The optimal dose of intravenous tranexamic acid for reducing blood loss in spinal surgery: a network meta-analysis

This study aims to evaluate the optimal dose of intravenous tranexamic acid (TXA) for reducing blood loss in spinal surgery.

A systematic search was conducted in the PubMed, Embase, Cochrane Library database from inception until November 2023. Randomized controlled trials (RCTs) incorporating diverse TXA dosing regimens for spinal surgery were included. The surface under the cumulative ranking curve (SUCRA) analysis was employed to determine ranking order. R software with gemtc package was used for all analyses, with a significance threshold set at P < 0.05.

Twenty-four RCTs were considered eligible and finally included. All TXA treatments demonstrated superior efficacy compared to the placebo, with statistically significant differences (P < 0.05). SUCRA values indicated that Treatment I (100 mg/kg + 10 mg.kg/h) claimed the top rank (SUCRA, 80.3%), followed by Treatment F (15 mg/kg + 2 mg.kg/h) in second place (SUCRA, 76.7%), and Treatment E (10 mg/kg + 2 mg.kg/h) in third place (SUCRA, 75.2%). Conversely, the placebo ranked the lowest (SUCRA, 0.3%). Additionally, Treatment I (100 mg/kg + 10 mg.kg/h) held the top rank (SUCRA, 95.6%), followed by Treatment N (30 mg/kg + 10 mg.kg/h) in second place (SUCRA, 81.0%), and Treatment K (15 mg/kg + 6 mg.kg/h) in third place (SUCRA, 74.8%). Importantly, no statistically significant differences were observed between any TXA treatments and the placebo concerning the occurrence of deep vein thrombosis (DVT) (P > 0.05).

This network meta-analysis underscores that intravenous TXA is associated with decreased overall blood loss in multilevel spine surgery. Notably, the highest dose in this network meta-analysis (100 mg/kg + 10 mg.kg/h) emerged as the only regimen demonstrating significant benefits in pairwise comparisons with other TXA doses. Although this regimen did not significantly increase DVT risk, careful consideration of safety data for higher doses remains essential.