PURPOSE: Deep vein thrombosis (DVT) in the critically ill is usually diagnosed by Doppler ultrasonography. Delays in performance are common and may affect patient outcomes. Two-point compression ultrasonography (CUS) of the common femoral and popliteal veins has been shown to accurately diagnose DVT when performed by critical care physicians. We sought to determine the utility of resident performed two-point CUS in critically ill patients.
METHODS: We performed a prospective study of two-point CUS by medical residents in consecutive, critically ill patients at a tertiary teaching hospital. A formal reading of the Doppler ultrasound by an attending radiologist was used as the reference standard. All residents received a 2-hour standardized training course in focused vascular ultrasonography. Resident studies were performed within 5 hours of the formal study. A linear 7.5 MHZ vascular probe was used. Results were categorized as compressible or noncompressible veins. The superficial femoral vein was not evaluated.
RESULTS: Seven residents performed seventy seven studies. Statistical analysis showed specificity of 96%, sensitivity of 75%, negative predictive value of 99%, positive predictive value of 50% and kappa of 0.62, considered good agreement. Three out of four DVT in the study were correctly identified. One isolated popliteal DVT was missed. The average time from the order of the examination until radiologist reading was 16.8 hours. Two out of three false positives occurred in patients with BMI greater than 30 and one lymph node was misidentified as a DVT.
CONCLUSIONS: Two-point CUS performed by medical residents has excellent negative predictive value and specificity for diagnosis of common femoral and popliteal DVT. Residents showed good interobserver agreement with radiologists. There was a tendency towards over diagnosing DVT. Assessment of venous compressibility may be more difficult in patients with BMI above 30.
CLINICAL IMPLICATIONS: Medical residents can perform and interpret CUS to exclude DVT in the critically ill in a timely fashion. Limitations include false positives in patients with BMI over 30 and difficulty in identifying isolated popliteal DVT.
DISCLOSURE: The following authors have nothing to disclose: Jonathan Caronia, Roman Reznik, Raghukumar Thirumala, Adrian Sarzynski, Bushra Mina
No Product/Research Disclosure Information