Abstract
Background
Upper extremity deep vein thrombosis (UEDVT) is common in patients with cancer. Data on recurrence and bleeding after cancer-associated UEDVT are limited.
Objectives
The objectives were to identify factors associated with venous thromboembolism (VTE) recurrence and bleeding in patients with cancer and UEDVT, overall and by cancer site, during and after anticoagulation.
Methods
We analyzed patients from the Registro Informatizado de Pacientes con Enfermedad TromboEmbólica (RIETE), an international, prospective, observational registry of objectively confirmed VTE. We estimated cumulative incidence functions at 6, 12, 18, and 24 months and used proportional subdistribution hazard models to assess associations. Analyses were performed during and after discontinuation, using a landmark at the day of discontinuation if anticoagulation was stopped within 180 days or at day 180 if anticoagulation continued more than 180 days.
Results
Of 5,195 patients with UEDVT, 2,210 had cancer. At 24 months, the cumulative incidence of recurrent VTE was 3.9% (95% CI: 3.1-4.9) in patients with cancer vs 2.5% (95% CI: 1.8-3.5) in those without cancer, and the cumulative incidence of bleeding was 6.8% (95% CI: 5.7-8.1) vs 4.3% (95% CI: 3.4-5.3), respectively (both P < 0.001). Among patients with cancer, 24-month recurrence varied by cancer site: lung, 6.4% (95% CI: 4.0-9.5); gastrointestinal, 4.0% (95% CI: 2.6-6.0); genitourinary, 3.6% (95% CI: 1.6-6.8); breast, 2.5% (95% CI: 1.2-4.5); and hematologic, 1.4% (95% CI: 0.5-3.4). During anticoagulation, lung cancer (subdistribution hazard ratio [sHR]: 10.1; 95% CI: 1.31-78.06) and younger age (per-year sHR: 0.97; 95% CI: 0.95-0.99) were associated with recurrence, whereas genitourinary cancer (sHR: 4.95; 95% CI: 1.16-21.17) and transient risk factors (sHR: 1.75; 95% CI: 1.16-2.64) were associated with bleeding. After anticoagulation discontinuation, 91 to 180 days (sHR: 0.49; 95% CI: 0.24-1.00) and more than 180 days (sHR: 0.35; 95% CI: 0.17-0.71) of anticoagulation were associated with lower recurrence vs 30 days or less.
Conclusions
Cancer-associated UEDVT carries higher risks of recurrence and bleeding than noncancer UEDVT. Lung cancer showed the greatest recurrence during treatment. Anticoagulation beyond 90 days was associated with a lower post-treatment recurrence. (Computerized Registry of Patients With Venous Thromboembolism [RIETE]; NCT02832245)
Key Words: bleeding, cancer, recurrence, upper extremity deep vein thrombosis, venous thromboembolism
Central Illustration
Cancer substantially increases the risk of venous thromboembolism (VTE), including recurrent events despite anticoagulation.1,2 Although lower extremity deep vein thrombosis (LEDVT) and pulmonary embolism (PE) are well characterized, upper extremity DVT (UEDVT) remains comparatively underdescribed in oncology. In contemporary practice, UEDVT accounts for a growing share of VTE, driven largely by central venous access devices (CVADs) used for systemic therapy and supportive care.3 Although UEDVT appears less frequently complicated by PE than lower limb DVT,4 affected patients often have complex morbidity, treatment interruptions, and meaningful mortality.5
Mechanistically, cancer-associated VTE reflects tumor-driven hypercoagulability, as well as endothelial injury and local stasis related to CVADs.6, 7, 8 Outcomes likely vary across tumor sites: some cancers have strong prothrombotic phenotypes (eg, lung cancer), whereas others confer higher mucosal bleeding risk during anticoagulation (eg, gastrointestinal and genitourinary cancers).9, 10, 11 Prior work in cancer-associated thrombosis (CAT) shows substantial heterogeneity in recurrence and bleeding by cancer site,9,10 yet most studies have pooled upper and lower extremity events or analyzed mixed VTE populations, limiting inferences specific to UEDVT.3,12 Furthermore, optimal treatment duration for cancer-associated VTE is uncertain, particularly in patients with CVADs and amid the evolving use of direct oral anticoagulants (DOACs).6, 7, 8,11
Robust estimation is methodologically challenging because the competing risk of death is high in cancer cohorts. Conventional Kaplan-Meier methods can overestimate event probabilities when death precludes observation of VTE or bleeding. Competing-risk approaches offer more appropriate probability estimates and effect measures in this setting.13
Using the Registro Informatizado de la Enfermedad TromboEmbólica (RIETE) registry, we aimed to define the clinical course of UEDVT in patients with and without cancer. Specifically, we estimated the incidence and timing of recurrent VTE and bleeding by cancer type during anticoagulation and after treatment discontinuation and identified predictors using proportional subdistribution hazard models. We also evaluated whether anticoagulation duration was associated with recurrence after treatment discontinuation across tumor sites.
Methods
Study design and data source
We performed a cohort analysis using RIETE (ClinicalTrials.gov NCT02832245), an ongoing international registry of objectively confirmed VTE. The registry methodology has been described elsewhere.14 All patients provided written informed consent. The study complied with the Declaration of Helsinki and was approved by local ethics committees at participating centers.
Participants
We included consecutive adults with an index UEDVT confirmed by Doppler ultrasonography or computed tomography. Patients who initiated anticoagulation more than 7 days before or after UEDVT diagnosis were excluded. Cancer-associated UEDVT was defined as UEDVT in patients with active cancer, defined as cancer diagnosed within the 6 months preceding or after VTE diagnosis, metastatic cancer, or any cancer treatment at the time of VTE, including surgery, chemotherapy, radiotherapy, immunotherapy, or hormone therapy. Patients with cancer in remission were classified in the noncancer group.
Outcomes and definitions
The primary outcomes were recurrent VTE, defined as objectively confirmed DVT in the upper or lower extremities or PE, excluding superficial vein thrombosis, and bleeding, classified as major or clinically relevant nonmajor according to the International Society on Thrombosis and Haemostasis criteria. Transient provoking factors for VTE included recent surgery within 2 months, immobilization for 4 days or longer, recent travel of more than 6 hours within 3 weeks, hormone therapy, pregnancy, or postpartum status within 2 months.
Anticoagulation
Initial and long-term regimens followed local practice and included low-molecular-weight heparin (LMWH), vitamin K antagonists, DOACs, or other agents. Treatment duration was categorized as 30 days or less, 31 to 90 days, 91 to 180 days, or more than 180 days.
Statistical analysis
Continuous variables are reported as medians with 25th-75th percentiles (Q1-Q3), and categorical variables are reported as counts and percentages. Between-group comparisons used Kruskal-Wallis tests for continuous variables and Fisher exact tests for categorical variables.
We conducted 2 time-to-event analyses. For the on-treatment analysis, follow-up began at anticoagulation initiation and ended at the first occurrence of the outcome of interest (recurrent VTE or bleeding), anticoagulation discontinuation, death, or end of follow-up. For the post-treatment landmark analysis, we defined a prespecified index date as the actual stop date for patients treated for 180 days or less, or day 180 for patients treated for more than 180 days. Patients who experienced recurrent VTE or bleeding before the index date were excluded from the post-treatment analysis by design. Anticoagulation duration category (≤30, 31-90, 91-180, or >180 days) was included as a covariate in post-treatment models, with 30 days or less as the reference.
To appropriately account for the competing risk of death, we used the Aalen-Johansen estimator to generate cumulative incidence functions (CIFs) and compute cumulative incidence estimates of recurrent VTE and bleeding at 6, 12, 18, and 24 months with 95% CIs. Differences in cumulative incidence between the groups were compared using Gray's test. Associations were assessed using proportional subdistribution hazard models, adjusting for age, sex, catheter-related thrombosis (CRT), transient provoking factors, cancer site (breast, lung, gastrointestinal, and genitourinary), and metastatic status. Subdistribution hazard ratios (sHRs) with 95% CIs are reported. The proportional subdistribution hazard assumption was assessed graphically using log-log plots of the CIFs, which revealed parallel curves across groups. Hematologic malignancies were excluded from multivariable models because metastatic status is not applicable and could not be harmonized with the solid-tumor covariate structure.
For cancer-site analyses, we displayed the 5 most frequent cancer categories separately: gastrointestinal, breast, lung, genitourinary, and hematologic malignancies. Patients with other cancer sites were included in the overall cancer analyses but were not displayed separately in site-stratified figures because of heterogeneous tumor types and small numbers. All tests were 2-sided with α = 0.05. Analyses were conducted with R Statistical Software (version 4.3.2; R Core Team, 2023). A P value of less than 0.05 was considered statistically significant.
Results
Study population
Between March 2001 and October 2024, a total of 5,195 patients with UEDVT were included in the RIETE registry. Among them, 2,210 (42.5%) had active cancer at UEDVT diagnosis. Of these, 1,951 (88.3%) had solid tumors and 286 (11.7%) had hematologic malignancies. The most frequent cancer sites were gastrointestinal (n = 684), breast (n = 408), lung (n = 363), genitourinary (n = 268), and hematologic (n = 286).
Compared with patients without cancer, those with cancer were older (median age: 63 [Q1-Q3: 52-70] vs 52 [Q1-Q3: 36-70] years), more likely to have catheter-related UEDVT (66.5% vs 30.6%), and more often diagnosed and managed as outpatients (65.3% vs 59.4%). Additional baseline characteristics by cancer status and site are presented in Table 1. Changes over time in the type of anticoagulant therapy administered for UEDVT, cancer treatments received at the time of UEDVT diagnosis, metastatic status, and the presence of a central venous catheter are shown in Supplemental Figure 1. In recent years, DOAC use increased to 36.8%, and 83.0% of cases involved the presence of a catheter and 57.1% involved metastatic disease at VTE diagnosis.
Table 1.
Baseline Characteristics and Treatment of Patients With Upper Extremity Deep Vein Thrombosis by Cancer Type
| No Cancer (n = 2,985) | Any Cancer (n = 2,210) | GI (n = 684) | Breast (n = 408) | Lung (n = 363) | GU (n = 268) | Hematologic (n = 286) | |
|---|---|---|---|---|---|---|---|
| Demographics | |||||||
| Age, y | 52 (36-70) | 63 (52-70) | 63 (56-70) | 58 (49-69) | 63 (54-69) | 64 (54-74) | 56 (39-69) |
| Male sex | 1,582 (53.0) | 1,131 (51.2) | 457 (66.8) | 8 (2.0) | 257 (70.8) | 153 (57.1) | 142 (49.7) |
| BMI, kg/m2 | 26 (23-29) | 25 (23-29) | 25 (23-28) | 27 (23-30) | 25 (22-28) | 26 (23-30) | 25 (23-28) |
| Outpatients at UEDVT diagnosis | 1,705 (59.4) | 1,394 (65.3) | 456 (69.2) | 303 (77.7) | 227 (64.5) | 156 (59.8) | 136 (48.6) |
| History of VTE | 318 (10.7) | 180 (8.1) | 51 (7.5) | 27 (6.6) | 29 (8.0) | 30 (11.2) | 21 (7.3) |
| Initial presentation | |||||||
| Asymptomatic UEDVT | 43 (1.4) | 69 (3.1) | 20 (2.9) | 12 (2.9) | 7 (1.9) | 7 (2.6) | 5 (1.7) |
| Symptomatic UEDVT | 2,672 (89.5) | 1,984 (89.8) | 623 (91.1) | 375 (91.9) | 316 (87.1) | 231 (86.2) | 268 (93.7) |
| UEDVT plus PE | 270 (9.0) | 157 (7.1) | 41 (6.0) | 21 (5.1) | 40 (11.0) | 30 (11.0) | 13 (4.5) |
| Transient risk factors | 1,095 (36.7) | 629 (28.5) | 199 (29.1) | 139 (34.1) | 78 (21.5) | 89 (33.2) | 49 (17.1) |
| Catheter related | 880 (30.6) | 1,434 (66.5) | 533 (80.4) | 284 (70.5) | 146 (41.6) | 154 (58.6) | 214 (76.2) |
| Metastases | – | 994 (51.7) | 347 (53.5) | 156 (40.0) | 253 (73.3) | 102 (43.4) | – |
| Recent surgerya | 376 (12.6) | 369 (16.7) | 143 (20.9) | 51 (12.5) | 36 (9.9) | 62 (23.1) | 17 (5.9) |
| Oncologic treatments | |||||||
| Chemotherapy | – | 1,385 (67.4) | 469 (72.4) | 270 (69.9) | 216 (63.7) | 141 (59.2) | 197 (74.3) |
| Radiotherapy | – | 263 (13.5) | 79 (12.8) | 41 (11.0) | 86 (26.7) | 24 (10.7) | 13 (5.3) |
| Hormone therapy | – | 126 (8.3) | 7 (1.5) | 94 (29.8) | 2 (0.8) | 21 (11.9) | 1 (0.6) |
| Immunotherapy | – | 82 (10.0) | 20 (7.5) | 12 (7.6) | 26 (22.2) | 4 (4.5) | 10 (10.0) |
| Anticoagulant drugs | |||||||
| Low-molecular-weight heparin | 880 (28.6) | 1,706 (77.5) | 557 (81.4) | 287 (70.3) | 309 (85.1) | 180 (67.2) | 219 (77.0) |
| Vitamin K antagonists | 1,403 (47.2) | 226 (10.3) | 54 (7.9) | 52 (12.7) | 15 (4.1) | 54 (20.1) | 28 (10.0) |
| Direct oral anticoagulants | 650 (21.8) | 209 (9.5) | 57 (8.3) | 56 (13.7) | 28 (7.7) | 27 (10.1) | 21 (7.5) |
| Days of AC | 124 (94-225) | 110 (86-200) | 117 (89-203) | 134 (94-257) | 95 (49-160) | 125 (92-213) | 95 (46-156) |
| ≤30 | 197 (6.7) | 256 (11.6) | 65 (9.6) | 22 (5.5) | 62 (17.1) | 19 (7.2) | 58 (20.3) |
| 31-90 | 294 (9.8) | 395 (17.9) | 122 (17.8) | 42 (10.3) | 93 (25.6) | 37 (13.8) | 65 (22.7) |
| 91-180 | 1,410 (47.2) | 906 (41.0) | 288 (42.1) | 185 (45.3) | 129 (35.5) | 118 (44.0) | 108 (37.8) |
| >180 | 1,084 (36.3) | 653 (29.5) | 209 (30.5) | 159 (38.9) | 79 (21.8) | 94 (35.0) | 55 (19.2) |
Data are presented as median (Q1-Q3) or n (%). Percentages were calculated using the number of patients with available data for each variable as the denominator. Q1-Q3 indicates 25th-75th percentiles.
AC = anticoagulation; BMI = body mass index; GI = gastrointestinal; GU = genitourinary; PE = pulmonary embolism; UEDVT = upper extremity deep vein thrombosis; VTE = venous thromboembolism.
Recent surgery was captured as a transient VTE risk factor and does not necessarily indicate oncologic surgery.
Recurrent VTE and bleeding during anticoagulation
Over 24 months of follow-up during anticoagulation, patients with cancer had a significantly higher cumulative incidence of recurrent VTE than patients without cancer: 3.9% (95% CI: 3.1-4.9) vs 2.5% (95% CI: 1.8-3.5). Similarly, the cumulative incidence of any bleeding during anticoagulation was higher in patients with cancer: 6.8% (95% CI: 5.7-8.1) vs 4.3% (95% CI: 3.4-5.3) (Table 2, Figure 1).
Table 2.
Clinical Outcomes by Cancer Status and Cancer Site
| No Cancer (n = 2,985) | All Cancer (n = 2,210) | GI (n = 684) | Breast (n = 408) | Lung (n = 363) | GU (n = 268) | Hematologic (n = 286) | |
|---|---|---|---|---|---|---|---|
| VTE recurrences | 151 (5.1) | 177 (8.0) | 52 (7.6) | 28 (6.9) | 37 (10.2) | 15 (5.6) | 23 (8.0) |
| Isolated DVT | 100 (66.2) | 114 (64.4) | 30 (57.7) | 21 (75.0) | 26 (70.3) | 10 (66.7) | 14 (60.9) |
| Concomitant DVT and PE | 9 (6.0) | 11 (6.2) | 3 (5.8) | 2 (7.1) | 4 (10.8) | 0 (0.0) | 0 (0.0) |
| Isolated PE | 25 (16.6) | 36 (20.3) | 15 (28.8) | 3 (10.7) | 6 (16.2) | 4 (26.7) | 4 (17.4) |
| Other sites/not reported | 3 (2.0) | 6 (3.4) | 3 (5.8) | 0 (0.0) | 0 (0.0) | 1 (6.7) | 1 (4.3) |
| Superficial vein thrombosis | 14 (9.3) | 10 (5.6) | 1 (1.9) | 2 (7.1) | 1 (2.7) | 0 (0.0) | 4 (17.4) |
| Days from DVT to recurrence | 249 (67-649) | 118 (40-361) | 132 (49-434) | 249 (46-648) | 68 (34-179) | 111 (42-303) | 50 (34-139) |
| Any bleeding | 131 (4.4) | 184 (8.3) | 61 (8.9) | 23 (5.6) | 42 (11.6) | 25 (9.3) | 18 (6.3) |
| Major bleeding | 46 (35.1) | 72 (39.1) | 24 (39.3) | 6 (26.1) | 19 (45.2) | 9 (36.0) | 7 (38.9) |
| Nonmajor bleeding | 85 (64.9) | 112 (60.9) | 37 (60.7) | 17 (73.9) | 23 (54.8) | 16 (64.0) | 11 (61.1) |
| Days from DVT to bleeding | 58 (14-231) | 60 (15-153) | 47 (15-199) | 73 (25-142) | 60 (15-138) | 64 (12-133) | 36 (15-101) |
| All-cause death | 197 (6.6) | 724 (32.8) | 227 (33.2) | 71 (17.4) | 208 (57.3) | 60 (22.4) | 87 (30.4) |
| Causes of death | |||||||
| PE | 7 (3.6) | 9 (1.2) | 3 (1.3) | 1 (1.4) | 2 (1.0) | 1 (1.7) | 1 (1.1) |
| Bleeding | 4 (2.0) | 16 (2.2) | 5 (2.2) | 0 (0.0) | 7 (3.4) | 1 (1.7) | 1 (1.1) |
| Days from DVT to death | 103 (32-567) | 119 (41-309) | 152 (52-366) | 167 (58-431) | 64 (32-184) | 220 (44-421) | 136 (50-329) |
Data are presented as median (Q1-Q3) or n (%). Q1-Q3 indicates 25th-75th percentiles. Values are crude counts and percentages over overall follow-up from upper extremity deep vein thrombosis diagnosis and are not cumulative incidence estimates.
DVT = deep vein thrombosis; GI = gastrointestinal; GU = genitourinary; PE = pulmonary embolism; VTE = venous thromboembolism.
Figure 1.
Cumulative Incidence of Recurrent VTE and Any Bleeding During Anticoagulation Patients With vs Without Cancer
Cumulative incidence of recurrent venous thromboembolism (VTE) (A) and any bleeding (B) during anticoagulation in patients with vs without cancer after anticoagulation initiation. Curves represent cumulative incidence functions estimated using the Aalen-Johansen method, accounting for the competing risk of death. P values were calculated using Gray's test.
Among all patients with UEDVT during 2017 to 2024, a total of 120 recurrent VTE events occurred during anticoagulation, including 68 UEDVT events (56.7%), 26 PE events (21.7%), 17 LEDVT events (14.2%), and 9 combined LEDVT/PE events (7.5%). Among patients with UEDVT and cancer, 67 recurrent VTE events occurred during anticoagulation, including 37 UEDVT events (55.2%), 13 PE events (19.4%), 11 LEDVT events (16.4%), and 6 combined LEDVT/PE events (9.0%). Among patients with UEDVT without cancer, 53 recurrent VTE events occurred, including 31 UEDVT events (58.5%), 13 PE events (24.5%), 6 LEDVT events (11.3%), and 3 LEDVT/PE events (5.7%).
During 2017-2024, the median time to recurrent UEDVT was 4 months (Q1-Q3: 1.0-12.5 months) among all patients with UEDVT, and recurrence was ipsilateral in 63.3% of cases. Among patients with UEDVT and cancer, the median time to recurrent UEDVT was 2 months (Q1-Q3: 1.0-5.0 months), and recurrence was ipsilateral in 57.4% of cases.
Outcomes by cancer site (during anticoagulation)
When stratified by cancer site, the 24-month cumulative incidence of recurrent VTE during anticoagulation was highest in patients with lung cancer (6.4% [95% CI: 4.0-9.5]), followed by gastrointestinal cancer (4.0% [95% CI: 2.6-6.0]), genitourinary cancer (3.6% [95% CI: 1.6-6.8]), breast cancer (2.5% [95% CI: 1.2-4.5]), and hematologic malignancies (1.4% [95% CI: 0.5-3.4]) (Figure 2A, Table 3). Bleeding risk during anticoagulation also varied by cancer site, with the highest 24-month cumulative incidences of any bleeding observed in patients with genitourinary cancer (10.3% [95% CI: 6.4-15.3]) and lung cancer (8.6% [95% CI: 5.6-12.3]) (Table 3); minor and major bleeding are shown separately in Figures 2B and 2C.
Figure 2.
Cumulative Incidence of Recurrent VTE and Bleeding During Anticoagulation and After Anticoagulation Discontinuation
Cumulative incidence of recurrent venous thromboembolism (VTE) (A), minor bleeding (B), and major bleeding (C) after upper extremity deep vein thrombosis according to cancer type during anticoagulation; cumulative incidence of recurrent VTE after anticoagulation discontinuation according to treatment duration before discontinuation (D); cumulative incidence of recurrent VTE after anticoagulation discontinuation according to cancer type (E); and cumulative incidence of any bleeding after anticoagulation discontinuation (F). Curves represent cumulative incidence functions estimated using the Aalen-Johansen method, accounting for the competing risk of death. P values were calculated using Gray's test. Cancer-site panels display the 5 most frequent cancer categories, comprising 2,009 of the 2,210 patients with active cancer; patients with other cancer sites were included in overall cancer analyses but not displayed separately. Post-treatment panels are restricted to the landmark population, which explains the lower number at risk. The total number at risk at each time point is shown below each panel.
Table 3.
Cumulative Incidence of VTE Recurrence and Bleeding During Anticoagulation and After Anticoagulation Discontinuation by Cancer Status, Cancer Site, and Anticoagulation Duration
| n | 6 mo | n | 12 mo | n | 18 mo | n | 24 mo | |
|---|---|---|---|---|---|---|---|---|
| During anticoagulation: cumulative incidence of recurrent VTE and bleeding | ||||||||
| No cancer (n = 2,985) | ||||||||
| Recurrent VTE | 40 | 1.4% (1.0-1.9) | 43 | 1.7% (1.2-2.3) | 47 | 2.1% (1.5-2.8) | 50 | 2.5% (1.8-3.5) |
| Any bleeding | 88 | 3.1% (2.5-3.8) | 92 | 3.4% (2.8-4.2) | 95 | 3.7% (3.0-4.5) | 99 | 4.3% (3.4-5.3) |
| Any cancer (n = 2,210) | ||||||||
| Recurrent VTE | 65 | 3.1% (2.4-3.9) | 73 | 3.7% (2.9-4.7) | 73 | 3.7% (2.9-4.7) | 74 | 3.9% (3.1-4.9) |
| Any bleeding | 113 | 5.4% (4.5-6.5) | 127 | 6.5% (5.5-7.7) | 129 | 6.8% (5.7-8.1) | 129 | 6.8% (5.7-8.1) |
| Gastrointestinal (n = 684) | ||||||||
| Recurrent VTE | 20 | 3.1% (2.0-4.7) | 22 | 3.6% (2.3-5.3) | 22 | 3.6% (2.3-5.3) | 23 | 4.0% (2.6-6.0) |
| Any bleeding | 28 | 4.7% (3.2-6.6) | 35 | 6.7% (4.7-9.1) | 37 | 7.6% (5.3-10.4) | 37 | 7.6% (5.3-10.4) |
| Breast (n = 408) | ||||||||
| Recurrent VTE | 8 | 2.1% (1.0-3.9) | 9 | 2.5% (1.2-4.5) | 9 | 2.5% (1.2-4.5) | 9 | 2.5% (1.2-4.5) |
| Any bleeding | 15 | 4.3% (2.5-6.9) | 17 | 5.1% (3.1-7.9) | 17 | 5.1% (3.1-7.9) | 17 | 5.1% (3.1-7.9) |
| Lung (n = 363) | ||||||||
| Recurrent VTE | 19 | 5.4% (3.3-8.1) | 21 | 6.4% (4.0-9.5) | 21 | 6.4% (4.0-9.5) | 21 | 6.4% (4.0-9.5) |
| Any bleeding | 23 | 7.4% (4.8-10.7) | 25 | 8.6% (5.6-12.3) | 25 | 8.6% (5.6-12.3) | 25 | 8.6% (5.6-12.3) |
| Genitourinary (n = 268) | ||||||||
| Recurrent VTE | 7 | 2.9% (1.3-5.7) | 8 | 3.6% (1.6-6.8) | 8 | 3.6% (1.6-6.8) | 8 | 3.6% (1.6-6.8) |
| Any bleeding | 18 | 8.8% (5.4-13.2) | 20 | 10.3% (6.4-15.3) | 20 | 10.3% (6.4-15.3) | 20 | 10.3% (6.4-15.3) |
| Hematologic (n = 286) | ||||||||
| Recurrent VTE | 4 | 1.4% (0.5-3.4) | 4 | 1.4% (0.5-3.4) | 4 | 1.4% (0.5-3.4) | 4 | 1.4% (0.5-3.4) |
| Any bleeding | 2 | 1.6% (0.3-5.1) | 2 | 1.6% (0.3-5.1) | 2 | 1.6% (0.3-5.1) | 2 | 1.6% (0.3-5.1) |
| After anticoagulation discontinuation: cumulative incidence of recurrent VTE and bleeding by cancer status and cancer site | ||||||||
| No cancer (n = 2,061) | ||||||||
| Recurrent VTE | 43 | 2.7% (2.0-3.6) | 58 | 4.1% (3.1-5.3) | 68 | 5.6% (4.3-7.1) | 76 | 7.0% (5.4-8.8) |
| Any bleeding | 49 | 2.6% (1.9-3.4) | 53 | 3.0% (2.2-3.9) | 57 | 3.6% (2.7-4.7) | 63 | 4.6% (3.5-6.1) |
| Any cancer (n = 1,458) | ||||||||
| Recurrent VTE | 65 | 5.0% (3.9-6.2) | 75 | 6.2% (4.9-7.7) | 87 | 8.0% (6.4-9.8) | 93 | 9.2% (7.4-11.3) |
| Any bleeding | 97 | 7.1% (5.8-8.5) | 106 | 8.2% (6.8-9.8) | 109 | 8.7% (7.1-10.4) | 112 | 9.3% (7.6-11.1) |
| Gastrointestinal (n = 464) | ||||||||
| Recurrent VTE | 17 | 4.0% (2.4-6.2) | 21 | 5.4% (3.4-8.1) | 25 | 7.3% (4.8-10.6) | 26 | 8.0% (5.2-11.5) |
| Any bleeding | 34 | 7.9% (5.6-10.8) | 39 | 9.9% (7.1-13.2) | 40 | 10.4% (7.5-13.9) | 41 | 11.1% (7.9-14.8) |
| Breast (n = 292) | ||||||||
| Recurrent VTE | 8 | 3.0% (1.4-5.6) | 9 | 3.5% (1.7-6.4) | 14 | 7.7% (4.2-12.6) | 16 | 9.9% (5.5-15.7) |
| Any bleeding | 10 | 3.5% (1.8-6.0) | 10 | 3.5% (1.8-6.0) | 10 | 3.5% (1.8-6.0) | 10 | 3.5% (1.8-6.0) |
| Lung (n = 198) | ||||||||
| Recurrent VTE | 19 | 10.4% (6.5-15.4) | 21 | 12.2% (7.7-17.8) | 23 | 14.2% (9.1-20.3) | 25 | 17.2% (11.0-24.6) |
| Any bleeding | 24 | 12.9% (8.5-18.2) | 25 | 13.7% (9.1-19.3) | 25 | 13.7% (9.1-19.3) | 26 | 15.0% (9.9-21.1) |
| Genitourinary (n = 179) | ||||||||
| Recurrent VTE | 3 | 2.0% (0.5-5.3) | 4 | 3.0% (1.0-7.2) | 5 | 4.8% (1.6-10.9) | 5 | 4.8% (1.6-10.9) |
| Any bleeding | 10 | 5.8% (3.0-10.0) | 10 | 5.8% (3.0-10.0) | 11 | 7.3% (3.6-12.7) | 11 | 7.3% (3.6-12.7) |
| Hematologic (n = 195) | ||||||||
| Recurrent VTE | 10 | 5.5% (2.8-9.5) | 11 | 6.3% (3.3-10.7) | 11 | 6.3% (3.3-10.7) | 11 | 6.3% (3.3-10.7) |
| Any bleeding | 10 | 5.2% (2.7-9.1) | 13 | 7.8% (4.3-12.7) | 13 | 7.8% (4.3-12.7) | 13 | 7.8% (4.3-12.7) |
| After anticoagulation discontinuation: cumulative incidence of recurrent VTE by anticoagulation duration | ||||||||
| ≤30 d | 14 | 10.8% (6.2-16.8) | 14 | 10.8% (6.2-16.8) | 15 | 12.2% (7.0-19.0) | 15 | 12.2% (7.0-19.0) |
| 31-90 d | 15 | 8.3% (4.9-13.0) | 16 | 9.1% (5.4-14.0) | 16 | 9.1% (5.4-14.0) | 16 | 9.1% (5.4-14.0) |
| 91-180 d | 17 | 4.3% (2.6-6.7) | 23 | 6.9% (4.4-10.1) | 26 | 8.5% (5.6-12.2) | 27 | 9.2% (6.0-13.1) |
| >180 d | 11 | 2.3% (1.2-4.0) | 13 | 2.9% (1.6-4.9) | 21 | 6.3% (3.9-9.5) | 25 | 8.8% (5.6-12.8) |
Values are number of events and cumulative incidence (95% confidence intervals).
Analysis period: March 2001 to October 2024.
VTE = venous thromboembolism.
Posttreatment outcomes (landmark analysis)
Among patients who remained event-free through anticoagulation and entered the post-treatment period, recurrent VTE continued to accrue. At 24 months after anticoagulation discontinuation, the cumulative incidence of recurrent VTE was 17.2% (95% CI: 11.0-24.6) in patients with lung cancer, 9.9% (95% CI: 5.5-15.7) in those with breast cancer, 8.0% (95% CI: 5.2-11.5) in those with gastrointestinal cancer, 6.3% (95% CI: 3.3-10.7) in those with hematologic malignancies, and 4.8% (95% CI: 1.6-10.9) in those with genitourinary cancer (Table 3, Figure 2E). At the same time point, the cumulative incidence of any bleeding was highest in patients with lung cancer (15.0% [95% CI: 9.9-21.1]) and gastrointestinal cancer (11.1% [95% CI: 7.9-14.8]) (Table 3, Figure 2F). Recurrence after anticoagulation discontinuation varied by prior treatment duration. At 6 months, the cumulative incidence of recurrent VTE was 10.8% (95% CI: 6.2-16.8) after 30 days or less of anticoagulation, 8.3% (95% CI: 4.9-13.0) after 31 to 90 days, 4.3% (95% CI: 2.6-6.7) after 91 to 180 days, and 2.3% (95% CI: 1.2-4.0) after more than 180 days. These trends persisted at 12 and 24 months, with the lowest recurrence observed after more than 180 days of therapy (Figure 2D, Table 3).
During 2017-2024, among all patients with UEDVT, 92 recurrent VTE events occurred after anticoagulation discontinuation, including 50 UEDVT events (54.3%), 20 PE events (21.7%), 13 LEDVT events (14.1%), and 9 LEDVT/PE events (9.8%). Among patients with UEDVT and cancer, 47 recurrent VTE events occurred after anticoagulation discontinuation, including 24 UEDVT events (51.1%), 9 PE events (19.1%), 8 LEDVT events (17.0%), and 6 LEDVT/PE events (12.8%). Among patients with UEDVT without cancer, 45 recurrent VTE events occurred, including 26 UEDVT events (57.8%), 11 PE events (24.4%), 5 LEDVT events (11.1%), and 3 LEDVT/PE events (6.7%).
Predictors of recurrent VTE and bleeding
In multivariable proportional subdistribution hazard models (Figure 3), lung cancer was independently associated with a higher risk of recurrent VTE during anticoagulation (sHR: 10.11; 95% CI: 1.31-78.06; P = 0.027). Younger age also modestly increased recurrence risk (per-year sHR: 0.97; 95% CI: 0.95-0.99; P = 0.012). For bleeding, genitourinary cancer (sHR: 4.95; 95% CI: 1.16-21.17; P = 0.031) and the presence of transient provoking factors (sHR: 1.75; 95% CI: 1.16-2.64; P = 0.008) were associated with increased risk.
Figure 3.
sHRs for Recurrent VTE and Bleeding in Patients With Cancer Enrolled in Registro Informatizado de la Enfermedad TromboEmbólica
Forest plots displaying subdistribution hazard ratios (sHRs) and 95% confidence intervals (CIs) from proportional subdistribution hazard models. Adjusted models include all variables listed in the figure. (A) Risk factors for recurrent VTE during anticoagulation (AC). (B) Risk factors for any bleeding during AC. (C) Risk factors for recurrent VTE after AC discontinuation. ∗Compared with AC <30 days. VTE = venous thromboembolism.
In the post-anticoagulation period, anticoagulation for more than 180 days was associated with a lower risk of recurrence compared with 30 days or less (sHR: 0.35; 95% CI: 0.17-0.71; P = 0.004), whereas anticoagulation for 91 to 180 days showed a borderline, nonsignificant association (sHR: 0.49; 95% CI: 0.24-1.00; P = 0.050) and no association for anticoagulation between 31 and 90 days (sHR: 0.74; 95% CI: 0.33-1.68; P = 0.48).
Anticoagulation and cancer therapies
Among patients with cancer, LMWH was the predominant initial therapy (77.5%), followed by vitamin K antagonists (10.3%) and DOACs (9.5%). The median anticoagulation duration was 110 days (Q1-Q3: 86-200 days). Concomitant cancer treatments included chemotherapy (67.4%), radiotherapy (13.5%), immunotherapy (10.0%), and hormone therapy (8.3%).
Mortality
At 24 months, all-cause mortality was significantly higher in patients with cancer than in those without cancer (43.7% [95% CI: 40.7-46.6] vs 8.2% [95% CI: 6.7-9.6]; P < 0.001). Among patients without cancer, fatal PE accounted for more deaths than fatal bleeding (7 vs 4 deaths), whereas among patients with cancer, fatal bleeding accounted for more deaths than fatal PE (16 vs 9 deaths). Although deaths directly attributed to PE or bleeding were rare in both groups, the pattern of fatal events differed by cancer status, suggesting a different risk-benefit balance for anticoagulation in cancer-associated UEDVT.
Discussion
In this large prospective cohort from RIETE, among patients with UEDVT, cancer was associated with a higher cumulative incidence of recurrent VTE and bleeding during anticoagulation and after treatment discontinuation (Central Illustration). These risks were not uniform across cancer sites: lung cancer showed the highest recurrence during anticoagulation, whereas genitourinary and lung cancers had the highest bleeding rates. Importantly, longer anticoagulation, particularly more than 180 days, was associated with a lower risk of post-treatment recurrence, with a borderline association observed for 91 to 180 days. These findings support tumor site and anticoagulation duration as important determinants of risk in cancer-associated UEDVT.
Central Illustration.
Upper Extremity Deep Vein Thrombosis in Patients With Cancer: Clinical Course and Risk Profile
sHR = subdistribution hazard ratio; VTE = venous thromboembolism.
Prior RIETE analyses and other large cohorts have documented tumor-specific variation in VTE outcomes among patients with LEDVT and PE.15 Our study extends this paradigm to UEDVT, a distinct phenotype frequently linked to CVADs and often under-represented in site-specific analyses. The high on-treatment recurrence observed in lung cancer is consistent with its prothrombotic biology, reinforcing that tumor site remains a relevant risk factor even when the index DVT involves the upper extremity. In contrast, patients with hematologic malignancies had comparatively lower recurrence during treatment, but still exhibited meaningful risk after discontinuation, underscoring the need for ongoing vigilance even after stopping therapy.
The increasing proportion of catheter-related UEDVT over time likely reflects the expanding use and longer duration of central venous access in oncology, including peripherally inserted central catheter lines, outpatient systemic therapy, and prolonged supportive care in patients living longer with cancer. However, RIETE does not capture catheter type, insertion technique, catheter care protocols, thromboprophylaxis policies, or institutional prevention strategies. Therefore, temporal trends cannot be attributed to a specific preventive strategy or to line use alone and may also reflect changes in cancer care pathways, survival, and detection practices.
Mechanistically, cancer-associated UEDVT may reflect the convergence of systemic hypercoagulability, inflammation, and local endothelial injury, often related to catheters. Lung cancer's thrombotic profile may be related to high tissue factor expression, circulating procoagulants, and heightened inflammatory activation. Elevated bleeding risk in genitourinary cancers and, to a lesser extent, lung cancer aligns with mucosal vulnerability and frequent instrumentation, which can be exacerbated by anticoagulant exposure. These dual risks of thrombosis and bleeding help explain the divergent cumulative incidence curves observed across tumor sites.
These findings may have clinical implications. First, longer anticoagulation may be considered for patients with cancer-associated UEDVT, especially those with lung cancer, if bleeding risk is acceptable. The observed inverse association between treatment duration and post-treatment recurrence suggests that stopping therapy before 90 days may be associated with higher recurrence risk. Although bleeding risk must always be balanced, these data support caution against premature discontinuation in patients with acceptable bleeding profiles. Second, in bleeding-prone cancers such as genitourinary cancer, clinicians should consider bleeding-aware strategies, including careful agent selection, minimization of concomitant antithrombotic therapy, attention to reversible bleeding sources, and use of LMWH when appropriate. When DOACs are chosen, tumor location and mucosal involvement should inform shared decision-making.
Third, although catheter-related UEDVT was common and adjusted for in our models, cancer site remained an independent predictor of both recurrence and bleeding, suggesting that tumor biology, not catheter mechanics alone, plays an important role. Catheter management, including functional assessment, prompt removal when no longer needed, and infection control, should complement anticoagulation. Finally, the landmark analysis showed that among patients who remained event-free through the initial period, recurrence risk increased in the months after anticoagulation discontinuation, particularly with lung cancer. This highlights the value of structured reassessment before stopping anticoagulation, incorporating cancer status, treatment response, ongoing catheter need, and bleeding trajectory.
From a methodological perspective, we addressed the competing risk of death, a major consideration in cancer populations, by reporting CIFs and applying proportional subdistribution hazard models to provide clinically interpretable probabilities and adjusted effect estimates. Our landmark analysis mitigated immortal-time bias by aligning post-treatment follow-up from a standardized index point. Together, these approaches enhance the applicability of our findings to real-world decision-making.
In CAT recent guidelines,6 LMWHs remain the standard for the initial 10 days and maintenance treatment of CAT, including catheter-related UEDVT. DOACs are now considered a valid alternative to LMWHs for initial and maintenance treatment in patients without a high risk of gastrointestinal or genitourinary bleeding, strong drug-drug interactions, or gastrointestinal absorption impairment. Anticoagulation with an LMWH or a DOAC is recommended for a minimum of 6 months for CAT, including CRT. After 6 months of treatment, dose reduction may be considered for patients receiving apixaban if anticoagulation needs to be continued.16 Although DOACs offer the convenience of oral administration and are noninferior to LMWHs in efficacy, their use should be individualized based on bleeding risk, drug interactions, and patient preference. LMWHs remain the standard in high-risk or unstable patients.
Management of CRT in patients with cancer is evolving, with increasing acceptance of DOACs as an alternative to LMWHs. However, the decision to retain or remove the catheter should be based on infection, catheter function, and patient-specific factors. As with VTE in any location, management of UEDVT requires weighing the antithrombotic benefits of anticoagulant therapy against bleeding risk. This balance is particularly important in UEDVT because the risk of PE is lower than with lower extremity DVT, and the 3-month benefit-risk balance in an oncology population with CRT warrants consideration, with a recurrence rate of 0.56% compared with a bleeding rate of 2.34%.17
Strengths of this study include its large, international, prospective design; focus on UEDVT rather than pooled VTE; stratification by cancer site; and clear separation of on-treatment vs post-treatment outcomes. This study has several limitations. First, the RIETE registry does not capture detailed oncologic data, including treatment regimens, response to therapy, or performance status, which may influence both thrombotic and bleeding risk. Second, the observational design introduces potential confounding by indication because treatment duration and anticoagulant selection were determined by clinical judgment. Third, the landmark analysis may introduce selection bias by restricting the analysis to patients who remained event-free during the initial treatment period. In addition, several subgroup estimates, particularly for recurrent VTE by cancer site, were based on a limited number of events, resulting in wide CIs; therefore, these estimates should be interpreted with caution.
Finally, residual confounding from unmeasured variables cannot be excluded. Therefore, the associations observed in this study should be interpreted as hypothesis-generating rather than causal.
In conclusion, UEDVT in patients with cancer is a high-risk condition with clinically significant variation in recurrence and bleeding outcomes by cancer site and anticoagulation duration. These findings highlight the importance of individualized management strategies, tailored to thrombotic and bleeding risk, tumor biology, and treatment goals. Future prospective trials are needed to define optimal treatment duration and anticoagulant choice in this increasingly common population, especially in the evolving landscape of cancer therapy and CVAD use.
Perspectives.
COMPETENCY IN MEDICAL KNOWLEDGE: Cancer-associated upper extremity deep vein thrombosis is associated with higher risks of recurrent venous thromboembolism and bleeding than upper extremity deep vein thrombosis in patients without cancer, with lung cancer showing the highest recurrence and genitourinary cancer the highest bleeding risk. Anticoagulation for more than 180 days was associated with lower post-treatment recurrence, with a borderline association for 91 to 180 days, supporting consideration of prolonged therapy in selected high-risk patients with acceptable bleeding risk. Tumor site may help inform individualized anticoagulation decisions, including closer attention to recurrence risk in lung cancer and bleeding-aware strategies in genitourinary cancer.
TRANSLATIONAL OUTLOOK: Prospective trials are needed to define optimal anticoagulation duration by cancer type, including the role of direct oral anticoagulants in high-risk tumors. Biomarker-based risk stratification and multidisciplinary collaboration may help personalize upper extremity deep vein thrombosis management and improve outcomes.
Funding Support and Author Disclosures
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Acknowledgments
We express our gratitude to ROVI for supporting this Registry with an unrestricted educational grant. We also thank the RIETE Registry Coordinating Center, S&H Medical Science Service, for their quality control data, and logistic and administrative support.
Footnotes
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
Appendix
For the supplemental figure and appendix, please see the online version of this paper.
Contributor Information
Olivier Espitia, Email: olivier.espitia@chu-nantes.fr.
the RIETE Investigators:
Paolo Prandoni, Benjamin Brenner, Dominique Farge-Bancel, Raquel Barba, Peter Verhamme, Hugo Hyung Bok Yoo, Radovan Malý, Laurent Bertoletti, Sebastian Schellong, Inna Tzoran, Pierpaolo Di Micco, Abilio Reis, Marijan Bosevski, Lucia Mazzolai, Joseph A. Caprini, My Hanh Bui, A. Abad-Fernández, M.D. Adarraga, J. Aibar, A. Alberich-Conesa, A. Alda, J. Alfonso, R. Alonso-Beato, J. Álvarez-Albarrán, P. Álvarez-Vega, C. Amado, M. Angelina-García, J.I. Arcelus, A. Ballaz, R. Barba, C. Barbagelata, M. Barrón, B. Barrón-Andrés, F. Beddar-Chaib, E.I. Bencosme de Méndez, C. Benítez-Castillo, Á. Blanco-Molina, N. Chamorro, L. Chasco, G., Claver, J. Criado, C. De Juana-Izquierdo, F. Del Molino, J. Del Toro, P. Demelo-Rodríguez, A.M. Díaz-Brasero, M.C. Díaz-Pedroche, J.A. Díaz-Peromingo, Á. Dubois-Silva, D. Durán, J.C. Escribano, A.R. Exposito-Villegas, C. Falgá, L.A. Fernández-Bermejo, C. Fernández-Martínez de Septién, J.L. Fernández-Reyes, M.Á. Fidalgo, I. Francisco, C. Gabara, A. García-Ortega, F. García-Bragado, O. Gavín-Sebastián, A. Gil-Díaz, C. Gómez-Cuervo, C. Gómez-Cepeda, A. González-Munera, C. González-García, J. González-Olmedo, J. Gorostidi-Pérez, L. Guirado, J. Gutiérrez-Guisado, C. Heredia, Hernández-Blasco, J. Hernández-Borge, M.J. Hernández-Vidal, D. Jiménez, I. Jou, M.D. Joya, S. Láinez-Justo, P. Llamas, H. López-Brull, L. López-Jiménez, P. López-Miguel, J.J. López-Núñez, M. López-Rubio, A. López-Ruiz, J.B. López-Sáez, A. Lorenzo, O. Madridano, A. Maestre, P.J. Marchena, A. Martínez-Sánchez, M.I. Mercado, J. Moisés-Lafuente, M. Monreal, L. Monzón-Escribano, A. Moreno-Fernández, M.S. Navas, J.A. Nieto, M.J. Núñez-Fernández, M. Olid, M. Ortiz, J. Osorio, S. Otálora, J. Pagán, P. Parra-Caballero, J.M. Pedrajas, M. Pérez-Pinar, C. Pérez-Ductor, M.L. Peris, M.L. Pesce, J.A. Porras, L.M. Prieto-Gañán, G. Puche, S. Raposo, A. Rivas, F. Rivera-Cívico, A. Rivera-Gallego, M. Roca-Toledo, A. Rodríguez-Cobo, M.D. Rojas-Piñero, P. Ruiz-Artacho, N. Ruiz-Giménez, P. Ruiz-Torregrosa, M. Sánchez-Camacho, T. Sancho, V. Sendín, G. Sierra-Palomares, P. Sigüenza, L. Sindín-Martín, A. Solé, S. Suárez-Fernández, R. Tascón-Rodríguez, R. Tirado, M.R. Toda, C. Tolosa, J. Trujillo-Santos, L. Tudela, R. Valle, J.F. Varona, E. Vega-Romero, G. Vidal, P. Villares, J.S. You, C. Ay, S. Nopp, I. Pabinger, M. Erard, Q. Van Thillo, P. Verhamme, A.T. Rocha, H.H.B. Yoo, S.E. Carbonell, J.E. Gómez-Mesa, A.C. Montenegro, J. Roa, P. Grenar, J. Hirmerova, R. Malý, S. Accassat, L. Bertoletti, A. Bura-Riviere, J. Catella, R. Chopard, F. Couturaud, B. Crichi, O. Espitia, R. Le Mao, B. Leclercq, I. Mahé, F. Moustafa, L. Plaisance, G. Poenou, G. Sarlon-Bartoli, P. Suchon, E. Versini, S. Schellong, Y. Jenab, A. Khodayari, P. Sadeghipour, S. Yadangi, B. Brenner, I. Tzoran, G. Barillari, M. Basaglia, F. Bilora, D. Bissacco, C. Bortoluzzi, M. Bortoluzzi, B. Brandolin, G. Buso, R. Casana, M.M. Ciammaichella, P. Di Micco, A. Guida, E. Imbalzano, D. Lambertenghi-Deliliers, A. Licci, D. Mastroiacovo, U. Mazzarelli, R. Pesavento, V. Pizzuti, A. Poz, P. Prandoni, M. Provisione, P. Scarinzi, C. Siniscalchi, A. Visonà, N. Vo Hong, B. Zalunardo, H. Cesnieks, A. Skride, M. Zicans, C. Flores-Ramírez, M.Á. Mendoza Romo-Ramírez, D. Tarín-Recéndez, S. Fonseca, M. Martins, C. Pereira-Fontes, B. Soeiro, M. Bosevski, I. Kuzmanoski, M. Zdraveska, S. Barco, L. Mazzolai, D.J. Angiolillo, A. Khalil, J.A. Caprini, L. Ortega-Paz, G.M. Parmar, A.J. Tafur, M.H. Bui, S.T. Nguyen, K.Q. Pham, G.B. Tran, and Manuel Monreal
Appendix
References
- 1.Heit J.A., Mohr D.N., Silverstein M.D., Petterson T.M., O’Fallon W.M., Melton L.J. Predictors of recurrence after deep vein thrombosis and pulmonary embolism: a population-based cohort study. Arch Intern Med. 2000;160:761–768. doi: 10.1001/archinte.160.6.761. [DOI] [PubMed] [Google Scholar]
- 2.Mahajan A., Brunson A., White R., Wun T. The epidemiology of cancer-associated venous thromboembolism: an update. Semin Thromb Hemost. 2019;45:321–325. doi: 10.1055/s-0039-1688494. [DOI] [PubMed] [Google Scholar]
- 3.Ploton G., Pistorius M.-A., Raimbeau A., et al. A STROBE cohort study of 755 deep and superficial upper-extremity vein thrombosis. Medicine (Baltimore) 2020;99 doi: 10.1097/MD.0000000000018996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Goldhaber S.Z., Visani L., De Rosa M. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER) Lancet. 1999;353:1386–1389. doi: 10.1016/s0140-6736(98)07534-5. [DOI] [PubMed] [Google Scholar]
- 5.Williams G.W., Giri S., Siwakoti K., Sodhi A., Kadaria D. Incidence, risk factors and outcomes of upper extremity deep venous thrombosis among hospitalized patients in the United States. Am J Respir Crit Care Med. 2016 [Google Scholar]
- 6.Farge D., Frere C., Connors J.M., et al. 2022 international clinical practice guidelines for the treatment and prophylaxis of venous thromboembolism in patients with cancer, including patients with COVID-19. Lancet Oncol. 2022;23:e334–e347. doi: 10.1016/S1470-2045(22)00160-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Elias A., Debourdeau P., Espitia O., et al. Central venous catheter associated upper extremity deep vein thrombosis in cancer patients: diagnosis and therapeutic management. Arch Cardiovasc Dis. 2024;117:72–83. doi: 10.1016/j.acvd.2023.11.011. [DOI] [PubMed] [Google Scholar]
- 8.Mahé I., Mayeur D., Couturaud F., et al. Anticoagulant treatment of cancer-associated thromboembolism. Arch Cardiovasc Dis. 2024;117:29–44. doi: 10.1016/j.acvd.2023.11.010. [DOI] [PubMed] [Google Scholar]
- 9.Trujillo-Santos J., Ay C., Bikdeli B., et al. Variation in venous thromboembolism outcomes based on primary cancer site: a RIETE registry analysis. JACC Adv. 2025;4 doi: 10.1016/j.jacadv.2025.102101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mahé I., Chidiac J., Bertoletti L., et al. The clinical course of venous thromboembolism may differ according to cancer site. Am J Med. 2017;130:337–347. doi: 10.1016/j.amjmed.2016.10.017. [DOI] [PubMed] [Google Scholar]
- 11.Ageno W., Vedovati M.C., Cohen A., et al. Bleeding with apixaban and dalteparin in patients with cancer-associated venous thromboembolism: results from the caravaggio study. Thromb Haemost. 2021;121:616–624. doi: 10.1055/s-0040-1720975. [DOI] [PubMed] [Google Scholar]
- 12.Espitia O., Raimbeau A., Planquette B., et al. A systematic review and meta-analysis of the incidence of post-thrombotic syndrome, recurrent thromboembolism, and bleeding after upper extremity vein thrombosis. J Vasc Surg Venous Lymphat Disord. 2024;12 doi: 10.1016/j.jvsv.2023.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.van Hylckama Vlieg M.A.M., Nasserinejad K., Visser C., et al. The risk of recurrent venous thromboembolism after discontinuation of anticoagulant therapy in patients with cancer-associated thrombosis: a systematic review and meta-analysis. EClinicalMedicine. 2023;64 doi: 10.1016/j.eclinm.2023.102194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bikdeli B., Jimenez D., Hawkins M., et al. Rationale, design and methodology of the computerized registry of patients with venous thromboembolism (RIETE) Thromb Haemost. 2018;118:214–224. doi: 10.1160/TH17-07-0511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Poh C., Brunson A., Keegan T., Wun T., Mahajan A. Incidence of upper extremity deep vein thrombosis in acute leukemia and effect on mortality. TH Open. 2020;4:e309–e317. doi: 10.1055/s-0040-1718883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mahé I., Carrier M., Mayeur D., et al. Extended reduced-dose apixaban for cancer-associated venous thromboembolism. N Engl J Med. 2025;392:1363–1373. doi: 10.1056/NEJMoa2416112. [DOI] [PubMed] [Google Scholar]
- 17.Wang T.-F., Kou R., Carrier M., Delluc A. Management of catheter-related upper extremity deep vein thrombosis in patients with cancer: a systematic review and meta-analysis. J Thromb Haemost. 2024;22:749–764. doi: 10.1016/j.jtha.2023.11.017. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.