Abstract
Purpose
To evaluate the accuracy of arthroscopic glenoid bone loss (GBL) assessment and its influencing factors, and to compare the consistency of three-dimensional computed tomography (3D CT) and 3D printed measurements.
Methods
3D printed models were created from the 3D CT images. Experts reviewed arthroscopic images to assess GBL, and a questionnaire was used to correlate assessment accuracy with surgeon age, experience, and annual surgical volume. Separately, ten residents measured each model three times using three methods: the linear method and best-fit circle method on 3D CT, and a manual method with vernier calipers on the physical models. Agreement was calculated using intra-group correlation coefficients (ICC).
Results
A statistically significant difference in assessment accuracy was found among surgeons with different annual surgical volumes (p = 0.029). Comparisons between the three measurement methods showed that the best-fit circle method yielded significantly higher values than both the linear and manual methods (p < 0.001). The manual method demonstrated the highest consistency across different times and observers (ICC 0.998 and 0.995, respectively), suggesting it is the most stable technique.
Conclusion
The accuracy of arthroscopic GBL judgment varies with the surgeon’s annual surgical volume. While all three measurement methods show excellent agreement, the best-fit circle method provides significantly larger GBL measurements compared to the linear and manual methods.
Levels of Evidence
Level II, Cross-Sectional Study.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12891-026-09709-w.
Keywords: Glenoid bone loss, 3D CT measurement, 3D printed measurement, Best-fit circle method, Linear method
Introduction
Bony Bankart lesion, a fracture or bone defect in the anteroinferior part of the glenoid rim with an associated labral detachment, is commonly associated with anterior instability of the glenohumeral joint. Although recommendations have been made for the treatment of glenoid bone loss (GBL), no actual consensus has been reached. Chuang et al. concluded that patients with bone loss less than 25% of the glenoid do well with arthroscopic Bankart repair only [5], while Provencher et al. believes soft-tissue stabilization is recommended for glenoid defects less than 15% [19]. However, this data is not well validated and the most appropriate value may be higher or lower [21]. Due to the ambiguity in the calculation of the GBL, there may be a huge discrepancy between the surgeon’s assessment of the defect area and the true defect area, with the arthroscopic estimate being 55% higher than that of three-dimensional computer tomography (3D CT) methods [2], according to Bakshi et al. Therefore, accurate bone defect assessment and proper surgical approach selection remain a great challenge in clinical practice. There are many different methods for evaluating GBL, such as the surface area method, the ratio method, the bare spot method, the best-fit circle method and the Pico method [22]. In the measurement of GBL, 3D CT has been considered as “gold standard” [22, 18, 3, 8] , some studies have also pointed out that 3D MRI shows comparable consistency with 3D CT in the measurement of GBL [22 ,23, 27] and even outperforms 3D CT in some aspects such as reduction of radiation and evaluation of soft tissues [11, 17, 25] . However, these methods are not very accurate due to objective factors such as the curvature of the glenoid.
Therefore, the purpose of this study was to evaluate the accuracy of arthroscopic GBL assessment and its influencing factors, and to compare the consistency of 3D CT and 3D printed measurements. Our hypotheses were that: (1) Surgeons’ accuracy in arthroscopic GBL judgment varies with seniority, and 3D printed models are superior to 3D CT for junior surgeons. (2) The best-fit circle method shows poor inter-observer consistency compared to the linear and manual methods. We therefore propose a novel, more stable measurement technique using 3D printed models.
Methods
Establishment and Measurement of the GBL Model
All patients provided informed consent. Ensured the availability of 3D CT images of both injured and uninjured sides in the study. The CT scans were performed with a slice thickness of 0.625 mm and a resolution of 512 × 512 pixels using the standard bone reconstruction sequence with 120 kV (peak). Hole-filling, void-removal and surface-smoothing tools were applied as needed for artifact reduction, leading to a 3D surface model [26]. Imported the images into the image-based modeling application Mimics (model specification: Mimics Medical 21.0). The 3D printer was used to create printed models, and each printed model was reviewed by one surgeon. Compared the measurements of printed models with on-screen measurements of the 3D CT images to verify the anatomical scale of the printed model [26]. All surgical procedures were performed by a surgeon with 12 years of upper limb clinical experience in sports medicine. Determined the standard view of the glenoid through a standard anterolateral portal. During the surgery, a C-arm machine was used for imaging, the spatial relative position between the scapula and the arthroscope was recorded and the images were retained (Fig. 1). However, since there is no constant and clear anatomical marker for arthroscopic judgment of GBL, and the position of the arthroscope is difficult to determine, this study cannot completely determine the exact position of the arthroscope, and can only reproduce the intraoperative position of the arthroscopy according to the X-ray pictures, which is also a constraint of this study. Similarly, when observed the bone defects of the 3D printed models with an arthroscope, the relative position of the scapula and the arthroscope was made as close as possible to the intraoperative position, and retained the picture (Fig. 2). These pictures were disrupted and designed into a questionnaire to determine GBL by sports medicine physicians, none of whom were involved in the treatment of these patients. However, despite the attempt to simulate intraoperative arthroscopic views in this study, the questionnaire results could only reflect experts’ interpretation of standardized images due to issues such as the absence of soft tissues and accurate joint spaces, rather than actual intraoperative arthroscopic decision-making. Since there is no recognized gold standard in clinical practice, we measured GBL using a variety of methods, including the best fit circle method, Pico method, linear method under 3D CT measurement and 2D CT measurement, and 3D printing measurement. Based on these calculated values, a range is determined as the standard answer of the questionnaire.
Fig. 1.
Intraoperative imaging of C-arm machine, the patient was placed in the lateral position, and intraoperative imaging with a C-arm machine was used to determine the position of the arthroscope
Fig. 2.
Arthroscopic picture of 3D printed models, the 3D printed model was secured with stents to simulate the intraoperative position and then take arthroscopic pictures
The area of the GBL was manually measured by ten residents, all of whom underwent protocol training before measuring. The bone defect measurements in this paper were recorded as defect length, taking into account that the calculation of bone defect area is more complex. One author reconstructed 3D images by mimics software (Fig. 3), and the glenoid widths were measured by ten residents on the computer and 3D printed models. Manual method was made with vernier calipers (Fig. 4). In the software, the GBL was measured by the linear method and the best-fit circle method, respectively, and the calculations were as follows: The linear method defined GBL as “a - b”, where “a” and “b” are the maximum widths of the lower 2/3 glenoid on the unaffected and affected sides, respectively [22]. The best-fit circle method calculated GBL as “c - d”, where “c” is the radius of the best-fit circle on the lower two-thirds glenoid, and “d” is the shortest distance from the circle’s center to the anterior glenoid rim. As for the manual method, the vernier caliper is placed on the surface of the glenoid and the GBL is calculated as “e - f”, with “e” and “f” representing the maximum width of the unaffected side and the affected side of the lower 2/3 glenoid. To assess consistency, the same residents repeated all measurements one month later in a randomized order.
Fig. 3.
3D reconstructed images of four patients. Three-dimensional reconstruction of the CT of the four patients was performed by the same physician using mimics software
Fig. 4.
Model measurements were made with vernier calipers, the stent secured the scapula and the width of the articular glenoid was measured using vernier calipers
Statistical analysis
The data were analyzed by IBM SPSS Statistics, version 26.0. The description of quantitative data was expressed as mean ± standard deviation, and t-test or analysis of variance (ANOVA) were used if it conformed to normal distribution and homogeneity of variance; otherwise, non-parametric test was used. The intra-class correlation coefficient (ICC) was used to evaluate the consistency of residents. The ICC ranged from 0.00 to 1.00—0.00 ~ 0.20 were considered to poor consistency; 0.20 ~ 0.40, fair; 0.40 ~ 0.60, moderate; 0.60 ~ 0.80, good; 0.80 ~ 1.00, excellent. P<0.05 was considered statistically significant.
Results
We selected four patients to undergo bilateral shoulder CT examination, the mean age of the patients was 30.00 years (range 26–35 years), of which 3 (75.00%) were male and 2 (50.00%) had dislocations of dominant hand, the average number of preoperative dislocations was 7.50 (range 4–10), and the average duration of symptoms was 25.50 months (range 12–48 months). All these patients underwent Bankart + Remplissage surgery. According to the follow-up results, none of these patients had any fear of dislocation during postoperative daily activities (e.g., playing badminton or doing housework), but a little abnormal sensation like crepitus or click was still left during high-intensive training (e.g., military training) with negtive apprehension-relocation test.
Online questionnaires were completed by 102 domestic surgeons. After excluding two with incomplete information, 100 responses were analyzed. The participants represented over 2/3 of provinces, ensuring reasonable geographic representation. Based on linear and manual measurement standards, the overall correct response rate was only 35.5% (142/400). Survey results indicated a strong preference for the lateral position (87.0%) and anterolateral approach (76.0%) during GBL assessment. Regarding the GBL threshold for soft tissue surgery, 46 surgeons selected < 20%, 29 opted for < 10%, and 22 supported < 25%. Notably, two respondents provided outliers: 40% and 5%.
The data was further analyzed (Table 1). There was a statistically significant difference in the number of questions answered correctly between the different surgical volume groups (p = 0.029). There was no significant difference between other factors (e.g., age, time to become a surgeon-in-chief, patient position, or portal for observing GBL) and the number of correct answered questions.
Table 1.
The relationship between age, time to become a chief surgeon, annual amount of arthroscopic surgery, patient position, portal for observing GBL and the number of correct answered questions
| Age | Total (n = 100) |
0 (n = 13) | 1 (n = 42) | 2 (n = 37) | 3 (n = 8) | F/χ² | P |
|---|---|---|---|---|---|---|---|
| 42.10 ± 5.08 | 42.23 ± 4.11 | 41.60 ± 5.82 | 42.41 ± 4.49 | 43.13 ± 5.59 | 0.28 | 0.834 | |
| Time to become a surgeon-in-chief | 12.45 ± 5.65 | 13.00 ± 5.52 | 12.29 ± 5.23 | 12.24 ± 6.38 | 13.38 ± 5.26 | 0.14 | 0.938 |
| Annual amount of arthroscopic surgery, n(%) | |||||||
|
<10 10–50 50–100 >100 |
2 (2.00) 9 (9.00) 15 (15.00) 74 (74.00) |
0 (0.00) 2 (2.00) 5 (5.00) 6 (6.00) |
2 (2.00) 6 (6.00) 7 (7.00) 27 (27.00) |
0 (0.00) 1 (1.00) 3 (3.00) 33 (33.00) |
0 (0.00) 0 (0.00) 0 (0.00) 8 (8.00) |
15.61 | 0.029 |
| Patient position, n(%) | |||||||
|
Beach chair position Lateral position |
13 (13.00) 87 (87.00) |
2 (2.00) 11 (11.00) |
4 (4.00) 38 (38.00) |
6 (6.00) 31 (31.00) |
1 (1.00) 7 (7.00) |
1.20 | 0.824 |
| Portal for observing GBL, n(%) | |||||||
|
Posterior portal Anterolateral portal |
24 (24.00) 76 (76.00) |
2 (2.00) 11 (11.00) |
12 (12.00) 30 (30.00) |
9 (9.00) 28 (28.00) |
1 (1.00) 7 (7.00) |
1.72 | 0.628 |
Bold indicates statistical significance
Abbreviations: GBL Glenoid bone loss
Ten sports medicine residents measured GBL using three methods on both 3D CT reconstructions and 3D printed models. Given the complexity of calculating defect area, bone defect length was used as a surrogate measurement in this study. A one-way ANOVA revealed homogeneous variance among the three methods, enabling post-hoc multiple comparisons (Tables 2 and 3). The Least Significant Difference (LSD) test demonstrated no significant difference between the manual and linear methods (p = 0.859), indicating comparable accuracy between 3D CT and 3D printed model measurements. However, significant differences were observed between the manual and best-fit circle methods, and between the linear and best-fit circle methods (p < 0.001). Both manual and linear methods yielded significantly smaller measurements than the best-fit circle method (mean differences: -2.53 mm and − 2.68 mm, respectively).
Table 2.
The one-way ANOVA between MM, LM and BM methods
| Methods (mean ± SD) | F | P | |||
|---|---|---|---|---|---|
| MM | LM | BM | |||
| The length of GBL | 4.46 ± 3.79 | 4.31 ± 3.82 | 6.98 ± 4.42 | 12.14 | <0.001 |
Bold indicates statistical significance
Abbreviations: GBL Glenoid bone loss, MM Manual method, LM Linear method, BM Best-fit circle method
Table 3.
Post hoc multiple comparisons between MM, LM and BM methods
| Methods | Methods | Mean difference | P | |
|---|---|---|---|---|
| The length of GBL | MM | LM | 0.15052 | 0.990 |
| MM | BM | -2.52604 | <0.001 | |
| LM | BM | -2.67656 | <0.001 |
Bold indicates statistical significance
Abbreviations: GBL Glenoid bone loss, MM Manual method, LM Linear method, BM Best-fit circle method
The consistency in this study was also analysed (Table 4). In general, the consistency between the different times, observers and methods are all excellent (ICC>0.9). However, the consistency between different times and different observers of manual method was higher than the other two methods (ICC 0.998 and 0.995, respectively), which may suggests that manual method is more stable.
Table 4.
The consistency between two measurements, different observers and methods
| ICC | P | 95%CI95% | |
|---|---|---|---|
| The consistency of the two measurements | |||
| MM | 0.998 | <0.001 | 0.995–0.999 |
| LM | 0.954 | <0.001 | 0.884–0.979 |
| BM | 0.911 | <0.001 | 0.824–0.953 |
| Intra-observer consistency | |||
|
MM LM BM |
0.998 0.965 0.918 |
<0.001 <0.001 <0.001 |
0.996–0.999 0.931–0.983 0.846–0.961 |
| Inter-method consistency | |||
|
MM vs. LMMM MM vs. BMMM LM vs. BMLM |
0.978 0.784 0.758 |
<0.001 0.036 0.037 |
0.970–0.987 -0.047-0.938 -0.048-0.925 |
Bold indicates statistical significance
Abbreviations: MM Manual method, LM Linear method, BM Best-fit circle method, ICC Intra-group correlation coefficients, 95%CI 95% confidence interval
Discussion
This study systematically evaluated the accuracy of GBL assessment by surgeons and compared three measurement methods. A key finding was that assessment accuracy varied significantly with the surgeon’s annual surgical volume, highlighting the inherent limitations of arthroscopic evaluation. Although all three methods—best-fit circle, linear, and manual—demonstrated excellent reliability, the best-fit circle method yielded significantly larger measurements than the other two.
Substantial controversy persists in GBL measurement. Current techniques include X-ray, CT, 3D CT, 3D MRI, and arthroscopy. However, arthroscopic assessment is often compromised by limited viewing angles and soft tissue obstruction, leading to inaccurate probe placement and GBL underestimation (Fig. 5). Although higher surgical volume is associated with better performance, overall accuracy remains low. This may be due to the defects in the arthroscopic evaluation itself and the fact that they were only given an arthroscopic picture and did not give them other methods such as 3D CT to assist them in their judgment. There is a clear need for a simpler and more stable assessment method for newly appointed surgeons. In this context, 3D printing technology emerges as a crucial tool for supporting clinical decision-making. The main advantages of the 3D printed model are as follows: The 3D printed model can determine the en-face view, which can be used to estimate the GBL area with an accuracy comparable to that of 3D CT and is a better aid for less experienced surgeons; in addition, it can simulate the preoperative operation of bony surgery, which is beneficial to the surgeon in designing the surgical plan [1]. The most accurate method for measuring GBL has not yet been determined, and even 3D CT, which is currently considered by most scholars to be the “gold standard”, has certain disadvantages. For example, since the average anteroposterior diameter of the glenoid being approximately 24.4 mm ± 2.9 mm [15], even slight discrepancies in measurement [7] or deviations from the en-face view [7, 12, 13] may have substantial clinical affects. This study demonstrated that all three measurement methods exhibited excellent consistency in results; however, the best-fit circle method yielded significantly larger measurements compared to the linear method and the manual method (p <0.001). It should be noted that this conclusion does not mean that the best fit circle method overestimates GBL when the gold standard has not yet been determined, and we cannot determine whether the best fit circle method overestimates the defect or the linear/manual measurement method underestimates the defect. This discrepancy may arise from multiple factors, including interobserver variability in manually defining the circle and geometric assumptions inherent to the fitting algorithm [6]. While no statistically significant difference was found between 3D CT and 3D printed measurements, we observed negative GBL values in 3D CT measurements, possibly due to image rotation artifacts. Converting the defect area to the length of the defect is a limitation of this study, but in fact, a 2.5 mm length defect can also reach a difference of about 5% after converting to a circular area, which may affect the progress of a surgical decision. Therefore, how to make a good preoperative prediction of GBL has become an urgent problem in the field of shoulder surgery: overestimation of GBL will lead to an increase in unnecessary surgical procedures and affect the postoperative rehabilitation of patients, whereas underestimation of GBL will lead to a higher risk of re-dislocation. The emergence of 3D printing technology can convert two-dimensional imaging data into touchable and measurable 3D structures. 3D printed models based on preoperative CT can provide characteristics of the patient’s anatomy [26], enabling surgeons to have a clearer understanding of the morphology and defect area of the glenoid. A cadaver study [29], comparing manual measurements with 3D CT measurements, demonstrated that there was no significant differences between them, it also showed that our study has some theoretical support, however, in terms of being able to accurately find the en-face view and the reliability of the method, the manual measurement method appears to be superior to 3D CT measurements. Although this technique is currently time-consuming, it allows for direct visual scanning of the scapula and is a valuable and practical tool for evaluating the GBL and measuring the glenoid [1]. 3D printing also allows for better measurement of the depth of the articular glenoid compared to traditional CT scans [29, 1]. Meanwhile, 3D printing can also simulate complex bone defect before surgery, design surgical procedures, and help doctors perform better surgery.
Fig. 5.
Probe placement affects the judgement of glenoid bone defects. Intraoperative real-life images suggest that the position of the probe can greatly influence the judgement of the articular glenoid bone defect
Arthroscopy examination can overestimate GBL, which has been confirmed by several studies [2, 16], and differences in arthroscopic portal and placement can affect the determination of GBL, with some studies believe that accuracy of measurement is higher near the three o’clock position [20, 10]. However, because the glenoid has a certain angle of inclination, it is difficult to reduce the error caused by the angle of inclination. The position of the arthroscopic lens can also affect the determination of GBL. For example, due to the principle of “near big and far small”, if the lens is placed close to the bone defect area, it will artificially enlarge the bone defect area, resulting in an overestimation of the GBL; the probe may not always be at a right angle to the long axis of the glenoid, and a deviation from the short axis may lead to underestimation of the GBL [9]. In addition, the lower 2/3 of the glenoid is not a flat surface and has a certain depth, making it more suitable to be considered a sphere or ellipsoid. Therefore, measuring the GBL in 3D view, constructing a “best fit sphere” 28, or calculating the concave surface area of the glenoid [14] is also a future research direction. It is believed that we should incorporate technologies such as mimics, 3D printing, and even artificial intelligence to enable surgeons to perform more complex and precise tasks in restricted spaces [4], increasing the accuracy of GBL determination.
Limitations
First, although this study has met the sample size requirement through previous sample size calculations, the overall sample size is still small, and future studies should focus on validating the conclusions of this study in a wider population. Second, the lack of adjacent soft tissue in 3D printing limits the simulation of normal anatomical structures [20], which can be further verified in the future by constructing 3D printed models containing soft tissues. Third, the findings of this study cannot be directly translated into surgical indication thresholds. We anticipate more clinical follow-up studies to confirm appropriate bony surgical thresholds. Finally, due to the lack of gold standards, it is difficult to judge which method is more in line with the actual situation, and each method has its own advantages and disadvantages, and should be judged together in clinical practice.
Supplementary Information
Acknowledgements
We would like to acknowledge the contributions of editors and reviewers for their time, valuable comments, and constructive suggestions. We would also like to acknowledge the contributions of CL.W and YF.Z to this study.
Abbreviations
- GBL
Glenoid bone loss
- 3D CT
Three-dimensional computed tomography
- ICC
Intra-group correlation coefficients
- MM
Manual method
- LM
Linear method
- BM
Best-fit circle method
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yuxin Xie, Shoulong Song and Fei Zhang. The first draft of the manuscript was written by Yuxin Xie and all authors commented on previous versions of the manuscript. All authors read and approved of the final manuscript.
Funding
This work was supported by National Natural Science Foundation of China (Grant numbers [82572718]).
Data availability
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Chinese PLA General Hospital (No.S2024-684-01).
Consent for publication
Informed consent was obtained from all individual participants included in the study. Additional informed consent was obtained from all individual participants for whom identifying information is included in this article. The authors affirm that human research participants provided informed consent for publication of the images in Figs. 1 and 3.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Yuxin Xie and Shoulong Song contributed equally to this work.
References
- 1.Al Najjar M, Mehta SS, Monga P. Three dimensional scapular prints for evaluating glenoid morphology: An exploratory study [published correction appears in J Clin Orthop Trauma. 2020 Nov-Dec;11(6):1176. 10.1016/j.jcot.2018.06.001 [DOI] [PMC free article] [PubMed]
- 2.Bakshi NK, Patel I, Jacobson JA, Debski RE, Sekiya JK. Comparison of 3-Dimensional Computed Tomography-Based Measurement of Glenoid Bone Loss With Arthroscopic Defect Size Estimation in Patients With Anterior Shoulder Instability. Arthroscopy. 2015;31(10):1880–5. 10.1016/j.arthro.2015.03.024. [DOI] [PubMed] [Google Scholar]
- 3.Bishop JY, Jones GL, Rerko MA, Donaldson C, MOON Shoulder Group. 3-D CT is the most reliable imaging modality when quantifying glenoid bone loss. Clin Orthop Relat Res. 2013;471(4):1251–6. 10.1007/s11999-012-2607-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bozkurt M, Apaydin N, Işik C, Bilgetekin YG, Acar HI, Elhan A. Robotic arthroscopic surgery: a new challenge in arthroscopic surgery Part-I: Robotic shoulder arthroscopy; a cadaveric feasibility study. Int J Med Robot. 2011;7(4):496–500. 10.1002/rcs.436. [DOI] [PubMed] [Google Scholar]
- 5.Chuang TY, Adams CR, Burkhart SS. Use of preoperative three-dimensional computed tomography to quantify glenoid bone loss in shoulder instability. Arthroscopy. 2008;24(4):376–82. 10.1016/j.arthro.2007.10.008. [DOI] [PubMed] [Google Scholar]
- 6.Do WS, Kim JH, Lim JR, Yoon TH, Chun YM. Disagreement Between the Accepted Best-Fit Circle Method to Calculate Bone Loss Between Injured and Uninjured Shoulders. Am J Sports Med. 2023;51(4):885–92. 10.1177/03635465221149743. [DOI] [PubMed] [Google Scholar]
- 7.Frank RM, Golijanin P, Vopat BG, et al. Impact of Sagittal Rotation on Axial Glenoid Width Measurement in the Setting of Glenoid Bone Loss. Am J Orthop (Belle Mead NJ). 2018;47(6). 10.12788/ajo.2018.0041. [DOI] [PubMed]
- 8.Green GL, Arnander M, Pearse E, Tennent D. CT estimation of glenoid bone loss in anterior glenohumeral instability: a systematic review of existing techniques. Bone Jt Open. 2022;3(2):114–22. 10.1302/2633-1462.32.BJO-2021-0163.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Griffith JF, Yung PS, Antonio GE, Tsang PH, Ahuja AT, Chan KM. CT compared with arthroscopy in quantifying glenoid bone loss. AJR Am J Roentgenol. 2007;189(6):1490–3. 10.2214/AJR.07.2473. [DOI] [PubMed] [Google Scholar]
- 10.Ji JH, Kwak DS, Yang PS, Kwon MJ, Han SH, Jeong JJ. Comparisons of glenoid bony defects between normal cadaveric specimens and patients with recurrent shoulder dislocation: an anatomic study. J Shoulder Elb Surg. 2012;21(6):822–7. 10.1016/j.jse.2011.10.016. [DOI] [PubMed] [Google Scholar]
- 11.Kompel AJ, Li X, Guermazi A, Murakami AM. Radiographic Evaluation of Patients with Anterior Shoulder Instability. Curr Rev Musculoskelet Med. 2017;10(4):425–33. 10.1007/s12178-017-9433-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kubicka AM, Stefaniak J, Lubiatowski P, et al. Reliability of measurements performed on two dimensional and three dimensional computed tomography in glenoid assessment for instability. Int Orthop. 2016;40(12):2581–8. 10.1007/s00264-016-3253-9. [DOI] [PubMed] [Google Scholar]
- 13.Lacheta L, Herbst E, Voss A, et al. Insufficient consensus regarding circle size and bone loss width using the ratio-best fit circle-method even with three-dimensional computed tomography. Knee Surg Sports Traumatol Arthrosc. 2019;27(10):3222–9. 10.1007/s00167-019-05391-9. [DOI] [PubMed] [Google Scholar]
- 14.Launay M, Choudhry MN, Green N, et al. Three-Dimensional Quantification of Glenoid Bone Loss in Anterior Shoulder Instability: The Anatomic Concave Surface Area Method. Orthop J Sports Med. 2021;9(6):23259671211011058. 10.1177/23259671211011058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lenart BA, Freedman R, Van Thiel GS, et al. Magnetic resonance imaging evaluation of normal glenoid length and width: an anatomic study. Arthroscopy. 2014;30(8):915–20. 10.1016/j.arthro.2014.03.006. [DOI] [PubMed] [Google Scholar]
- 16.Magarelli N, Milano G, Baudi P, et al. Comparison between 2D and 3D computed tomography evaluation of glenoid bone defect in unilateral anterior gleno-humeral instability. Radiol Med. 2012;117(1):102–11. 10.1007/s11547-011-0712-7. [DOI] [PubMed] [Google Scholar]
- 17.Markenstein JE, Jaspars KC, van der Hulst VP, Willems WJ. The quantification of glenoid bone loss in anterior shoulder instability; MR-arthro compared to 3D-CT. Skeletal Radiol. 2014;43(4):475–83. 10.1007/s00256-013-1780-7. [DOI] [PubMed] [Google Scholar]
- 18.Min KS, Sy JW, Mannino BJ. Area Measurement Percentile of 3-Dimensional Computed Tomography Has the Highest Interobserver Reliability When Measuring Anterior Glenoid Bone Loss. Arthroscopy. 2023;39(6):1394–402. 10.1016/j.arthro.2022.12.035. [DOI] [PubMed] [Google Scholar]
- 19.Provencher MT, Bhatia S, Ghodadra NS, et al. Recurrent shoulder instability: current concepts for evaluation and management of glenoid bone loss. J Bone Joint Surg Am. 2010;92(Suppl 2):133–51. 10.2106/JBJS.J.00906. [DOI] [PubMed] [Google Scholar]
- 20.Provencher MT, Detterline AJ, Ghodadra N, et al. Measurement of glenoid bone loss: a comparison of measurement error between 45 degrees and 0 degrees bone loss models and with different posterior arthroscopy portal locations. Am J Sports Med. 2008;36(6):1132–8. 10.1177/0363546508316041. [DOI] [PubMed] [Google Scholar]
- 21.Provencher MT, Ferrari MB, Sanchez G, Anavian J, Akamefula R, LeBus GF. Current Treatment Options for Glenohumeral Instability and Bone Loss: A Critical Analysis Review. JBJS Rev. 2017;5(7):e6. 10.2106/JBJS.RVW.16.00091. [DOI] [PubMed] [Google Scholar]
- 22.Rayes J, Xu J, Sparavalo S, Ma J, Jonah L, Wong I. Calculating glenoid bone loss based on glenoid height using ipsilateral three-dimensional computed tomography [published correction appears in Knee. Surg Sports Traumatol Arthrosc. 2023;31(1):177. 10.1007/s00167-022-07020-4. [DOI] [PubMed] [Google Scholar]
- 23.Rosenthal Y, Samim M, Gyftopoulos S, et al. 3D-MRI versus 3D-CT in the evaluation of glenoid deformity in glenohumeral arthritis using Dixon 3D FLASH sequence. Skeletal Radiol. 2022;51(12):2281–9. 10.1007/s00256-022-04086-6. [DOI] [PubMed] [Google Scholar]
- 24.Rossi LA, Frank RM, Wilke D et al. Evaluation and Management of Glenohumeral Instability With Associated Bone Loss: An Expert Consensus Statement Using the Modified Delphi Technique [published correction appears in Arthroscopy. 2025;41(2):532. 10.1016/j.arthro.2020.12.237 [DOI] [PubMed]
- 25.Sgroi M, Huzurudin H, Ludwig M, Zippelius T, Reichel H, Kappe T. MRI Allows Accurate Measurement of Glenoid Bone Loss. Clin Orthop Relat Res. 2022;480(9):1731–42. 10.1097/corr.0000000000002215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang KC, Jones A, Kambhampati S, et al. CT-Based 3D Printing of the Glenoid Prior to Shoulder Arthroplasty: Bony Morphology and Model Evaluation. J Digit Imaging. 2019;32(5):816–26. 10.1007/s10278-019-00177-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Yanke AB, Shin JJ, Pearson I, et al. Three-Dimensional Magnetic Resonance Imaging Quantification of Glenoid Bone Loss Is Equivalent to 3-Dimensional Computed Tomography Quantification: Cadaveric Study. Arthroscopy. 2017;33(4):709–15. 10.1016/j.arthro.2016.08.025. [DOI] [PubMed] [Google Scholar]
- 28.Zhang H, Zhu Y, Lu Y, Li F, Jiang C. Establishment of a True En Face View in the Evaluation of Glenoid Morphology for Treatment of Traumatic Anterior Shoulder Instability. Arthroscopy. 2020;36(3):668–79. 10.1016/j.arthro.2019.10.036. [DOI] [PubMed] [Google Scholar]
- 29.Zhou J, Zhong B, Qu R, et al. Anatomic measurement of osseous parameters of the glenoid. Sci Rep. 2022;12(1):13424. 10.1038/s41598-022-17783-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.