Assessing Extravascular Lung Water in Critically Ill Patients Using Lung Ultrasound: A Systematic Review on Methodological Aspects in Diagnostic Accuracy Studies.
Neuteboom OB., Heldeweg ML., Pisani L., Smit MR., Lagrand WK., Cherpanath TG., Dondorp AM., Schultz MJ., Tuinman PR.
Lung ultrasound (LUS) is a non-invasive bedside method used to quantify extravascular lung water (EVLW). To evaluate the methodology and diagnostic accuracy of LUS in studies assessing EVLW in intensive care unit patients, PubMed and Embase were searched for studies comparing LUS with imaging modalities. In 14 relevant studies a wide variety of equipment used and training of examiners were noted. Four scoring systems were reported: (i) a binary score (the presence of three or more B-lines); (ii) a categorical score; (iii) a numerical score; (iv) a quantitative LUS score using software. The diagnostic accuracy of LUS varied: sensitivity ranged from 50%-98%, specificity from 76%-100% and r² from 0.20-0.91. Methodology and diagnostic accuracy varies substantially in published reports. Further research is needed to correlate methodological factors with diagnostic accuracy. Hospitals should standardize LUS methodology. Consensus is needed to harmonize LUS methodology for lung water assessment.