Cost-effective fetal lung volumetry for assessment of congenital diaphragmatic hernia

(1) To investigate the reproducibility of total fetal lung volume (TFLV) measurements using a free 3D modeling software (3DSlicer); (2) To correlate these measurements with lung-to-head ratio (LHR) or TFLV measured using PACS and; (3) To determine the role of 3DSlicer in predicting perinatal outcomes in cases with congenital diaphragmatic hernia (CDH) who had fetal tracheal occlusion (FETO).

Retrospective cohort study between 2012 and 2017 at Texas Children's Hospital (2011-2017), including all patients who underwent FETO for CDH. LHR was measured by ultrasound and TFLV was measured by MRI at the time of referral and 6 weeks after FETO using 3DSlicer and PACS. We evaluated intra- and inter-rater reliability of TFLV measurement using 3DSlicer, infant survival to 1 year, need for ECMO and pulmonary hypertension.

The intra- and inter-rater reliability of TFLV measured with 3DSlicer was excellent before and after FETO (Intra-class correlation coefficient: 0.98-0.99 and 0.94-0.99, respectively). There was a good correlation between TFLV measured with PACS and with 3DSlicer before and after FETO (r = 0.78 and r = 0.99, respectively). Similarly, there was a good correlation between TFLV measurements using PACS or 3DSlicer and LHR after FETO (r = 0.86 and r = 0.88, respectively). Infants who survived to 1 year had a significantly higher TFLV evaluated with 3DSlicer before FETO compared to non-surviving infants (OR = 1.16[1.1-1.3], p = 0.03) as well as a significantly higher TFLV evaluated by 3DSlicer after FETO (OR = 1.2[1-1.4], p = 0.04).

Lung volume measurements using free 3DSlicer in infants with severe CDH who underwent FETO are reproducible and reliable, and have comparable predictive capability for survival as those measured using conventional software.