Transfusion practices in intensive care units: An Australian and New Zealand point prevalence study

Objective

To describe current transfusion practices in intensive care units (ICUs) in Australia and New Zealand, compare them against national guidelines, and describe how viscoelastic haemostatic assays (VHAs) are used in guiding transfusion decisions.

Design, setting and participants

Prospective, multicentre, binational point-prevalence study. All adult patients admitted to participating ICUs on a single day in 2021.

Main outcome measures

Transfusion types, amounts, clinical reasons, and triggers; use of anti-platelet medications, anti-coagulation, and VHA.

Results

Of 712 adult patients in 51 ICUs, 71 (10%) patients received a transfusion during the 24hr period of observation. Compared to patients not transfused, these patients had higher Acute Physiology and Chronic Health Evaluation II scores (19 versus 17, p = 0.02), a greater proportion were mechanically ventilated (49.3% versus 37.3%, p < 0.05), and more had systemic inflammatory response syndrome (70.4% versus 51.3%, p < 0.01). Overall, 63 (8.8%) patients received red blood cell (RBC) transfusions, 10 (1.4%) patients received platelet transfusions, 6 (0.8%) patients received fresh frozen plasma (FFP), and 5 (0.7%) patients received cryoprecipitate. VHA was available in 42 (82.4%) sites but only used in 6.6% of transfusion episodes when available. Alignment with guidelines was found for 98.6% of RBC transfusions, but only 61.6% for platelet, 28.6% for FFP, and 20% for cryoprecipitate transfusions.

Conclusions

Non-RBC transfusion decisions are often not aligned with guidelines and VHA is commonly available but rarely used to guide transfusions. Better evidence to guide transfusions in ICUs is needed.