Morbidity and mortality associated with brachial vein thrombosis.

We have noted a significant incidence of pulmonary embolism (PE) and mortality associated with upper extremity deep venous thrombosis (UEDVT). Since there is an association between site of lower extremity DVT (LEDVT) and PE, we hypothesized that there might also be a correlation between site of UEDVT and PE with associated mortality. To further elucidate this hypotheses, we analyzed the mortality and incidence of PE diagnosed with subclavian/axillary/internal jugular vein thrombosis during an 11-year period at our institution and compared the data to those of patients diagnosed with brachial DVT. We studied 598 patients diagnosed with acute internal jugular, subclavian, axillary, or brachial DVT by duplex scanning. The patients were divided into three groups based on the most proximal location of the thrombus: group I, UEDVT involving the subclavian or axillary veins (n = 467); group II, isolated internal jugular DVT (n = 80); group III, brachial DVT alone (n = 52). Mortality rates at 2 months were 29%, 25%, and 21% for each group, respectively. The number of patients diagnosed with PE by ventilation/perfusion scans in groups I, II, and III, respectively, were 5%, 6.25% and 11.5% (p = 0.13). Furthermore, stratification by risk factors failed to demonstrate factors associated with increased 2-month mortality. Contrary to the initial hypothesis of a relationship between the site of thrombosis and the incidence of PE and mortality, these data demonstrated no statistical differences in mortality or incidence of PE among the groups studied. Additionally, these data suggest that brachial vein thrombosis is a disease process related to comparable associated mortality and morbidity similar to other forms of UEDVT. Based on these data, we suggest that UEDVT may be thought of as a marker for the severity of systemic illness of the patient rather than just as a cause of venous thromboembolism.