Influence of renin–angiotensin system blockades on vascular access survival in patients on maintenance hemodialysis

Results on the association between the use of renin–angiotensin system blockades (RASBs) and vascular access‐related outcomes are inconsistent. We aimed to compare vascular access‐related outcomes according to the use of RASBs in hemodialysis patients.

Group 3 was prescribed RASBs (n = 16,811).Vascular access-related outcomes were classified into intervention-free survival (IFS), thrombosis-free survival (TFS), and vascular access survival (VAS).Results: No significant difference in the three access survival rates was identified among the three groups.The multivariate Cox regression analyses indicated that Group 3 had better outcomes in IFS and TFS than Group 1.The numbers of angioplasties performed were significantly greater in Group 1 than in the other two groups.The numbers of thrombectomies performed were significantly the lowest in Group 3 among all the groups.Conclusions: Our study revealed different results according to types of access survival in univariate or multivariate analyses.The association of RASBs with favorable outcomes in vascular access remains unclear.

| INTRODUCTION
Hemodialysis (HD) is the most commonly used modality among three renal replacement therapies including HD, peritoneal dialysis, and kidney transplantation. 1 Its prevalence increases with population aging and increase in comorbidities, such as diabetes mellitus (DM) or hypertension.Vascular access for sufficient blood flow rate is required to maintain adequate HD at limited intervals or times.HD can be performed using temporary or permanent vascular access.Temporary vascular access includes non-tunneled or tunneled catheter and permanent vascular access, includes arteriovenous fistula (AVF) and arteriovenous graft (AVG).Proper vascular access needs to be maintained because it is essential in performing HD.
Previous studies have evaluated the efficacy of some medications in improving vascular access-related outcomes; however, evidence of their efficacy is insufficient. 2,3Renin-angiotensin system blockades (RASBs) are well-known blood pressure-lowering drugs.5][6][7][8] As RASBs have such effects on blood vessels, they may have beneficial effects in vascular accessrelated outcomes in patients on HD.0][11][12][13] Thus, we aimed to compare vascular accessrelated outcomes according to the use of RASBs using a representative sample of patients on maintenance HD.

| Data source and study population
This retrospective study used the laboratory and clinical data from a national HD quality assessment program and the claims data from the Health Insurance Review and Assessment (HIRA) of the Republic of Korea. 14,15Briefly, the 4th, 5th, and 6th HD quality assessment programs were performed from July 2013 to December 2013, July 2015 to December 2015, and March 2018 to August 2018, respectively.
The programs included patients on maintenance HD (≥3 months), those undergoing HD at least twice a week (≥8 per month), and those aged ≥18 years.We analyzed HD quality assessment data and the claims data of all patients on HD who had undergone HD quality assessment from the HIRA.
The numbers of patients included in the 4th, 5th, and 6th HD quality assessment program were 21,846, 35,538, and 31,294, respectively.Among these, we excluded repeated participants or participants with an insufficient dataset (n = 32,459) and who underwent HD using a catheter (n = 1316).Finally, 54,903 were included in our study.This study was approved by the Institutional Review Board (IRB) of the Yeungnam University Medical Center (approval no.: YUMC 2022-01-010).Informed consent was not obtained from the patients since the records and information of the participants were anonymized and de-identified before the analysis.

| Variables
Data including age, sex, underlying disease of end-stage renal disease, HD vintages (days), and type of vascular access were collected.Laboratory data during the assessment included hemoglobin (g/dL), Kt/V urea , serum albumin (g/dL), serum calcium (mg/dL), serum phosphorus (mg/dL), serum creatinine (mg/dL), pre-dialysis systolic blood pressure (mmHg), pre-dialysis diastolic blood pressure (mmHg), and ultrafiltration volume (L/session).These data were collected monthly, and all laboratory values were averaged from the monthly collected values.Kt/V urea was calculated using Daugirdas' equation. 16e codes for medication are listed in Table S1.The use of RASBs was defined as more than one prescription during the 6 months of each HD quality assessment period.The patients were divided into Groups 1, 2, and 3. Group 1 was not prescribed any blood pressurelowering drugs.Group 2 was prescribed other blood pressurelowering agents except for RASBs.Group 3 was prescribed RASBs.
The medications were also evaluated for aspirin or statin.Prescription of one or more medications for a year before the evaluation of the HD quality assessment program was considered the use of medication.
The HD quality assessment program included a questionnaire for monitoring and surveillance.Monitoring was evaluated according to the presence of four checklists including inspection, palpation, auscultation, and cannulation distance.Surveillance was evaluated with multiple responses among static intra-access pressure, duplex ultrasound, ultrasound dilution technique, or angiography.The presence of comorbidities was analyzed for a year before the evaluation of the HD quality assessment program.Comorbidity was defined using the codes utilized by Quan et al. 17,18 Finally, the Charlson comorbidity index (CCI) score was calculated.
Vascular access-related outcomes were classified into intervention-free survival (IFS), thrombosis-free survival (TFS), and vascular access survival (VAS).The event was evaluated between end date of each HD quality assessment program and date of the end point of follow-up.The intervention was defined as the presence of procedure code for percutaneous transluminal angioplasty (M6597).Thrombectomy was defined as the presence of procedure codes for percutaneous or surgical thrombectomy (M6632, M6633, or M6639 for percutaneous thrombectomy or O2083 for surgical thrombectomy or revision).New vascular access formation was defined as the presence of procedure code for new VA formation (O2011, O2012, and O2081 for new AVF or O2082 for new AVG).
The intervention-free interval was defined as the interval between end-date of each HD quality assessment program and the performance of intervention or thrombectomy.The thrombosis-free interval was defined as the interval between the end date of each HD quality assessment program and the performance of thrombectomy.The VAS interval was defined as the interval between the end date of each HD quality assessment program and new VA formation.Further, we evaluated the cases of angioplasty or thrombectomy divided by 1000 person-years.Person-year was calculated using the interval between the end date of each HD quality assessment program and new VA formation or censoring.
The outcomes were followed up until April 2022.If the patient was transferred for peritoneal dialysis, underwent kidney transplantation, or died during follow-up, that date was considered the end point of follow-up, and the data were censored.During follow-up, clinical outcomes except death were defined using the electronic data.The codes were O7072, O7071, and O7061 for peritoneal dialysis and R3280 for kidney transplantation.The data for patient's death were obtained from the HIRA database.Multivariate Cox regression analyses were performed using the enter mode.Statistical significance was set at P < 0.05.

| RESULTS
The numbers of patients in Groups 1, 2, and 3 were 28,521, 9571, and 16,811, respectively (Table 1).The patients in Group 2 were the oldest and had the greatest HD vintage and KtV urea among the three groups.One of the three groups, Group 3 had the highest number of patients who were male, had DM, used statins, or used aspirin; had the highest CCI scores, ultrafiltration volume, systolic blood pressure, and serum creatinine; and had the highest proportion of patients without monitoring.The 5-year survival rates in Groups 1, 2, or 3 were 71.2%, 69.6%, and 69.9%, respectively, for IFS rates (Figure 1A); 79.8%, 78.6%, and 79.3%, respectively, for TFS rates (Figure 1B); and 90.8%, 90.1%, and 90.2%, respectively, for VAS rates (Figure 1C).
No significant difference in the three access survival rates was identified among the three groups.Statistical significance was not obtained in the univariate Cox regression analyses.However, the multivariate Cox regression analyses indicated that Group 3 had better outcomes in IFS and TFS than Group 1 (Table 2).No significant differences in IFS, TFS, and VAS rates according to the type of vascular access were observed among the three groups in the subgroup analyses (Table 3).Group 2 or 3 showed better trends for vascular access-related outcomes compared to Group 1, without statistical significance.
We performed subgroup analyses based on the timing of the HD quality assessment program and HD vintage (Tables S2 and S3).Most  respectively.Thrombectomy cases in Groups 1, 2, and 3 were 227, 236, and 212 cases/1000 person-years, respectively.The numbers of angioplasties performed were significantly greater in Group 1 than in the other two groups.The numbers of thrombectomies performed were significantly the lowest in Group 3 among all the groups.

| DISCUSSION
We analyzed the data of 54,903 patients who underwent HD and the HD quality assessment program in South Korea.The Kaplan-Meier curves did not indicate any significant differences in IFS, TFS, and VAS rates among the three groups.The multivariate Cox regression analyses also did not demonstrate significant differences in IFS, TFS, and VAS rates among the three groups except for the marginal benefit for IFS and TFS in Group 3 compared to Group 1.These trends were also similar to those from analyses according to the type of vascular access.However, Group 3 had lower trend in the number of angioplasties or thrombectomies received than the other groups.
5][6][7][8] Hence, the use of RASBs as blood pressure-lowering drugs may be associated with additional benefits for vascular access-related outcomes.However, clinical studies have yielded inconsistent results on the efficacy of RASBs in vascular access-related outcomes.Heine et al included 137 patients with AVF creation and compared unassisted access patency between patients with and without RASBs. 10The patients were followed up for 12 months; however, no significant difference in unassisted access patency was observed between the two groups.Further, no significant association between polymorphism of the angiotensinconverting enzyme and vascular access outcomes was identified.
Wang et al analyzed the data from a prospective multicenter cohort in the United States and included 602 patients with newly created AVFs. 11They compared the access maturation according to various antihypertensive medications and revealed that calcium channel blockades resulted in better overall AVF maturation than other medications.However, the use of RASBs was not associated with AVF maturation.A single-center study including 349 patients with newly created AVFs also failed to identify the beneficial effect of RASBs on AVF patency. 2 In a study of 121 patients with AVG, Gradzki observed better primary access failure in patients with angiotensin-converting enzyme inhibitors than those without the drug. 12Saran et al analyzed the data of the Dialysis Outcomes and Practice Patterns Study cohort including 900 AVF and 1944 AVG. 13 They noted better secondary AVF patency in patients with the angiotensin-converting enzyme inhibitors than those without the drug.These inconsistent results indicate the need for further research to identify the effects of RASBs on vascular access-related outcomes.
Our results suggest two aspects: access survival and the number of interventions during follow-up.Previous guidelines have recommended the use of primary patency, assisted primary patency, and cumulative patency for vascular access-related outcomes. 9These were defined as intervals between vascular access creation and first intervention, thrombectomy, or abandonment of vascular access, respectively.However, our study did not include the data for the time of vascular access creation and interventions before each HD quality assessment program.Therefore, definitions for vascular access-related survival were revised from the previous guidelines.Definitions of IFS, TFS, or VAS in our study were similar to those of primary patency, assisted primary patency, or cumulative patency.For access survival, we failed to identify a significant association in the three survival rates among the three groups in the univariate analyses.However, the multivariate analyses demonstrated a beneficial effect of the use of RASBs for IFS and TFS in Group 3 compared to that in Group 1.
Differences in IFS or TFS between the two groups were marginally significant.Additionally, we did not identify significant differences in the three vascular access-related survivals in the subgroup analyses according to AVF or AVG among the three groups.
Data were analyzed using the SAS Enterprise Guide version 7.1 (SAS Institute, Cary, NC, USA) or R version 3.5.1 (R Foundation for Statistical Computing, Vienna).Categorical variables are presented as numbers and percentages, whereas continuous variables are presented as means ± standard deviations.Pearson's χ 2 test or Fisher's exact test was used to analyze the categorical variables.For continuous variables, the means were compared using a one-way analysis of variance, followed by the Tukey post hoc test.The survival estimates were calculated using the Kaplan-Meier curve and Cox regression analyses.P-values for the comparison of survival curves were determined using the log-rank test.Multivariate Cox regression analyses were adjusted for age, sex, type of vascular access, underlying disease of end-stage renal disease, monitoring, surveillance, CCI score, HD vintage, ultrafiltration volume, Kt/V urea , hemoglobin, serum albumin, serum creatinine, serum phosphorus, serum calcium, systolic blood pressure, diastolic blood pressure, use of aspirin, and use of statin.

F I G U R E 1
Kaplan-Meier curves of vascular access-related outcomes according to groups.(A) Intervention-free survival curve (P = 0.800).(B) Thrombosis-free survival curve (P = 0.500).(C) Vascular access survival curve (P = 0.700).T A B L E 2 Cox regression analyses for vascular access-related outcomes.
Meanwhile, we evaluated the number of interventions during follow-up and calculated the number of interventions according to groups at 1000 person-years.Each patient had a different follow-up duration, and we used the number of interventions at 1000 personyears.The number of angioplasties performed in Groups 2 and 3 was similar, although that in Group 1 had 1.13 times higher than those in the other two groups.The number of thrombectomies in Group 2 was 1.11 times higher than that in Group 3. Significant results were obtained in some vascular access-related outcomes, although the findings remain insufficient to confirm the evident efficacy of RASB in vascular access-related outcomes.Our study has the following strengths.First, our study included a large sample size.The HD quality assessment program was planned to improve the outcomes of patients on maintenance HD.The HIRA, as a Korean government agency, performed HD quality assessments in almost all HD facilities at regular intervals.The total number of patients who underwent HD in South Korea was approximately 52,378 in 2013, 62,634 in 2015, and 77,617 in 2018.1 In total, we analyzed 46.0% of patients on maintenance HD in South Korea at the time of the relevant HD quality assessment.Second, the study included various clinical and laboratory data despite using a representative sample.We analyzed the following two datasets that were merged as one: data from each HD quality assessment and claims data with medications, procedures, or comorbidities of relevant patients from the HIRA.In this study, a positive association was not observed between RASB use and vascular access outcomes.This may be associated with the factor that the vascular access outcome is a result of the complex interplay of several risk factors, and patients on HD have heterogeneous characteristics.Vascular access outcomes are known to be influenced by various factors, including demographic factors such as age, sex, race/ethnicity; clinical comorbidities like cardiac, pulmonary, or peripheral vascular diseases, diabetes, obesity; as well as technical factors like surgical experience, vascular access care, intervention methods, needling practices, and post-needling care.[19][20][21] Cox regression analyses for vascular access-related outcomes according to type of vascular access.Multivariate analysis was adjusted for age, sex, underlying disease of end-stage renal disease, Charlson comorbidity index score, hemodialysis vintage, ultrafiltration volume, Kt/V urea , hemoglobin, serum albumin, serum creatinine, serum phosphorus, serum calcium, systolic blood pressure, diastolic blood pressure, monitoring, surveillance, use of aspirin, and use of statin and was performed using enter mode.
Note: Multivariate analysis was adjusted for age, sex, type of vascular access, underlying disease of end-stage renal disease, Charlson comorbidity index score, hemodialysis vintage, ultrafiltration volume, Kt/V urea , hemoglobin, serum albumin, serum creatinine, serum phosphorus, serum calcium, systolic blood pressure, diastolic blood pressure, monitoring, surveillance, use of aspirin, and use of statin and was performed using enter mode.Abbreviations: CI, confidence interval; HR, hazard ratio; IFS, intervention-free survival; TFS, thrombosis-free survival; VAS, vascular access survival.analysesdid not show a statistically significant association between RASB use and vascular access outcomes and these were similar to those from the total cohort.Total person-year values in Groups 1, 2, and 3 were 103,274, 40,756, and 71,375 person-years, respectively.Angioplasty cases in Groups 1, 2, and 3 were 443, 391, and 391 cases/1,000 person-years, T A B L E 3 Abbreviations: CI, confidence interval; HR, hazard ratio; IFS, intervention-free survival; TFS, thrombosis-free survival; VAS, vascular access survival.