Kaveh R. Sajadi, MD, FAAOS
Read complete study: Computer navigation of the glenoid component in reverse total shoulder arthroplasty: a clinical trial to evaluate the learning curve.
Accurate placement of the glenoid component in reverse total shoulder arthroplasty (rTSA) is important to reduce component loosening, scapular notching, instability and to maximize impingement-free range of motion. Initial stability of the implant is critical for bony in-growth and is dependent upon optimal screw placement and maximizing screw length. Glenoid exposure and anatomy can be difficult, and in the presence of glenoid wear or deformity, placement of the implant and screws may be challenging.
Computer navigation has been shown to improve the ability to place the central glenoid post/cage fully within the glenoid vault and to optimize screw placement and length (Nashikkar et al). Computer navigation has also been shown to improve accuracy in achieving correct version and inclination of the glenoid component (Nguyen et al; Verborgt et al; Kircher et al). Some surgeons may be concerned about the perception of increased surgical time associated with computer navigation as well as any learning curve that comes with adopting a new procedure. This study by Wang et al investigated the learning curve of computer navigated rTSA, the accuracy of implant placement, and the effect on surgical time.
This well-designed study was a prospective case-series of a single experienced shoulder surgeon evaluating his first 24 consecutive navigated rTSAs. Although he had extensive experience with the arthroplasty system, his only prior experience with this navigation system was a sawbone workshop, a single cadaveric case, and a single clinical case to allow himself and his surgical team to be familiar with the setup and workflow. All cases underwent preoperative planning using Blue Ortho CT protocol (Blue Ortho, Grenoble, France). The goal of planning was to attain a glenoid implant version as close to 00 as possible using reaming or augments, attain 00 of inferior inclination, maximize bone-implant contact, and keep the glenoid central cage fully contained in the glenoid vault. Preoperative planning took 3-5 minutes per patient, a step I would argue should be done whether navigation is planned or not.
Surgical time was recorded from skin incision to closure. Stages of the navigation process were separately timed as well, involving coracoid dissection and tracker placement, bony landmark registration, glenoid preparation, and screw placement. For a control group, times were compared to the prior 24 non-navigated reverse arthroplasties. Postoperative 3-D CT scans were obtained 10-12 weeks after surgery and used to compare accuracy of implant placement with regard to the preoperative plan. In 2 of the 24 cases, complications occurred. In one case, the coracoid tracker malfunctioned and had to be replaced, adding approximately 10 minutes to the case; in another, the tracker was not able to be fixed to the coracoid and navigation was abandoned.
The mean overall surgical time for the navigated cases was 77.3 minutes, compared to the previous non-navigated cases which had a mean time of 78.5 minutes. In addition, using regression analysis, a curve of best fit showed a clear significant downward trend which flattened after 8 cases, indicated a surgical learning curve of 8 cases. No indication for a learning curve was identified in the individual stages, though the first stage, coracoid tracker placement, showed the greatest variation in time. Finally, accuracy of implant placement by postoperative CT showed the mean deviation of version of 30 with a range of 0-70. The inclination showed a mean deviation of 50 and range of 0-110. Neither version nor inclination showed a learning curve.
This timely study addresses several important points regarding computer navigation in shoulder arthroplasty. First, it confirms previous studies in showing that computer navigation improves accuracy of implant placement and decreases variability. Implant placement here was within 30 of version and 50 of inclination with tight ranges. This is consistent with the previous study by Kircher et al, which showed improved accuracy of glenoid placement, as well as numerous studies from the hip and knee literature which show improved implant placement and decreased variability.
Second, this study shows that the learning curve for adding computer navigation to the armamentarium of an experienced shoulder surgeon is relatively short–approximately 8 cases. This echoes my personal learning experience with computer navigation. Though I have not precisely measured times as nicely as Dr. Wang and his colleagues, my comfort and speed significantly improved after approximately 4-5 cases. As the authors further note, this learning curve compares very favorably with other surgical procedures, such as the arthroscopic Latarjet which has a reported learning curve of up to 20 cases.
Finally, this study refutes any concerns about increased operative time by adding computer navigation. The overall surgical time for navigated cases (77.3 minutes) was almost identical to the surgical time for non-navigated cases (78.5 minutes). Although there are additional steps involved in navigation (exposing the coracoid, placing the coracoid tracker, and registering the bony landmarks), this time is likely saved during reaming and implant placement due to increased confidence in position and improved visualization of trajectories.
One important point from this study is that the results are not necessarily generalizable to other computer navigation systems. This study specifically evaluated the ExactechGPS® computer-assisted surgical technology. The previous study by Kircher et al, using a different system, showed similar accuracy but a significantly longer operative time associated with computer navigation.
Computer navigation improves the accuracy of glenoid implant placement and increases the length of screws placed (Nashikkar et al), which also improves initial implant stability, which improves the likelihood of bony in-growth for long-term stability. This study shows that navigation is easy to adopt with a relatively short learning curve and no significant additional time requirement during surgery. I have adopted computer navigated shoulder arthroplasty, as I believe this is the best way to address complex glenoids during surgery.
Kaveh Sajadi, MD, FAAOS, practices orthopaedics with Kentucky Bone and Joint Surgeons and is an instructor in the University of Kentucky’s residency program. He completed his residency at the Campbell Clinic and his fellowship at the NYU Langone Hospital for Joint Diseases. Dr. Sajadi is a frequent instructor at Exactech domestic and international shoulder courses.
References:
- Wang AW, Hayes A, Gibbons R, Mackie KE. Computer navigation of the glenoid component in reverse total shoulder arthroplasty: a clinical trial to evaluate the learning curve. J Shoulder Elbow Surg 2020; 29(3):617-623. https://doi.org/10.1016/j.jse.2019.08.012
- Kircher J, Wiedemann M, Magosch P, Lichtenberg S, Habermayer P. Improved accuracy of glenoid positioning in total shoulder arthroplasty with intraoperative navigation: a prospective-randomized clinical study. J Shoulder Elbow Surg 2009; 18:515-20. https://doi.org/10.1016/j.jse.2009.03.014
- Nashikkar P, Scholes C, Haber M. Role of intraoperative navigation in the fixation of the glenoid component in reverse total shoulder arthroplasty: a clinical case-control study. J Shoulder Elbow Surg 2019; 28(9):1685-1691. https://doi.org/10.1016/j.jse.2019.03.013
- Nguyen D, Ferreira LM, Brownhill JR, King GJ, Drosdowech DS, Faber KJ, et al. Improved accuracy of computer assisted glenoid implantation in total shoulder arthroplasty: an in-vitro randomized controlled trial. J Shoulder Elbow Surg 2009; 18:907-14. https://doi.org/10.1016/j.jse.2009.02.022
- Verbogt O, De Smedt T, Vanhees M, Clockaerts S, Parizel PM, Van Glabbeek F. Accuracy of placement of the glenoid component in reversed shoulder arthroplasty with and without navigation. J Shoulder Elbow Surg 2011; 20:21-6. https://doi.org/10.1016/j.jse.2010.07.014