Purpose of the study: Total knee arthroplasty (TKA) is associated with important intra- and postoperative bleeding often requiring transfusions, leading to certain risks despite classical methods of haemostasis. The purpose of this study was to estimate the efficacy of a fibrin glue for reducing postoperative blood loss after TKA.

Material and methods: This was a single centre single operator randomised prospective study. Two groups of patients received a first-intention TKA implanted without cement. For a control group (n=24) classical surgical haemostasis was performed; in another group of 22 patients, 4 ml of fibrin glue were vaporised on the internal structures and the subcutaneous tissue intraoperatively, just after insertion of the implants and associated with classical haemostasis. In all cases, the patients had gravity drainage for 48 h; the tourniquet was not removed before dressing. All patients had preventive anticoagulation postoperatively on day 1.

Results: Blood loss was calculated from d−1 to d+4 and was expressed in grams and in percentage in order to overcome the bias of body weight and gender. Mean blood loss was 17 g in the fibrin glue group (raw data) or 24% of the blood mass while it was 211 g in the control group, i.e. 31%. The percentage of blood loss was 31 and 24% with fibrin (p=0.05). Three of 22 patients, 13%, required transfusion in the fibrin glue group versus 11 of 24, 46%, in the control group. We did not take into consideration the rate of seroconversion at three and six months postoperatively.

Conclusion: Use of a fibrin glue can significantly reduce the need for transfusion and reduce blood loss postoperatively after first-intention unilateral uncemented TKA.