Because life expectancy is increasing, the number of primary knee arthroplasties performed is projected to increase 673% by 2030, according to Westrich et al. Also, Toulson et al. in a recent study predict that the incidence of deep infection associated with primary total knee arthroplasty ranges from 1% to 2%. Periprosthetic knee infection is one of the most dramatic and difficult to manage complications following total knee arthroplasty. Therefore, periprosthetic knee infection will continue to be a significant complication and an economic burden in the future. Our objective was to identify the risk factors that may provide greater likelihood of infection and thus select high-risk patients and to take maximum prevention strategies.

Case-control study, between infected and non infected patients, undergoing primary total knee arthroplasty between January 2008 and January 2013. The risk factors evaluated were: duration of hospital stay, surgery duration, prophylactic antibiotics and timing for administration, volume of blood transfusion, autologous blood recovery system use, anesthetic technique, ASA classification, Diabetes Mellitus, Obesity (BMI>30), immunosuppression and history of any infection in the month preceding surgery. The presence of infection was defined by the criteria of the Center for Disease Control for Nosocomial Surgical Site Infections1. Statistical analysis IBM SPSS Statistics 20 (Fisher's exact test, Mann-Whitney U test and Student's t-test). Statistical significance for p ≤ 0.05.

We evaluated 540 patients with a mean follow-up of 56 months. We identified 21 deep infections (3,8%), and 35 superficial wound infections and found a positive correlation between infection and obesity (p <0.01), immunosuppression (p <0.01), volume of blood transfusion (p=0.02), history of any infection in the month preceding surgery (p <0.01). We found a negative correlation with the use of a autologous blood recovery system (p <0.01). Other factors, commonly referred in the literature, showed no association or did not reach statistical significance.

The incidence of periprosthetic knee infection after primary total knee arthroplasty stays high. The presence of obesity, immunosuppression, blood transfusion, history of any infection in the month preceding surgery were identified as significant risk factors for infection to occur. The identification, modification or eviction of the risk factors implied are essential to reduce and prevent infection in arthroplasty.