Skip to main content
Download PDF
  • Research article
  • Open access
  • Published:

Investigating the reliability and validity of subacromial space measurements using ultrasound and MRI

Journal of Orthopaedic Surgery and Research volume 18, Article number: 986 (2023) Cite this article

Abstract

Background

While ultrasound (US) measures of the subacromial space (SAS) have demonstrated excellent reliability, measurements are typically captured by experts with extensive ultrasound experience. Further, the agreement between US measured SAS width and other imaging modalities has not been explored. This research evaluated the agreement between SAS measures captured by novice and expert raters and between US and magnetic resonance imaging (MRI). This study also evaluated the effect of US transducer tilt on measured SAS.

Methods

Nine men and nine women participated in this study. US images were captured by a novice and expert with the participant in both seated and supine positions. An inclinometer was fixed to the US probe to measure transducer tilt. SAS width was measured in real time from freeze framed images. MRI images were captured, and the humerus and acromion manually segmented. The SAS width was measured using a custom algorithm.

Results

Intraclass correlation coefficients (ICCs) between novice and expert raters were 0.74 and 0.63 for seated and supine positions, respectively. Intra-rater agreement was high for both novice (ICC = 0.83–0.84) and expert (ICC ≥ 0.94) raters. Agreement between US and MRI was poor (ICC = 0.21–0.49) but linearly related.

Conclusions

Moderate agreement between novice and expert raters was demonstrated, while the agreement between US and MRI was poor. High intra-rater reliability within each rater suggests that US measures of the SAS may be completed by a novice with introductory training.

Background

The subacromial space (SAS) is comprised of the humeral head inferiorly, and the anteroinferior surface of the anterior third of the acromion, the coracoacromial ligament and acromioclavicular joint superiorly [1]. Reduction of the SAS can lead to subacromial impingement syndrome (SAIS), whereby the tissues occupying the space, notably the supraspinatus tendon, become compressed and subsequently damaged [2]. SAIS is one of the most common shoulder disorders, accounting for 48% of all clinical diagnoses [3]. Narrowing of the SAS can result from musculoskeletal exposures that modify healthy joint kinematics [2, 4,5,6] or innate morphological parameters [7,8,9]. Both mechanisms pose risk of SAIS and/or rotator cuff tears and subsequent pain, functional limitations, and further tissue injury [10,11,12]. Thus, accurate measurement of the SAS width is beneficial for risk identification and prevention.

A variety of imaging-based methods have been used to quantify the SAS. Radiography has been used to capture the acromiohumeral interval (minimum SAS width), demonstrating high reliability [8, 13,14,15,16]. However, changes in arm position and X-ray beam orientation have been shown to overestimate SAS measurements [14, 17]. While radiography captures a single static image, biplane fluoroscopy enables investigators to view the structures dynamically across simultaneous images [18]. This method is particularly advantageous as it offers the same benefits as conventional radiography, while allowing for three-dimensional measurement of the SAS in both static and dynamic states [18, 19]. Researchers investigating the SAS have also used computed tomography (CT) scans, which yields greater precision compared to conventional radiography [20, 21]. Bey and colleagues [19, 22,23,24] developed a model-based motion tracking technique that uses CT scans alongside biplane radiographs to track in vivo glenohumeral kinematics with a high degree of accuracy compared to dynamic radiostereometric analysis (RMS errors < 0.385 mm for the scapula and < 0.374 mm for the humerus) [24]. This method has been subsequently adopted by other research groups and deemed the gold standard for quantifying the SAS [25]. However, while deemed noninvasive, these techniques still expose participants to radiation, albeit the effective dosage is reportedly low [25]. Magnetic resonance imaging (MRI) has been regarded as the most appropriate imaging technique for complex anatomical structures [26,27,28]. Research using MRI to capture the SAS width reports high precision, reliability, and accuracy of the anatomical models reconstructed from MRI images [27, 29]. While these methods have been deemed effective for measuring the SAS, they are not readily accessible and can be expensive. Ultrasound (US) is another imaging modality that has been used to measure the SAS. US is efficient, noninvasive, relatively easy to administer, and inexpensive. However, US may overestimate the minimum SAS width, as the lateral positioning of the probe may capture the distance between the acromion and lateral humerus rather than the minimum distance between the coracoacromial arch and the superior-most point on the humeral head.

Several studies have investigated the inter- and intra-rater reliability of SAS measures using US. Among studies where the agreement in SAS measures between experienced clinicians was quantified, excellent intra- (ICC: ≥ 0.92) and inter- (ICC: ≥ 0.88) rater reliability was reported [30,31,32]. In these studies, participants included asymptomatic individuals, and patients with SAIS and/or rotator cuff tendinopathy. Further, excellent within-day (ICC ≥ 0.98) and day-to-day (ICC ≥ 0.96) intra-rater reliability has been reported for expert raters [33]. Other studies have compared the agreement of SAS measurements captured by both novice and expert raters. Novice raters included physical therapists, physiotherapy students, and orthopedic residents who received practice and/or training prior to commencing the study [34,35,36,37]. These studies revealed that the novice raters demonstrated good to excellent inter-rater reliability among one another (ICC: 0.79) [35] and compared to an expert (0.70–0.77) [34, 36, 37]. Intra-rater reliability for the novice rater was reported to be excellent (ICC ≥ 0.84) [35, 37]. SAS measurement agreement between expert clinicians and non-clinician novice raters, as well as the agreement between their US measures and that quantified from MRI, has not been studied.

In an effort to improve reliability of US-based measures, researchers have implemented various tools and methodologies. Bulbrook et al. [38] introduced a 3D-printed novel transducer attachment to improve the reliability of capturing muscle architecture parameters in the thigh. In the shoulder, an inclinometer has been attached to the US probe in effort to assist in maintaining a consistent probe angle between repeated images [39]. However, researchers have not studied whether SAS measures captured by two raters are in better agreement when the same transducer angle is adopted. Although measurements performed by expert raters have previously demonstrated excellent reliability, minimal research exists that examines the validity of SAS compared to other imaging modalities and the sensitivity to transducer tilt variation. It is also important to determine whether a novice non-clinician rater with minimal US training can reliably and accurately measure the SAS. Confirmed reliability of a novice researcher, as compared to an expert clinician, and validity, as compared to MRI, would enable more feasible and accessible laboratory-based assessments of SAIS risk.

The primary purposes of this study were to: (1) evaluate the agreement of SAS measures captured using US between a novice with introductory training and an expert and (2) evaluate the agreement between US and MRI measures of the SAS. We hypothesized that the SAS measurements of novice and expert raters would have moderate to excellent agreement. We also expected the US measures to overestimate the minimum SAS captured from MRI due to the imaging capabilities of US. The secondary purpose of this research was to determine whether inter-rater reliability was improved when the transducer tilt angle was consistent between raters. We hypothesized that inter-rater reliability would be improved when a consistent transducer tilt angle was adopted.

Methods

Participants

Eighteen right-hand dominant, healthy young adults (9 M, 9W; 23.6 (2.6) years; 24.9 (3.4) kg/m2) free from self-reported shoulder pain or injury in their right shoulder were recruited from the university population. Participants were excluded if they had positive signs for any of three clinical impingement tests (Hawkins-Kennedy, Painful Arc Sign and/or Infraspinatus Muscle Strength Tests) [40]. In addition, participants were excluded if they had a Disabilities of the Arm, Shoulder, and Hand (DASH) score > 25 [41]. Participants were also excluded if it was deemed unsafe for them to enter the MRI room; this included having any metal within their body (i.e., implanted clips or devices) or were pregnant. This study received ethical approval through the institutional research ethics board (York University Office of Research Ethics #e2019-387).

MRI: acquisition and analysis

MRI was acquired by a certified MRI technologist using a 3 T Siemens Tim Trio scanner (Siemens Healthineers, Erlangen, Germany) with a Siemens 18 channel body coil. Participants were lying in a supine position, while coronal images of their right shoulder were captured using a T1 VIBE sequence (FOV = 200 mm, slice thickness = 1 mm, matrix = 256 × 256, voxel size = 0.8 × 0.8x1mm, TR/TE = 11 ms/4.81 ms, flip angle = 10 degrees, average = 1, time = 340 s) (Fig. 1). The humerus and acromion were manually segmented using open-source software (3D Slicer Image Computing Platform, https://www.slicer.org/) [42]. To measure the subacromial space, segmentation post-processing was performed using pyMSKT (https://github.com/gattia/pymskt). Bone surfaces were created by applying a Gaussian filter (σ2 = 1.56 mm) to binary bone segmentation masks and surfaces extracted using Marching Cubes [43, 44]. Bone surfaces were resampled to 10,000 points using Voronoi clustering [45]. For each vertex on the acromion surface a vector was projected 50 mm normal to the surface, if the vector intersected the humerus the Euclidean distance from the originating point to the intersection was calculated and recorded at the respective vertex. Heatmaps of continuous SAS distances are shown in Fig. 2. Two SAS measures were extracted. First, the mean distance of the 10% closest points (MRI_Hum10) (Fig. 2). Second, the minimum distance from the most lateral point on the acromion to the humerus was calculated as a measure more representative of US-captured SAS width (MRI_Lat) (Fig. 2).

Fig. 1
figure 1

MRI image of the right shoulder using a T1 VIBE sequence in the coronal plane

Full size image
Fig. 2
figure 2

Surface meshes of the humerus and acromion process with heat maps of the distance between the bone surfaces (darker = narrower SAS width; lighter = wider SAS width). The arrows represent the normal projection from the acromion to the closest humeral point (MRI_Hum10; medial-most arrow (green)) and the closest distance between the lateral acromion to the humerus (MRI_Lat; lateral-most arrow (blue))

Full size image

Ultrasound: training, capture, and outcome measurement

A high frequency (12 MHz) linear array US transducer (GE LOGIQ e, GE HealthCare, Chicago, IL, USA) was used to record US images (Depth, 5 cm; 1 focus point). Images were captured of the right shoulder with the participant in both sitting and supine positions. In the seated position, participants sat with their body erect, arm down by their side in neutral, with their elbow flexed 90°, such that the dorsal aspect of the hand was resting on the ipsilateral thigh [31]. Images were also captured with the participants lying supine with their arm resting in neutral, down by the side with the palm facing medially. Supine images were captured to more closely resemble positioning during MRI acquisition [12]. For each position (seated, supine), the transducer was placed at the most anterolateral aspect of the acromion, with the probe oriented along the long axis of the humerus and perpendicular to the skin [46]. Additionally, a digital inclinometer within a 3D-printed case was affixed to the transducer with Velcro straps [39] (Fig. 3). This permitted capturing the transducer tilt angle of both raters relative to vertical.

Fig. 3
figure 3

Digital inclinometer housed within a 3D-printed case affixed to the US transducer with Velcro straps

Full size image

US images were captured by one expert rater and one novice rater. The expert rater practiced as a sports medicine physician for 15 years and earned a doctoral degree in musculoskeletal biomechanics specific to the shoulder. The expert also had > 9 years of experience using US to image the shoulder. The novice rater was a graduate student investigator in the field of kinesiology/biomechanics with limited to no experience collecting US images. Prior to participant recruitment, the novice completed introductory US training specific to the shoulder, which included SAS measurements, supraspinatus tendon measurements and supraspinatus and infraspinatus muscle imaging. The training consisted of a written manual and video modules (participant positioning, transducer placement, outcome measurements) prepared by the expert rater and senior author, in conjunction with published research and guidelines. The novice carefully reviewed all written and video modules (~ 4.5 h, which included all tissues), following which 5 h of hands-on practice was completed across four non-consecutive days.

The raters each performed three repetitions of the SAS measurements with the participant in both seated and supine positions. In addition, the novice rater performed additional SAS measurements at the expert’s transducer angle. All trials were block randomized with raters blinded to each other’s measurements. The novice measurements captured at the expert’s transducer angle were always completed last. The SAS width was measured as the smallest distance between the inferior edge of the acromion and the superior edge of the humerus [46]. All measurements were captured in real time from freeze-framed images (Fig. 4).

Fig. 4
figure 4

Ultrasound image of the subacromial space. Crosses (+) show the superior (acromion) and inferior (humerus) landmarks of the subacromial space

Full size image

Statistical analysis

Data were analyzed across raters (novice, expert), imaging modalities (US, MRI), body positions (seated, supine), and MRI-based calculation methods (MRI_Hum10, MRI_Lat). The mean of the three repeated US measurements was used for analysis. Intraclass correlation coefficients (ICC) were used to assess absolute agreement of SAS measurements. Specifically, two-way random effects models (ICC (2,1)) assessed absolute agreement between (1) self-selected expert and novice measurements (seated, supine), (2) US and MRI measures (novice, expert; MRI_Hum10, MRI_Lat; seated, supine), and (3) self-selected expert and novice with expert angle (seated). Bland–Altman analyses were also conducted to evaluate agreement between raters and imaging modalities [47]. As well, eight linear regression models ([novice, expert] x [seated, supine] × [MRI_Hum10, MRI_Lat]) were used to determine whether US-based measurements (independent variable) accounted for significant variance in MRI-based measurements (dependent variable). ICCs (2,1) were also used to assess intra-rater reliability of repeated measurements within raters. ICCs were interpreted using classifications previously defined (< 0.50 = poor, 0.50–0.75 = moderate, 0.75–0.90 = good, > 0.90 = excellent) [48]. The magnitude of SAS differences between raters and imaging modalities was additionally calculated. Further, the post hoc power (p2) for ICCs between novice and expert raters (seated, supine) and between ultrasound and MRI-measured SAS was calculated using a two-tailed test with a null hypothesis of p = 0, raters (k) = 2, and α = 0.05 [49]. All statistical analyses were conducted using StataIC 16.0 (StataCorp LLC, College Station, TX), MATLAB (Mathworks, Natick, MA, USA) and R v4.3.2 [50].

Results

SAS: novice versus expert rater

The average SAS magnitudes across participants for the novice rater were 12.7 (2.2) mm and 12.4 (2.0) mm for seated and supine positions, respectively, and for the expert rater were 11.4 (2.0) mm and 11.0 (1.8) mm for seated and supine positions, respectively. SAS measurements captured by the novice rater generally overestimated those of the expert rater in both seated and supine positions by 1.3 (1.0) mm and 1.4 (1.3) mm, respectively (Fig. 5). The average absolute differences between the novice and expert were 1.3 (max, 3.0; min, 0.3) and 1.6 (max, 3.8; min, 0.3) for seated and supine, respectively. Bland–Altman analysis confirmed the systematic overestimation of SAS measurements by the novice rater (Fig. 6). ICCs ([95% CI], p2) between the novice and expert were moderate, with higher agreement in seated (0.74 [0–0.92], 0.97) compared to supine (0.63 [0–0.88], 0.86).

Fig. 5
figure 5

Box plots of the US SAS measures by novice and expert raters for both seated (left) and supine (right) body positions

Full size image
Fig. 6
figure 6

Bland–Altman plots to evaluate agreement between US SAS measurements for novice and expert raters for the seated (top, left) and supine (top, right) body positions, and of the expert (self-selected transducer angle) and novice (with expert transducer angle) in a seated body position (bottom, middle)

Full size image

Both novice and expert raters demonstrated good to excellent intra-rater reliability between the three repeated measurements for each condition. ICCs [95% CI] were generally higher for the expert in both seated (0.94 [0.87–0.97]) and supine (0.97 [0.93–0.99]), compared to the novice (seated, 0.83 [0.68–0.93]; supine, 0.84 [0.68–0.93]). The range of measurements about the participant-specific mean for the three repeated measures (absolute value – participant specific mean) was 0–2.3 mm and 0–2.60 mm for the novice rater in seated and supine, respectively, and 0–1.2 mm and 0–0.80 mm for the expert rater in seated and supine, respectively (Fig. 7). Further, both novice and expert raters demonstrated good agreement between their seated and supine measurements, with ICCs of 0.87 [0.68–0.95] and 0.78 [0.51–0.91] for novice and expert, respectively.

Fig. 7
figure 7

Difference (spread) of repeated ultrasound measurements from the participant-specific mean for each participant for the novice (top) and expert (bottom)

Full size image

Transducer angle variation

Self-selected transducer tilt angle between the novice and expert raters differed by an average of 9.3 (5.7)° for the seated position and 11.2 (7.1)° for the supine position. Inter-rater agreement was higher when the novice captured measurements using the expert’s mean participant-specific transducer angle (0.79 [0.48–0.92]) compared to their own independent angle (0.74 [0–0.92]) (Fig. 6). The mean difference between raters was 0.7 (1.3) mm.

Ultrasound and MRI agreement

The mean SAS measured from the MRI was 9.5 (1.5) mm for MRI_Hum10, and 10.6 (1.4) mm for MRI_Lat. The agreement between MRI methods was excellent (ICC, 0.98 [0.85–0.99]). Agreement between MRI and US measures was poor for both raters, body positions and MRI-based calculation methods (Table 1). Generally, the US measurements overestimated MRI (Table 1). Bland–Altman analyses revealed a systematic overestimation of MRI SAS magnitudes by US of 1.5–3.2 mm using the MRI_Hum10 method and 0.4–2.1 mm using the MRI_Lat method (Figs. 8, 9).

Table 1 Intraclass correlation coefficients [95% confidence interval], post hoc power (p2) and mean differences (standard deviation) between MRI and Ultrasound
Full size table
Fig. 8
figure 8

Bland–Altman plots to evaluate agreement between MRI_Hum10 and (1) novice US SAS measurement in seated body position (top, left), (2) novice US SAS measurement in supine body position (top, middle), (3) novice US SAS measurement with expert transducer angle in seated body position (top, right), (4) expert US SAS measurement in seated body position (bottom, left), (5) expert US SAS measurement in supine body position (bottom, right)

Full size image
Fig. 9
figure 9

Bland–Altman plots to evaluate agreement between MRI_Lat and (1) novice US SAS measurement in seated body position (top, left), (2) novice US SAS measurement in supine body position (top, middle), (3) novice US SAS measurement with expert transducer angle in seated body position (top, right), (4) expert US SAS measurement in seated body position (bottom, left), (5) expert US SAS measurement in supine body position (bottom, right)

Full size image

Linear regression between ultrasound and MRI

Regression models generally showed US measures of the SAS to be significant linear predictors of MRI (Table 2). The model that accounted for the most variance (R2 = 0.48) was the expert rater, with a supine body position and an MRI_Hum10 calculation method.

Table 2 Regression analysis results (Prob > F, R2) and regression equation for each model (rater, body position, MRI-based calculation method)
Full size table

Discussion

Moderate agreement was demonstrated between novice and expert raters with the measurements of the novice rater generally overestimating that of the expert. As expected, adopting a consistent transducer tilt angle improved the inter-rater reliability, however, the improvement was small in magnitude. Further, SAS captured using ultrasound overestimated the SAS width measured from MRI. A significant linear relationship was revealed between US and MRI for most measurement conditions, suggesting that US systematically overestimated MRI (0.4–3.2 mm) and prediction models can be generated.

Novice versus expert SAS measurements

This research evaluated the agreement in SAS measures captured both within and between novice and expert raters using ultrasound. Moderate agreement was demonstrated between raters with higher agreement shown for measurements captured in the standard seated position (ICC = 0.74) compared to supine (ICC = 0.63). The findings of this study are comparable to previous research. Cavaggion et al. [34] found moderate to good inter-rater reliability in both healthy participants (ICC = 0.77) and those with subacromial shoulder pain (ICC = 0.61). The novice rater was a physical therapist with 3 years of research US experience, while the expert was also a physical therapist with 9 years of clinical and US experience [34]. Similarly, Kumar and Attwood [36] found comparable agreement between the novice and expert raters for both left (ICC = 0.70) and right shoulders (ICC = 0.75). The experience of the raters was comparable to the current study, with the novices being physiotherapy students and the expert a registered physiotherapist. An earlier study reported slightly lower inter-rater agreement (ICC = 0.70) between the novice and expert raters when the shoulder position was neutral. The novice rater was an orthopedic resident with limited US experience, while the expert was a musculoskeletal radiologist with 6 years of US experience [37].

While the ICCs between novice and expert raters were classified as modest, the absolute agreement between raters was considerably variable with differences between raters as low as 0.27 mm and as high as 3.77 mm across seated and supine positions (Fig. 6). Other studies have similarly reported variable differences between the novice and expert raters, albeit lower than reported in the current study [34, 36, 37]. Thus, the higher variability demonstrated in this study may be attributed to the relative experience level of the novice rater. While it is possible that variation in transducer positioning may have affected the agreement between raters, the current research found that transducer angle agreement did not have a considerable impact on measurement reliability. While this suggests that transducer tilt variation was not a primary contributor to measurement disparity, the modest improvement in agreement with consistent angles between raters suggests that future research should continue to explore the impact of transducer positioning on measured outcomes.

Ultrasound versus MRI-based measurements

This study explored the agreement between SAS measured using ultrasound to that captured from MRI. The agreement between imaging modalities was found to be poor (ICCs: 0.21–0.49), with the highest agreement between modalities found between the expert rater (participant lying supine) and the MRI_Hum10 outcome (ICC = 0.49). This finding may have resulted from the better consistency in participant positioning (supine position) during US acquisition [12]. Generally, the US systematically overestimated the MRI SAS measurements for both raters (Figs. 8, 9). Researchers have similarly shown US to overestimate musculoskeletal parameters compared to MRI [51, 52]. For US measurements of the SAS, only the lateral-most borders of the acromion and humerus are visible (Fig. 4). This may not be consistent with the absolute minimum distance between bone surfaces that can be acquired using 3D surfaces reconstructed from MRI. However, interestingly, the agreement between the US measured SAS and the MRI_Lat outcome, which is theoretically more similar to the US measurement, was comparable to MRI_Hum10 located more medially.

Implications for laboratory-based SAS capture by non-expert

Overall, the current research demonstrated moderate inter-rater reliability, good to excellent intra-rater reliability and poor validity as compared to MRI. The high intra-rater reliability, which is consistent with previous literature [30, 31, 34,35,36,37], showed that novice (ICC: 0.83–0.84) and expert (ICC: 0.94–0.97) raters were capable of consistently measuring the SAS in both seated and supine positions, with similar agreement to that reported throughout the literature (ICCs: 0.84–0.99) [33, 35]. However, the poor agreement with MRI suggests that US may not be capable of capturing the true minimum SAS distance, as even a small change in the subacromial space (~ 1 mm) can be clinically meaningful [53]. Despite the poor agreement, the regression and Bland–Altman analyses revealed that a systematic difference between US and MRI exists, with the ability to predict MRI SAS from regression equations (Table 2; Figs. 8, 9). However, it is important to consider that, despite yielding a significant model, the coefficient of determination was low for all models (R2 = 0.21–0.32), with the exception of the US measure of the expert rater in the supine position, which yielded a stronger relationship with MRI_Hum10 measures (R2 = 0.48). Thus, with high intra-rater reliability demonstrated, this research suggests that a novice with introductory training may be capable of consistently measuring the SAS width within a testing session. This enables the study of SAS changes as a function of fatigue, exercise, occupational exposures or activities of daily living, sub-maximal and maximal exertions, among others. Further, the true minimum distance, while different in magnitude from that captured using US, may be calculated using regression equations.

Limitations

Certain limitations of this research should be considered. First, participant positioning during US imaging was standardized across all participants and raters. As participants may have had to change positions (seated, prone) between raters, it is possible that minor differences in posture existed between the two raters, which may have affected the SAS width. As well, with measurements captured from real-time freeze-framed images, it is uncertain whether differences between raters were due to differences in the US image or the measurement protocol. Further, intra-rater reliability of US imaging was assessed from the three repeated measurements in each position. The three measures were captured consecutively as the entire US imaging protocol was blocked randomized. This may have affected the reliability. Lastly, post hoc power calculations for ICC measures revealed a high power between ultrasound measures of novice and expert raters for both seated (p2 = 0.97) and supine (p2 = 0.86). However, power was lower for ICCs between ultrasound and MRI (p2 = 0.14–0.60). Thus, future investigations of image modality agreement should consider a larger sample size.

Conclusions

Novice and expert raters demonstrated moderate agreement in SAS measurements, which modestly improved with a consistent transducer tilt position. While the agreement between the US and MRI was poor, suggesting that US is not reporting the true minimum SAS, MRI SAS may be predicted from regression equations. Further, high intra-rater reliability suggests that the novice rater with introductory training is capable of capturing SAS measurements. Future research should examine this agreement in older adults with rotator cuff-related pain and/or SAIS where the SAS may be smaller and demonstrate less variability.

Availability of data and materials

Data available upon reasonable request.

Abbreviations

CT:

Computed tomography

DASH:

Disabilities of the Arm, Shoulder, and Hand questionnaire

ICC:

Intraclass correlation coefficient

MRI:

Magnetic resonance imaging

MRI_Hum10:

Mean distance of the 10% closest points

MRI_Lat:

Minimum distance from the most lateral point on the acromion to the humerus

SAIS:

Subacromial impingement syndrome

SAS:

Subacromial space

US:

Ultrasound

References

  1. Neer CS. Acromioplasty for the chronic impingement syndrome in the shoulder: a preliminary report. J Bone Joint Surg. 1972;54:41–50.

    Article  PubMed  Google Scholar 

  2. Michener LA, McClure PW, Karduna AR. Anatomical and biomechanical mechanisms of subacromial impingement syndrome. Clin Biomech. 2003;18:369–79.

    Article  Google Scholar 

  3. van der Windt DA, Koes BW, De Jong BA, Bouter LM. Shoulder disorders in general practice: incidence, patient characteristics, and management. Ann Rheumat Disea. 1995;95:959–64.

    Article  Google Scholar 

  4. Chopp-Hurley JN, O’Neill JM, McDonald AC, Maciukiewicz JM, Dickerson CR. Fatigue-induced glenohumeral an scapulothoracic kinematic variability: implications for subacromial space reduction. J Electromyograph Kinesiol. 2016;29:55–63.

    Article  Google Scholar 

  5. Ludewig PM, Braman JP. Shoulder impingement: biomechanical considerations in rehabilitation. Man Ther. 2011;16:33–9.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Ludewig PM, Cook TM. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder impingement. Phys Ther Rehab J. 2000;80:276–91.

    CAS  Google Scholar 

  7. Balke M, Schmidt C, Dedy N, Banerjee M, Bouillon B, Liem D. Correlation of acromial morphology with impingement syndrome and rotator cuff tears. Acta Orthop. 2012;84:178–83.

    Article  Google Scholar 

  8. Chopp-Hurley JN, O’Neill JM, Dickerson CR. Distribution of bone and tissue morphological properties related to subacromial space geometry in a young, healthy male population. Surg Radiol Anat. 2016;38:135–46.

    Article  PubMed  Google Scholar 

  9. Hughes RE, Bryant CR, Hall JM, Wening J, Huston LJ, Kuhn JE, et al. Glenoid inclination is associated with full-thickness rotator cuff tears. Clin Orthop Relat Res. 2003;407:86–91.

    Article  Google Scholar 

  10. Greenberg DL. Evaluation and treatment of shoulder pain. Med Clin. 2014;98:487–504.

    Google Scholar 

  11. Lin JC, Weintraub N, Aragaki DR. Nonsurgical treatment for rotator cuff injury in the elderly. J Am Med Direct Assoc. 2008;9:626–32.

    Article  Google Scholar 

  12. McCreesh KM, Crotty JM, Lewis JS. Acromiohumeral distance measurement in rotator cuff tendinopathy: Is there a reliable, clinically applicable method? A systematic review. Brit J Sports Med. 2015;49:298–305.

    Article  Google Scholar 

  13. Gruber G, Bernhardt GA, Clar H, Zachert M, Glehr M, Wurnig C. Measurement of the acromiohumeral interval on standardized anteroposterior radiographs: a prospective study of observer variability. J Should Elb Surg. 2010;19:10–3.

    Article  Google Scholar 

  14. Mirayan R, Donohoe S, Batech M, Suh BD, Acevedo DC, Singh A. Is there a difference in the acromiohumeral distances measured on radiographic and magnetic resonance images of the same shoulder with a massive rotator cuff tear? J Should Elb Surg. 2020;29:1145–51.

    Article  Google Scholar 

  15. Sanguanjit P, Apivatgaroon A, Boonsun P, Srimongkolpitak S, Chernchujit B. The differences of the acromiohumeral interval between supine and upright radiographs of the shoulder. Sci Rep. 2022;12:9404.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Saupe N, Pfirrmann CWA, Schmid MR, Werner CML, Zanetti M. Association between rotator cuff abnormalities and reduced acromiohumeral distance. Am J Roentgenol. 2006;187:376–82.

    Article  Google Scholar 

  17. Fehringer EV, Rosipal CE, Rhodes DA, Lauder AJ, Puumala SE, Feschuk CA, et al. The radiographic acromiohumeral interval is affected by arm and radiographic beam position. Skel Radiol. 2008;37:535–9.

    Article  Google Scholar 

  18. Bushberg JT, Seibrt JA, Leidholdt EM, Boone JM. Fluoroscopy. In: Bushberg JT, editor. The essentials of physics of medical imaging. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2012. p. 282–311.

    Google Scholar 

  19. Bey MJ, Brock SK, Beierwaltes WN, Zauel R, Kolowich PA, Lock TR. In vivo measurement of subacromial space width during shoulder elevation: technique and preliminary results in patients following unilateral rotator cuff repair. Clin Biomech. 2007;22:767–73.

    Article  Google Scholar 

  20. Lochmuller EM, Meier U, Anetzberger H, Habermeyer P, Muller-Gerbl M. Determination of subacromial space width and inferior acromial mineralization by 3D CT: preliminary data from patients with unilateral supraspinatus outlet syndrome. Surg Radiol Anat. 1997;19:329–37.

    Article  CAS  PubMed  Google Scholar 

  21. Werner CML, Conrad SJ, Meyer DC, Keller A, Hodler J, Gerber C. Intermethod agreement and interobserver correlation of radiologic acromiohumeral distance measurements. J Should Elb Surg. 2008;17:237–40.

    Article  Google Scholar 

  22. Bey MJ, Kline SK, Zauel R, Kolowich PA, Lock TR. In vivo measurement of glenohumeral joint contact patterns. J Adv Sign Proc. 2010. https://doi.org/10.1155/2010/162136.

    Article  Google Scholar 

  23. Bey MJ, Kline S, Zauel R, Lock TR, Kolowich PA. Measuring in-vivo glenohumeral joint kinematics: technique and preliminary results. J Biomech. 2008;41:711–4.

    Article  PubMed  Google Scholar 

  24. Bey MJ, Zauel R, Brock S, Tashman S. Validation of a new model-based tracking technique for measuring three-dimensional, in vivo glenohumeral joint kinematics. J Biomech Eng. 2006;128:604–9.

    Article  PubMed  Google Scholar 

  25. Mozingo JD, Akbari-Shandiz M, Marquez FM, Schueler BA, Holmes DR, McCollough CH, et al. Validation of imaging-based quantification of glenohumeral joint kinematics using an unmodified clinical biplane fluoroscopy system. J Biomech. 2018;71:306–12.

    Article  PubMed  PubMed Central  Google Scholar 

  26. De Wilde L, Plasschaert F, Berghs B, van Hoecke M, Verstraete K, Verdonk R. Quantified measurement of subacromial impingement. J Should Elb Surg. 2003;12:346–9.

    Article  Google Scholar 

  27. Lawrence RL, Schlangen DM, Schneider KA, Schoenecker J, Senger AL, Starr WC, et al. Effect of glenohumeral elevation on subacromial supraspinatus compression risk during simulated reaching. J Orthopaed Res. 2017;35:2329–37.

    Article  Google Scholar 

  28. Tasaki A, Nimura A, Nozaki T, Yamakawa A, Niitsu M, Morita W, et al. Quantitative and qualitative analyses of subacromial impingement by kinematic open MRI. Knee Surg Sport Traumatol Arthro. 2015;23:1489–97.

    Article  Google Scholar 

  29. Graichen H, Bonel H, Stammberger T, Heuck A, Englmeier KH, Reiser M, et al. A technique for determining the spatial relationship between the rotator cuff and the subacromial space in arm abduction using MRI and 3D image processing. Mag Res Med. 1998;40:640–3.

    Article  CAS  Google Scholar 

  30. Bagcier F, Kulcu DG, Yorulmaz E, Altunok EC. Intra- and inter-rater reliability of ultrasound measurements of supraspinatus tendon thickness, acromiohumeral distance, and occupation ratio in patients with shoulder impingement syndrome. Arch Rheumatol. 2020;35:385–93.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Hougs Kjær B, Ellegaard K, Wieland I, Warming S, Juul-Kristensen B. Intra-rater and inter-rater reliability of the standardized ultrasound protocol for assessing subacromial structures. Physiol Theor Pract. 2017;33:398–409.

    Article  Google Scholar 

  32. McCreesh KM, Anjum S, Crotty JM, Lewis JS. Ultrasound measures of supraspinatus tendon thickness and acromiohumeral distance in rotator cuff tendinopathy are reliable. J Clin Ultras. 2016;44:159–66.

    Article  Google Scholar 

  33. Kumar P, Bradley M, Swinkels A. Within-day and day-to-day intrarater reliability of ultrasonographic measurements of acromion-greater tuberosity distance in healthy people. Physiol Theor Pract. 2010;26:347–51.

    Article  Google Scholar 

  34. Cavaggion C, Navarro-Ledesma S, Luque-Suarez A, Juul-Kristensen B, Voogt L, Struyf F. Subacromial space measured by ultrasound imaging in asymptomatic subjects and patients with subacromial shoulder pain: an inter-rater reliability study. Physiol Theor Prac. 2022. https://doi.org/10.1080/09593985.2022.2072251.

    Article  Google Scholar 

  35. Kumar P, Chetwynd J, Evans A, Wardle G, Crick C, Richardson B. Interrater and intrarater reliability of ultrasonographic measurements of acromion-greater tuberosity distance in healthy people. Physiol Theor Pract. 2011;27:172–5.

    Article  Google Scholar 

  36. Kumar P, Attwood M. Inter-rater reliability of ultrasound measurements of acromion-greater tuberosity distance between experienced and novice raters in healthy people. Musculoskel Care. 2017;16:163–6.

    Article  Google Scholar 

  37. Pijls BG, Kok FP, Penning LIF, Guldemond NA, Arens HJ. Reliability study of the sonographic measurements of the acromiohumeral distance in symptomatic patients. J Clin Ultras. 2010;38:128–34.

    Article  CAS  Google Scholar 

  38. Bulbrook BB, Chopp-Hurley JN, Wiebenga EG, Pritchard JM, Gatti AA, Keir PJ, et al. Muscle architecture and subcutaneous fat measurements of rectus femoris and vastus lateralis at optimal length aided by a novel ultrasound transducer attachment. Physiol Can. 2023;75:74–82.

    Article  Google Scholar 

  39. McDonald AC. Understanding the response of the shoulder complex to the demands of repetitive work [Doctoral Dissertation]. 2017. McMaster University.

  40. Park HB, Yokota A, Gill HS, El Rassi G, McFarland EG. Diagnostic accuracy of clinical tests for the different degrees of subacromial impingement syndrome. J Bone Joint Surg. 2005;87:1446–55.

    PubMed  Google Scholar 

  41. Institute for Work & Health: Disabilities of the arm, shoulder, and hand (2022). http://www.dash.iwh.on.ca. Accessed Jul 2022.

  42. Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin J-C, Pujol S, et al. 3D Slicer as an image computing platform for the quantitative imaging network. Mag Res Imag. 2012;30:1323–41.

    Article  Google Scholar 

  43. Lorensen WE, Cline HE. Marching cubes: a high resolution 3D surface construction algorithm. ACM SIGGRAPH Comput Graph. 1987;1987(21):163–9.

    Google Scholar 

  44. Mun D, Kim BC. Three-dimensional solid reconstruction of a human bone from CT images using interpolation with triangular Bézier patches. J Mech Sci Technol. 2017;31:3875–86.

    Article  Google Scholar 

  45. Valette S, Chassery JM, Prost R. Generic remeshing of 3D triangular meshes with metric-dependent discrete voronoi diagrams. IEEE Trans Visual Comput Graph. 2008;14:369–81.

    Article  Google Scholar 

  46. Desmeules F, Minville L, Riederer B, Cote CH, Fremont P. Acromio-humeral distance variation measured by ultrasonography and its association with the outcome of rehabilitation for shoulder impingement syndrome. Clin J Sport Med. 2004;14:197–205.

    Article  PubMed  Google Scholar 

  47. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Int J Nurs Stud. 2010;47:931–6.

    Article  Google Scholar 

  48. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15:155–63.

    Article  PubMed  PubMed Central  Google Scholar 

  49. Zou GY. Sample size formulas for estimating intraclass correlation coefficients with precision and assurance. Stat Med. 2012;31(29):3972–81.

    Article  CAS  PubMed  Google Scholar 

  50. R Core Team. 2023. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.r-project.org/.

  51. Donohue KW, Little KJ, Gaughan JP, Kozin SH, Norton BD, Zlotolow DA. Comparison of ultrasound and MRI for the diagnosis of glenohumeral dysplasia in brachial plexus birth palsy. J Bone Join Surg. 2017;99:123–32.

    Article  Google Scholar 

  52. Renfree KJ, Dahiya N, Kransdorf MJ, Zhang N, Patel KA, Drace PA. Comparative accuracy of 1.5T MRI, 3T MRI, and static ultrasound in diagnosis of small gaps in repaired flexor tendons: a cadaveric study. J Hand Surg. 2021;46:287–94.

    Article  Google Scholar 

  53. Chopp JN, O’Neill JM, Hurley K, Dickerson CR. Superior humeral head migration occurs after a protocol designed to fatigue the rotator cuff: A radiographic analysis. J Should Elb Surg. 2010;19:1137–44.

    Article  Google Scholar 

Download references

Acknowledgements

The authors would like to thank Manoj Singh for his assistance capturing MRI data.

Funding

Funding for this study was provided by a York University Junior Faculty Funds and Minor Research Grant, and a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant held by Dr. Jaclyn N. Chopp-Hurley. The authors have no other competing interests to declare that are relevant to the content of this article.

Author information

Authors and Affiliations

  1. School of Kinesiology and Health Science, York University, 4700 Keele Street, Toronto, ON, M3J 1P3, Canada

    Stephen M. Boulanger, Alexandra Mahna, Oriana Culig, Loriann M. Hynes & Jaclyn N. Chopp-Hurley

  2. Department of Kinesiology, University of Waterloo, Waterloo, ON, Canada

    Talia Alenabi

  3. Department of Radiology, Stanford University, Stanford, CA, USA

    Anthony A. Gatti

  4. NeuralSeg Ltd., Hamilton, ON, Canada

    Anthony A. Gatti

Authors
  1. Stephen M. Boulanger
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Alexandra Mahna
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Talia Alenabi
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Anthony A. Gatti
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Oriana Culig
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Loriann M. Hynes
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jaclyn N. Chopp-Hurley
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

SMB completed the data analysis and prepared the manuscript. AM carried out project administration, participant recruitment, and data collection. TA was involved in project development and data collection. AAG provided support with data analysis and statistical procedures. OC assisted in the data collection and analysis. LMH provided equipment support. JNC-H was involved with the project development, administration, and manuscript writing, as well as securing funding for this project.

Corresponding author

Correspondence to Jaclyn N. Chopp-Hurley.

Ethics declarations

Ethics approval and consent to participate

This study received approval from the York University Office of Research Ethics (e2019-387). Written consent was obtained from all participants prior to data collection.

Competing interests

AAG is a shareholder of NeuralSeg Ltd. a medical image analysis company.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Boulanger, S.M., Mahna, A., Alenabi, T. et al. Investigating the reliability and validity of subacromial space measurements using ultrasound and MRI. J Orthop Surg Res 18, 986 (2023). https://doi.org/10.1186/s13018-023-04482-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13018-023-04482-1

Keywords

Journal of Orthopaedic Surgery and Research

ISSN: 1749-799X