Background Non-vitamin K antagonist oral anticoagulants provide a valuable alternative option to warfarin for thromboprophylaxis for non-valvular atrial fibrillation. NICE has recently released a national consensus statement which will guide local protocol/guideline development. Currently, their use is often governed by physician preference combined with patient factors. We explored the pattern of prescribing in the pre-guideline era.

Method Retrospective analysis of admissions from June–October 2014. Student T-test and Fischer's exact test with Bonferroni corrections were used to compare demographics between groups. A creatinine clearance of <60 umol/L defined renal impairment.

Results 61 patients, 37 female, mean age 80.8 years were audited. 35 were prescribed rivoroxaban 20 mg, 7 rivoroxaban 15 mg, 11 apixaban 5 mg and 8 apixaban 2.5 mg. None were prescribed dabigatran. Age, CHADS2 and HAS-BLED scores did not show statistically significant variability between subgroups. Renal impairment was signficantly commoner in the apixaban 2.5 mg group compared to the 5mg group (p<0.001) and the rivoroxaban 15mg compared to the 20mg group (P<0.01). There was no difference between the rivoroxaban and apixaban groups.

Conclusion Currently, renal impairment is the main determinant for thromboprophylaxis choice. Local guidelines have been created with physician education underway and re-audit planned in 3 months. The authors will present the full data.