Determining incidence and predictors of deep vein thrombosis in patients with non-small cell lung cancer

Abstract

7159 Background: The risk of deep vein thrombosis (DVT) among patients with non-small cell lung cancer (NSCLC) has not been well studied. We conducted a retrospective cohort study of patients with NSCLC to determine the incidence of DVT and characterize predictors of DVT in NSCLC patients. Methods: The pulmonary oncology database of the SMBD-Jewish General Hospital contains prospectively collected clinical data on all lung cancer patients followed in the pulmonary oncology clinic since January 1, 1997. We identified all consecutive patients with a histologically confirmed new diagnosis of NSCLC between January 1, 1997 and December 31, 2004 and determined the occurrence of an objectively defined DVT. Data on age, gender, NSCLC type and stage, Eastern Cooperative Oncology Group (ECOG) performance status, exposure to surgery and chemotherapy, and death was collected and compared among patients with DVT and patients without DVT. Results: Of the 493 NSCLC patients included in the cohort for a total of 634 person-years, 67 (13.6%) patients developed an objectively confirmed DVT. We calculated an incidence rate of 110 cases (95% confidence interval (CI) 80, 130) per 1000 person-years. Risk factors associated with occurrence of DVT were advanced stage (p = 0.0006) and male sex (p = 0.04). A multivariable regression analysis adjusted for recent surgery and performance status showed that advanced stage (Rate ratio 2.55, 95% CI 1.33–4.89) and male sex (Rate ratio 1.75, 95% CI 1.03–2.94) were independent predictors of DVT. Age, type of NSCLC, and chemotherapy did not predict DVT. The risk of dying was 1.7-fold increased (Hazard Ratio 1.73, 95% CI 1.29–2.32, adjusted for age, sex, stage, surgery, performance status, and date of lung cancer diagnosis) among patients with DVT compared to patients without DVT. Conclusions: Our results show a high incidence of DVT in NSCLC patients and that advanced stage and male sex are important predictors of DVT. Moreover, NSCLC patients with DVT have a 1.7 fold increased risk of dying than patients without DVT. Confirmation of our results by prospective studies may provide the necessary evidence for targeted use of prophylactic anticoagulants in NSCLC patients to prevent development of DVT and improve related survival. No significant financial relationships to disclose.