Deep venous thrombosis (DVT) is typically caused by one or more of Virchow's triad: stasis, hypercoagulability, and/or endothelial damage. DVT may occur in ambulatory patients presenting to the emergency department (ED) with leg pain and/or swelling, and is also a frequent complication of critical illness due to multiple and often coexisting risk factors, including immobility, surgery, trauma, indwelling devices, malignancy, and inflammatory states. A vexing problem is the unreliability of the symptoms and signs of DVT in the critical care setting, which are often limited by obesity, edema, and surgical dressings. In the intensive care unit (ICU), 10%–100% of DVTs are clinically unsuspected, and pulmonary embolism is the most frequent incidental autopsy finding, directly contributing to death in approximately 5% of all cases.
The true incidence and prevalence of lower extremity DVT (LEDVT) in the critical care setting is unknown, and patients may often be asymptomatic. Based on screening studies using ultrasonography for the diagnosis, incidence rates vary considerably as a result of differences in patient population, adequacy of prophylaxis, sonographic technique, and sonographer skills. Reported incidences range from as low as 8% to as high as 18% for proximal LEDVT, with the majority of cases occurring within the first week of an ICU stay. In the ambulatory and ED setting a DVT is typically symptomatic, and ultrasound is a reliable way to exclude one.
Clinically, LEDVTs are classified according to their embolic risk. Proximal (popliteal and higher veins) DVTs present a significant embolic risk, while isolated calf vein thrombosis is unlikely to embolize. Given that only 20% of calf DVTs will extend proximally, anticoagulation may be held in the ICU setting to avoid unnecessary complications. Therefore, calf veins are not routinely examined by ultrasound in the ICU setting. In the ambulatory and ED setting, if calf veins are not imaged it is recommended that patients return in 5–7 days to be reimaged in case a calf vein thrombus has propagated proximally.
Bedside ultrasonography can play a pivotal role in the diagnostic algorithms of venous thromboembolic disease. In order for the clinician to incorporate ultrasound into the timely diagnosis of DVT, a thorough understanding of the strength and limitations, clinical applications, and technical performance of lower extremity sonography is necessary.
Given the rapid growth in availability of portable ultrasound units in many EDs and ICUs, there are an increasing number of clinicians performing bedside diagnostic venous sonograms. While the examination is the same, whether it is performed in the ED or ICU, the ability of ultrasound to reliably exclude a DVT depends on the prevalence and/or pretest probability of a DVT. Ultrasound tends to perform much better in patients who are symptomatic in either setting.
Accuracy studies of ED and hospitalist physicians with variable amounts of ultrasound training (2–30 hours) reveal sensitivities between 70% and 100%, and specificities between 76% and 100% when compared to the "official" radiology study. Although these results ...