BACKGROUND: The frequency of chronic postsurgical pain (CPSP) after knee replacement remains high, but might be decreased by improvements to prevention. OBJECTIVES: To identify pre- and postsurgical factors predictive of CPSP 6 months after knee replacement. STUDY DESIGN: Single-center prospective observational study. SETTING: An orthopedic unit in a French hospital. METHODS: Consecutive patients referred for total or unicompartmental knee arthroplasty from March to July 2013 were prospectively invited to participate in this study. For each patient, we recorded preoperative pain intensity, anxiety and depression levels, and sensitivity and pain thresholds in response to an electrical stimulus. We analyzed OPRM1 and COMT single-nucleotide polymorphisms. Acute postoperative pain (APOP) in the first 5 days after surgery was modeled by a pain trajectory. Changes in the characteristics and consequences of the pain were monitored 3 and 6 months after surgery. Bivariate analysis and multivariate logistic regression were conducted to identify predictors of CPSP. RESULTS: We prospectively evaluated 104 patients in this study, 74 (28.8%) of whom reported CPSP at 6 months. Three preoperative factors were found to be associated with the presence of CPSP in multivariate logistic regression analysis: high school diploma level (OR = 3.83 [1.20 - 12.20]), consequences of pain in terms of walking ability, as assessed with the Brief Pain Inventory short form "walk" item (OR = 4.06 [1.18 - 13.94]), and a lack of physical activity in adulthood (OR = 4.01 [1.33 - 12.10]). One postoperative factor was associated with the presence of CPSP: a high-intensity APOP trajectory. An association of borderline statistical significance was found with the A allele of the COMT gene (OR = 3.4 [0.93 - 12.51]). Two groups of patients were identified on the basis of their APOP trajectory: high (n = 28, 26%) or low (n = 80, 74%) intensity. Patients with high-intensity APOP trajectory had higher anxiety levels and were less able to walk before surgery (P < 0.05). LIMITATIONS: This was a single-center study and the sample may have been too small for the detection of some factors predictive of CPSP or to highlight the role of genetic factors. CONCLUSION: Our findings suggest that several preoperative and postoperative characteristics could be used to facilitate the identification of patients at high risk of CPSP after knee surgery. All therapeutic strategies decreasing APOP, such as anxiety management or performing knee replacement before the pain has a serious effect on ability to walk, may help to decrease the risk of CPSP. Further prospective studies testing specific management practices, including a training program before surgery, are required.