Background: Aging is associated with increased prevalence of atrial fibrillation and thrombophlebitis. Anticoagulation has been shown to be effective in preventing thromboembolic events in AF patients and treatment of thrombophlebitis. Non-vitamin k antagonist oral anticoagulants (NOACs) have been shown to be non-inferior or even superior to warfarin for long-term stroke prevention and thrombophlebitis treatment. In Meta-analysis of clinical trials, NOACs was not associated with excess bleeding in the elderly participant. However, studies showed that elderly patients are undertreated by anticoagulation due to concerns about the higher risk of bleeding and fall in elderly. Few studies have provided real-life data on the outcome of elderly patients on NOACs especially those presented with fall while have been anticoagulated.

Methods: We performed a systematic search of hospital electronic medical records using ICD coding 9 and 10 for accidental fall and chronic anticoagulation. We selected elderly patients (> 65 years old) who came to the emergency department with the complaint of fall and have been treated with NOACs or warfarin prior to admission from August 2014 to September 2017. Patients who transferred to outside facility were excluded. CHADS2 score, BMI, albumin level, nursing home residence, major bleeding, intracranial bleeding, blood product transfusion, reversal factor administration, length of stay, ICU admission, discharge to institutional care, inpatient mortality, discontinuation of anticoagulation at discharge were collected.

Results: 144 patients met the criteria for the study, (warfarin = 116 and NOACs = 28). The average age was 81.2 ± 8.5 years, 52.1% were female, 23.6% were nursing home residents and 52.77% had previous fall. Patients were taking average 9.5 ± 4.6 medications prior to admission and 37.5% were taking at least one sedative medication. CHADS2 score greater than 3 was observed in 57.45% and 74.07% of patients. (Warfarin vs NOACs, p value= 0.18). INR was in the therapeutic range in 38.59% of patients taking warfarin. Major bleeding, intracranial bleeding, ICU admission, death and length of stay at the hospital was 27.6%, 17.8%, 23%, 3.2% and 5.16± 3.42 days in the warfarin group and 7.7%, 10.7%, 25%, 0% and 4.11 ± 2.13 days in the NOACS group, respectively (p value = 0.34, 0.70, 0.81, 0.59 and 0.1). Length of stay in ICU was longer in the warfarin group (2.8 ± 2.7 days vs 1.4 ± 0.5 days; p value = 0.017). Anticoagulation was continued in 84.7%of patients upon discharge.

Conclusions: NOACs did not have more adverse effects on the studied patients in comparison to warfarin. The analysis even showed a shorter ICU stay in patients taking NOACs. The significant number of patients on warfarin were not properly anticoagulated. Factors contributing to fall in elderly like polypharmacy and sedative medications can be optimized to minimize fall and subsequent outcome especially in patients who will be anticoagulated. Larger studies are needed to further evaluation of differences between NOACs and warfarin in elderly patients prone to fall.