3D-printed soft-tissue physical models of renal malignancies for individualized surgical simulation: a feasibility study

Abstract

To construct patient-specific physical three-dimensional (3D) models of renal units with materials that approximates the properties of renal tissue to allow pre-operative and robotic training surgical simulation, 3D physical kidney models were created (3DSystems, Rock Hill, SC) using computerized tomography to segment structures of interest (parenchyma, vasculature, collection system, and tumor). Images were converted to a 3D surface mesh file for fabrication using a multi-jet 3D printer. A novel construction technique was employed to approximate normal renal tissue texture, printers selectively deposited photopolymer material forming the outer shell of the kidney, and subsequently, an agarose gel solution was injected into the inner cavity recreating the spongier renal parenchyma. We constructed seven models of renal units with suspected malignancies. Partial nephrectomy and renorrhaphy were performed on each of the replicas. Subsequently all patients successfully underwent robotic partial nephrectomy. Average tumor diameter was 4.4 cm, warm ischemia time was 25 min, RENAL nephrometry score was 7.4, and surgical margins were negative. A comparison was made between the seven cases and the Tulane Urology prospectively maintained robotic partial nephrectomy database. Patients with surgical models had larger tumors, higher nephrometry score, longer warm ischemic time, fewer positive surgical margins, shorter hospitalization, and fewer post-operative complications; however, the only significant finding was lower estimated blood loss (186 cc vs 236; p = 0.01). In this feasibility study, pre-operative resectable physical 3D models can be constructed and used as patient-specific surgical simulation tools; further study will need to demonstrate if this results in improvement of surgical outcomes and robotic simulation education.