Although not readily accessible yet to many community and hospital pharmacists, fuse deposition modelling (FDM) is a 3D printing technique that can be used to create a 3D pharmaceutical dosage form by employing drug loaded filaments extruded via a nozzle, melted and deposited layer by layer. FDM requires printable filaments, which are commonly manufactured by hot melt extrusion, and identifying a suitable extrudable drug-excipient mixture can sometimes be challenging. We propose here the use of passive diffusion as an accessible loading method for filaments that can be printed using FDM technology to allow for the fabrication of oral personalised medicines in clinical settings. Utilising Hansen Solubility Parameters (HSP) and the concept of HSP distances (Ra) between drug, solvent, and filament, we have developed a facile pre-screening tool for the selection of the optimal combination that can provide a high drug loading (a high solvent-drug Ra, >10, and an intermediate solvent–filament Ra value, ~10). We have identified that other parameters such as surface roughness and stiffness also play a key role in enhancing passive diffusion of the drug into the filaments. A predictive model for drug loading was developed based on Support Vector Machine (SVM) regression and indicated a strong correlation between both Ra and filament stiffness and the diffusion capacity of a model BCS Class II drug, nifedipine (NFD), into the filaments. A drug loading, close to 3% w/w, was achieved. 3D printed tablets prepared using a PVA-derived filament (Hydrosupport, 3D Fuel) showed promising characteristics in terms of dissolution (with a sustained release over 24 h) and predicted chemical stability (>3 years at 25 °C/60% relative humidity), similar to commercially available NFD oral dosage forms. We believe FDM coupled with passive diffusion could be implemented easily in clinical settings for the manufacture of tailored personalised medicines, which can be stored over long periods of time (similar to industrially manufactured solid dosage forms).
Background: Simulation-based techniques using three-dimensional models are gaining popularity in neurosurgical training. Most pre-existing models are expensive, so we felt a need to develop a real-life model using 3D printing technology to train in endoscopic third ventriculostomy. Methods: The brain model was made using a 3D-printed resin mold from patient-specific MRI data. The mold was filled with silicone Ecoflex™ 00-10 and mixed with Silc Pig® pigment additives to replicate the color and consistency of brain tissue. The dura mater was made from quick-drying silicone paste admixed with gray dye. The blood vessels were made from a silicone 3D-printed mold based on magnetic resonance imaging. Liquid containing paprika oleoresin dye was used to simulate blood and was pumped through the vessels to simulate pulsatile motion. Results: Seven residents and eight senior neurosurgeons were recruited to test our model. The participants reported that the size and anatomy of the elements were very similar to real structures. The model was helpful for training neuroendoscopic 3D perception and navigation. Conclusions: We developed an endoscopic third ventriculostomy training model using 3D printing technology that provides anatomical precision and a realistic simulation. We hope our model can provide an indispensable tool for young neurosurgeons to gain operative experience without exposing patients to risk.