Abstract
BACKGROUND: This study aimed to determine the diagnostic accuracy of the Diaphragmatic Thickening Fraction (DTF) and Diaphragmatic Excursion (DE) in predicting extubation success. METHODS: This study was a systematic review with meta-analysis of observational studies. We searched the MEDLINE/PubMed, Embase, LILACS, CINAHL, Cochrane Central, PEDro, Web of Science, and SCOPUS databases, with no restrictions on period or language. The risk of bias and quality of the studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2), Newcastle Ottawa Score, and GRADE tools. RevaMan version 5.4 was used for the meta-analysis. RESULTS: A total of 14 studies were included in the meta-analysis, which included 657 patients. DTF showed low overall accuracy AUC 0.63 (0.57-0.69), but performed better in infants AUC 0.82 (0.74-0.89) with a mean difference of 11.92 (7.73-16.11) mm between success and failure. DE showed greater diagnostic accuracy, with AUC 0.72 (0.58-0.85) and a mean difference of 2.21 (1.44-2.98) mm in infants. Assessment by the left hemithorax is still limited. The extubation failure rate was 26.9%, and the success group had shorter mechanical ventilation (-4.5 days) and hospitalisation (-12.2 days) times. CONCLUSIONS: Diaphragm ultrasound shows promise in predicting extubation success in children, especially in the assessment of diaphragmatic excursion. The thickening fraction showed better accuracy in infants, and the right hemithorax was the most evaluated. However, the heterogeneity of the studies limits the interpretation of the findings.