Objective To test the feasibility of conducting a randomised controlled trial to evaluate the impact of a closed-loop blood sampling system and blood conservation bundle.

Methods Single site, parallel group, pilot randomised control trial comparing open system sampling to closed system sampling and conservation bundle aligned with national guidelines. Randomisation sequence was generated by an independent statistician and allocation concealment maintained via sealed opaque envelopes.

The study setting was the general intensive care unit of a major metropolitan public hospital in Queensland, Australia. Participants were ≥ 18 years who had an arterial catheter inserted in intensive care.

Main outcome measures included trial feasibility, blood sample loss, haematocrit (HCT) change, and packed red blood cell transfusion use.

Results Eighty patients were randomised (n = 39 open group, n = 41 closed group). Characteristics in each group were equal at baseline with overall median age 60 years (IQR 48.6–70.4), 58 % male, and median APACHE II score 16 (IQR 11–22). The proportion of patients eligible was 29 % and missed eligible was 65 %. Otherwise, feasibility criteria were met with proportion of eligible patients agreeing to enrolment 99 %, 100 % of patients receiving allocated treatment; only 1 % of data missing. Analysis demonstrated a significant reduction in mean daily blood sample losses (open 32.7 (SD 1.58) mL vs closed 15.5 (SD 5.79) mL, t = −8.454, df = 78, p < 0.001).

Conclusions A large, multi-site trial is feasible with enhanced eligibility criteria, increased recruitment support. The intervention reduced daily blood sample volumes and transfusion use. Further trials are required to provide both effectiveness and implementation outcomes.