Antenatal Fetal Lung Volume for Predicting Neonatal Respiratory Distress Syndrome: A Systematic Review and...

Objective: To determine the diagnostic accuracy of antenatal ultrasonographic fetal lung volume (FLV) for predicting neonatal respiratory distress syndrome (RDS) and to identify clinically relevant sources of heterogeneity. Methods: A systematic review and diagnostic test accuracy meta-analysis was conducted in accordance with PRISMA-DTA and registered in PROSPERO (CRD420251150707). MEDLINE, Scopus, CINAHL Complete, and CENTRAL were searched from inception to 15 November 2025, supplemented by manual searching of reference lists and gray literature. Studies reporting antenatal ultrasonographic FLV with postnatal diagnosis of neonatal RDS were included. Diagnostic accuracy outcomes were synthesized using a bivariate random-effects model to estimate pooled sensitivity, specificity, likelihood ratios, diagnostic odds ratio (DOR), and hierarchical summary receiver operating characteristic (HSROC) curves. For studies reporting FLV as a continuous measure without prespecified cutoffs, mean differences were pooled using random-effects models. Prespecified subgroup analyses evaluated fetal population and scan-to-delivery interval. A sensitivity analysis excluding studies with non-standard RDS definitions was conducted to assess the robustness of pooled estimates. Results: Seven studies (n = 900) met the inclusion criteria. The pooled sensitivity and specificity of FLV for predicting neonatal RDS were 88.00% (95% CI: 78.00 to 94.00) and 82.00% (95% CI: 70.00 to 90.00), respectively, with an HSROC area under the curve of 0.93. Substantial heterogeneity was observed (I² > 75% for both sensitivity and specificity). Subgroup analyses suggested lower heterogeneity in studies including mixed term and preterm fetuses and in studies assessing FLV within 24 h before delivery; however, these findings should be interpreted cautiously because of the small number of studies. In continuous outcome analyses, neonates without RDS had significantly larger FLV than those with RDS (mean difference 7.12 cm³, 95% CI: 5.90 to 8.34). Reporting of measurement reproducibility was limited across studies. Conclusions: Ultrasonographic FLV shows moderate diagnostic performance for predicting neonatal RDS and may provide adjunctive value for perinatal risk stratification, especially when measured close to delivery. However, the current evidence is limited by small study numbers, single-country data, post hoc ROC-derived thresholds, inconsistent RDS definitions, unmeasured confounding, and limited reproducibility reporting. Therefore, FLV should be regarded as an exploratory adjunctive biomarker rather than a validated stand-alone clinical test. Standardized acquisition and segmentation protocols, gestational-age-adjusted or fetal-size-adjusted thresholds, mandatory reliability metrics, and external validation are needed before routine implementation.