Khorana score as a predictive tool of venous thromboembolism in cancer patients in rural practice.

Abstract

e24049 Background: The Centers for Disease Control and Prevention (CDC) estimates that 60,000 to 100,000 Americans die from venous thromboembolism (VTE) annually, costing the U.S. healthcare system $7–10 billion each year. VTE is a leading cause of mortality in cancer patients, and risk is particularly elevated in rural areas, where cancer death rates are higher than in metropolitan areas. The Khorana score has been shown to predict VTE incidence before starting treatment and as such identify those patients that would benefit most from VTE prophylaxis. We aimed to replicate the utility of the Khorana score on a rural population of cancer patients to help address VTE in advance. Methods: From our electronic health records system, we extracted data from cancer patients (age ≥ 18 years) who had begun chemotherapy from 2017-January through 2019-August. Patients who were pregnant, had undergone hormonal therapy or major surgeries within three months prior to VTE diagnosis, or had a history of coagulation disorders or VTE prior to starting chemotherapy were excluded. A Khorana score (0—6, interval = 1) was calculated for each prior to their starting chemotherapy and was used for VTE risk stratification, 0 – low; 1,2 – intermediate; ≥3 – high. We hypothesized that VTE occurrence during the first 6 months following the start of treatment would be positively associated with risk category. Results: The 144, mostly female (64%), cancer patients meeting inclusion had a mean age of 75 years (range 48—95) and VTE risk stratification: 10% (15) high, 67% (96) intermediate, and 23% (33) low. In the first 6 months of starting treatment, 6% (9/144) developed VTE, with a prevalence of 20% (3/15) amongst patients stratified as high risk, 5% (5/96) of intermediate, and 3% (1/33) of the low risk group. A linear-by-linear association of the observed Khorana scores with VTE occurrence implies a meaningful positive association between VTE risk stratification by Khorana score and VTE, chi-square (1,144) = 4.018, p = .045. Conclusions: The trend of higher VTE prevalence in intermediate and high risk patients supports the observed positive association between Khorana score VTE risk category and a greater percentage of patients experiencing VTEs with increasing risk. This trend may likely be amplified with application to a larger number of patients. The study limitations include the lack of diversity of organ affection and sample size, specifically upon risk stratification. However, initiating a discussion about the risks and benefits of VTE prophylaxis prior to starting treatment is of outmost importance.