Impact of Femoral Medullary Canal Preservation During Primary Total Knee Arthroplasty on Early Perioperative Outcomes

Purpose- Intramedullary (IM) femoral referencing during total knee arthroplasty (TKA) has been associated with increased perioperative blood loss. This study aimed to evaluate whether a simplified extramedullary (EM) free-hand referencing technique, designed to preserve the femoral medullary canal, is associated with differences in perioperative bleeding-related parameters and alignment accuracy compared with conventional IM referencing. Methods- A retrospective cohort study was conducted including patients who underwent primary TKA using either IM or EM femoral referencing by a single surgeon. Perioperative hemoglobin change, calculated estimated blood loss, and postoperative D-dimer levels were analyzed as comparative bleeding-related parameters. Operative time, transfusion rates, early recovery indicators, and radiographic alignment outcomes were also assessed. All patients were managed under a standardized perioperative blood management protocol, including routine tranexamic acid administration. Results- A total of 125 TKAs were analyzed (IM: 60; EM: 65). Operative time was comparable between groups. The EM group demonstrated a smaller postoperative hemoglobin decrease (1.92 ± 0.8 vs. 2.35 ± 1.0 g/dL; p = 0.022) and lower calculated estimated blood loss (618.4 ± 291.5 vs. 755.8 ± 272.1 mL; p = 0.008). Postoperative D-dimer levels at 24 hours were lower in the EM group (p = 0.015) and were analyzed as an exploratory marker of perioperative physiological response. Transfusion rates were low and comparable between groups. Radiographic alignment accuracy did not differ significantly between techniques. Conclusion- Femoral medullary canal preservation using a simplified extramedullary free-hand referencing technique was associated with favorable perioperative bleeding-related parameters without compromising alignment accuracy or operative efficiency. This approach represents a practical alternative to intramedullary referencing in primary TKA and warrants further prospective evaluation. Level of Evidence III