Evaluation of point-of-care haemoglobin measurement accuracy in surgery (PREMISE) and implications for transfusion practice: a prospective cohort study.

Summary

In a two-centre prospective cohort (1139 patients; 1735 intraoperative samples), none of the point-of-care haemoglobin devices achieved agreement within ±4 g/L of laboratory values in the clinically crucial 60–100 g/L range. These findings question reliance on point-of-care haemoglobin for transfusion decisions during noncardiac surgery.

Key Findings

• Prospective two-centre cohort with 1139 patients and 1735 intraoperative samples assessed three point-of-care haemoglobin devices.
• No device achieved limits of agreement within ±4 g/L compared to laboratory haemoglobin, including in 680 samples with lab Hgb <100 g/L.
• Findings specifically challenge device accuracy in the 60–100 g/L range where transfusion decisions are most sensitive.

Clinical Implications

Avoid sole reliance on point-of-care haemoglobin for transfusion decisions near thresholds; confirm with laboratory measurements or use decision strategies that incorporate uncertainty. Reassess institutional patient blood management algorithms and device procurement.

Limitations

• Did not evaluate downstream patient outcomes based on transfusion decisions altered by device readings.
• Potential variation by sampling site and device-specific performance was not fully detailed in the abstract.

Future Directions

Conduct outcome-focused trials integrating device uncertainty into transfusion algorithms; evaluate device-specific calibration or hybrid strategies combining point-of-care and lab measurements.