Glenoid bone loss is a common cause of morbidity after reverse total shoulder arthroplasty (RTSA) in patients with inflammatory arthritis. Glenoid bone loss can lead to insufficient bone stock that makes successful implantation of glenoid baseplates difficult, if not impossible, because of inadequate bony fixation. The frequency of glenoid bone loss in patients with inflammatory arthritis, based on preoperative imaging, has been reported to range from 37% to 78%,1–4 and early studies investigating RTSA in patients with inflammatory arthritis showed glenoid mechanical failure rates as high as 25%.5 To combat this problem, many surgeons supplement the eroded glenoid with intraoperative bone grafting or modify baseplate placement to maximize the bone-implant interface. With the use of modern bone-grafting and placement techniques and innovative implants, surgeons have improved glenoid baseplate fixation overall and particularly in patients with inflammatory arthritis. Recent systematic reviews of RTSA in patients with inflammatory arthritis have shown glenoid loosening and fracture rates of approximately 4.9% and 2.9%, respectively,6,7 giving a mechanical glenoid failure rate of approximately 7.8% or one in 13 patients.
Despite the prevalence of glenoid bone loss seen on preoperative imaging in patients with inflammatory arthritis, the authors found no study specific to these patients that has demonstrated a correlation between the amount of glenoid bone loss and the frequency of mechanical glenoid failure. The purpose of our study was to determine and compare the glenoid vault volumes of patients with inflammatory arthritis undergoing RTSA to determine if a threshold exists where baseplate fixation is at risk for failure. To determine glenoid vault volumes, we used three-dimensional (3-D) volumetric modeling based on preoperative CT scans. Statistical analysis was then performed to compare glenoid vault volumes. We hypothesized that a lower glenoid vault volume would be associated with an increased risk of glenoid baseplate failure because of inadequate fixation.
MATERIALS AND METHODS
Ethical Review and Study Design
This study was approved by the institutional review board of the University of Tennessee (approval no. 15-024214-XM). All patients provided informed consent.
A search of our institutional billing database using the Current Procedural Terminology (American Medical Association, Chicago, IL, USA) code 23472 identified all primary shoulder arthroplasty procedures performed by a single fellowship-trained shoulder and elbow surgeon between 2010 and 2015. Within this initial search group, a chart review identified patients with a diagnosed history of inflammatory arthritis. Fourteen subjects (five men and nine women) met these inclusion criteria. All 14 had primary RTSA with the use of the Comprehensive Shoulder System (Zimmer Biomet, Warsaw, IN, USA), including the mini-glenosphere baseplate.
Treatment and Outcome Measurements
Humeral head autografting to supplement glenoid bone stock was done at the time of surgery if deemed necessary by the operating surgeon. Immediate perioperative baseplate or glenoid component failure (defined by glenoid fracture or component loosening requiring revision or conversion to hemiarthroplasty) was then determined to stratify patients into failures and nonfailures.
Preoperative shoulder CT scans were imported into Mimics (Materialise, Leuven, Belgium) to measure preoperative glenoid vault volumes. We used a method similar to that described by Codsi et al.8 to determine the glenoid vault boundary: drawing a line along the plane of the glenoid fossa and then drawing a second line perpendicular to the fossa line and tangential to the endosteal surface at the spinoglenoid notch. A mask was made of the space defined by the endosteal surface and the second tangential line for each slice of the glenoid vault. A 3-D model of the vault was then constructed from the mask, and the glenoid vault volume was retrieved from this model.
Glenoid vault volumes (cm3) were calculated and compared for failures and nonfailures. Gender-based subgroup comparisons of failure and nonfailure vault volumes were done to assess male-specific and female-specific differences. A two-sided t-test was performed to compare each group. Differences with P<0.05 were considered statistically significant.
Of the 14 patients, 10 (four men and six women) required glenoid bone stock supplementation with humeral head autograft. Four patients, (two men and two women), experienced baseplate failures secondary to glenoid component loosening or fracture. All four patients with failures required glenoid bone grafting at the time of surgery.
The average glenoid vault volume for all patients who experienced failures was 7.64 cm3 with a range from 2.06 cm3 to 13.2 cm3 and standard deviation of 5.05 cm3. The average glenoid vault volume for all patients who did not experience failures was 11.51 cm3 with a range from 5.68 cm3 to 19.81 cm3 and standard deviation of 5.13 cm3. When comparing glenoid vault volumes of all failures to all nonfailures, no statistical significance was found (P=0.226).
The average glenoid vault volume for failures in male patients was 11.78 cm3 with a range from 10.35 cm3 to 13.20 cm3 and standard deviation of 2.02 cm3. The average glenoid vault volume for nonfailures in men was 18.31 cm3 with a range from 17.14 cm3 to 19.81 cm3 and standard deviation of 1.37 cm3. When comparing failures to nonfailures in men, a statistically significant difference was found (P=0.021).
The average glenoid vault volume for failures in female patients was 3.52 cm3 with a range from 2.06 cm3 to 4.98 cm3 and standard deviation of 2.06 cm3. The average glenoid vault volume in female patients with nonfailure was 8.59 cm3 with a range from 5.68 cm3 to 11.90 cm3 and standard deviation of 2.39 cm3. When comparing failures to nonfailures in women, a statistically significant difference was found (P=0.031).
There were no baseplate failures in men with glenoid vault volumes of more than 13.5 cm3 or in women with volumes of more than 5.0 cm3.
Mechanical glenoid failure continues to be a significant cause of morbidity in patients with inflammatory arthritis who have RTSA, with approximately one in 13 patients sustaining this complication. This is particularly relevant in patients with inflammatory arthritis in whom insufficient bone stock often is magnified by poor bone quality, and determining which patients are at increased risk for mechanical glenoid failure may be a helpful preoperative surgical planning tool.
The results of this study demonstrate that baseplate failures, in both male and female patients with inflammatory arthritis, occurred in patients with significantly lower average glenoid vault volumes when compared to their gender-specific counterparts. Furthermore, these results suggest a level of glenoid vault volume below which baseplate fixation is potentially compromised. Specifically, male patients with glenoid vault volumes below 13.5 cm3 and female patients with glenoid vault volumes of less than 5.0 cm3 may be at higher risk for baseplate failure even with the use of bone grafting to augment glenoid deficiency.
Our study has several limitations. First, it does not account for bone removed during the reaming process, which could have had an additional effect on failure; however, we believe that the reaming process would not have caused a significant difference in bone loss between failures and nonfailures because all surgeries were performed by a single fellowship-trained surgeon, thus standardizing the reaming process. Also, six of 10 patients who required humeral autografting at the time of surgery did not sustain a baseplate failure, suggesting that preoperative bone volume is more indicative of failure than post-reaming residual glenoid volume. Although our study defines overall glenoid vault volumes, it does not specify where in the glenoid vault that volume is located. A study that defines both the amount and location of glenoid vault volume may be more predictive of which patients are truly at risk for failure. Although this exclusion may seem a disadvantage to our study, we believe that measuring volume alone, as opposed to volume and location, is a more practical preoperative evaluation because of the simplicity of having only one measurable variable.
In individuals with inflammatory arthritis, male patients with glenoid vault volumes of 13.5 cm3 or less and female patients with glenoid vault volumes of 5 cm3 or less may be at higher risk of baseplate failures because of insufficient bone stock despite the use of bone grafting. Further studies are needed to validate our results and determine whether measuring glenoid vault volumes would be a helpful preoperative surgical planning tool.
1. Ekelund A, Nyberg R. Can reverse shoulder arthroplasty be used with few complications in rheumatoid arthritis? Clin Orthop Relat Res. 2010; 469:2483–2488.
2. Hattrup S, Sanchez-Sotelo J, Sperling J, et al. Reverse shoulder replacement for patients with inflammatory arthritis
. J Hand Surg Am. 2012; 37:1888–1894.
3. Holcomb J, Hebert D, Mighell M, et al. Reverse shoulder arthroplasty in patients with rheumatoid arthritis. J Shoulder Elbow Surg. 2010; 19:1076–1084.
4. Young A, Smith M, Bacle G, et al. Early results of reverse shoulder arthroplasty in patients with rheumatoid arthritis. J Bone Joint Surg Am. 2011; 93:1915–1923.
5. Rittmeister M, Kerschbaumer F. Grammont reverse total shoulder arthroplasty
in patients with rheumatoid arthritis and nonreconstructible rotator cuff lesions. J Shoulder Elbow Surg. 2001; 10:17–22.
6. Cho C, Kim D, Song K. Reverse shoulder arthroplasty in patients with rheumatoid arthritis: a systematic review. Clin Orthop Surg. 2017; 9:325–331.
7. Gee E, Hanson E, Saithna A, et al. Reverse shoulder arthroplasty in rheumatoid arthritis: a systematic review. Open Orthop J. 2015; 9:237–245.
8. Codsi M, Bennetts C, Gordiev K, et al. Normal glenoid vault anatomy and validation of a novel glenoid implant shape. J Shoulder Elbow Surg. 2008; 17:471–478.