ObjectiveTo clarify the effects of tourniquet use on pain, early rehabilitation, blood loss, incidence rate of thrombosis in primary total knee arthroplasty (TKA) through a randomized controlled trial. MethodBetween Janurary 2014 and August 2015, 168 patients with knee osteoarthritis undergoing primary TKA were randomly allocated to tourniquet group (n=84) or non-tourniquet group (n=82) . There was no significant difference in gender, age, body mass index, affected side, osteoarthritis grading, disease duration, preoperative range of motion (ROM), visual analogue scale (VAS), Hospital for Special Surgery (HSS) score, and hemoglobin (Hb) between 2 groups (P>0.05) . The operation time, hospitalization time, 90°knee flexion time, straight leg lifting time, and ambulation time were compared between 2 groups. Intraoperative blood loss, Hb decrease, postoperative VAS score, HSS score, ROM, and postoperative complications were recorded and compared. ResultsThere was no significant difference in operation time (t=-1.353, P=0.178) . The patients were followed up 3-20 months (mean, 12 months) in tourniquet group, and 3-22 months (mean, 13 months) in non-tourniquet group. No significant difference was found in Hb decrease (t=-1.855, P=0.066) and transfusion rate (23.81% of tourniquest group vs. 25.61% of non-tourniquest group) (χ2=0.072, P=0.788) between 2 groups. Significant difference was found in the incidence rate of thrombosis between tourniquet and non-tourniquet groups (10.71% vs. 2.44%) (χ2=4.592, P=0.032) , and the intraoperative blood loss of tourniquet group was significantly less than that of non-tourniquet group (t=-16.066, P=0.000) . The 90°knee flexion time, straight leg lifting time, ambulation time, and hospitalization time of tourniquet group were significantly later than those of non-tourniquet group (P<0.05) . The tourniquet group had significantly higher VAS score at 3, 5, 7, and 14 days after operation (P<0.05) and lower HSS score at 28 days after operation (t=-4.192, P=0.000) than non-tourniquet group, but there was no significant difference in the ROM between 2 groups (t=0.676, P=0.500) . ConclusionsThe use of a tourniquet during TKA will increase knee pain and thrombotic events, but can not decrease total blood loss and transfusion rate. A tourniquet use during TKA is unfavorable for early rehabilitation progress.