Clinical course of upper extremity deep vein thrombosis in patients with or without cancer: a systematic review

• The incidence of upper extremity deep vein thrombosis (UEDVT) is increasing
• Only a few large studies have prospectively evaluated the natural history of UEDVT
• Cancer patients with UEDVT have a higher risk of recurrence, mortality and bleeding
• Due to absence of direct evidence, treatment strategies are extrapolated from lower extremity DVT or PE
• A large prospective registry with long term follow-up of patients with UEDVT is desirable

ABSTRACTBackgroundThe incidence of upper extremity deep vein thrombosis (UEDVT) is increasing. Information on the clinical course of UEDVT is scarce, especially in cancer patients.AimTo summarize the clinical evidence regarding long-term clinical outcomes of UEDVT, in terms of recurrent venous thromboembolism (VTE), mortality, and anticoagulant-related bleeding, in patients with or without concomitant cancer.MethodsA systematic search of the literature was conducted in MEDLINE, EMBASE and BIOSIS Previews. Incidence rates for all outcome variables were calculated.ResultsIn total, 45 studies comprising 4580 patients were included. No randomized controlled trials were identified. In most studies, patients were treated solely with anticoagulants. Among the prospective studies, the incidences of recurrent VTE and bleeding complications averaged 5.1% and 3.1% respectively, during 3 to 59 months of follow-up. In the retrospective studies these figures were 9.8% and 6.7% respectively. Among the prospective studies, the mortality rate was 24% after one year. In the retrospective studies this rate was 35%. Cancer patients were found to have a 2- to 3-fold higher risk of recurrent VTE, an 8-fold increased risk of mortality, and a 4-fold increased risk of bleeding during anticoagulant therapy, compared to non-cancer patients.ConclusionsStudies were very heterogeneous in terms of study design, study populations and treatment approaches. Follow-up durations varied greatly, hampering combined analyses of average incidence rates. There is a need for large prospective studies to provide information on the best management of this disease, especially in high risk groups such as those with cancer.