Four-extremity venous duplex ultrasound for suspected deep venous thrombosis is an anachronism

Background: Duplex ultrasound is the "gold standard" for diagnosis of acute deep venous thrombosis (DVT) because of its high specificity, sensitivity, safety, and portability. However, unnecessary testing epitomizes inefficient use of scarce health care resources. Here we hypothesize that the majority of simultaneous four-extremity duplex ultrasound (FED) examinations are unnecessary. By analyzing clinical factors of patients with acute DVT found on FED, we aimed to identify a subset of high-risk patients who may have a valid indication for four-extremity testing. Methods: We retrospectively reviewed all venous duplex ultrasound examinations performed in our Intersocietal Accreditation Commission-accredited vascular laboratory from January 1, 2009, to December 31, 2016. Patients with duplex ultrasound scans of all four limbs were included. DVT risk factors and indication for duplex ultrasound examination were recorded. The primary outcome was finding of acute DVT. Results: There were 188 patients who met our search criteria, of whom 31 patients (16.5%) had acute DVT (11 upper extremity, 16 lower extremity, and 4 upper and lower extremity). Fever of unknown origin (FUO) was the main indication for requesting FED (53.7%). Patients who underwent FED for FUO had a significantly lower likelihood of DVT (odds ratio, 0.21; P= .01). DVT was rarely the proximate cause (<1% of all cases) as follow-up culture results and clinical course most often revealed other sources of fever. Only patients with an upper extremity central venous catheter (CVC; n =103) with at least two associated risk factors had an upper extremity DVT, which was usually line associated (93%). Only patients with at least two associated risk factors had a lower extremity DVT. Conclusions: FED for FUO is inefficient, given that DVT was rarely the proximate cause of fever. Acute upper extremity DVT was found only in patients with an upper extremity CVC, demonstrating that patients without upper extremity CVC do not benefit from upper extremity duplex ultrasound examination. Upper extremity DVT is usually line associated and dependent on the number of cumulative risk factors present, suggesting that only the extremity associated with the CVC in the right clinical context should be imaged. Lower extremity DVT is also dependent on the number of cumulative risk factors present, and testing should be reserved for patients according to the clinical context. Our results indicate that a restrictive strategy can reduce testing inefficiency and health care cost without compromising patients' safety.