Current State of Surgical Simulation Training in Otolaryngology : Systematic Review of Simulation Training Models

The review article presents an expansive list of otolaryngology-specific surgical simulation training models as described in Otolaryngology literature as well as evaluates recent advances in simulation training in Otolaryngology.


INTRODUCTION
Despite the changing landscape of surgical education with work hour restrictions and decreased independence of trainees in clinical activities [1], most training programs uphold a traditional, dogmatic approach in training surgical residents that is based on a hierarchical, apprenticeship model.This model often presents a time-intensive learning curve that leaves trainees, who are often first responders, with limited knowledge, skills, and confidence to competently deal with clinical and surgical scenarios of varying complexity [2].
Given that surgical simulation training (SST) engages the two most important tenets of adult learning by permitting moderated practical experience in the setting of guided reflection, it holds great potential in the training of otolaryngologists [3].Although surgical simulation activities can be resource intensive [4,5], the availability of an arena for deliberate practice in a risk-free, low stress environment is an effective way to acquire skills specific to the practice of a surgical subspecialty [6].
Simulation models remain a keystone in the design of high-yield simulation activities by providing apparatus for the acquisition of surgical skills which can then be transferred to patient care [7].Simulation models fulfill the role of physical vessels that afford trainees an opportunity to hone their psychomotor and decision-making skills without the loom of patient risk [2].
With the exponential advancement of computing and manufacturing technologies, several simulation models and platforms have recently been developed and deployed in the training of surgical residents.On a broad scale, these simulation models can be categorized into synthetic bench models, computer-based models (virtual reality or web-based), animal models (tissue or live), and human cadaveric models [8], while a combination of any of these constitutes a hybrid model.
In this study, we aim to provide a systematically reviewed list of otolaryngology simulators that are documented in Otolaryngology literature and discuss recent updates in simulation training in the field of otolaryngology.

Study Selection
With the assistance of an information specialist, an a priori research protocol was designed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology [9].Subsequently, a sensitive systematic review was performed to obtain as many articles from five databases namely Ovid/Medline, PubMed, Embase, Web of Science, and Cochrane.The databases were searched from their inception through July 18 th , 2018.Two investigators (M.A. and M.L.) conducted the search and reviewed selected articles.Subject headings for the search included otolaryngology, otology, airway, laryngeal, rhinology, reconstructive, facial plastics, and head and neck oncology cross-referenced with the terms simulation, simulation training, and simulation models.Bibliographies were manually searched to identify studies that met inclusion criteria.Inter-investigator discordances in the review process were resolved by consensus.

Eligibility Criteria
For the systematic review, all articles in the English literature reporting on simulation training in otolaryngology were eligible for inclusion.After elimination of duplicates, all articles were subjected to a title and abstract screen.Articles were excluded if they (1) did not report adequate data regarding the simulation model/platform, (2) did not report on otolaryngology-specific simulation, (3) were not unique (i.e., reporting simulation models already described at an earlier date in other included records), (4) were review articles not describing original simulation models and/or (5) were not reported in English language.
Full texts of the remaining articles were then comprehensively reviewed.A flow chart of the systematic review design with complete numeric details is provided in Figure 1.

Data Extraction and Statistical Analysis
All data was systematically aggregated using Microsoft Excel software, version 16.12 (Microsoft).Extracted data end points included: author names, year of publication, type of simulation model, brief description of the simulation model, and key training objective of the simulation model.Data was analyzed using SPSS software, version 25 (IBM).Descriptive statistics were applied in the analysis of all studies that met inclusion criteria.

Overall Characteristics
The systematic review selection process and its results are captured in Figure 1.The aforementioned search strategy was applied to Ovid/Medline, PubMed, Embase, Web of Science, and Cochrane databases which yielded an output of 178 records.After elimination of duplicates, a total of 165 articles remained, which were subsequently subjected to title and abstract screening.The remaining 112 articles were subjected to comprehensive, full-text analysis.Ultimately 83 articles met final inclusion criteria.A manual bibliography screen of included articles yielded an additional 20 articles.A total of 103 records were deemed eligible for inclusion.A single record [10] was utilized as a double entry in two separate categories, yielding a grand total of 104 described otolaryngologic surgical simulation models.Of the records included, only 8 simulation models were reported in or before the year 2004, 20 were reported between 2005 and 2009, 34 models were reported between 2010 and 2014, and 42 models were described in or after the year 2015 (Figure 2).There were a total of 50 synthetic, 21 computer-based, 19 animal cadaver, 6 human cadaver, and 8 hybrid models described.Synthetic simulators were the most common type of simulators in all categories with the exception of oncology/facial plastics/

REVIEW
reconstructive category, where animal cadaver models were more common.
Otology had the highest number (n = 37) of reported simulation models while oncology/facial plastics/reconstruction had the lowest number (n = 16).A total of 18 of 50 synthetic simulators were purely formulated using 3D-printing techniques (36%), with only 5 such models described prior to 2015 (27.78%).A majority of 3D-printed synthetic simulators were described in or after the year 2015.Hybrid models most often constituted animal cadaver models combined with synthetic models (75%), with the remainder comprised of synthetic models combined with computer-based platforms (25%).

Airway/Laryngeal Surgical Simulation Models
A total of 34 airway and laryngeal surgical simulation models were identified (Table 1).These models consisted of 18 synthetic, 4 computer-based, 8 animal cadaver, 4 human cadaver, and 4 hybrid models.Of these 34 models, 15 models were aimed at establishing emergency surgical airways.Only 6 articles captured in our review reported on pediatric airway/laryngeal simulation models.Of all the airway/laryngeal surgical simulators, 9 models were manufactured by 3D printing (26.47%).

Oncology/Facial Plastics/Reconstruction Simulation Models
A total of 16 oncology/facial plastics/reconstructive simulation models were identified (Table 2).These models consisted of 5 synthetic, 1 computer-based, 10 animal cadaver, and 2 hybrid models.Of the 16 simulation models, 4 were manufactured by 3D-printing (25%).Animal cadaver models were the most common type of surgical simulation platform.Chicken cadaver parts remained the most widely reported simulation model for microvascular anastomosis training in otolaryngology literature.

Rhinology
A total of 17 rhinology surgical simulation models were identified in our systematic review (Table 3).These models consisted of 6 synthetic, 6 computer-based, 5 animal cadaver, and 2 hybrid models.Of these 17 models, 14 articles described simulation models aimed at enhancing overall skills required for Endoscopic Sinus and Skull Base Surgery (ESSS) (88%).Only 3 out of 17 simulation models were designed for specific endoscopic tasks (12%), and only 2 out of 17 models were manufactured by 3D-printing (11.76%).

Otology
A total of 37 neuro-otologic simulation models were identified (Table 4).These models consisted of 21 synthetic, 12 computer-based, 2 animal cadaver, and 2 human cadaver models.All computer-based models utilized virtual reality (VR) technology except for one web-based otoscopy simulator aimed for otoscopic examination described by Wickens et al. [11].Of these 37 models, 13 were intended for temporal bone drilling (35%), 11 were aimed for ventilation tube insertion (30%), and others included a variety of other ear surgical procedures.Only 3 out of 37 simulation models were manufactured by 3D-printing (8.11%).

DISCUSSION
The implicit appeal of using simulation platforms in training is that mistakes on a simulation platform have no real-world consequences other than to serve as a marker of different degrees of task achievement [12].Perhaps the

REVIEW
best example of the successful use of simulation in training is that of the flight simulator created by Edward Link in 1929 to train novice pilots [13].In time, simulation-based training has grown to be the industry standard in avionics and is currently used for a variety of reasons from the training of novice pilots to flight testing of new aircraft systems [12].
The parallels to the use of simulation in surgical training are striking.With the increasing awareness of ethical concerns, complexity of surgical procedures, healthcare costs, and clinical governance, surgical trainees are faced with new-age hurdles to achieve proficiency and competency within the confines of a structured timeframe.Additionally, limited availability of time for teaching and learning due to work hour restrictions has led to a detachment from the traditional Halstedian dogma of "master and apprentice" [14].Given that simulation provides a tool for aptitude testing, early skills acquisition, and advanced skills training [12], the development and use of surgical simulation training models have more recently gained exponent popularity as demonstrated by Figure 2. In their cross-sectional survey-based study, Deutsch et al managed to investigate interest amidst 43 US otolaryngology residency programs in advancing simulation training, with 92.9% of respondents confirming the presence of a simulation center or program at their institution and 83.8% of respondents indicating interest in participating in multi-center simulation initiatives [15].These findings are reflective of a transition in the core philosophy of surgical education.
Historically, human and animal cadaver models and live animal models provided the mainstay raw material for simulation activities.As noted by Musbahi et al, the authors agree that it remains difficult to surpass the ability of human and animal cadaver models to provide anatomic accuracy, tissue con-  As shown by our results, an ever-increasing number of anatomically accurate, customized 3D-printed models are being created as this technology becomes more available, accessible, and user-friendly.We are inclined to agree with VanKoevering and Malloy [16] in that 3D-printing provides surgical educators a unique advantage by affording an opportunity to rapidly create complex head and neck anatomical models that can be utilized for procedural training.
These 3D-printed models hold an advantage over computer-based platforms because they permit tactile sensation and the use of real instruments.Given that these factors are intrinsic components in psychomotor skills training, 3D-printed models hold a distinct advantage over computer-based simulation platforms [16].
In the last five years, the arena of surgical simulation has seen a sharp increase in VR simulation platforms [17].With the advancement in digital 3D-

REVIEW
visualization along with haptic sensory technology, VR simulation models are providing a more interactive experience than ever before.Our findings are in line with the findings of Arora et al in that VR simulators appear to be most commonly employed in the subspecialties of rhinology and otology [17].
Although prior review articles have attempted to provide a database of otolaryngologic simulation models, the recent rapid increase in the number of documented simulation models warranted an updated review of otolaryngology simulation models.To the best of our knowledge, our paper presents the most expansive database of otolaryngology-specific simulation models, such that we report a total of 104 simulation models compared to the 60 models reported by Javia et al. [118] and the 64 models reported by Musbahi et al [8].Additionally, in contrast to works such as that of Bhutta et al. [119] and Chan et al. [120], our article discusses simulation models in all divisions of otolaryngology, rather than addressing simulation training in only a single otolaryngologic sub-specialty.Our article also serves as the first to describe otolaryngology-specific simulators under a dedicated Oncologic/Facial Plastics/Reconstruction category in addition to other more commonly described categories such as Airway/Laryngeal, Pediatric, Rhinology, and Otology.
Due to the vast and ever-growing number of documented simulation models, the systematic review performed may not have captured a comprehensive list of all available surgical simulators.Evaluation of quality and validity of individual simulation platforms was also not conducted given that this be beyond the scope of this paper.The data presented by the authors is instead intended to provide an expansive list that contains simulation options that would adequately suffice the breadth of otolaryngologic training for medical students and residents alike.Future projects can be geared towards the expansion and validity testing of this dynamic and constantly growing list, and utilize it to create a standard, uniform, cost-effective, and high-fidelity simulation curriculum that can be employed by otolaryngology training programs.

CONCLUSION
Current literature shows the availability of several otolaryngology-specific simulation models that have proven beneficial in otolaryngologic surgical training.Recent advancements in manufacturing and computing technologies are contributing to a paradigm shift in surgical simulation education.
With the availability of these options, there exists the potential to establish a well-structured and standardized approach to simulation activities across otolaryngology training programs.

Figure 1 .
Figure 1.Systematic review flowchart based on the PRISMA methodology.PRISMA, preferred reporting items for systematic reviews and meta-analyses.

Figure 2 .
Figure 2. Number of simulation models described in literature over time.

Table 1 .
Summary of Airway/Laryngeal Surgical Simulation Models

Table 3 .
Summary of Endoscopic Sinus and Skull Base Surgical Simulation Models

Table 4 .
Summary of Otology Surgical Simulation Models