Nonvitamin K Anticoagulant Agents in Patients With Advanced Chronic Kidney Disease or on Dialysis With AF

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 67, NO. 24, 2016

ª 2016 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER

http://dx.doi.org/10.1016/j.jacc.2016.02.082

REVIEW TOPIC OF THE WEEK

Nonvitamin K Anticoagulant Agents in Patients With Advanced Chronic Kidney Disease or on Dialysis With AF Kevin E. Chan, MD, MSCI,a,b Robert P. Giugliano, MD, SM,c Manesh R. Patel, MD,d Stuart Abramson, MD,e Meg Jardine, MBBS, PHD,f Sophia Zhao, MD, PHD,a Vlado Perkovic, MBBS, PHD,f Franklin W. Maddux, MD,b Jonathan P. Piccini, MD, MHSCd

ABSTRACT Nonvitamin K-dependent oral anticoagulant agents (NOACs) are currently recommended for patients with atrial fibrillation at risk for stroke. As a group, NOACs significantly reduce stroke, intracranial hemorrhage, and mortality, with lower to similar major bleeding rates compared with warfarin. All NOACs are dependent on the kidney for elimination, such that patients with creatinine clearance <25 ml/min were excluded from all the pivotal phase 3 NOAC trials. It therefore remains unclear how or if NOACs should be prescribed to patients with advanced chronic kidney disease and those on dialysis. The authors review the current pharmacokinetic, observational, and prospective data on NOACs in patients with advanced chronic kidney disease (creatinine clearance <30 ml/min) and those on dialysis. The authors frame the evidence in terms of risk versus benefit to bring greater clarity to NOAC-related major bleeding and efficacy at preventing stroke specifically in patients with creatinine clearance <30 ml/min. (J Am Coll Cardiol 2016;67:2888–99) © 2016 by the American College of Cardiology Foundation.

P

atients with atrial fibrillation (AF) are in-

little experience with reversal agents, as well as

creasingly being treated with nonvitamin

higher costs. Currently, all NOACs depend to some

K-dependent

agents

extent on renal function for clearance. Consequently,

(NOACs) because these drugs are at least as effective

NOACs may potentially accumulate in patients with

as warfarin, are safer, do not require routine moni-

renal dysfunction, leading to an increased risk for

toring, and are simpler to use (1–5). In 2013, NOACs

bleeding.

accounted for 62% of new prescriptions of anticoagu-

chronic kidney disease (CKD) (creatinine clearance

lant agents in the United States, and their adoption is

[CrCl] <30 ml/min) are already at increased risk for

increasing (1).

bleeding from uremia-induced platelet dysfunction

oral

anticoagulant

Additionally,

patients

with

Although NOACs hold great promise, they also

(6,7). Patients on hemodialysis have added bleeding

have drawbacks, such as limited availability of and

risks from repeated vascular access cannulation,

From the aFresenius Medical Care North America, Waltham, Massachusetts; bDivision of Nephrology, Massachusetts General Hospital, Boston, Massachusetts; cCardiovascular Division, Brigham and Women’s Hospital, Boston, Massachusetts; dDuke Clinical Listen to this manuscript’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster.

advanced

Research Institute, Duke University Medical Center, Durham, North Carolina; eMaine Medical Center, Tufts University School of Medicine, Portland, Maine; and the fGeorge Institute for Global Health, University of Sydney, Sydney, Australia. Dr. Chan is a steering committee member of the RENAL-AF trial, which is funded by Bristol-Myers Squibb; and is an employee of Fresenius Medical Care. Dr. Piccini receives grants for clinical research from ARCA Biopharma, the Agency for Healthcare Research and Quality, Gilead, Johnson & Johnson, and St. Jude Medical; and serves as a consultant to Janssen Pharmaceuticals, Pfizer-BMS, and Medtronic. Dr. Giugliano was the principal investigator of the ENGAGE AF–TIMI 48 trial, which was supported by a research grant from Daiichi-Sankyo to the Brigham and Women’s Hospital, where he is employed; and has received honoraria for consulting or continuing medical education activities from Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, Janssen, The American College of Cardiology, Pfizer, and Portola. Dr. Maddux is an employee of Fresenius Medical Care and has stock ownership. Dr. Jardine is supported by a Career Development Fellowship from the National Health and Medical Research Council of Australia and

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

2889

NOACs in Advanced CKD and Dialysis

dialysis membrane interactions, higher than average

RENAL PHARMACOKINETICS OF

ABBREVIATIONS

blood pressures, and considerable heparin adminis-

ANTICOAGULANT AGENTS

AND ACRONYMS

tration during treatment. Because patients with advanced CKD and end-stage renal disease (ESRD)

Uremia affects every organ system in the

were excluded from all of the pivotal phase 3 NOAC

body and influences the pharmacokinetics of

trials, no randomized controlled trial data guide

many drugs. Uremia can impair plasma pro-

NOAC use when the CrCl is <25 ml/min, although

tein binding, which may lead to increased

the U.S. Food and Drug Administration (FDA)

free drug levels in the blood, and in turn,

label

renal failure can also impair nonrenal drug

supports

NOAC

dosing

for

patients

with

CrCl #15 ml/min.

metabolism by reduction and hydrolysis re-

The most recent 2014 American Heart Association

actions (10,11).

ACC = American College of Cardiology

AF = atrial fibrillation AHA = American Heart Association

CI = confidence interval CKD = chronic kidney disease CrCl = creatinine clearance ESRD = end-stage renal

(AHA), American College of Cardiology (ACC), and

Renal elimination of drugs occurs predom-

Heart Rhythm Society (HRS) guideline for the man-

inantly through glomerular filtration and oc-

agement of patients with AF recommends warfarin as

casionally through tubular secretion and

Administration

the anticoagulant agent of choice in patients with

reabsorption. When the glomerular filtration

HR = hazard ratio

advanced CKD or ESRD (8). Despite these recommen-

rate and tubular function are impaired by

dations and the lack of prospective trials to validate the

renal disease, the clearance of drugs elimi-

efficacy and safety of NOACs in this patient population,

nated by these mechanisms will be decreased,

the prescription of these drugs is becoming more

and the plasma half-lives of the drugs will be

prevalent, as we showed in our previous publication

prolonged. This can result in increased total

disease

FDA = U.S. Food and Drug

HRS = Heart Rhythm Society NOAC = nonvitamin Kdependent anticoagulant agent

PK = pharmacokinetic

(9). Our updated analysis, as of October 2015 (unpub-

drug exposure, as quantified by the area under the

lished data from Fresenius Medical Care), indicates

curve (Figure 1). Without appropriate dose adjustment,

that 23.5% of patients with advanced CKD and 11.6% of

repeated dosing of a drug that is inadequately cleared

those on dialysis with AF who are anticoagulated were

could lead to bioaccumulation over time and toxicity

prescribed NOACs (Central Illustration). Among the

from supratherapeutic levels of the drug (Figure 2). To

NOAC-treated patients with CKD (n ¼ 102,504), the

counteract decreased renal drug elimination, medi-

most commonly used drug was apixaban (10.4%), fol-

cations can be given at a lower dose or frequency to

lowed by rivaroxaban (9.5%), dabigatran (3.5%), and

prevent unintended drug accumulation (11).

edoxaban (0.1%). Among the NOAC-treated dialysis

Anticoagulant drug dosing in patients with renal

patients (n ¼ 140,918), the most commonly used drug

impairment is based on glomerular filtration rate,

was apixaban (10.5%), followed by rivaroxaban (0.8%),

which is impractical to measure directly in an office

dabigatran (0.3%), and edoxaban (0.01%).

setting. Instead, the glomerular filtration rate can

This review summarizes the available evidence

be estimated by measuring the 24-h urine CrCl,

regarding the efficacy (prevention of stroke) and

because almost all creatinine is eliminated through

safety (major bleeding) of the 4 currently available

the glomerulus. Because collecting urine for 24 h is

NOACs

a tedious process, CrCl is typically estimated by

(apixaban,

rivaroxaban,

dabigatran,

and

edoxaban), with particular emphasis on the outcomes

an

in subgroups of patients with AF with CKD. Pharma-

measurement.

equation,

using

a

single

serum

creatinine

cokinetic (PK) data for patients with advanced CKD

For NOAC dosing, CrCl should be calculated using

and those on dialysis will be discussed, as well as the

the Cockcroft-Gault formula, because the pivotal

data on reversal agents. Finally, we outline the steps

phase 3 NOAC trials have reported outcomes by CrCl

required to determine if NOACs are appropriate

using that formula (12). This is despite the fact that

treatment options in patients with AF with advanced

the Cockcroft-Gault formula can overestimate true

CKD or on dialysis.

CrCl by 10% to 40%, and newer estimating equations,

the National Heart Foundation; has received speaking fees from Amgen and Roche; has received advisory board funding from Boehringer Ingelheim; has received unrestricted research funding from Gambro, Baxter, CSL, Amgen, Eli Lilly and Merck; and has served on a steering committee for a trial funded by Janssen; all honoraria are contributed directly to clinical research programs. Dr. Patel is supported by research grants from AstraZeneca, Bayer, Procyrion, Jansen, and the National Heart, Lung, and Blood Institute; and is a consultant advisory board member to AstraZeneca, Jansen, Bayer, Genzyme, and Merck. Dr. Perkovic chairs the steering committee of a trial of an unrelated compound funded by Janssen; has consulted for Bristol-Myers Squibb in an unrelated area; and has a policy of honoraria going to his institution. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Manuscript received January 20, 2016; revised manuscript received February 16, 2016, accepted February 23, 2016.

2890

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

C E NT R AL IL L U STR AT IO N Use of Nonvitamin K-Dependent Oral Anticoagulant Agents in Patients With Advanced Chronic Kidney Disease and on Dialysis: Substantial and Growing

Chan, K.E. et al. J Am Coll Cardiol. 2016;67(24):2888–99.

Prevalence of nonvitamin K-dependent oral anticoagulant agent (NOAC) use is rising among patients with advanced chronic kidney disease (CKD) and those on dialysis anticoagulated for atrial fibrillation, despite the most recent American Heart Association, American College of Cardiology, and European Heart Rhythm Society guideline, which discourages the use of NOACs when creatinine clearance (CrCl) is <30 ml/min. There are few randomized trial data on NOACs in this population. All NOACs depend on the kidney for elimination, and it is unclear if severe renal impairment leads to drug bioaccumulation to precipitate inadvertent bleeding.

F I G U R E 1 Pharmacokinetic Curves for Renally Cleared Drugs in Patients With Normal and Impaired Kidney Function

Drug Concentration (ng/mL)

A

B 100

100

80

80

60

60

40

40

20

20

0

0 0

6

12

18

24

0

6

Time (Hours)

12

18

24

Time (Hours)

Pharmacokinetic (PK) curves illustrating the peak, trough, and total drug exposure for a renally cleared drug in a patient with normal kidney function (A) and in a patient with impaired kidney function (B). Decreased drug clearance resulted in increased peak level, trough level, and total drug exposure. Total drug exposure is quantified by the area under the PK curve, in blue.

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

such as the MDRD (Modification of Diet in Renal Disease)

and

2891

NOACs in Advanced CKD and Dialysis

CKD-EPI

(Chronic

Kidney Disease

Epidemiology Collaboration) equations, have been

F I G U R E 2 Pharmacokinetic Curve Illustrating the Effect of Drug Multidosing in a

Patient With Impaired Renal Function, Resulting in Bioaccumulation of the Drug Level

shown to be more accurate (13–15). The Cockcroft100

CrCl ðml=minÞ ¼

ð140  ageÞ  weight ðkgÞ  sex 72  serum creatinine ðmg=dlÞ

where sex ¼ 1 in men and 0.85 in women. It is also important to note that CrCl estimations derived from these equations are valid only when a patient’s creatinine level is in a steady state. It is a mistake to extrapolate CrCl calculations using creatinine values

Drug Concentration (ng/mL)

Gault equation is as follows:

80 60 40 20 0 0

from acutely ill patients with acute kidney injury.

12

24

36

48

60

72

Time (Hours)

Drug dosing in patients with acute kidney injury should be individualized until the creatinine level stabilizes, which may take weeks or longer (11). Drug removal through hemodialysis is determined

A mismatch between drug dose and drug clearance results in the carryover of excess drug level to the next dosing interval, which occurs every 24 h.

primarily by molecular size. Small molecules (<1,500 Da) are amenable to removal through high-flux dialysis membranes. Protein-bound drugs are not cleared by dialysis, whereas unbound drugs in the intravascular space may be cleared. A drug with a high volume of distribution, arbitrarily defined at >0.7 l/kg, indicates a highly tissue-bound drug, where a significant amount of drug is distributed in the extravascular space that is not directly accessible to the extracorporeal circulation for removal by dialysis. Measured drug levels are expected to rebound after hemodialysis is completed and are the result of drug concentrations in the intra- and extravascular compartment reequilibrating after dialysis (11).

anticoagulation levels from the drug. Warfarin has the advantages of clinical familiarity, low cost, and widespread availability of reversal agents. Reversal of warfarin’s anticoagulant effect can be accomplished with vitamin K or fresh-frozen plasma for nonemergent situations, with 4-factor prothrombin complexes recommended for life-threatening bleeding (22). Warfarin significantly increases the risk for bleeding as CrCl decreases. This is likely secondary to superimposed platelet dysfunction from worsening uremia. A patient on warfarin with CrCl <30 ml/min has a 4.9-fold (95% confidence interval [CI]: 2.6 to 9.1) increased

relative

risk

for

bleeding

compared

WARFARIN F I G U R E 3 Pharmacokinetics of Warfarin

Warfarin, originally used as a rodenticide in 1948, was warfarin

later found to be effective and relatively safe for preventing thrombosis and was approved for use as an anticoagulant agent in 1954 (16). Warfarin is not

parent drug (99% bound)

dialyzable, because 99% is bound to plasma proteins, and elimination of the drug is almost entirely by hepatic metabolism into an inactive metabolite CYP2C9 metabolized

(Figure 3). Very little of the parent compound is excreted in the urine (11,17,18). Studies also suggest that patients with advanced CKD and on dialysis require reduced doses of warfarin. This may be in part due to alterations of hepatic metabolism of warfarin secondary to renal failure (19). Animal studies in CKD have shown a significant down-regulation (40% to

Inactive metabolite

85%) of hepatic cytochrome P-450 metabolism, which renal elimination

corroborates with clinical data (17,19–21). Warfarin is minimally dependent on the kidney for elimination, such that progression of CKD or acute changes

in

creatinine

can

minimally

influence

Warfarin does not depend on the kidney for elimination.

2892

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

they met the same criteria used in the AVERROES trial

T A B L E 1 Estimated Major Bleeding Rates for Patients on

for dose reduction, 4.7% of apixaban patients

Warfarin by Creatinine Clearance

received a lower dose of 2.5 mg twice per day. Overall, Kidney Function

Major Bleeding (Events per 100 Patient-Years)

apixaban significantly decreased the risk for stroke

6.2 (4.1–8.9)

and systemic embolism (HR: 0.79; 95% CI: 0.66 to

CrCl $60 ml/min CrCl 30–59 ml/min CrCl <30 ml/min

8.3 (5.1–12.8)

0.95) and major bleeding (HR: 0.69; 95% CI: 0.60 to

30.5 (17.0–50.3)

0.80) compared with warfarin (2). These trends per-

54–100

sisted in post hoc analyses among patients with CrCl

Dialysis

of 25 to 50 ml/min, in whom apixaban demonstrated a Data from Limdi et al. (17) and Elliott et al. (64).

nonsignificantly decreased risk for stroke (HR: 0.79;

CrCl ¼ creatinine clearance.

95% CI: 0.55 to 1.14) and a significantly decreased risk for bleeding (HR: 0.50; 95% CI: 0.38 to 0.66) referent with patients on warfarin without renal disease

to warfarin (24,25). Because of multiple criteria for

(Table 1) (17).

dose reduction, it is uncertain how many patients with CrCl of 25 to 50 ml/min received the 2.5-mg

NOACs

versus the 5-mg apixaban dose and whether dose affects these HR estimates.

Currently, there are 4 FDA-approved NOACs for

Apixaban was FDA approved in December 2012 for

stroke prevention in patients with AF: apixaban,

the prevention of stroke and systemic embolism in

rivaroxaban, dabigatran, and edoxaban. Table 2

patients with nonvalvular AF at a dose of 5 mg twice

summarizes the renal clearance, effect of dialysis,

per day and with a 2.5-mg twice-daily dose in patients

and reversal agents for these medications; in addi-

with 2 of the following: serum creatinine between

tion, it outlines the hazard ratios (HRs) for stroke and

1.5 and 2.5 mg/dl, age $80 years, and body weight

bleeding from the pivotal phase 3 trials among sub-

#60 kg. In the original labeling, the drug was not

jects with CrCl <50 ml/min (referent to warfarin).

recommended for patients with CrCl <25 ml/min.

APIXABAN. Apixaban is a direct factor Xa inhibitor

The drug label was amended in January 2014 for

with 27% renal elimination (Figure 4A). In the

patients with renal impairment, including those with

AVERROES (Apixaban Versus Acetylsalicylic Acid to

ESRD maintained on hemodialysis. For these pa-

Prevent Stroke in Atrial Fibrillation Patients Who

tients, no dose reduction (5 mg twice daily) was

Have Failed or Are Unsuitable for Vitamin K Antago-

suggested unless patients were also $80 years of age

nist Treatment) trial (n ¼ 5,599 subjects with

or had body weight #60 kg, in which case the reduced

creatinine <2.5 mg/dl), subjects with AF who were

dose of 2.5 mg twice per day could be used. Patients

unable to take warfarin were randomized to apixaban

with creatinine >2.5 mg/dl, with CrCl <25 ml/min, or

5 mg twice per day or aspirin. A reduced dose of

on long-term dialysis were excluded from the

apixaban 2.5 mg twice per day was given to patients

ARISTOTLE trial; therefore, these dosing suggestions

who met at least 2 of the following criteria: serum

were based partially on a single-dose (not multidose)

creatinine between 1.5 and 2.5 mg/dl, age $80 years,

PK study in 8 hemodialysis subjects matched to

and body weight #60 kg. The study showed that

8 subjects with normal renal function (26). In this

apixaban was superior to aspirin in preventing stroke

study, the post-hemodialysis administration of 5 mg

or systemic embolization (HR: 0.32; 95% CI: 0.19 to

apixaban resulted in 36% higher drug exposure

0.56), and the trial was stopped prematurely by the

compared with healthy subjects with normal renal

Data and Safety Monitoring Board because of over-

function. In another 10-mg single-dose PK study of 24

whelming efficacy (23). No significant difference in

subjects with mild and moderate CKD compared with

the rate of major bleeding between the groups was

8 subjects with normal kidney function, total apix-

observed (HR: 1.06; 95% CI: 0.58 to 1.93). The supe-

aban exposure was estimated via regression models

riority of apixaban over aspirin in stroke prevention

to be 44% greater in subjects with CrCl of 15 ml/min

and a similar incidence of major bleeding were pre-

than in subjects with normal kidney function (27).

served among patients with CrCl of 25 to 50 ml/min in

Thus, under current dosing suggestions on the apix-

the study.

aban label, patients with advanced and end-stage

In the ARISTOTLE (Apixaban for Reduction in

kidney disease could be exposed to 40% more drug.

Stroke and Other Thromboembolic Events in Atrial

Further studies are needed to establish the optimal

Fibrillation)

trial

(n

¼

18,201

subjects

with

apixaban dose in this population.

creatinine <2.5 mg/dl), patients were randomized to

In terms of dialysis clearance, only 6.7% of apix-

apixaban 5 mg twice per day or warfarin. Because

aban is cleared by a 4-h hemodialysis session

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

T A B L E 2 Characteristics of Warfarin and Nonvitamin K-Dependent Oral Anticoagulant Agents

Warfarin

Apixaban

Rivaroxaban

Dabigatran

Edoxaban

Renal clearance of parent drug

<1%

27%

36%

80%

50%

Removal with 4 h of hemodialysis

<1%

7%

<1%

50%–60%

9%

8

21

50

50–10

107

Reversal agent

Vitamin K, FFP, 4F-PCC

4F-PCC

4F-PCC

Idarucizumab

4F-PCC

Lowest CrCl drug can be prescribed per FDA label, ml/min

Can be used on dialysis

<15*

15

15

15

HR (95% CI) of stroke referent to warfarin, CrCl <50 ml/min

Reference

0.79 (0.55–1.14)

0.88 (0.65–1.19)

0.56 (0.37–0.85)

0.87 (0.65–1.18)†

HR (95% CI) of major bleeding referent to warfarin, CrCl <50 ml/min

Reference

0.50 (0.38–0.66)

0.98 (0.84–1.14)

1.01 (0.79–1.30)

0.76 (0.58–0.98)†

Volume of distribution, l (66)

*A 5-mg twice-daily dose of apixaban is suggested for patients with CrCl <15 ml/min. This dosing suggestion was based on a small single-dose pharmacokinetic and pharmacodynamic (anti-Xa activity) study. Clinical efficacy and long-term safety studies have not been done in this population; therefore, use apixaban with caution in patients with advanced or end-stage chronic kidney disease. †30–50 ml/min for the comparison of edoxaban versus warfarin (37). CI ¼ confidence interval; CrCl ¼ creatinine clearance; FDA ¼ U.S. Food and Drug Administration; FFP ¼ fresh-frozen plasma; 4F-PCC ¼ 4-factor prothrombin complex concentrate; HR ¼ hazard ratio.

(Optiflux F180NR dialyzer, dialysate flow rate 500

RIVAROXABAN. Rivaroxaban is also a direct factor

ml/min, blood flow rate 350 to 500 ml/min, no hepa-

Xa inhibitor, and the kidneys eliminate 36% of the

rin) (26); therefore, in patients who have overdosed or

parent drug and the remainder is hepatically metab-

are having life-threatening bleeding, dialysis would

olized into an inactive form (Figure 4B) (28). In

not be an effective means to remove apixaban from

ROCKET-AF (Rivaroxaban Once Daily Oral Direct

the circulation.

Factor Xa Inhibition Compared With Vitamin K

F I G U R E 4 Pharmacokinetics of Nonvitamin K-Dependent Oral Anticoagulant Agents

A

B

apixaban

rivaroxaban

parent drug (95% bound)

parent drug (87% bound) 6% cleared with dialysis

CYP3A4/5 and P-glycoprotein

inactive metabolite

~50% in feces

27% renal elimination

C

CYP3A4/5 and CYP2J2 metabolized

51% inactive metabolite

7% feces

36% renal elimination

D dabigatran etexilate

edoxaban 10% metabolite dabigatran active (35% bound)

CYP3A4 parent drug (55% bound)

50-60% cleared with dialysis

80% renal elimination

(A) Apixaban, (B) rivaroxaban, (C) dabigatran, and (D) edoxaban.

9% cleared with dialysis

40% bile elimination

50% renal elimination

2893

2894

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

Antagonism for Prevention of Stroke and Embolism

dependent on the kidney for removal from the body

Trial in Atrial Fibrillation), a randomized controlled

(Figure 4C). In the RE-LY (Randomized Evaluation of

trial of 14,264 patients with AF at moderate to high risk

Long-Term Anticoagulation Therapy) trial, dabigatran

for stroke (mean CHADS2 score 3.5  0.9), all of whom

150 mg twice daily was superior to warfarin for the

had CrCl >30 ml/min, rivaroxaban was noninferior to

prevention

warfarin at preventing stroke and systemic embolism

(HR: 0.66; 95% CI: 0.53 to 0.82) in patients with AF

(3). Additionally, the occurrence of nonmajor clini-

and 1 or more additional risk factors for stroke;

cally relevant or major bleeding was not significantly

furthermore, no increase in major bleeding was seen

different between the 2 groups. Patients with CrCl

with the drug (HR: 0.93; 95% CI: 0.81 to 1.07) (4). In a

>50 ml/min received 20 mg rivaroxaban once daily,

separate analysis in relation to renal function, cubic

whereas subjects with CrCl of 30 to 49 ml/min received

splines were used to model the rate of major bleeding

15 mg once daily. There was no evidence of heteroge-

by CrCl for warfarin and 150 mg of dabigatran. Here,

neity in outcomes in patients treated with the 15- or

there was no statistical difference in bleeding between

20-mg dose compared with warfarin (29); further-

the 2 drugs; however, the rate of major hemorrhage

more, the effect estimates for stroke prevention were

among dabigatran-treated patients accelerated and

similar for rivaroxaban subjects with CrCl of 50 to

surpassed warfarin when the CrCl fell below 50 ml/min

80 ml/min (HR: 0.85; 95% CI: 0.67 to 1.08) and 30 to

(31). Thus, multiple dosing guidelines advise caution

50 ml/min (HR: 0.88; 95% CI: 0.65 to 1.19) relative to

with dabigatran when CrCl is between 30 and

warfarin. For major bleeding, there was no difference

50 ml/min (32). Accordingly, the U.S. label prohibits

in bleeding risk in the CKD population between

dabigatran coadministration with drugs that inhibit

rivaroxaban and warfarin overall.

P-glycoprotein. Dabigatran is also the only dialyzable

of

stroke

and

systemic

embolism

The FDA approved a 20-mg once-daily rivarox-

NOAC. A 4-h hemodialysis session will remove 50% to

aban dose in November 2011. The label included a

60% of plasma dabigatran, with a 10% rebound in

recommended 15-mg once-daily dose for patients

dabigatran levels post-dialysis (33).

with CrCl between 15 and 50 ml/min. Of note, Kubitza

The FDA approved a low dose of dabigatran (75 mg

et al. (28) reported a 52% increase in drug exposure and

twice daily) for patients with CrCl between 15 and

a 26% increase in peak concentration after a single

30 ml/min in 2010. The approval of a 75-mg dose in

10-mg dose among 8 patients with a mean CrCl of

patients with stage 4 CKD was on the basis of a phase I

43 ml/min. These data suggest that the dose of rivar-

PK study of 29 subjects, 11 of whom had CrCl

oxaban in patients with moderate CKD needed to

<30 ml/min. Model simulations showed that dabiga-

match the exposure of patients with no renal impair-

tran 75 mg daily in patients with advanced CKD had a

ment may have to be lower than 15 mg; however, these

32% lower steady-state peak and 42% lower trough

PK findings were not validated by outcomes from the

level compared with patients with moderate CKD on

ROCKET-AF subgroup analysis of patients with CKD.

150 mg of dabigatran twice per day (34,35). This is

No increased bleeding was seen in patients with

in contrast to another study that found a 6-fold in-

moderate CKD on 15 mg of rivaroxaban (HR: 0.98; 95%

crease in dabigatran exposure in patients with

CI: 0.84 to 1.14) referent to warfarin, which was the

CrCl <30 ml/min compared with healthy subjects (33).

same as the main conclusion of the full study. Rivar-

EDOXABAN. Edoxaban is the most recent factor Xa

oxaban is not recommended for use in patients with

inhibitor approved for use in the United States. The

CrCl <15 ml/min or those on dialysis. An inconse-

kidneys clear approximately 50% of the unmetabo-

quential amount of rivaroxaban is cleared by dialysis

lized drug (Figure 4D). Total drug exposure was found

(30). A 7-day trial of rivaroxaban 10 mg daily among 18

to increase by 32%, 74%, and 72% in patients with

maintenance hemodialysis patients (FX60, 4 h, blood

mild,

flow rate 400 ml/min, dialysate flow rate 500 ml/min,

moderate,

and

severe

renal

impairment,

respectively (36).

citrate anticoagulation) resulted in drug exposure

In the ENGAGE AF–TIMI 48 (Effective Anti-

levels similar to healthy volunteers given 20 mg;

coagulation With Factor Xa Next Generation in Atrial

however, the coefficient of variation was twice as high,

Fibrillation–Thrombolysis In Myocardial Infarction

suggesting moderate interpatient variability in the

48) trial of 21,105 patients with AF with CrCl

clearance or metabolism of the drug.

>30 ml/min, subjects received 60 mg of edoxaban, or

DABIGATRAN. Dabigatran is a direct thrombin inhib-

30 mg if their CrCl was between 30 and 50 ml/min,

itor and was the first FDA-approved NOAC, at a dose of

body weight was <60 kg, or a strong P-glycoprotein

150 mg twice per day for CrCl >30 ml/min. The drug

inhibitor was coadministered. Edoxaban was non-

has a renal clearance of at least 80% and is thus highly

inferior at preventing stroke and systemic embolism

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

and superior at preventing major bleeding (HR: 0.80;

Interpretation of these findings is challenging, as the

95% CI: 0.71 to 0.91) compared with warfarin. In

results may be confounded by indication and the

subgroup analyses among patients with CrCl between

inclination to treat sicker patients.

30 and 50 ml/min, edoxaban remained noninferior for

stroke

prevention

compared

with

Overall,

it

remains

uncertain

whether

anti-

warfarin.

coagulation provides more benefit than harm when

Bleeding was also significantly decreased (HR: 0.76;

used to prevent stroke in patients with AF with

95% CI: 0.58 to 0.98) with edoxaban (37).

advanced CKD or on dialysis. The 2014 AHA, ACC, and

A post hoc analysis of the ENGAGE AF–TIMI 48 trial

HRS guideline for the management of patients with AF

for patients on edoxaban (vs. warfarin) whose CrCl

states that it is reasonable to prescribe warfarin

fell below 30 ml/min after baseline (n ¼ 1,202) showed

(international normalized ratio 2.0 to 3.0) to patients

that stroke (2.36 vs. 1.89 events per 100 patient-years)

with nonvalvular AF with CHA 2DS2-VASc scores of 2 or

and major bleeding (6.83 vs. 6.49 events per 100

greater who have CrCl <15 ml/min or are on hemodi-

patient-years) rates were similar (38). Another pro-

alysis (8). In contrast, the KDIGO (Kidney Disease:

spective safety study of 93 patients found similar

Improving Global Outcomes) guidelines state that

3-month bleeding rates in patients with stage 4 CKD

routine anticoagulation of dialysis patients with AF for

on 15 mg of edoxaban versus 30 or 60 mg of edoxaban

the primary prevention of stroke is not indicated (46).

with mean CrCl of 70 ml/min (39). Both of the studies

Taken together, the aggregated data suggest that

mentioned in the preceding text were underpowered,

anticoagulation increases the risk for bleeding by at

and these results should be regarded as hypothesis

least 20% in patients with advanced CKD or on dial-

generating, given how few events occurred in these

ysis, but what remains unclear is the degree to which

studies.

warfarin and NOACs reduce the risk for stroke in

Edoxaban is poorly cleared by dialysis: 4 h of he-

patients with advanced CKD or ESRD (41,42,45).

modialysis decreased total drug exposure by only 9%

Randomized clinical trials in the general population

(F180NR, blood flow rate 350 ml/min, dialysate flow

have suggested that warfarin reduces the risk for

rate 500 ml/min) (40).

stroke by 64% in patients with AF compared with

SHOULD WE ANTICOAGULATE PATIENTS WITH AF WITH ADVANCED OR END-STAGE KIDNEY DISEASE TO PREVENT STROKE? Patients with AF with advanced CKD or on dialysis are at high risk for adverse events and are complicated to manage. Most of these patients will have CHA2DS2 VASc scores >2, and studies indicate that the rates of AF, stroke, and bleeding all increase as kidney function worsens (25). There are no prospective randomized studies on the efficacy and safety of anticoagulation to prevent stroke in this population. Thus, we do not know the magnitude of stroke prevention conferred by anticoagulation in patients with advanced CKD or on dialysis, nor whether the stroke prevention benefit of warfarin or NOACs outweighs the increased risk for bleeding. The risk for anticoagulantrelated bleeding remains a serious clinical concern. Dialysis patients are more likely to die of bleeding events than from embolic stroke. An unpublished analysis from the Fresenius Medical Care Research database showed that 2.7% of all deaths were secondary to hemorrhage, and 1% were from embolic stroke. Multiple observational studies in CKD and dialysis have identified an association between warfarin use

placebo (47). There is some evidence that anticoagulation does not lead to similar risk reduction in the advanced CKD and ESRD populations (48). Although patients with advanced CKD and ESRD are at higher risk for stroke, this risk for stroke may not be significantly modifiable by anticoagulation in patients with advanced renal disease. There are several reasons to explain why vitamin K antagonism may be associated with a lower margin of benefit in patients with advanced CKD or dialysis. First, uremia-induced platelet dysfunction may protect against thrombosis. Second, competing comorbid risks may attenuate the opportunity for benefit. More specifically, patients with CKD have numerous competing comorbidities that shorten their life expectancies and the follow-up time for stroke events. The average life expectancy of a dialysis patient is 5 years; moreover, 24% of CKD stage 4 and 45% of CKD stage 5 patients will die within 5 years (49,50). Until well-done, prospective, randomized studies are available, physicians must individually balance the risk for stroke in each patient against the perceived magnitude of stroke prevention anticoagulant agents may provide.

NOAC USE IN PATIENTS WITH LATEOR END-STAGE KIDNEY DISEASE

and increased risk for embolic stroke and bleeding (41–44), whereas other studies suggest that warfarin

If oral anticoagulation is indicated and the perceived

is protective against stroke in this population (45).

benefits outweigh the risks for therapy, the 2014

2895

2896

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

AHA, ACC, and HRS guidelines endorse warfarin

monitoring of renal function is advised, perhaps

as the first-line therapy in patients with CKD

every 2 to 4 months and during acute illness. In these

stage 4 (CrCl 15 to 30 ml/min) or CKD stage 5

rare instances in which NOACs are prescribed, a sug-

(CrCl <15 ml/min) or on dialysis, because all NOACs

gested dose of apixaban 2.5 to 5 mg twice daily can be

partially rely on the kidney for elimination (8).

given to patients on dialysis, but rivaroxaban or

Similarly, the 2015 updated European Heart Rhythm

edoxaban at a reduced dose could also be effective.

Association practical guide on the use of nonvitamin

Dabigatran is less favorable because the risk for

K antagonist anticoagulant agents in patients with

bleeding substantially increases when CrCl drops

nonvalvular AF recommends refraining from NOAC

below 50 ml/min. In dialysis patients, drug levels

use in dialysis patients and patients with CKD with

would substantially fluctuate with dialysis treatment,

CrCl <30 ml/min, given that there are few random-

given that dialysis clears 50% to 60% of the drug.

ized trial data in this population (32). Although the

Drug bioaccumulation would occur when patients

U.S. label for NOACs supports their use when CrCl is

miss their dialysis treatment, which could lead to

as low as 15 ml/min, extreme caution should be used,

serious bleeding. Finally, patients who are intolerant

as these recommendations were based mostly on PK

of warfarin may consider nonpharmacological ther-

data, and limited efficacy data support this practice.

apy for stroke prevention with percutaneous left

PK studies are only moderately reliable for quanti-

atrial appendage closure (53). Occlusion of the left

fying the relationship between NOAC dose with total

atrial appendage may be particularly attractive in

drug exposure and anticoagulation level (e.g., factor

patients with ESRD, given the observed large reduc-

Xa activity), which are used as surrogates of efficacy

tion in bleeding (54).

and safety in PK studies (51). Renal disease progres-

Last, reducing the intradialytic heparin dose merits

sion, hemodynamically related fluctuations in creat-

consideration in patients initiated on warfarin or a

inine, and acute kidney injury commonly occur in

NOAC to potentially decrease the risk for peritreat-

patients with poor kidney function. This can lead to

ment bleeding. There is little evidence to guide such

unexpected decreases in CrCl and create a mismatch

practices, but stopping or reducing the heparin dose

between NOAC dose and renal function to increase

by at least 50% appears reasonable. Patients can be

the risk for bleeding. Ultimately, randomized, pro-

rechallenged on heparin if they experience clotting

spective studies are needed to establish the efficacy

during treatment.

and safety of NOACs in the advanced CKD and dial-

MAJOR HEMORRHAGE IN THE SETTING OF

ysis population. Despite the paucity of evidence and guidelines that clearly recommend against the use of NOACs in

NOACs: ISSUES SPECIFIC TO ADVANCED CKD AND DIALYSIS PATIENTS

ESRD, observational data have shown NOAC use and

Dabigatran is the only NOAC that is very dependent

increasing; furthermore, 1 study found that many

on the kidney for elimination and is also highly

patients were prescribed higher doses of rivaroxaban

dialyzable, with 50% to 60% of plasma dabigatran

and dabigatran that were not properly reduced to

removed with a 4-h hemodialysis session (55).

account for any renal impairment (9). There was

Hemodialysis for dabigatran removal should be

evidence of higher bleeding risk associated with

considered when the severity of bleeding necessitates

dabigatran (relative risk: 1.48; 95% CI: 1.21 to 1.81)

faster drug clearance, and it would be unsafe to wait

and rivaroxaban (relative risk: 1.38; 95% CI: 1.03 to

for the drug to be completely eliminated by the native

1.83) compared with warfarin, yet there was inade-

kidneys. A patient with CrCl of 55 ml/min could clear

quate power to compare stroke outcomes between

88% of total body dabigatran in 45 h (3 half-lives;

the groups.

Table 3) through intrinsic kidney function or with 3

among

dialysis

patients

to

be

substantial

NOACs may be considered first-line therapy in pa-

serial 4-h hemodialysis sessions (1 half-life per ses-

tients who need anticoagulation with concurrent

sion). It may be reasonable to wait 45 h if bleeding is

calcific uremic arteriolopathy, a rare but devastating

controlled, but active bleeding that does not respond

calcific disease of the small vessels, where warfarin

to conventional measures merits consideration for

has been strongly associated with disease progression

more aggressive measures.

(52). Patients who are intolerant of warfarin and have

In patients with life-threatening bleeding who

histories of warfarin skin necrosis or protein C/S

need immediate dabigatran reversal, physicians may

deficiency might benefit from off-label NOAC ther-

consider

apy; however, this is highly speculative. If NOACs are

neutralize the anticoagulant effect of dabigatran

used

within 30 min in a study of 90 patients with median

in

patients

with

advanced

CKD,

close

idarucizumab,

which

was

shown

to

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

effects carried over to subgroup analyses among

T A B L E 3 Dabigatran Half-Lives by Renal Function

Creatinine Clearance, ml/min

patients with CrCl between 25 and 50 ml/min who Half-Life (h)

were enrolled in the phase III AF trials and that the

$80

13

effects are similar in other high-risk groups, such as

50–80

15

those with advanced age or low body weight.

30–49

18

<30

28

All NOACs are dependent on the kidney for elimination. Thus, the most critical development issue for NOAC use in the advanced CKD and

CrCl of 58 ml/min (56). Idarucizumab is partially dependent on the kidney for elimination (32% within the first 6 h) but is approved for patients with kidney disease. The anticoagulant effects of warfarin, apixaban, rivaroxaban, and edoxaban can be reversed by 4-factor prothrombin complex concentrate (57–59), which can replete coagulation factors much faster than fresh-frozen plasma, with less volume overload, which is favorable to patients with advanced CKD and on dialysis who are prone to volume overload (60). Last, patients with advanced CKD or on dialysis are prone

to

uremic

platelet

dysfunction.

Actively

bleeding patients can be given desmopressin. Dialysis can also partially correct the bleeding time and other in vitro tests of platelet function in approximately two-thirds of uremic patients (61–63).

dialysis population is to determine the appropriate dose reduction that optimizes the prevention of stroke yet minimizes the risk of bleeding. Well-done multidose PK studies are necessary to estimate the appropriate NOAC dose, recognizing that PK models have limited accuracy in predicting population pharmacokinetics. Randomized

controlled

trials

are

ultimately

needed to validate PK-driven hypotheses by demonstrating clinical efficacy (stroke prevention) and safety (bleeding) of NOACs versus placebo in patients with AF with advanced CKD and on dialysis. Prevention of stroke with NOACs in patients with AF would become the standard of care if such a trial indicated superiority at stroke prevention, with acceptable

bleeding

risks

referent

to

placebo.

Comparative efficacy trials between NOACs and other anticoagulant agents would be less informative. This

SUMMARY AND FUTURE DIRECTIONS

is because warfarin and aspirin have not been proved to be effective and safe for preventing stroke in

NOAC use in patients with advanced CKD and on

patients with AF with severe renal impairment, as no

dialysis is substantial and increasing, despite AHA,

phase III efficacy trials have been conducted in

ACC, and HRS and European Heart Rhythm Associa-

patients with CrCl <25 ml/min; furthermore, some

tion guidelines that endorse warfarin as the antico-

observational studies suggest that warfarin may

agulant of choice when CrCl is <30 ml/min. There are

cause harm in this population.

few randomized trial data on NOACs among patients

In conclusion, the improved safety profile of NOACs compared with warfarin for the general pop-

with advanced CKD or on dialysis. Phase 3 AF trials in the general population have

ulation raises the prospect of an improved anti-

shown that NOACs decrease both stroke and bleeding

coagulation strategy for patients with advanced renal

in comparison with warfarin, whereas patients with

impairment and an indication for anticoagulation.

advanced CKD and ESRD have a baseline increased

High-quality randomized trials should be conducted

risk for both stroke and bleeding (64,65). Because of

in these populations, given their altered drug meta-

their high risk, patients with advanced CKD and ESRD

bolism profile and high rates of both clotting and

could potentially derive great benefit from NOACs if

bleeding events.

the benefits for stroke and cardiovascular events prove to be similar to or greater than those shown in

REPRINT REQUESTS AND CORRESPONDENCE: Dr.

the general population with warfarin and NOACs,

Kevin E. Chan, Fresenius Kidney Care, 920 Winter

with

Street, Waltham, Massachusetts 02451. E-mail: kevin.

a

bleeding

risk

that

remains

acceptable.

Furthermore, it is encouraging that these favorable

[email protected].

REFERENCES 1. Desai NR, Krumme AA, Schneeweiss S, et al. Patterns of initiation of oral anticoagulants in pa-

2. Granger CB, Alexander JH, McMurray JJV, et al., for the ARISTOTLE Committees and Investigators.

3. Patel MR, Mahaffey KW, Garg J, et al., for the ROCKET AF Steering Committee. Rivaroxaban

tients with atrial fibrillation—quality and cost implications. Am J Med 2014;127:1075–82.

Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981–92.

versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883–91.

2897

2898

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

NOACs in Advanced CKD and Dialysis

4. Connolly SJ, Ezekowitz MD, Yusuf S, et al., for the RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361:1139–51. 5. Giugliano RP, Ruff CT, Braunwald E, et al., for the ENGAGE AF–TIMI 48 Investigators. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013;369:2093–104. 6. Sreedhara R, Itagaki I, Lynn B, et al. Defective platelet aggregation in uremia is transiently worsened by hemodialysis. Am J Kidney Dis 1995; 25:555–63. 7. Majerus P, Tollefson DM. Blood coagulation and anticoagulant, thrombolytic, and antiplatelet drugs. In: Brunton LL, Lazo JS, Parker KL, editors. Goodman and Gilman’s the Pharmacological Basis of Therapeutics. 11th ed. New York: McGraw-Hill, 2006:1467–88. 8. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64:2246–80. 9. Chan KE, Edelman ER, Wenger JB, et al. Dabigatran and rivaroxaban use in atrial fibrillation patients on hemodialysis. Circulation 2015;131: 972–9. 10. Reidenberg MM. The biotransformation of drugs in renal failure. Am J Med 1977;62:482–5. 11. Aronoff GR, Bennett WM, Berns JS, et al., editors. Drug Prescribing in Renal Failure: Dosing Guidelines for Adults and Children. 5th edition. Philadelphia, PA: American College of Physicians, 2007. 12. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31–41. 13. Pfeffer MA, Burdmann EA, Chen CY, et al., for the TREAT Investigators. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease. N Engl J Med 2009;361:2019–32. 14. Levey AS, Bosch JP, Lewis JB, et al., for the Modification of Diet in Renal Disease Study Group. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130: 461–70. 15. Inker LA, Perrone RD. Assessment of kidney function. UpToDate. Available at: http://www. uptodate.com/contents/assessment-of-kidneyfunction?source¼search_result&search¼estimationþofþ gfr&selectedTitle¼6w150#H20659776. Accessed April 7, 2016. 16. Holbrook AM, Pereira JA, Labiris R, et al. Systematic overview of warfarin and its drug and food interactions. Arch Intern Med 2005;165: 1095–106. 17. Limdi NA, Beasley TM, Baird MF, et al. Kidney function influences warfarin responsiveness and hemorrhagic complications. J Am Soc Nephrol 2009;20:912–21. 18. Holford NH. Clinical pharmacokinetics and pharmacodynamics of warfarin. Understanding the

Pharmacokinet

valvular atrial fibrillation. Europace 2015;17: 1467–507.

19. Pichette V, Leblond FA. Drug metabolism in chronic renal failure. Curr Drug Metab 2003;4: 91–103.

33. Stangier J, Rathgen K, Stähle H, et al. Influence of renal impairment on the pharmacokinetics and pharmacodynamics of oral dabigatran etexilate: an open-label, parallel-group, single-centre

dose-effect relationship. 1986;11:483–504.

Clin

20. Leblond F, Guévin C, Demers C, et al. Downregulation of hepatic cytochrome P450 in chronic renal failure. J Am Soc Nephrol 2001;12:326–32. 21. Dreisbach AW, Japa S, Gebrekal AB, et al. Cytochrome P4502C9 activity in end-stage renal disease. Clin Pharmacol Ther 2003;73:475–7. 22. Sarode R, Milling TJ Jr., Refaai MA, et al. Efficacy and safety of a 4-factor prothrombin complex concentrate in patients on vitamin K antagonists presenting with major bleeding: A randomized, plasma-controlled, phase IIIb study. Circulation 2013;128:1234–43. 23. Coppens M, Synhorst D, Eikelboom JW, et al. Efficacy and safety of apixaban compared with aspirin in patients who previously tried but failed treatment with vitamin K antagonists: results from the AVERROES trial. Eur Heart J 2014;35:1856–63. 24. Levey AS, Stevens LA, Schmid CH, et al., for the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150:604–12. 25. Hohnloser SH, Hijazi Z, Thomas L, et al. Efficacy of apixaban when compared with warfarin in relation to renal function in patients with atrial fibrillation: insights from the ARISTOTLE trial. Eur Heart J 2012;33:2821–30. 26. Wang X, Tirucherai G, Marbury TC, et al. Pharmacokinetics, pharmacodynamics, and safety of apixaban in subjects with end-stage renal disease on hemodialysis. J Clin Pharmacol 2016;56: 628–36. 27. Chang M, Yu Z, Shenker A, et al. Effect of renal impairment on the pharmacokinetics, pharmacodynamics, and safety of apixaban. J Clin Pharmacol 2016;56:637–45. 28. Kubitza D, Becka M, Mueck W, et al. Effects of renal impairment on the pharmacokinetics, pharmacodynamics and safety of rivaroxaban, an oral, direct factor Xa inhibitor. Br J Clin Pharmacol 2010;70:703–12. 29. Fox KAA, Piccini JP, Wojdyla D, et al. Prevention of stroke and systemic embolism with rivaroxaban compared with warfarin in patients with non-valvular atrial fibrillation and moderate renal impairment. Eur Heart J 2011;32:2387–94. 30. De Vriese AS, Caluwé R, Bailleul E, et al. Dosefinding study of rivaroxaban in hemodialysis patients. Am J Kidney Dis 2015;66:91–8. 31. Hijazi Z, Hohnloser SH, Oldgren J, et al. Efficacy and safety of dabigatran compared with warfarin in relation to baseline renal function in patients with atrial fibrillation: a RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy) trial analysis. Circulation 2014;129: 961–70. 32. Heidbuchel H, Verhamme P, Alings M, et al. Updated European Heart Rhythm Association practical guide on the use of non-vitamin K antagonist anticoagulants in patients with non-

study. Clin Pharmacokinet 2010;49:259–68. 34. Hariharan S, Madabushi R. Clinical pharmacology basis of deriving dosing recommendations for dabigatran in patients with severe renal impairment. J Clin Pharmacol 2012;52:119S–25S. 35. Beasley BN, Unger EF, Temple R. Anticoagulant options—why the FDA approved a higher but not a lower dose of dabigatran. N Engl J Med 2011; 364:1788–90. 36. Parasrampuria DA, Truitt KE. Pharmacokinetics and pharmacodynamics of edoxaban, a nonvitamin k antagonist oral anticoagulant that inhibits clotting factor Xa. Clin Pharmacokinet 2016; 55:641–55. 37. Bohula EA, Giugliano RP, Ruff CT, et al. The impact of renal function on outcomes with edoxaban in the ENGAGE AF-TIMI Trial. Circulation 2015;132:A17169. 38. Daiichi-Sankyo, Inc. Savaysa (endoxaban) NDA 206316. FDA Advisory Committee briefing document. Available at: http://www.fda.gov/ downloads/advisorycommittees/committeesmeeting materials/drugs/cardiovascularandrenaldrugsadvisory committee/ucm420705.pdf. Accessed April 7, 2016. 39. Koretsune Y, Yamashita T, Kimura T, et al. Short-term safety and plasma concentrations of edoxaban in Japanese patients with non-valvular atrial fibrillation and severe renal impairment. Circ J 2015;79:1486–95. 40. Parasrampuria DA, Marbury T, Matsushima N, et al. Pharmacokinetics, safety, and tolerability of edoxaban in end-stage renal disease subjects undergoing haemodialysis. Thromb Haemost 2015;113:719–27. 41. Chan KE, Lazarus JM, Thadhani R, et al. Warfarin use associates with increased risk for stroke in hemodialysis patients with atrial fibrillation. J Am Soc Nephrol 2009;20:2223–33. 42. Shah M, Avgil Tsadok M, Jackevicius CA, et al. Warfarin use and the risk for stroke and bleeding in patients with atrial fibrillation undergoing dialysis. Circulation 2014;129:1196–203. 43. Winkelmayer WC, Liu J, Setoguchi S, et al. Effectiveness and safety of warfarin initiation in older hemodialysis patients with incident atrial fibrillation. Clin J Am Soc Nephrol 2011;6:2662–8. 44. Li J, Wang L, Hu J, et al. Warfarin use and the risks of stroke and bleeding in hemodialysis patients with atrial fibrillation: a systematic review and a meta-analysis. Nutr Metab Cardiovasc Dis 2015;25:706–13. 45. Olesen JB, Lip GYH, Kamper AL, et al. Stroke and bleeding in atrial fibrillation with chronic kidney disease. N Engl J Med 2012;367:625–35. 46. Herzog CA, Asinger RW, Berger AK, et al. Cardiovascular disease in chronic kidney disease. A clinical update from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2011;80: 572–86.

Chan et al.

JACC VOL. 67, NO. 24, 2016 JUNE 21, 2016:2888–99

47. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857–67. 48. van Walraven C, Hart RG, Singer DE, et al. Oral anticoagulants vs aspirin in nonvalvular atrial fibrillation: an individual patient meta-analysis. JAMA 2002;288:2441–8. 49. Keith DS, Nichols GA, Gullion CM, et al. Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 2004;164:659–63. 50. U.S. Renal Data System, USRDS 2009 Annual Data Report Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. Bethesda, Maryland: National Institute of Diabetes and Digestive and Kidney Diseases, 2009. 51. Patel MR, Washam JB. Edoxaban and the need for outcomes-based NOAC dosing. Lancet 2015; 385:2232–3. 52. Nigwekar SU, Kroshinsky D, Nazarian RM, et al. Calciphylaxis: risk factors, diagnosis, and treatment. Am J Kidney Dis 2015;66:133–46. 53. Reddy VY, Möbius-Winkler S, Miller MA, et al. Left atrial appendage closure with the Watchman device in patients with a contraindication for oral anticoagulation: the ASAP study (ASA Plavix Feasibility Study With Watchman Left Atrial Appendage Closure Technology). J Am Coll Cardiol 2013;61:2551–6.

NOACs in Advanced CKD and Dialysis

54. Holmes DR Jr., Doshi SK, Kar S, et al. Left atrial appendage closure as an alternative to warfarin for stroke prevention in atrial fibrillation: a patient-level meta-analysis. J Am Coll Cardiol 2015;65:2614–23. 55. Khadzhynov D, Wagner F, Formella S, et al. Effective elimination of dabigatran by haemodialysis. A phase I single-centre study in patients with end-stage renal disease. Thromb Haemost 2013; 109:596–605. 56. Pollack CV Jr., Reilly PA, Eikelboom J, et al. Idarucizumab for dabigatran reversal. N Engl J Med 2015;373:511–20. 57. Zahir H, Brown KS, Vandell AG, et al. Edoxaban effects on bleeding following punch biopsy and reversal by a 4-factor prothrombin complex concentrate. Circulation 2015;131:82–90. 58. Eerenberg

ES,

Kamphuisen

PW,

Sijpkens MK, et al. Reversal of rivaroxaban and dabigatran by prothrombin complex concentrate: a randomized, placebo-controlled, crossover study in healthy subjects. Circulation 2011; 124:1573–9. 59. Baumann Kreuziger LM, Keenan JC, Morton CT, et al. Management of the bleeding patient receiving new oral anticoagulants: a role for prothrombin complex concentrates. Biomed

US/Kcentra/EN/Kcentra-Prescribing-Information.pdf. Accessed April 7, 2016. 61. Stewart JH, Castaldi PA. Uraemic bleeding: a reversible platelet defect corrected by dialysis. Q J Med 1967;36:409–23. 62. Nenci GG, Berrettini M, Agnelli G, et al. Effect of peritoneal dialysis, haemodialysis and kidney transplantation on blood platelet function. I. Platelet aggregation by ADP and epinephrine. Nephron 1979;23:287–92. 63. Lindsay RM, Friesen M, Aronstam A, et al. Improvement of platelet function by increased frequency of hemodialysis. Clin Nephrol 1978;10: 67–70. 64. Elliott MJ, Zimmerman D, Holden RM. Warfarin anticoagulation in hemodialysis patients: a systematic review of bleeding rates. Am J Kidney Dis 2007;50:433–40. 65. McMahan DA, Smith DM, Carey MA, et al. Risk of major hemorrhage for outpatients treated with warfarin. J Gen Intern Med 1998;13:311–6. 66. Gonzalez-Quesada CJ, Giugliano RP. Comparison of the phase III clinical trial designs of novel oral anticoagulants versus warfarin for the treatment of nonvalvular atrial fibrillation: implications for clinical practice. Am J Cardiovasc Drugs 2014; 14:111–27.

Res Int 2014;2014:583794. 60. CSL Behring. Kcentra prescribing information. Available at: http://labeling.cslbehring.com/PI/

KEY WORDS apixaban, dabigatran, edoxaban, renal dialysis, rivaroxaban, stroke

2899