JOA - 2026-07-23 - Journal Article
Finding the Sweet Spot: A Mixed-Effects Analysis of Four Tourniquet Strategies and Blood Conservation in 15,394 Primary Total Knee Arthroplasties.
Mallett KE, Weinblatt AI, Ondek NT, Öettl FC, Lan R, Lyman S, Chalmers BP, Della Valle AG
Topics
Key Takeaway
No-tourniquet TKA was associated with +211.3 mL greater calculated blood loss and 3.05x higher transfusion odds versus incision-to-cementation tourniquet use, though NNT to prevent one transfusion was approximately 250 in the TXA era.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study compared four tourniquet strategies—no tourniquet, cementation-only, incision-to-cementation, and incision-to-closure—across 15,394 primary TKAs using mixed-effects regression with surgeon as a random intercept. No-tourniquet use produced the highest median CBL (785 mL), a +0.46 g/dL Hgb drop, and OR 3.05 for transfusion versus incision-to-cementation. Incision-to-cementation demonstrated the most favorable blood conservation profile; incision-to-closure offered no additional benefit.
Key Limitation
Retrospective design means tourniquet strategy was not randomly assigned, and surgeons likely selected no-tourniquet for specific patient or anatomic reasons that residual confounding may not fully capture despite multivariable adjustment.
Original Abstract
BACKGROUND
Tourniquet use in total knee arthroplasty (TKA) remains variable, and its effect on perioperative blood loss in the era of tranexamic acid (TXA) is debated. This study compared four contemporary tourniquet strategies and their associations with hemoglobin drop, calculated blood loss, and transfusion risk.
METHODS
We retrospectively reviewed 15,394 primary TKAs performed from 2019 to 2023 at a high-volume academic center. Tourniquet use was categorized as: no-tourniquet, selective use during cementation only, incision-to-cementation (reference), or incision-to-closure. Outcomes included perioperative hemoglobin (Hgb) drop, calculated blood loss (CBL) using Nadler and Mercuriali's equations, and postoperative transfusion. Mixed-effects linear and logistic regression models quantified associations after adjustment for age, sex, body mass index, surgical time, Charlson Comorbidity Index, fixation method, preoperative hemoglobin, American Society of Anesthesiologists classification, and postoperative venous thromboembolism prophylaxis, with surgeon as a random intercept.
RESULTS
Compared with incision-to-cementation, the no-tourniquet group experienced a greater Hgb drop (+0.46 g/dL; 95% confidence interval (CI) 0.38 to 0.54; P < 0.001) and the highest median CBL (785 mL), with an adjusted increase of +211.3 mL (95% CI 169.8 to 252.7; P < 0.001). Selective tourniquet use was associated with a smaller, but significant increase in Hgb drop (+0.13 g/dL; P < 0.001). Incision-to-closure did not significantly differ. Transfusion was uncommon, but more frequent in the no-tourniquet group, with three times the adjusted odds compared with incision-to-cementation (odds ratio (OR) 3.05; 95% CI 1.72 to 5.41; P < 0.001).
CONCLUSIONS
Forgoing a tourniquet was associated with increased blood loss and higher odds of transfusion, though the absolute effect was modest in the contemporary TXA era (number needed to treat approximately 250 to prevent one transfusion). Tourniquet use from incision-to-cementation demonstrated the most favorable blood conservation profile; selective use performed similarly. These findings support a tailored approach to tourniquet use in TKA.