Background: After total knee arthroplasty (TKA), 17%-60% of the patients do not or only partially return to work (RTW). Reasons for no or partial RTW remain unclear, warranting further research. Physical activity (PA) has proven beneficial effects on work participation. Therefore, we hypothesized that preoperative PA is associated with RTW after TKA. Methods: Working TKA patients participating in an ongoing prospective cohort study were included. Preoperatively and 1 year postoperatively, patients were asked to define their work status and PA level according to the Dutch Recommendation for Health-Enhancing PA and the Fitnorm. Multivariate logistic regression analysis was performed to assess the effect of PA on RTW, taking into account established prognostic factors for RTW among TKA patients. Results: Of 283 eligible patients, 266 (93%) completed the questionnaires sufficiently. Preoperatively, 141 patients (54%) performed moderate PA for ≥5 d/wk and 42 (16%) performed intense PA for ≥3 d/wk. Concerning RTW, 178 patients (67%) reported full RTW, 59 patients (22%) partial RTW, and 29 patients (11%) no RTW. Preoperative PA was not associated with RTW. Patients who reported that their knee symptoms were not or only partially work-related had lower odds of no RTW (odds ratio 0.37, 95% confidence interval 0.17-0.81). Also, for each additional week patients expected to be absent from work, the likelihood of no RTW increased (odds ratio 1.11, 95% confidence interval 1.03-1.18). Conclusion: No association between preoperative PA and RTW after TKA was found. Patient beliefs and preoperative expectations did influence RTW and should be addressed to further improve RTW after TKA.