Drain use in total knee arthroplasty is neither associated with a greater transfusion rate nor a longer hospital stay.

PURPOSE: In recent years, the exclusion of a drain in total knee arthroplasty (TKA) is gaining popularity. This retrospective study aims to investigate a tertiary hospital's experience with the use of a drain in TKA. The authors hypothesise that the use of a drain will: (1) increase the peri-operative total blood loss (TBL) and transfusion rate; (2) increase the length of hospital stay (LOS); (3) reduce the 30-day readmission rate and incidence of additional surgical procedure performed.

METHODS: Patients who underwent a unilateral primary TKA in 2012 were included. Seven surgeons performed 575 TKAs with the use of drains, while nine other surgeons performed 902 TKAs without the use of drains. The patients were prospectively followed-up for two years. Peri-operative TBL was calculated using the haemoglobin balance method. All patients followed the hospital's transfusion and post-operative rehabilitation protocol.

RESULTS: There was a bigger drop in haemoglobin level by 0.5 g/dl (95 % CI, 0.4, 0.6) and greater TBL by 169 ml (95 % CI, 126, 181) in the drain group (both p < 0.001). However, the transfusion rate was 37/575 (6.4 %) and 48/902 (5.3 %) in the drain and no drain groups respectively (p = 0.370), while the LOS was four (IQR, 4, 5) and four (IQR 3, 5) days respectively (p = 0.228). The 30-day readmission rate was 10/575 (1.7 %) in the drain group, compared with 26/902 (2.9 %) in the no-drain group (p = 0.165). The incidence of additional surgical procedure performed was 5/575 (0.9 %) in the drain group, compared with 15/902 (1.7 %) in the no-drain group (p = 0.198).

CONCLUSIONS: Although the use of a drain in TKA is associated with greater peri-operative TBL, this additional amount of blood loss does not translate into an increased transfusion rate or a longer LOS. It also does not reduce the 30-day readmission rate and incidence of additional surgical procedure performed on the same knee.