According to Fedor Lurie (Jobst Vascular Institute, Toledo, USA), who presented at the European Venous Forum Annual Meeting (EVF; 28–30 June, Athens, Greece), patients with a high probability of iliofemoral deep vein thrombosis (DVT) are not diagnosed early enough in the USA, which can become problematic. The research presented by Lurie investigated a means of identifying the anatomical location of DVT using existing risk stratifications, with a view to expedite future diagnoses and treatment of iliofemoral DVT.
In the USA, current guidelines recommend risk stratification of patients with suspected DVT, using a clinical decision rule (such as Well’s score) and D-dimer level. However, Lurie pointed out that this strategy results in early anticoagulation and a delay in imaging tests in high risk patients.
Additionally, specific management of patients with confirmed DVT depends on the anatomical location of the thrombus (iliofemoral, femoropopliteal, or calf), which, when known, can accelerate appropriate investigation and treatment.
Given that larger thrombus volume and acuteness of thrombosis in iliofemoral DVT should result in higher concentrations of circulating products of thrombus degradation, as well how acute iliofemoral DVT is usually more symptomatic compared to femoropopliteal or calf DVT, which should increase the value of the Well’s score, Lurie hypothesised that acute iliofemoral DVT is associated with higher levels of D-dimer and a higher Well’s score compared to femoropopliteal and calf DVT.
A nested case–control study was performed, identifying a prospective cohort of 221 patients who had a positive duplex ultrasound performed within 48 hours from onset of symptoms, who were admitted to emergency departments of seven hospitals in Northwest Ohio and Southeast Michigan from 2014 to 2016. Each patient was matched by age and gender with a control who presented negative for DVT in a 1:5 ratio, meaning the total number of controls was 1103.
The relationship between the anatomical location of DVT, D-dimer levels, and Well’s score were examined using parametric and non-parametric (for Well’s score) tests. Receiver operating characteristic (ROC) curves were analysed to identify the best cut-off values for both Well’s score and D-dimer.
Findings of the study show that among patients with positive DVT ultrasound scans, iliofemoral DVT was found in 75, femoropopliteal in 92, and calf DVT in 54 patients. Both D-dimer and Well’s score showed high diagnostic value for identification of iliofemoral DVT with areas under the ROC curves 0.9 and 0.8, respectively (p<0.0001).
Regarding the risk stratification instruments, only D-dimer was sufficiently sensitive to identify femoropopliteal DVT (ROC area 0.8, p<0.0001), whilst neither D-dimer nor Well’s score had sufficient sensitivity to identify calf DVT (ROC area 0.6, p=0.1). Patients who had D-dimer of 700ng/ml or more were 22.5 times more likely to have iliofemoral DVT (OR=22.5, 95%, CI 11.2–45.1) and those who had D-dimer level between 500 and 700 were 9.5 times more likely to have femoropopliteal DVT (OR=9.5, 95%, CI 5.7–16.0).
Lurie concluded that the D-dimer is more sensitive to DVT level compared to Well’s score. Importantly, he reiterated that it is possible to identify patients at high risk for iliofemoral DVT based on D-dimer level, as the findings suggested that the D-dimer level of 700ng/ml or higher is more likely to be associated with iliofemoral DVT. Lurie pointed to the practical implications of these findings, in terms of the possibility of immediate intervention, given that the cut-off point for D-dimer can be used to identify high risk patients for iliofemoral DVT.