https://orcid.org/0000-0003-1831-4166
TO THE EDITOR:
Guidelines recommend direct oral anticoagulants (DOACs) as preferred initial therapy for most venous thromboembolic events (VTEs) and DOAC dose-reduction (DR) for extended secondary prophylaxis.1-3 There has been an ongoing debate regarding the use of DOACs in patients with obesity. Obesity elevates both the baseline and the recurrent VTE risk.4,5 In 2021, the International Society on Thrombosis and Hemostasis recommended the use of DOACs for primary therapy in patients with body mass index (BMI) >40 kg/m2 or weight >120 kg.6,7 Data on the use of dose-reduced anticoagulation as secondary prophylaxis in patients with obesity, however, remain limited.8
To address this gap, we conducted a single-center analysis of adult patients with a documented visit between 1 January 2015 and 31 December 2023 within Montefiore Einstein Medical Center. We queried our hospital-wide database, using SQL (Structured Query Language), for patients with documented acute VTE. Patients were required to have received DR-apixaban, defined as 2.5 mg apixaban twice daily, for ≥3 months after a period of therapeutic dose anticoagulation lasting 3 to 12 months.
Clinical data were extracted from the electronic medical record system. Medical records of SQL-selected patients were manually reviewed, and information was individually assessed by 2 reviewers. Patients were included if their BMI at DR was ≥39 kg/m2. Patients were excluded if they lacked adequate documentation of the VTE, had received DR-AC (dose-reduced anticoagulation) for a different indication (atrial fibrillation, primary prophylaxis, etc), renal dysfunction (creatinine ≥2 mg/dL), if doses could not be confirmed, or if there was significant nonadherence. Patients were followed for 12 months from the date of DR or until therapy was discontinued.
Two hundred and fifty-five patient records were identified by data query of which 43 patients were eligible for inclusion (Figure 1). The final cohort included 9 males and 34 females. The majority self-identified as Black (53.5%), followed by White (25.6%) and Hispanic/Latino (18.6%). Average age at time of clot was 55.3 ± 13 years. Average BMI at DR was 45.7 ± 5.4 kg/m2. Ten patients had BMI ≥50 kg/m2.
CONSORT diagram of included patients. AC, anticogulation; CONSORT, Consolidated Standards of Reporting Trials.
CONSORT diagram of included patients. AC, anticogulation; CONSORT, Consolidated Standards of Reporting Trials.
Of 26 patients with pulmonary embolisms (PEs), 2 were massive, 8 submassive, and 6 were saddle PEs. Fourteen of the 26 were bilateral and 12 were unilateral. There were 11 PEs with sPESI (Simplified PE Severity Index) ≥1. Five patients required intervention including 4 who received thrombolysis and 1 who underwent thrombectomy. Of the 25 patients with deep vein thrombosis (DVT), 21 had proximal lower extremity DVTs, with 2 having phlegmasia cerulea dolens. The DVT group also contained 2 internal jugular DVTs and 2 isolated distal DVTs. Eight patients had both PEs and DVTs.
In addition to obesity, risk factors included 4 patients with previous VTE, 5 with active cancer, and 4 who had undergone bariatric surgery years before VTE. Antiphospholipid antibody status was negative in all 18 patients tested. One patient had factor V Leiden and 1 patient had prothrombin gene mutation, both heterozygous.
Patients received initial therapeutic anticoagulation for 7.8 ± 3.1 months (Table 1). During the therapeutic treatment phase, 12 patients had documented bleeding events. Four were major bleeds: 2 with menorrhagia, 1 occult gastrointestinal bleed from a large tubular adenoma, and 1 without identifiable source. There were 4 clinically relevant nonmajor bleeds (CRNMBs), and 4 minor bleeds. In 4 of these 12 patients, the bleeding explicitly influenced the decision of DR. Of the 4 major bleeds, the patient who had had a major bleed related to the tubular adenoma while on therapeutic anticoagulation, had recurrent bleeding on reduced dose. Despite this recurrent bleed, the patient who had had a saddle PE with an sPESI of 5, continued DR-apixaban treatment. The other 11 patients with bleeding on therapeutic anticoagulation did not have recurrent hemorrhage. Only 1 other patient bled on DR-AC. This patient was found to have an acute on chronic subdural hematoma of unknown chronicity; apixaban (started for a unilateral DVT) was discontinued 5.5 months after DR. This patient had no expansion of the hematoma on serial imaging and no further complications. No CRNMBs or minor bleeds occurred in the DR-apixaban period. There were no recurrent VTEs in either the therapeutic anticoagulation or the DR-AC periods.
Bleeding and VTE recurrence events in initial therapeutic and follow-up dose reduction period
| Initial therapeutic period | Mean (SD) | Follow-up DR period | Mean (SD) |
|---|---|---|---|
| Duration, mo | 7.8 (3.1) | Duration, mo | 10.8 (2.4) |
| Complications | n | Complications | n |
| Recurrent VTE | 0 | Recurrent VTE | 0 |
| Any bleeds | 12 | Any bleeds | 2 |
| Major bleeds | 4 | Major bleeds | 2∗ |
| CRNM bleeds | 4 | CRNM bleeds | 0 |
| Minor bleeds | 4 | Minor bleeds | 0 |
| Initial therapeutic period | Mean (SD) | Follow-up DR period | Mean (SD) |
|---|---|---|---|
| Duration, mo | 7.8 (3.1) | Duration, mo | 10.8 (2.4) |
| Complications | n | Complications | n |
| Recurrent VTE | 0 | Recurrent VTE | 0 |
| Any bleeds | 12 | Any bleeds | 2 |
| Major bleeds | 4 | Major bleeds | 2∗ |
| CRNM bleeds | 4 | CRNM bleeds | 0 |
| Minor bleeds | 4 | Minor bleeds | 0 |
SD, standard deviation.
Two major bleeds on dose reduced apixaban include a patient with a gastrointestinal bleed from a large tubular adenoma, who, after initially bleeding on therapeutic anticoagulation, was found once again to have a drop in hemoglobin 8.5 months after dose reduction. The other major bleed was a patient with an acute on chronic subdural hematoma of unknown chronicity 5.5 months after dose reduction.
In this study examining the safety and efficacy of DR-apixaban in patients with severe obesity for secondary prevention of recurrent VTE, we found a low bleeding rate and no recurrent VTEs. To our knowledge, this is the largest study examining DR-apixaban in this extreme setting.
The recent RENOVE trial examined DR-DOACs as secondary prophylaxis and analyzed a subgroup of patients with obesity with BMIs >30 kg/m2.9 Their data suggested no significant difference in a composite end point of bleeding and clinically relevant bleeds in this subgroup (∼30% of their study population). No data was presented for the severely obese.9 The Apixaban Cancer Associated Thrombosis (API-CAT) trial similarly explored secondary VTE prophylaxis with apixaban in patients with cancer and found no significant difference in recurrent thrombosis or bleeding in a small cohort of patients with BMI ≥35 kg/m2.10 Schaefer et al11 examined DR-DOACs in a population with 52% having BMIs >30 kg/m2. They found a nonsignificant decrease in thrombosis in the reduced-dose group and a significant decrease in bleeding.11 This study did have a few patients with BMIs >40 kg/m2 who were on apixaban or rivaroxaban. Vlanzy et al12 examined DR-apixaban in a small group (n = 13) of a more similar obesity level (>120kg, BMI not noted) and found no increase in VTE but a suggestion of increased bleeding. Our data explores a larger population with specifically severe obesity, including a significant portion above 50 kg/m2. Our data suggest that DR-apixaban is a reasonable option in this population.
Notably, in our data there was a complete absence of recurrent VTE. Given the rate of recurrence of VTE of the general population on DOACs is about 1% to 4% per year, this negative signal is reassuring. Further, many patients in our study were at an additional high risk for recurrence due to malignancy, thrombophilia, or severe VTE events.13,14 Given the elevated risk of many of these patients, the lack of VTE recurrence is further reassuring.
The bleeding risk in our DR-apixaban study during the secondary prophylactic phase was 4.7% (5.2% per patient year), higher than the generally accepted risk of bleeding of about 0.5% to 2% annually in patients with obesity.15-21 Several retrospective studies have noted higher bleeding risk, especially when including CRNMBs.19,20,22 Other studies have excluded patients who have previously bled on therapeutic AC but our study did not, because this represents real-world practices. Many physicians try DR as a “middle ground” to balance anticoagulation against bleeding in their risk assessment strategies. This is exemplified in our study by the very high bleeding rate (9.3% or 14.3% per patient year for CRNMB or major bleed, respectively) in the therapeutic anticoagulation phase. Another contributing factor is that our study was performed at a low-income urban center with primarily Hispanic and Black patients, an understudied group and with a large proportion of females. These factors have each been found individually to be associated with elevated bleeding risk.23-25
Despite the limitations, we believe these results are hopeful. They provide initial evidence of the efficacy of DR-apixaban in a population with severe obesity. Because current practice patterns do use apixaban in patients with severe obesity during the therapeutic period, this aids in addressing a critical need in the literature regarding management of anticoagulation after this initial period.2,3,12 Given the high-risk nature of this population, a large prospective multicenter trial examining this topic is critical.
The project was approved under expedited review by our institutional review board.
Contribution: E.A.R. participated in data collection, case review, data processing, and manuscript drafting and revision; N.W.T. participated in project design, SQL query, data collection, data processing, and manuscript drafting and revision; N.F. and P.P. participated in data collection and provided final approval of the manuscript; and H.H.B. contributed to project design, SQL query, case review, data processing, manuscript drafting and revision.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Eliot Rapoport, Department of Medicine, Montefiore Einstein Medical Center, 1825 Eastchester Rd, Bronx 10467, NY; email: erapoport@montefiore.org.
