The Hematologist

November-December 2017, Volume 14, Issue 6

Use of Direct Oral Anticoagulants in Patients on Hemodialysis

Stephan Moll, MD Professor of Medicine, Division of Hematology/Oncology
University of North Caroline School of Medicine, Chapel Hill, NC

Published on: October 11, 2017

Mavrakanas TA, Samer CF, Nessim SJ, et al. Apixaban pharmacokinetics at steady state in hemodialysis patients. J Am Soc Nephrol. 2017;28:2241-2248.

Four direct oral anticoagulants (DOACs) are presently approved by the U.S. Food and Drug Administration (FDA) for therapeutic anticoagulation in atrial fibrillation and venous thromboembolism. All are, to varying degrees, renally cleared. Dose reduction is recommended for dabigatran, edoxaban, and rivaroxaban in patients with moderate renal impairment, and avoidance altogether is recommended in patients with severe renal impairment and on dialysis. Only apixaban can be used in patients with severely decreased renal function and those on hemodialysis. The FDA-approved prescription information states that full dose apixaban (5 mg [bid]) can be used in such patients, unless at least two of the following characteristics apply: patient age is 80 years or older, body weight is 60 kg or less, and serum creatinine is 1.5 mg/dL or higher, in which case dose reduction to 2.5 mg bid is indicated.1 However, the prescription information acknowledges that “clinical efficacy and safety studies with Eliquis did not enroll patients with end-stage renal disease (ESRD) on dialysis”.1

Many clinicians have been hesitant to use apixaban in patients with severely impaired renal function or on hemodialysis due to limited available data and the fact that small pharmacokinetics/pharmacodynamics (PK/PD) studies have shown a higher apixaban drug exposure in patients with renal failure treated with even a single dose of standard-dose apixaban compared to patients with normal renal function.2,3 The present article is the first study assessing multiple-dose administration of apixaban in patients on hemodialysis and provides important additional data on the PK/PD of apixaban in patients on hemodialysis.

The purpose of the present study was to determine the PK/PD of 2.5 mg bid and of 5 mg bid apixaban in patients on hemodialysis. In part A of this study, seven patients received 2.5 mg bid of apixaban for eight days. Blood samples were drawn immediately before dosing and at 11 time points during the 24 hours after dosing. Significant accumulation of drug between days 1 and 8 was demonstrated, with the area under the curve (AUC) increasing two to 5.4 times (p<0.001), reflecting a markedly higher exposure at steady state compared with exposure after a single first dose. Furthermore, the study showed that the trough (Cmin) and peak (Cmax) drug levels significantly increased between days 1 and 8, highlighting that previous single-dose PK/PD studies need to be interpreted with caution.2,3 In part B of the study, on day 9 after a 2.5 mg dose, apixaban levels were monitored hourly during hemodialysis in six of the patients. Only 4 percent of the drug was removed by hemodialysis, confirming that the procedure has no appreciable effects on apixaban plasma concentration. This is not surprising, given that apixaban is highly protein bound. In part C of this study, after a five-day wash-out period, five patients received 5 mg apixaban bid for eight days and multiple blood draws allowed determination of PK/PD for this dosing regimen. Again, pronounced elevations in AUC, Cmin and Cmax were observed between day 1 and day 8. Most noteworthy is that the AUC and the peak and the trough apixaban levels at steady state (day 8) with the dose of 5 mg bid were markedly higher than in patients with normal renal function on 5 mg bid. Furthermore, the PK/PD data with the dose of 2.5 mg bid was comparable with those of standard dose (5 mg bid) in patients with preserved renal function.

The study’s three major findings were: 1) Significant accumulation of apixaban drug was observed between day 1 and 8; 2) apixaban levels at steady state with the 2.5 mg bid dose were close to the expected median values in patients who qualified for a reduced apixaban dose in previously conducted randomized, controlled trials and above the 10th percentile of the expected values with the full dose (5 mg bid); and 3) supratherapeutic drug exposure and trough levels with the 5 mg bid dose were demonstrated (i.e. levels above the 90th percentile of the predicted levels for this dose).

This study has significant limitations. Only a few patients on hemodialysis were studied (7 patients in part A, 6 in part B, 5 in part C). Additionally, there were large interindividual difference in the PK data, as reflected by wide coefficients of variation for AUC, Cmax and Cmin, which limits the confidence in generalizability of the data presented and the conclusions. For instance, several patients on the 5-mg bid dose had PK data that were within the expected ranges of patients with preserved renal function on the same dose. Finally, although published in the Journal of the American Society of Nephrology, a reputable journal, there are some data discrepancies between the article’s tables and the abstract, inconsistencies with how data are presented, and a lack of clarity as to whether data presented show mean or median values, 10th to 90th or fifth to 95th percentiles, making data interpretation difficult and somewhat limiting my confidence in the data presented.

Given the paucity of PK/PD and safety and efficacy data of DOACs in patients with severe renal failure or on hemodialysis, my preference is to see such patients on warfarin if they require anticoagulation. However, if warfarin is not a treatment option, then I feel fairly comfortable using apixaban. I would not use any of the other DOACs in these patients given the current lack of published data with DOACs other than apixaban. In respect to apixaban dose, in most patients I would use a 2.5 mg bid dose, based on published PK/PD data, particularly the present study; I would stay away from using the 5 mg bid dosing regimen that the FDA-approved prescribing information suggests, unless the patient is particularly prothrombotic (high CHADS-2 score, recent venous thromboembolism).1-3 I would feel particularly strongly about using the lower dose in the elderly, the patients with a lower body weight, and those at increased risk for bleeding. Finally, I would make sure that the patient is well informed and involved in the decision-making process, as anticoagulation treatment decisions in these patients are not based on a wealth of data.

Renal Excretion25%80%35%33%
CrCl > 955 mg bid

2.5 bid if ≥ 2 of following characteristics: age ≥ 80 years, weight ≤ 60 kg, creatinine ≥ 1.5 mg/dL
10 mg bid for 1 week, then 5 mg bid.

No reference on dosing in renal impairment1
150 mg bid150 mg bidDo not use60 mg qd20 mg qd15 mg bid for 21 days, then 20 mg qd
CrCl > 50-9560 mg qd
CrCl 30-5030 mg qd30 mg qd15 mg qd
CrCl 15-3075 mg bidavoidavoid
CrCl < 15avoidavoidavoidavoid

Abbreviations: A Fib = atrial fibrillation; VTE = venous thromboembolism; CrCl = Creatinine Clearance; qd = once daily; bid = twice daily.



  1. Bristol-Myers Squibb Company. ELIQUIS (apixaban). Last revised July 2016.
  2. 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;56:628-636.
  3. Chang M, Yu Z, Shenker A, et al. Effect of renal impairment on the pharmacokinetics, pharmacodynamics, and safety of apixaban. J Clin Pharmacol. 2016;56:637-645.

Conflict of Interests

Dr. Moll has consulted for Boehringer-Ingelheim, Janssen Pharmaceuticals, and Portola. back to top