ABSTRACT
Based on astute clinical observation alone, a strong link between the presence of cancer and the development of symptomatic venous thromboembolism (VTE) was established nearly a century and a half ago (1). Since that time, numerous autopsy series, epidemiologic studies, and results of clinical trials have provided overwhelming evidence for this association (2-6). Despite this abundant evidence, many gaps remain in our knowledge linking malignancies and VTE. To date, all of the epidemiologic observations have had limitations or biases that have prevented the accurate assessment of the true incidence of symptomatic VTE, such as reporting the incidence of VTE in selected cases that have undergone autopsy (7), incomplete cancer case ascertainment (8,9), low power (10-12), and use of surrogate endpoints such as ultrasound evidence of VTE rather than clinically symptomatic VTE (13). Furthermore, the incidence, time course, and associated risk factors that contribute to VTE in particular tumor types have only recently been studied.
