Lung ultrasound allows the diagnosis of weaning-induced pulmonary oedema.

Abstract

Detecting weaning-induced pulmonary oedema (WIPO) is important because its treatment might prompt extubation. For this purpose, lung ultrasound might be an attractive tool, since it demonstrates pulmonary oedema through the appearance of B-lines. To test the ideal profile (increase in the number of B-lines) for diagnosing WIPO. Before and at the end of 62 spontaneous breathing trials (SBT) performed in 42 patients, we prospectively assessed lung ultrasound on four anterior chest wall points. B-lines were counted before and at the end of SBT. We looked for the threshold of B-line increase (Delta-B-lines) that provided the best diagnostic accuracy, compared to the reference diagnosis of WIPO established by experts blinded to lung ultrasound. SBT failed in 33 cases. WIPO occurred in 17 cases and all failed. The best diagnostic accuracy was reached with a Delta-B-lines ≥6. Among WIPO, the number of B-lines increased by ≥ 6 in 15 cases (including 13 cases with an increase of ≥ 8 B-lines). Among the 16 cases with SBT failure but without WIPO, the Delta-B-lines was ≥ 6 in two cases. Among the 33 cases with SBT failure, this profile diagnosed WIPO with a sensitivity of 88% (64-98) and a specificity of 88% (62-98) [area under the receiver operating characteristic curve 0.91 (0.75-0.98)]. Among the 29 cases with SBT success, a Delta-B-lines ≥ 6 occurred in two cases. This study suggests that a Delta-B-lines ≥ 6 on four anterior points allows the diagnosis of WIPO with the best accuracy. This should be confirmed in larger populations.