Blood loss is a major concern in total knee arthroplasty (TKA) along with postoperative knee function. The present study explores the impact of tourniquet and closed-suction drains on blood loss as well as knee function in TKA.

A prospective clinical trial was conducted on 111 patients admitted for TKA. Subjects were divided into three groups based on duration of tourniquet use (T+: whole-course tourniquet, T-: cementation only tourniquet) and usage of closed-suction drain (D+: drain use, D-: no drain). Thirty-six subjects were included in group T+D+, 42 in T-D+ and 33 in T-D-. Data from study population was analysed for pre and post-operative hemoglobin level (Hb), perioperative and total blood loss, blood transfusion rates, knee range of motion (ROM), and pain level assessment. Direct and indirect costs associated to nursing time and drains were calculated. Results are presented in mean ± SD.

No statistically significant differences were observed among the three groups (T+D+, T-D+ and T-D-) concerning total blood lost (calculated using Gross' formula), Hb levels over the first six postoperative weeks, blood transfusion rates and intra-articular hematomas. Intraoperative bleeding was significantly reduced in T+ subjects compared to T- subjects (100 ± 88 mL vs. 279 ± 235 mL respectively, p < 0.001), yet length of surgery was unaffected by the different tourniquet inflation strategies. Hidden blood loss was lower in D+ subjects compared to D- subjects (1161 ± 554 mL vs. 1667 ± 554 mL respectively, p < 0.001), but it was compensated by the blood loss in the drains. Early post operative ROM (flexion: 79.1 ± 14.8°, extension: −5.0 ± 6.7°) was superior in group T-D- compared to group T+D+ (flexion: 71.9 ± 17.1°, p = 0.071°; extension: −9.9 ± 6.4°, p = 0.004). Nevertheless, ROM six weeks postoperative was not statistically different between groups (flexion: 114.0 ± 13.3°, extension: 0.4 ± 5.2°). Patient-reported postoperative pain was also similar in all groups. Nursing time dedicated to drain management was 30 min/patient (330h total for 660 patients/year). Total costs related to drains were $31.92CAD/patient ($21,067CAD total for 660 patients/year).

Our results suggest that whole-course tourniquet and closed-suction drain use in TKA do not yield beneficial results in total blood loss, blood transfusion rates, complication rates and knee rehabilitation. Their clinical relevance in TKA is questionable. Moreover, nursing time and costs related to drains should have been allocated elsewhere in patient care.