% CI: 0.662–0.800, p < 0.0001) for the patients (drug users) who received cDDDs of 0–30, 30–90, and >90 compared with the drug nonusers. In this stratified analysis, prevention of early AVF failure was associated with a cDDD of antiplatelet agent use of ≥30 (Table 4). The antiplatelet agents conferred a similar protective effect in the AVG-group patients (Table 5).
The key new findings of this nationwide cohort study are as follows. (1) The number of patients with the first VA dysfunction and the male-to-female ratio were higher in the AVF group than in the AVG group. (2) The clinical predictors of the first VA dysfunction were advanced age, hyperlipidemia, DM, CVD, CHF, PAD, and sepsis or SIRS. (3) The additional risk factors for the first AVF failure were male sex, HTN, cancer, and PUD. (4) After clinical risk adjustment, the use of antiplatelet agents at cDDD ≥30 prevented loss of VA patency.
Maintaining VA patency remains one of the most crucial challenges in patients who receive HD. The pathogenesis of very early AVF failure (<3 months) remains poorly understood . By excluding this group of patients, we ensured that the study did not include patients with immature fistulas; thus, we improved the reliability of our findings. Furthermore, the strength and advantage of our study is that we used data of almost the entire national population of Taiwan. Because our study involved a large number of participants who were followed-up for a long period, our results provide real-world evidence . The Taiwan NHIRD dataset has been proven to be highly reliable; therefore, our results are highly convincing [31, 32].
The most common cause of VA dysfunction is venous stenosis, caused by venous neointimal hyperplasia within the juxta-anastomotic region (for an AVF) or at the graft–vein junction (for an AVG). The following factors contribute to neointimal hyperplasia: (1) surgical trauma at the time of VA creation, (2) hemodynamic shear stress at the artery–vein or graft–vein junctions, (3) bioincompatability of the AVG, (4) vessel injury caused by dialysis needle punctures, and (5) endothelial dysfunction caused by uremia toxin [14, 33]. A review article divided the risk factors for AVF dysfunction into two categories, namely nonmodifiable factors, such as advanced age, DM, hypotension, arteriosclerosis, and artery and venous anatomical factors, and modifiable factors, such as smoking, anastomosis type, cannulation technique, and first cannulation being performed within 14 days of VA creation . Although numerous studies have examined the clinical, anatomical, and technical factors affecting VA dysfunction, the causes of individual variation in numerous patients remain unknown [34–38]. Additional studies on comprehensive risk identification and adjustment are essential for improving the accuracy of comparisons.
Reports have revealed that female sex is an independent risk factor for primary AVF failure, which is contrary to our results [37, 39, 40]. In addition, female sex was associated with AVF maturation failure in one European cohort study . Women exhibit smaller vessel diameters and higher vasomotor activity than men; these findings can explain the poorer outcome of AVFs in women than in men [37, 39, 40]. However, some data have suggested that the 1-year patency rates in women are similar to those in men for AVFs [34, 42]. Our study revealed a higher proportion of AVG creation in women than in men (Table 1). The discrepancy between the results of the previous studies and those of the present study is possibly caused by the exclusion of patients with very early VA failure (<3 months) from our analysis as well as differences in selection criteria for VA creation and demographics.
The role of adjuvant medications in prolonging VA patency is a crucial research topic. Several clinical investigations and trials have reported that statins, anticoagulants, or antiplatelet drugs can reduce thrombosis and improve VA outcomes [9–11, 15, 16, 20, 43–45]. In this study, only the use of antiplatelet agents was associated with longer VA patency durations among patients receiving HD after adjustment for multiple clinical risk factors, thus supporting the evidence obtained from several randomized studies and one related study in the Cochrane database of systematic review [10, 11, 15, 45]. If the dose of the antiplatelet agent was >30 cDDD, the aOR for the ability of the antiplatelet agent to reduce VA dysfunction risk in patients receiving HD was significant. This is also the first study to propose that antiplatelet agents exert effects in a dose-responsive manner in patients receiving HD who are antiplatelet users.
Some neutral or contradictory results have been reported. Meta-analyses by Coleman et al. and Palmer et al. have shown that antiplatelet therapy prevented only AVF thrombosis and did not exert an effect on AVG thrombosis [17, 18]. However, the quality of the evidence was low because of the heterogeneity among the trials, small number of studies used in each comparison, short follow-up periods, and moderate methodological quality of the studies resulting from incomplete reporting. The USRDS Dialysis Mortality and Morbidity Wave II Study investigated 901 patients in a retrospective study and concluded that the risk of AVF failure associated with the use of antiplatelet agents was significantly higher than that associated with other medications [16, 21]. However, the doses and dates for initiation and withdrawal of the antiplatelet drugs were not recorded. Furthermore, patients who receive antiplatelet drugs usually exhibit more severe comorbidities than do patients who do not receive this treatment, which could have introduced bias. The clinical risk adjustment was limited to age, sex, race, CAD history, and PAD history.
Several possible factors may explain the beneficial effect of antiplatelet agents. The pathogenesis of venous neointimal hyperplasia has been elucidated previously; oxidative stress, platelet activation, endothelial dysfunction, and inflammation play a crucial role in the development of venous neointimal hyperplasia [13, 14, 21, 35, 38, 46]. Aspirin not only acts as an antiplatelet agent but also reduces oxidative stress and inflammation. Antiplatelet agents may be associated with superior cardiovascular outcomes than other medications in patients receiving HD, thus reducing intradialytic hypotension and the stasis and hypercoagulability of blood. Additional studies are warranted to verify the effects of antiplatelet agents on VA patency and to explore the underlying mechanism.
Several drugs in addition to antiplatelet agents have been reported to maintain VA patency. Treatment with dipyridamole plus low-dose aspirin substantially reduced the risk of stenosis and increased the duration of AVG patency . However, one conflicting result of the effect of the same drug regimen on graft patency was reported . Studies have revealed that statins reduced neointimal proliferation and vascular inflammation [9, 19, 26]. Until recently, direct evidence of a beneficial effect of statins on VA patency rate was not available . In addition, low-dose warfarin does not appear to prolong AVG patency [27, 48]. Our findings revealed that statins, anticoagulants, and dipyridamole are ineffective at maintaining long-term AVF or AVG patency; these findings are consistent with those of some of the aforementioned studies. Nevertheless, a randomized controlled trial is required to firmly establish the role of these agents in VA outcomes.
Our study, similar to all database studies, had several inherent limitations. First, the data of several unmeasured confounders, such as drug compliance, use of self-paid medications, blood pressure, body mass index, cigarette smoking, physical activity, different types and combinations of antiplatelet agents, location of the VA (forearm or upper arm), and cannulation technique, are not available in the NHIRD. Second, previous studies have demonstrated the beneficial effects of some antihypertensive agents in prolonging primary VA patency, which were not included in our analysis . However, we adjusted the comorbidity HTN in each group; consequently, balanced prescriptions of antihypertensive agents were anticipated. Third, laboratory data of inflammatory marker levels, lipid profiles, hemoglobin levels, and calcium–phosphate product levels were unavailable. Failure to consider the aforementioned variables may have caused some residual bias. However, because of the significance and magnitude of the observed effects, these limitations are unlikely to have noticeably affected the results. In addition, because the study cohort included only Taiwanese patients who received HD, the results may not be generalizable to other populations. Finally, our investigation was a retrospective observational study that had some methodological drawbacks and selection bias despite the case–control study design. The clinical relevance of this study must be further verified through large-scale prospective trials.
The most noteworthy clinical implications of this study are that some risk factors are associated with early VA failure, and the use of antiplatelet agents protects VA from patency loss in a dose–response manner. Thus, active identification and treatment of risk factors and early administration of antiplatelet drugs are necessary in patients receiving dialysis that have multiple clinical risk factors.
This study was based, in part, on data obtained from the Taiwan’s NHI claims database provided by the HWDC, Ministry of Health and Welfare. The authors acknowledge the assistance of the HWDC, Ministry of Health and Welfare and the support of the Research Center of Biostatistics, College of Management, Taipei Medical University for the statistical analysis, study design, and interpretation of data. This manuscript was edited by Wallace Academic Editing.
- 1. Rayner HC, Pisoni RL, Bommer J, Canaud B, Hecking E, Locatelli F, et al. Mortality and hospitalization in haemodialysis patients in five European countries: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrology, dialysis, transplantation: official publication of the European Dialysis and Transplant Association—European Renal Association. 2004;19(1):108–20. Epub 2003/12/13. pmid:14671046.
- 2. Santoro D, Benedetto F, Mondello P, Pipito N, Barilla D, Spinelli F, et al. Vascular access for hemodialysis: current perspectives. International journal of nephrology and renovascular disease. 2014;7:281–94. Epub 2014/07/22. pmid:25045278.
- 3. Lok CE, Foley R. Vascular access morbidity and mortality: trends of the last decade. Clinical journal of the American Society of Nephrology: CJASN. 2013;8(7):1213–9. pmid:23824198.
- 4. Righetti M, Ferrario G, Serbelloni P, Milani S, Tommasi A. Some old drugs improve late primary patency rate of native arteriovenous fistulas in hemodialysis patients. Annals of vascular surgery. 2009;23(4):491–7. Epub 2008/11/01. pmid:18973987.
- 5. Group VAW. III. NKF-K/DOQI Clinical Practice Guidelines for Vascular Access: update 2000. Am J Kidney Dis. 2001;37(1 Suppl 1):S137–81. Epub 2001/03/07. pmid:11229969.
- 6. Hemachandar R. Analysis of Vascular Access in Haemodialysis Patients—Single Center Experience. Journal of clinical and diagnostic research: JCDR. 2015;9(10):Oc01–4. Epub 2015/11/12. pmid:26557553.
- 7. Lin CF, Chiou HY, Chang YH, Liu JC, Hung YN, Chuang MT, et al. Risk of arteriovenous fistula failure associated with hypnotic use in hemodialysis patients: a nested case-control study. Pharmacoepidemiol Drug Saf. 2016;25(8):889–97. pmid:26799147.
- 8. Chang CJ, Ko PJ, Hsu LA, Ko YS, Ko YL, Chen CF, et al. Highly increased cell proliferation activity in the restenotic hemodialysis vascular access after percutaneous transluminal angioplasty: implication in prevention of restenosis. Am J Kidney Dis. 2004;43(1):74–84. Epub 2004/01/09. pmid:14712430.
- 9. Yeh CH, Huang TS, Wang YC, Huang PF, Huang TY, Chen TP, et al. Effects of Antiplatelet Medication on Arteriovenous Fistula Patency After Surgical Thrombectomy. Current vascular pharmacology. 2016;14(4):353–9. Epub 2016/03/01. pmid:26924326.
- 10. Tanner NC, Da Silva A. Medical adjuvant treatment to increase patency of arteriovenous fistulae and grafts. Cochrane Database Syst Rev. 2015;7:CD002786. pmid:26184395.
- 11. Dixon BS, Beck GJ, Vazquez MA, Greenberg A, Delmez JA, Allon M, et al. Effect of dipyridamole plus aspirin on hemodialysis graft patency. The New England journal of medicine. 2009;360(21):2191–201. Epub 2009/05/22. pmid:19458364.
- 12. Roy-Chaudhury P, Kelly BS, Miller MA, Reaves A, Armstrong J, Nanayakkara N, et al. Venous neointimal hyperplasia in polytetrafluoroethylene dialysis grafts. Kidney international. 2001;59(6):2325–34. Epub 2001/05/31. pmid:11380837.
- 13. Dukkipati R, Molnar MZ, Park J, Jing J, Kovesdy CP, Kajani R, et al. Association of vascular access type with inflammatory marker levels in maintenance hemodialysis patients. Seminars in dialysis. 2014;27(4):415–23. Epub 2013/10/15. pmid:24118625.
- 14. Lee T, Roy-Chaudhury P. Advances and new frontiers in the pathophysiology of venous neointimal hyperplasia and dialysis access stenosis. Advances in chronic kidney disease. 2009;16(5):329–38. Epub 2009/08/22. pmid:19695501.
- 15. Dember LM, Beck GJ, Allon M, Delmez JA, Dixon BS, Greenberg A, et al. Effect of clopidogrel on early failure of arteriovenous fistulas for hemodialysis: a randomized controlled trial. JAMA. 2008;299(18):2164–71. pmid:18477783.
- 16. Paraskevas KI, Mikhailidis DP, Roussas N, Giannoukas AD. Effect of antiplatelet agents, statins, and other drugs on vascular access patency rates. Angiology. 2012;63(1):5–8. pmid:22144688.
- 17. Coleman CI, Tuttle LA, Teevan C, Baker WL, White CM, Reinhart KM. Antiplatelet agents for the prevention of arteriovenous fistula and graft thrombosis: a meta analysis. International journal of clinical practice. 2010;64(9):1239–44. Epub 2010/05/12. pmid:20455955.
- 18. Palmer SC, Di Micco L, Razavian M, Craig JC, Ravani P, Perkovic V, et al. Antiplatelet Therapy to Prevent Hemodialysis Vascular Access Failure: Systematic Review and Meta-analysis. American Journal of Kidney Diseases. 2013;61(1):112–22. http://dx.doi.org/10.1053/j.ajkd.2012.08.031. pmid:23022428
- 19. Janardhanan R, Yang B, Vohra P, Roy B, Withers S, Bhattacharya S, et al. Simvastatin reduces venous stenosis formation in a murine hemodialysis vascular access model. Kidney international. 2013;84(2):338–52. Epub 2013/05/03. pmid:23636169.
- 20. Chang HH, Chang YK, Lu CW, Huang CT, Chien CT, Hung KY, et al. Statins Improve Long Term Patency of Arteriovenous Fistula for Hemodialysis. Scientific reports. 2016;6:22197. Epub 2016/02/24. pmid:26902330.
- 21. Yevzlin AS, Conley EL, Sanchez RJ, Young HN, Becker BN. Vascular access outcomes and medication use: a USRDS study. Seminars in dialysis. 2006;19(6):535–9. Epub 2006/12/08. pmid:17150056.
- 22. Chen CI, Kao PF, Wu MY, Fang YA, Miser JS, Liu JC, et al. Influenza Vaccination is Associated with Lower Risk of Acute Coronary Syndrome in Elderly Patients with Chronic Kidney Disease. Medicine. 2016;95(5):e2588. Epub 2016/02/06. pmid:26844466.
- 23. Fang YA, Chen CI, Liu JC, Sung LC. Influenza Vaccination Reduces Hospitalization for Heart Failure in Elderly Patients with Chronic Kidney Disease: A Population-Based Cohort Study. Acta Cardiologica Sinica. 2016;32(3):290–8. Epub 2016/06/09. pmid:27274169.
- 24. Kuo TH, Tseng CT, Lin WH, Chao JY, Wang WM, Li CY, et al. Association Between Vascular Access Dysfunction and Subsequent Major Adverse Cardiovascular Events in Patients on Hemodialysis: A Population-Based Nested Case-Control Study. Medicine. 2015;94(26):e1032. pmid:26131808.
- 25. Chu YT W S, Lee YC, et al. Assessing measures of comorbidity using National Health Insurance Databases. Taiwan J Public Health. 2010;29(3):191–200.
- 26. Pisoni R, Barker-Finkel J, Allo M. Statin therapy is not associated with improved vascular access outcomes. Clinical journal of the American Society of Nephrology: CJASN. 2010;5(8):1447–50. Epub 2010/05/29. pmid:20507962.
- 27. Crowther MA. Low-Intensity Warfarin Is Ineffective for the Prevention of PTFE Graft Failure in Patients on Hemodialysis: A Randomized Controlled Trial. Journal of the American Society of Nephrology. 2002;13(9):2331–7. pmid:12191977
- 28. Yeh C-H, Huang T-S, Wang Y-C, Huang P-F, Huang T-Y, Chen T-P, et al. Effects of Antiplatelet Medication on Arteriovenous Fistula Patency After Surgical Thrombectomy. Current vascular pharmacology. 2016;14(4):353–9. pmid:26924326
- 29. Birch N, Fillaus J, Florescu MC. The effect of statin therapy on the formation of arteriovenous fistula stenoses and the rate of reoccurrence of previously treated stenoses. Hemodialysis international International Symposium on Home Hemodialysis. 2013;17(4):586–93. Epub 2012/10/20. pmid:23078106.
- 30. Liu JC, Yang TY, Hsu YP, Hao WR, Kao PF, Sung LC, et al. Statins dose-dependently exert a chemopreventive effect against lung cancer in COPD patients: a population-based cohort study. Oncotarget. 2016. Epub 2016/08/16. pmid:27517752.
- 31. Database NHIR. Taiwan, http://nhird.nhri.org.tw/en/index.html (cited in 2015).
- 32. Peng YC, Lin CL, Yeh HZ, Tung CF, Chang CS, Kao CH. Diverticular disease and additional comorbidities associated with increased risk of dementia. Journal of gastroenterology and hepatology. 2016;31(11):1816–22. Epub 2016/04/01. pmid:27029523.
- 33. Roy-Chaudhury P, Lee T, Woodle B, Wadehra D, Campos-Naciff B, Munda R. Balloon-assisted maturation (BAM) of the arteriovenous fistula: the good, the bad, and the ugly. Seminars in nephrology. 2012;32(6):558–63. Epub 2012/12/12. pmid:23217336.
- 34. Smith GE, Gohil R, Chetter IC. Factors affecting the patency of arteriovenous fistulas for dialysis access. Journal of vascular surgery. 2012;55(3):849–55. pmid:22070937.
- 35. Tanaka A, Ito Y, Tanaka T, Satozaki S, Hayashi F, Tsuda I. Blood monocyte count may be a predictor of vascular access failure in hemodialysis patients. Therapeutic apheresis and dialysis: official peer-reviewed journal of the International Society for Apheresis, the Japanese Society for Apheresis, the Japanese Society for Dialysis Therapy. 2013;17(6):620–4. Epub 2013/12/18. pmid:24330557.
- 36. Verest S, Logghe P, Claes K, Kuypers D, Fourneau I. Effect of clinical examination and anatomical location on native arteriovenous fistula maturation rate in high risk patients. Acta chirurgica Belgica. 2014;114(5):324–31. Epub 2015/05/30. pmid:26021537.
- 37. Miller CD, Robbin ML, Allon M. Gender differences in outcomes of arteriovenous fistulas in hemodialysis patients. Kidney international. 2003;63(1):346–52. pmid:12472802.
- 38. De Marchi S, Falleti E, Giacomello R, Stel G, Cecchin E, Sepiacci G, et al. Risk factors for vascular disease and arteriovenous fistula dysfunction in hemodialysis patients. Journal of the American Society of Nephrology: JASN. 1996;7(8):1169–77. Epub 1996/08/01. pmid:8866409.
- 39. Huijbregts HJ, Bots ML, Moll FL, Blankestijn PJ, members C. Hospital specific aspects predominantly determine primary failure of hemodialysis arteriovenous fistulas. Journal of vascular surgery. 2007;45(5):962–7. pmid:17466788.
- 40. Yap YS, Chuang HY, Wu CH, Chi WC, Lin CH, Liu YC. Risk Factors for Early Failure of Arteriovenous Vascular Access Among Patients With Type 2 Diabetes Mellitus. Therapeutic apheresis and dialysis: official peer-reviewed journal of the International Society for Apheresis, the Japanese Society for Apheresis, the Japanese Society for Dialysis Therapy. 2016;20(2):112–7. Epub 2016/02/27. pmid:26916506.
- 41. Masengu A, Maxwell AP, Hanko JB. Investigating clinical predictors of arteriovenous fistula functional patency in a European cohort. Clinical kidney journal. 2016;9(1):142–7. Epub 2016/01/23. pmid:26798475.
- 42. Rooijens PP, Tordoir JH, Stijnen T, Burgmans JP, de Smet AA, Yo TI. Radiocephalic wrist arteriovenous fistula for hemodialysis: meta-analysis indicates a high primary failure rate. European journal of vascular and endovascular surgery: the official journal of the European Society for Vascular Surgery. 2004;28(6):583–9. pmid:15531191.
- 43. Sreedhara R, Himmelfarb J, Lazarus JM, Hakim RM. Anti-platelet therapy in graft thrombosis: results of a prospective, randomized, double-blind study. Kidney international. 1994;45(5):1477–83. Epub 1994/05/01. pmid:8072261.
- 44. Torsney E, Mayr U, Zou Y, Thompson WD, Hu Y, Xu Q. Thrombosis and neointima formation in vein grafts are inhibited by locally applied aspirin through endothelial protection. Circulation research. 2004;94(11):1466–73. Epub 2004/05/01. pmid:15117816.
- 45. Harter HR, Burch JW, Majerus PW, Stanford N, Delmez JA, Anderson CB, et al. Prevention of thrombosis in patients on hemodialysis by low-dose aspirin. The New England journal of medicine. 1979;301(11):577–9. Epub 1979/09/13. pmid:112475.
- 46. Wu CC, Wen SC, Yang CW, Pu SY, Tsai KC, Chen JW. Baseline plasma glycemic profiles but not inflammatory biomarkers predict symptomatic restenosis after angioplasty of arteriovenous fistulas in patients with hemodialysis. Atherosclerosis. 2010;209(2):598–600. Epub 2009/11/27. pmid:19939386.
- 47. Tayebi P, Kazemzadeh G, Banihashem A, Ravari H. Effect of low dose aspirin and dipyridamole on primary patency of arteriovenous grafts in hemodialysis patients: a randomized double-blind placebo-controlled trial. Electron Physician. 2018;10(1):6135–9. pmid:29588811.
- 48. Andreucci VE, Fissell RB, Bragg-Gresham JL, Ethier J, Greenwood R, Pauly M, et al. Dialysis Outcomes and Practice Patterns Study (DOPPS) data on medications in hemodialysis patients. Am J Kidney Dis. 2004;44(5 Suppl 2):61–7. Epub 2004/10/16. pmid:15486876.
- 49. Chen FA, Chien CC, Chen YW, Wu YT, Lin CC. Angiotensin Converting-Enzyme Inhibitors, Angiotensin Receptor Blockers, and Calcium Channel Blockers Are Associated with Prolonged Vascular Access Patency in Uremic Patients Undergoing Hemodialysis. PloS one. 2016;11(11):e0166362. Epub 2016/11/11. pmid:27832203.