Aim: To assess the blood loss in patients undergoing total knee arthroplasty and to determine the effects of surgical technique, duration of surgery and use of tourniquet. To look at the value of patient factors including gender, age, weight, pre-operative haemoglobin as predictive factors for future transfusion. We assessed the complications of wound healing, regaining the range of motion and thromboembolic effects.

Patients and Methods: A prospective review of 150 patients undergoing knee arthroplasty, with a minimum follow-up of 2 years (2–8 years). We divided the patients into three groups (n=50). Group A had no tourniquet applied and haemostasis secured before skin closure, Group B had tourniquet released after cementation to secure haemostasis before skin closure and Group C had tourniquet applied until after the skin closure.

Patients were matched for age, gender, pathology, weight, implant type, pre operative haemoglobin and senior operator in all three groups. We assessed intra-operative and total blood loss, transfusions requirements, postoperative wound complication, regaining of the range of motion, incidence of systemic effects of tourniquet and duration of hospital stay. We also looked at the effects of NSAIDS on blood loss and compared the validity of various factors reported in the literature to be predictive of future transfusion after the surgery. Statistical analysis used were, student’s t-test, univariate and multivariate analysis and regression statistical analysis.

Results: Group A had maximum blood loss (mean 1374 mls.) followed by Group B (mean 774 mls.) The mean blood loss of Group C was 550 mls (p< 0.001, 95% confidence interval range of 527843 mls). Statistically the duration of operation was the most important factor in minimising the blood loss (p< . 0001 R₂ =0.68). The association of pre operative haemoglobin and weight as predictors of future transfusions statistically did not show a strong relationship (R2 = 0.17, R₂ = 0.13 respectively). Statistically no significant difference was found in wound complications, hospital stay, post operative pain and regaining the range of motion in three groups (R₂ 0.58). Patients on NSAIDS did not loose more bloods than patients not on NSAIDS.

Conclusions: The use of tourniquet until after the skin closure is statistically the best method for reducing blood loss in total knee arthroplasty. It does not cause significant wound problems and does not significantly affect the regaining of range of motion. Furthermore the usage of NSAIDS does not cause excessive postoperative bleeding.