Diagnostic Performance of Wells Score Combined With Point‐of‐care Lung and Venous Ultrasound in Suspected Pulmonary Embolism

• Published in: Academic emergency medicine 2017-03, Vol.24 (3), p.270-280
• Main Authors: Nazerian, Peiman, Volpicelli, Giovanni, Gigli, Chiara, Becattini, Cecilia, Sferrazza Papa, Giuseppe Francesco, Grifoni, Stefano, Vanni, Simone, Mark Courtney, D.
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Cohort Studies; Embolisms; Emergency medical care; Emergency Service, Hospital; Female; Fibrin Fibrinogen Degradation Products; Humans; Lungs; Male; Medical diagnosis; Middle Aged; Point-of-Care Systems; Pulmonary arteries; Pulmonary Embolism - diagnosis; Pulmonary Embolism - diagnostic imaging; Pulmonary embolisms; Risk; Sensitivity and Specificity; Ultrasonic imaging; Ultrasonography - methods; Venous Thrombosis - diagnostic imaging
• ISSN: 1069-6563; 1553-2712
• DOI: 10.1111/acem.13130

Objective Lung and venous ultrasound are bedside diagnostic tools increasingly used in the early diagnostic approach of suspected pulmonary embolism (PE). However, the possibility of improving the conventional prediction rule for PE by integrating ultrasound has never been investigated. Methods We performed lung and venous ultrasound in consecutive patients suspected of PE in four emergency departments. Conventional Wells score (Ws) was adjudicated by the attending physician, and ultrasound was performed by one of 20 investigators. Signs of deep venous thrombosis (DVT) at venous ultrasound and signs of pulmonary infarcts or alternative diagnoses at lung ultrasound were considered to recalculate two items of the Ws: signs and symptoms of DVT and alternative diagnosis less likely than PE. The diagnostic performances of the ultrasound‐enhanced Ws (USWs) and Ws were then compared after confirmation of the final diagnosis. Results A total of 446 patients were studied. PE was confirmed in 125 patients (28%). USWs performed significantly better than Ws, with a sensitivity of 69.6% versus 57.6% and a specificity of 88.2% versus 68.2%. In combination with D‐dimer, USWs showed an optimal failure rate (0.8%) and a significantly superior efficiency than Ws (32.3% vs. 27.2%). A strategy based on lung and venous ultrasound combined with D‐dimer would allow to avoid CT pulmonary angiography in 50.5% of patients with suspected PE, compared to 27.2% when the rule without ultrasound is applied. Conclusions A pretest risk stratification enhanced by ultrasound of lung and venous performs better than Ws in the early diagnostic process of PE.