Objective: To determine the risk factors associated with prosthetic knee infection in elderly patients in a referral hospital in Peru. Patients and methods: A case and control study was performed. The calculated sample was 44 cases and 132 controls. The data were collected retrospectively from clinical records. U-Mann Whitney and Chi-square tests were performed in the comparison of cases and controls. Odds ratios (OR) were calculated in a binary logistic regression analysis to identify the risk factors, a P < .05 and a 95% confidence interval (CI) were considered significant. Results: Significant (P < .05) risk factors evidenced in the bivariate analysis were obesity (OR = 9.72; 95% CI: 4.47-21.14), smoking (OR = 4.06; 95% CI: 1.59-10.39), rheumatoid arthritis (OR = 4.66; 95% CI: 1.52-14.32), diabetes mellitus type 2 (OR = 5.63; 95% CI: 2.69-11.78), persistent drainage (OR = 9.27; 95% CI: 3.85-22.31), superficial infection (OR = 6.87; 95% CI: 3.25-14.49) and prolonged hospital stay (OR = 4.67; 95% CI: 2.26-9.64). In the multivariate analysis where it was adjusted for confounding variables, it was determined that risk factors were obesity (ORa = 9.14; 95% CI: 3.28-25.48), diabetes mellitus (ORa = 3.77; 95% CI: 1.38-10.32), persistent drainage (ORa = 4.64; 95% CI: 1.03-20.80) and superficial wound infection (ORa = 27.35; 95% CI: 2.57-290.64). Conclusions: Risk factors for prosthetic knee infection identified in this study are preventable. The main risk factors were obesity, diabetes mellitus type 2, superficial wound infection and persistent drainage, which were considered together or separately to be risk factors in the population studied.