Background and purpose — Venous thromboembolism (VTE) is a serious postoperative complication after total knee arthroplasty (TKA). Use of a tourniquet has shown conflicting results for risk of VTE after TKA. We aimed to investigate the associated risk of VTE after TKA using tourniquet in a fast-track set-up as no previous data exists. Patients and methods — We performed an observational cohort study from 9 fast-track centers including unilateral primary TKA from 2010–2017 with prospective collection of preoperative risk-factors and complete 90-day follow-up. Use of a tourniquet was registered in the Danish Knee Arthroplasty Register. Postoperative VTE was identified from health records. We performed risk analyses using a mixed-effects logistic regression model adjusting for previously identified risk factors. Results — Of the 16,250 procedures (39% males, mean age 67.9 [SD 10.0] years, median LOS 2 [interquartile range 2–3]) 12,518 (77%) were performed with a tourniquet. The annual tourniquet usage varied greatly between departments from 0% to 100%, but also within departments from 0% to 99%. There was no significant difference between the 2 groups with 52 (0.42%) VTEs in the tourniquet group vs. 25 (0.67%) in the no-tourniquet group (p = 0.06 for cumulative 90-day incidence of VTE). This association remained statistically insignificant for VTE using tourniquet after adjustment for previously identified risk factors. Conclusion — We found no association between the use of a tourniquet and increased risk of 90-day VTE after primary fast-track TKA, irrespective of the length of time for which the tourniquet was applied.