Abstract
INTRODUCTION
There are few prediction tools for estimating the risk of thrombosis but they are based on studies performed on hospitalized medical patients without cancer or on hospitalized neutropenic cancer patients without special consideration to lymphoma patients.
AIM
Aim of our study was to determine incidence of thromboembolic (TE) events in patients with non Hodgkin lymphoma (NHL), Hodgkin lymphoma (HL) and chronic lymphocytic leukemia (CLL)/ small lymphocytic lymphoma (SLL) who were hospitalized to the lymphoma department in the Clinic of hematology, Clinical Center Serbia, Belgrade and Clinic of hematology, Clinical Center Kragujevac. Also, we assessed 2 predictive models (Padua and Khorana score) and create new model for the identification of lymphoma patients at risk for thromboembolism.
MATERIALS AND METHODS
We reviewed all medical records of patients with with NHL, HL and CLL/SLL diagnosed and treated at two previously mentioned institution between January 2006 and December 2014.
RESULTS
The study population included 1820 eligible lymphoma patients. Of all the patients included in the study, 99 (5.4%) developed at least one TE during a follow-up period of 3 months from the end of therapy. In the final multivariate analysis, the following variables were independently associated with risk of TE: previous VTE and/or arterial events, reduced mobility (ECOG 2-4), obesity (BMI >30 kg/m(2)), extranodal localization, mediastinum involvement, development of neutropenia during therapy and hemoglobin level less than 100g/L. Subsequently, we assigned points for the risk model based on the regression coefficients obtained from the final model and developed Thrombosis Lymphoma (ThroLy) score consisting of all significant variables from the multivariate analysis. The Throly score was arrived at by assigning 2 points for all parameters with an OR >5 in multivariate regression analyses (e.g., previous VTE and arterial events, mediastinum involvement, and BMI) and 1 point for rest all other significant variables. Finally, population were divided into 3 risk categories for TE based on the score from the risk model: low (score 0-1), intermediate (score 2-3) and high (score >3). High risk score had a positive predictive value (probability of TE in those designated high risk) of 65.2%.
CONCLUSIONS
Significance of our investigation is development of score that help phisicians to recruit lymphoma patients at risk for development of thromboembolic complications. Also, we can say that our score is dynamic allowing us to change approach during different phase of therapy and is not limited to outpatient settings or with some complicated laboratory analysis.
Collapse