Kanagaraj Uthaya Kumaran, Castaldo Michael, Braschel Melissa, Harris Anna, Mitra Souvik
Division of Neonatology, Department of Pediatrics, University of British Columbia, Vancouver, BC, Canada.
Clinical Research Support Unit, British Columbia Children's Hospital Research Institute, Vancouver, BC, Canada.
Pediatr Res. 2025 Sep 18. doi: 10.1038/s41390-025-04389-z.
In extremely preterm neonates, left atrium to aortic root ratio (LA:Ao) and fractional shortening (FS) are commonly used to assess left atrial dilation and left ventricle (LV) function in targeted neonatal echocardiography (TNE), respectively. This study aims to compare the interrater reliability of two-dimensional (2D) and motion mode (M-mode) echocardiography for LA:Ao and FS in preterm infants.
A retrospective study was conducted on infants born <29 weeks gestational age who underwent TNE between October 2020 and September 2023. Interrater reliability for LA:Ao and FS was assessed using intraclass correlation coefficients (ICC) and Bland-Altman plots.
A total of 79 scans were assessed independently by 3 observers. Median (IQR) gestational age was 25.9 (25-27.3) weeks and birth weight was 797 (627.5-916.8) grams. The ICC for LV FS was 0.72 (95% CI: 0.40-0.85) for M-mode and 0.73 (95% CI: 0.56-0.83) for 2D-mode. The ICC for LA: Ao ratio was 0.87 for both M-mode (95% CI: 0.73-0.93) and 2D-mode (95% CI: 0.80-0.92).
Both 2D and M-modes are reliable markers for measuring LA:Ao ratio in extremely preterm infants. FS should be interpreted cautiously as a marker of LV systolic function due to moderate reliability.
Both two-dimensional (2D) and motion mode (M-mode) showed good interrater reliability for left atrium and aortic root (LA:Ao) ratio measurements. Findings support the interchangeable use of either modality in Targeted Neonatal Echocardiography (TNE) protocols for assessing left atrial dilation. Fractional shortening (FS) demonstrated only moderate reliability across both modes, reinforcing the need for caution when using FS alone to evaluate left ventricular function in extremely preterm infants. Results highlight the importance of standardized TNE-specific training and routine quality assurance (QA) processes to minimize variability and ensure consistency in clinical and research echocardiographic assessments.
在极早产儿中，左心房与主动脉根部比值（LA:Ao）和缩短分数（FS）通常分别用于在目标新生儿超声心动图（TNE）中评估左心房扩张和左心室（LV）功能。本研究旨在比较二维（2D）和M型超声心动图在早产儿中测量LA:Ao和FS的观察者间可靠性。
对2020年10月至2023年9月期间接受TNE的孕周小于29周的婴儿进行回顾性研究。使用组内相关系数（ICC）和Bland-Altman图评估LA:Ao和FS的观察者间可靠性。
3名观察者独立评估了总共79次扫描。中位（IQR）孕周为25.9（25 - 27.3）周，出生体重为797（627.5 - 916.8）克。M型超声心动图测量LV FS的ICC为0.72（95%CI：0.40 - 0.85），2D模式为0.73（95%CI：0.56 - 0.83）。M型（95%CI：0.73 - 0.93）和2D模式（95%CI：0.80 - 0.92）测量LA: Ao比值的ICC均为0.87。
2D和M型模式都是测量极早产儿LA:Ao比值的可靠指标。由于可靠性中等，FS作为LV收缩功能指标应谨慎解读。
二维（2D）和M型模式在测量左心房与主动脉根部（LA:Ao）比值方面均显示出良好的观察者间可靠性。研究结果支持在目标新生儿超声心动图（TNE）方案中可互换使用这两种模式来评估左心房扩张。缩短分数（FS）在两种模式下仅显示出中等可靠性，这强化了在极早产儿中单独使用FS评估左心室功能时需谨慎的必要性。结果强调了标准化TNE特定培训和常规质量保证（QA）流程的重要性，以尽量减少变异性并确保临床和研究超声心动图评估的一致性。
