Force measurement metrics for simulated elbow arthroscopy training

Force measurement metrics for simulated elbow arthroscopy training Background: Elbow arthroscopy is a difficult surgical technique. Objective metrics can be used to improve safe and effective training in elbow arthroscopy. Force exerted on the elbow tissue during arthroscopy can be a measure of safe tissue manipulation. The purpose of this study was to determine the force magnitude and force direction used by experts during arthroscopic elbow navigation in cadaveric specimens and assess their applicability in elbow arthroscopy training. Methods: Two cadaveric elbows were mounted on a Force Measurement Table (FMT) that allowed 3-dimensional measurements (x-, y-, and z-plane) of the forces exerted on the elbow. Five experts in elbow arthroscopy performed arthroscopic navigation once in each of two cadaveric elbows, navigating through the posterior, posterolateral and anterior compartment in a standardized fashion with visualization of three to four anatomic landmarks per compartment. Thetotal absoluteforce(F ) and force direction exerted (α and β) on the elbow during arthroscopy were recorded. α abs being the angle in the horizontal plane and β being the angle in the vertical plane. The 10th–90th percentiles of the data were used to set threshold levels for training. Results: The median F was24N(19N – 30 N), 27N (20N – 33 N) and29N(23N – 32 N) for the posterior, abs posterolateral and anterior compartment, respectively. The median α was - 29° (- 55° – 5°), - 23° (- 56° – -1°) and 4° (- 22° – -18°) for the posterior, posterolateral and anterior compartment, respectively. The median β was - 71° (- 80° – -65°), - 76° (- 86° – -69°) and - 75° (- 81° – -71°) for the posterior, posterolateral and anterior compartment, respectively. Conclusion: Expert data on force magnitude and force direction exerted on the elbow during arthroscopic navigation in cadaveric specimens were collected. The proposed maximum allowable force of 30 N (smallest 90th percentile of F ) abs exerted on the elbow tissue, and the 10th–90th percentile range of the force directions (α and β) for each compartment may be used to provide objective feedback during arthroscopic skills training. Keywords: Elbow, Arthroscopy, Navigational forces, Experts, Skills assessment, Education, Cadaver Background Arthroscopy requires excellent visual spatial awareness to Over the past decades elbow arthroscopy has become a sur- mentally recreate a 3-dimenionsal environment from gical tool due to better understanding of the neurovascular 2-dimensional images. This cannot be learned by assisting anatomy, technical advancements, and broadening range of and observing in the operating theatre alone (Aggarwal et indications (Hilgersom et al., 2018;Yeohetal., 2012). An in- al., 2004; Aim et al., 2016; Rosenthal et al., 2006;Tashiro et crease in elbow arthroscopy use is expected to raise the al., 2009). Moreover, elbow arthroscopy specifically is tech- number of complications, which emphasizes the importance nically challenging due to limited working space and close of training in portal placement and arthroscopic skills to de- proximity of neurovascular structures (Hilgersom et al., liver safe surgical care (Rose & Pedowitz, 2015). 2017;Marshalletal., 1993; Miller et al., 1995;Omidet al., 2012; Stothers et al., 1995). Further distinguishing elbow * Correspondence: n.f.hilgersom@amc.uva.nl arthroscopy is the need for mirrored hand-eye coordination Department of Orthopaedic Surgery, Amsterdam University Medical Centres, in the lateral decubitus position when compared to most University of Amsterdam, Amsterdam Movement Sciences, Meibergdreef 9, other arthroscopic modalities; and overhand versus 1105 AZ Amsterdam, the Netherlands Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 2 of 9 underhand holding of instruments. All above, in combin- (90th percentile) to cover for the effects of other condi- ation with the lower frequency compared to knee or shoul- tions when setting the metrics’ threshold. der arthroscopy, makes it apparent that elbow arthroscopy has a longer learning curve in time. Cadaveric specimen Currently, no consensus exists on the minimal number of Two fresh-frozen right-handed upper limb cadaver spec- elbow arthroscopies that must be performed to become an imens without evidence of previous trauma, surgery or expert. Savoie states that a minimal number of 100 per- deformity were prepared to mimic an arthroscopic formed elbow arthroscopies is necessary (Savoie 3rd, 2007). setting. These specimens were derived from bodies that Furthermore, Claessen et al. (Claessen et al., 2017)observed entered the department of anatomy through a donation a 30% complication rate in portal placement by novice sur- program. From these persons written consent was geons, which was significantly higher compared to experi- obtained during life that allowed the use of their entire enced elbow arthroscopists (Elfeddali et al., 2013; Marti et bodies for educational and research purposes. Specimens al., 2013). These numbers make clear that elbow arthros- were stored at − 20 °C and thawed 24 h before use. The copy (simulated) training is essential (Claessen et al., 2017; upper limb cadaveric specimens arms were dissected Rose & Pedowitz, 2015). transversely 15–20 cm proximal of the humeral epicon- Cadaveric training is still the preferred training method dyles and mounted onto the custom-made static arm to improve arthroscopic skills because it provides the most holder of the force measurement table (FMT) with the realistic setting (Camp et al., 2016;Hui et al., 2013;Koehler posterior humerus facing superiorly and the humeral et al., 2015). Objective performance measurement by using epicondyles orientated horizontally, mimicking a lateral metrics is preferred over global rating scales such as Ob- decubitus position (Horeman et al., 2016). jective Structured Assessment of Technical Skills (OSATS) (Horeman et al., 2016;Martin etal., 1997; van Hove et al., Force measurement table 2010). Such metrics have yet to be defined in elbow For the interested readers, the force measurement table is arthroscopy, but have been defined in knee and shoulder described in detail by Horeman et al. (Horeman et al., arthroscopy, for example to differentiate between levels of 2016). In short, when a cadaveric specimen is firmly fixated experience and to set thresholds for safe tissue manipula- in the vice of the FMT, it measures the forces in x-, y-, tion (Stunt et al., 2014;Tashiro et al., 2009;Tuijthofetal., z-direction during arthroscopic skills training, enabling 2011). Recently, Obdeijn et al. (Obdeijn et al., 2016)defined objective performance tracking of the trainees. The FMT a maximum allowable force magnitude of 7.3 N (90th per- consists of three squared frames, each connected to one centile) using expert data derived thresholds and demon- another by four beams that bend upon loading (Fig. 1). The strated that force direction is equally important as force three frames displace independently; each in a single direc- magnitude for safe wrist arthroscopy to prevent cartilage tion (i.e. x-, y-, or z-direction) (Fig. 1). Theapplied forceon damage. Similarly, forces exerted on the elbow by experts each frame is calculated by measuring the relative displace- during elbow arthroscopy may also be valuable indicators ment of the four bending beams and multiplication with of a safe elbow arthroscopy. the bending beams’ known stiffness (Fig. 1). Bending beam The purpose of this study was to determine the force displacements were measured using Linear Hall effect magnitude and force direction used by experts during sensors and Neodymium disk magnet built into the bend- arthroscopic elbow navigation in cadaveric specimens ing beams (Horeman et al., 2016). The FMT allowed con- and assess their applicability in elbow arthroscopy tinuous recording of the forces exerted on the cadaver training. elbow by the instruments in a range of 0 N to 750 N in three loading directions, with an accuracy of 0.1 N and a Methods sample frequency of 24 Hz (Horeman et al., 2016). To pos- The study was designed to fit within the set time sched- ition the elbow above the FMT, a custom-made stand with ule of the two day-26th annual international Arthros- vice was mounted on the FMT. The vice allows fast mount- copy & Arthroplasty Courses Utrecht. This implied that ing of the prepared humerus bone in a 45-degree angle to we could perform data acquisition with five experts mimic the actual procedure (Fig. 1). operating on two cadaveric specimens. This approach A camera tracking system using two digital video cam- was suitable to meet the study goal, because a similar eras set up on both sides of the operator and the arthro- strategy was followed for assessing a threshold naviga- scope camera was set up for monitoring of instrument tion force for wrist arthroscopy (Obdeijn et al., 2016): a) use, capture ‘occurrences’ (e.g.; probing of the predefined focus on experts and recruit as many as possible to landmarks) and adequate postprocessing of the data determine if their navigation force variation is acceptably acquired with the FMT. small to set a safety threshold; b) keep other conditions Qualitative analyses of the individual contribution of as constant as possible; and c) propose a safety margin the arthroscope and probe on the total forces exerted on Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 3 of 9 Fig. 1 Force Measurement Table. The FMT with custom-made stand with vice attached is shown. The FMT consists of the three squared frames and bending beams with Hall effect sensors and magnets in the x-, y- and z-plane of the FMT. The design of the FMT allows continuous recordings of the forces exerted on the cadaver specimen attached to the vice in three loading directions the cadaver elbow was performed by combining the data portals. During the arthroscopic navigation experts from the FMT and camera tracking system. consecutively visualized the posterior, posterolateral and anterior compartment and were asked to determine the Experts predefined landmarks (Fig. 2). In the posterior compart- The expert group consisted of five upper limb surgeons ment the landmarks were the olecranon tip, olecranon specialized in elbow arthroscopy and instructors at the fossa, medial gutter, and lateral gutter (Fig. 2a). In the 26th annual international Arthroscopy & Arthroplasty posterolateral compartment the landmarks were the Courses Utrecht. The experts filled out a questionnaire radial head, capitellum, and proximal radioulnar joint to document their demographic data (Table 1). Prior to the experiment, expert one created the follow- Table 1 Demographic data and experience of the five participants ing arthroscopic portals in both cadaveric specimens; proximal anteromedial, proximal anterolateral, midtrici- Expert 1 2 3 4 5 pital, posterolateral and soft spot portal, and as routinely Age (years) 42 38 44 50 48 is performed with elbow arthroscopy, shaved fibrous Gender Male Male Male Female Female tissue blocking the view. The midtricipital, posterolateral Dexterity Right Right Left Right Right and proximal anteromedial portal served as viewing Expertise Expert Expert Expert Expert Expert portals for the posterior, posterolateral and anterior Exp EA (years) 2 4 8 16 15 compartment, respectively. NR EA (year) 100 100 20–25 100 10–15 Each expert performed an arthroscopic navigation once on each cadaveric elbow using the above-described Exp Experience, EA Elbow arthroscopy,NR Number Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 4 of 9 Fig. 2 Arthroscopic views of predefined landmarks per compartment. a Posterior compartment. b Posterolateral compartment. c Anterior compartment (Fig. 2b). In the anterior compartment the landmarks the total absolute force (F ) per sample was calcu- abs were the radial head, capitellum, coronoid tip, and lated by summation of the force measurements in the coronoid fossa (Fig. 2c). Each landmark had to be x-, y-, and z-plane after the force measurement in touched by the probe and visualised in the centre of z-direction was compensated for the mass of the specimen thearthroscopicimage.Once a landmark wasvisual- and holder. In addition, the force direction in the horizon- ized per protocol, as visually verified by one of the tal plane (α) could be derived from the force magnitude in researchers, the expert could proceed to the next ana- the x- and y-plane, and the force direction in the vertical tomic landmark. Arthroscopic elbow navigation was plane (β), which is aligned with the humerus mounted on performed in the same consecutive order of experts the set-up, from the force magnitude in the x- and on both elbow specimens. All measurements were z-plane. A positive α-angle implies a direction of force to performed on the same day. During the experiment the lateral side and a negative α-angle implies direction of experts could extend the elbow as they felt necessary force to the medial side. A positive β-angle implies upward for proper portal placement and instrument use. All direction of force and a negative β-angle implies down- arthroscopic tasks were performed using an arthro- ward direction of force. scopic probe and a 30°-angle 4 mm arthroscope from Karl Storz (Tuttlingen, Germany). Statistical analysis The experts were asked to perform the tasks as they The presence of normal distributions for F , α and β was abs would be performing live surgery on an actual patient. determined with the Kolmogorov-Smirnov test per com- partment. As the data were not normally distributed, F , abs Data processing α and β were expressed in terms of median (10th–90th The data gathered with the FMT and camera tracking percentile). A Mann-Whitney U-test was performed to system were processed using Matlab (version R2014a, compare the F measurements for the anterior compart- abs The Mathworks, Natick, MA, USA) and IBM SPSS ment between the two cadaveric specimens (p < 0.05). statistics (version 22, SPSS, Chicago, IL, USA). All Prior to this study, Obdeijn et al. (Obdeijn et al., 2016) raw voltage data were filtered with a low-pass Butter- successfully applied the 10th and 90th percentiles to set worth filter with a cut-off frequency of 24 Hz to sup- thresholds for safe tissue manipulation and force direc- press high-frequency noise. For each compartment tion in wrist arthroscopy. Therefore, we used a similar Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 5 of 9 strategy in this study: the 10th and 90th percentiles of force (Fig. 4a). The Mann-Whitney U-test indicated a F α and β were used to set threshold levels for safe significant difference between the values of F for the abs, abs tissue manipulation and force direction that can be used anterior compartment between the two cadaveric speci- during elbow arthroscopy training. mens (p <0.05). Results Horizontal angle (α) and vertical angle (β) Figure 3 shows an example of the force measurement in The median α, force direction in the horizontal plane, time of one navigation task performed by one expert in is −29° forthe posteriorcompartment with arange the posterior compartment. A qualitative initial analysis of 60°, is − 23° for the posterolateral compartment combining the force data and video footage showed with a range of 55° and is 4° for the anterior com- force fluctuations in a similar direction during probing partment with a range of 40° (Fig. 4b). Notable is the of a landmark, force fluctuations in an opposite direction more medial direction and smaller range of α in the during elbow flexion, and only marginal variation in anterior compartment compared to the posterior and forces during instrument changes when only the arthro- posterolateral compartment (Figs.4b and 6b). The me- scope was in place (Fig. 3). dian β, force direction in the vertical plane, is − 71° The histograms of F , α and β for the posterior, pos- for the posterior compartment with a range of 15°, − abs terolateral and anterior compartment of both cadaveric 76° for the posterolateral compartment with a range elbows are presented in Fig. 4. of 17° and − 75° for the anterior compartment with a range of 10° (Fig. 4c). The median β remains fairly Total absolute of force F constant with a maximum difference of 5° and max- abs The median F is similar for each compartment, imum range of 17° (Fig. 6c). Figure 5 provides a sche- abs being 24 N (range 19 N – 30 N) for the posterior matic representation of the median β for all compartment, 27 N (20 N – 33 N) for the posterolateral compartments combined. compartment, and 29 N (23 N – 32 N) for the anterior Comparison of the force direction between the differ- compartment (Fig. 4a). In the anterior compartment, two ent compartments showed a second smaller peak around peaks of F are observed in the histogram, one around − 90 degrees for α and β in the posterior and posterolat- abs 23 N of absolute force and one around 30 N of absolute eral compartment (Fig. 4b and c). Expert 1, who created Fig. 3 Example of force measurement in time of one navigation task performed by one expert. This example shows force measurement in time of one navigation task performed by one expert in the posterior compartment. In the upper graph the individual force components as well as the overall combined force F are illustrated. In the lower graph the stars indicate the moments of touching and displaying the assigned abs landmark. In this 2D representation, the first Area (A) represents a location were elbow flexing occurs indicated by an oppositely directed change in Force. The following area’s (B) represent instrument bone/tissue interaction with force fluctuations in similar direction Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 6 of 9 Fig. 4 Histograms presenting the data points for F , alfa and beta of each compartment. Histograms presenting the median, 10th and 90th abs percentile for frequency of total absolute force (F ), horizontal angle (α) and vertical angle (β) data points in the posterior, posterolateral and abs anterior compartment of both cadaver elbows. a Histogram showing F . The two black arrows point to two separate peaks in frequency of F , abs abs around 23 N and 30 N in the anterior compartment. b Histogram showing α. α is positive to the right, and negative to the left. The two black arrows point to a smaller peak around − 90° for α and β in the posterior and posterolateral compartment. c Histogram showing β. β is positive upward, and negative downward. The two black arrows point to a smaller peak around − 90° for α and β in the posterior and posterolateral compartment the portals, had a substantial share in this peak, particu- found for knee joint distraction (mean F of 43-50 N) abs larly when performing the task in the first cadaveric (Stunt et al., 2014). specimen. A possible contributing factor to the overall higher force load is that manoeuvring the arthroscope to a compartment Safe zone – Metric threshold is performed primarily by knowing the correct orientation Finally, a graphical interpretation is given in Fig. 6 of the of the arthroscope and by haptic feedback using the bony median values of the F , α and β for each of the three structures for guidance, such as sliding along the anterior abs compartments (posterior, posterolateral and anterior) as face of the humerus to create the proximal anteromedial well as the set safe zone using the 10th and 90th percentile portal, or using bony structures as a support point/wedge force values from Fig. 4. The 90th percentile values indi- to take a corner while navigating around the elbow (Fig. 5). cate the set maximum threshold for the metric. In addition to a lack of joint distraction, this relative high bone-instrument loading may cause a higher overall loading Discussion on the elbow. The consequences of the relative high force This study shows that median loads of 24-29 N are may be limited, because surgical procedures during elbow exerted on the elbow by experts during arthroscopic arthroscopy are primarily performed outside of the articu- navigation in a cadaveric elbow. These loads represent lating surfaces of the elbow joint (i,e. synovectomy, capsular the combined forces exerted by the arthroscope and the release, loose body removal). This reduces the chance of in- probe on the anatomic structures of the elbow. The jury to delicate tissues inside the joint such as the poorly overall measured forces are considerably higher than ex- healing articular cartilage. pert force data for wrist arthroscopy (median F of The arthroscope assembly (e.g. arthroscope, cables, abs 3.8 N) (Obdeijn et al., 2016) and probing of meniscal tis- camera) and supporting hand plus arm most likely have sue in the knee (mean F ranging 2.8–3.9 N) (Tuijthof the highest contribution in the total force as combined abs et al., 2011), but they are lower than expert force data analysis of video footage and force data with the aim to Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 7 of 9 The median force direction and range during arthro- scopic navigation in the elbow is similar for all compart- ments in the vertical plane (β) (Figs. 4c, 5 and 6c). The median force direction of the anterior compartment in the horizontal plane (α) is more medial compared to the pos- terior and posterolateral compartment, and the range of force direction is smaller when compared to the posterior and posterolateral compartment (Fig. 4b and 6b). These findings can be related to working through the proximal anteromedial portal, the anatomical location of the anter- ior compartment and the anatomical distance between the landmarks in the anterior compartment, respectively. The second smaller peak in force direction in both planes (α and β) observed around − 90° in the posterior and posterolateral compartment (Fig. 4b and c) seems attributable to suboptimal portal placement as expert 1 who created the portals had a substantial share in this second peak, particularly in the first specimen. This is supported by the lower median F used by expert 1 in abs cadaver 1. Expert 1 created the portals and as such knew the exact orientation of the portals resulting in a lower median F compared to the other experts. abs Elbow arthroscopy, when performed with the patient in a Fig. 5 Median force direction in the vertical plane (β) for all lateral decubitus position, requires a mirrored way of in- compartments combined. Schematic representation of the median strument handling with a 30° arthroscope when compared direction of Fabs in the vertical plane for all compartments to performing arthroscopy of most other joints. The force combined represented by β. The blue lines represent the lowest and direction in the vertical plane (β)shows minor variation highest values of the 10th and 90th percentiles of β. An elbow x-ray has been superimposed over the cadaveric elbow to further clarify (Figs. 5 and 6c), which is a sign that this range may be used the correlation between the overall median force direction and for novice surgeons to strive for. The latter is strengthened elbow joint by Obdeijn et al. (Obdeijn et al., 2016; Obdeijn et al., 2014) who showed that force direction is equally as important as correlate force direction and variation in force magnitude force magnitude, and found that novices showed consider- to instrument use showed hardly any variation in forces able variation in loading direction compared to experts during probing of landmarks or instrument changes with when performing wrist arthroscopy. The force direction the arthroscope in place (Figs. 3 and 5). This is a possible area defined by the 10th–90th percentile of expert thresh- assumption as the FMT measured the total combined olds for α and β (Figs. 5 and 6) may be used to adjust the forces exerted on the cadaveric elbow and is unable to direction of the arthroscope to properly navigate through quantitatively assess the individual contribution of the the complex elbow anatomy. To be of assistance for the arthroscope or probe used during elbow arthroscopy. trainees, it is necessary to visualize the direction of force on Based on expert data, the 10th and 90th percentiles of the video screen via augmented reality. Implementing this the exerted force have been used to determine force in a meaningful way is a challenging task, as is shown by thresholds in wrist arthroscopy and probing of menisci theworkofSmitetal. (Smitetal., 2017). (Obdeijn et al., 2016; Tuijthof et al., 2011). Utilizing the There are limitations to this study. First, although the same strategy on current expert data, we propose a max- number of data points per surgeon was high, the num- imum allowable force load of 30 N to be exerted on the ber of experts and cadavers was small, but feasible elbow during arthroscopic navigation, which is the smal- within the set time frame of the advanced elbow course. lest value of 90th percentiles of the force magnitude of Besides the variation amongst the experts, other condi- all three compartments (Fig. 4a). This threshold level tions (cadavers, the joint status in time and portal place- should be demonstrated in elbow arthroscopy training ment) do effect the forces. Since our aim was not to to let novices experience the feel of the magnitude of a assess the individual contribution of each condition, but load around 30 N, as this is a most likely a lot higher rather set an overall safety threshold, we argue that the than novices expect (Obdeijn et al., 2014; Tuijthof et al., small group of surgeons conducting the trials on two ca- 2011). This can help students to train their haptic senses davers should represent the entire group of expert elbow in a safe way by preventing them to use higher loads. arthroscopists sufficiently. This is supported by the Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 8 of 9 Fig. 6 Safe force zone (magnitude and direction) for all compartments. The red dots indicate the median value of F , α and β for each abs compartment. The thick black lines with white dots at their respective ends represent the 10th and 90th percentile value of F in the median abs force direction. The grey boxes surrounding the thick black lines indicate the combined boundaries of the 10th -90th percentiles for F , α and β. abs The origin is taken at the same position for each compartment. a 3D graphical representation. For reference an elbow in the lateral decubitus position is added. So the forces are directed towards the surgeon, b Top view, c Sagittal view narrow range of the 10th–90th percentile of the median and second cadaver elbow, respectively. Consequently, values of F , α and β. Second, the data was collected threshold levels as determined in this study should be ad- abs from cadaveric specimens that are usually stiffer than el- justed per cadaver and training time on the cadaver (swell- bows from live patients. Therefore, one can reason that ing due to irrigation). Therefore, we recommend starting higher forces will be observed when performing arthros- training basic elbow arthroscopic skills on a simulator. This copy on living patients. However, this may be partly will provide the same standard for all trainees at any time, compensated as cadaveric specimen are commonly ob- and allows adequate objective feedback by setting one tained from elderly people with usually lesser tissue threshold value and facilitates observation of training pro- quality than young people. gress of participants compared to their peers. After obtain- Although cadaveric training provides the most realistic ing proficiency in basic arthroscopic skills on a simulator, a experience, cadaveric training is not the preferred method trainee may advance to cadaveric skills training to become to start training elbow arthroscopy skills. First, because ca- acquainted with the feeling and effect of the loads on hu- daveric training is expensive and there is limited availability man tissues along with learning to adapt to anatomic varia- (Camp et al., 2016; Stirling et al., 2014). Moreover, as was tionsasisthe case in livesurgery. also shown in this study, the anatomic variation amongst Nonetheless, this study shows that force data can be cadaveric specimen as well as their joint status in due to accurately and reliably recorded in three loading direc- time compromises similar training conditions for a certain tions using the FMT (Horeman et al., 2016), allowing amount of repetitions or trainees. For example, in the expert thresholds to be defined for force magnitude and present study two peaks of F were observed during navi- force directions that can be used for objective feedback abs gation of the anterior compartment, around 23 N and 30 N during elbow arthroscopy training. (Fig. 4a), which were attributable to the use of two cadaver elbows (Mann-Whitney U, p < 0.05). In addition, due to Conclusions continuous water irrigation of the elbow for a long duration Expert data on force magnitude and force direction (five elbow arthroscopies) the soft tissues would swell, pos- exerted on the elbow during arthroscopic navigation in sibly making portal insertion, gaining orientation and work- cadaveric specimens was collected. The proposed max- ing inside the joint more difficult. In this study, this was imum allowable force of 30 N (smallest 90th percentile observed as moderate differences in the median F of of F ) exerted on the elbow tissue, and the 10th–90th abs abs 8.2 N and 3.5 N between the first and last expert in the first percentile range of the force directions (α and β) for Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 9 of 9 each compartment may be used to provide objective Claessen F, Kachooei AR, Kolovich GP, Buijze GA, Oh LS, van den Bekerom MPJ, Doornberg JN (2017) Portal placement in elbow arthroscopy by novice feedback during arthroscopic skills training. surgeons: cadaver study. Knee Surg Sports Traumatol Arthrosc 25:2247–2254 Elfeddali R, Schreuder MH, Eygendaal D (2013) Arthroscopic elbow surgery, is it Abbreviations safe? J Shoulder Elb Surg 22:647–652 F : Total absolute force; FMT: Force Measurement Table; OSATS: Objective abs Hilgersom NF, Oh LS, Flipsen M, Eygendaal D, van den Bekerom MP (2017) Tips Structured Assessment of Technical Skills to avoid nerve injury in elbow arthroscopy. World J Orthop 8:99–106 Hilgersom NFJ, Molenaars RJ, van den Bekerom MPJ, Eygendaal D, Doornberg JN Acknowledgements (2018) Review of Poehling et al (1989) on elbow arthroscopy: a new We are grateful to the Elbow Study Collaborative for their assistance technique. J ISAKOS 3:116–124 during the elbow course, and participating as elbow experts in our study. Horeman T, Tuijthof GJM, Wulms PB, Kerkhoffs GMMJ, Gerards RM, Karahan M (2016) Elbow Study Collaborative Bertram The, Carina L.E. Gerritsma, Lex A force measurement system for training of arthroscopic tissue manipulation Boerboom, Tom Roeling, Marco van der Pluijm, Michel P.J. van den Bekerom, skills on cadaveric specimen. J Med Devices 10:044508–044501/044507 and Denise Eygendaal. Hui Y, Safir O, Dubrowski A, Carnahan H (2013) What skills should simulation training in arthroscopy teach residents? A focus on resident input. Int J Availability of data and materials Comput Assist Radiol Surg 8:945–953 The datasets used and/or analysed during the current study are available Koehler R, John T, Lawler J, Moorman C 3rd, Nicandri G (2015) Arthroscopic from the corresponding author on reasonable request. training resources in orthopedic resident education. J Knee Surg 28:67–74 Marshall PD, Fairclough JA, Johnson SR, Evans EJ (1993) Avoiding nerve damage during elbow arthroscopy. J Bone Joint Surg Br 75:129–131 Authors’ contributions Marti D, Spross C, Jost B (2013) The first 100 elbow arthroscopies of one surgeon: NFJH was responsible for data interpretation, manuscript preparation, analysis of complications. J Shoulder Elb Surg 22:567–573 manuscript design, creation of figures and editing. GJMT and TH were Martin JA, Regehr G, Reznick R, MacRae H, Murnaghan J, Hutchison C, Brown M responsible for the study design, data collection, data analysis, data (1997) Objective structured assessment of technical skill (OSATS) for surgical interpretation, and providing feedback on the manuscript. RLWAB was residents. Br J Surg 84:273–278 responsible for the availability of cadaveric specimens, study design and Miller CD, Jobe CM, Wright MH (1995) Neuroanatomy in elbow arthroscopy. J providing feedback on the manuscript. DE and MPJB were responsible for Shoulder Elb Surg 4:168–174 data interpretation, clinical insight and providing feedback on the Obdeijn MC, van Baalen SJ, Horeman T, Liverneaux P, Tuijthof GJ (2014) The use manuscript. All authors read and approved the final manuscript. of navigation forces for assessment of wrist arthroscopy skills level. J Wrist Surg 3:132–138 Ethics approval Obdeijn MC, Horeman T, de Boer LL, van Baalen SJ, Liverneaux P, Tuijthof GJ The cadaveric specimens used in this study were derived from bodies that (2016) Navigation forces during wrist arthroscopy: assessment of expert entered the department of anatomy, University of Utrecht, through a donation levels. Knee Surg Sports Traumatol Arthrosc 24:3684–3692 program. From these persons written consent was obtained during life that Omid R, Hamid N, Keener JD, Galatz LM, Yamaguchi K (2012) Relation of the allowed the use of their entire bodies for educational and research purposes. radial nerve to the anterior capsule of the elbow: anatomy with correlation to arthroscopy. Arthroscopy 28:1800–1804 Consent for publication Rose K, Pedowitz R (2015) Fundamental arthroscopic skill differentiation with Not applicable. virtual reality simulation. Arthroscopy 31:299–305 Rosenthal R, Gantert WA, Scheidegger D, Oertli D (2006) Can skills assessment on Competing interests a virtual reality trainer predict a surgical trainee's talent in laparoscopic The authors declare that they have no competing interests. surgery? Surg Endosc 20:1286–1290 Savoie FH 3rd (2007) Guidelines to becoming an expert elbow arthroscopist. Arthroscopy 23:1237–1240 Publisher’sNote Smit D, Spruit E, Dankelman J, Tuijthof G, Hamming J, Horeman T (2017) Springer Nature remains neutral with regard to jurisdictional claims in Improving training of laparoscopic tissue manipulation skills using various published maps and institutional affiliations. visual force feedback types. Surg Endosc 31:299–308 Stirling ER, Lewis TL, Ferran NA (2014) Surgical skills simulation in trauma and Author details orthopaedic training. J Orthop Surg Res 9:126 Department of Orthopaedic Surgery, Amsterdam University Medical Centres, Stothers K, Day B, Regan WR (1995) Arthroscopy of the elbow: anatomy, portal University of Amsterdam, Amsterdam Movement Sciences, Meibergdreef 9, sites, and a description of the proximal lateral portal. Arthroscopy 11:449–457 1105 AZ Amsterdam, the Netherlands. Department of Biomechanical Stunt JJ, Wulms PH, Kerkhoffs GM, Sierevelt IN, Schafroth MU, Tuijthof GJ (2014) Engineering, Delft University of Technology, Mekelweg 2, 2628 CD Delft, the Variation in joint stressing magnitudes during knee arthroscopy. Knee Surg Netherlands. Department of Human Movement Sciences, Vrije Universiteit Sports Traumatol Arthrosc 22:1529–1535 Amsterdam, De Boelelaan 1117, 1081 HV Amsterdam, the Netherlands. Tashiro Y, Miura H, Nakanishi Y, Okazaki K, Iwamoto Y (2009) Evaluation of skills in Department of Anatomy, University Medical Centre Utrecht, Heidelberglaan arthroscopic training based on trajectory and force data. Clin Orthop Relat 100, 3584 CX Utrecht, the Netherlands. Department of Orthopaedic Surgery, Res 467:546–552 Amphia Hospital, Molengracht 21, 4818 CK Breda, the Netherlands. Tuijthof GJ, Horeman T, Schafroth MU, Blankevoort L, Kerkhoffs GM (2011) Department of Orthopaedic Surgery, Onze Lieve Vrouwe Gasthuis, Probing forces of menisci: what levels are safe for arthroscopic surgery. Knee Oosterpark 9, 1091 AC Amsterdam, the Netherlands. Zuyd University of Surg Sports Traumatol Arthrosc 19:248–254 Applied Science, Nieuw Eyckholt 300, 6419 DJ Heerlen, the Netherlands. van Hove PD, Tuijthof GJ, Verdaasdonk EG, Stassen LP, Dankelman J (2010) Objective assessment of technical surgical skills. Br J Surg 97:972–987 Received: 16 May 2018 Accepted: 20 September 2018 Yeoh KM, King GJ, Faber KJ, Glazebrook MA, Athwal GS (2012) Evidence-based indications for elbow arthroscopy. Arthroscopy 28:272–282 References Aggarwal R, Moorthy K, Darzi A (2004) Laparoscopic skills training and assessment. Br J Surg 91:1549–1558 Aim F, Lonjon G, Hannouche D, Nizard R (2016) Effectiveness of virtual reality training in Orthopaedic surgery. Arthroscopy 32:224–232 Camp CL, Krych AJ, Stuart MJ, Regnier TD, Mills KM, Turner NS (2016) Improving resident performance in knee arthroscopy: a prospective value assessment of simulators and cadaveric skills laboratories. J Bone Joint Surg Am 98:220–225 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Experimental Orthopaedics Springer Journals

Force measurement metrics for simulated elbow arthroscopy training

Free
9 pages

Loading next page...
 
/lp/springer-journals/force-measurement-metrics-for-simulated-elbow-arthroscopy-training-mnCeS0oiNb
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Orthopedics
eISSN
2197-1153
D.O.I.
10.1186/s40634-018-0157-1
Publisher site
See Article on Publisher Site

Abstract

Background: Elbow arthroscopy is a difficult surgical technique. Objective metrics can be used to improve safe and effective training in elbow arthroscopy. Force exerted on the elbow tissue during arthroscopy can be a measure of safe tissue manipulation. The purpose of this study was to determine the force magnitude and force direction used by experts during arthroscopic elbow navigation in cadaveric specimens and assess their applicability in elbow arthroscopy training. Methods: Two cadaveric elbows were mounted on a Force Measurement Table (FMT) that allowed 3-dimensional measurements (x-, y-, and z-plane) of the forces exerted on the elbow. Five experts in elbow arthroscopy performed arthroscopic navigation once in each of two cadaveric elbows, navigating through the posterior, posterolateral and anterior compartment in a standardized fashion with visualization of three to four anatomic landmarks per compartment. Thetotal absoluteforce(F ) and force direction exerted (α and β) on the elbow during arthroscopy were recorded. α abs being the angle in the horizontal plane and β being the angle in the vertical plane. The 10th–90th percentiles of the data were used to set threshold levels for training. Results: The median F was24N(19N – 30 N), 27N (20N – 33 N) and29N(23N – 32 N) for the posterior, abs posterolateral and anterior compartment, respectively. The median α was - 29° (- 55° – 5°), - 23° (- 56° – -1°) and 4° (- 22° – -18°) for the posterior, posterolateral and anterior compartment, respectively. The median β was - 71° (- 80° – -65°), - 76° (- 86° – -69°) and - 75° (- 81° – -71°) for the posterior, posterolateral and anterior compartment, respectively. Conclusion: Expert data on force magnitude and force direction exerted on the elbow during arthroscopic navigation in cadaveric specimens were collected. The proposed maximum allowable force of 30 N (smallest 90th percentile of F ) abs exerted on the elbow tissue, and the 10th–90th percentile range of the force directions (α and β) for each compartment may be used to provide objective feedback during arthroscopic skills training. Keywords: Elbow, Arthroscopy, Navigational forces, Experts, Skills assessment, Education, Cadaver Background Arthroscopy requires excellent visual spatial awareness to Over the past decades elbow arthroscopy has become a sur- mentally recreate a 3-dimenionsal environment from gical tool due to better understanding of the neurovascular 2-dimensional images. This cannot be learned by assisting anatomy, technical advancements, and broadening range of and observing in the operating theatre alone (Aggarwal et indications (Hilgersom et al., 2018;Yeohetal., 2012). An in- al., 2004; Aim et al., 2016; Rosenthal et al., 2006;Tashiro et crease in elbow arthroscopy use is expected to raise the al., 2009). Moreover, elbow arthroscopy specifically is tech- number of complications, which emphasizes the importance nically challenging due to limited working space and close of training in portal placement and arthroscopic skills to de- proximity of neurovascular structures (Hilgersom et al., liver safe surgical care (Rose & Pedowitz, 2015). 2017;Marshalletal., 1993; Miller et al., 1995;Omidet al., 2012; Stothers et al., 1995). Further distinguishing elbow * Correspondence: n.f.hilgersom@amc.uva.nl arthroscopy is the need for mirrored hand-eye coordination Department of Orthopaedic Surgery, Amsterdam University Medical Centres, in the lateral decubitus position when compared to most University of Amsterdam, Amsterdam Movement Sciences, Meibergdreef 9, other arthroscopic modalities; and overhand versus 1105 AZ Amsterdam, the Netherlands Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 2 of 9 underhand holding of instruments. All above, in combin- (90th percentile) to cover for the effects of other condi- ation with the lower frequency compared to knee or shoul- tions when setting the metrics’ threshold. der arthroscopy, makes it apparent that elbow arthroscopy has a longer learning curve in time. Cadaveric specimen Currently, no consensus exists on the minimal number of Two fresh-frozen right-handed upper limb cadaver spec- elbow arthroscopies that must be performed to become an imens without evidence of previous trauma, surgery or expert. Savoie states that a minimal number of 100 per- deformity were prepared to mimic an arthroscopic formed elbow arthroscopies is necessary (Savoie 3rd, 2007). setting. These specimens were derived from bodies that Furthermore, Claessen et al. (Claessen et al., 2017)observed entered the department of anatomy through a donation a 30% complication rate in portal placement by novice sur- program. From these persons written consent was geons, which was significantly higher compared to experi- obtained during life that allowed the use of their entire enced elbow arthroscopists (Elfeddali et al., 2013; Marti et bodies for educational and research purposes. Specimens al., 2013). These numbers make clear that elbow arthros- were stored at − 20 °C and thawed 24 h before use. The copy (simulated) training is essential (Claessen et al., 2017; upper limb cadaveric specimens arms were dissected Rose & Pedowitz, 2015). transversely 15–20 cm proximal of the humeral epicon- Cadaveric training is still the preferred training method dyles and mounted onto the custom-made static arm to improve arthroscopic skills because it provides the most holder of the force measurement table (FMT) with the realistic setting (Camp et al., 2016;Hui et al., 2013;Koehler posterior humerus facing superiorly and the humeral et al., 2015). Objective performance measurement by using epicondyles orientated horizontally, mimicking a lateral metrics is preferred over global rating scales such as Ob- decubitus position (Horeman et al., 2016). jective Structured Assessment of Technical Skills (OSATS) (Horeman et al., 2016;Martin etal., 1997; van Hove et al., Force measurement table 2010). Such metrics have yet to be defined in elbow For the interested readers, the force measurement table is arthroscopy, but have been defined in knee and shoulder described in detail by Horeman et al. (Horeman et al., arthroscopy, for example to differentiate between levels of 2016). In short, when a cadaveric specimen is firmly fixated experience and to set thresholds for safe tissue manipula- in the vice of the FMT, it measures the forces in x-, y-, tion (Stunt et al., 2014;Tashiro et al., 2009;Tuijthofetal., z-direction during arthroscopic skills training, enabling 2011). Recently, Obdeijn et al. (Obdeijn et al., 2016)defined objective performance tracking of the trainees. The FMT a maximum allowable force magnitude of 7.3 N (90th per- consists of three squared frames, each connected to one centile) using expert data derived thresholds and demon- another by four beams that bend upon loading (Fig. 1). The strated that force direction is equally important as force three frames displace independently; each in a single direc- magnitude for safe wrist arthroscopy to prevent cartilage tion (i.e. x-, y-, or z-direction) (Fig. 1). Theapplied forceon damage. Similarly, forces exerted on the elbow by experts each frame is calculated by measuring the relative displace- during elbow arthroscopy may also be valuable indicators ment of the four bending beams and multiplication with of a safe elbow arthroscopy. the bending beams’ known stiffness (Fig. 1). Bending beam The purpose of this study was to determine the force displacements were measured using Linear Hall effect magnitude and force direction used by experts during sensors and Neodymium disk magnet built into the bend- arthroscopic elbow navigation in cadaveric specimens ing beams (Horeman et al., 2016). The FMT allowed con- and assess their applicability in elbow arthroscopy tinuous recording of the forces exerted on the cadaver training. elbow by the instruments in a range of 0 N to 750 N in three loading directions, with an accuracy of 0.1 N and a Methods sample frequency of 24 Hz (Horeman et al., 2016). To pos- The study was designed to fit within the set time sched- ition the elbow above the FMT, a custom-made stand with ule of the two day-26th annual international Arthros- vice was mounted on the FMT. The vice allows fast mount- copy & Arthroplasty Courses Utrecht. This implied that ing of the prepared humerus bone in a 45-degree angle to we could perform data acquisition with five experts mimic the actual procedure (Fig. 1). operating on two cadaveric specimens. This approach A camera tracking system using two digital video cam- was suitable to meet the study goal, because a similar eras set up on both sides of the operator and the arthro- strategy was followed for assessing a threshold naviga- scope camera was set up for monitoring of instrument tion force for wrist arthroscopy (Obdeijn et al., 2016): a) use, capture ‘occurrences’ (e.g.; probing of the predefined focus on experts and recruit as many as possible to landmarks) and adequate postprocessing of the data determine if their navigation force variation is acceptably acquired with the FMT. small to set a safety threshold; b) keep other conditions Qualitative analyses of the individual contribution of as constant as possible; and c) propose a safety margin the arthroscope and probe on the total forces exerted on Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 3 of 9 Fig. 1 Force Measurement Table. The FMT with custom-made stand with vice attached is shown. The FMT consists of the three squared frames and bending beams with Hall effect sensors and magnets in the x-, y- and z-plane of the FMT. The design of the FMT allows continuous recordings of the forces exerted on the cadaver specimen attached to the vice in three loading directions the cadaver elbow was performed by combining the data portals. During the arthroscopic navigation experts from the FMT and camera tracking system. consecutively visualized the posterior, posterolateral and anterior compartment and were asked to determine the Experts predefined landmarks (Fig. 2). In the posterior compart- The expert group consisted of five upper limb surgeons ment the landmarks were the olecranon tip, olecranon specialized in elbow arthroscopy and instructors at the fossa, medial gutter, and lateral gutter (Fig. 2a). In the 26th annual international Arthroscopy & Arthroplasty posterolateral compartment the landmarks were the Courses Utrecht. The experts filled out a questionnaire radial head, capitellum, and proximal radioulnar joint to document their demographic data (Table 1). Prior to the experiment, expert one created the follow- Table 1 Demographic data and experience of the five participants ing arthroscopic portals in both cadaveric specimens; proximal anteromedial, proximal anterolateral, midtrici- Expert 1 2 3 4 5 pital, posterolateral and soft spot portal, and as routinely Age (years) 42 38 44 50 48 is performed with elbow arthroscopy, shaved fibrous Gender Male Male Male Female Female tissue blocking the view. The midtricipital, posterolateral Dexterity Right Right Left Right Right and proximal anteromedial portal served as viewing Expertise Expert Expert Expert Expert Expert portals for the posterior, posterolateral and anterior Exp EA (years) 2 4 8 16 15 compartment, respectively. NR EA (year) 100 100 20–25 100 10–15 Each expert performed an arthroscopic navigation once on each cadaveric elbow using the above-described Exp Experience, EA Elbow arthroscopy,NR Number Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 4 of 9 Fig. 2 Arthroscopic views of predefined landmarks per compartment. a Posterior compartment. b Posterolateral compartment. c Anterior compartment (Fig. 2b). In the anterior compartment the landmarks the total absolute force (F ) per sample was calcu- abs were the radial head, capitellum, coronoid tip, and lated by summation of the force measurements in the coronoid fossa (Fig. 2c). Each landmark had to be x-, y-, and z-plane after the force measurement in touched by the probe and visualised in the centre of z-direction was compensated for the mass of the specimen thearthroscopicimage.Once a landmark wasvisual- and holder. In addition, the force direction in the horizon- ized per protocol, as visually verified by one of the tal plane (α) could be derived from the force magnitude in researchers, the expert could proceed to the next ana- the x- and y-plane, and the force direction in the vertical tomic landmark. Arthroscopic elbow navigation was plane (β), which is aligned with the humerus mounted on performed in the same consecutive order of experts the set-up, from the force magnitude in the x- and on both elbow specimens. All measurements were z-plane. A positive α-angle implies a direction of force to performed on the same day. During the experiment the lateral side and a negative α-angle implies direction of experts could extend the elbow as they felt necessary force to the medial side. A positive β-angle implies upward for proper portal placement and instrument use. All direction of force and a negative β-angle implies down- arthroscopic tasks were performed using an arthro- ward direction of force. scopic probe and a 30°-angle 4 mm arthroscope from Karl Storz (Tuttlingen, Germany). Statistical analysis The experts were asked to perform the tasks as they The presence of normal distributions for F , α and β was abs would be performing live surgery on an actual patient. determined with the Kolmogorov-Smirnov test per com- partment. As the data were not normally distributed, F , abs Data processing α and β were expressed in terms of median (10th–90th The data gathered with the FMT and camera tracking percentile). A Mann-Whitney U-test was performed to system were processed using Matlab (version R2014a, compare the F measurements for the anterior compart- abs The Mathworks, Natick, MA, USA) and IBM SPSS ment between the two cadaveric specimens (p < 0.05). statistics (version 22, SPSS, Chicago, IL, USA). All Prior to this study, Obdeijn et al. (Obdeijn et al., 2016) raw voltage data were filtered with a low-pass Butter- successfully applied the 10th and 90th percentiles to set worth filter with a cut-off frequency of 24 Hz to sup- thresholds for safe tissue manipulation and force direc- press high-frequency noise. For each compartment tion in wrist arthroscopy. Therefore, we used a similar Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 5 of 9 strategy in this study: the 10th and 90th percentiles of force (Fig. 4a). The Mann-Whitney U-test indicated a F α and β were used to set threshold levels for safe significant difference between the values of F for the abs, abs tissue manipulation and force direction that can be used anterior compartment between the two cadaveric speci- during elbow arthroscopy training. mens (p <0.05). Results Horizontal angle (α) and vertical angle (β) Figure 3 shows an example of the force measurement in The median α, force direction in the horizontal plane, time of one navigation task performed by one expert in is −29° forthe posteriorcompartment with arange the posterior compartment. A qualitative initial analysis of 60°, is − 23° for the posterolateral compartment combining the force data and video footage showed with a range of 55° and is 4° for the anterior com- force fluctuations in a similar direction during probing partment with a range of 40° (Fig. 4b). Notable is the of a landmark, force fluctuations in an opposite direction more medial direction and smaller range of α in the during elbow flexion, and only marginal variation in anterior compartment compared to the posterior and forces during instrument changes when only the arthro- posterolateral compartment (Figs.4b and 6b). The me- scope was in place (Fig. 3). dian β, force direction in the vertical plane, is − 71° The histograms of F , α and β for the posterior, pos- for the posterior compartment with a range of 15°, − abs terolateral and anterior compartment of both cadaveric 76° for the posterolateral compartment with a range elbows are presented in Fig. 4. of 17° and − 75° for the anterior compartment with a range of 10° (Fig. 4c). The median β remains fairly Total absolute of force F constant with a maximum difference of 5° and max- abs The median F is similar for each compartment, imum range of 17° (Fig. 6c). Figure 5 provides a sche- abs being 24 N (range 19 N – 30 N) for the posterior matic representation of the median β for all compartment, 27 N (20 N – 33 N) for the posterolateral compartments combined. compartment, and 29 N (23 N – 32 N) for the anterior Comparison of the force direction between the differ- compartment (Fig. 4a). In the anterior compartment, two ent compartments showed a second smaller peak around peaks of F are observed in the histogram, one around − 90 degrees for α and β in the posterior and posterolat- abs 23 N of absolute force and one around 30 N of absolute eral compartment (Fig. 4b and c). Expert 1, who created Fig. 3 Example of force measurement in time of one navigation task performed by one expert. This example shows force measurement in time of one navigation task performed by one expert in the posterior compartment. In the upper graph the individual force components as well as the overall combined force F are illustrated. In the lower graph the stars indicate the moments of touching and displaying the assigned abs landmark. In this 2D representation, the first Area (A) represents a location were elbow flexing occurs indicated by an oppositely directed change in Force. The following area’s (B) represent instrument bone/tissue interaction with force fluctuations in similar direction Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 6 of 9 Fig. 4 Histograms presenting the data points for F , alfa and beta of each compartment. Histograms presenting the median, 10th and 90th abs percentile for frequency of total absolute force (F ), horizontal angle (α) and vertical angle (β) data points in the posterior, posterolateral and abs anterior compartment of both cadaver elbows. a Histogram showing F . The two black arrows point to two separate peaks in frequency of F , abs abs around 23 N and 30 N in the anterior compartment. b Histogram showing α. α is positive to the right, and negative to the left. The two black arrows point to a smaller peak around − 90° for α and β in the posterior and posterolateral compartment. c Histogram showing β. β is positive upward, and negative downward. The two black arrows point to a smaller peak around − 90° for α and β in the posterior and posterolateral compartment the portals, had a substantial share in this peak, particu- found for knee joint distraction (mean F of 43-50 N) abs larly when performing the task in the first cadaveric (Stunt et al., 2014). specimen. A possible contributing factor to the overall higher force load is that manoeuvring the arthroscope to a compartment Safe zone – Metric threshold is performed primarily by knowing the correct orientation Finally, a graphical interpretation is given in Fig. 6 of the of the arthroscope and by haptic feedback using the bony median values of the F , α and β for each of the three structures for guidance, such as sliding along the anterior abs compartments (posterior, posterolateral and anterior) as face of the humerus to create the proximal anteromedial well as the set safe zone using the 10th and 90th percentile portal, or using bony structures as a support point/wedge force values from Fig. 4. The 90th percentile values indi- to take a corner while navigating around the elbow (Fig. 5). cate the set maximum threshold for the metric. In addition to a lack of joint distraction, this relative high bone-instrument loading may cause a higher overall loading Discussion on the elbow. The consequences of the relative high force This study shows that median loads of 24-29 N are may be limited, because surgical procedures during elbow exerted on the elbow by experts during arthroscopic arthroscopy are primarily performed outside of the articu- navigation in a cadaveric elbow. These loads represent lating surfaces of the elbow joint (i,e. synovectomy, capsular the combined forces exerted by the arthroscope and the release, loose body removal). This reduces the chance of in- probe on the anatomic structures of the elbow. The jury to delicate tissues inside the joint such as the poorly overall measured forces are considerably higher than ex- healing articular cartilage. pert force data for wrist arthroscopy (median F of The arthroscope assembly (e.g. arthroscope, cables, abs 3.8 N) (Obdeijn et al., 2016) and probing of meniscal tis- camera) and supporting hand plus arm most likely have sue in the knee (mean F ranging 2.8–3.9 N) (Tuijthof the highest contribution in the total force as combined abs et al., 2011), but they are lower than expert force data analysis of video footage and force data with the aim to Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 7 of 9 The median force direction and range during arthro- scopic navigation in the elbow is similar for all compart- ments in the vertical plane (β) (Figs. 4c, 5 and 6c). The median force direction of the anterior compartment in the horizontal plane (α) is more medial compared to the pos- terior and posterolateral compartment, and the range of force direction is smaller when compared to the posterior and posterolateral compartment (Fig. 4b and 6b). These findings can be related to working through the proximal anteromedial portal, the anatomical location of the anter- ior compartment and the anatomical distance between the landmarks in the anterior compartment, respectively. The second smaller peak in force direction in both planes (α and β) observed around − 90° in the posterior and posterolateral compartment (Fig. 4b and c) seems attributable to suboptimal portal placement as expert 1 who created the portals had a substantial share in this second peak, particularly in the first specimen. This is supported by the lower median F used by expert 1 in abs cadaver 1. Expert 1 created the portals and as such knew the exact orientation of the portals resulting in a lower median F compared to the other experts. abs Elbow arthroscopy, when performed with the patient in a Fig. 5 Median force direction in the vertical plane (β) for all lateral decubitus position, requires a mirrored way of in- compartments combined. Schematic representation of the median strument handling with a 30° arthroscope when compared direction of Fabs in the vertical plane for all compartments to performing arthroscopy of most other joints. The force combined represented by β. The blue lines represent the lowest and direction in the vertical plane (β)shows minor variation highest values of the 10th and 90th percentiles of β. An elbow x-ray has been superimposed over the cadaveric elbow to further clarify (Figs. 5 and 6c), which is a sign that this range may be used the correlation between the overall median force direction and for novice surgeons to strive for. The latter is strengthened elbow joint by Obdeijn et al. (Obdeijn et al., 2016; Obdeijn et al., 2014) who showed that force direction is equally as important as correlate force direction and variation in force magnitude force magnitude, and found that novices showed consider- to instrument use showed hardly any variation in forces able variation in loading direction compared to experts during probing of landmarks or instrument changes with when performing wrist arthroscopy. The force direction the arthroscope in place (Figs. 3 and 5). This is a possible area defined by the 10th–90th percentile of expert thresh- assumption as the FMT measured the total combined olds for α and β (Figs. 5 and 6) may be used to adjust the forces exerted on the cadaveric elbow and is unable to direction of the arthroscope to properly navigate through quantitatively assess the individual contribution of the the complex elbow anatomy. To be of assistance for the arthroscope or probe used during elbow arthroscopy. trainees, it is necessary to visualize the direction of force on Based on expert data, the 10th and 90th percentiles of the video screen via augmented reality. Implementing this the exerted force have been used to determine force in a meaningful way is a challenging task, as is shown by thresholds in wrist arthroscopy and probing of menisci theworkofSmitetal. (Smitetal., 2017). (Obdeijn et al., 2016; Tuijthof et al., 2011). Utilizing the There are limitations to this study. First, although the same strategy on current expert data, we propose a max- number of data points per surgeon was high, the num- imum allowable force load of 30 N to be exerted on the ber of experts and cadavers was small, but feasible elbow during arthroscopic navigation, which is the smal- within the set time frame of the advanced elbow course. lest value of 90th percentiles of the force magnitude of Besides the variation amongst the experts, other condi- all three compartments (Fig. 4a). This threshold level tions (cadavers, the joint status in time and portal place- should be demonstrated in elbow arthroscopy training ment) do effect the forces. Since our aim was not to to let novices experience the feel of the magnitude of a assess the individual contribution of each condition, but load around 30 N, as this is a most likely a lot higher rather set an overall safety threshold, we argue that the than novices expect (Obdeijn et al., 2014; Tuijthof et al., small group of surgeons conducting the trials on two ca- 2011). This can help students to train their haptic senses davers should represent the entire group of expert elbow in a safe way by preventing them to use higher loads. arthroscopists sufficiently. This is supported by the Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 8 of 9 Fig. 6 Safe force zone (magnitude and direction) for all compartments. The red dots indicate the median value of F , α and β for each abs compartment. The thick black lines with white dots at their respective ends represent the 10th and 90th percentile value of F in the median abs force direction. The grey boxes surrounding the thick black lines indicate the combined boundaries of the 10th -90th percentiles for F , α and β. abs The origin is taken at the same position for each compartment. a 3D graphical representation. For reference an elbow in the lateral decubitus position is added. So the forces are directed towards the surgeon, b Top view, c Sagittal view narrow range of the 10th–90th percentile of the median and second cadaver elbow, respectively. Consequently, values of F , α and β. Second, the data was collected threshold levels as determined in this study should be ad- abs from cadaveric specimens that are usually stiffer than el- justed per cadaver and training time on the cadaver (swell- bows from live patients. Therefore, one can reason that ing due to irrigation). Therefore, we recommend starting higher forces will be observed when performing arthros- training basic elbow arthroscopic skills on a simulator. This copy on living patients. However, this may be partly will provide the same standard for all trainees at any time, compensated as cadaveric specimen are commonly ob- and allows adequate objective feedback by setting one tained from elderly people with usually lesser tissue threshold value and facilitates observation of training pro- quality than young people. gress of participants compared to their peers. After obtain- Although cadaveric training provides the most realistic ing proficiency in basic arthroscopic skills on a simulator, a experience, cadaveric training is not the preferred method trainee may advance to cadaveric skills training to become to start training elbow arthroscopy skills. First, because ca- acquainted with the feeling and effect of the loads on hu- daveric training is expensive and there is limited availability man tissues along with learning to adapt to anatomic varia- (Camp et al., 2016; Stirling et al., 2014). Moreover, as was tionsasisthe case in livesurgery. also shown in this study, the anatomic variation amongst Nonetheless, this study shows that force data can be cadaveric specimen as well as their joint status in due to accurately and reliably recorded in three loading direc- time compromises similar training conditions for a certain tions using the FMT (Horeman et al., 2016), allowing amount of repetitions or trainees. For example, in the expert thresholds to be defined for force magnitude and present study two peaks of F were observed during navi- force directions that can be used for objective feedback abs gation of the anterior compartment, around 23 N and 30 N during elbow arthroscopy training. (Fig. 4a), which were attributable to the use of two cadaver elbows (Mann-Whitney U, p < 0.05). In addition, due to Conclusions continuous water irrigation of the elbow for a long duration Expert data on force magnitude and force direction (five elbow arthroscopies) the soft tissues would swell, pos- exerted on the elbow during arthroscopic navigation in sibly making portal insertion, gaining orientation and work- cadaveric specimens was collected. The proposed max- ing inside the joint more difficult. In this study, this was imum allowable force of 30 N (smallest 90th percentile observed as moderate differences in the median F of of F ) exerted on the elbow tissue, and the 10th–90th abs abs 8.2 N and 3.5 N between the first and last expert in the first percentile range of the force directions (α and β) for Hilgersom et al. Journal of Experimental Orthopaedics (2018) 5:45 Page 9 of 9 each compartment may be used to provide objective Claessen F, Kachooei AR, Kolovich GP, Buijze GA, Oh LS, van den Bekerom MPJ, Doornberg JN (2017) Portal placement in elbow arthroscopy by novice feedback during arthroscopic skills training. surgeons: cadaver study. Knee Surg Sports Traumatol Arthrosc 25:2247–2254 Elfeddali R, Schreuder MH, Eygendaal D (2013) Arthroscopic elbow surgery, is it Abbreviations safe? J Shoulder Elb Surg 22:647–652 F : Total absolute force; FMT: Force Measurement Table; OSATS: Objective abs Hilgersom NF, Oh LS, Flipsen M, Eygendaal D, van den Bekerom MP (2017) Tips Structured Assessment of Technical Skills to avoid nerve injury in elbow arthroscopy. World J Orthop 8:99–106 Hilgersom NFJ, Molenaars RJ, van den Bekerom MPJ, Eygendaal D, Doornberg JN Acknowledgements (2018) Review of Poehling et al (1989) on elbow arthroscopy: a new We are grateful to the Elbow Study Collaborative for their assistance technique. J ISAKOS 3:116–124 during the elbow course, and participating as elbow experts in our study. Horeman T, Tuijthof GJM, Wulms PB, Kerkhoffs GMMJ, Gerards RM, Karahan M (2016) Elbow Study Collaborative Bertram The, Carina L.E. Gerritsma, Lex A force measurement system for training of arthroscopic tissue manipulation Boerboom, Tom Roeling, Marco van der Pluijm, Michel P.J. van den Bekerom, skills on cadaveric specimen. J Med Devices 10:044508–044501/044507 and Denise Eygendaal. Hui Y, Safir O, Dubrowski A, Carnahan H (2013) What skills should simulation training in arthroscopy teach residents? A focus on resident input. Int J Availability of data and materials Comput Assist Radiol Surg 8:945–953 The datasets used and/or analysed during the current study are available Koehler R, John T, Lawler J, Moorman C 3rd, Nicandri G (2015) Arthroscopic from the corresponding author on reasonable request. training resources in orthopedic resident education. J Knee Surg 28:67–74 Marshall PD, Fairclough JA, Johnson SR, Evans EJ (1993) Avoiding nerve damage during elbow arthroscopy. J Bone Joint Surg Br 75:129–131 Authors’ contributions Marti D, Spross C, Jost B (2013) The first 100 elbow arthroscopies of one surgeon: NFJH was responsible for data interpretation, manuscript preparation, analysis of complications. J Shoulder Elb Surg 22:567–573 manuscript design, creation of figures and editing. GJMT and TH were Martin JA, Regehr G, Reznick R, MacRae H, Murnaghan J, Hutchison C, Brown M responsible for the study design, data collection, data analysis, data (1997) Objective structured assessment of technical skill (OSATS) for surgical interpretation, and providing feedback on the manuscript. RLWAB was residents. Br J Surg 84:273–278 responsible for the availability of cadaveric specimens, study design and Miller CD, Jobe CM, Wright MH (1995) Neuroanatomy in elbow arthroscopy. J providing feedback on the manuscript. DE and MPJB were responsible for Shoulder Elb Surg 4:168–174 data interpretation, clinical insight and providing feedback on the Obdeijn MC, van Baalen SJ, Horeman T, Liverneaux P, Tuijthof GJ (2014) The use manuscript. All authors read and approved the final manuscript. of navigation forces for assessment of wrist arthroscopy skills level. J Wrist Surg 3:132–138 Ethics approval Obdeijn MC, Horeman T, de Boer LL, van Baalen SJ, Liverneaux P, Tuijthof GJ The cadaveric specimens used in this study were derived from bodies that (2016) Navigation forces during wrist arthroscopy: assessment of expert entered the department of anatomy, University of Utrecht, through a donation levels. Knee Surg Sports Traumatol Arthrosc 24:3684–3692 program. From these persons written consent was obtained during life that Omid R, Hamid N, Keener JD, Galatz LM, Yamaguchi K (2012) Relation of the allowed the use of their entire bodies for educational and research purposes. radial nerve to the anterior capsule of the elbow: anatomy with correlation to arthroscopy. Arthroscopy 28:1800–1804 Consent for publication Rose K, Pedowitz R (2015) Fundamental arthroscopic skill differentiation with Not applicable. virtual reality simulation. Arthroscopy 31:299–305 Rosenthal R, Gantert WA, Scheidegger D, Oertli D (2006) Can skills assessment on Competing interests a virtual reality trainer predict a surgical trainee's talent in laparoscopic The authors declare that they have no competing interests. surgery? Surg Endosc 20:1286–1290 Savoie FH 3rd (2007) Guidelines to becoming an expert elbow arthroscopist. Arthroscopy 23:1237–1240 Publisher’sNote Smit D, Spruit E, Dankelman J, Tuijthof G, Hamming J, Horeman T (2017) Springer Nature remains neutral with regard to jurisdictional claims in Improving training of laparoscopic tissue manipulation skills using various published maps and institutional affiliations. visual force feedback types. Surg Endosc 31:299–308 Stirling ER, Lewis TL, Ferran NA (2014) Surgical skills simulation in trauma and Author details orthopaedic training. J Orthop Surg Res 9:126 Department of Orthopaedic Surgery, Amsterdam University Medical Centres, Stothers K, Day B, Regan WR (1995) Arthroscopy of the elbow: anatomy, portal University of Amsterdam, Amsterdam Movement Sciences, Meibergdreef 9, sites, and a description of the proximal lateral portal. Arthroscopy 11:449–457 1105 AZ Amsterdam, the Netherlands. Department of Biomechanical Stunt JJ, Wulms PH, Kerkhoffs GM, Sierevelt IN, Schafroth MU, Tuijthof GJ (2014) Engineering, Delft University of Technology, Mekelweg 2, 2628 CD Delft, the Variation in joint stressing magnitudes during knee arthroscopy. Knee Surg Netherlands. Department of Human Movement Sciences, Vrije Universiteit Sports Traumatol Arthrosc 22:1529–1535 Amsterdam, De Boelelaan 1117, 1081 HV Amsterdam, the Netherlands. Tashiro Y, Miura H, Nakanishi Y, Okazaki K, Iwamoto Y (2009) Evaluation of skills in Department of Anatomy, University Medical Centre Utrecht, Heidelberglaan arthroscopic training based on trajectory and force data. Clin Orthop Relat 100, 3584 CX Utrecht, the Netherlands. Department of Orthopaedic Surgery, Res 467:546–552 Amphia Hospital, Molengracht 21, 4818 CK Breda, the Netherlands. Tuijthof GJ, Horeman T, Schafroth MU, Blankevoort L, Kerkhoffs GM (2011) Department of Orthopaedic Surgery, Onze Lieve Vrouwe Gasthuis, Probing forces of menisci: what levels are safe for arthroscopic surgery. Knee Oosterpark 9, 1091 AC Amsterdam, the Netherlands. Zuyd University of Surg Sports Traumatol Arthrosc 19:248–254 Applied Science, Nieuw Eyckholt 300, 6419 DJ Heerlen, the Netherlands. van Hove PD, Tuijthof GJ, Verdaasdonk EG, Stassen LP, Dankelman J (2010) Objective assessment of technical surgical skills. Br J Surg 97:972–987 Received: 16 May 2018 Accepted: 20 September 2018 Yeoh KM, King GJ, Faber KJ, Glazebrook MA, Athwal GS (2012) Evidence-based indications for elbow arthroscopy. Arthroscopy 28:272–282 References Aggarwal R, Moorthy K, Darzi A (2004) Laparoscopic skills training and assessment. Br J Surg 91:1549–1558 Aim F, Lonjon G, Hannouche D, Nizard R (2016) Effectiveness of virtual reality training in Orthopaedic surgery. Arthroscopy 32:224–232 Camp CL, Krych AJ, Stuart MJ, Regnier TD, Mills KM, Turner NS (2016) Improving resident performance in knee arthroscopy: a prospective value assessment of simulators and cadaveric skills laboratories. J Bone Joint Surg Am 98:220–225

Journal

Journal of Experimental OrthopaedicsSpringer Journals

Published: Oct 11, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off