The predictive value of newborn and infant lung ultrasound score for mechanical ventilation needs: a systematic review and meta-analysis

Objective To investigate the predictive value of neonatal and infant lung ultrasound scores for the need for mechanical ventilation through meta-analysis. Methods Literature up to October 1, 2024, on neonatal and infant lung ultrasound scores and mechanical ventilation was searched in PubMed, Web of Science, Embase, and The Cochrane Library databases. The diagnostic accuracy of the included studies was evaluated using the Quality Assessment tool for Diagnostic Accuracy Studies. Revman5.4 and StataSE-64 software were employed to calculate the pooled sensitivity, specificity and AUC value of neonatal and infant lung ultrasound scores for predicting the need of mechanical ventilation. Results The meta-analysis comprised 9 studies (7 prospective studies and 2 retrospective), including a total of 1,746 patients. The LUS score predicted the need for mechanical ventilation. Overall sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio were 74% (95% CI: 66%–81%), 81% (95% CI: 71%–87%), 3.8 (2.4–6.1), 0.32 (0.23–0.46) and 12 (5–6), respectively. The forest plots indicated significant heterogeneity for sensitivity ( p = 0.81, I 2 = 82.36%, 95%CI: 72.82%–91.90%) and specificity ( p = 0.74, I 2 = 51.27%, 95%CI: 17.74%–84.80%). Conclusion Meta-analysis of multivariate categorical variables indicated that the higher the LUS scores, the greater the risk of mechanical ventilation. The combined results of meta-analysis of diagnostic data suggest that LUS score has high accuracy in predicting the need for mechanical ventilation. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/ , PROSPERO CRD420251029542.