Edinburgh Research Explorer

A restrictive versus liberal transfusion strategy to prevent myocardial injury in patients undergoing surgery for fractured neck of femur: a feasibility randomised controlled trial (RESULT-NOF)

Research output: Contribution to journalArticle

Original languageEnglish
JournalBritish Journal of Anaesthesia
Early online date20 Jul 2020
Publication statusE-pub ahead of print - 20 Jul 2020


Background: Optimum transfusion strategy in patients with fractured neck of femur (NOF) is uncertain, particularly if there is coexisting cardiovascular disease.
Methods: Prospective randomised feasibility trial of two transfusion strategies in a single hospital. We randomised patients undergoing surgery for NOF to a restrictive (haemoglobin 70-90 g.L-1) or liberal (Haemoglobin 90-110 g.L-1) transfusion strategy for duration of hospitalisation. Feasibility outcomes included: enrolment rate, protocol compliance, difference in haemoglobin and blood exposure. The primary clinical outcome was myocardial injury using troponin. Secondary outcomes included major adverse cardiac events, postoperative complications, duration of hospitalisation, mortality, quality of life.
Results: 200/907 (22% of eligible patients) were enrolled. 62 (31%) decreased haemoglobin to 90 g.L-1 or less and were exposed to the intervention. Overall protocol compliance was 81% in the liberal group and 64% in the restrictive group. Haemoglobin concentrations were similar preoperatively and at postoperative day (POD) 1 but lower in the restrictive group at POD 2 (mean difference (MD) 7.0 g.L-1 [95% CI 1.6, 12.4]). Lowest haemoglobin within 30 days/before discharge was lower in the restrictive group (MD 5.3 g.L-1 [95% CI 1.7, 9.0]). 58% of patients in the restrictive group received no transfusion compared with 4% in the liberal group (difference in proportion 54.5% [95% CI 36.8%, 72.2%]). The proportion with the primary clinical outcome was 14/26 (54%, liberal) vs 24/34 (71%, restrictive), difference in proportion -16.7% [95% CI -41.3%, 7.8%] p=0.182.
Conclusion: A study of two transfusion strategies with a clinically relevant cardiac outcome is feasible.

ID: 154166577