Date of Award

January 2024

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Cassius I. Ochoa Chaar


In patients undergoing revascularization for peripheral arterial disease (PAD), low-dose Factor Xa inhibitors (FXaI) taken with aspirin improved limb and cardiovascular outcomes compared to aspirin alone. Further, in atrial fibrillation and venous thromboembolism, FXaI are recommended over vitamin K antagonists (VKA) for chronic anticoagulation. While studies have evaluated different perioperative antithrombotic (antiplatelet + anticoagulation) regimens in patients treated for PAD, the role of chronic anticoagulation in patients undergoing lower extremity peripheral vascular interventions (PVI) has not been determined. Moreover, combination antithrombotic regimens are common in current clinical practice; however, limited evidence exists to endorse them in societal guidelines. More specifically, the effectiveness of FXaI and clopidogrel (FXaI+C) in comparison to dual antiplatelet therapy (DAPT, aspirin and clopidogrel) following PVI remains unexplored.

The objective of this thesis is to investigate outcomes of PVI for PAD in patients treated with different regimens of FXaI. The primary aim is to compare PVI outcomes in patients chronically anticoagulated with FXaI or VKA. In a secondary analysis, this study assesses the outcomes of patients discharged on FXaI+C and DAPT.

The Vascular Quality Initiative-PVI database was used between 2004 and 2023. In the primary analysis, patients consistently treated with FXaI or VKA before the procedure, at discharge and on long-term follow up were defined as those receiving chronic anticoagulation. Patient demographics, procedural details, and perioperative and long-term outcomes were compared between FXaI and VKA chronic anticoagulation groups. In the secondary analysis, patients were grouped based on discharge medications alone. Patients discharged on FXaI and clopidogrel without aspirin were separated from those discharged on DAPT. Nearest neighbor propensity score matching was used to generate patient populations with similar comorbidities and procedural characteristics. Patient demographics, procedural details, and long-term outcomes following PVI were compared between the two groups.

In the primary analysis, 6,885 were chronically anticoagulated with FXaI (N = 2,427) or VKA (N = 4,458). Compared to patients taking VKA, those receiving FXaI had lower rates of major amputation (6.7% vs 8.4%, P=0.020) and mortality (7.6% vs 15.2%, P<0.001). Using Kaplan-Meier analysis, patients treated with FXaI had improved amputation-free survival (AFS) after PVI compared to patients receiving VKA. Adjusting for significant patient and procedural characteristics, Cox proportional hazard regression demonstrated that there is an increased risk for major amputation or mortality in patients using VKA compared to FXaI (HR 1.61, [1.36-1.90]).

In the secondary analysis, a total of 161,550 patients underwent primary PVI, and 32.5% (N=52,584) were discharged on either FXaI+C or DAPT. After four-to-one matching, 8,356 patients receiving DAPT were compared with 2,089 patients taking FXaI+C. After a mean follow up of 400 days, patients discharged on FXaI and clopidogrel had significantly higher rates of reinterventions (20.4% vs. 15.9%, P<0.001) and one-year mortality (13.3% vs. 11.3%, P<0.001) as compared to patients discharged on DAPT. Cox Proportional-hazard regression analysis showed that discharge on FXaI+C was independently associated with increased mortality compared to DAPT (HR=1.22, [1.01-1.47]) after PVI for PAD.

In conclusion, chronic anticoagulation with FXaI as compared to VKA was associated with superior perioperative and long-term outcomes in patients with PAD undergoing PVI. FXaI should be the preferred agents over VKA for chronic anticoagulation after PVI in patients with PAD. However, in patients with PAD not receiving chronic anticoagulation, DAPT seems to be associated with improved limb outcomes and survival compared to FXaI+C at discharge after PVI. Further investigation is needed to determine optimal combination antithrombotic pharmacotherapy regimens after PVI for PAD.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.