Deep vein thrombosis remains a significant and common complication following joint replacement and debate exists over which contributing factors are important. This study investigates the effect of a number of variables on the incidence of symptomatic deep vein thrombosis in knee arthroplasty surgery. Data was collected prospectively on 3449 knee replacements including procedure type, tourniquet time, surgeon, patient age, and gender. These variables were assessed by the use of generalised linear modelling against the presence or absence of symptomatic deep vein thrombosis demonstrated on duplex ultrasonography. The overall deep vein thrombosis rate was 1.6%. The only variable which had an association with confirmed symptomatic DVT was operation type with total knee replacements having a higher incidence than unicompartmental knee replacements (2.2% vs 0.3% p=0.0003). Tourniquet time did not exhibit a statistically significant effect (p=0.63) These data show that the DVT rate in unicompartmental knee replacement is statistically significantly lower than that of total knee replacement. They do not support the notion that increased tourniquet time is associated with an increased risk of DVT.