PT - JOURNAL ARTICLE AU - Sheweidin Aziz AU - Krishan Almeida AU - Grahame Taylor TI - How should we manage hip fracture patients on direct oral anticoagulants? AID - 10.1136/bmjebm-2019-111317 DP - 2020 Apr 24 TA - BMJ Evidence-Based Medicine PG - bmjebm-2019-111317 4099 - http://ebm.bmj.com/content/early/2020/04/24/bmjebm-2019-111317.short 4100 - http://ebm.bmj.com/content/early/2020/04/24/bmjebm-2019-111317.full AB - Objective Currently, there are no national protocols in place for managing hip fracture patients on direct oral anticoagulants (DOACs). Hence, various local management protocols exist. We compared three different local protocols and a control group to assess blood loss and time delay to theatre.Methods Sequential data were collected for 120 hip fracture patients in four groups; wait 24 hours from last dose, wait 48 hours, perform DOAC levels and control.Results DOAC use in our hip fracture patients was 14%. Median haemoglobin (Hb) drop between the three protocol groups showed no significant difference (13.5, 21.5 and 16.0 g/L) (Kruskal-Wallis, p=0.9). Median Hb drop in the control group was 16.0 g/L versus 17.5 g/L in the protocol groups combined (Mann Whitney-U, p=0.7). Average Hb drop in the control group was 19.2 g/L and in the protocol groups was 22.1 g/L; a 15% greater blood loss with DOACs. The frequency distribution of blood loss was different between the control and protocol groups, but not between the protocol groups. The highest Hb drop in the control group was 3.4 g/L, while in the protocol groups, it was 7.8 g/L. Median Hb on arrival to hospital was higher in the control group (124 g/L) compared with the three protocol groups (87 g/L) (t-test p<0.0001). Transfusion rates of up to 40% were observed within the DOAC groups compared with zero in the control group.Median time to theatre between the three protocol groups was significantly different at 17.5, 53.3 and 22.5 hours, respectively (Kruskal-Wallis, p<0.0001).Conclusion DOACs caused increased bleeding and delays to theatre in hip fracture patients, however the largest Hb difference was already apparent on arrival. Subsequent blood loss was minimal on average; a few patients bled heavily. Prolonged waiting made no significant difference to blood loss, but caused delay to theatre leading to financial losses from best practice tariff.