Blood loss estimation during posterior spinal fusion for adolescent idiopathic scoliosis.

PURPOSE: Blood loss (BL) during posterior spinal fusion for adolescent idiopathic scoliosis (AIS) may be estimated using a variety of unproven techniques. Patient care and research on BL are likely impacted by a lack of standardization. A novel FDA-approved blood volume (BV) analysis system (BVA-100 Blood Volume Analyzer) allows rapid processing with > 97% accuracy. The purpose of this study was to investigate common methods for BL estimation. METHODS: BV assessment was performed with the BVA-100. After obtaining a baseline sample of 5 mL of blood, 1 mL of I-131-labeled albumin was injected intravenously over 1 min. Five milliliter blood samples were then collected at 12, 18, 24, 30, and 36 min post-injection. Intravenous fluid was minimized to maintain euvolemia. Salvaged blood was not administered during surgery. BL was estimated using several common techniques and compared to the BV measurements provided by the BVA-100 (BVABL). RESULTS: Thirty AIS patients were prospectively enrolled with major curves of 54° and underwent fusions of 10 levels. BL based on the BVA-100 (BVABL) was 519.2 [IQR 322.9, 886.2] mL. Previously published formulas all failed to approximate BVABL. Multiplying the cell saver volume return by 3 (CS3) approximates BVABL well with a Spearman correlation coefficient and ICC of 0.80 and 0.72, respectively. An extrapolated cell salvage-based estimator also showed high intraclass correlation coefficient (ICC) and Spearman coefficients with less bias than CS3. CONCLUSION: Published formulaic approaches do not approximate true blood loss. Multiplying the cell saver volume by 3 or using the cell salvage-based estimator had the highest correlation coefficient and ICC. LEVEL OF EVIDENCE: Prospective cohort Level 2.