Training in Endoscopic Endonasal Neurosurgical Procedures: A Systematic Review of Available Models

Background  Proficiency in endoscopic endonasal neurosurgery (EEN) requires a unique skill set and is associated with a steep learning curve. Endoscopic simulation models present an opportunity for trainees to enhance their surgical skills and anatomical knowledge in a risk-free environment. To date, four main categories of training modalities have been recorded: cadaveric, synthetic, animal, and virtual reality (VR) models. Objective  To analyze the features, advantages, and disadvantages of each EEN training model, describe the stages of simulation training, and propose future directions and an algorithm for EEN training. Methods  A comprehensive search was performed including studies published before August 2023. Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines were followed in the search and selection of studies included. Terms for searching title and abstract were ("simulation") AND ("endoscopic endonasal") AND ("neurosurgery"). Articles were screened for duplicates. Results  We identified 176 references, and 43 studies met the inclusion criteria. Twelve articles described cadaveric models, with 7 of them studying internal carotid artery injury management and 3 of them cerebrospinal leak management; 14 articles illustrated synthetic models, 1 article described animal models, and 16 articles outlined VR simulations. Conclusion  Cadaveric models provide realistic hands-on surgical training with acceptable cost and remain the gold standard method for training. Synthetic models are most effective for preoperative surgical planning and animal models are best suited for testing of hemostatic techniques. VR simulations may become the primary training tool for novices to acquire basic anatomical knowledge as their role in advanced surgical planning is evolving.