Effectiveness of closed blood-sampling devices in critically Ill adults: A feasibility trial

Abstract

Background: Anaemia in critically ill patients is often worsened by diagnostic blood sampling. Closed blood sampling devices(CBSDs) may reduce iatrogenic blood loss and transfusion needs. However, robust evidence of their effectiveness and feasibilityin the intensive care unit (ICU) setting is lacking.Aim: To assess the feasibility of conducting a multicentre randomised controlled trial (RCT) evaluating the effectiveness ofCBSDs in reducing transfusion requirements in critically ill adults.Study Design: A two-centre, open-label, feasibility RCT was conducted in two Spanish university hospitals between November2024 and March 2025. Patients with an ICU stay of 24 h who were expected to have an arterial catheter for at least 72 h longerwere randomised to either a CBSD group (intervention) or standard practice (control). Primary outcome was feasibility (recruitmentrate, intervention fidelity and dropout). Secondary outcomes included transfusion rates, discard volume and catheter-relatedadverse events. Data were collected over a maximum of 21 days or until ICU discharge or catheter removal. As this was afeasibility study with a small sample size, no statistical inference analyses were performed.Results: Of 678 patients screened, 9.3% (n = 63) were eligible, and of those eligible, 76.2% (n = 48) were enrolled. After exclusionsand losses, 31 patients were analysed (12 intervention, 19 control). Low eligibility was mainly due to short catheter dwell timesand advanced monitoring needs with a Flotrac system. Intervention fidelity was suboptimal in four patients in the interventiongroup because one of the ICUs routinely used a venous route for blood gas sampling, despite patients having the CBSD in anarterial line. Thirteen patients (27.1%) were lost to follow-up for the same reasons as the low eligibility findings. There were nomissing data (0%). Transfusion rates per 100 catheter days were lower in the intervention group (5.2 vs. 15.6). Discard volume per100 catheter days was substantially reduced (53.1 mL vs. 970.7 mL). No catheter-related bacteraemias occurred.Conclusions: A full-scale RCT evaluating CBSDs in ICU patients is feasible with protocol modifications, including broader siteparticipation and improved device integration with advanced monitoring systems. Preliminary data suggest that CBSDs mayreduce transfusion requirements and blood loss in critically ill patients.Relevance to Clinical Practice: Nurses’ involvement in the prevention of iatrogenic anaemia is key. CBSDs appear to be afeasible strategy for reducing blood loss associated with blood tests, with preliminary results linking their use to a decrease inthe need for transfusions. If this is confirmed in the final clinical trial, it would mean greater safety for patients by avoidingcomplications linked to transfusions and would contribute to environmental sustainability by reducing the production of carbondioxide associated with the process of collecting, storing and distributing red blood cells.