Persistent wound drainage after total joint arthroplasty (TJA) has been associated with a higher incidence of superficial and deep periprosthetic infection but the predictors for prolonged drainage and its outcome have not been thoroughly studied.

A consecutive series of 7,153 TJA cases performed between 2000 and 2006 at our institute, were recruited into this study. There were 301 cases (4.2%) of persistent wound drainage, defined as discharge from the wound for > 48 hours. The cases were matched in a 2:1 ratio for type of surgery, joint replaced, and date of surgery.

This study identified higher BMI (p< 0.005), malnutrition as defined by serum albumin< 3.4g/dl (p< 0.04), longer operative time (p< 0.01), and higher medical comorbidities, in particular diabetes (p< 0.001) as important risk factors for persistent wound drainage. In addition, patients in the drainage group were more likely to have a peak INR of > 1.5 (p< 0.001) during their hospital stay. Patients with wound drainage had a significantly lower hemoglobin postoperatively (p< 0.01) that necessitated greater number of postoperative allogenic transfusions (p=0.004). The hospital length of stay for the drainage group was also significantly higher (p< 0.005). One of the major risk factors for development of deep infection was prolonged drainage (> 7 days). In the deep hematoma and periprosthetic subgroups, the mean of delay in treatment was 6 days in those with retention of the prosthesis and successful outcome, and 9.5 days for those with failure of incision and drainage leading to resection arthroplasty (p= 0.03).

72% of the patient were successfully treated by oral or intravenous antibiotics. 27% required at least one re-operation for deep hematoma and 13% developed deep periprosthetic infection, resulting in 6% rate of resection arthroplasty. 1.5% of those with drainage remained in girdlestone status.

This study suggests early surgery for persistent drainage and avoidance of aggressive anticoagulation.