Bedside lung ultrasound, mobile radiography and physical examination: a comparative analysis of diagnostic tools in the critically ill

• Published in: Critical care and resuscitation Vol. 18; no. 2; p. 124
• Main Authors: Inglis, Andrew J, Nalos, Marek, Sue, Kwan-Hing, Hruby, Jan, Campbell, Daniel M, Braham, Rachel M, Orde, Sam R
• Format: Journal Article
• Language: English
• Published: Australia 01.06.2016
• Subjects: Australia; Critical Care - methods; Critical Care - statistics & numerical data; Critical Illness; Female; Humans; Lung - diagnostic imaging; Lung Diseases - diagnosis; Male; Middle Aged; Physical Examination - methods; Physical Examination - statistics & numerical data; Point-of-Care Systems - statistics & numerical data; Prospective Studies; Reproducibility of Results; Tomography, X-Ray Computed - statistics & numerical data; Ultrasonography
• ISSN: 1441-2772

To compare lung ultrasonography (LUS), chest xray (CXR) and physical examination (Ex) for the detection of pathological abnormalities in the lungs of critically ill patients. A prospective cohort study of 145 patients in the intensive care unit of a tertiary teaching hospital who were undergoing echocardiography for a clinical indication. Each patient was independently assessed by Ex, CXR and LUS on the same day. Examiners were asked to comment on the presence or absence and severity of pleural effusion, lung consolidation and alveolar interstitial syndrome (AIS). Independent expert examiners performed the LUS and an independent radiologist reported on the CXR. Ex, CXR and LUS were in fair agreement with each other in detecting a pulmonary abnormality (CXR v LUS, κ = 0.31; CXR v Ex, κ = 0.29; LUS v Ex, κ = 0.22). LUS detected more abnormalities than did CXR (16.2%; χ(2) = 64.1; P < 0.001) or Ex (23.5%; χ(2) = 121.9; P < 0.001). CXR detected more pleural effusions than LUS (9.3%; χ(2) = 7.6; κ = 0.39), but LUS detected more pleural effusions than Ex (22.8%; χ(2) = 36.4; κ = 0.18). There was no significant difference in the performance of LUS and CXR in quantifying the size of a pleural effusion (Z = -1.2; P = 0.23). Ex underestimated size compared with CXR or LUS. LUS detected more consolidation than CXR (17%; χ(2) = 115.9; P < 0.001) and Ex (16.2%; χ(2) = 90.3; P < 0.001). We saw no difference in performance between CXR and Ex in detecting lung consolidation (0.9%; χ(2) = 0.51; P < 0.48). LUS detected more cases of AIS than CXR (5.5%; χ(2) = 7.9; P = 0.005) and Ex (13%; χ(2) = 25.8; P < 0.001). There was only fair-to-moderate agreement between LUS, CXR and Ex in detecting pulmonary abnormalities, including pleural effusion, lung consolidation and AIS. The higher rate of detection from LUS, combined with its ease of use and increasing accessibility, makes for a powerful diagnostic tool in the ICU.