Deep vein thrombosis (DVT) is a common condition with a high risk of post-thrombotic morbidity, especially in patients with a proximal thrombus. Successful iliofemoral (IF) clot removal has been shown to reduce the severity of post thrombotic syndrome (PTS). It is assumed that earlier thrombus lysis is associated with better outcome. Generally, the earlier IFDVT is confirmed, the earlier thrombus lysis could be performed. D-dimer levels and Wells score are currently used to assess the pre-duplex probability for DVT, however, some studies indicate that D-dimer value varies depending on the thrombus extent and localization. Using D-dimer and other risk factors might facilitate development of a model selecting those with an increased risk of IFDVT that might benefit from early referral for additional analysis and adjunctive IF thrombectomy.
All consecutive adult patients from a retrospective cohort of STAR diagnostic center (primary care) in Rotterdam suspected of DVT between September 2004 and August 2016 were assessed for this retrospective study. Diagnostic workup for DVT including Wells score and D-dimer were performed as well as complete duplex ultrasonography (CDUS). Patients with an objective evidence of DVT were categorized according to thrombus localization using the Lower Extremity Thrombolysis (LET)-classification. Logistic regression analysis was done for a model predicting IFDVT. Cut-off value of the model was determined using an ROC-curve.
A total of 3381 patients were eligible for study recruitment, of whom 489 (14.5%) had confirmed DVT. We developed a multivariate model (sensitivity of 77% and specificity of 82%, area under the curve (AUC)=0.90, 0.86-0.93) based on D-dimer, Wells score, age, and anticoagulation use which is able to distinguish IFDVT patients from all patients suspected of DVT.
This multivariate model will adequately distinguish IFDVT among all suspected DVT patients. Practically, this model could give each patient a pre-duplex risk score which could be used to prioritize suspected IFDVT patients for an immediate imaging test to confirm or exclude IFDVT. Further validation studies are needed to confirm potential of this prediction model for IFDVT.

Copyright © 2021. Published by Elsevier Inc.

Author