Pierre-Henri Flurin, MD
Laurent Angibaud, Dipl. Ing.
Computer-assisted orthopedic surgery (CAOS) systems, such as navigation platforms, have been developed for various applications since the 1990s as an opportunity to improve the accuracy of the alignment of implants.1,2,3,4,5 CAOS application dedicated to the field of total shoulder arthroplasty (TSA) has been more difficult to achieve, hampered by technical difficulties related to the reliability of the equipment6 and the challenges of reduced visibility due to the shoulder’s limited exposure.
However, the clinical result of a TSA highly depends on the quality of anatomical and biomechanical restitution of the glenohumeral joint.7,8,9 The difficulties of proper appreciation of the anatomical landmarks combined with the narrowness of the surgical field tend to challenge the accuracy of the bone cut and preparation, resulting in a crucial need for TSA-dedicated guidance solutions such as personalized surgical instrumentation (PSI) and CAOS systems.10,11,12,13,14,15,16
Combined advances in computed tomography (CT) medical imaging and technical improvements in computer and tracking systems have enabled the development of a new generation of image-based CAOS system for shoulder prosthetic application.17,18,19,20,21
There is currently only one widely used computer-aided navigation system on the market for TSA application, whose technical and clinical evaluations are encouraging.22,23,24,25,26,27 This initial system is expected to be joined by other technological platforms over the upcoming years, all further enhanced by the development of augmented and/or virtual reality, robotic, sensor, and machine learning. In this chapter, we will describe the history as well as the general principles of CAOS in the field of orthopedics. Then, we will describe, more precisely, the globally launched computer-aided navigation dedicated to TSA. Finally, we will present the results of validation studies and the current clinical evaluation of this technology.
HISTORY OF NAVIGATION
The initial concepts of CAOS emerged almost 30 years ago, with the first clinical applications1,2,3,4 being reported in the mid-1990s. From these early days, numerous technologies in the field of CAOS were developed for a large range of clinical applications such as total hip arthroplasty, total knee arthroplasty (TKA), spine, and a variety of other procedures. The overachieving goal behind each of these technologies was to provide guidance to the surgeon to improve the accuracy of implant positioning and reproducibility as an attempt to ultimately improve clinical outcomes.
Over the years, CAOS has morphed into different types of technology. The initial development related to active robotic systems intended to perform the machining of the envelope of the expected femoral stem based on preoperative CT-based plan.5 The inherent complexity and cost of these active robots limited their adoption in the operating room.
In parallel, semiactive systems were developed, where the robot prepositions a cutting instrument according to a preoperative CT-based plan, but the bone cut/preparation is ultimately performed by the surgeon.28
The most widely used family of CAOS systems relates to the passive systems such as computer-aided navigation, where the surgeon follows guidance and/or information displayed on a screen at the time of surgery.29,30,31,32,33,34 These navigation systems rely on surgical plans that, in turn, may be based on intraoperative measurements (aka imageless system) or on preoperative three-dimensional (3D) models derived from CT (aka image-based system).
Image-based navigation systems are based on two key technologies: 3D model reconstruction from a CT scan and a real-time surgical guidance tracking system, used at the preoperative and intraoperative stage, respectively.
At the preoperative stage, a CT of the patient is performed, and a reconstructed 3D model is established. A planning application allows the surgeon to establish a surgical plan by selecting the proper size and type of implant and then planning the position and orientation of the selected implant relative to the reconstructed 3D model.
At the time of surgery, key anatomical landmarks are acquired in order to establish the relationship between the patient’s anatomy and the reconstructed 3D model. Once the verification of the registration is completed, the surgeon follows the displayed information to guide the orientation and position of the cutting tools to prepare the bone according to the preestablished surgical plan.
In their current form, navigation systems have demonstrated superior results in terms of alignment by substantially reducing the occurrence of outliers.29,30,31,32,33,34 Also, because of their real-time guidance, these systems represent an effective educational tool to upskill the most junior surgeons.35,36 However, usual navigation systems tend to be perceived as compromising the efficiency in the operating room by adding operative time.30,33,34,37 In addition, some of these systems may appear as being too complex, which directly impacts the user experience. Finally, the most active debate relates to the benefit of using CAOS in terms of clinical outcomes, which may explain its somewhat limited adoption over the years. A recent study using data from the Australian Orthopaedic Association National Joint Replacement Registry showed that the revision rate for TKA patients less than 65 years old was reduced by ˜20% if computer-assisted navigation was used.38 Notably, the authors found that its usage led to a significant reduction in the rate of revision due to loosening/lysis (P = 0.001), which is the leading reason for revision TKA. Along the same lines, another study demonstrated that CAOS produced better clinical outcomes compared to traditional surgery in TKA after 1 year.39 Such evidence tends to demonstrate that greater accuracy of implantation results in an improved survival rate. Finally, it can be hypothesized that navigation systems provide added value for difficult indications such as those associated with reduced exposure or bone defects and deformity when visual landmarks are limited, such as in TSA.
DESCRIPTION OF A CONTEMPORARY CAOS PLATFORM FOR TSA
The presented CAOS system (ExactechGPS [eGPS], Blue Ortho, Gières, France) has been developed with a particular focus on enriching the user experience through added efficiency and efficacy. The first clinical application for TKA was released in 2010 and then seconded by an application dedicated to TSA in 2016. Since then, the application has been used in almost 30,000 surgeries. The eGPS is considered as a closed platform, being only compatible with the Equinoxe shoulder implant from the same manufacturer.
The eGPS includes (1) a display unit composed of a proprietary infrared charge-coupled device camera and a touchscreen tablet intended to be located in the sterile field and directly accessible by the surgeon during the surgery and (2) a set of wireless active trackers. The camera is intended to define the 3D position and orientation within 6° of freedom of the trackers rigidly attached to patient bone and surgical instruments as well as the system-specific probe used to acquire the anatomical landmarks during the registration phase. The infrared camera provides an extra-large field of vision (>135°) for maintaining the line of sight during the surgery.40 Based on the limited distance between the localizer and the camera, the high frequency of the camera, and other proprietary attributes (eg, algorithms), this hardware has been proven to offer both accurate and precise surgical resection measurements during simulated TKA.41
The eGPS for TSA includes two dedicated software applications. The preoperative planning application (Equinoxe Planning App, Blue Ortho, Gières, France) allows surgeons to establish a surgical plan regarding the preparation of the glenoid based on a reconstructed 3D model of the scapula referencing the Friedman axis connecting the center of the glenoid to the trigonum.42 The intraoperative application (ExactechGPS Shoulder Application, Blue Ortho, Gières, France) is intended to provide real-time visual guidance and alignment in order to execute the surgical plan at the time of the surgery.
Finally, the last aspect of the eGPS relates to the navigated mechanical instrument, which includes a navigated modular driver for a series of drills as well as side-specific blocks intended to rigidly attach the reference tracker to the scapula at the level of the coracoid.
Preoperative Navigation Steps
A preoperative CT scan is performed for each patient requiring a TSA according to a precise protocol with 1-mm-thick sections of the entire scapula.
Subsequently, the CT is uploaded to the preoperative application with the objective of performing surgical planning according to two distinct options:
If the planning is only intended for visualization, then the immediate reconstruction feature of the application is used to establish an automated 3D-reconstructed scapula on which selected implants can be assessed in terms of position and/or orientation relative to a referential based on the Friedman axis connecting the center of the glenoid to the trigonum established by the user.
If the planning is intended to be used for surgical navigation, then a manual 3D reconstruction performed by Blue Ortho specialists is required.
After completion of the manual reconstruction, a detailed 3D rendering of the scapula featuring the Friedman axis is available for planning purposes using the dedicated application. In this application, the surgeon plans the glenoid portion of an anatomical TSA (ATSA) or a reverse TSA (RTSA) in order to optimize the choice and position of the implant.
The inclination of the selected glenoid implant is calculated relative to the frontal plane of the scapula. This plane is defined by Friedman axis and the most inferior point of the scapula.
The version of the selected glenoid implant is calculated relative to a plane passing through Friedman axis, orthogonal to the frontal plane of the scapula.
Preoperative planning allows the user to visualize the deformity and/or bone defect in 3D and to select the most suitable implant (eg, size, length of the central peg, posterior or superior augment, standard or expanded glenosphere for RTSA). The possibility of adjusting the position and orientation of the glenoid implant in terms of inclination, version, and depth is even more useful in cases of substantial deformity or a small glenoid, which allows the user to anticipate eccentric reaming or the use of augmented implants.