Venous Thromboembolism (VTE) and Survival in a Cancer Chemotherapy Outpatient Clinic: A Retrospective Chart Review Validation of a VTE Predictive Model.

Because venous thromboembolism (VTE) often occurs as a complication of cancer and chemotherapy, cancer patients have a 47-fold elevation of risk of VTE compared to the general population (Khorana et al., Journal of Thrombosis and Haemostasis 2007). However, until recently there was not a model to accurately predict which cancer patients would experience a VTE during their treatment. Khorana et al. have developed such a model (Khorana et al., Blood 2008). We conducted a retrospective study to test the effectiveness of Khorana and colleagues' clinical model at predicting risk of VTE in patients with metastatic cancer. We collected data from an outpatient oncology clinic for 112 patients with solid tumors or malignant lymphoma who had undergone active chemotherapy in the last two years. The data included the five predictive variables outlined in Khorana et al.'s model: site of cancer (2 points for a very high-risk site, 1 point for a high-risk site); a platelet count of 350 × 10⁹/L or more; hemoglobin less than 100g/L and or/use of erythropoiesis-stimulating agents; leukocyte count more than 11 × 10⁹/L; and body mass index of 35kg/m² or more (1 point each) (Khorana et al., Blood 2008). All data was collected on the first day of chemotherapy or the most recent labs before the start of chemotherapy. A score of 0 points represented a low risk of VTE, 1-2 points represented an intermediate risk and 3-4 points represented a high risk. There were 20 low risk patients, 63 intermediate risk and 29 high risk. We found that the clinical model accurately predicted which patients would experience a VTE (Table 1). Of the total 112 patients, 23 experienced a VTE. We determined that patients considered high risk were 8-fold more likely to experience a VTE as compared to low risk patients (see Table 1). Kuderer et al. demonstrated that this risk model also predicts overall mortality for patients receiving chemotherapy (Kuderer et al., ASH 2009). The results of our study also showed a strong correlation between overall mortality and a higher risk score. The mortality rate of high risk patients during the trial period was 69.0% compared to 30.0% of low risk patients and 42.9% of intermediate risk patients. Our retrospective review validates Khorana and colleagues' clinical model to predict VTE and survival rates in cancer patients. Our results, combined with data from previous studies, indicate that patients who are at risk for VTE also have a higher mortality rate. With an accurate predictive model, primary prophylaxis treatment for VTE could be considered for high risk patients.

No relevant conflicts of interest to declare.