Predictors of active cancer thromboembolic outcomes: mortality associated with calf deep vein thrombosis.

BACKGROUND: Venous thromboembolism (VTE) is a leading cause of mortality in patients with cancer. Outcomes, including mortality data, in the cancer population are limited in those with calf deep vein thrombosis (CDVT) compared to individuals with proximal deep vein thrombosis. The aim of this study was to assess the prognosis among patients with active cancer and CDVT.

METHODS: Single institution inception cohort of cancer-associated CDVT patients who presented with thrombosis distal to popliteal level confirmed objectively by ultrasound. We defined active cancer as metastatic disease or use of chemotherapy at diagnosis. Clinical and laboratory data were extracted from the electronic health records. The Khorana Risk Score (KRS) was extracted based on data at entry. Institutional review board approval was obtained prior to the analysis. Categorical variables are expressed as percentages and continuous variables as median (interquartile range). Kaplan-Meier method, Pearson's χ2, Mann-Whitney U and Cox proportional hazard were applied. SPSS software version 22 was used for all statistics.

RESULTS: One hundred nine patients (men=44 [40%], Age>65=89 [82%], BMI>30=25 [23%], Smoker=59 [54%]) were included. The majority had a low (30%) or intermediate KRS (64%) at diagnosis. Forty-seven percent died during a median follow-up time of 2.5 years (0.5-3.1). After multivariate analysis, the predictors of mortality were: smoking (hazard ratio [HR] 2.3; 95% CI: 1.2-4.7), metastasis (HR=5.8; 95% CI: 2.9-11.7), gastrointestinal cancer (HR=3.9; 95% CI: 1.8-8.5), and lung cancer (HR=4.1 95% CI: 1.7-10.3). VTE specific variables not associated with mortality included: bilateral CDVT, concomitant pulmonary embolism, multiple vein involvement, filter placement, or a surgery-associated event. The KRS was not predictive of death.

CONCLUSIONS: Cancer-specific variables and smoking predicted mortality among CDVT patients in this cohort, rather than VTE characteristics at the time of CDVT diagnosis.