Oral anticoagulation for stroke prevention in atrial fibrillation and advanced kidney disease

Ellen Linnea Freese Ballegaard; Jonas Bjerring Olesen; Anne-Lise Kamper; Bo Feldt-Rasmussen; Gunnar Gislason; Christian Torp-Pedersen; Nicholas Carlson

doi:10.1016/j.rpth.2024.102350

. 2024 Feb 15;8(2):102350. doi: 10.1016/j.rpth.2024.102350

Oral anticoagulation for stroke prevention in atrial fibrillation and advanced kidney disease

Ellen Linnea Freese Ballegaard ^1,^2,^∗, Jonas Bjerring Olesen ³, Anne-Lise Kamper ¹, Bo Feldt-Rasmussen ^1,², Gunnar Gislason ^2,^3,⁴, Christian Torp-Pedersen ⁵, Nicholas Carlson ¹

PMCID: PMC10933578 PMID: 38481950

Abstract

Background

The net benefit of oral anticoagulation (OAC) with vitamin K antagonists or direct oral anticoagulants in patients with advanced chronic kidney disease and atrial fibrillation remains uncertain.

Objectives

We examined the use, efficacy, and safety of OAC in patients with estimated glomerular filtration rate (eGFR) of <30 mL/min/1.73 m² (including dialysis-treated patients) and atrial fibrillation.

Methods

In a retrospective cohort study, patients diagnosed with atrial fibrillation and eGFR of <30 mL/min/1.73 m² were identified in national Danish registers between 2010 and 2022. Initiation of OAC was identified based on redemption of a relevant prescription. One-year risks of thromboembolic event, major bleeding, and death associated with OAC and no treatment were computed and standardized to the distribution of risk factors in the sample based on hazards determined in multiple Cox regression models adjusted for age and sex.

Results

A total of 3208 patients were included (mean age 80 years, 52.8% males, 20.9% chronic dialysis). OAC was initiated in 1375 (42.9%) patients, of whom 48.1% were vitamin K antagonists and 51.9% were direct oral anticoagulants. One-year risks in nontreated and anticoagulated patients were 4.8% (95% CI, 3.8%-5.7%) and 3.6% (95% CI, 2.8%-4.6%; P = .028) for thromboembolic event, 7.6% (95% CI, 6.6%-8.7%) and 10.5% (95% CI, 9.3%-12.1%; P < .001) for major bleeding, and 36.3% (95% CI, 34.2%-38.3%) and 29.6% (95% CI, 27.6%-31.6%; P < .001) for death, respectively.

Conclusion

In a retrospective study on patients with advanced chronic kidney disease and atrial fibrillation, OAC was associated with overall decreased 1-year risk of thromboembolic event and death offset by increased 1-year risk of major bleeding.

Keywords: anticoagulants, atrial fibrillation, hemorrhage, renal dialysis, thromboembolism

Graphical abstract

Essentials

•
Benefit of anticoagulation in atrial fibrillation and advanced kidney disease is unclear.
•
Cohort study on risks and benefits of anticoagulation in patients with advanced kidney disease.
•
Anticoagulation is associated with a reduced risk of stroke and death and an increased risk of bleeding.
•
Results lend support for the overall benefit of anticoagulation in advanced kidney disease.

1. Introduction

Prevalence of atrial fibrillation is inversely correlated with renal function, increasing to 20% in patients receiving dialysis treatment due to end-stage kidney disease (ESKD) [[1], [2], [3], [4]]. Atrial fibrillation is associated with an increased risk of mortality and a 2- to 3-fold increased risk of stroke in patients with ESKD compared with general populations [1]. In patients with preserved renal function and atrial fibrillation, trials with oral anticoagulation treatment have demonstrated a >50% reduction in the risk of stroke [5]. However, patients with advanced chronic kidney disease (CKD) were broadly excluded from these studies. Hence, no randomized controlled trial has evaluated the safety and efficacy of oral anticoagulation compared with placebo in patients with an estimated glomerular filtration rate (eGFR) of <30 mL/min/1.73 m² [6].

Advanced CKD infers concomitant increased risk of stroke and bleeding due to platelet dysfunction and impaired platelet–endothelium interaction [7,8]. Taken together, this generates uncertainty as to the risk-benefit ratio of oral anticoagulation in patients with advanced CKD and atrial fibrillation. Due to the paucity of clinical data and uncertainties related to results from cohort studies, primarily in patients on chronic dialysis or with a definition of CKD based on diagnosis codes without eGFR [[9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22]], current guidelines remain ambiguous, and management strategies divergent [[23], [24], [25]]. Utilizing the unique and comprehensive Danish health care registers, the aim of this study was to evaluate the use, efficacy, and safety of oral anticoagulation in patients with incident atrial fibrillation and eGFR of <30 mL/min/1.73 m² (dialysis and predialysis), with emphasis on net clinical benefit and the attempted balance between risks of major bleeding and thromboembolic event.

2. Methods

2.1. Data availability

As detailed patient data hold potential for reidentification, data for this study were governed by the Danish Data Protection Agency and can only be made available to additional researchers upon formal request to the Danish Authorities. All code is shared for review and reuse under the Statistics Denmark license.

2.2. Data registers

The Danish healthcare system is tax-funded and provides healthcare for all Danish residents. All Danish residents have a unique identification number, which permits cross-linkage of nationwide healthcare registers on an individual basis. The Danish National Patient Registry holds information on all discharge diagnoses and surgical procedures based on the 10th edition of the International Classification of Diseases (ICD-10) and the Nordic Medico-Statistical Committee Classification of Surgical Procedures code [26]. The Danish Prescription Registry provides information on all drugs dispensed from Danish pharmacies in accordance with the Anatomical Therapeutic Chemical Classification System [27]. Biochemistry data from 4 out of 5 Danish regions are available from the Danish National Laboratory Database [28], and the Danish Civil Registration System contains information on causes of death [29]. Loss to follow-up is practically nonexistent in the registers.

2.3. Study population

All patients with incident atrial fibrillation between January 1, 2010, and June 2, 2022, were identified in Danish registers. This included both outpatient and inpatient hospital contacts. In outpatients, the date of discharge was defined as the date of the first hospital contact with a diagnosis code for atrial fibrillation. The diagnosis of atrial fibrillation has been validated in the Danish National Patient Registry with a positive predictive value of 92% between the registry and patients’ charts [30]. Estimation of eGFR was based on the last measurement of creatinine within 2 years to 7 days prior to inclusion using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation [31]. Assessment of eGFR based on the last recorded plasma creatinine has previously been assessed with an intraclass correlation coefficient of 0.88 (95% CI, 0.85-0.91) [32].

Oral anticoagulation treatment was defined as redemption of a prescription for any anticoagulant (Anatomical Therapeutic Chemical Classification System code B01AA, B01AE, or B01AF, Supplementary Table S1) within −30 days to +14 days from discharge, which permitted correct identification of 67.0% of patients initiating treatment. Time from discharge to first prescription is shown in Supplementary Figure S1.

Patients <18 years, patients with preexisting valvular heart disease, and patients on preceding oral anticoagulation treatment as defined by a redeemed prescription between 365 and 30 days prior to discharge were excluded. To avoid immortal time bias and events that could affect treatment strategy, we further excluded nonsurvivors and patients with thromboembolic or bleeding events in the fortnight after discharge. Hence, the index date, ie, the beginning of follow-up, was defined as the date of discharge +14 days.

2.4. Baseline characteristics

Baseline pharmacologic treatments were determined based on prescriptions redeemed in the 6 months preceding the diagnosis of atrial fibrillation. Comorbidities were identified from hospital diagnoses within 5 years before inclusion, with augmentation of hypertension [33] and diabetes diagnoses based on prescription refills. Baseline biochemistry data were defined by the latest record in the year before admission. Dialysis was defined based on registration of any administrative code denoting dialysis due to chronic renal failure within 3 months before the index. All administrative codes used to define the dataset are provided in Supplementary Table S1.

2.5. Outcomes and follow-up

Patients were followed until either emigration, death, thromboembolic event, major bleeding, or June 16, 2022 (end of data). The primary efficacy endpoint was thromboembolic event defined as ischemic or unclassified stroke, transient ischemic attack, or systemic arterial thromboembolism (ICD-10 codes: G458-G459, I63-I64, and I74). Identification of ischemic or unspecified stroke based on ICD-10 code in the Danish Patient Registry has previously been validated with a positive predictive value of 93.5% (95% CI, 82.5%-97.8%) and a negative predictive value of 71.8% (95% CI, 62.8%-79.4%) [26]. The secondary safety endpoints were death or major bleeding defined as gastrointestinal, urogenital, airway, intraocular, cardiac, intradural, retroperitoneal, or intracranial bleeding causing hospitalization or as anemia caused by bleeding (ICD-10 codes: D500, D62, G951A, H313, H431, H356, H450, H052A, I60-62, I312, I850, I864A, J942, K228F, K250, K252, K254, K256, K260, K262, K264, K266, K270, K272, K274, K276, K280, K282, K284, K286, K298A, K638B, K638C, K661, K868G, K920-K922, N02, R04, R31, S064-S066, and S368D). Identification of major bleeding was based on a number of validated administrative codes congruous with preexisting standards [26]. ICD-10 codes defining all endpoints are provided in Supplementary Table S1.

2.6. Statistical analysis

Continuous covariates are presented according to normal distribution as means with SDs or medians with IQR. Comparisons were performed with Student’s t-test or 1-way anova for normally distributed continuous data and Mann–Whitney U-test or Kruskal–Wallis test for skewed continuous data. Categorical covariates are presented as frequencies and percentages and compared with chi-squared test. Patients with missing descriptive data, ie, missing hemoglobin level and/or missing platelet count, were included in all analyses. Calculation of median follow-up was performed based on the reverse Kaplan–Meier estimator.

One-year standardized risks of thromboembolic event, major bleeding, and death were calculated with the comparison of oral anticoagulation treatment and no oral anticoagulation treatment based on Cox regression models adjusted for age and sex with G-computation of 1-year risks standardized to the distribution of risk factors in the sample. The assumption of proportional hazards for the Cox regression models was tested graphically with plots of cumulative martingale residuals. To evaluate robustness, principal analyses were additionally tested in multiple Cox regression models adjusted for the parameters in the age, sex, congestive heart failure, hypertension, diabetes, previous stroke, and peripheral vascular disease (CHA₂DS₂-VASc) score, previous major bleeding, and antiplatelet treatment, and in inverse-probability-weighted analyses based on person-specific propensities for anticoagulation estimated in multiple logistic regression models adjusted for age and sex alone and for CHA₂DS₂-VASc score parameters, previous major bleeding, and antiplatelet treatment [34]. Subgroup analyses stratified on dialysis treatment, sex, age categories (<70 years, 70-79 years, and ≥80 years), diabetes, previous stroke, and CHA₂DS₂-VASc scores of ≤2, 3, and ≥4 were performed. Furthermore, the associated effects of antiplatelet treatment were studied in patients treated with and without oral anticoagulation.

To compare risks associated with treatment with direct oral anticoagulants (DOACs) and vitamin K antagonists (VKA), subgroup analyses were performed in patients with eGFR ≥ 15 mL/min/1.73 m² only. DOACs have not been approved by the European Medicines Agency for patients with eGFR < 15 mL/min/1.73 m² [[35], [36], [37], [38]] and are, therefore, not recommended in this patient group in the 2020 European Society of Cardiology guideline [24]. All analyses were performed as intention-to-treat.

To evaluate the quality of VKA treatment on outcomes, analyses were performed in event-free 1-year survivors comparing risks of outcomes between patients without oral anticoagulation, patients on DOAC treatment, and patients on VKA treatment dichotomized by time in therapeutic range (TTR) (ie, <70% and ≥70% as defined by an international normalized ratio [INR] between 2.0 and 3.0). VKA-treated patients were only included, providing ≥1 INR measurement with an INR value >1.3 within the period 6 to 12 months following index. Treatment was defined by the latest redeemed prescription 6 months preceding index.

Several sensitivity analyses were performed. First, to address risk of misclassification of eGFR due to acute kidney injury related to onset of atrial fibrillation, a redefined cohort requiring 2 distinct eGFR measurements <30 mL/min/1.73 m² separated by >90 days prior to diagnosis of atrial fibrillation was identified. Second, to address risk of misclassification of treatment due to delayed redemption of oral anticoagulation prescription, a redefined cohort with postponement of inclusion until discharge +60 days was identified to permit greater capture of redeemed prescriptions for anticoagulants. Third, to address impact of change of baseline treatment, ie, initiation or discontinuation of oral anticoagulation, 2 sensitivity analyses were performed with censoring of follow-up time after 365 and 730 days, respectively.

To evaluate the impact of treatment on mortality following thromboembolic and bleeding events, standardized 90-day risk of death was computed based on Cox regression models adjusted for age and sex in patients with events.

A 2-sided P value < .05 was considered to indicate statistical significance. All analyses were performed with SAS version 9.4 (SAS Institute) and R version 4.0.3 (R Core Team).

2.7. Ethics approval and reporting guidelines

Retrospective studies do not require informed consent or prior approval from ethics committees in Denmark. The use of study data was approved through the Danish Data Protection Agency (reference P-2019-191). For preparation of this manuscript, the Strengthening the Reporting of Observational Studies in Epidemiology cohort reporting guidelines were used [39].

3. Results

A total of 129,528 patients with incident atrial fibrillation and a preceding plasma creatinine measurement were identified between 2010 and 2022, of which 3208 patients had advanced CKD, as defined by an eGFR < 30 mL/min/1.73 m². An overview of the study inclusion is provided in Figure 1. Overall, median age was 80 years (IQR, 72-87), 52.8% were male, and 20.9% received chronic dialysis therapy. Oral anticoagulation was initiated in 1375 patients (42.9%), with VKA accounting for 48.1% of treated patients. Baseline data stratified on oral anticoagulation treatment are presented in the Table. Initiation of oral anticoagulation was associated with increased age, female sex, less advanced CKD, lower frequency of both previous bleeding and ischemic stroke and higher CHA₂DS₂-VASc score. The only missing data in the study were descriptive data on baseline hemoglobin level (n = 452) and platelet count (n = 402). Patients who missed these values were included in all analyses. Baseline characteristics stratified on strata of eGFR are provided in Supplementary Table S2. Among patients treated with dialysis, median age was 71 years (IQR, 63-78), 65.7% were male, 19.1% initiated treatment with oral anticoagulation, and frequency of previous major bleeding was higher than in patients with nondialysis CKD, whereas frequency of previous ischemic stroke was similar.

Study population flowchart. eGFR, estimated glomerular filtration rate.

Table.

Baseline data on patients without and patients with oral anticoagulation.

Characteristic	No anticoagulation (n = 1833), n (%)^d	Anticoagulation (n = 1375), n (%)^d	P value
Age, median (IQR), y	79 (70-87)	81 (74-88)	<.001
<70	456 (24.9)	207 (15.1)	<.001
70-79	544 (29.7)	416 (30.3)
≥80	833 (45.4)	752 (54.7)
Male sex	1007 (54.9)	687 (50.0)	.006
eGFR, median (IQR), mL/min/1.73 m²	18 (9-25)	23 (17-27)	<.001
Chronic kidney disease
eGFR 15-29 mL/min/1.73 m²	1013 (55.3)	1085 (78.9)	<.001
eGFR < 15 mL/min/1.73 m²	277 (15.1)	162 (11.8)
Dialysis	543 (29.6)	128 (9.3)
Hemoglobin level^a, mean (SD), mmol/L	6.9 (1.1)	7.3 (1.1)	<.001
Platelet count^b, median (IQR), x 10⁹/L	245 (193-212)	240 (193-301)	.45
CHA₂DS₂-VASc score^c
≤2	466 (25.4)	272 (19.8)	<.001
3	457 (24.9)	336 (24.4)
≥4	910 (49.6)	767 (55.8)
Median score (IQR)	3 (2-5)	4 (3-5)	.001
Previous ischemic stroke	128 (7.0)	71 (5.2)	.04
Previous major bleeding	327 (17.8)	177 (12.9)	<.001
Hypertension	976 (53.2)	795 (57.8)	.01
Diabetes	576 (31.4)	463 (33.7)	.19
Insulin-treatment	337 (18.4)	250 (18.2)	.92
Antiplatelet agents
Acetylsalicylic acid	665 (36.3)	535 (38.9)	.14
Other	326 (17.8)	269 (19.6)	.22
Lipid-lowering treatment	757 (41.3)	670 (48.7)	<.001

Open in a new tab

P values are calculated according to normal distribution with the Student’s t-test or Mann–Whitney U-test for continuous data and chi-squared test for categorical data.

eGFR, estimated glomerular filtration rate.

Missing observations: n = 452.

Missing observations: n = 402.

CHA₂DS₂-VASc score calculated as congestive heart failure (1 point), hypertension (1 point), age ≥75 years (2 points), diabetes (1 point), previous stroke/transient ischemic attack/thromboembolism (2 points), vascular disease (1 point), age 65-74 years (1 point), and female sex (1 point).

Unless otherwise specified.

3.1. One-year risks of thromboembolic event, major bleeding, and death

Median follow-up time was 5.7 years (IQR, 3.5-7.6). During follow-up, 262 (8.2%) and 510 (15.9%) patients were hospitalized due to thromboembolic and bleeding events, respectively. Crude 1-year mortality was 33.5%. Overall, oral anticoagulation compared with no oral anticoagulation was associated with a decreased rate of a thromboembolic event (hazard ratio [HR], 0.75; 95% CI, 0.58-0.97; P = .03), an increased rate of major bleeding (HR, 1.41; 95% CI, 1.18-1.68; P < .001), and a decreased rate of death (HR, 0.77; 95% CI, 0.71-0.84; P < .001) in models adjusted for age and sex. Corresponding 1-year risks of thromboembolic event, major bleeding, and death in patients without and with oral anticoagulation were 4.8% (95% CI, 3.8%-5.7%) and 3.6% (95% CI, 2.8%-4.6%), 7.6% (95% CI, 6.6%-8.7%) and 10.5% (95% CI, 9.3%-12.1%), and 36.3% (95% CI, 34.2%-38.3%) and 29.6% (95% CI, 27.6%-31.6%), respectively. Standardized 1-year risks of thromboembolic event, major bleeding, and death are shown in Figure 2A–C. Comparable standardized 1-year risks were estimated in models adjusted for parameters of the CHA₂DS₂-VASc score, previous major bleeding, and antiplatelet treatment (Supplementary Table S3). Similarly, hazards from inverse-probability-weighted models remained unchanged, both in models with propensities based on age and sex alone and in models with propensities based on CHA₂DS₂-VASc score, previous major bleeding, and antiplatelet treatment (Supplementary Table S4). An overview of outcomes stratified by treatment strategy is provided in Figure 3, displaying that both overall and in the separate groups, death was the most common outcome, succeeded by no event, major bleeding, and thromboembolic event.

Oral anticoagulation treatment strategy and associated age and sex-adjusted standardized risk of (A) thromboembolic event, (B) major bleeding, and (C) death during the first year after index date. Standardized risks were obtained from G-computation standardized to the distribution of risk factors in the sample based on Cox regression models adjusted for age and sex.

Alluvial diagram visualizing outcomes during first year after index in patients treated with or without oral anticoagulation.

One-year standardized risk of thromboembolic event was substantially increased in patients with prior stroke. Oral anticoagulation was, however, associated with nonsignificant and comparable hazards in both patients with and without prior stroke. Across strata of CHA₂DS₂-VASc score (≤2, 3, and ≥4), oral anticoagulation was associated with a reduced 1-year standardized risk of mortality, an increased risk of bleeding, but only a reduced risk of thromboembolic event in patients with a score of 3 (P = .03) or ≥4 (P = .08). However nonsignificant, oral anticoagulation in dialysis-treated patients was associated with an increased risk of death, whereas the risk of thromboembolic events and major bleeding followed the pattern of the overall analysis as the patients with predialysis CKD. HRs and standardized 1-year risks from prespecified subgroups are provided in Figure 4A–C. Furthermore, there was no time-period effect in the outcomes when comparing patients included from 2010 to 2016 and 2017 to 2022, respectively (Supplementary Table S5). The percentage of patients treated with oral anticoagulation was 34.7% in the first period and 54.0% in the second period (P < .001). VKA accounted for 67.6% and 34.7% of these prescriptions in the 2 time periods, respectively (P < .001).

Hazard ratios and 1-year standardized risks of (A) thromboembolic events, (B) major bleeding, and (C) death in prespecified subgroups, including P for interaction. (A–C) Standardized risks were obtained from G-computation standardized to the distribution of risk factors in the sample based on Cox regression models adjusted for age and sex. CHA₂DS₂-VASc score calculated as: congestive heart failure (1 point), hypertension (1 point), age ≥75 years (2 points), diabetes (1 point), previous stroke/transient ischemic attack/thromboembolism (2 points), vascular disease (1 point), age 65-74 years (1 point), and female sex (1 point).

3.2. Subgroup analysis: VKA vs DOAC

One-year standardized risks and corresponding HRs for all outcomes in patients with eGFR 15 to 29 mL/min/1.73 m² comparing no treatment, VKA, and DOAC are presented in Supplementary Table S6. For thromboembolic event, the risk differences were -0.6% (95% CI, -2.3% to 0.8%) for VKA vs no treatment, -1.8% (95% CI, -3.2% to -0.5%) for DOAC vs no treatment, and -1.2% (95% CI, -2.8% to 0.4%) for DOAC vs VKA. For major bleeding, the risk differences were 4.3% (95% CI, 1.9%–7.2%) for VKA vs no treatment, 2.8% (95% CI, 0.8%–4.8%) for DOAC vs no treatment, and -1.5% (95% CI, -4.5% to 1.1%) for DOAC vs VKA. For death, the risk differences were -3.4% (95% CI, -6.8% to 0.5%) for VKA vs no treatment, -8.9% (95% CI, -11.8% to -5.8%) for DOAC vs no treatment, and -5.5% (95% CI, -9.2% to -2.0%) for DOAC vs VKA.

3.3. Subgroup analysis: antiplatelet treatment

Antiplatelet treatment in patients without oral anticoagulation was associated with an increased risk of thromboembolic events and unchanged risk of major bleeding and death (Supplementary Table S7). Risk differences were 3.1% (95% CI, 1.6%-4.9%) for thromboembolic event, 1.0% (95% CI, -0.8% to 2.8%) for major bleeding, and 1.1% (95% CI, -1.5% to 3.7%) for death. In patients treated with oral anticoagulation, antiplatelet treatment was not associated with difference in risk of all outcomes. Risk differences were 1.2% (95% CI, -0.1% to 2.6%) for thromboembolic event, -0.2% (95% CI, -2.5% to 2.2%) for major bleeding, and 0.0% (95% CI, -3.1% to 2.8%) for death.

3.4. Subgroup analysis: time in therapeutic range

The impact of TTR was evaluated in 1548 (48.3%) event-free patients at 1 year. Baseline characteristics are provided in Supplementary Table S8, showing that patients with sufficient and nonsufficient VKA were overall comparable, while patients on DOAC were older and had a higher CHA₂DS₂-VASc score and a higher eGFR than nontreated and VKA-treated patients. Standardized 1-year risks of all outcomes stratified on treatment and HRs and subgroup analyses are provided in Figure 5A–C and Supplementary Table S9, respectively. The percentage of measurements within therapeutic range is shown in Supplementary Figure S2. Mean TTR was 49.7% during the first 12 months after initiation of treatment and 52.6% when omitting the first 3 months after initiation of treatment.

Standardized age and sex-adjusted risk of (A) a thromboembolic event, (B) major bleeding, and (C) death in event-free 1-year survivors stratified by no anticoagulation, nonsufficient VKA, sufficient VKA, and DOAC. (A–C) Treatment was defined by the latest prescription refill in the 6 to 12 months after index date. VKA-treated patients had to have at least 1 international normalized ratio measurement >1.3 to be included. Sufficient VKA was defined as ≥70% of international normalized ratio measurements between 2.0 and 3.0. DOAC, direct oral anticoagulant; TTR, time in therapeutic range; VKA, vitamin K antagonist.

3.5. Sensitivity analyses addressing misclassification

A total of 2378 (74.1%) patients had 2 distinct eGFR measurements <30 mL/min/1.73 m², separated by ≥90 days prior to diagnosis of atrial fibrillation. Principal results remained unchanged, with exception of the 1-year standardized risk of thromboembolic event, which was not significantly different between patients without and with oral anticoagulation: 5.1% (95% CI, 4.1%-6.1%) and 4,1% (95% CI, 3.1%-5.3%), P = .15 for thromboembolic event; 7.8% (95% CI, 6.4%-9.1%) and 10.3% (95% CI, 8.6%-12.0%), P = .005 for major bleeding; and 35.9% (95% CI, 33.8%-38.4%) and 30.5% (95% CI, 28.2%-32.6%), P < .001 for death (Supplementary Table S10).

Following postponement of index to discharge +60 days, a total of 2728 (85.0%) patients were identified. Principal results remained unchanged in patients without and with oral anticoagulation with 1-year standardized risks of 4.1% (95% CI, 3.3%-5.1%) and 3.0% (95% CI, 2.3%-3.8%), P = .02, for thromboembolic event; 6.8% (95% CI, 5.7%-7.8%) and 8.6% (95% CI, 7.3%-10.0%), P = .009, for major bleeding; and 26.4% (95% CI, 24.4%-28.4%) and 22.3% (95% CI, 20.3%-24.3%), P < .001, for death (Supplementary Table S11).

3.6. Sensitivity analyses addressing treatment change

An overview of patients still adherent to the baseline treatment strategy as defined by one or more prescription redemptions for anticoagulants during 6-month intervals is presented in Supplementary Figure S3. After 24 months, 80% of event-free patients remained nonanticoagulated, and 60% of event-free patients remained anticoagulated. Regarding antiplatelets, 87% of event-free patients remained nontreated, and 97% of event-free patients remained treated after 1 year (defined by a prescription redemption within 180 to 365 days after index).

Censoring of follow-up time after 365 and 730 days confirmed the findings of the main analysis (Supplementary Tables S12 and S13). Censoring after 365 days gave 1-year standardized risks in patients without and with oral anticoagulation of 4.8% (95% CI, 4.0%-5.9%) and 3.3% (95% CI, 2.3%-4.3%), P = .04 for thromboembolic event; 7.4% (95% CI, 6.3%-8.7%) and 10.1% (95% CI, 8.1%-11.7%), P = .008 for major bleeding; and 36.5% (95% CI, 34.5%-38.7%) and 27.5% (95% CI, 25.2%-30.1%), P < .001 for death. While censoring after 730 days resulted in 1-year standardized risks in patients without and with oral anticoagulation of 4.5% (95% CI, 3.8%-5.2%) and 3.8% (95% CI, 3.2%-4.3%), P = .004 for thromboembolic event; 7.7% (95% CI, 6.8%-8.9%) and 9.7% (95% CI, 8.2%-11.2%), P = .03 for major bleeding; and 36.7% (95% CI, 34.6%-39.0%) and 27.1% (95% CI, 24.9%-29.2%), P < .001 for death.

3.7. Ninety-day risk of death after thromboembolic event or bleeding

Ninety-day standardized risks and corresponding HRs of death following a thromboembolic event or major bleeding are provided in Supplementary Tables S14 and S15. Oral anticoagulation was not associated with harm or benefit on risk of 90-day mortality following thromboembolic or bleeding events overall. Case fatality rates were 29.9% vs 31.4% and 30.3% vs 23.2% in treated vs nontreated patients for thromboembolic event and major bleeding, respectively.

4. Discussion

In a comprehensive retrospective cohort study of patients with advanced CKD and incident atrial fibrillation, oral anticoagulation for mitigation of risk of stroke was initiated in 40% of patients. Overall, oral anticoagulation treatment was associated with decreased standardized 1-year risks of thromboembolic event and death and increased standardized 1-year risk of major bleeding compared with no treatment.

Current guidelines governing the use of oral anticoagulation for stroke-risk reduction in patients with advanced CKD remain divided [23,25,40]. No clinical trials investigating the net benefit of oral anticoagulation compared with no treatment in patients with eGFR < 30 mL/min/1.73 m² have been completed. Furthermore, retrospective data remain limited in predialysis patients [41], and interpretation continues to be abstruse, with heterogeneous results providing no conclusive evidence of harm or benefit. As such, existing data have reported oral anticoagulation to be associated with increased [[9], [10], [11], [12], [13]], unchanged [[14], [15], [16]], and decreased [[17], [18], [19]] risk of stroke, and increased [[12], [13], [14], [15], [16], [17],20,21] and unchanged [10,18,19,22] risk of bleeding. Unsurprisingly, oral anticoagulation remains inconsistently used [42,43].

Randomized clinical trials have previously demonstrated noninferiority of DOAC treatment compared with VKA treatment in patients with preserved kidney function [[44], [45], [46], [47]]. Comparisons of DOAC and VKA treatment in ESKD remain limited to 3 small clinical trials: the RENal hemodialysis patients ALlocated apixaban versus warfarin in Atrial Fibrillation (RENAL-AF) trial (n = 154) comparing apixaban 5 mg twice a day with VKA [48]; the Compare Apixaban and Vitamin K Antagonists in Patients With Atrial Fibrillation and End-Stage Kidney Disease (AXADIA-AFNET8) trial (n = 97) comparing apixaban 2.5 mg twice a day with VKA [49]; and the Valkyrie trial (n = 132) comparing rivaroxaban 10 mg every day with VKA [50]. Overall, results remain divergent, with all trials underpowered for definite conclusions. Comparably, our results demonstrated similar benefit-to-harm ratios regarding thromboembolic and bleeding risks associated with DOAC and VKA treatment and marginal benefit on 1-year survival associated with DOACs. As such, our results corroborate existing observations of effects from nondiscriminating populations of patients with eGFR < 30 mL/min/1.73 m², including ESKD [51].

Adequacy of VKA treatment in advanced CKD remains a challenge. TTR in the RENAL-AF, AXADIA-AFNET8, and Valkyrie trials ranged from 44% to 55%, while in the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY), Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE), Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism (ROCKET AF), and Efficacy and Safety of Edoxaban Versus Warfarin in Subjects With Atrial Fibrillation (ENGAGE AF-TIMI48) trials, TTR ranged from 55% to 65% [[44], [45], [46], [47]]. Overall, our results remained underpowered for evaluation of efficacy and safety outcomes in explorative analyses dependent on the quality of VKA treatment. Nonetheless, suboptimal TTR was associated with increased risk of death compared with VKA with optimal TTR and DOAC treatment. Interpretation is, however, uncertain, with plausible interference biases due to the impact of nonrelated hospitalization and illness on delivery of VKA treatment. Furthermore, patients assessed as having a high risk of thromboembolic event might have more frequent evaluation of INR, which can induce ascertainment bias and explain the (nonsignificant) association between optimal TTR and highest risk of thromboembolic events in the subgroups.

4.1. Strengths and limitations

Overall, results are supported by the comprehensive nature of the multiple national healthcare databases in Denmark, permitting access to updated and validated data on numerous aspects of healthcare with minimal loss to follow-up [[26], [27], [28],52]. Nonetheless, although the nationwide scope effectively minimizes selection biases due to geographic and demographic variation, unmeasured residual confounding cannot be excluded. Limitations apply. The decision to initiate or withhold treatment remains unaddressed, with plausible implications on treatment-related outcomes. However, we evaluated parameters likely to affect choice of treatment strategy and prognosis in subgroup analyses and confirmed the main results in robustness analyses, ie, multiple Cox regression models adjusted for parameters of the CHA₂DS₂-VASc score, previous major bleeding, and antiplatelet treatment and inverse-probability-weighted models. Lipid-lowering treatment was not included in the analyses since the association between hyperlipidemia and risk of cardiovascular events is less pronounced in advanced CKD [53]. Another challenge is accurate classification of kidney function in retrospective studies. However, our study employed a validated algorithm [32] for identification of eGFR < 30 mL/min/1.73 m² with principal results unchanged in robustness analyses entailing confirmatory creatinine measurement. Treatment allocation was determined based on redemption of a prescription within a predefined time window permitting accurate classification of 67.0% of prescriptions within the first year. Principal results remained unchanged in sensitivity analyses employing an expanded time window permitting identification of 78.7% of treated patients. Of note, adherence to baseline treatment strategy was 80% in patients without oral anticoagulation and 60% in patients with oral anticoagulation after 24 months. In other studies, >50% of treated patients discontinued the treatment within 1 year [22,54,55].

Finally, the inherently observational and retrospective nature of the study precludes causal inference, and conclusions remain exploratory.

5. Conclusion

Our results demonstrate overall benefit of oral anticoagulation for mitigation of risk of thromboembolic events and death in patients with eGFR < 30 mL/min/1.73 m² despite an associated increase in risk of major bleeding. Benefit was, however, most apparent in nondialysis-treated patients, with a treatment-associated nonsignificant increase in mortality noted in dialysis-treated patients with ESKD.

Acknowledgments

Funding

This work was supported by grants from Helsefonden (21-B-0387), Arvid Nilssons Fond, Augustinusfonden (19-2397), and the Danish Society of Nephrology. The funders had no role in the design and conduct of the study, collection management, analysis and interpretation of data, preparation, review, or approval of the manuscript, and decision to submit the manuscript for publication.

Author contributions

E.L.F.B. and N.C. had full access to all data in the study and take responsibility for the integrity of the data and accuracy of the analysis. E.L.F.B., J.B.O., C.T.-P., and N.C. conceived and designed the study. E.L.F.B., J.B.O., A.-L.K., B.F.-R., G.G., C.T.-P., and N.C. contributed to the acquisition, analysis, or interpretation of data, critically revised the manuscript for important intellectual content, and provided administrative, technical, or material support. E.L.F.B. and N.C. drafted the manuscript and obtained funding. E.L.F.B., C.T.-P., and N.C. performed statistical analysis.

Relationship Disclosure

E.L.F.B., A.-L.K., B.F.-R., and G.G. have no conflicts of interest. J.B.O. received speaker honoraria or consultancy fees from Bayer, Bristol-Myers Squibb, and Pfizer. C.T.-P. received grants from Bayer and Novo Nordisk. N.C. received lecture fees from Bristol-Myers Squibb and AstraZeneca.

Footnotes

Handling Editor: Dr Kristen Sanfilippo

The online version contains supplementary material available at https://doi.org/10.1016/j.rpth.2024.102350.

Supplementary material

Supplementary Table S1

Diagnoses, surgical procedures and pharmacotherapy used for defining the population, comorbidity and outcomes

Supplementary Table S2 Baseline data on patients by stage of chronic kidney disease

Supplementary Table S3 Hazard ratios and corresponding one-year risks of outcomes in Cox regression models adjusted for parameters of the CHADS-VASc score, previous major bleeding, and antiplatelet treatment

Supplementary Table S4 Hazard ratios (95% CI) for all outcomes estimated with propensity score models with inverse probability weighting adjusted for age and sex

Supplementary Table S5 Subgroup analyses of hazard ratio and one-year standardized risk of thromboembolic event, major bleeding and death in patients with and without oral anticoagulation during the first and second half of the study period (2010-2016 and 2017-2022)

Supplementary Table S6 Subgroup analyses of hazard ratio and one-year standardized risk of thromboembolic event, major bleeding and death in patients with eGFR > 15 mL/min/1.73 m²

Supplementary Table S7 Subgroup analyses of patients with or without antiplatelet treatment

Supplementary Table S8 Baseline characteristics on one-year survivors grouped by no oral anticoagulation, non-sufficient VKA treatment, sufficient VKA treatment, and DOAC treatment

Supplementary Table S9 Subgroup analyses of hazard ratio and one-year standardized risk of thromboembolic event, major bleeding and death in one-year survivors stratified on treatment

Supplementary Table S10 Sensitivity analyses including only patients with ≥2 measurements of eGFR < 30 ml/min/1.73 m² >90 days apart prior to diagnosis of atrial fibrillation

Supplementary Table S11 Sensitivity analyses with postponement of index to discharge +60 days.

Supplementary Table S12 Overall analyses with censoring of follow-up at index+365 days

Supplementary Table S13 Overall analyses with censoring of follow-up at index+730 days

Supplementary Table 14 Subgroup analyses of 90-days hazard ratio and standardized risk of death following a thromboembolic event in patients without and with oral anticoagulation.

Supplementary Table S15 Subgroup analyses of 90-days hazard ratio and standardized risk of death following a major bleeding in patients without and with oral anticoagulation.

Supplementary Figure S1 Time to redemption of prescription for anticoagulants

Supplementary Figure S2 Bar chart of INR measurements during first year

Supplementary Figure S3 Bar chart of event free patients still adherent to baseline treatment during first 24 months after index

mmc1.docx^{(351.3KB, docx)}

References

1.Zimmerman D., Sood M.M., Rigatto C., Holden R.M., Hiremath S., Clase C.M. Systematic review and meta-analysis of incidence, prevalence and outcomes of atrial fibrillation in patients on dialysis. Nephrol Dial Transplant. 2012;27:3816–3822. doi: 10.1093/ndt/gfs416. [DOI] [PubMed] [Google Scholar]
2.Liao J.-N., Chao T.-F., Liu C.-J., Wang K.-L., Chen S.-J., Lin Y.-J., et al. Incidence and risk factors for new-onset atrial fibrillation among patients with end-stage renal disease undergoing renal replacement therapy. Kidney Int. 2015;87:1209–1215. doi: 10.1038/ki.2014.393. [DOI] [PubMed] [Google Scholar]
3.Genovesi S., Pogliani D., Faini A., Valsecchi M.G., Riva A., Stefani F., et al. Prevalence of atrial fibrillation and associated factors in a population of long-term hemodialysis patients. Am J Kidney Dis. 2005;46:897–902. doi: 10.1053/j.ajkd.2005.07.044. [DOI] [PubMed] [Google Scholar]
4.Goldstein B.A., Arce C.M., Hlatky M.A., Turakhia M., Setoguchi S., Winkelmayer W.C. Trends in the incidence of atrial fibrillation in older patients initiating dialysis in the United States. Circulation. 2012;126:2293–2301. doi: 10.1161/CIRCULATIONAHA.112.099606. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Hart R.G., Benavente O., McBride R., Pearce L.A. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med. 1999;131:492–501. doi: 10.7326/0003-4819-131-7-199910050-00003. [DOI] [PubMed] [Google Scholar]
6.Charytan D., Kuntz R.E. The exclusion of patients with chronic kidney disease from clinical trials in coronary artery disease. Kidney Int. 2006;70:2021–2030. doi: 10.1038/sj.ki.5001934. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Saran R., Robinson B., Abbott K.C., Bragg-Gresham J., Chen X., Gipson D., et al. US renal data system 2019 annual data report: epidemiology of kidney disease in the United States. Am J Kidney Dis. 2020;75:A6–A7. doi: 10.1053/j.ajkd.2019.09.003. [DOI] [PubMed] [Google Scholar]
8.Sohal A.S., Gangji A.S., Crowther M.A., Treleaven D. Uremic bleeding: pathophysiology and clinical risk factors. Thromb Res. 2006;118:417–422. doi: 10.1016/j.thromres.2005.03.032. [DOI] [PubMed] [Google Scholar]
9.Chan K.E., Lazarus J.M., Thadhani R., Hakim R.M. Warfarin use associates with increased risk for stroke in hemodialysis patients with atrial fibrillation. J Am Soc Nephrol. 2009;20:2223–2233. doi: 10.1681/ASN.2009030319. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Wizemann V., Tong L., Satayathum S., Disney A., Akiba T., Fissell R.B., et al. Atrial fibrillation in hemodialysis patients: clinical features and associations with anticoagulant therapy. Kidney Int. 2010;77:1098–1106. doi: 10.1038/ki.2009.477. [DOI] [PubMed] [Google Scholar]
11.Wiesholzer M., Harm F., Tomasec G., Barbieri G., Putz D., Balcke P. Incidence of stroke among chronic hemodialysis patients with nonrheumatic atrial fibrillation. Am J Nephrol. 2001;21:35–39. doi: 10.1159/000046216. [DOI] [PubMed] [Google Scholar]
12.Kumar S., de Lusignan S., McGovern A., Correa A., Hriskova M., Gatenby P., et al. Ischaemic stroke, haemorrhage, and mortality in older patients with chronic kidney disease newly started on anticoagulation for atrial fibrillation: a population based study from UK primary care. BMJ. 2018;360:k342. doi: 10.1136/bmj.k342. [DOI] [PubMed] [Google Scholar]
13.Sy J., Wenziger C., Marroquin M., Kalantar-Zadeh K., Kovesdy C., Streja E. Warfarin use, stroke, and bleeding risk among pre-existing atrial fibrillation US veterans transitioning to dialysis. Nephron. 2022;146:360–368. doi: 10.1159/000521494. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Winkelmayer W.C., Liu J., Setoguchi S., Choudhry N.K. Effectiveness and safety of warfarin initiation in older hemodialysis patients with incident atrial fibrillation. Clin J Am Soc Nephrol. 2011;6:2662–2668. doi: 10.2215/CJN.04550511. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Keskar V., McArthur E., Wald R., Harel Z., Zimmerman D., Molnar A.O., et al. The association of anticoagulation, ischemic stroke, and hemorrhage in elderly adults with chronic kidney disease and atrial fibrillation. Kidney Int. 2017;91:928–936. doi: 10.1016/j.kint.2016.10.017. [DOI] [PubMed] [Google Scholar]
16.Pokorney S.D., Black-Maier E., Hellkamp A.S., Friedman D.J., Vemulapalli S., Granger C.B., et al. Oral anticoagulation and cardiovascular outcomes in patients with atrial fibrillation and end-stage renal disease. J Am Coll Cardiol. 2020;75:1299–1308. doi: 10.1016/j.jacc.2020.01.019. [DOI] [PubMed] [Google Scholar]
17.Olesen J.B., Lip G.Y.H., Kamper A.L., Hommel K., Køber L., Lane D.A., et al. Stroke and bleeding in atrial fibrillation with chronic kidney disease. N Engl J Med. 2012;367:625–635. doi: 10.1056/NEJMoa1105594. [DOI] [PubMed] [Google Scholar]
18.Lai H.M., Aronow W.S., Kalen P., Adapa S., Patel K., Goel A., et al. Incidence of thromboembolic stroke and of major bleeding in patients with atrial fibrillation and chronic kidney disease treated with and without warfarin. Int J Nephrol Renovasc Dis. 2009;2:33–37. doi: 10.2147/ijnrd.s7781. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Jun M., James M.T., Ma Z., Zhang J., Tonelli M., McAlister F.A., et al. Warfarin initiation, atrial fibrillation, and kidney function: comparative effectiveness and safety of warfarin in older adults with newly diagnosed atrial fibrillation. Am J Kidney Dis. 2017;69:734–743. doi: 10.1053/j.ajkd.2016.10.018. [DOI] [PubMed] [Google Scholar]
20.Genovesi S., Rossi E., Gallieni M., Stella A., Badiali F., Conte F., et al. Warfarin use, mortality, bleeding and stroke in haemodialysis patients with atrial fibrillation. Nephrol Dial Transplant. 2015;30:491–498. doi: 10.1093/ndt/gfu334. [DOI] [PubMed] [Google Scholar]
21.Agarwal M.A., Potukuchi P.K., Sumida K., Naseer A., Molnar M.Z., George L.K., et al. Clinical outcomes of warfarin initiation in advanced chronic kidney disease patients with incident atrial fibrillation. JACC Clin Electrophysiol. 2020;6:1658–1668. doi: 10.1016/j.jacep.2020.06.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Shen J.I., Montez-Rath M.E., Lenihan C.R., Turakhia M.P., Chang T.I., Winkelmayer W.C. Outcomes after warfarin initiation in a cohort of hemodialysis patients with newly diagnosed atrial fibrillation. Am J Kidney Dis. 2015;66:677–688. doi: 10.1053/j.ajkd.2015.05.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.January C.T., Wann L.S., Calkins H., Chen L.Y., Cigarroa J.E., Cleveland J.C., et al. 2019 AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS Guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. Circulation. 2019;140:e125–e151. doi: 10.1161/CIR.0000000000000665. [DOI] [PubMed] [Google Scholar]
24.Hindricks G., Potpara T., Dagres N., Arbelo E., Bax J.J., Blomström-Lundqvist C., et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS) Eur Heart J. 2021;42:373–498. doi: 10.1093/eurheartj/ehaa612. [DOI] [PubMed] [Google Scholar]
25.Wanner C., Herzog C.A., Turakhia M.P., Blankestijn P.J., Carrero J.J., Clase C.M., et al. Chronic kidney disease and arrhythmias: highlights from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2018;94:231–234. doi: 10.1016/j.kint.2018.05.005. [DOI] [PubMed] [Google Scholar]
26.Schmidt M., Schmidt S.A.J., Sandegaard J.L., Ehrenstein V., Pedersen L., Sørensen H.T. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol. 2015;7:449–490. doi: 10.2147/CLEP.S91125. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Wallach Kildemoes H., Toft Sørensen H., Hallas J. The Danish National Prescription Registry. Scand J Public Health. 2011;39:38–41. doi: 10.1177/1403494810394717. [DOI] [PubMed] [Google Scholar]
28.Arendt J.F.H., Hansen A.T., Ladefoged S.A., Sørensen H.T., Pedersen L., Adelborg K. Existing data sources in clinical epidemiology: laboratory information system databases in Denmark. Clin Epidemiol. 2020;12:469–475. doi: 10.2147/CLEP.S245060. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Pedersen C.B. The Danish Civil Registration System. Scand J Public Health. 2011;39:22–25. doi: 10.1177/1403494810387965. [DOI] [PubMed] [Google Scholar]
30.Rix T.A., Riahi S., Overvad K., Lundbye-Christensen S., Schmidt E.B., Joensen A.M. Validity of the diagnoses atrial fibrillation and atrial flutter in a Danish patient registry. Scand Cardiovasc J. 2012;46:149–153. doi: 10.3109/14017431.2012.673728. [DOI] [PubMed] [Google Scholar]
31.Levey A.S., Stevens L.A., Schmid C.H., Zhang Y.L., Castro A.F., Feldman H.I., et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:604–612. doi: 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Siew E.D., Ikizler T.A., Matheny M.E., Shi Y., Schildcrout J.S., Danciu I., et al. Estimating baseline kidney function in hospitalized patients with impaired kidney function. Clin J Am Soc Nephrol. 2012;7:712–719. doi: 10.2215/CJN.10821011. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Ekholm O., Hesse U., Davidsen M., Kjøller M. The study design and characteristics of the Danish national health interview surveys. Scand J Public Health. 2009;37:758–765. doi: 10.1177/1403494809341095. [DOI] [PubMed] [Google Scholar]
34.Robins J.M. Marginal structural models. Proc Am Stat Assoc Sect Bayesian Stat Sci. 1998:1–10. [Google Scholar]
35.European Medicines Agency Eliquis product information. 2021. https://www.ema.europa.eu/en/documents/product-information/eliquis-epar-product-information_en.pdf./
36.European Medicines Agency Xarelto product information. 2018. https://www.ema.europa.eu/en/documents/product-information/xarelto-epar-product-information_en.pdf./
37.European Medicines Agency Pradaxa product information. 2018. https://www.ema.europa.eu/en/documents/product-information/pradaxa-epar-product-information_en.pdf./
38.European Medicines Agency Lixiana product information. 2020. https://www.ema.europa.eu/en/documents/product-information/lixiana-epar-product-information_en.pdf./
39.von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P., et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61:344–349. doi: 10.1016/j.jclinepi.2007.11.008. [DOI] [PubMed] [Google Scholar]
40.Kirchhof P., Benussi S., Kotecha D., Ahlsson A., Atar D., Casadei B., et al. 2016 ESC Guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Eur Heart J. 2016;37:2893–2962. doi: 10.1093/eurheartj/ehw210. [DOI] [PubMed] [Google Scholar]
41.Parker K., Hartemink J., Saha A., Mitra R., Lewis P., Power A., et al. A systematic review of the efficacy and safety of anticoagulants in advanced chronic kidney disease. J Nephrol. 2022;35:2015–2033. doi: 10.1007/s40620-022-01413-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Randhawa M.S., Vishwanath R., Rai M.P., Wang L., Randhawa A.K., Abela G., et al. Association between use of warfarin for atrial fibrillation and outcomes among patients with end-stage renal disease: a systematic review and meta-analysis. JAMA Netw Open. 2020;3 doi: 10.1001/jamanetworkopen.2020.2175. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Dahal K., Kunwar S., Rijal J., Schulman P., Lee J. Stroke, major bleeding, and mortality outcomes in warfarin users with atrial fibrillation and chronic kidney disease: a meta-analysis of observational studies. Chest. 2016;149:951–959. doi: 10.1378/chest.15-1719. [DOI] [PubMed] [Google Scholar]
44.Granger C.B., Alexander J.H., McMurray J.J. V., Lopes R.D., Hylek E.M., Hanna M., et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981–992. doi: 10.1056/NEJMoa1107039. [DOI] [PubMed] [Google Scholar]
45.Patel M.R., Mahaffey K.W., Garg J., Pan G., Singer D.E., Hacke W., et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–891. doi: 10.1056/NEJMoa1009638. [DOI] [PubMed] [Google Scholar]
46.Giugliano R.P., Ruff C.T., Braunwald E., Murphy S.A., Wiviott S.D., Halperin J.L., et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369:2093–2104. doi: 10.1056/NEJMoa1310907. [DOI] [PubMed] [Google Scholar]
47.Connolly S.J., Ezekowitz M.D., Yusuf S., Eikelboom J., Oldgren J., Parekh A., et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361:1139–1151. doi: 10.1056/NEJMoa0905561. [DOI] [PubMed] [Google Scholar]
48.Pokorney S.D., Chertow G.M., Al-Khalidi H.R., Gallup D., Dignaco P., Mussina K., et al. Apixaban for patients with atrial fibrillation on hemodialysis: a multicenter randomized controlled trial. Circulation. 2022;146:1735–1745. doi: 10.1161/CIRCULATIONAHA.121.054990. [DOI] [PubMed] [Google Scholar]
49.Reinecke H., Engelbertz C., Bauersachs R., Breithardt G., Echterhoff H.-H., Gerβ J., et al. A randomized controlled trial comparing apixaban to the vitamin K-antagonist phenprocoumon in patients on chronic hemodialysis: the AXADIA-AFNET 8 study. Circulation. 2023;147:296–309. doi: 10.1161/CIRCULATIONAHA.122.062779. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.De Vriese A.S., Caluwé R., Van Der Meersch H., De Boeck K., De Bacquer D. Safety and efficacy of vitamin K antagonists versus rivaroxaban in hemodialysis patients with atrial fibrillation: a multicenter randomized controlled trial. J Am Soc Nephrol. 2021;32:1474–1483. doi: 10.1681/ASN.2020111566. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Chen C., Cao Y., Zheng Y., Dong Y., Ma J., Zhu W., et al. Effect of rivaroxaban or apixaban in atrial fibrillation patients with stage 4-5 chronic kidney disease or on dialysis. Cardiovasc Drugs Ther. 2021;35:273–281. doi: 10.1007/s10557-021-07144-8. [DOI] [PubMed] [Google Scholar]
52.Dalgaard F., Pieper K., Verheugt F., Camm A.J., Fox K.A., Kakkar A.K., et al. GARFIELD-AF model for prediction of stroke and major bleeding in atrial fibrillation: a Danish nationwide validation study. BMJ Open. 2019;9 doi: 10.1136/bmjopen-2019-033283. [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Baigent C., Landray M.J., Reith C., Emberson J., Wheeler D.C., Tomson C., et al. The effects of lowering LDL cholesterol with simvastatin plus ezetimibe in patients with chronic kidney disease (Study of Heart and Renal Protection): a randomised placebo-controlled trial. Lancet. 2011;377:2181–2192. doi: 10.1016/S0140-6736(11)60739-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Siontis K.C., Zhang X., Eckard A., Bhave N., Schaubel D.E., He K., et al. Outcomes associated with apixaban use in patients with end-stage kidney disease and atrial fibrillation in the United States. Circulation. 2018;138:1519–1529. doi: 10.1161/CIRCULATIONAHA.118.035418. [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Tan J., Bae S., Segal J.B., Zhu J., Alexander G.C., Segev D.L., et al. Warfarin use and the risk of stroke, bleeding, and mortality in older adults on dialysis with incident atrial fibrillation. Nephrology (Carlton) 2019;24:234–244. doi: 10.1111/nep.13207. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials