Specialised Surgical Instruments for Endoscopic and Endoscope-Assisted Neurosurgery: A Systematic Review of Safety, Efficacy and Usability

Simple Summary The greatest technical barrier in endoscopic and endoscope-assisted neurosurgery is the instruments employed for these approaches. This systematic review aimed to identify specialised instruments for this type of surgery and evaluate their safety, efficacy and usability. We identified 50 instruments over 60 studies that were broadly safe and effective and generally considered to be ergonomic, though the learning curve was often noted as a disadvantage. Only eight studies compared the new instrument to standard instruments and comparisons were generally favourable to the new instrument. The development of novel and specialised instruments for endoscopic and endoscope-assisted neurosurgery is an area of interest for the field, but these instruments do not meet the need for improved articulation and future development should be based on established guidelines for neurosurgical innovation. Abstract While there have been great strides in endoscopic and endoscope-assisted neurosurgical approaches, particularly in the treatment of deep-sited brain and skull base tumours, the greatest technical barrier to their adoption has been the availability of suitable surgical instruments. This systematic review seeks to identify specialised instruments for these approaches and evaluate their safety, efficacy and usability. Conducted in accordance with the PRISMA guidelines, Medline, Embase, CENTRAL, SCOPUS and Web of Science were searched. Original research studies that reported the use of specialised mechanical instruments that manipulate tissue in human patients, cadavers or surgical models were included. The results identified 50 specialised instruments over 62 studies. Objective measures of safety were reported in 32 out of 62 studies, and 20 reported objective measures of efficacy. Instruments were broadly safe and effective with one instrument malfunction noted. Measures of usability were reported in 15 studies, with seven reporting on ergonomics and eight on the instruments learning curve. Instruments with reports on usability were generally considered to be ergonomic, though learning curve was often considered a disadvantage. Comparisons to standard instruments were made in eight studies and were generally favourable. While there are many specialised instruments for endoscopic and endoscope-assisted neurosurgery available, the evidence for their safety, efficacy and usability is limited with non-standardised reporting and few comparative studies to standard instruments. Future innovation should be tailored to unmet clinical needs, and evaluation guided by structured development processes.


Introduction
In Lewis Carroll's 1871 novel Through The Looking Glass, and What Alice Found There, Alice falls into a new world that shifts her perception of what she thought she knew and was possible [1]. Over the past 100 years neurosurgeons have been through their own looking glass: the endoscope. In the last 30 years, many new endoscopic and endoscopeassisted neurosurgical approaches and procedures have been developed, facilitated by a rise in complimentary innovation in endoscopic instruments and surgical techniques [2].
The development of endoscopic and endoscope-assisted approaches has played a key role in treating and improving the outcomes of brain tumours. The endoscopic endonasal transsphenoidal approach, for instance, may decrease the incidence of surgical complications when compared with traditional microsurgical cases in treating pituitary adenomas, likely a direct result of the improved visualisation of anatomy this approach provides [3]. Utilisation of endoscopic approaches in treating patients with sinonasal and ventral skull base cancers has also been found to significantly improve patient quality of life scores within the first postoperative year [4].
However, despite the great strides made in endoscopy, it is the use of surgical instruments in endoscopic approaches that has remained the greatest technical barrier to their adoption [5]. Presently, many of the instruments used in endoscopic and endoscope-assisted neurosurgical approaches have been adopted from the armamentarium of neighbouring specialties such as rhinology and urology. These include instruments that grasp, such as forceps, and instruments that cut and divide, such as scissors, knives and punches [6]. Although most such instruments are straight, some angled instruments are also in popular use, particularly in combination with angled endoscopes. Alongside these traditional instruments special classes have been developed, including microdebriders for precise tissue removal, ultrasonic devices for removal of firm tumours that are not easily suctioned and mono-and bipolar forceps for coagulation.
Historically, the introduction of new instruments such as these has been unstructured and variable. Though assessment frameworks for innovation in neurosurgery, including the development of novel neurosurgical instruments, have been developed [7,8], an instrument may be used for the first time in the form of a dedicated research study but, more often, may be published as a non-comparative trial without special institutional board review. Although many such instruments are safe and effective, the dangers of this process are obvious and have been frequently reported in the literature [9]. The usability of a new instrument, including its ergonomics and learning curve, is also an important factor when considering its adoption within the surgical community.
The aims of this review were therefore first, to identify specialised instruments that have been developed for endoscopic and endoscope-assisted neurosurgery; second, to assess the evidence for the safety and efficacy of these instruments; and third, to assess their usability.

Materials and Methods
This review was conducted and the results are presented in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [10]. The review protocol was registered on the PROSPERO database (CRD42020209146).

Literature Search
A search of the Cochrane Controlled Register of Trials (CENTRAL), MEDLINE (via Ovid), Embase (via Ovid), SCOPUS and Web of Science databases from their inception was performed on 15 June 2020. No filters or limitations were applied to the search. Keywords, Medical Subject Heading (MeSH) terms and their synonyms for "surgical instrument", "endoscopic surgery" and "neurosurgery" were combined. The full search strategies for each database are available in Appendix A. Following screening and selection, an additional manual search of the reference lists of included studies was conducted.

Eligibility Criteria
The following 'PICO' framework was employed to assess the eligibility of articles: Population: human patients, surgical models, cadavers; Intervention: specialised instruments for endoscopic or endoscope-assisted neurosurgery; Comparison: standard instruments for neurosurgery; Outcome: safety measures (blood loss, intraoperative complications, postoperative complications, mortality rate), efficacy measures (EOR, operative time), assessments of ergonomics, assessments of learning curve. Eligible articles were included in the review if they met the following criteria: the specialised instrument is mechanical with the aim of manipulating tissue; the instrument was applied to endoscopic or endoscope-assisted neurosurgery; and the study was an original research paper, including preclinical and clinical studies.
Eligible articles were excluded if they met any of the following criteria: the study evaluated a device or system that was not a mechanical instrument with the aim of manipulating tissue; the instrument was not applied to endoscopic or endoscope-assisted neurosurgery; or the study was not an original research paper.

Study Screening and Selection
Results of the final searches of all databases were imported to EndNote X9.3.2 (Philadelphia, PA, USA). After removal of duplicates, an initial screening of the title and abstract of each article using the eligibility criteria was performed. The full text of articles that remained included following this screen were then read to ensure all eligibility, inclusion and exclusion criteria were met before extracting data. All stages of screening were carried out independently by two reviewers (H.A. and J.C.). Any discrepancies between the two screenings were discussed and resolved with a third reviewer (H.J.M.).

Data Extraction and Analysis
The following data were extracted from included articles: study design; study population and sample size; pathology; procedure; instrument name; company that produced the instrument and their location; type of surgery the instrument was for; instrument features; instrument function; reports of instrument malfunction; patient outcomes measures that provided evidence for safety; general comments on safety; surgical outcomes that provided evidence for efficacy; general comments on efficacy; any comparison to standard neurosurgical instruments and which instruments; any ergonomic assessment or comments on instrument ergonomics; any learning curve assessment or comment on instrument learning curve; and limitations of the instrument. Summary statistics with accompanying narrative synthesis were generated for the identified available instruments by safety, efficacy and usability metrics used. Comparative statistical analyses were not performed due to the heterogeneity of the underlying metrics.

Search Results
Following the search strategy detailed in Section 2 and Appendix A, a total of 5757 articles were identified ( Figure 1). After removal of duplicates 4143 titles and abstracts were screened. Following the full-text screen, 62 articles were identified as fully meeting the eligibility and inclusion criteria and were included in the final analysis. The reference lists of these 62 articles were then screened for other potentially relevant articles that were not identified through the database search. This screening yielded zero results.
Novel instruments in included studies were applied to a variety of specialities ( Figure 3). The most common application of a novel instrument was for the treatment of pituitary tumours using the endoscopic endonasal transsphenoidal approach (eight instruments), followed by endoscopic third ventriculostomy for hydrocephalus (seven instruments).

Available Instruments
A total of 50 specialised instruments for endoscopic or endoscope-assisted neurosurgery were identified (Table 1). Most available specialised instruments were for general use in all endoscopic and endoscope-assisted neurosurgical approaches, but served a single function. The most common function of these instruments was resection (28 instruments), followed by dissection (5), coagulation (5), electrocautery (4) and retraction (4) (Figure 4).  An additional nine instruments were reported for specific approaches or procedures and capable of performing specialised single-functions: two instruments were designed for endoscopic third ventriculostomy to dilate the floor of the third ventricle; one instrument for craniosynostosis surgery; one as suction for intraventricular haematoma procedures; one to retrieve resected sections of colloid cysts; one for retraction during laminectomy; one for retraction during the endoscopic endonasal transsphenoidal approach; one for severing the sympathetic nerve of thoracic vertebrae; and one as a cannula for resection instruments.

Evidence for Safety
Reporting of evidence for the safety of available instruments is varied with nonstandardised outcomes, with 11 studies not reporting on safety at all (Supplementary  Table S1).
An additional 19 out of 62 studies only provide a general comment with no objective measures such as "according to our experience in more than 100 applications this bipolar instrument is safe" [16] or report comments on objective measurements such as a decrease in operative blood loss but do not report data [59].
The highest rate of intraoperative complications is 23%, reported for the BoneScalpel and Piezoelectric System in 13 patients with craniosynostosis undergoing endoscopeassisted surgery [18]. In this study the intraoperative complication rate is combined for the use of both instruments so it is not possible to accurately assess the complication rate of each individually. The intraoperative complication rate for the NeuroBalloon [43] also cannot be calculated as the authors report "more than 1000 cases" with no definitive number given.
The highest post-operative complication rate is 50% reported for the Micro ENP Ultrasonic Handpiece in two cases of endoscopic third ventriculostomy for intraventricular tumours [36]. However, it is unclear if this complication, ventriculitis, is directly related to the instrument or the operation given that ventriculitis is a known complication [73]. The only other instruments to report a postoperative complication rate over 10% were the 6.3 mm Percutaneous Endoscopic Instrument [13] utilised to treat migrated disk herniation in 22 adult patients, the 980 nm Diode Laser [14] utilised in nine adult patients with hydrocephalus, the OmniGuide CO2 laser [55] utilised in a case series of 16 adult patients with pituitary lesions that underwent surgery using the endoscopic endonasal transsphenoidal approach and the SONOCA Ultrasonic Aspirator [63] utilised in nine adult patients with supratentorial intraventricular tumours that underwent endoscopic resection.

Evidence for Efficacy
Reporting of the efficacy of available instruments is also varied and non-standardised (Supplementary Table S2). There is no reporting on efficacy in nine out of 62 studies and 33 only report a general comment such as "proven to be a valid and effective instrument" [41] with no outcome measures.
Mean operative time was reported in 10 studies, ranging from less than 10 min for the Artemis Neuro Evacuation Device for the treatment of a large suprasellar tumour with cavernous sinus invasion using the endoscopic endonasal approach [15] to 114 min for the Bipolar Microscissors, the mean operative time over 100 cases of intracranial astrocytoma treated with neuroendoscopic surgeries [17]. EOR, either gross, subtotal or partial was also reported in nine studies with GTR ranging from 33.33% for the SONOCA Ultrasonic Aspirator using endoscopic approaches for intraventricular tumours [62] to 100% for both the Self-Retaining Retractor used for the endoscopic endonasal transsphenoidal approach in pituitary adenomas [60] and the Sonopet Ultrasonic Bone Aspirator when operating on skull base tumours using the endoscopic endonasal approach [67].

Ergonomic Assessment
Overall, only 7 out of 62 studies report outcomes or comments relating to the ergonomics of the instrument (Supplementary Table S3). Comment on instrument ergonomics featured in six studies of six instruments: the Suction Device made of Shape Memory Alloy connected to the ATOM5 Record 55 DDS [68], the Malleable Endoscope Suction Instrument [34], the Monoshaft Bipolar Cautery [42], the Bipolar Microscissors [17], the Sonopet Ultrasonic Bone Aspirator [66] and the Handpiece, Keyhold and Needle-Type Probes and Probe Sheaths for use with the Ultrasonic Surgical Unit [31]. One study reported administering a seven-question survey in which surgeons rated the ergonomics of the final instrument design as "very good" [34].

Learning Curve Assessment
Comments on the learning curve of the available instruments were reported in eight out of 62 studies (Supplementary Table S4). The learning curve is reported to be improved or similar to standard instruments for the 2.0 µm Diode Pumped Solid State (DPSS) Laser, "steep and quick" [12], the Bipolar Microscissors, "no special training is needed" [17], NeuroBalloon, "use of the NeuroBalloon catheter in our experience may shorten this learning process" [43], the Sonopet Ultrasonic Bone Aspirator, "there did not appear to be a difference in learning curve" [64] and the ZESSYS, "could reduce the technique difficulties" [72].
A study of the NICO Myriad reports the learning curve as a disadvantage [46], however they note this is the case for all devices. Though the learning curve is commented on for the SONOCA Ultrasonic Bone Aspirator, the comment pertains to the difficulty of the surgical technique rather than the use of the instrument [63].

Comparison to Standard Instruments
Comparison to standard neurosurgical instruments was part of the study design of eight out of 62 studies for five instruments (Supplementary Table S4). Kawamata, Amano, and Hori [24] comment that the Flexible Forceps used during endoscopic endonasal approaches for pituitary tumours were "able to access sites where regular dedicated instruments... could not easily reach" facilitating improved manipulation of tissue in comparison to standard excision instruments, while Kawamata and colleagues [30] report that the Harmonic Scalpel for dissection of intracranial tumours demonstrates a "much lower temperature and much smaller area of elevated temperature than monopolar or bipolar diathermy" devices. Additionally, Nakamura et al. [44] comment that the "clinical and radiological results in the NAC [New Angled Chisel] group were better" in patients undergoing the procedure with the novel instrument in comparison to patients who underwent the same procedure without the NAC. These three studies do not report any objective outcomes.
The Bipolar Microscissors used to treat intracranial astrocytomas show reduced mean blood loss and operative time in comparison to standard instruments, though whether or not the difference is significant is not reported (361 mL vs. 278 mL and 145 min vs. 114 min) [17]. Mean operative time was improved when comparing a standard cylindrical tubular retractor to the Novel Rectangular Tubular Retractor [54], standard instruments to the Piezoelectric System [56], and TESSYS instrument to the novel ZESSYS [72]. The Modified Neuroendoscope Technology (MNT): a transparent sheath and haematoma smashing aspirator utilised for endoscopic evacuation of cerebral haemorrhages reduced mean ICU monitoring time (5 days vs. 16.2 days, p = 0.000), improved 6-month mortality rate (6.7% vs. 22.5%, p = 0.036) and improved 6-month Glasgow Outcome Scale result (3.7 vs. 2.9, p = 0.021) in comparison to procedures using an existing external drainage and monitoring system [38]. Mean operative time increased (86.6 min vs. 39.7 min, p = 0.000), though given the improved safety markers this does not appear to be a significant limitation. Similarly, the Sonopet Ultrasonic Bone Aspirator improved mean blood loss (16.55 mL vs. 22.58 mL, p = 0.000) and mean operative time (31.92 min vs. 41.33 min, p = 0.000) in comparison to standard Jansen-Middleton kerrison, thru-cut and backbiter instruments for endoscopically resected supratentorial intraventricular tumours [64].

Principal Findings
This systematic review identified 50 specialised instruments for endoscopic and endoscope-assisted neurosurgery across 62 studies. Notably, there was a steep rise in included papers published in the last decade.
The novelty of development in almost all instruments was in what they do, such as a new way to resect a skull base tumour, rather than in how they do it, i.e., the articulation of the instrument. The evidence for the safety and efficacy of these instruments was limited with non-standardised reporting across studies. Very few studies reported on usability assessments and in those that did all but one reported only subjective comments with no objective measures.
The development of specialised instruments for endoscopic and endoscope-assisted neurosurgery is a growing area fuelled by technological innovation. However, the innovation demonstrated in the identified studies within this review does not solve the established issues with standard instruments: namely, issues of triangulation [74] and articulation that would enhance flexibility and mobility of the instrument [75].

Safety
It is essential that the field finds a way to evaluate safety objectively in order to demonstrate that novel instruments have merit in their use and that they are an improvement on currently available instruments. This comparison is the "gold standard" in pharmacology and with adoption of the IDEAL-D and IDEAL frameworks [7,8] could become the gold standard for the development of surgical instruments in endoscopic neurosurgery. While this lack of comparison exists it is difficult to establish if the outcome measures reported for safety are a result of the instrument causing improvement or deterioration, or if this is a result of the pathology or surgical approach.

Comparison with Other Studies
Our review found that no studies used either an already established structured framework for any assessment of a novel instrument or devised a new one. While the development of such frameworks is a recent advance in improving the design and evaluation process for surgical instruments, over half of the studies identified were published after the design criteria for instruments for endoscopic neurosurgery had been reported [47,75] and just over a third of studies after publication of the IDEAL framework, developed specifically for neurosurgical instruments, in 2013 [7].
Our findings are in keeping with the wider literature on neurosurgical innovation. A recent systematic review of the use of the IDEAL framework in innovation for the endoscopic endonasal approach for skull base meningiomas identified 26 studies, none of which could be classified on the IDEAL framework, with some first-in-man studies being reported before pre-clinical studies of the same instrument [76]. However, there is some data to suggest that the IDEAL and similar frameworks are being adopted, albeit slowly, within the neurosurgical community. In a recent bibliometric analysis Ota et al. found 51 studies that cited IDEAL with the overwhelming majority being positive [77].

Strengths and Limitations
The most significant limitation of this review is that the nature of the literature required the pooling of results of hugely disparate outcomes. This invites bias in the interpretation and conclusions drawn, particularly as it was not possible to conduct comparative statistical tests of quantitative measures such as mean blood loss or operative duration due to the small sample size of studies that reported the same objective measures. However, this review was systematic and exhaustive. The search of the literature was broad, with no limiters placed on the databases, in order to ensure all the relevant literature was considered for inclusion based upon pre-established and clear criteria.

Conclusions
While there are many specialised instruments for endoscopic and endoscope-assisted neurosurgery available, the reporting of their safety and efficacy is limited. Specialised instruments are infrequently compared to standard instruments, meaning accurate conclusions to establish if they provide improved, cost-effective, surgical options for patients cannot be drawn. The usability of these instruments is also not regularly or objectively assessed, raising concerns about the adoption of new technologies and the effect of these measures on safety and efficacy.
This review shows that novel specialised instruments for endoscopic and endoscopeassisted neurosurgery are an area of interest for the field with many devices being developed in the past few years to overcome procedural issues with standard instruments that make them sub-optimal for these approaches. Despite the increase in new instruments, almost all of the identified devices are novel in what they do rather than in how they do it, mostly innovating the end effectors of the instrument but not the unmet need for improved articulation.
Available instruments define the applications and limitations of endoscopic and endoscope-assisted neurosurgery. Given that the identifed studies are inconsistent in their reporting of safety, efficacy and usability future research should seek to establish guidelines for surgical innovation, developing the IDEAL and IDEAL-D frameworks towards a more practical criteria that can be applied successfully in future studies of specialised instruments.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/cancers14122931/s1. Table S1: Summary of safety reporting in identified studies of available instruments for endoscopic or endoscope-assisted neurosurgery. Table S2: Summary of operative outcome reporting in identified studies of available instruments for endoscopic or endoscope-assisted neurosurgery. Table S3: Summary of reporting of ergonomic and learning curve assessment in identified studies of available instruments for endoscopic or endoscope-assisted neurosurgery. Table S4: Summary of studies reporting comparison to standard instruments.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.