Development of a patient specific cartilage graft using magnetic resonance imaging and 3D printing

Abstract

Objectives

The goal of this project was to develop and validate a patient-specific, anatomically correct graft for cartilage restoration using magnetic resonance imaging (MRI) data and 3-dimensional (3D) printing technology. The specific aim was to test the accuracy of a novel method for 3D printing and implanting individualized, anatomically shaped bio-scaffolds to treat cartilage defects in a human cadaveric model. We hypothesized that an individualized, anatomic 3D-printed scaffold designed from MRI data would provide a more optimal fill for a large cartilage defect compared to a generic flat scaffold.

Methods

Four focal cartilage defects (FCDs) were created in paired human cadaver knees, age <40 years, in the weight-bearing surfaces of the medial femoral condyle (MFC), lateral femoral condyle (LFC), patella, and trochlea of each knee. MRIs were obtained, anatomic grafts were designed and 3D printed for the left knee as an experimental group, and generic flat grafts for the right knee as a control group. Grafts were implanted into corresponding defects and fixed using tissue adhesive. Repeat post-implant MRIs were obtained. Graft step-off was measured as the distance in mm between the surface of the graft and the native cartilage surface in a direction perpendicular to the subchondral bone. Graft contour was measured as the gap between the undersurface of the graft and the subchondral bone in a direction perpendicular to the joint surface.

Results

Graft step-off was statistically significantly better for the anatomic grafts compared to the generic grafts in the MFC (0.0 ​± ​0.2 ​mm vs. 0.7 ​± ​0.5 ​mm, p ​< ​0.001), LFC (0.1 ​± ​0.3 ​mm vs. 1.0 ​± ​0.2 ​mm, p ​< ​0.001), patella (−0.2 ​± ​0.3 ​mm vs. −1.2 ​± ​0.4 ​mm, p ​< ​0.001), and trochlea (−0.4 ​± ​0.3 vs. 0.4 ​± ​0.7, p ​= ​0.003). Graft contour was statistically significantly better for the anatomic grafts in the LFC (0.0 ​± ​0.0 ​mm vs. 0.2 ​± ​0.4 ​mm, p ​= ​0.022) and trochlea (0.0 ​± ​0.0 ​mm vs. 1.4 ​± ​0.7 ​mm, p ​< ​0.001). The anatomic grafts had an observed maximum step-off of −0.9 ​mm and a maximum contour mismatch of 0.8 ​mm.

Conclusion

This study validates a process designed to fabricate anatomically accurate cartilage grafts using MRI and 3D printing technology. Anatomic grafts demonstrated superior fit compared to generic flat grafts.

Level of evidence

Level IV.