Assessment and Classification of Glenohumeral Deformity



Assessment and Classification of Glenohumeral Deformity


Richard B. Jones, MD

Ari R. Youderian, MD



Shoulder arthroplasty has seen an enormous evolution in the last 2 decades. Subsequently, techniques to evaluate the glenohumeral joint and its related structures have evolved as well. In the past, surgeons relied heavily on physical examination and standard radiographs to evaluate disorders of the shoulder. This was followed by a progression to more advanced imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI). Using these modalities, surgeons were able to identify more complex patterns of glenoid or humeral deformity, leading to classification systems and an enhanced ability to “quantify” deformities. Consequently, our understanding of how various glenohumeral deformities and anatomic parameters contribute to glenohumeral disorders and disease processes has improved. Similarly, our increased knowledge base has changed our approach to shoulder arthroplasty and allowed us to achieve better outcomes. Subsequent development of three-dimensional (3D) technology continues to advance the evaluation of bony deformity and has further progressed with preoperative planning software and navigation techniques.

Recognition of glenohumeral deformity or anatomic parameters that may impact patient outcomes after surgery is of paramount importance. Failure to do so may compromise the chances of successful surgical treatment of the patient’s shoulder disorder. Hasan et al1 evaluated characteristics of unsatisfactory shoulder arthroplasties and found that malposition of the components occurred in 28% of the cases. Others have shown that placement of an anatomic glenoid component in excessive retroversion leads to significantly decreased glenohumeral contact area, increased contact pressure, increased micromotion of the implant, and increased osteolysis around the pegs.2,3,4

Preoperative assessment of the extent and location of bony deformity, associated soft tissue pathology, and other scapular anatomic parameters is commonly performed with standard radiographs, axial CT scans through the glenohumeral joint, MRI, or 3D CT reconstructions.


IMAGING MODALITIES


Standard Radiographs

Despite the many advanced imaging techniques available for the shoulder, the initial evaluation usually includes standard radiographs. A shoulder radiographic series may vary among ordering physicians and can be tailored to address the various conditions being evaluated. A common shoulder series for glenohumeral arthritis includes a true scapular anteroposterior (AP) (Grashey) view, outlet view, and axillary view. Other common views include true AP views with the humerus in internal and external rotation.

While a standard AP view is a common component of a basic shoulder series, it may not be the best to evaluate the glenohumeral joint in arthritic conditions. Since the glenohumeral joint is often angled anteriorly 30° to 40°, this view results in overlap of the glenoid and humeral head. The Grashey view, on the other hand, is a true AP of the shoulder in the plane of the scapula (FIGURE 6.1). By rotating the patient posteriorly or angling the beam laterally, the overlap of the glenoid and humeral head is eliminated. This allows a more accurate view of the glenohumeral joint space and provides a better assessment of glenohumeral arthritis, humeral head subluxation, and joint congruity.5 This view may be further modified to evaluate joint space narrowing by adding a weight and abduction, which applies an axial load to the joint.6 The Grashey view allows an evaluation of glenoid inclination (GI), superior migration of the humeral head, as well as a general assessment of the acromioclavicular joint.

The axillary view is essential and provides an assessment of abnormal glenoid wear patterns, glenoid retroversion, and posterior subluxation of the humeral head. Since soft tissue can interfere with the appreciation of the glenohumeral joint line on an axillary view, a projection that shows a continuous line from the coracoid base and glenoid articular surface should be obtained.7 (FIGURE 6.2)

The supraspinatus outlet view (FIGURE 6.3) is a modification of a standard scapular Y view with caudal tilt of the x-ray beam. It allows evaluation of the acromion and the subacromial space. An os acromiale, which may be of clinical importance during reverse shoulder
arthroplasty (RSA), can also be visualized on this view as well as on the axillary view. Both the scapular Y and outlet views are also useful to detect humeral head subluxation or dislocation.






FIGURE 6.1 Anteroposterior (AP) Grashey radiograph.

Characteristic radiographic changes occur as a result of the degenerative processes that impact the glenohumeral joint. Osteoarthritis (OA) leads to alterations in glenohumeral biomechanics, which can transform the anatomic features of the humerus and glenoid. Development of anterior contractures can lead to increased peripheral contact stresses on the glenoid followed by increased glenoid posterior erosion and increased glenoid retroversion. This may lead to ultimate decentering and posterior subluxation of the humeral head.8 Findings consistent with OA include marginal osteophytes around the anatomic neck of the humerus and glenoid, joint space narrowing, subchondral cysts, and subchondral sclerosis (FIGURE 6.4). Post-traumatic arthritis may demonstrate more severe deformity of the humeral head or glenoid.






FIGURE 6.2 Axillary radiograph.






FIGURE 6.3 Scapular outlet radiograph.

Rheumatoid arthritis (RA) may produce periarticular osteopenia, marginal erosions, and joint space narrowing with the absence of osteophytes (FIGURE 6.5). Concentric erosion of the glenoid may occur causing medial migration of the humeral head. Concomitant rotator cuff disease is common in RA as well, which may lead to superior migration of the humeral head. This causes eventual erosion and thinning of the acromion and may lead to acetabularization of the glenohumeral joint and acromion.5

Rotator cuff tear arthropathy can demonstrate similar radiographic findings as OA or RA. Narrowing of the acromiohumeral interval in a nonrheumatoid patient indicates a large and chronic rotator cuff tear. The humeral head may articulate with the undersurface of the acromion, the superior portion of the glenoid, and the base of the coracoid process possibly leading to acetabularization of the glenohumeral joint and acromion. The head may eventually lose the contour of the greater tuberosity (FIGURE 6.6).

While radiographs are routinely performed on patients considering shoulder arthroplasty, they do
contain limitations. Nyffeler et al9 showed that glenoid retroversion was overestimated on axillary radiographs 86% of the time with a maximum difference of 21°. The authors noted three factors that may contribute to this: (1) Most axillary radiographs are not aligned properly with the superior and inferior margins of the glenoid superimposed; (2) Often, on axillary views, the medial border of the scapula is not visualized due to the patient’s neck preventing sufficient medial placement of the cassette; and (3) The quality of the images depends on the orientation of the beam relative to the patient’s scapula. Small changes in this relationship can alter the version measurement. The authors concluded that glenoid version could not be accurately determined on standard radiographs and recommended a more reproducible imaging modality such as CT scans.






FIGURE 6.4 AP Grashey view demonstrating findings consistent with osteoarthritis including marginal osteophytes around the anatomic neck of the humerus and glenoid, joint space narrowing, subchondral cysts, and subchondral sclerosis.






FIGURE 6.5 AP Grashey (A) and axillary (B) views demonstrating findings consistent with rheumatoid arthritis (RA) including periarticular osteopenia, marginal erosions, and joint space narrowing with the absence of osteophytes.


Computed Tomography

It is now common for shoulder surgeons to employ more advanced imaging modalities such as CT or MRI in the evaluation of the glenohumeral joint. CT scans have been used in standard two-dimensional (2D) format, 3D reformatted 2D scans in the plane of the scapula, and more recently, 3D CT imaging often combined with preoperative surgical planning software. This has prompted abundant investigation as to which modality is most accurate to assess the glenoid.

Friedman10 first introduced the technique of using 2D CT scans to measure glenoid version. However, the accuracy of traditional 2D clinical CT scans has been questioned. 2D clinical CT scans are axial cuts traditionally aligned to the patient’s torso. Because of the natural tilt of the scapula, which is typically oriented in 20° to 30° of anteversion with respect to the coronal plane of the body, these axial slices are rarely aligned perpendicular to the scapula body. The mean angle between the direction of axial scans and the scapular body is approximately 35°, which may lead to inaccurate measurement of glenoid version.11 (FIGURE 6.7) The average error in version measurement using this technique has been demonstrated to be ±5°; however, the maximum error is even higher at 16°.11 Subsequent research has shown that obtaining accurate and reproducible glenoid measurements, at least
of version, requires CT scans using axial slices that have been reoriented in the plane of the scapula.11,12,13,14,15,16 CT scans that are not reoriented have shown significant overestimation of version and inclination.11,12,13,15






FIGURE 6.6 This Grashey view shows, in rotator cuff tear arthropathy, the humeral head may articulate with the undersurface of the acromion, the superior portion of the glenoid, and the base of the coracoid process possibly leading to acetabularization of the glenohumeral joint and acromion.

Rotation of the scapula in relation to the CT orientation can also affect measured version. Minor rotation of the scapula can alter the accuracy of glenoid version measurement by up to 10°. It is recommended that the glenoid articular surface should be perpendicular to plane of CT cut.14

More recently, much investigation has focused on 3D reconstructed CT scans in the evaluation of the glenohumeral joint. 3D reconstruction of CT scan images offers the potential to evaluate the bony erosion without positional errors. 3D reconstructed CT scans have shown improved assessment of glenoid bone loss in patients with instability17,18 as well as glenohumeral arthritis.19,20 (FIGURE 6.8) 3D CT images also have shown better interrater reliability19 and more accurate measurements of version and inclination21 compared to 2D measurements. 3D CT reconstruction techniques have furthermore led to preoperative planning techniques allowing for templating of implant size and positioning and has been shown to be reproducible and beneficial for surgical decision-making.19,21,22






FIGURE 6.7 The mean angle between the direction of axial scans and the scapular body is approximately 35°, which may lead to inaccuracy in version measurement.






FIGURE 6.8 Three-dimensional reconstruction of computed tomography (CT) scan showing Friedman line.


Magnetic Resonance Imaging

While CT scans remain the gold standard for advanced imaging of the glenohumeral joint, MRI can be useful and has certain advantages. MRI allows superior imaging of the soft tissues resulting in better assessment of the rotator cuff, deltoid, and other soft tissue structures around the shoulder, which can assist decision-making in shoulder arthroplasty (FIGURE 6.9). While CT scans are still the standard, the capacity of MRI to visualize bone morphology has improved with advancing technology. Compared to radiographs, MRI has been shown to provide a more accurate and reproducible assessment of glenoid version23 with better inter- and intrarater reliability in identifying glenoid morphology and classification.23,24
MRI has also been shown to be comparable to CT in the assessment of glenoid bone loss25,26 and version.27 Lowe et al,27 in their study comparing MRI to CT scan in the evaluation of the glenoid, found that MRI is largely comparable to CT scan for evaluation of the glenoid, with similar measurements of version and identification of less severe glenoid deformity. However, MRI is less accurate at distinguishing between Walch type B2 and C glenoids.






FIGURE 6.9 Axial magnetic resonance image (MRI).


SCAPULAR ANATOMIC PARAMETERS AND THEIR IMPLICATIONS

Much work has been done evaluating various scapular anatomic parameters and their influence on shoulder disorders, most commonly glenohumeral OA and rotator cuff tears (RCT). Version and inclination of the glenoid, size and shape of the glenoid vault, and acromial morphology including lateral extension, size, tilt, and rotation have been investigated. Identifying anatomic parameters and their causal relationship to shoulder disorders may allow the surgeon to identify patients at risk for particular problems or modify these factors at surgery to improve outcomes.


Glenoid Version

Abnormal glenoid version can have adverse consequences in shoulder arthroplasty. In anatomic total shoulder arthroplasty (ATSA), numerous studies have demonstrated increased forces in the cement mantle, glenoid implant, and glenoid bone as well as increased micromotion with implantation in excessive retroversion, all of which contribute to early glenoid loosening.2,4,28,29 During reverse total shoulder arthroplasty (RTSA), abnormal placement of the glenoid component can adversely affect ROM, stability, and scapular notching.30,31,32,33 Therefore, proper determination of the patient’s preoperative version can be crucial to a successful arthroplasty.






FIGURE 6.10 A, In a scapula with no deformity, Friedmans line and the scapular body line will be the same. B, In a more deformed scapula, Friedmans line and the scapular body line may differ causing variability in version calculation between the two methods. (A and B, Redrawn with permission from Friedman RJ, Hawthorne KB, Genez BM. The use of computerized tomography in the measurement of glenoid version. J Bone Joint Surg Am. 1992;74(7):1032-1037, and Randelli M, Gambrioli PL. Glenohumeral osteometry by computed tomography in normal and unstable shoulders. Clin Orthop Relat Res. 1986;(208):151-156.)

Axillary radiographs have been demonstrated to be inaccurate in the measurement of glenoid version.9 Friedman et al10 were among the first to use CT scans to characterize the relationship between glenoid retroversion and OA. They described a technique to calculate glenoid version using the transverse axis of the scapula, defined as a line drawn from the tip of the medial border of the scapula to the midpoint of the glenoid fossa. The first slice distal to the coracoid is usually chosen for measurement. Randelli and Gambrioli34 also proposed a technique that uses the scapular body line to calculate glenoid version (scapular body method). In both of these methods, a line drawn perpendicular to the axis along the glenoid surface was defined as neutral version, and the angle from either the posterior margin or the anterior margin of the glenoid defined the native version (FIGURE 6.10). Rouleau et al35 compared the validity of both of these measurement methods for determining glenoid version. The authors showed that both methods demonstrated excellent reliability but suggested the use of the Friedman method citing it as more user-friendly in the presence of a curved scapula for all glenoid types.
The authors also introduced new reference lines to characterize a glenoid surface with posterior wear and eccentric erosion. These included the paleoglenoid (original glenoid surface), intermediate glenoid (line from anterior and posterior edge), and neoglenoid (posterior erosion surface) (FIGURE 6.11). They indicated that in the presence of a B2 glenoid and posterior erosion, the choice of an intermediate glenoid line is more reliable for measurement. This line also represents the surface that could be obtained with minimal bone loss after limited reaming of the glenoid surface during total shoulder arthroplasty (TSA). Poon et al36 introduced a third method of version measurement using a 2D CT slice focusing solely on the glenoid vault. The authors felt their technique was more relevant to glenoid implant placement and was generally more simple and accessible to everyone (FIGURE 6.12). However, the vault method has been shown to have less reliability and more variability according to CT slice height or angulation than the Friedman method in glenoid version measurement.37






FIGURE 6.11 A, Arthritic shoulder showing glenoid with posterior wear and eccentric erosion. B, Paleoglenoid (original glenoid surface), (C) intermediate glenoid (line from anterior and posterior edge), (D) neoglenoid (posterior erosion surface). (Redrawn with permission from Rouleau DM, Kidder JF, Pons-Villanueva J, Dynamidis S, Defranco M, Walch G. Glenoid version: how to measure it? Validity of different methods in two-dimensional computed tomography scans. J Shoulder Elbow Surg. 2010;19(8):1230-1237.)






FIGURE 6.12 Poon two-dimensional (2D) glenoid vault method. (Reproduced with permission from Poon PC, Ting FS. A 2-dimensional glenoid vault method for measuring glenoid version on computed tomography. J Shoulder Elbow Surg. 2012;21(3):329-335.)

3D CT reconstructions are quickly becoming the standard for assessing the glenohumeral joint, often in
concert with preoperative planning software. Kwon et al38 used 3D CT imaging of cadaveric specimens and showed that on average, the glenoid version angles measured from the 3D CT images were within 1.0° ± 0.7° of those from the actual specimen. The measurements from the 3D CT images also showed high interobserver and intraobserver reliability. Beuckelaers et al39 described a 3D technique to quantify glenoid erosion and measure version in a population with primary glenohumeral arthritis and posterior subluxation of the humeral head. They concluded that the amount of erosion in glenoids with a Walch type B1 morphology can be underestimated using 2D CT evaluation because the orientation of the maximum erosion in type B1 glenoids is situated more posteroinferiorly (FIGURE 6.13).






FIGURE 6.13 Three-dimensional (3D) measurement technique showing maximum erosion in type B1 glenoids is situated more posteroinferiorly. (Reproduced with permission from Beuckelaers E, Jacxsens M, Van Tongel A, De Wilde LF. Three-dimensional computed tomography scan evaluation of the pattern of erosion in type B glenoids. J Shoulder Elbow Surg. 2014;23(1):109-116.)


Inclination of the Glenoid

The inclination of the glenohumeral joint is an important variable. Increased GI has been associated with RCT40,41,42 and superior migration of the humeral head.43,44 Inclination of the glenoid can also play an important role in TSA. Proper positioning of the glenoid component is the most important factor for implant longevity and avoiding clinical failure.45 Excessive superior or inferior placement of the glenoid component has been shown to create the highest stress on the cement mantle.46 Downward tilting of the glenoid component has been shown to decrease superior migration of the humeral head,47 subluxation of the humeral head and implant tilting,48 as well as balance supraspinatus insufficiency.49 In RTSA, excessive superior inclination can lead to increased risks of baseplate loosening, whereas excessive inferior tilt may increase impingement on the neck of the scapula leading to scapular notching.50






FIGURE 6.14 The beta angle is defined as the angle formed between the floor of the supraspinatus fossa and the glenoid fossa measured on an anteroposterior (AP) Grashey view.

Maurer et al51 described the most commonly accepted technique for accurate and reproducible measurement of GI on standardized AP radiographs and CT images. The authors described the beta angle as being the most reproducible measurement for GI on conventional AP radiographs, providing a resistance to positional variability of the scapula and good interrater reliability. Radiographically, the beta angle is defined as the angle formed between the floor of the supraspinatus fossa and the glenoid fossa measured on an AP Grashey view (FIGURE 6.14). On CT images, the coronal section at the deepest point of the supraspinatus fossa is used. The measurement line is placed along the cortical margin of the floor of the supraspinatus fossa. As with glenoid version, measurements of GI have been shown to be most accurate using 3D reformatted scans in the plane of the scapula. Nonreformatted CT scans as well as standard radiographs are less accurate.52 More recently, MRI has shown to be comparable to CT scan in accurately measuring GI using the beta angle.42

Boileau et al53 drew our attention to problems using the beta angle technique to measure inclination prior to RTSA. Baseplates are routinely placed on the inferior portion of the glenoid with slight overhang to reduce scapular notching. The authors point out that
in patients with central erosion of the glenoid (Favard E1), there is risk for placing the baseplate in superior tilt if this is not taken into account. The beta angle will underestimate the superior orientation of the baseplate by using the entire glenoid fossa in the measurement. They introduce a new measurement for inclination called the RSA angle (FIGURE 6.15). The RSA angle was defined as the angle between the inferior part of the glenoid fossa and the perpendicular to the floor of the supraspinatus fossa. In their investigation, the RSA angle averaged 20° ± 5° versus the TSA (or beta) angle, which was on average 10° ± 5° lower. This difference needs to be corrected to achieve neutral inclination of the baseplate.

Only gold members can continue reading. Log In or Register to continue

Jun 23, 2022 | Posted by in ORTHOPEDIC | Comments Off on Assessment and Classification of Glenohumeral Deformity
Premium Wordpress Themes by UFO Themes