Tranexamic acid use decreases transfusion rate in children with cerebral palsy undergoing proximal femoral varus derotational osteotomy

Previous studies demonstrated the safety of tranexamic acid (TXA) use in cerebral palsy (CP) patients undergoing proximal femoral varus derotational osteotomy (VDRO), but were underpowered to determine if TXA alters transfusion rates or estimated blood loss (EBL). The purpose of this study was to investigate if intraoperative TXA administration alters transfusion rates or EBL in patients with CP undergoing VDRO surgery. We conducted a retrospective review of 390 patients with CP who underwent VDRO surgery between January 2004 and August 2019 at a single institution. Patients without sufficient clinical data and patients with preexisting bleeding or coagulation disorders were excluded. Patients were divided into 2 groups: those who received intraoperative TXA and those who did not. Out of 390 patients (mean age 9.4 +/- 3.8 years), 80 received intravenous TXA (TXA group) and 310 did not (No-TXA group). There was no difference in mean weight at surgery (P = .25), Gross Motor Function Classification System level (P = .99), American Society of Anesthesiologist classification (P = .50), preoperative feeding status (P = .16), operative time (P = .91), or number of procedures performed (P = .12) between the groups. The overall transfusion rate was lower in the TXA group (13.8%; 11/80) than the No-TXA group (25.2%; 78/310) (P = .04), as was the postoperative transfusion rate (7.5%; 6/80 in the TXA group vs 18.4%; 57/310 in the No-TXA group) (P = .02). The intraoperative transfusion rate was similar for the 2 groups (TXA: 7.5%; 6/80 vs No-TXA: 10.3%; 32/310; P = .53). The EBL was slightly lower in the TXA group, although this was not significant (TXA: 142.9 +/- 113.1 mL vs No-TXA: 177.4 +/- 169.1 mL; P = .09). The standard deviation for EBL was greater in the No-TXA group due to more high EBL outliers. The percentage of blood loss based on weight was similar between the groups (TXA: 9.2% vs No-TXA: 10.1%; P = .40). The number needed to treat (NNT) with TXA to avoid one peri-operative blood transfusion in this series was 9. The use of intraoperative TXA in patients with CP undergoing VDRO surgery lowers overall and postoperative transfusion rates. Level of evidence: III, Retrospective Comparative Study.