MRI prediction of fetal lung volumes and the impact on counselling

Development of a 3D-printed neonatal congenital diaphragmatic hernia model and standardisation of intra-operative measurement

INTRODUCTION

Three-dimensional (3D) printing is frequently used for surgical simulation and training, however, no widely available model exists for neonatal congenital diaphragmatic hernia (CDH). The aim of this study was to develop a 3D-printed model of CDH and test interobserver variability in the simulated model for obtaining measurements of the diaphragmatic defect and ipsilateral diaphragm.

METHODS

A term fetal MRI (3.5 kg) of thorax, diaphragm and defect (15 mm × 5 mm) were delineated and segmented after parental consent to produce 3D-printed models. Consultant and trainee paediatric surgeons were invited to measure the posterior-lateral diaphragmatic defect and ipsilateral diaphragm. Mean measurement error was calculated (millimetres). Data are presented as median (range) and number/total (%).

RESULTS

An abdominal and thoracoscopic model were produced and tested by 52 participants (20 consultants and 32 trainees). Diaphragmatic defect via laparotomy measured 15 (10-20) mm (AP) × 16 (10-25) mm (ML) and thoracoscopically 14 (11-19) mm (AP) × 15 (11-22) mm (ML). Mean error per measurement was 4 (1-17) mm via laparotomy vs. 3 (0.5-9.5) mm thoracoscopically. Mean error was similar between consultants and trainees via laparotomy (4.3 vs. 3.9 mm, p = 0.70) and thoracoscopically (3 vs. 3 mm, p = 0.79). Error did not correlate with experience as operating surgeon via laparotomy (β = 13.0 [95% CI - 55.9 to 82.0], p = 0.71) or thoracoscopically (β = 1.4[95% CI - 6.4 to 9.2], p = 0.73.

CONCLUSIONS

We have designed and built simulation models for CDH repair via laparotomy and thoracoscopically. Operators can reliably measure the diaphragmatic defect and ipsilateral diaphragm, regardless of surgical experience and operative approach.