What is the Optimal Management of Upper-Extremity Superficial and Deep Vein Thrombosis?

The incidence of upper extremity deep vein thrombosis (UEDVT) is increasing, most commonly due to more widespread use of central venous catheters (CVCs), but no randomized, controlled trials have evaluated the treatment for UEDVT or upper extremity superficial vein thrombosis (UESVT), leaving several questions unanswered about the optimal management of these conditions.

To explore the current strategies used to manage UEDVT and UESVT, and to assess rates of recurrent VTE, mortality, thrombotic symptoms, and treatment-related bleeding events in patients with these conditions, Suzanne M. Bleker, MD, from the Department of Vascular Medicine at the Academic Medical Center in Amsterdam, the Netherlands, and authors collected follow-up data from the prospective, international, multicenter, diagnostic management ARMOUR (Diagnostic Algorithm in Suspected Upper Extremity Deep Vein Thrombosis) study that was published in the Journal of Thrombosis and Haemostasis. They found that the risk of recurrent VTE was low in patients with UEDVT and negligible in those with UESVT, while patients with cancer had a significant risk of recurrent VTE.

“Currently, physicians are left with expert opinions based on indirect evidence that is extrapolated from studies on DVT of the leg or pulmonary embolism,” Dr. Bleker told ASH Clinical News. “It can be reassuring to know that the risk of recurrent VTE seems relatively low after having an UEDVT and that the risk of developing post-thrombotic symptoms also seems to be quite low.”

The ARMOUR study included 406 patients with a clinical suspicion of UEDVT enrolled from 16 centers in six countries between January 2010 and June 2012. Follow-up data were collected from medical charts and through visits or telephone calls between August 2014 and January 2015 using a structured questionnaire. Dr. Bleker and colleagues analyzed outcomes data from medical charts or from the patients’ primary-care practitioners if the patient could not be contacted.

Of the entire cohort:

  • 104 patients were diagnosed with UEDVT (26%)
  • 57 patients were diagnosed with isolated UESVT (14%)

Over a median of 3.5 years, the researchers collected follow-up information for 98 percent of patients with UEDVT (n=102; mean age = 54±17 years; 43% men) and 97 percent of patients with UESVT (n=55; mean age = 56±17 years; 56% men).

One hundred patients with UEDVT (98%) and 40 patients with UESVT (73%) were started on anticoagulant treatment; the median treatment durations were 182 days and 42 days, respectively.

Characteristics of post-thrombotic management are detailed in TABLE. Low-molecular-weight heparin (LMWH) and fondaparinux were the most frequently prescribed initial anticoagulation therapies. Long-term anticoagulation therapy usually consisted of vitamin K antagonists (VKAs) for UEDVT patients, while LMWH and fondaparinux remained the preferred anticoagulants for UESVT patients.

In interviews, patients were asked to score the presence of the following post-thrombotic symptoms of the affected arm on a scale from zero (absent) to three (severe): pain, cramps, heaviness, pruritus, and paresthesia. Based on those scores, patients were grouped as having:

  • No symptoms (0)
  • Mild symptoms (1-5)
  • Moderate symptoms (6-10)
  • Severe symptoms (11-15)

Eight percent of patients with UEDVT (n=6 of 72) had moderate post-thrombotic symptoms and five percent (n=5) had major bleeding episodes. No patients with UESVT, however, experienced post-thrombotic symptoms or major bleeding.

“Post-thrombotic symptoms after UEDVT and UESVT were relatively infrequent and mostly mild,” the authors wrote, adding that “anticoagulant therapy of UEDVT appeared to carry a substantial risk of major bleeding.”

Of the 137 cancer patients included in the ARMOUR study, 43 patients were diagnosed with UEDVT and 19 with UESVT; the most common cancers in this population were breast cancer, gastrointestinal cancers, and hematologic malignancies. Compared with the 75 cancer patients in the ARMOUR trial for whom thrombosis was not detected, cancer patients with UEDVT more often had a CVC (74% vs. 40%; p<0.01) or were receiving ongoing cancer therapy at the time of enrollment (81% vs. 60%; p=0.02).

In the group of patients with UEDVT and cancer, 78 percent were managed with LMWH monotherapy for the complete treatment period, while 81 percent of patients without cancer were switched to VKAs for the long-term treatment.

Cancer patients were treated for a median of 182 days (range = 42–339), and 38 percent were treated indefinitely, compared with 13 percent of patients without cancer. Notably, 30 percent of these patients received elastic compression stockings for the arm – a management strategy that the American College of Chest Physicians does not recommend. No patients with cancer experienced anticoagulant-related major bleeding, though six patients had a clinically relevant non-major bleed after a median of 3.6 months following the start of anticoagulation.

Recurrent VTE occurred in nine patients (9%) with UEDVT – five of which occurred during anticoagulation – and one patient had UESVT. Twenty-six patients (26%) with UEDVT and 18 patients with UESVT (33%) died during study follow-up. In both cohorts, the majority of deaths were related to cancer (70% of UEDVT patients and 56% of UESVT patients).

“The data suggest a significant risk of recurrent VTE [in cancer patients], with rates that appear two-fold higher compared to non-cancer patients (18% vs. 7.5%; hazard ratio [HR] = 2.2; 95% CI 0.6-8.2 adjusted for age and sex),” Dr. Bleker and colleagues wrote. “These results support the guidelines [from the AACP] to use anticoagulant treatment for at least three to six months in patients with active cancer and UEDVT and to consider prolonging anticoagulant therapy when the cancer is not cured or the patient is still receiving cancer treatment.”

The study is limited by the retrospective data collection for certain patients and its small patient population, which did not allow the researchers to compare recurrence rates between patients with distal versus proximal UEDVT or among those whose UEDVT was related or not related to CVC. In addition, the centers participating in the trial were specialized in the diagnosis and management of VTE, thus treatment strategies captured in this study may not be generalizable to daily practice in other settings.

“Further studies are needed to inform physicians on the best management strategies for this disease, particularly in high-risk subgroups, such as cancer patients,” Dr. Bleker told ASH Clinical News. “Our results should ideally be confirmed in a prospective registry study with standardized treatment strategies, a large sample size, a long follow-up duration, and central adjudication of outcomes.”


Bleker SM, Van Es N, Kleinjan A, et al. Current management strategies and long-term clinical outcomes of upper extremity venous thrombosis. J Thromb Haemost. 2016;14:973-81.

TABLE. Type of Initial and Long-Term Treatment for Upper Extremity Venous Thrombosis
Characteristic Patients with UEDVT (n=102) Patients with UESVT (n=55)
Anticoagulant treatment started, n (%) 100 (98) 49 (73)
Initial treatment, n (%)
LMWH 88 (88) 28 (69)
UFH 3 (3) 0
Fondaparinux 8 (8) 11 (28)
DOAC (rivaroxaban) 1 (1) 0
Unknown 0 1 (3)
Long-term treatment, n (%)
LMWH 41 (41) 27 (67)
Fondaparinux 2 (2) 10 (25)
VKA 56 (56) 2 (5)
DOAC (rivaroxaban) 1 (1) 0
Unknown 1 (3)
Additional treatments, n (%)
Systemic thrombolysis 1 (1) 0
Catheter-directed thrombolysis 0 0
First rib resection 1 (1) 0
Vena cava superior filter placement 0 0
Stent placement 1 (1) 0
CVC removal within 2 weeks 2 (5) 0
Elastic compression stockings 31 (30) 17 (31)
UEDVT = upper extremity deep vein thrombosis; LMWH = low-molecular-weight heparin; UFH = unfractionated heparin; DOAC = direct oral anticoagulants; VKA = vitamin K antagonist; CVC = central venous catheters