Background
Rivaroxaban and apixaban are two direct oral anticoagulants (DOACs) targeting Factor Xa. Each DOAC was separately proven effective and safe when compared to standard treatment (heparin followed by warfarin) in patients diagnosed with venous thromboembolism (VTE). Several retrospective cohort analyses suggest apixaban may be superior to rivaroxaban due to less bleeding rates. One recent study showed the safety of apixaban and rivaroxaban for acute VTE were comparable. Also, long-term anticoagulation with low molecular weight heparin (LMWH), has never been directly compared with Factor Xa inhibitors. Given the patient population at our facility includes a significant percentage of elderly from nursing homes with multiple comorbidities and significantly higher bleeding risks than the general population, we aimed to evaluate the local real-world DOACs and heparin use with particular focus on safety.
Methods
A retrospective study was conducted at Kingsbrook Jewish Medical Center. Demographics, relevant laboratory/ imaging studies for patients admitted from 1/2016 to 12/2018 with the diagnosis of VTE based on the ICD 9/10 codes were collected from the IT dept. VTE patients who had bleeding events during the same admission for VTE or were admitted for relevant bleeding events based on ICD 9/10 codes within 6 months from the diagnosis of VTE were identified. Major bleeding events were defined as requiring hospitalization, blood transfusion or a significant drop in hemoglobin (more than 2 g/dl). The rest of the bleeding events were classified as minor. Demographics and clinical characteristics were summarized with means/median for continuous variables and with proportions for categorical variables. The differences in covaries were assessed with chi-square, Fisher exact test or t-tests.
Results
A total of 177 acute VTE patients were identified in the study. 37.9% (n=67) and 32.8% (n=58) patients were started on rivaroxaban and apixaban, respectively, as monotherapy. 29.3% (n=52) patients were given unfractionated heparin (23%, n=12) or LMWH (77%, n=40) based on the renal function. The bleeding rate in the apixaban group (4/58, 6.9%) was slightly higher than that in the rivaroxaban group (3/67, 4.5%), however, there was no statistical significance. Compared to patients received heparin (11/52, 21.2%), patients started on rivaroxaban (p=0.008) or apixaban (p=0.049) had a significantly lower rate of bleeding. Of note, majority of the bleeding event (16/18) was captured during the same admission when VTE was diagnosed. VTE patients in the apixaban group were older (p = 0.007) and had a longer length of stay (LOS, p= 0.024) compared to the ones in the rivaroxaban group.
We then combined rivaroxaban and apixaban into DOAC group. Compared to Heparin group (n=11, 21.1%), the bleeding events in DOAC group (n=7, 5.6%) were significantly less (P=0.0045). Heparin group (n=21, 40.4%) included more patients with cancer than the DOAC group (n=9, 7.2%, p <0.0001), suggesting active malignancy may be correlating with higher bleeding risk.
We then looked at the bleeding risk in non-cancer patients. Similarly, we didn't observe any superiority between rivaroxaban (n=2, 3.1%) and apixaban (n=3, 5.8%) regarding bleeding events in non-cancer patients with acute VTE. The average LOS in the apixaban group was significantly longer than that in the rivaroxaban group with non-cancer patients. It may be associated with relatively older age in the apixaban group compared to that in the rivaroxaban group. Most importantly, in patients without active malignancy, we found that the bleeding rate in the DOAC group was only 4.3% while the heparin group had a much high bleeding rate of 16.1% (p=0.035).
Discussion
Our study suggested that the safety of apixaban and rivaroxaban are comparable in VTE patients. In contrast, heparin including LMWH had much higher bleeding risk compared to either DOAC, especially in the beginning. For patients who are hospitalized for acute VTE, heparin intravenously or subcutaneously are usually initiated while the decision for oral anticoagulants are still pending. However, the benefits of such "bridging with heparin" strategy are not warranted, given the high bleeding risk associated with heparin treatment as shown in our study. Except for hemodynamically instability, excessive burden or clots, or impeding procedures, we recommend that DOACs should be used as monotherapy in VTE patients.
Disclosures
No relevant conflicts of interest to declare.
Incidence of symptomatic venous thromboembolism following hospitalization for coronavirus disease 2019: Prospective results from a multi-center study