Assessment of the 3D-recontructed images suggested that the rotational component to the deformities of the radius and ulna was more minimal, that is, the distal and proximal landmarks of each bone appeared to have maintained their usual rotational relations.Preoperative planning (Figure 3) used paper, scissors and tape and would have been facilitated by 3D prototypes of the bones involved.Figure 3Preoperative planning��defining the osteotomies.The operation involved two closing wedge osteotomies (uniplanar for the ulna and biplanar for the radius). The excised radial wedge was used as bone graft at the ulna and direct compression plates (3.5mm) were used in compression mode. Nonsteroidal anti-inflammatory drugs were excluded postoperatively.

A cast was used post-operatively for 4 weeks, and radiographs were obtained at 6 weeks and 12 weeks (Figure 4). At 12 weeks, he had clinical and radiographic union, and pronation/supination had improved to 75��/90��.Figure 4Postoperative results��orthogonal views at twelve weeks.Whilst dealing with this case of malunion of the shaft of radius and ulna, CT scanning with 3D formatting was performed to aid preoperative planning. These images suggested that there was no rotational component to the malunion and that accordingly the osteotomies could be planned relatively easily in a conventional manner. However, as an adjunct it was hoped to use the DICOM data to build a model of the malunited bones.

A university research department was involved, and several problems were encountered: (i) time delay due to file conversion problems, (ii) only a truncated model was produced, and (iii) the angle of malunion did not conform to that obtained on CT scan, that is, the model was not an accurate representation.The cost of RP is primarily determined by the amount of the material used, the cost of the 3D printer, and the cost of the software licenses required to process the DICOM data to a suitable format to print. With this in mind, our aim was to find an appropriate method to produce an acceptable orthopaedic RP model (i) at minimal cost, (ii) without access to a university research department, and (iii) using open-source software and a public-access 3D printing service. A secondary aim was to compare the cost involved against quotations from established companies.3.

MethodBefore models were produced or the patient’s images processed, informed consent was obtained for use of case material and CT images/data for research and publication.The CT scans of the patient’s forearm were then processed using open-source software OsiriX (DICOM image processing software for OS X) and MeshLab Brefeldin_A (a system for the processing and editing of unstructured 3D triangular meshes). Both packages are distributed under open-source licensing��Lesser General Public Licence (LGPL)��and are therefore free.