IntroductionTwo-point compression ultrasound is purportedly a simple and accurate means to diagnose proximal lower extremity deep vein thrombosis (DVT), but the pitfalls of this technique have not been fully elucidated. The objective of this study is to determine the accuracy of emergency medicine resident-performed two-point compression ultrasound, and to determine what technical errors are commonly made by novice ultrasonographers using this technique.MethodsThis was a prospective diagnostic test assessment of a convenience sample of adult emergency department (ED) patients suspected of having a lower extremity DVT. After brief training on the technique, residents performed two-point compression ultrasounds on enrolled patients. Subsequently a radiology department ultrasound was performed and used as the gold standard. Residents were instructed to save videos of their ultrasounds for technical analysis.ResultsOverall, 288 two-point compression ultrasound studies were performed. There were 28 cases that were deemed to be positive for DVT by radiology ultrasound. Among these 28, 16 were identified by the residents with two-point compression. Among the 260 cases deemed to be negative for DVT by radiology ultrasound, 10 were thought to be positive by the residents using two-point compression. This led to a sensitivity of 57.1% (95% CI [38.8–75.5]) and a specificity of 96.1% (95% CI [93.8–98.5]) for resident-performed two-point compression ultrasound. This corresponds to a positive predictive value of 61.5% (95% CI [42.8–80.2]) and a negative predictive value of 95.4% (95% CI [92.9–98.0]). The positive likelihood ratio is 14.9 (95% CI [7.5–29.5]) and the negative likelihood ratio is 0.45 (95% CI [0.29–0.68]). Video analysis revealed that in four cases the resident did not identify a DVT because the thrombus was isolated to the superior femoral vein (SFV), which is not evaluated by two-point compression. Moreover, the video analysis revealed that the most common mistake made by the residents was inadequate visualization of the popliteal vein.ConclusionTwo-point compression ultrasound does not identify isolated SFV thrombi, which reduces its sensitivity. Moreover, this technique may be more difficult than previously reported, in part because novice ultrasonographers have difficulty properly assessing the popliteal vein.