Although drains date back to the Hippocratic era, their routine use remains controversial in total hip arthroplasty. The literature suggests that they can provide a retrograde route for infection as well as decreasing the organism count required to develop an infection. The use of drains has not decreased the size of wound haematomas at day five on ultrasound or the incidence of massive wound haematomas. Neither have they been shown to significantly decrease wound infections. This consecutive prospective randomised study was designed to evaluate what role drains have in the management of patients undergoing hip arthroplasty.

A total of 577 patients undergoing unilateral or bilateral hip arthroplasty were evaluated in a randomised prospective trial of drain versus no drain, between September 1997 and December 2000. All patients had a standardised pre, inter and post operative regime and were independently assessed using the Harris hip score and SF36 pre-operatively, at discharge and at six months post surgery.

The superficial and deep infection rate of 6. 4% and 0. 4% was seen in those drained and 7. 1% and 0. 7% in the non-drained group. Only one patient sustained a clinical haematoma that did not requiring drainage or transfusion in the non-drain group. The transfusion rate in those drained was 33. 0% compared to 26. 4% in those not drained. There was no statistical advantage in using a drain P> 0. 05 regarding these variables or in the length of stay, SF36 or Harris hip scores at pre-op and six months. Using a drain did significantly increase the likelihood of requiring a transfusion P< 0. 05.

In conclusion drains provide no statistical advantage whilst represent an additional cost and expose hip arthroplasty patients to an unacceptable risk of infection and transfusion.