Prediction of postdischarge venous thromboembolism using a risk assessment model

Abstract

The risk of postdischarge venous thromboembolism (VTE) (either deep vein or pulmonary embolism) is increasingly recognized yet the prescription of postdischarge thromboprophylaxis is inconsistent. There is a paucity of information to aid clinicians in identifying surgical patients who are at increased risk for postdischarge VTE. This study aimed to determine the incidence and risk factors associated with symptomatic postdischarge VTE and develop a risk score to identify patients who may benefit from extended duration thromboprophylaxis. This was a retrospective study. All nonorthopedic cases in which the patient was discharged alive without inpatient VTE were selected from the 2005-2009 National Surgical Quality Improvement Program database. A multivariate logistic regression was used to create a risk score for postdischarge VTE prediction. The dataset was split into two-thirds for risk score development and validated in the remaining one-third. The overall incidence of early postdischarge VTE for 2005-2009 National Surgical Quality Improvement Program was 0.3%. The risk score stratified patients into low, moderate, and high risk for postdischarge VTE with the incidence based on the risk score ranging from 0.07% to 2.2%. The risk score had good predictive ability with c-statistic = 0.72 for model development and c-statistic = 0.71 in the validation dataset. Factors associated with postdischarge VTE on multivariate analysis included race, increasing age, steroid use, body mass index ≥30, malignancy, higher American Society of Anesthesiologists class, increasing operative time, length of postsurgical stay, and major postoperative complication. This novel postdischarge VTE prediction score utilizes patient, operative, and early outcome factors to accurately identify patients at increased risk of a postdischarge thromboembolic event. The development of a patient- specific postdischarge VTE risk profile may help address the challenge of determining postdischarge prophylaxis requirements.