Avoiding Femoral Canal Instrumentation in Computer-Assisted Total Knee Arthroplasty With Contemporary Blood Management Had Minimal Differences in Blood Loss and Transfusion Rates Compared to Conventional Techniques

Abstract

BackgroundComputer-assisted surgery that does not utilize femoral canal instrumentation is theorized to have less blood loss. However, there is a paucity of data on this, particularly in the era of tranexamic acid use. We sought to analyze the association of computer navigation with total calculated blood loss and transfusion rate in patients undergoing primary total knee arthroplasty (TKA). MethodsWe identified 14,890 patients who underwent unilateral primary TKA at a single institution from 2016 to 2020. Computer-assisted surgery in the form of an accelerometer or robotics was utilized in 4,165 TKAs (28%). Drains were utilized in 4,860 TKAs (32%). We used multivariate logistic regression analysis to determine if computer navigation reduced the rate of blood transfusion and linear regression analysis to determine the impact of computer navigation on blood loss. ResultsIn total, 542 patients (3.6%) underwent a transfusion. The average change in hemoglobin (Hgb) was 2.1 g/dL (standard deviation [SD] 0.91) and average total calculated blood loss was 310 mL (SD = 154). In a multivariate regression model, computer navigation was not protective of transfusion (odds ratio [OR] 1.04, P = .73). Preoperative Hgb <10 (OR 10.5, P < .0001) and drain use (OR 2.25, P < .0001) were the most significant risk factors for transfusion. In a linear regression model, computer navigation reduced blood loss by 19 mL (SD 2.94, P < .0001) per case. ConclusionIn this large retrospective cohort analysis of contemporary TKA patients, computer-assisted surgery that eliminates intramedullary femoral canal instrumentation during primary TKA was not associated with reduced transfusion rates and had minimal differences in overall blood loss.