Artículo del NEJM

1. The n e w e ng l a n d j o u r na l of m e dic i n e
original article
Stroke and Bleeding in Atrial Fibrillation
with Chronic Kidney Disease
Jonas Bjerring Olesen, M.D., Gregory Y.H. Lip, M.D.,
Anne-Lise Kamper, M.D., D.M.Sc., Kristine Hommel, M.D.,
Lars Køber, M.D., D.M.Sc., Deirdre A. Lane, Ph.D.,
Jesper Lindhardsen, M.D., Gunnar Hilmar Gislason, M.D., Ph.D.,
and Christian Torp-Pedersen, M.D., D.M.Sc.
A BS T R AC T
BACKGROUND
Both atrial fibrillation and chronic kidney disease increase the risk of stroke and From the Department o
Dr. of antithrombotic penhagen (J.B.O., J.L.,
systemic thromboembolism. However, these risks, and the effects Abraham Villalobos F RMi
Hellerup
University H
treatment, have not been thoroughly investigated in patients with both conditions. and the Department
(A.-L.K., K.H.) and the
METHODS penhagen University Ho

2. Introducción
• La prevalencia de fibrilación auricular (FA) en personas ≥40
años es de 2.3% y 5.9% en ≥65 años.
• La prevalencia de enfermedad renal crónica (ERC) aumenta
en 3.5% en personas entre los 45 y 64 años de edad hasta
6% en aquellos ≥75 años.
• El riesgo de EVC:
• FA: 5 veces más.
• ERC: 3.7 veces más
• Etapa V: 5.8 veces más.

3. Introducción
• La warfarina aumenta el riesgo de EVC isquémico en
pacientes en hemodiálisis.
• El riesgo de sangrado está aumentado en pacientes con
FA y ERC en tratamiento con warfarina.
• La mayoría de los estudios de terapia antitrobótica
excluyen pacientes con ERC.

4. Atrial fibrillation in hemodialysis patients: clinical
features and associations with anticoagulant therapy
Volker Wizemann1, Lin Tong2, Sudtida Satayathum2, Alex Disney3, Takashi Akiba4, Rachel B. Fissell5,
Peter G. Kerr6, Eric W. Young7,8 and Bruce M. Robinson2,7
1
Georg Haas Dialysezentrum, Giessen, Germany; 2Arbor Research Collaborative for Health, Ann Arbor, Michigan, USA; 3Renal Unit,
Queen Elizabeth Hospital, Woodville, South Australia; 4Tokyo Women’s Medical University, Tokyo, Japan; 5Cleveland Clinic, Cleveland,
Ohio, USA; 6Monash Medical Centre, Melbourne, Victoria, Australia; 7University of Michigan, Ann Arbor, Michigan, USA and 8Veterans
Administration Medical Center, Ann Arbor, Michigan, USA
original article http://www.kidney-international.org
Using data from the international Dialysis Outcomes and Atrial fibrillation (AF) is the most commonInternational Society of Nephrology
& 2010 cardiac
Practice Patterns Study (DOPPS), we determined incidence, dysrhythmia requiring clinical attention and is associated
prevalence, and outcomes among hemodialysis patients with with age and cardiovascular morbidity.1 In the general
atrial fibrillation. Cox proportional hazards models, to identify population, AF is a potent risk factor for stroke (cerebrovas-
•
associations with newly diagnosed atrial fibrillation and cular events)2 and death.3 AF is often associated with
DOPPS: Dialysis Outcomes and Practice Patterns Study.
clinical outcomes, were stratified by country and study phase impaired cardiac performance and lower quality of life.4
Atrial fibrillation in hemodialysis patients: clinical
and adjusted for descriptive characteristics and Given the aging populations and associated rising burden of
comorbidities. Of 17,513 randomly sampled patients, 2188 cardiovascular disease in industrialized countries, the pre-
features and associations with anticoagulant therapy
•
had preexisting atrial fibrillation, with wide variation in valence of AF is expected to increase by 2.5-fold over the next
17,513 pacientes escogidos al azar:
prevalence across countries. Advanced age, non-black race,
higher facility mean dialysate calcium, prosthetic heart
five decades.5
Decreased glomerular filtration rate and chronic kidney
Volker Wizemann1, Lin Tong2, Sudtida Satayathumare Alex Disney3, Takashi Akiba4, Rachel B. Fissell5,
valves, and valvular heart disease were associated with disease
6
2
, independently associated with cardiovascular
Peter G. Kerr positivelyW. Young all-cause mortality M. Robinson2,7 group; the incidence of ESRD has
6 7,8
higher risk of new atrial fibrillation. Atrial fibrillation at study events. Older patients with end-stage renal disease (ESRD)
•
, Eric associated with and Bruce are a rapidly growing
2188 con FA preexistente.
enrollment was
1 and stroke. The CHADS2 score identified approximately tripled in the last decade.7 Cardiovascular prognosis is even
Georg Haas Dialysezentrum, Giessen, Germany; 2Arbor Research Collaborative for Health, Ann Arbor, Michigan, USA; 3Renal Unit,
equal-size groups of hemodialysis patients with atrial worse in patients with coexisting cardiovascular disease and
4 5
Queen Elizabeth low (less than 2) and higher risk (more thanTokyo Women’s Medical University,therapy.8 In addition to
fibrillation with Hospital, Woodville, South Australia; ESRD requiring hemodialysis (HD) Tokyo, Japan; Cleveland Clinic, Cleveland,
•
6 7 8
Ohio,for subsequent strokes on a Centre, Melbourne, basis.
4) USA; Monash Medical per 100 patient-year Victoria, Australia; University of Michigan, with AF, the Michigan, USA and Veterans
these patient-based factors associated Ann Arbor, specific
El uso de warfarina demostró un aumento en el riesgo para EVC, particularmente en mayores
Administration Medical Center, Ann Arbor, Michigan, USA pattern of intermittent HD may raise the risk of HD patients
Among patients with atrial fibrillation, warfarin use was
associated with a significantly higher stroke risk, particularly developing AF. Differences in the incidence of AF between
de 75 años. in those over 75 years of age. Our study shows that atrial
fibrillation is common and associated with elevated risk of
HD and peritoneal dialysis patients9 and the occurrence of
supraventricular arrhythmias in the last hours of HD10 point
Using data from the international is even higher among and an associationfibrillation practice patterns andmost common cardiac
adverse clinical outcomes, and this risk Dialysis Outcomes toward Atrial between HD (AF) is the the
elderly patients prescribed warfarin. The effectiveness and development of AF in HD patients.
•
original article V dysrhythmia requiring clinical attention and is associated
Practice Patterns Study (DOPPS), we determined incidence, Wizemann et al.: Atrial fibrillation in DOPPS hemodialysis patients
safety of warfarin in hemodialysis patients require additional In the general population, clinical trial data support the
A mayor duración de la ERC mayor models, to identify y prevalencia de stroke risk stroke (cerebrovas-
incidencia population, AF is a potent risk FA. for
prevalence, and outcomes among hemodialysis patients with
investigation.
atrial fibrillation. Cox proportional hazards
with stroke prevention in many patients with 1
use of warfarin for age and cardiovascular morbidity. In the general
non-rheumatic AF; risk stratification according to factor
Kidney International (2010) 77, 1098–1106; doi:10.1038/ki.2009.477;
cular events)2 11–13
associations with newly diagnosed atrial fibrillation and example, by the CHADS2and death.guides recom-
published online 6 January 2010 (for score) now 3 AF is often associated with
clinical outcomes, patients withwarfarin
KEYWORDS: atrial fibrillation; DOPPS; outcomes;
country and study mendations impaired cardiac performance use in
Table 4 | Stroke rates among HDwere stratified byhistory of atrial fibrillation, by CHADS2 scorea are few,andany,
phase for warfarin use. Though warfarin lower quality of life.4
dialysis patients with AF is common, there if
and adjusted for descriptive characteristics and data supporting its efficacy and there are and associated rising burden of
Given the aging populations widespread
(A) Patients with history of non-valvular concerns cardiovascular disease All patients with history
(B) in of clinical trial
comorbidities. Of 17,513 randomly sampled patients, 2188 about its safety.14–16 In the absence industrialized countries, the pre-
AF—prescribed aspirin but notwide variation in in dialysis patients, evaluation ofof to increase byAFb
had preexisting atrial fibrillation, with warfarinb data non-valvular
valence of AF is expectedtools such as the 2.5-fold over the next
5 a useful step toward informed
CHADS2 prevalence across countries. AdvancedStroke rate per CHADS2 score may provide
Patients Stroke age, non-black 100 race, five Patients
decades. Stroke Stroke rate per 100
score c
higher facility mean dialysate calcium, prosthetic heart
(n) events (n) patient-yearsdecisions about Decreased glomerular filtration rate andpatient-yearsd
d warfarin use.
(n) events (n) chronic kidney
Correspondence: Bruce M. Robinson, Arbor Research associated with
Collaborative for
This study reports several findings: the prevalence of AF in
0
valves, and valvular heart disease were
Health, 315 West Huron Street, 0
16 0.0
Suite 360, Ann Arbor, Michigan, MI 48103, an international, representative sample of HD associated with cardiovascular
disease are independently patients; the
141 1 0.5
higher risk bruce.robinson@arborresearch.org Atrial fibrillation atassociations events.6 Older patients with end-stage renal disease (ESRD)
USA. E-mail: of new atrial fibrillation. study of AF with comorbid conditions, laboratory
1 enrollment February 2009; revised 18 Septemberwith accepted mortality and cardiovascular growing group; the incidence of 2.1
Received 25 was positively associated 2009; all-cause
114 1 0.6 measures, are a rapidly medications; and the relative
568 18 ESRD has
2 13 stroke. The CHADS2 score identified approximately
174
and October 2009; published online 6 January 2010
6 2.3 risks of adverse outcomes including mortality and strokes
tripled 875the last decade. 23
in 7
Cardiovascular prognosis is even
1.9
equal-size groups of hemodialysis patients with atrial worse in patients with coexisting1098–1106 cardiovascular disease and
3 167
1098 8 3.8 882 Kidney International45
(2010) 77, 3.9
4
fibrillation with low (less than 2) and higher risk (more than
92 6 5.6
ESRD requiring hemodialysis (HD) therapy.8 In addition to
403 27 6.0
5 4) 70 subsequent strokes on a per 100 patient-year basis.
for 6 6.1 these patient-based factors associated with AF, the specific
311 25 6.5
6 Among patients with atrial fibrillation, warfarin use was
10 2 19.0 pattern of intermittent HD 9
70 may raise the risk of HD patients
12.7
Overall associated with a significantly higher stroke risk, particularly
643 29 3.3 developing AF. Differences in the incidence of AF between
3250 148 3.4
a in those over among patients withOur study shows that atrialat DOPPS enrollment, excluding 177 patients with9mechanical occurrence of
Stroke rates during DOPPS follow-up
75 years of age. history of atrial fibrillation (AF) HD and peritoneal dialysis patients and the heart valves.
b fibrillation generally comparable with those used to evaluate CHADS2 supraventricular arrhythmias on heparin hours of HD10 point
Column A includes dialysis patientsis common and associated with elevated risk of in the general population; 27 patients in the last between dialysis sessions

5. REVIEW Table 4 Risk stratification for warfarin use in stroke
Europace (2010) 12, 1666–1672
doi:10.1093/europace/euq387 Warfarin and vascular calcification prevention in dialysis patients with atrial fibrillation
Warfarin has been linked to ectopic calcification, which may Risk stratification Description
adversely affect vascular health. Warfarin, a vitamin K antagonist, ................................................................................
Warfarin in haemodialysis patients with atrial prevents the hepatic formation of clotting factors. However, Favours warfarin Known atrial thrombus
Prosthetic heart valve
fibrillation: what benefit? vitamin K-dependent proteins occur in a number of extrahepatic
tissues including arterial walls and bone. There is increasing evi- CHADS2 score greater than or equal to the
OBRI score by two points
dence that subclinical deficiency of vitamin K has an effect on
Felix Yang 1*, Denise Chou 2, Paul Schweitzer 3, and Sam Hanon 3
51 Mitral stenosis
Division of Cardiology/Arrhythmia Offices Forman 2, bone health Montefiore Medical Center, calcification.
1
Department of Medicine, and vascular 111 E. 210th Street, Bronx, NY 10467, USA; Department of
Warfarin, through its 2
3
Neurology, New York Presbyterian Hospital—Cornell, New York, NY, USA; and Division of Cardiology, Department of Medicine, Beth Israel Medical Center, New York, NY, USA Previous TIA or stroke
ability to interfere with vitamin K remodelling, is therefore a
Downloaded from http://europace.oxfordjournals.org/ at HINARI Cuba Administrative Account on August 20, 2012
Received 24 July 2010; accepted after revision 19 September 2010; online publish-ahead-of-print 2 November 2010 Patient preference
model of peripheral vitamin K deficiency.18
Favours no warfarina Age ,65 years with no risk factors
In the hyperphosphataemic environment such as renal failure,
•
Warfarin is commonly used to prevent stroke in patients with atrial fibrillation; however, patients on haemodialysis may not derive the same Uncontrolled hypertension
vascular smooth muscle cells (VSMCs) have the capacity to trans-
La warfarina aumenta el riesgo de sangrado.
benefit from warfarin as the general population. There are no randomized controlled studies in dialysis patients which demonstrate the effi-
form into osteoblast-like increased risk for haemorrhagic stroke ectopic ischae- 52
cacy of warfarin in preventing stroke. In fact, warfarin places the dialysis patient at cells capable of producing and possibly bone.
Concurrent antiplatelet use
mic stroke. Additionally, warfarin increases the risk of major bleeding and has been associated with vascular calcification. Routine use of History of active calciphylaxis
These VSMCs initiate and regulate vascular calcification. Matrix Gla
•
warfarin in dialysis for stroke prevention should be discouraged, and therapy should only be reserved for dialysis patients at high risk for
Previous life-threatening haemorrhage
Se debe reservar para aquellos pacientes con alto
thrombo-embolic stroke and carefully monitored if implemented. inhibits the calcification process above; however,
protein (MGP)
----------------------------------------------------------------------------------------------------------------------------------------------------------- Severe malnutrition
Atrial fibrillation † the protein Stroke
Haemodialysis † is activated by a process which requires vitamin
Keywords
riesgo de eventos tromboembólicos y el INR debe ser
Warfarin in dialysis patients K.53 – 55 Warfarin therefore may lead to vascular calcification. In
Non-compliance
1669
monitorizado de manera más estricta.
the murine model, MGP-deficient mice develop extensive vascular
Frequent falls
calcification in the aorta and die early from aortic vessel walls.10 Admin-
altered interactions between the platelet and rupture. In par-
Introduction Adapted from Sood et al. 41
Table 3 Studies of among haemodialysis patients with a
istration of vitamin ticular, uraemia causes altered arachidonic acid metabolism which
K antagonists in rodents also induces vascular a
Atrial fibrillation is common warfarin in dialysis patients witha atrial fibrillation decreased thromboxane A2
leads to multitude of abnormalities:
Consider the use of antiplatelet agents in patients not suitable for warfarin.
prevalence of 11 –27% in cross-sectional studies.1 – 3 However, Even production, abnormal intracellular calcium mobilization, and K
calcification. it in non-CKD human patients, vitamin
Study known how the risk
is not (year, design) of stroke in dialysis of dialysis patients with Mean epinephrine, and serotonin production.
Number patients with decreased platelet ADP, Major findings
atrial fibrillation compares to people not on dialysis. The reported Uraemia also impairs binding between IIb –IIIa receptors and the
AF (no. of patients with AF on follow-up
annual rates of stroke vary widely between 1 and 15%.4 – 7 Vazquez von Willebrand factor, leading to impaired platelet aggregation.11
8 warfarin)
. .et. al. . .report. .that. the .presence .of. .atrial. .fibrillation. .increased. the . . . . . . Finally, .uraemia. results.in.increased.endothelial. .production. of. .pros-. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. .. ..... ... ... ........ . ... ........ ........ ... ...... ...... ...... . ........ ......... ......... . ....
risk of stroke in incident dialysis patients 9.8-fold. Despite this see- taglandin I2 and nitric oxide, agents which have vasodilatory and
9
To et al. (2007, stroke in atrial fibrillation, 75% of patients
mingly high rate of 40 (10) 26 months
antiplatelet properties.10 Cerebrovascular events did not differ between patients with AF
1,9
retrospective)
with atrial fibrillation on dialysis are not anticoagulated. Here, Patients with very low GFR or on dialysis are at increased riskAF (5.0%/year vs. 2.4%/year; NS)
from those without
we review the literature for and against the use of warfarin in hae-
46 for haemorrhagic stroke. In a study by Iseki et al. 12 of 1609 patients
Genovesi et al. 127 (31 at enrolment) 36 months No difference in stroke incidence when comparing an
modialysis patients with atrial fibrillation. over 4 years, the relative risk increase in dialysis patients vs. the
(2008, prospective
We conducted a literature search of Medline through Ovid undertreated population of dialysis patients with AF (only
general population for cerebral haemorrhage was 10.7. The rela-
multicentre)
(1966 to April 2010). The Medical Subject Heading terms ‘war- tive risk for subarachnoid bleed was24% of AF patients were on warfarin at enrolment) compared
4.0. Additionally, these
farin’, ‘atrial fibrillation’, ‘bleeding’, and ‘stroke’ were combined bleeds occurred 10 years earlier than in the patients without AF (15.4 vs. 12.4%; P ¼ 0.4).
with general population.
with ‘end-stage renal disease’, ‘dialysis’, ‘haemodialysis’, and The Rotterdam Study demonstrated similar increased risks for hae-
DOPPS3 (2010, 3245 (509) Not Warfarin use was associated with higher stroke risk; significantly
‘kidney failure’. Additional searches were also conducted for ‘calci- morrhagic stroke.13 The age- and sex-adjusted hazard ratio for
retrospective) uraemic arteriopathy.’
phylaxis’ and ‘calcific reported (95% CI, patients .75 years of age (HR ¼ 2.17; 95% CI 1.04–4.53,
haemorrhagic stroke was 4.1 in 1.25 –13.42) for the
lowest quartile of the estimated GFR (,53.90.04).
P ¼ mL/min/1.73 m2 for
Downloaded from http://
Haemodialysis patients and the
Chan et al. (2010, 42
1671 (746) 19 months
2
men; ,50.4 mL/min/1.73 m for women) vs. the highest quartile haemorrhagic stroke risk (1.2%/year
Warfarin use increased
of estimated GFR (.72 mL/min/1.73 m2 for men; .70.1 mL/
baseline risk of bleeding
retrospective) among warfarin users vs. 0.5%/year among non-users) and
min/1.73 m2 for women). The presumed mechanism is the effect
Patients on dialysis have an increased risk of bleeding at baseline of uraemia on platelet function or perhaps the stroke risk (5.8%/year among warfarin users vs.
ischaemic relationship
due to multiple factors. There is an acquired defect in primary hae- 2.3%/year among non-users) without increasing all-cause
between GFR and cerebral small-vessel disease.14 The devastating
mostasis as a result of defects in platelet secretion, aggregation, and mortality or hospitalization
effects of cerebral haemorrhage are evident in the fact that among
* Corresponding author. Tel: +1 718 920 7383; fax: +1 646 292 5187, Email: fey2002@gmail.com

6. Objetivos
• Determinar el riesgo EVC o tromboembolismo sistémico
y sangrado asociado con ERC en pacientes con FA y
determinar si el efecto de la warfarina y aspirina difieren
entre pacientes con o sin enfermedad renal crónica.

7. Métodos
• Cohorte Danesa y registros Nacionales de pacientes.
• Estudio retrospectivo.
• Pacientes con FA no valvular diagnosticados en hospital
entre 1997 y 2008.
• El periodo de seguimiento comenzó 7 días después del
egreso hospitalario.
• Se excluyeron los que murieron, tuvieron un evento
tromboembólico o un sangrado mayor dentro de los
primeros 7 días después del egreso del hospital.

8. 146,251 Patients were discharged with nonvalvular
atrial fibrillation (1997–2008)
13,879 Were excluded
187 Were discharged ≤7 days before end of study
2506 Died ≤7 days after discharge
1617 Had thromboembolism or major bleeding
≤7 days after discharge
9569 Were treated with platelet inhibitors other
than aspirin
127,884 (96.6%) Did not have 3587 (2.7%) Received a diagnosis of 901 (0.7%) Underwent renal-
renal disease non–end-stage chronic kidney disease replacement therapy
4538 (3.5%) Received a diagnosis of 228 (6.4%) Underwent renal-replace-
non–end-stage chronic kidney ment therapy during follow-up
disease during follow-up
215 (4.7%) Underwent renal-
replacement therapy
34 Underwent renal-replacement
therapy during follow-up
Figure 1. Study Population, According to Status with Respect to Renal Disease.
Of the 132,372 patients who were discharged from the hospital with a diagnosis of nonvalvular atrial fibrillation during
the study period and who were included in the study, 4488 had renal disease at baseline. The majority of the study
population had no renal disease at baseline, although non–end-stage chronic kidney disease or a disorder necessi-
tating renal-replacement therapy developed in 4572 patients during follow-up.

9. Métodos
• Se estratificó a los pacientes en forma dependiente del
tiempo en base a las dosis de diurético de asa que
recibían.
• Se estudió la influencia de la enfermedad causante de la
falla renal y se dividió en grupos:
• Enfermedad poliquística renal AD.
• Glomerulonefritis crónica.
• Nefropatía diabética.
• Nefropatía tubulointersticial crónica.
• Nefropatía hipertensiva.
• Otras.

10. Métodos
• EVC y Sangrado:
• Escala CHA2DS2-VASc (no se incluyeron: presencia de falla renal
y nivel de INR)
• Escala HAS-BLED
• Resultados:
• Hospitalización o muerte por EVC o tromboembolismo
sistémico.
• Sangrado: GI, intracraneal, tracto urinario o vía aérea.
• Infarto al miocardio.
• Otras causas.

11. Table S1. The CHA2DS2-VASc score7
Letter Risk factor Points
C Congestive heart failure 1
H Hypertension 1
A Age 2
D Diabetes mellitus 1
S Stroke or transient ischaemic attack or systemic thromboembolism 2
V Vascular disease (myocardial infarction, peripheral artery disease, aortic plaque) 1
A Age 65-74 years 1
Sc Sex category (female sex) 1 9
Table S2. The HAS-BLED score8
Letter Risk factor Points
H Hypertension (i.e. uncontrolled blood pressure) 1
A Abnormal liver or renal function 1 or 2
S Stroke or transient ischaemic attack or systemic thromboembolism 1
B Bleeding tendency or predisposition 1
L Labile international normalized ratio (INR) 1
E Elderly (i.e. >65 years) 1
D Drugs (antiplatelet agents, non-steroidal anti-inflammatory drugs) or alcohol abuse 1 or 2

12. Análisis Estadístico
• Comparación de características:
• Chi-cuadrada para variables categóricas.
• Test Kruskal-Wallis o T de Student para variables continuas.
• Riesgo de EVC o tromboembolismo sistémico:
• Modelos de riego proporcional de Cox.

13. Table 1. Baseline Characteristics of the Patients.*
Non–End-Stage Disease Requiring
Chronic Kidney Renal-Replacement
No Renal Disease Disease Therapy
Characteristic (N = 127,884) (N = 3587) (N = 901) P Value
Age — yr 73.2±12.9 76.5±11.0 66.8±11.7 <0.001
Risk factors for stroke or thromboembo-
lism — no. (%)
Congestive heart failure 22,073 (17.3) 1284 (35.8) 171 (19.0) <0.001
Hypertension 53,917 (42.2) 1952 (54.4) 486 (53.9) <0.001
Age
≥75 yr 66,675 (52.1) 2277 (63.5) 267 (29.6) <0.001
65–74 yr 31,245 (24.4) 809 (22.6) 293 (32.5) <0.001
Diabetes mellitus 10,920 (8.5) 885 (24.7) 129 (14.3) <0.001
History of stroke or systemic throm- 17,928 (14.0) 644 (18.0) 133 (14.8) <0.001
boembolism
Vascular disease 18,174 (14.2) 1034 (28.8) 248 (27.5) <0.001
Female sex 59,930 (46.9) 1472 (41.0) 303 (33.6) <0.001
Risk factors for bleeding — no. (%)
Hypertension 53,917 (42.2) 1952 (54.4) 486 (53.9) <0.001
Abnormal liver function 2,070 (1.6) 106 (3.0) 36 (4.0) <0.001
History of stroke or systemic throm- 17,928 (14.0) 644 (18.0) 133 (14.8) <0.001
boembolism
History of bleeding 8,969 (7.0) 584 (16.3) 137 (15.2) <0.001
Age ≥65 yr 95,418 (74.6) 3035 (84.6) 533 (59.2) <0.001
Use of NSAIDs 26,592 (20.8) 843 (23.5) 99 (11.0) <0.001
Alcohol abuse 4,552 (3.6) 145 (4.0) 43 (4.8) 0.05
Antithrombotic medication — no. (%)
Warfarin only 36,638 (28.6) 609 (17.0) 178 (19.8) <0.001
Aspirin only 23,952 (18.7) 879 (24.5) 153 (17.0) <0.001
Warfarin and aspirin 10,745 (8.4) 290 (8.1) 45 (5.0) <0.001
CHA2DS2-VASc score† <0.001
0 11,720 (9.2) 70 (2.0) 42 (4.7)
1 16,926 (13.2) 251 (7.0) 165 (18.3)
≥2 99,238 (77.6) 3266 (91.1) 694 (77.0)
HAS-BLED score‡ <0.001
0 or 1 51,262 (40.1) 883 (24.6) 390 (43.3)
2 46,159 (36.1) 1336 (37.2) 312 (34.6)
≥3 30,463 (23.8) 1368 (38.1) 199 (22.1)
* Plus–minus values are means ±SD. NSAIDs denotes nonsteroidal antiinflammatory drugs.

14. Resultados
• Se incluyeron a 132,372 pacientes con FA no valvular:
• 127,884 (96.6%) no tenían enfermedad renal de base.
• 3,587 (2.7%) tenían ERC no terminal.
• 901 (0.7%) requerían terapia de reemplazo.
• 4538 (3.5%) progresaron a enfermedad renal no terminal.
• 477 (0.4%) requirieron terapia sustitutiva.
• De los 1378 pacientes con ERC ya en reemplazo:
• 1074 (77.9%) hemodiálisis.
• 212 (15.4%) diálisis peritoneal.
• 92 (6.7%) transplantados.

15. risk factors, antithrombotic the
Table 2. Event Rates, According to Status with Respect to Renal Disease.*
inclusion. As compared with pat
Event Rate per 100 have renal disease, the risk of s
No. of No. of Person-yr thromboembolism was increase
Event Person-yr Events (95% CI)
with non–end-stage chronic kid
Stroke or thromboembolism ard ratio, 1.49; 95% confidence in
No renal disease 461,734 16,648 3.61 (3.55–3.66) 1.59; P<0.001) and among those
Non–end-stage CKD 13,078 842 6.44 (6.02–6.89) replacement therapy (hazard ra
Disease requiring renal- 2,922 164 5.61 (4.82–6.54) 1.57 to 2.14; P<0.001). The resul
replacement therapy a sensitivity analysis for the out
Bleeding systemic thromboembolism that
No renal disease 457,605 16,195 3.54 (3.48–3.59) ischemic attack (Table 5 in th
Appendix). Among patients wit
Non–end-stage CKD 12,515 1,097 8.77 (8.26–9.30)
chronic kidney disease, no signi
Disease requiring renal- 2,734 243 8.89 (7.84–10.08)
replacement therapy
between the risk of stroke or sy
embolism and dose of loop diu
Myocardial infarction
(Table 6 in the Supplementary A
No renal disease 480,745 9,037 1.88 (1.84–1.92)
the diagnostic groups analyzed,
Non–end-stage CKD 13,500 784 5.81 (5.41–6.23) or systemic thromboembolism w
Disease requiring renal- 2,925 175 5.98 (5.16–6.94) greatest degree among patients
replacement therapy nephropathy and to the lowest de
Death with chronic tubulointerstitial
No renal disease 493,305 55,297 11.21 (11.12–11.30) whom the increase in risk was n
Non–end-stage CKD 14,052 5,431 38.65 (37.63–39.69) Table 6 in the Supplementary A
Disease requiring renal- 3,114 914 29.35 (27.51–31.32) Warfarin treatment was as
replacement therapy significantly decreased risk of s
thromboembolism overall and
* A patient’s renal status could change during follow-up. CI denotes confidence requiring renal-replacement the
interval, and CKD chronic kidney disease.
nonsignificantly decreased risk
with non–end-stage chronic kid
tients (0.4%) after a median of 217 days (interquar- ble 3). In an analysis that comp
tile range, 33 to 681). Of the 1378 patients requir- who had any renal disease with

16. Table 3. Hazard Ratios for Stroke or Systemic Thromboembolism.*
Non–End-Stage Chronic Kidney Disease Requiring Renal-
Total Population
No Renal Disease Disease Replacement Therapy
(N = 132,372)
Characteristic (N = 127,884)† (N = 3587)† (N = 901)†
Hazard Ratio Hazard Ratio Hazard Ratio Hazard Ratio
(95% CI) P Value (95% CI) P Value (95% CI) P Value (95% CI) P Value
All participants 1.00 1.49 (1.38–1.59) <0.001 1.83 (1.57–2.14) <0.001
Antithrombotic therapy
None 1.00 1.00 1.00 1.00
Warfarin 0.59 (0.57–0.62) <0.001 0.59 (0.56–0.61) <0.001 0.84 (0.69–1.01) 0.07 0.44 (0.26–0.74) 0.002
Aspirin 1.11 (1.07–1.15) <0.001 1.10 (1.06–1.14) <0.001 1.25 (1.07–1.47) 0.01 0.88 (0.59–1.32) 0.54
Warfarin and aspirin 0.70 (0.65–0.75) <0.001 0.69 (0.64–0.74) <0.001 0.76 (0.56–1.03) 0.08 0.82 (0.37–1.80) 0.62
Risk factors for thromboembolism‡
Congestive heart failure 1.03 (0.99–1.07) 0.18 1.03 (0.99–1.08) 0.11 0.98 (0.84–1.14) 0.78 0.96 (0.64–1.43) 0.84
Hypertension 1.06 (1.03–1.09) <0.001 1.05 (1.02–1.09) 0.002 1.13 (0.98–1.30) 0.10 1.05 (0.76–1.45) 0.78
Age
≥75 yr 3.48 (3.31–3.66) <0.001 3.56 (3.38–3.76) <0.001 1.87 (1.48–2.36) <0.001 2.46 (1.60–3.79) <0.001
65–74 yr 2.02 (1.91–2.14) <0.001 2.03 (1.92–2.16) <0.001 1.52 (1.18–1.94) 0.001 2.18 (1.46–3.24) <0.001
Diabetes 1.32 (1.26–1.38) <0.001 1.32 (1.25–1.39) <0.001 1.16 (0.99–1.36) 0.07 1.41 (0.95–2.10) 0.09
History of stroke or systemic 3.20 (3.10–3.31) <0.001 3.24 (3.14–3.35) <0.001 2.71 (2.34–3.15) <0.001 1.99 (1.36–2.91) <0.001
thromboembolism
Vascular disease 1.10 (1.06–1.15) <0.001 1.12 (1.07–1.16) <0.001 0.89 (0.76–1.05) 0.17 1.11 (0.78–1.58) 0.57
Female sex 1.12 (1.08–1.15) <0.001 1.12 (1.08–1.15) <0.001 1.06 (0.92–1.22) 0.44 1.34 (0.97–1.85) 0.08
* Results from the time-dependent Cox regression analyses were adjusted for year of inclusion.
† Numbers of patients are from baseline data. Because the analyses were time-dependent, these numbers changed during follow-up.
‡ The reference group for the hazard ratio for each thromboembolism risk factor is the group of all patients in the study without that risk factor.

17. Table 4. Hazard Ratios for Bleeding.*
Disease Requiring Renal-
Total Population No Renal Disease Non–End-Stage Chronic Replacement Therapy
Characteristic (N = 132,372) (N = 127,884)† Kidney Disease (N = 3587)† (N = 901)†
Hazard Ratio Hazard Ratio Hazard Ratio Hazard Ratio
(95% CI) P Value (95% CI) P Value (95% CI) P Value (95% CI) P Value
All participants 1.00 2.24 (2.10–2.38) <0.001 2.70 (2.38–3.07) <0.001
Antithrombotic therapy
None 1.00 1.00 1.00 1.00
Warfarin 1.28 (1.23–1.33) <0.001 1.28 (1.23–1.33) <0.001 1.36 (1.17–1.59) <0.001 1.27 (0.91–1.77) 0.15
Aspirin 1.21 (1.16–1.26) <0.001 1.21 (1.16–1.26) <0.001 1.12 (0.96–1.30) 0.14 1.63 (1.18–2.26) 0.003
Warfarin and aspirin 2.15 (2.04–2.26) <0.001 2.18 (2.07–2.30) <0.001 1.63 (1.32–2.02) <0.001 1.71 (0.98–2.99) 0.06
Risk factors for bleeding‡
Hypertension 1.01 (0.98–1.04) 0.52 1.01 (0.98–1.04) 0.58 0.99 (0.87–1.11) 0.81 0.92 (0.71–1.20) 0.55
Abnormal liver function 1.37 (1.23–1.52) <0.001 1.40 (1.25–1.57) <0.001 1.31 (0.90–1.91) 0.16 0.74 (0.34–1.64) 0.46
History of stroke or systemic 1.23 (1.18–1.28) <0.001 1.24 (1.19–1.30) <0.001 1.04 (0.89–1.22) 0.62 0.93 (0.63–1.36) 0.70
thromboembolism
History of bleeding 2.44 (2.33–2.55) <0.001 2.54 (2.42–2.67) <0.001 1.70 (1.45–1.99) <0.001 2.09 (1.50–2.91) <0.001
Age ≥65 yr 2.09 (2.00–2.17) <0.001 2.12 (2.03–2.20) <0.001 1.61 (1.35–1.92) <0.001 1.36 (1.03–1.80) 0.03
Use of NSAIDs 1.12 (1.08–1.16) <0.001 1.12 (1.08–1.17) <0.001 1.10 (0.96–1.26) 0.19 0.91 (0.62–1.33) 0.63
Alcohol abuse 1.40 (1.30–1.52) <0.001 1.43 (1.32–1.56) <0.001 1.01 (0.73–1.39) 0.97 1.33 (0.70–2.54) 0.39
* Results from the time-dependent Cox regression analyses were adjusted for year of inclusion. NSAIDs denotes nonsteroidal antiinflammatory drugs.
† Numbers of patients are from baseline data. Because the analyses were time-dependent, these numbers changed during follow-up.
‡ The reference group for the hazard ratio for each bleeding risk factor is the group of all patients in the study without that risk factor.

18. 14
Table S6. Risk of stroke or systemic thromboembolism and bleeding associated with level and type of non-end-stage chronic kidney
disease in patients with non-valvular atrial fibrillation (n=132,372)
Stroke or systemic
Frequency Bleeding
thromboembolism
N (%) HR (95% CI) P value HR (95% CI) P value
Patients with no renal disease 127,884 (96.6)* 1 (reference) 1 (reference)
Non-end-stage CKD (overall) 3,587 (2.7)* 1.49 (1.38-1.59) <0.001 2.24 (2.10-2.38) <0.001
Non-end-stage CKD by loop-diuretics†
Furosemide 1,820 (22.4)‡ 1.40 (1.26-1.54) <0.001 1.93 (1.76-2.11) <0.001
Furosemide 40-160 mg/daily 2,760 (34.0)‡ 1.60 (1.43-1.80) <0.001 2.41 (2.17-2.68) <0.001
Furosemide 3,545 (43.6)‡ 1.52 (1.30-1.79) <0.001 2.85 (2.51-3.23) <0.001
Non-end-stage CKD by underlying disease
Adult polycystic kidney disease 224 (2.8)§ 1.22 (0.88-1.69) 0.24 1.78 (1.32-2.40) <0.001
Chronic glomerulonephritis 584 (7.2)§ 1.30 (1.03-1.64) 0.03 2.81 (2.40-3.30) <0.001
15
Diabetic nephropathy 1,526 (18.8)§ 1.41 (1.21-1.65) <0.001 1.96 (1.70-2.25) <0.001
Chronic tubulointerstitial nephropathy 875 (10.8)§ 1.18 (0.97-1.44) 0.09 1.32 (1.08-1.62) 0.01
Hypertensive nephropathy 282 (3.5)§ 1.92 (1.46-2.54) <0.001 1.82 (1.36-2.43) <0.001
Other etiology 184 (2.3) 1.43 (0.96-2.14) 0.08 1.38 (0.91-2.10) 0.13
Nephropathy of unknown etiology 4,625 (56.9) 1.60 (1.45-1.76) <0.001 2.49 (2.29-2.71) <0.001
Results from time-dependent Cox regression analyses adjusted for CHA2DS2-VASc risk factors or HAS-BLED risk factors, respectively,
antithrombotic treatment, and year of inclusion. CI: confidence interval; CKD: chronic kidney disease; HR: hazard ratio.
* Numbers at baseline. As the analyses were time-dependent, these numbers changed during follow-up.
† Patients with non-end-stage CKD were time-dependently stratified according to loop-diuretic treatment dose.
‡ Number of patients with maximum treatment dose during follow-up

19. Figure S1
Risk with non-end-stage chronic kidney disease (CKD) and renal replacement therapy (RRT)
in patients with non-valvular atrial fibrillation
Stroke or TE
Non-end-stage CKD
RRT
Bleeding
Non-end-stage CKD
RRT
Myocardial infarction
Non-end-stage CKD
RRT
Death from all causes
Non-end-stage CKD
RRT
0,8 1 1,5 2 2,5 3
Hazard ratio
Results from multivariate (adjusted for all baseline characteristics) time-dependent Cox regression
analyses. Results only shown for non-end-stage CKD and RRT, the reference was patients with no

20. Discusión
• La enfermedad renal no terminal y la terminal se asocian a
un riesgo aumentado de EVC, tromboembolismo
sistémico y sangrado.
• El riesgo de eventos isquémicos-embólicos no se modifica
de acuerdo a la severidad de la enfermedad renal.
• El riesgo de sangrado sí se asocia a la severidad de la ER y
a la causa de ésta.
• El riesgo de IAM y muerte se ve incrementado en
pacientes con FA y ERC.

21. Discusión
• Warfarina reduce el riesgo de EVC y tromboembolismo
sistémico en toda la población mientras Aspirina no.
• El tratamiento más efectivo en este tipo de pacientes es la
anticoagulación oral.
• Se necesitan más ECA que incluyan pacientes con ERC ya
que sólo se han estudiado en aquellos con TFG ≥30ml/
min.
• Estudio limitado por ser cohorte observacional.
• Pacientes hospitalizados.
• Aspirina.