Low-dose apixaban on dialysis: no major bleeds, but more strokes, in a small series

Design

Retrospective single-centre service evaluation

Population

88 patients on renal replacement therapy (33 apixaban 2.5 mg twice daily, 55 vitamin K antagonist)

Primary outcome

Major bleeding and thrombotic events

Effect

Major bleeds 0 vs 4 (p=0.10); stroke 7.1 vs 1.1 per 100 patient-years (p=0.07)

Outside the United States, apixaban is not licensed for people on dialysis. This retrospective service evaluation at a UK tertiary anticoagulation clinic compared 33 patients on renal replacement therapy taking apixaban 2.5 mg twice daily with 55 taking a vitamin K antagonist.

There were no major bleeds with apixaban and 4 (7.3%) with vitamin K antagonists (p=0.10). Strokes were 3 with apixaban and 1 with vitamin K antagonists (7.1 vs 1.1 per 100 patient-years, p=0.07), and only one stroke was related to atrial fibrillation. Of 85 trough anti-Xa levels on apixaban, 84 were within the upper limit of the general population range.

Neither difference was statistically significant, and the numbers are too small to judge either drug. The anti-Xa data suggest trough levels on 2.5 mg twice daily stayed below the upper limit of the general-population range in these patients.

• Consider apixaban 2.5 mg twice daily as a second-line option on dialysis when a vitamin K antagonist is unsuitable, recognising it is off-label outside the US.
• Measure trough anti-Xa levels where available, calibrated for apixaban.
• Weigh the higher stroke count on apixaban in this series, though it was not significant.
• Revisit whether anticoagulation is still indicated at every review; both bleeding and stroke risks are high on dialysis.

Why it matters

Clinicians already use apixaban off-label on dialysis, and this adds anti-Xa data suggesting trough levels stayed below the upper limit of the general-population range.

Don't overread it

The small, non-randomised series cannot show that apixaban is safer or less effective than a vitamin K antagonist.

The statistics, in plain English

With 88 patients and 8 major events in total, p values of 0.10 and 0.07 mean neither difference can be distinguished from chance. The groups were not randomised and may differ in ways that affected outcomes.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.