Background: Blood management strategies in total hip arthroplasty (THA) are essential in reducing intraoperative blood loss, blood transfusion and associated complications. This study investigates whether using intraoperative cell-salvage (ICS) with tranexamic acid (TXA) has additional effects on blood loss and allogeneic transfusion in primary THA. Additionally, we evaluated the financial impact of using ICS on our institution.
Methods: Using an institutional database, 1171 cases of primary unilateral THA performed between May 2015 and January 2016 were identified. Subjects were separated into those who received only TXA (n = 323) and those who received TXA and ICS (n = 848). Calculated blood loss and post-operative blood transfusions were assessed using logistic regression. Drop in hematocrit was assessed using linear regression. Multivariable models adjusted for intraoperative blood transfusions, pre-operative autologous blood donation, anticoagulation medications, sex, and body mass index. Pricing data was used to calculate the costs associated with these interventions.
Results: The likelihood of post-operative allogeneic blood transfusion was similar for the combined group relative to the TXA group (OR = 0.63; 95% CI: 0.26, 1.54), as was the likelihood of any post-operative blood transfusion (OR = 1.13; 95% CI: 0.63, 2.01). There was no correlative relationship between use of ICS and hematocrit drop when accounting for baseline hematocrit (R2 = 0.118). Factoring in rental, service fees, and disposable equipment, the utilization of ICS added $146 to each case, resulting in a gross expenditure of over $123,000 during the study period.
Conclusions: The combination of ICS with TXA for primary unilateral THA did not improve blood loss or transfusion outcomes compared to TXA alone. As there was no observed clinical benefit to combined treatment, additional costs associated with routine usage of ICS may not be justifiable. Our institution would have reduced expenditures for blood loss management products by 85% during the study period if all patients had only received TXA.
