Background/Purpose: Total knee replacement (TKR) patients are routinely prescribed anticoagulation therapy to prevent deep vein thrombosis (DVT) and pulmonary embolism (PE). Clinical guidelines are ambiguous regarding the specific agent and duration of prophylaxis. We conducted a cost-effectiveness analysis to evaluate the benefits and risks of 14- and 35-day therapy with the most commonly prescribed anticoagulants post-TKR.

Methods: We built a probabilistic, state-transition computer simulation model to assess clinical and economic outcomes of 14-day and 35-day anticoagulation therapy after TKR. Complications of TKR and therapy included DVT, PE, bleeding, and prosthetic joint infection (PJI). DVTs could progress to PE and were classified as symptomatic or asymptomatic, with asymptomatic carrying a greater risk of PE. Operative site bleeds were associated with an increased risk of PJI (RR = 11), while non-operative site bleeds (CNS, GI) carried a quality of life decrement and risk of death. We evaluated 5 anticoagulation agents: rivaroxaban, low molecular weight heparin (LMWH), fondaparinux, warfarin, and aspirin. Each was associated with a unique reduction in DVT risk and an increased risk of bleeding compared with no anticoagulation; these risks were estimated from published literature. Costs included the agent and, when applicable, injection administration or monitoring costs (Table). We assumed a 1 year horizon and a willingness to pay (WTP) threshold of $100,000 per quality adjusted life year (QALY). Strategies with incremental cost-effectiveness ratios (ICERs) below WTP were deemed cost-effective.

Results: Aspirin at any duration was associated with the highest incidence of DVT and PE (28% and 5%, respectively). Extended fondaparinux resulted in the largest reduction in thromboses (DVT + PE) and greatest increase in bleeds (14% and 3%, respectively). Extended rivaroxaban reduced DVT incidence to 18% while increasing bleeds to 6%. Extended LMWH was associated with DVT and bleeding incidence of 19% and 4%, respectively. Both extended fondaparinux and rivaroxaban resulted in 0.74 QALYs; all other strategies, including no prophylaxis, resulted in fewer QALYs and higher costs and were therefore dominated (Table). The ICER for extended fondaparinux ($16.3M) greatly exceeded WTP; thus, extended rivaroxaban was the preferred strategy.

Conclusion: Extended rivaroxaban therapy after TKR is a cost-effective strategy to prevent DVT and PE while minimizing bleeding risk. The high cost and risk of bleeding of fondaparinux precluded it from being cost-effective. While there has been increased interest in using lower potency therapies, such as aspirin, these results demonstrate that aspirin’s low bleeding risk and low cost do not compensate for its poor efficacy in preventing DVT post-TKR.

Table.