Introduction

Understanding the morphological features of the proximal humerus is essential not only for the anatomical design of prosthetics in arthroplasty but also for enhancing preoperative planning to improve clinical outcomes for patients. In clinical practice, accurately measuring the humeral head diameter (HHD), posterior offset (PO), and retroversion angle is crucial for successful total shoulder arthroplasty1. Similarly, accurately determining the shoulder acromiohumeral contact surface (AHCS) arc length is crucial for managing tendon-graft length in superior capsule reconstruction2. The performance of postoperative shoulder functionality may be compromised by the inadequate assessment of the AHCS arc length of the proximal humerus. Despite anatomical deviations in the humeral head between females and males, there has been limited research on sex-specific differences in AHCS arc length3.

Measurements of the AHCS arc length may be biased by different surgeons during superior capsule reconstruction surgery. To accurately investigate the morphometric features of the AHCS arc length, measurements should be conducted at a shoulder abduction angle of 20–30°4. Tibone et al.. found that the AHCS arc length measured at 20° or 40° of glenohumeral abduction angle resulted in reduced postoperative subacromial pressure5. However, little is known about the correlation between shoulder abduction and the AHCS arc length in females and males6.

The objectives of this study were to evaluate sex differences in the AHCS arc length measurements of the shoulder and to characterize their correlation with shoulder abduction as assessed through three-dimensional (3D) computed tomography (CT) imaging.

Materials and methods

Ethics approval

All procedures involving human participants and/or human medical records were conducted in accordance with the ethical standards set forth by the local institutional ethics committee, as well as the 1975 Helsinki Declaration and its subsequent amendments or comparable ethical guidelines. The protocol received prior review and approval from the local Ethics and Research Committees.

Study design and patients

This retrospective observational study analyzed upper limb CT images obtained from January 2011 to January 2021 at the radiology department of an academic medical center in China. Eligible participants included all examinations conducted through the Picture Archiving and Communication System (PACS) during this period. Exclusion criteria encompassed patients with shoulder fractures, shoulder osteoarthritis, rotator cuff tears, shoulder deformities, shoulder instability, scoliosis, upper limb bone tumors, immunological disorders, and poor-quality imaging. Notably, patients aged under 20 years or over 49 years were also excluded. Additionally, images which were unable to perform shoulder abduction angle measurements (defined as the angle between the axis of the proximal humerus and the line of the spinous process of the spine) were excluded (Fig. 1). Following the application of these inclusion and exclusion criteria, a total of 169 cases from the normal Chinese population were included in the final analysis. All images were reviewed by two senior orthopedic surgeons, with independent examination results documented. Any discrepancies in interpretation were resolved by consensus. The study cohort was categorized into three interval of shoulder abduction angle groups (Table 1).

Fig. 1
figure 1

Flowchart diagram of the study. A total of 169 patients were included in the final analysis.

Table 1 The number of observed shoulders categorized by abduction angle.

CT analysis parameters, methods of measurement and definition of shoulder AHCS arc length

Shoulder CT scans were performed using a 64-row CT scanner (Aquilion; Toshiba Medical Systems Corporation, Otawara, Japan). The CT data were stored in the Digital Imaging and Communications in Medicine (DICOM) format. The shoulder AHCS was measured using 3-Matic software (version 12.0; Materialise, Leuven, Belgium)6,7. We first imported three-dimensional surface models of the scapula and humerus into the 3-Matic software and selected six landmarks located around the articular surfaces to define the anatomical neck plane (the plane that best fits the selected landmarks). The epiphyseal sphere was defined as the sphere that best fits the articular surface, and the humeral head diameter (HHD) was defined as the diameter of this epiphyseal sphere. A coordinate system was then established based on the metaphyseal cylinder (the cylinder that best fits the proximal part of the humeral shaft). The proximal humeral shaft axis was designated as the inferior-superior axis (X-axis), the anterior-posterior axis (Z-axis) was defined as the axis on the plane of the anatomical neck perpendicular to the X-axis, and the mediolateral axis (Y-axis) was defined as the axis perpendicular to the X-Z plane. Finally, an arc was drawn in the X-Y plane connecting the superior glenoid tubercle, the superior vertex of the humeral head surface, and the lateral edge of the greater tuberosity, with the length of this arc defined as the shoulder AHCS arc length (Fig. 2). Additionally, we measured the acromiohumeral distance (AHD) as a reference value for established anatomical parameters of the shoulder joint between females and males7. AHD was used in relation to acomiohumeral distance as opposed to AHCS arc length. The AHD was measured using the same method as reported in previous studies8.

All imaging parameters were measured independently and in a standardized manner by two senior orthopedic surgeons. After a two-week interval from the initial measurement, both investigators conducted independent re-measurements using the same procedures and techniques to evaluate the reliability of all parameters.

Fig. 2
figure 2

Creation of the coordinate system for representative 3D surface models and measurement of AHCS arc length. (a) The plane of the anatomical neck was established using six landmarks located along the perimeter of the articular surface. The epiphyseal sphere is defined as the sphere that best fits the articular surface (center, O) and has a diameter equivalent to the humeral head diameter (HHD). (b) The coordinate system was defined based on the axis of the proximal humeral shaft and the plane of the anatomical neck. (c) The shoulder AHCS arc length, defined as the distance from the superior glenoid tubercle to the lateral edge of the greater tuberosity (DSGT), was measured in the X-Y plane (blue arc). (Not to scale).

Data process and statistical analysis

The data for both males and females were found to be non-normally distributed according to the Shapiro-Wilk test (p > 0.05). The Mann-Whitney U test was employed to evaluate the differences in AHCS arc length at the same abduction angle between sexes, as well as the differences across various abduction angles within the same sex. Pearson’s correlation coefficient and linear regression analyses were conducted to assess the relationship between shoulder AHCS arc length and shoulder abduction angles. Intra- and interobserver reliability were determined using intraclass correlation coefficients (ICC), with a 95% confidence interval. Results are presented as mean ± standard deviation. Statistical analyses were performed using SPSS Statistics version 22.0.0.0 (IBM Corp., Armonk, NY, USA), with statistical significance set at p < 0.05.

Results

Study population

This study included 169 participants, with a median age of 35.02 [26.29–43.75] (males: 33.42 ± 8.60 years, females: 37.46 ± 8.42 years, comprising 67 females (39.64%). The final analysis included a total of 77 left shoulders (45.56%) and 92 right shoulders (54.44%) (Table 2). The humeral head diameter (HHD) of the participants was measured as 43.67 ± 3.27 mm [38.40–50.50 mm] (males: 45.30 ± 2.87 mm, females: 41.07 ± 1.89 mm). The shoulder abduction angle was 8.52 ± 8. 18° [0.02–33.62] (males: 7.91 ± 7.83°, females: 9.44 ± 8.66°). The shoulder abduction group (0–10] ° accounted for 71.01% of the total participants (Table 1). The intra-observer reliability, as measured by the intraclass correlation coefficient (ICC), ranged from 0.896 to 0.971, indicating excellent reliability, while the inter-observer reliability ranged from 0.954 to 0.984 (Table 3).

Table 2 Characteristics of the study population.

Sex differences of the AHCS arc length

The mean reference value for the AHCS arc length was 45.09 ± 4.20 mm. We found no statistically significant difference in AHCS arc length between the left (45.10 ± 4.31 mm) and right (45.07 ± 4.13 mm) shoulders. However, the AHCS arc length was significantly greater in males (46.99 ± 3.69 mm) compared to females (42.19 ± 3.15 mm) (p < 0.001). Notably, significant differences were observed between females and males across all three abduction angle groups (all p < 0.01) (Fig. 3; Table 3). Additionally, the acromiohumeral distance (AHD) was assessed to provide reference values for established acromiohumeral anatomical measurements of the shoulder joint. The AHD was significantly greater in males (7.85 ± 0.95 mm) compared to females (7.40 ± 0.85 mm) (p = 0.016) in the abduction group (0–10].

Fig. 3
figure 3

Statistical analysis between the abduction angle and the AHCS arc length. The letters ‘a’ and ‘b’ denote the marking methods used for comparisons among different abduction angle groups. An asterisk (*) indicates a significant difference in comparisons within the female group, with p < 0.05.

Table 3 Sex differences in the AHCS arc length across three abduction groups.

The correlation between abduction angle and AHCS arc length

For subgroups analysis, a negative linear correlation was observed between the abduction angle and the AHCS arc length in both males (R2 = 0.436, AHCS arc length = 49.450 − 0.312 × abduction angle, p < 0.001) and females (R2 = 0.434, AHCS arc length = 44.457 − 0.240 × abduction angle, p < 0.001) (Fig. 4). In addition, a weak positive correlation (R² = 0.144, p = 0.020) was observed between the AHCS arc length and the humeral head diameter in females. However, this correlation was not present in males.

Fig. 4
figure 4

Linear regression analyses of the abduction angles and the AHCS arc length. Line-regression analysis was conducted to illustrate the relationship between abduction angles and AHCS arc length in both sexes.

Discussion

Accurate understanding of the anatomical acromiohumeral contact surface (AHCS) arc length is critical for diagnosing shoulder disorders, planning surgical interventions, and developing rehabilitation protocols aimed at restoring normal shoulder function9. Abnormalities or changes in AHCS arc length may indicate shoulder pathologies such as impingement syndrome, rotator cuff tears, or osteoarthritis10. However, little is known about sex differences in AHCS arc length. In this study, we performed 3D reconstructions of 169 CT images from healthy Chinese participants and measured anatomical parameters to establish reference values for the AHCS arc length (Fig. 5). Our findings revealed that the AHCS arc length was significantly shorter in females compared to males. Additionally, the AHCS arc length was negatively correlated with the abduction angle.

Fig. 5
figure 5

The study flowchart illustrates sex differences in AHCS arc length, which may inform future preoperative planning for shoulder surgeries.

Research has demonstrated sex-specific differences in musculoskeletal anatomy11. The acromion tends to be smaller and more hooked in females compared to males, potentially influencing the AHCS arc length and the subacromial space12. From an evolutionary perspective, males may have developed larger, stronger shoulders to support physically demanding tasks, while females may have evolved greater flexibility and range of motion for activities requiring dexterity. Males typically have a larger humeral head, which can result in variations in how the humerus interacts with the acromion. In our study, the AHCS arc length was significantly greater in men (46.99 ± 3.69 mm) than in women (42.19 ± 3.15 mm) (p< 0.001). This finding suggests that, in general, females may have a narrower subacromial space13. The reduced space could impact the arc length and contact surface during shoulder movements, potentially contributing to a higher incidence of impingement syndromes. Presumably if the subacromial space is narrower and the humerus lies superior then the AHCS arc length will be larger. Additionally, females typically exhibit greater shoulder flexibility and range of motion, which may further influencing the dynamics of the AHCS arc length during various activities14.

Biomechanical analyses suggest that males generally possess greater muscle mass and strength around the shoulder, which may influencing joint stability and the forces exerted on the AHCS15. These differences in muscle strength and mechanics can alter loading patterns, thereby affecting the AHCS. Another important factor is the effect of sex hormones. Estrogen may play a role in the development and maintenance of musculoskeletal tissues. Hormonal differences can impact bone density, ligament laxity, and muscle composition, contributing to variations in shoulder anatomy and biomechanics16.

Understanding sex differences is essential for developing personalized treatment plans. For instance, surgical techniques in superior capsule reconstruction may need to be tailored to account for anatomical variations. Notably, no geographically related differences were observed in the AHCS values (Table 4). Lee et al. and Cline et al. measured the AHCS arc length (at 30° shoulder abduction) on the surgical side in Korean and American patients, respectively17,18. No statistical differences were found between the AHCS arc lengths. Interestingly, the reported AHCS arc lengths in previous studies were similar to those in the current study at 0°, 15°, and 30° shoulder abduction angles, with variations ranging from 3.25 mm to 4.65 mm. Our results indicate that the reference value for AHCS arc length is linearly and negatively correlated with the abduction angle in both sexes, suggesting that preoperative planning should select the appropriate abduction angle for measuring the AHCS arc length19. Interestingly, our results revealed a weak positive correlation (R² = 0.144, p= 0.020) between the reference AHCS arc length and humeral head diameter in female participants. However, this correlation was not observed in males, suggesting that skeletal differences between male and female specimens may not directly account for the variation in AHCS arc length. Additionally, Pennington et al. reported that the AHCS arc length in patients undergoing superior capsule reconstruction was significantly smaller than before surgery20. Therefore, we hypothesize that the difference in AHCS arc length pre- and postoperatively may be attributed to the inferior displacement of the humeral head following surgery21.

Table 4 Summary of previous studies reporting intraoperative AHCS arc length and preoperative AHD measurements in superior capsule reconstruction.

We further analyzed AHD as a reference value for established acromiohumeral anatomical measurements in the shoulder joint between females and males. Notably, previous studies have suggested that thicker tendon grafts in superior capsule reconstruction patients may result in favorable outcomes24. However, Baek et al. found that using a tendon graft thicker than a patient’s preoperative AHD may increase the risk of graft retear and subacromial erosion post-surgery25. As the shoulder abduction angle increases from 0° to 34°, the AHD progressively decreases, consistent with the findings of Giphart et al.26. This may be attributed to the continuous change in the coordinates of the closest distance between the humeral head and the acromion, with muscle dynamics also playing a role. Overall, the similarity between AHD values reported in previous studies and our findings suggests that our measurement methods are reliable for further application.

This study has several limitations. First, we reported AHCS arc length measurements from a relatively small sample of 169 patients, which may not fully capture the sex differences across different age groups. However, it is worth noting that the sample included both male and female participants from a wide age range, indicating reasonable representation. Additionally, the unequal age distribution between male and female participants may have introduced bias, potentially influencing the interpretation of the retrospective findings. Considering the sex-based variances in bone structure may be the contributing factor to the difference in AHCS arc length, personal measurements maybe have a better predictive value of AHCS arc length than adjusting sex alone. Future studies should aim to include larger sample sizes and a more diverse population to improve the reliability and validity of the results. Finally, since the participants were recruited from a single academic medical center, further multi-institutional studies are needed to confirm these findings.

Conclusions

In summary, our study found that women had a shorter AHCS arc length compared to men across all three abduction angle intervals. Additionally, the reference values for AHCS arc length were linearly and negatively correlated with the abduction angle in both sexes. These findings support the need for sex-specific standardization of AHCS arc length measurements, which is crucial for individualized preoperative planning in shoulder joint surgery.