Practical considerations for determination of scapular internal rotation and its value in reverse total shoulder arthroplasty

doi:10.21203/rs.3.rs-2209106/v1

Download PDF

Research Article

Practical considerations for determination of scapular internal rotation and its value in reverse total shoulder arthroplasty

https://doi.org/10.21203/rs.3.rs-2209106/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 05 Apr, 2023

Read the published version in Journal of Orthopaedic Surgery and Research →

You are reading this latest preprint version

Background

Scapulothoracic orientation, especially scapular internal rotation (SIR) may influence range of motion in reverse total shoulder arthroplasty (RTSA) and is subjected to body posture. Clinical measurements of SIR rely on apical bony landmarks which are subjected to changes in scapulothoracic orientation, while radiographic measurements are often limited by the restricted field of view (FOV) in CT scans. Therefore, the goal of this study was (1) to determine whether the use of CT scans with a limited FOV to measure SIR is reliable and (2) if a clinical measurement could be a valuable alternative.

Methods

This anatomical study analyzed the whole-body CT-scans of 100 shoulders in 50 patients (32 male and 18 female) with a mean age of 61.2 ± 20.1 years (range 18; 91). (1) CT-scans were rendered into 3D models and SIR was determined as previously described. Results were compared to measurements taken in 2D CT-scans with a limited FOV. (2) Three apical bony landmarks were defined: (the angulus acromii (AA), the midpoint between the AA and the coracoid process tip (C) and the acromioclavicular (AC) joint. The scapular axis was determined connecting the trigonum scapulae with these landmarks and referenced to the glenoid center. The measurements were repeated with 0°, 10°, 20°, 30° and 40° anterior scapular tilt.

Results

Mean SIR was 44.8° ± 5.9° and 45.6° ± 6.6° in the 3D and 2D model, respectively (p < 0.371). Mean difference between the measurements was 0.8°± 2.5° with a maximum of 10.5°. A Bland-Altman-Plot showed 5% outliers. Midpoint AA/C showed no significant difference to the scapular axis at 0° (p = 0.203) as did the AC joint at 10° anterior scapular tilt (p = 0.949). All other points showed a significant difference from the scapular axis at all degrees of tilt.

Conclusion

2D CT-scans are reliable to determine SIR, even if the spine is not depicted. Position of the patient in the CT scanner needs to be considered and supine positioning presents a limitation. Clinical measurements using apical superficial scapula landmarks are a possible alternative, however anterior tilt influenced by posture alters measured SIR.

Posture Types

scapulothoracic orientation

reverse total shoulder arthroplasty

scapular internal rotation

Impaired internal and external rotation pose a challenge in reverse total shoulder arthroplasty (RTSA).(1–3) Scapulothoracic orientation has been shown to be an important factor influencing simulated range of motion (ROM) in RTSA.(2) A CT-based study showed that with increasing thoracic kyphosis the scapula protracts, internally rotates and anteriorly tilts relative to the body axes.(4) Based on these findings, a classification of shoulder arthroplasty patients into different Posture Types has been suggested. Patients are categorized from Type A with upright posture and retracted scapulae over average Type B to Type C with kyphotic posture with protracted and internally rotated scapulae.(4) (Fig. 1) Moroder et al.(2) further investigated the impact of those static scapula orientation changes on ROM in RTSA, using a modified arthroplasty planning software that accounts for scapulothoracic orientation.(2) Scapular internal rotation (SIR) was found to play a major role affecting ROM, especially rotational movement.(2, 4)

Although geometrical considerations of these studies seem valid, measurements were conducted based on 3D-CT models of the entire torso, which is neither practical nor feasible due to radiation exposure. In a clinical setting surgeons rely on plain-radiographic imaging, clinical examination and possibly CT-scans or MRI of the affected shoulder. For the determination of SIR, two obstacles need to be faced: First, the field of view (FOV) is often limited, and the spine not sufficiently depicted in the CT/MRI-scans, and therefore body axes cannot easily be determined. Furthermore, obtained values for scapulothoracic orientation might be affected by the supine position. Secondly, SIR is defined by the relation of the scapular axis (line from the trigonum scapulae to the center of the glenoid) to the transversal body axis. While clinically the trigonum scapulae can be accessed non-invasively through palpation, the center of the glenoid cannot be determined. Therefore, apical landmarks like the acromion, the coracoid process tip or the acromioclavicular (AC) joint are used to determine scapulothoracic orientation. (5–10) Considering, that with increasing thoracic kyphosis the scapula shifts anteriorly, subsequent anterior tilt around the scapular axis is seen which could possibly alter the measured SIR based on apical landmarks.

Therefore, the two main objectives of this study are 1) to determine the difference between measurements of SIR in 2D CT-scans without depicted spine and standard 3D measurements. And 2) to determine if a clinical measurement based on apical scapula landmarks could be a valuable alternative. We hypothesize that measurement of SIR on 2D CT-scans without depiction of the spine is reliable. We further hypothesize that the measurement of SIR in respect to the body axes is overestimated with progressive scapular tilt, regardless of the chosen apical landmark.

We performed a search of our institutional database for patients who had received a positron emission tomography (PET)–computed tomography (CT) backdated from January 2020 until we had identified 50 patients who met the following inclusion criteria: (1) age 18 years or older; (2) supine positioning of the patient with both arms at the side and elbows resting on the examination table; (3) complete depiction of the trunk from the base of the skull to the pelvis; and (4) sufficient CT quality for 3-dimensional rendering. Patients with visual pathologies of the upper extremities or thorax that potentially could alter scapular orientation, or scapulothoracic dimensions (e.g. fractures, prostheses, or dysplasia) were excluded. For each patient, both shoulders were analyzed as individual cases, which led to a total of 100 shoulders. CT imaging was performed with identical scan parameters and a primary slice thickness of 1.25 mm using a single scanner (Discovery MI; GE Healthcare, Chalfont St Giles, UK). The resulting study cohort consisted of 32 male and 18 female patients with a mean age of 61.2 ± 20.1 years (range 18; 91).

Measurement of scapula orientation

Whole body-CT scans were exported, anonymized, and rendered into 3D models using Horos software (Horosproject.org; Nimble Co LLC d/b/a Purview in Annapolis, MD USA). As previously described (4), SIR was defined as an angle between a perpendicular line to the best-fit sagittal axis (midpoint from the vertebra body Th1 and midpoint of the sternum) and a line from the medial root of the scapular spine (trigonum scapulae) to the center of the glenoid (Fig. 2A). Anterior scapular tilt was defined as an angle between the midpoint of the glenoid and the inferior angulus and a line drawn vertically along the fixed examination table (Fig. 2B) on a parasagittal view in a 3D model. For independent determination of 2D scapular internal rotation, whole body CT scans were cropped on an axial view leaving only the scapula and humerus from the acromion to the level of the inferior angulus and excluding visualization of the spine, to mimic a standard clinical CT-scan field of view. Internal scapular rotation was then measured at the level of the deepest glenoid concavity as the angle between the midpoint of the glenoid to the medial root of the scapular spine and the standardized CT reference line for the transversal axis. (Fig. 2C). In a sub analysis all patient (n = 5) who showed differences between 2D and 3D measurements outside the confidence interval were evaluated for thoracic scoliosis. A COBB angle was measured between the most tilted thoracic vertebrae above and below the apex of deformity.

Influence of anterior tilt on measurement of scapular internal rotation

To determine the influence of anterior scapular tilt on the clinical measurement of SIR through palpable bony landmarks, 3D models of each scapula were created, removing the thorax and humerus. First, the following landmarks were set to define the scapular plane: center of the glenoid, medial root of scapular spine (trigonum scapulae) and the inferior angulus. (Fig. 3A, B, C and D) Then bony landmarks were set, which could be used in a clinical examination: (1) the angulus acromii(11) (AA), (2) the most apical point of the posterior acromioclavicular (AC)- joint on the clavicular side(12), (3) and the most apical point located on a perpendicular line from the midpoint between the coracoid process tip (C) and the angulus acromii(13). To determine the scapular axis, the marks of the glenoid and the medial root of the scapular spine are aligned on a parasagittal view and the inferior angulus is set vertically with 0° angle to mimic a neutral state. The scapula is then tilted anteriorly around the scapular axis in 10°, 20°, 30° and 40°. (Fig. 4) On a transverse plane, a scapular axis is drawn and the angles to the AA, AC-joint and midpoint between AA and C are measured. This is repeated for each step of scapular tilt. (Fig. 5)

Statistical Analysis

For statistical analyses and descriptive data, we used SPSS Statistics Version 24.0 (IBM) software. All outcome variables were tested for normal distribution using the Kolmogorov-Smirnov test and showed a normal distribution. To compare 3D and 2D measurements of SIR an independent t-test was used. Furthermore, a linear regression was calculated, and a Bland-Altman-Plot created. Correlations between parameters were analyzed using the Pearson correlation coefficient (R). Sex differences were calculated by means of the independent t-test; differences between both shoulders, invariant analysis of variance. A p-value < 0.05 was considered significant.

Scapular internal rotation (SIR)

Mean SIR was 44.8° ± 5.9° and 45.6° ± 6.6° in the 3D and 2D model, respectively (p < 0.371). There was a linear regression between 3D and 2D measurements for SIR (R = 0.856; p < 0.001).

Mean difference between the measurements was 0.8°± 2.5° with a maximum of 10.5°. A Bland-Altman-Plot showed 5% outliers (± 1.96 SD). (Fig. 6) These 5 outliers were found in 4 patients. They showed a mean thoracic scoliosis of 18.8° ± 2.6° (range 11.1°; 17.2°).

Scapular tilt

Mean anterior scapular tilt was 20.9° ± 5.0° (range 12.7; 39.5). There was a positive correlation between SIR and scapular tilt (correlation coefficient R = 0.471; p < 0.001). There was no significant difference between sexes for SIR and scapular tilt.

Influence of anterior tilt on measurement of scapular internal rotation

Results for measurement of scapular internal rotation based on different landmarks (AA, AC- joint and midpoint AA/C) dependent on progressive anterior scapular tilt in 0°, 10°, 20°, 30° and 40° are shown in Fig. 7. Midpoint AA/C showed no significant difference to scapular axis a 0° (p = 0.203) and AC at 10° anterior tilt (p = 0.949), all other point showed a significant difference from the scapular axis at all degrees of tilt. There were no sex related differences for scapular tilt measurements with the landmarks of the AA and midpoint AA/C, but significant differences for the AC-joint, independent of the degree of tilt. We found no differences between both shoulders in any of the measurements.

Scapular internal rotation (SIR) has been recognized an important factor affecting shoulder movement in RTSA.(1, 2, 4) This study showed that even with a limited field of view in clinical CT-scans, SIR can be reliably measured. However, values need to be carefully considered due to supine position of the patient during examination. When using apical superficial landmarks to determine SIR in the upright standing position we showed that anterior scapular tilt needs to be considered, as progressive changes alter the measurements.

The position of the scapula relative to the torso is defined by three motions: scapula internal/external rotation, anterior/posterior tilt and upward/downward rotation, as well as combined forms with translational changes like protraction.(9) Rotational movement of the arm, following RTSA, is subjected to changes of scapular orientation in the transverse plane, meaning that an increase of SIR favors internal arm rotation and vice versa.(1, 2, 4) In a computer simulation study, body axes were used as a reference coordinate system to account for these changes.(2) It was shown that body posture influences the scapulothoracic orientation and therefore alters the simulated ROM. Patients with Posture Type C (poor posture) showed a lower simulated range of motion. However, some disadvantages of poor posture could be counteracted with a modification of the prosthetic components by means of a lower neck-shaft angle and higher retrotorsion of the humeral component, as well a larger or inferior eccentric glenosphere.(2)

Static SIR can be determined by measuring an angle between the scapular axis to the transversal body axis.(10, 14) Using CT examination, the glenoid center and medial border of the scapula can easily be identified to determine the scapular axis, but the spine is often not depicted in the image making it difficult to determine body axes. In this study it was shown that there was less than one degree mean difference in measured SIR between the established use of 3D whole body CT scans, and 2D measurements with a limited FOV. However, the maximum difference between 2D and 3D measurement for SIR was 10.5°. While revaluating those outlier cases we noticed a difference between both shoulders especially in the 3D measurements. We saw that there was a great discrepancy between the body axis marked by a line through the first thoracic vertebra and the sternum and the standard sagittal axis of the scanner. Therefore, we measured thoracic scoliosis in those patients and found mean values of 18.8°. It seems as if advanced thoracic scoliosis changes the alignment of the spine in a way that it potentially alters measured SIR for both sides. (Fig. 8). This phenomenon though is only seen in the 3D measurements, as in 2D the reference line is determined by the position in the scanner. Arguably we can infer that for the evaluation of possible ROM the 2D measurements show more clinically relevant data.

Even though measurements in this study were conducted using CT scans, it can be assumed, that this method can be translated to MRI which typically also offers a limited FOV and comparable bony landmarks identification.(15)

While the transversal body axis can be easily determined clinically, the measurement of the scapular axis is limited, because the center of the glenoid cannot be assessed and therefore the bony landmarks used for CT- measurements cannot be utilized. However, superficial apical landmarks can be used as surrogates to analyze changes in scapulothoracic orientation and evaluate scapulohumeral rhythm.(16, 17) Due to skin shifting during movement invasive procedures like pin insertion into bony landmarks seem to be most accurate but certainly practically not feasible, which is why less-invasive procedures like motion sensors, biplane fluoroscopy and simple goniometer measurements are being considered. (5–9) All in common, superficial scapular landmarks need to be defined.(10) A study by Ludewig et al.(7) investigated historical and current local scapular coordinate systems. The angulus acromii (AA) is a commonly used landmark to reference scapular orientation.(10, 11) Generally it is easily palpable even with thicker overlaying soft tissue and therefore suitable for detection of changes in scapular movement. However, our study shows that SIR is underestimated by around 19° when scapular tilt is at 0°. While they AA can be easily identified, the AC- joint and the midpoint between the coracoid process tip and the angulus acromii were described more reliable to match the center of the glenoid.(12, 13) The midpoint between the AA and the C was anatomically the landmark closest to the center of the glenoid when scapular tilt was at 0°. This is an interesting finding for motion sensor technology. By using this point, the trigonum scapulae and the inferior scapular angle, a scapular plane could be reconstructed. The AC-joint, on the other hand was closest to the scapular axis at around 10° of anterior tilt, with only about 5° variability between 10° and 20° anterior scapular tilt, which means that it would be suitable to measure SIR in Type A and B patients. However, in Type C patients with progressive anterior scapular tilt, scapular internal rotation would be highly overestimated with both landmarks but could still be underestimated with AA.

This study has some limitations. As described before, CT scans for analysis of SIR and anterior tilt were performed in the supine position, which could alter static orientation. Matsumura et al. investigated the difference between scapulothoracic orientation in supine and standing position using CT-scans.(18) Interestingly, they found significantly less upward rotation, internal rotation and anterior tilt of the scapula in the standing position, although it was suspected that gravity led to an increase in internal rotation and anterior tilt. Nevertheless, this study underlines the need of an investigation of scapulothoracic orientation in the standing or sitting position. Although not seen in our study cohort some patients might not be in a straight position on the examination table, which could alter the 2D measurement of SIR. We believe that this error can be avoided by paying attention to patient positioning in the gantry. Landmarks were set on 3D-CT models directly on the bone and soft tissue was not considered.

Clinical and radiographic determination of scapula internal rotation remains challenging. 2D CT-scans are reliable to determine scapular internal rotation, even if the spine is not depicted, however, position of the patient and scoliosis needs to be considered and supine positioning presents a limitation. Clinical measurement using apical scapula landmarks is possible however anterior tilt possibly alters measured SIR with progressive changes in scapulothoracic orientation. In athletic and average patients with low anterior scapular tilt (Posture Type A and B) the AC-joint is the landmark most reliable to match the scapular axis.

AA	Angulus acromii
AC	Acromioclavicular
C	Coracoid process tip
FOV	Field of view
ROM	Range of motion
RTSA	Revere total shoulder arthroplasty
SIR	Scapular internal rotation

Ethical Approval

The local ethics committee approved this study prior to investigation (EA4/119/19). A consent for publication is not applicable.

Competing interests

PM is a consultant for Arthrex. All other authors declare that they have no competing interests.

Authors' contributions

All listed authors have contributed substantially to this work: P.S., P.M. and D.A. for the study conception and design; P.S., D.M. and A.P.O. for the data collection; P.S., D.M. and G.K. for the data analysis; P.S., D.M. and C.A. for the data interpretation; P.S., J.G.H., G.K. and D.M. for the first drafting of the manuscript, the figures, and the literature research. All authors contributed to the writing and critical reading of the final manuscript and approved it for submission.

Funding

There was no funding for this study.

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Karelse AT, Bhatia DN, De Wilde LF. Prosthetic component relationship of the reverse Delta III total shoulder prosthesis in the transverse plane of the body. J Shoulder Elbow Surg. 2008;17(4):602–7.
Moroder P, Urvoy M, Raiss P, Werthel JD, Akgun D, Chaoui J, et al. Patient Posture Affects Simulated ROM in Reverse Total Shoulder Arthroplasty: A Modeling Study Using Preoperative Planning Software. Clin Orthop Relat Res. 2022;480(3):619–31.
Stephenson DR, Oh JH, McGarry MH, Rick Hatch GF, Lee TQ. Effect of humeral component version on impingement in reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2011;20(4):652–8.
Moroder P, Akgun D, Plachel F, Baur ADJ, Siegert P. The influence of posture and scapulothoracic orientation on the choice of humeral component retrotorsion in reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2020.
Lempereur M, Brochard S, Leboeuf F, Remy-Neris O. Validity and reliability of 3D marker based scapular motion analysis: a systematic review. J Biomech. 2014;47(10):2219–30.
Borstad JD, Ludewig PM. Comparison of scapular kinematics between elevation and lowering of the arm in the scapular plane. Clin Biomech (Bristol, Avon). 2002;17(9–10):650–9.
Ludewig PM, Hassett DR, Laprade RF, Camargo PR, Braman JP. Comparison of scapular local coordinate systems. Clin Biomech (Bristol, Avon). 2010;25(5):415–21.
Ludewig PM, Phadke V, Braman JP, Hassett DR, Cieminski CJ, LaPrade RF. Motion of the shoulder complex during multiplanar humeral elevation. J Bone Joint Surg Am. 2009;91(2):378–89.
McClure PW, Michener LA, Sennett BJ, Karduna AR. Direct 3-dimensional measurement of scapular kinematics during dynamic movements in vivo. J Shoulder Elbow Surg. 2001;10(3):269–77.
Wu G, van der Helm FC, Veeger HE, Makhsous M, Van Roy P, Anglin C, et al. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion–Part II: shoulder, elbow, wrist and hand. J Biomech. 2005;38(5):981–92.
van der Helm FC, Pronk GM. Three-dimensional recording and description of motions of the shoulder mechanism. J Biomech Eng. 1995;117(1):27–40.
de Groot JH. The variability of shoulder motions recorded by means of palpation. Clin Biomech (Bristol, Avon). 1997;12(7–8):461–72.
Pearl ML, Jackins S, Lippitt SB, Sidles JA, Matsen FA, 3rd. Humeroscapular positions in a shoulder range-of-motion-examination. J Shoulder Elbow Surg. 1992;1(6):296–305.
Damas CN, Silva J, Sa MC, Torres J. Computed tomography morphological analysis of the scapula and its implications in shoulder arthroplasty. Eur J Orthop Surg Traumatol. 2016;26(2):127–32.
Graichen H, Stammberger T, Bonel H, Haubner M, Englmeier KH, Reiser M, et al. Magnetic resonance-based motion analysis of the shoulder during elevation. Clin Orthop Relat Res. 2000(370):154–63.
Matsuki K, Hoshika S, Ueda Y, Tokai M, Takahashi N, Sugaya H, et al. Three-dimensional kinematics of reverse shoulder arthroplasty: a comparison between shoulders with good or poor elevation. JSES Int. 2021;5(3):353–9.
Walker D, Matsuki K, Struk AM, Wright TW, Banks SA. Scapulohumeral rhythm in shoulders with reverse shoulder arthroplasty. J Shoulder Elbow Surg. 2015;24(7):1129–34.
Matsumura N, Yamada Y, Oki S, Yoshida Y, Yokoyama Y, Yamada M, et al. Three-dimensional alignment changes of the shoulder girdle between the supine and standing positions. J Orthop Surg Res. 2020;15(1):411.

No competing interests reported.

Download PDF

Journal Publication

published 05 Apr, 2023

Read the published version in Journal of Orthopaedic Surgery and Research →

Editorial decision: Major revision
22 Nov, 2022
Reviews received at journal
14 Nov, 2022
Reviewers agreed at journal
03 Nov, 2022
Reviewers invited by journal
03 Nov, 2022
Editor assigned by journal
03 Nov, 2022
Submission checks completed at journal
03 Nov, 2022
First submitted to journal
27 Oct, 2022

You are reading this latest preprint version

Practical considerations for determination of scapular internal rotation and its value in reverse total shoulder arthroplasty

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Methods

Measurement of scapula orientation

Influence of anterior tilt on measurement of scapular internal rotation

Statistical Analysis

Results

Scapular internal rotation (SIR)

Scapular tilt

Influence of anterior tilt on measurement of scapular internal rotation

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1