Upper extremity deep venous thrombosis and its impact on morbidity and mortality rates in a hospital-based population

Although much attention has been focused on lower extremity deep venous thrombosis (LEDVT), there is a relative paucity of data regarding the impact of upper extremity deep venous thrombosis (UEDVT) on morbidity and mortality rates. To increase our knowledge with the latter disease, we have reviewed our experience at our institution with 170 patients who had brachial, axillary, and subclavian vein thromboses.

Over the past 5 years, UEDVT was diagnosed in 170 patients by duplex scanning. The indications for duplex examination were either upper extremity swelling (95%) or as part of the workup for pulmonary embolism (5%). There were 103 women (61%) and 67 men (39%), with ages ranging from 9 to 101 years (mean, 68 +/- 17 years). The diagnosis was made in 152 patients (89%) while they were admitted to the hospital and in 18 patients (11%) in the outpatient clinic. Risk factors included presence of a central venous catheter or pacemaker in 110 patients (65%), malignancy in 63 patients (37%), concomitant LEDVT in 19 patients (11%), and history of LEDVT in 18 patients (11%). Fifty-six patients (33%) had multiple risk factors, whereas 36 patients (21%) had no obvious risk factor.

The 1-month and 3-month mortality rates for the entire study group were 16% and 34%, respectively. Patients who had concomitant LEDVT, were 75 years of age or older, and were not treated with anticoagulation medication had a significantly higher 1-month mortality rate. Patients whose diagnoses were made in the outpatient setting were statistically younger and had a lower 3-month mortality rate when compared with the patients whose diagnoses were made as inpatients. Pulmonary embolism was documented by ventilation/perfusion scan in 12 patients (7%). Although no patient in the group in which UEDVT was diagnosed on an outpatient basis was documented to have a pulmonary embolism and 12 patients (8%) in the inpatient group had pulmonary emboli, this difference was not statistically significant. Anticoagulation medication did not totally prevent pulmonary embolism in this review. All patients were followed-up for between 0 to 49 months (mean, 13 +/- 1 months). No swelling of the affected arm was observed in 145 patients (94%); four patients complained of mild intermittent swelling (2%), and seven patients reported significant swelling (4%).

Contrary to previous reports, these data suggest that UEDVT is associated with a low incidence of postthrombotic upper extremity swelling, but a significant incidence of pulmonary embolism and rate of mortality. This review suggests that UEDVT is at least as serious a disease entity as LEDVT and should be managed as aggressively as LEDVT.