Navigation Improves Our Ability to Reach That Goal.
Gregory S. Bauer, MD
Read complete study: Three-dimensional preoperative planning and patient-specific instrumentation for total shoulder arthroplasty
Computer navigation leads to more accurate glenoid targeting during Total Shoulder Arthroplasty (TSA) compared with three-dimensional (3D) preoperative planning alone.
As shoulder surgeons, it is our goal to provide our patients with the best possible outcomes after they choose to undergo shoulder arthroplasty. There are patient factors we cannot control, such as tissue quality, bone loss, compliance, and overall health. However, we can control how we implant the prosthesis. The development of 3D planning software has helped us develop a better plan before we go to the operating room. It is up to us to execute the plan.
Studies have shown that there is a concern for increased loosening rates if glenoid components for anatomic shoulders are placed in excessive retroversion.1 We try to correct that with selective reaming and possibly using augmented components. The goal is to implant the glenoid closer to neutral version and to avoid superior inclination in reverse shoulder replacements. With all the innovations we have seen surrounding shoulder arthroplasty, is there anything that can help us execute these plans at a higher level of precision?
Planning software allows us to create a plan optimizing implant selection, position and degree of bony support. However, the accuracy of implantation is still a concern. Iannotti et al. demonstrated that without 3D planning software, inclination and version were off by an average over 10° each in a sawbones model. The utilization of 3D planning software improved accuracy. Yet, version was still off by an average of 6.7°, and inclination was off by 9.3° from the preoperative plan.2
In the same article, the authors looked at their abilities to execute their preoperative plans that had been created using 3D planning software. The procedures were to be performed with the assistance of computer navigation. With this technology, the 3D plan with CT data is loaded into the device. A line–of–sight camera is used to view trackers that are affixed to the scapula and to surgical instruments. The trackers and instruments are used to register multiple points on the scapula to the navigation system that can now provide real-time 3D information and feedback to the surgeon for this patient’s scapula.
In this study, they retrospectively reviewed 50 primary shoulder arthroplasties. All cases had been planned utilizing 3D planning software. During the case, the trackers were secured to the coracoid and the scapula data points were registered to the navigation system. Then, both an attending surgeon and a fellow attempted to place the guide pin on the glenoid face in the position, version and inclination that they had planned preoperatively. They could not see the screen to provide feedback as to the accuracy of their placement. This was recorded and later evaluated. Malposition was determined to have taken place if the starting point was >4mm displaced from the planned starting point or there was a >10° variance in version or inclination from the plan.
The study group included 25 men and 25 women. Thirteen (13) shoulders had a TSA implanted and 37 had a Reverse Shoulder Arthroplasty (RSA) implanted. For the TSA group, three (3) patients had standard glenoids placed and 13 had 8° posterior augmented glenoid components placed. For the RSA group, four (4) patients had standard baseplates implanted, and the remainder received an augmented baseplate. Glenoids were classified as Walch A1 in 23 patients, A2 in six (6) patients, B1 in five (5) patients, B2 in 13 patients, and B3 in three (3) patients. For the RSA patients, glenoids were also classified with Favard classification as E0 in 22 patients, E1 in nine (9) patients, and E2 in six (6) patients.
The results showed a mean displacement of the starting point of 3.2mm (range, 0.3-10.3mm) with 18 measurements greater than 4mm. Error in glenoid version averaged 6.4° (range, -24 to 24). Forty-nine (49) % of cases exceeded 5°, and 25% exceeded more than 10° from the plan. The mean error in inclination was 6.6° (range, -13 to 19). Fifty (50) % exceeded 5°. Overall, 48% of the time the placement of the guide pin would have been off by more than 4mm or greater than 10° in version or inclination from the preoperative plan. The attendings were more accurate with their pin placement than the fellows with the attendings having an error rate of 38% and the fellows an error rate of 58%.
With 20 years of experience as a fellowship–trained shoulder surgeon, I came to one conclusion very quickly when I adopted intraoperative navigation: I was not as good as I hoped I was. When I looked straight on at the glenoid face and was about to drill a center hole–just like I had done a thousand times before–I realized that often what I would have done was different than what I had preoperatively planned. The real-time feedback of the navigation system changed the way I would have performed the procedure to the way I had actually planned the procedure many times. This personal experience is consistent with the findings of this study. I felt as though the GPS intraoperative navigation got me within 2mm and 2° of my plan.3 The fellowship–trained attendings involved in this study are all high–volume surgeons with vast experience with complex shoulder arthroplasties. Despite their vast experience, they would have made an error, based upon the study criteria, executing their preoperative plan by more than 4mm or greater than 10° of version or inclination 38% of the time without the aid of navigation.
Performing shoulder arthroplasties can be challenging. How we implant the prostheses has a direct result on our patients’ short– and long–term outcomes. Navigation improves our ability to accurately execute our plans for our patients with the goal of maximizing outcomes.
References:
- Farron A, Terrier A, Büchler P. Risks of loosening of a prosthetic glenoid implanted in retroversion. J Shoulder Elbow Surg. 2006 Jul-Aug;15(4):521-6. doi: 10.1016/j.jse.2005.10.003.
- Iannotti J, Baker J, Rodriguez E, Brems J, Ricchetti E, Mesiha M, Bryan J. Three-dimensional preoperative planning software and a novel information transfer technology improve glenoid component positioning. J Bone Joint Surg Am. 2014 May 7;96(9):e71. doi: 10.2106/JBJS.L.01346.
- Data on file at Exactech, Inc.
Greg Bauer, MD, a shoulder specialist at Goldsboro Orthopaedics in North Carolina, completed his internship and orthopaedic residency at Washington University School of Medicine and conducted his shoulder fellowship at Columbia-Presbyterian Medical Center. Dr. Bauer is an accomplished presenter and author.