Lung ultrasound aeration score as a predictor of extubation failure in neonates: an updated systematic review with meta-analysis and meta-regression.

This study aimed to evaluate the performance of LUS aeration score in predicting extubation failure among mechanically ventilated neonates. This review was conducted in accordance with PRISMA-DTA 2018. Electronic searches were performed in PubMed, ScienceDirect, Wiley, Nature, and Springer. Methodological quality was assessed using QUADAS-3. Diagnostic meta-analysis was performed in Stata 17.0 using MIDAS, while univariable bivariate meta-regression was conducted in R using the mada package. Pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), log diagnostic odds ratio (logDOR), and area under the curve (AUC) were calculated. Clinical applicability was assessed using Fagan nomograms, while sensitivity analysis, subgroup analysis, and Deeks' funnel plot asymmetry test were conducted to evaluate robustness, heterogeneity, and publication bias. Ten studies involving 859 neonates were included. LUS aeration score showed high diagnostic accuracy, with a pooled sensitivity of 0.87, specificity of 0.86, PLR of 6.13, NLR of 0.15, logDOR of 3.71, and AUC of 0.93. Heterogeneity was moderate for sensitivity and substantial for specificity. Anatomical scanning coverage was the only significant moderator in meta-regression (p = 0.049), with posterior-inclusive protocols associated with a lower false-positive rate. Deeks' test showed no significant publication bias. LUS aeration score demonstrates good diagnostic performance for predicting neonatal extubation failure and may support clinical decision-making as an adjunct to conventional extubation readiness assessment. Standardized, validated scoring protocols and age-specific external validation are required before a universal cutoff can be recommended. What is Known: • Extubation failure remains a clinically important complication in mechanically ventilated neonates and is associated with increased morbidity. • Lung ultrasound (LUS) is a bedside, radiation-free imaging method, whereas the LUS aeration score is a quantitative measure of regional aeration loss; reported accuracy for predicting extubation failure varies across studies. What is New: • This updated diagnostic meta-analysis of 10 studies involving 859 neonates found strong pooled diagnostic performance of the LUS aeration score for predicting extubation failure, with a sensitivity of 0.87, specificity of 0.86, and AUC of 0.93. • Exploratory univariable bivariate meta-regression identified anatomical scanning coverage as the only significant overall moderator, with posterior-inclusive protocols associated with a lower false-positive rate.