Systemic lupus erythematosus diagnosis impacts clinical outcomes of arteriovenous fistulas in comparison to other end-stage renal disease etiologies

Abstract

Objectives

Arteriovenous fistulas primary patency at one-year occurs in 43–85% of the patients with end-stage renal disease. The diagnosis attributable to end-stage renal disease has been suggested to impact arteriovenous fistulas outcomes. The objective was to compare primary patency at one week, 1, 3, 6, and 12 months of follow-ups, among systemic lupus erythematosus patients and two control groups; additionally, we evaluated the impact of systemic lupus erythematosus to predict early patency loss.

Methods

A retrospective review of charts from arteriovenous fistulas created between 2008 and 2017 was performed. One-hundred thirty-four patients were identified and classified according to end-stage renal disease attributable diagnosis as: systemic lupus erythematosus cases (N = 14), control-group-1 (91 patients with primarily diabetes and hypertension), and control-group-2 (29 patients with idiopathic end-stage renal disease). A case–control matched design (1:2:1) was proposed. Logistic regression analysis and Kaplan–Meier curves were used. Institutional Review Board approval was obtained.

Results

More systemic lupus erythematosus patients lost primary patency at 3 (28.6%) and 12 months (71.4%) than patients from control-groups-1 (vs. 3.6% and 35.7%, respectively) and -2 (vs. 0% and 14.3%, respectively), (p ≤ 0.011 for both). Days of primary patency survival were shorter in systemic lupus erythematosus patients (p = 0.003). Systemic lupus erythematosus diagnosis was the only factor associated with early patency loss, HR: 3.141, 95%CI: 1.161–8.493 (systemic lupus erythematosus diagnosis vs. control-group-1) and HR: 12.582, 95%CI: 1.582–100.035 (systemic lupus erythematosus diagnosis vs. control-group-2).

Conclusions

Diagnosis attributable to end-stage renal disease has a major impact on arteriovenous fistula outcomes in patients. Systemic lupus erythematosus patients have an increased risk of arteriovenous fistulas patency loss within the first six months of follow-up. Patients with idiopathic end-stage renal disease had an excellent one year arteriovenous fistula patency survival.

Keywords

Arteriovenous fistulas systemic lupus erythematosus primary patency

Introduction

The number of patients with end-stage renal disease (ESRD) rises by about 20,000 patients per year and substantially contributes to overall mortality.^1,2 Many ESRD patients are candidates for kidney transplantation, and they sometimes require the creation of arteriovenous fistulas (AVFs) while waiting for a kidney. Since 1997, different vascular access guidelines have been published in an effort to increase the placement of AVFs as the primary modality of hemodialysis,^3,4 which is nowadays considered the optimal access method for hemodialysis in ESRD patients.⁵ Multiple risk factors may play a role in AVF patency. Among the most studied ones are age, sex, diabetes mellitus (DM), atherosclerosis, smoking, and the concomitant use of medications.⁶ Besides DM diagnosis which has been associated to poor AVF-related outcomes,^6–10 few studies have evaluated whether the diagnosis attributable to ESRD is an important risk factor for AVF patency.

Approximately 10–15% of cases of ESRD are classified as idiopathic,^11,12 and such cases had shown a dramatic increase in the Central America region¹³; it seems important to know how this group behaves in terms of AVF patency. On the other hand, systemic lupus erythematosus (SLE) is an autoimmune disease, with frequent kidney involvement due to lupus nephritis, which progresses to ESRD in 10–30% of the patients, even with appropriate medical interventions.^14–16 Hispanic SLE patients are characterized by a more severe disease presentation, and their renal manifestations are of particular concern.¹⁴ There is limited published information regarding AVF patency in SLE patients. One study that involved primarily African-American patients from the USA showed that SLE patients undergoing hemodialysis were more likely to develop vascular access thrombosis than those with additional diagnoses, the most frequent being glomerular disease, DM, and hypertension.¹⁷

The present study was conducted at the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, an experienced center regarding SLE management, with more than 4000 SLE patients registered and 1500 SLE patients attended yearly.^18,19 The primary objective was to compare primary patency at 12 months of follow-up among SLE patients and two control groups. Secondary objectives were to compare primary patency at one week and one, three, and six months of follow-up among the three groups and to investigate the impact of SLE diagnosis as a risk factor for early patency loss.

Methods

Ethical consideration

This was a retrospective study based on chart review. All data from patients have been de-identified and no personal information is available at any point. The study was approved by the Institutional Review Board (Reference number: SCI-2533–18-19–1).

Population study and group definitions

For the current report, we identified AVFs created between 2008 and 2017, as well as the diagnoses that defined the cause of ESRD. A total of 134 patients were identified during the study period. Three groups were defined based on the diagnosis attributable to ESRD. Group 1 (cases, N = 14) consisted of patients with ESRD secondary to SLE. SLE diagnosis was established by a rheumatologist, and American College of Rheumatology (ACR)/Systemic Lupus International Collaborating Clinics (SLICC) 2012 criteria were additionally met.²⁰ Control-group-1 (N = 91) consisted of patients with chronic non-autoimmune diseases, basically, those with DM and hypertension. Finally, control-group-2 (N = 29) consisted of patients with idiopathic ESRD. Idiopathic etiology was considered after a rational evaluation for the potential causes of ESRD was discarded by both an internist and a nephrologist.

Outcome definitions

Our primary outcome was defined as loss of AVF primary patency at 12 months of follow-up, according to the standard definition.²¹ Secondary outcomes were established as loss of primary patency at one week and one, three, and six months of follow-up. In addition, for each patient, the effective period of AVF patency was calculated.

Finally, early patency loss was defined as loss of AVF patency in the first six months of follow-up.

Data collection

A retrospective chart review was performed. A standardized format was previously designed and included the following variables: sociodemographic characteristics, primary diagnosis attributable to ESRD, comorbid conditions established according to the Charlson comorbidity index,²² type and localization of AVFs, treatment at AVF intervention, and disease severity (evaluated according to patients’ emergency room visits and/or previous hospitalizations). Two data abstractors were trained to ensure homogeneity during data collection. In addition, an independent observer blinded to the abstractors’ classifications confirmed the primary and secondary outcomes. We excluded two patients with incomplete information regarding their primary outcomes.

Statistical analysis

Descriptive statistics were used. The distribution of variables was first observed. The X² test was used to compare non-normally distributed dichotomous variables, and Kruskal–Wallis test was used to compare non-normally distributed continuous variables.

A case–control study (1:2:1) was designed. Cases (N = 14, SLE patients) were compared to patients from control-groups-1 (N = 28) and -2 (N = 14).

After a careful literature review,^4,6 the following variables were considered for matching: sex, age at fistula creation, type of fistula (autologous vs. grafts), localization of fistula, use of anticoagulants and/or antiplatelet agents at surgery, and Charlson index. Due to our patients’ characteristics, matching was possible only for individual variables from 50% (sex matching) to 100% (AVF type matching).

A Cox regression model was used to define SLE diagnosis as a risk factor for early patency loss. Reference groups were control groups. Due to the limited number of outcomes (early patency loss), only one additional variable was included in the models tested to avoid overfeeding. This additional variable was selected based on its statistical significance in the univariate analysis (age at AVF creation, years of formal education, and medium-low socioeconomic status). Finally, Kaplan–Meier curves were used to define 12-month patency survival in each group. A log-rank test was used to compare primary patency among cases and both control groups. Analysis were repeated considering one reference group, which was integrated with 42 control patients, 28 patients with ESRD related to chronic non-autoimmune diseases, and 14 patients with idiopathic ESRD; results were similar and are presented as Supplementary material.

Due to the limited number of patients with SLE-ESRD diagnosis included, we estimated sample size power to detect differences among groups, assuming that SLE patients will present twice as much risk as control patients for AVFs patency loss¹⁷ and found power (1-β error probability) of 0.839, with a critical z value of 1.64 (one tail).

All statistical tests were two-sided and evaluated at the 0.05 significance level. Statistical analysis was performed using SPSS (v.17.0; SPSS, IBM, Armonk, NY).

Results

Population characteristics

Table 1 summarizes data from the overall population and compares sociodemographic characteristics, diagnosis attributable to ESRD, anticoagulant and antiplatelet usage at procedure, Charlson score, and disease severity before the procedure between cases and controls. In the entire population, the majority of the patients (82.1%) were female, but the percentage decreased to 50% in patients from control-group-2. This percentage was lower than cases (vs. 92.9%) and patients from control-group-1 (vs. 92.9%) (p = 0.003). In addition, the median (IQR) age at AVF creation in the entire population was 32.4 years (25.5–44.6). Patients from control-group-1 (42 years (28–51.4)) were significantly older than cases (vs. 28 years (23–37.8)) and patients from control-group-2 (vs. 28.7 years (21.3–35)), respectively, (p = 0.009).

Table 1.

Comparison of baseline characteristics between groups (G).

Variables	Overall populationN = 56	CasesN = 14	Control-G-1N = 28	Control-G-2N = 14	p Values
Sociodemographic characteristics
Female gender	46 (82.1)	13 (92.9)	26 (92.9)	7 (50)	0.003
Age at ESRD attributable diagnosis, years^a	23.4 (14.8–32.2)	20.9 (17–26.9)	28.6 (15.7–35.7)	21 (12.9–26.8)	0.274
Age at surgery, years^a	32.4 (25.5–44.6)	28 (23–37.8)	42 (28–51.4)	28.7 (21.3–35)	0.009
Years of formal education^a	12 (9–17)	14.5 (12–17)	12 (9–12)	12 (9–12)	0.108
Low-medium socioeconomic level	34 (60.7)	12 (85.7)	12 (42.9)	10 (71.4)	0.018
Hispanic ethnicity	53 (94.6)	14 (100)	25 (89.3)	14 (100)	0.205
Diagnosis attributable to ESRD
DM	13 (23.2)	0	13 (46.4)	0	0.000
Hypertension	13 (23.2)	0	13 (46.4)	0	0.000
SLE	14 (25)	14 (100)	0	0	0.000
Other diagnosis	13 (23.2)	0	13 (46.4)	0	0.000
Anticoagulation	6 (10.7)	5 (35.7)	1 (3.6)	0	0.002
Antiagreggation	3 (5.4)	0	2 (7.1)	1 (7.1)	0.59
Charlson score^a	4 (3–5)	3 (3–4.3)	5 (2–5.8)	2 (2–3)	0.001
Disease severity previous to the procedure
Previous hospitalizations	9 (16.4)	4 (28.6)	5 (18.5)	0	0.113
Emergency department visit in the previous six months to surgery	5 (9.1)	3 (21.4)	1 (3.7)	1 (7.1)	0.166

Note: Data presented as N (%) of patients.

^aMedian (IQR).

ESRD: end-stage renal disease; DM: diabetes mellitus; SLE: systemic lupus erythematosus.

In the entire population, 60.7% of patients had a medium-low socioeconomic status. A higher percentage of cases (85.7%) and patients from control-group-2 (71.4%) were of medium-low socioeconomic status than patients from control-group-1 (vs. 42.9%) (p = 0.018). As expected, significant differences were found regarding patients’ main diagnosis attributable to ESRD. In the entire population, only a minority of the patients (10.7%) were on anticoagulant medication previous to the AVF procedure. This percentage was higher among cases when compared to both control groups (35.7%, 3.6%, and 0%, respectively) (p = 0.002). Finally, the median Charlson score in the overall population was 4 (3–5). Patients from control-group-1 had higher Charlson scores (5 (2–5.8)) than cases (vs. 3 (3–4.3)) and patients from control-group-2 (vs. 2 (2–3)), (p = 0.001).

Characteristics particular to SLE patients

In addition to demographic characteristics described in Table 1, SLE patients had moderate disease activity according to the SLE Disease Activity Index (SLEDAI)²³ and important damage according to the 2012 SLICC criteria.²⁴ The mean SLEDAI score was 7.8 ± 5.6, and the mean SLICC score was 6.8 ± 2. The majority of the patients were on corticosteroids (78.6%) and their mean dose of daily oral prednisone was 10 ± 13.3 mg. Finally, up to 42.9% of the patients were on immunosuppressive drugs, meanwhile five patients were on anticoagulants. All patients on anticoagulant drugs were diagnosed with secondary antiphospholipid syndrome. None of the SLE patients was on biologic drugs.

AVF types and localization

Table 2 describes AVF types and sites of creation in the entire population, and in cases and control groups. Among the three groups, different AVF configurations were created, according to individual arterial and venous anatomies. The majority of the patients underwent autologous fistula creation (91.1%), and this percentage was similar among the groups. In addition, the most frequent site of fistula creation was brachiocephalic (60.7%), followed by radiocephalic (17.9%), without significant differences in their distribution among the groups.

Table 2.

Comparison of AVFs procedures features between groups (G).

Variables	Overall populationN = 56	CasesN = 14	Control-G-1N = 28	Control-G-2N = 14	p Values
AVFs type
Arteriovenous fistula	51 (91)	14 (100)	24 (85.7)	13 (92.9)	0.299
Arteriovenous graft	5 (8.9)	0	4 (14.3)	1 (7.1)
Localization of AVFs
Radio-cephalic	10 (17.9)	2 (14.3)	6 (21,4)	2 (14.3)	0.8
Brachiocephalic	34 (60.7)	9 (64.3)	15 (53.6)	10 (71.4)
Brachiobasilic	6 (10.7)	2, (14.3)	3 (10.7)	1 (7.1)
Brachioaxilar	3 (5.4)	0	3 (5.4)	0
Period of time from diagnosis to fistula creation, median (IQR)	8.7 (3.4–17.2)	7.4 (3.4–12.6)	12.1 (5–21.5)	5 (2.4–13)	0.126

Note: Data presented at N (%), unless otherwise indicated.

AVF: arteriovenous fistula.

Outcomes

Table 3 and Figure 1 summarizes the comparison of primary and secondary outcomes among groups. As shown, more SLE patients lost primary patency at 3 months (28.6%) and 12 months of follow-up (71.4%) than patients from control-groups-1 (vs. 3.6% and 35.7%, respectively) and -2 (vs. 0% and 14.3%, respectively) (p = 0.011 and p = 0.007, respectively). In accordance, the median (IQR) days of primary patency survival was shorter in SLE patients than in patients from control-groups-1 and -2: 97.5 days (49–360), 360 days (168.8–360), and 360 days (360–360), respectively, p = 0.003. Figure 2(a) shows the comparison of Kaplan–Meier curves representing AVF patency survival from maturation up to 12 months of follow-up. Of note, SLE patients showed a dramatic decline in survival; meanwhile, the opposite was true for patients assigned to control-group-2 (log-rank test, p = 0.002).

Table 3.

Comparison of primary and secondary outcomes between groups (G).

	Overall populationN = 56	CasesN = 14	Control-G-1N = 28	Control-G-2N = 14	p Values
Loss of primary patency at 12 months	22 (39)	10 (71.4)	10 (35.7)	2 (14.3)	0.007
Days of primary patency survival median (IQR)	360 (135–360)	97.5 (49–360)	360 (168.8–360)	360 (360–360)	0.003
Loss of primary patency at one week	1 (1.8)	0	1 (3.6)	0	0.601
Loss of primary patency at one month	5 (8.9)	3 (21.4)	2 (7.1)	0	0.124
Loss of primary patency at three months	5 (8.9)	4 (28.6)	1 (3.6)	0	0.011
Loss of primary patency at six months	6 (10.7)	2 (14.3)	3 (10.7)	1 (7.1)	0.83

Note: Data presented as N (%) of patients, unless otherwise indicated.

Figure 1.

Forest Plots of individual hazard ratio to predict early patency loss. I bars indicate 95% confidence intervals.

Figure 2.

(a) Kaplan–Meier estimates of primary patency survival in cases (14 SLE patients) and both control groups. (b) Kaplan–Meier estimates of primary patency survival in 14 SLE patients and 42 control patients.

Similar results were obtained when patients from both control groups were pooled and their outcomes compared to those from the cases, as summarized in Supplementary Table 1 and Figure 2(b).

SLE diagnosis as a risk factor for early patency loss

In the overall population, early patency loss occurred in 17 patients; among them, 9 were assigned to the SLE group. Meanwhile, only one early patency loss occurred in control-group-2.

In the different Cox-regression models tested, SLE diagnosis was the only variable consistently associated with early patency loss as summarized in Table 4 and results were reproduced when patients from both control groups were pooled (HR: 4.356, 95%CI: 1.663–11.357, p = 0.003). Interestingly, patients diagnosed with SLE had almost 12 times increased risk of early patency loss compared to patients from control-group-2 and two times as much risk as patients from control-group-1.

Table 4.

Cox-regression model to predict early patency loss.

	Hazard ratio	95% CI	p Values
SLE diagnosis versus control-group-1	3.141	1.161–8.493	0.024
SLE diagnosis versus control-group-2	12.579	1.582–100.035	0.017

SLE: systemic lupus erythematosus; CI: confidence interval.

Discussion

In the present study, SLE diagnosis was the only factor associated with early patency loss, which was defined as AVF patency loss in the first six months of follow-up after vascular access maturation. The magnitude of risk of patency loss in SLE patients was overwhelming compared to patients with idiopathic ESRD and substantial compared to patients with chronic non-autoimmune diseases. In accordance, days of primary patency survival in SLE patients were reduced to a median of 97.5 days compared to those in patients in both control groups.

Few studies have assessed vascular access implementation¹⁷ and vascular access survival^17,25 in SLE patients, with conflicting results. The paucity of published studies may be related to the fact that SLE is a rare autoimmune disease, with variable clinical courses, substantial comorbidities, and frequently requiring intensive immunosuppressive treatment to achieve disease control.^14,15,18 In addition, SLE has also been considered a hypercoagulable state, particularly when associated with secondary antiphospholipid syndrome.^17,25,26 All characteristics mentioned may affect surgical indication, including vascular access creation, due to concerns for increased complication rates.^15,16,27 Similar to our results, Shafi and Gupta¹⁷ demonstrated that SLE patients on hemodialysis were more likely to develop vascular access thrombosis than non-SLE patients. In particular, the thrombosis rate was significantly higher in women, African-Americans, and patients with AV grafts in the SLE cohort, than in their corresponding subgroups of the non-SLE population. Similar to patients of Hispanic ethnicity, those of the African-American origin have been associated with a more severe presentation and poor outcomes in the context of SLE patients,^14,15 although a multi-ethnic US cohort of SLE patients failed to identify race as a risk factor for arterial or venous thrombotic events.²⁸ Authors also included a control group of non-SLE patients, matched with relevant demographic variables such as age, sex, and ethnicity, but ESRD-associated diagnosis was not specified.¹⁷ On the contrary, Plantinga et al.²⁵ compared early loss of patency between SLE-ESRD and other ESRD patients, including patients with DM and hypertension. They defined early patency loss as if patients lost vascular access patency during the first year of follow-up. Rates of access patency loss were similar between both groups after adjustments. Differences between Plantinga et al.’s study²⁵ and ours may be related to a limited representation in their study of Hispanic white patients (13.9%) and patients without insurance (14.5%), both of whom had been associated with poor outcomes.^14,15 Also, the presence or absence of loss of patency was defined based on claims data, while we performed a retrospective review of the charts and corroborated primary outcome with an independent observer.

A second important finding was the variability in the patency survival among the three groups of patients. SLE patients represented the worst scenario with 28.6% of patency survival at 12 months of follow-up, which is slightly lower than the 34.4% described by Shafi and Gupta¹⁷ and substantially lower than the 56.2% described by Plantinga et al.²⁵ As we mentioned previously, differences in demographic and clinical characteristics among SLE subpopulations may account for the different outcomes. Also, flares may develop during the course of SLE patients’ follow-up, and vascular access implementation, instead of being elective, may become an emergency procedure, making it impossible to adjust treatment before AVF creation.¹⁸ Meanwhile, the opposite was true for idiopathic ESRD patients who showed a patency survival of 85.7% at 12 months of follow-up. This nosologic entity is clinically silent and carries a poor prognosis because it remains undetected until advanced stages of renal failure.^13,29–31 Our study adds important information relevant to the topic, because poor outcomes had been primarily attributed to the lack of access to renal replacement therapy of the affected patients, which are patients from low-income communities in Central America and South Asia. Nonetheless, if universal healthcare is adopted, patients may benefit from renal transplantation. Many of these potential candidates will need AVFs while awaiting for a kidney transplant³¹ and our study supports this recommendation. Finally, intermediate outcomes were found in patients assigned to control-group-1 (64.3%), mainly consisting of patients with DM and patients with hypertension. Similar results have been described in the literature, about one year AVF primary patency rates ranging from 43–85%.^32–34

There are limitations of this study that need to be addressed. First, this is a retrospective study and potential prognostic variables may not have been reliably identified. Also, the majority of the patients had medium-low socioeconomic status which precludes solid analysis regarding the impact of social determinants on diseases outcomes. In addition, a priori matching was incomplete due to differences in the populations’ characteristics. Second, we had a reduced number of patients with early patency loss which limited the number of variables to be included in the regression models and the ability to consider additional potential confounding variables. Third, Latin American patients with SLE had shown unique behaviors in terms of prognosis and outcomes and this limits the generalization of the results.

Our study has strengths that need to be emphasized. We present the largest series of AVFs created in Hispanic SLE patients and compared their AVF-related outcomes with two control groups, and exhaustive matching was attempted although difficult to achieve. One of the control groups consisted of patients with idiopathic ESRD, in whom published information regarding outcomes is lacking. Also, a standardized format was designed to gather information from charts, and data retrievers were previously trained. In addition, an independent observer corroborated the primary outcomes.

Conclusions

The diagnosis attributable to ESRD has a major impact on AVF outcomes in Hispanic patients from a tertiary care level center. In particular, SLE patients have an increased risk of AVF patency loss within the first six months of follow-up compared to patients with chronic diseases such as diabetes and hypertension and those with idiopathic ESRD. This group of patients appears to have excellent one year AVF patency survival, which may be related to a younger age at surgery and a larger proportion of men. The high rate of patency loss in SLE patients is a matter that needs to be deeply discussed with a multidisciplinary team. Further research is necessary to increase our understanding of factors associated with AVF failure and to provide optimal access to this complex group of patients. This may be more easily achieved through cooperation between centers in charge of these patients, in order to achieve a large cohort sample to support solid statistical analysis.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Carlos A Hinojosa

Supplemental material

Supplemental material for this article is available online.

References

United States Renal Data System (USRDS). Incidence, prevalence, patient characteristics, and treatment modalities. In: USRDS 2015 annual report: epidemiology of kidney disease in the United States , National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disease, Bethesda, MD, 2015, www.usrds.org/2015/view/v2_01.aspx (accessed 25 May 2016).

GBD 2016 Causes of Death Collaborators. Global, regional, and national age-sex specific mortality for 264 causes of death, 1980–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017; 390: 1151–1210.

Gold

Hoffman

Fistula first: the national vascular access improvement initiative. WMJ 2006; 105: 71–73.

Sidawy

Spergel

Besarab

, et al. The society for vascular surgery: clinical practice guidelines for the surgical placement and maintenance of arterio-venous hemodialysis access. J Vasc Surg 2008; 48: 2S–25S.

Allon

Vascular access for hemodialysis patients, new data should guide decision making. Clin J Am Soc Nephrol 2019; 14: 954–961.

Smith

Gohil

Chetter

IC.

Factors affecting the patency of arteriovenous fistulas for dialysis access. J Vasc Surg 2012; 55: 849–855.

Sedlacek

Teodorescu

Falk

, et al. Hemodialysis access placement with preoperative noninvasive vascular mapping: comparison between patients with and without diabetes. Am J Kidney Dis 2001; 38: 560–564.

Sesso

Melaragno

Luconi

, et al. Survival of dialyzed diabetic patients. Rev Assoc Med Bras 1995; 41: 178–182.

Resic

Sahovic

Mesic

Predictors of AV fistula adequacy in haemodialysed patients. Med Arh 2005; 59: 177–178.

10.

Konner

Hulbert-Shearon

Roys

, et al. Tailoring the initial vascular access for dialysis patients. Kidney Int 2002; 62: 329–338.

11.

PK.

Chronic kidney disease epidemic: how do we deal with it?

Nephrology (Carlton) 2018; 23: 116–120.

12.

Saran

Robinson

Abbott

, et al. US Renal Data System 2016 Annual Data Report: epidemiology of kidney disease in the United States. Am J Kidney Dis 2017; 69: S1–S688.

13.

Pearce

Caplin

Gunawardena

, et al. CKD of unknown cause: a global epidemic? Kidney Int Rep 2018; 4: 367–369.

14.

Gergianaki

Bortoluzzi

Bertsias

Update on the epidemiology, risk factors, and disease outcomes of systemic lupus erythematosus. Best Pract Res Clin Rheumatol 2018; 32: 188–205.

15.

Inda-Filho

Neugarten

Putterman

, et al. Improving outcomes in patients with lupus and end stage renal disease. Semin Dial 2013; 26: 590–596.

16.

Costenbader

Desai

Alarcón

, et al. Trends in the incidence, demographics, and outcomes of end-stage renal disease due to lupus nephritis in the US from 1995 to 2006. Arthritis Rheum 2011; 63: 1681–1688.

17.

Shafi

Gupta

Risk of vascular access thrombosis in patients with systemic lupus erythematosus on hemodialysis. J Vasc Access 2007; 8: 103–108.

18.

Barrera-Vargas

Quintanar-Martínez

Merayo-Chalico

, et al. Risk factors for systemic lupus erythematosus flares in patients with end-stage renal disease: a case-control study. Rheumatology (Oxford) 2016; 55: 429–435.

19.

Mejía-Vilet

Arreola-Guerra

Córdova-Sánchez

, et al. Comparison of lupus nephritis induction treatments in a hispanic population: a single-center cohort analysis. J Rheumatol 2015; 42: 2082–2091.

20.

Petri

Orbai

Alarcón

, et al. Derivation and validation of Systemic Lupus International Collaborating Clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum 2012; 64: 2677–2686.

21.

Lee

Mokrzycki

Moist

, et al. Standardized definitions for hemodialysis vascular access. Semin Dial 2011; 24: 515–524.

22.

Quan

Couris

, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol 2011; 173: 676–682.

23.

Gladman

Ibanez

Urowitz

MB.

Systemic lupus erythematosus disease activity index 2000. J Rheumatol 2002; 29: 288–291.

24.

Bakula

Cikes

Anic

Validation of the new classification criteria for systemic lupus erythematosus on a patient cohort from a national referral center: a retrospective study. Croat Med J 2019; 60: 333–344.

25.

Plantinga

Lim

Patzer

, et al. Comparison of vascular access outcomes in patients with end-stage renal disease attributed to systemic lupus erythematosus vs. other causes: a retrospective cohort study. BMC Nephrol 2016; 17: 64.

26.

Abu-Shakra

Gladman

Urowitz

, et al. Anticardiolipin antibodies in systemic lupus erythematosus: clinical and laboratory correlations. Am J Med 1995; 99: 624.

27.

Haviv

YS.

Association of anticardiolipin antibodies with vascular access occlusion in hemodialysis patients: cause or effect?

Nephron 2000; 86: 447–454.

28.

Alarcón

Calvo-Alen

McGwin

Jr , et al. Systemic lupus erythematosus in a multiethnic cohort: LUMINA XXXV.Predictive factors of high disease activity over time. Arthritis Rheum Dis 2006; 65: 1168–1174.

29.

Whelan

The global epidemic of chronic kidney disease: a call for action. Occup Environ Med 2016; 73: 499–500.

30.

Wesseling

Van Wendel De Joode

Crowe

, et al. Mesoamerican nephropathy: geographical distribution and time trends of chronic kidney disease mortality between 1970 and 2012 in Costa Rica. Occup Environ Med 2015; 72: 714–721.

31.

Raines

González

Wyatt

, et al. Risk factors for reduced glomerular filtration rate in a Nicaraguan community affected by Mesoamerican nephropathy. MEDICC Rev 2014; 16: 16–22.

32.

Choi

Lal

Cerveira

, et al. Durability and cumulative functional patency of transposed and nontransposed arteriovenous fistulas. J Vasc Surg 2003; 38: 1206–1212.

33.

Huber

Carter

, et al. Patency of autogenous and polytetrafluoroethylene upper extremity arteriovenous hemodialysis accesses: a systematic review. J Vas Surg 2003; 38: 1005–1011.