Surgeons rate 3D printed heart models higher than imaging cardiologists did

A feasibility study from The Hospital for Sick Children in Toronto has revealed that pediatric cardiothoracic surgeons scored 3D printed heart models built from three fused imaging sources higher than imaging cardiologists did. Surgeons gave the valve-integrated models a median rating of 5 out of 5, against a median of 4 from imaging cardiologists.

The models merged cardiac CT, MRI and 3D echocardiography into a single printed anatomy for each of 10 pediatric patients with complex congenital heart disease, including atrioventricular septal defects, functionally single-ventricle hearts and complex congenital mitral valve disease.

Each imaging method captured a different part of the anatomy well and other parts poorly. Valve tissue moved continuously through the cardiac cycle, which left it poorly resolved in models derived from a single modality. 

Datasets from each modality were segmented and combined so the printed models carried valve structures alongside the chamber and vessel anatomy that CT and MRI already handled.

A 10-member faculty panel of seven pediatric cardiologists and three pediatric cardiothoracic surgeons assessed the models on Likert scales covering anatomical understanding and preoperative surgical planning. The gap between the two groups pointed the value of the added valve detail toward the operating room rather than the reading room, which matters for hospital point-of-care AM programs deciding which cases justify the segmentation and printing time. 

Congenital heart disease affects roughly 1 in 100 newborns worldwide, giving pediatric cardiac programs a steady case volume against which to weigh that cost.

The study was not without its flaws, with its authors noting limitations which included 10 patients and 10 evaluators, an imbalance of seven imagers to three surgeons, subjective rather than blinded assessment, and no downstream data on operative time or complications.