Use of postoperative venous thromboembolism to predict disease-specific survival in cancer patients.

Abstract

9032 Background: Oncology patients who develop a spontaneous venous thromboembolic event (VTE) have an inferior cancer prognosis. It is not known whether oncology patients who develop a postoperative VTE following complete surgical resection are at the same survival disadvantage as oncology patients who develop a spontaneous VTE. Methods: An institutional database at the Memorial Sloan-Kettering Cancer Center was used to identify all cancer patients who underwent abdominal, pelvic, thoracic or soft tissue procedures between January 1, 2000 and December 31, 2005. Patients who developed a VTE within 30 days of the procedure were identified from a prospective morbidity and mortality database. Stage I-III patients with a VTE following complete resection were then matched 1:10 to controls by age, sex, cancer type, stage, and surgical procedure. Overall (OS) and Disease Specific Survival (DSS) were compared for the entire cohort and for the cohort of matched patients using the Kaplan-Meier method. Results: Of the 23,541 cancer patients who underwent surgery, 474 (2%) had a postoperative VTE. The median follow-up was 24.9 months. In the entire cohort, VTE patients had a significantly worse OS at 5 years compared to controls (49.9% vs 61.1%; p <.0001). The survival difference was seen in stage I-III but not in stage IV patients. The type of VTE did not impact OS and patients with a deep vein thrombosis had a similar survival to patients with a pulmonary embolism. A total of 205 Stage I-III VTE patients who underwent a complete resection were then matched with 2050 controls by age, sex, cancer type, stage, and surgical procedure. In this matched analysis VTE patients continued to demonstrate a significantly worse prognosis with an inferior 5 year OS (54.7% vs. 66.2%; p <.0001) and DSS (67.8% vs. 80.3%; p =.0007) for VTE patients as compared to controls. Conclusions: Postoperative VTE in oncology patients with limited disease and a complete surgical resection is associated with an inferior cancer survival. A postoperative VTE remains a poor prognostic factor, even when controlling for age, stage, cancer type and surgical procedure further supporting an independent link between hypercoagulability and cancer survival. No significant financial relationships to disclose.