Abstract
Objectives:
To investigate the clinical utility of additional axillary or Velpeau views in evaluating potential shoulder trauma following a standard radiograph series of anteroposterior, Grashey, and/or trans-scapular views.
Design:
Retrospective study.
Setting:
Level I academic medical center.
Patients:
All patients in a 10-year span who received an initial shoulder radiograph series followed by additional axillary/Velpeau views within 24 hours.
Main Outcome Measurements:
The clinical utility of the additional axillary/Velpeau views, including the final diagnosis and treatment plan, as ascertained through examination of radiology reports, progress notes, and radiograph images.
Results:
A total of 271 cases were reviewed, with 35 patients being excluded from the final cohort because they received post-treatment radiographs to confirm a successful therapeutic outcome. The additional axillary/Velpeau views did not affect clinical decision making in 230 (97.5%) of the remaining 236 cases. All 6 patients whose care benefitted from the additional views carried the diagnosis of shoulder instability, accounting for 40% of this diagnostic group. The additional views confirmed an equivocal finding in 5 of these 6 cases and changed the diagnosis (demonstrating a posterior dislocation that was not evident on initial radiographs) and treatment plan (leading to a closed glenohumeral reduction procedure) in the other case.
Conclusions:
Additional axillary/Velpeau views of suspected shoulder trauma rarely led to a change in the final treatment plan, except in patients in which a definitive diagnosis of stability or instability could not be made based on initial radiographs. A cost/benefit analysis is required to weigh the cost of additional radiographs with the benefit of capturing infrequent yet serious dislocations (usually posterior).
Keywords: shoulder, radiograph, x-ray, axillary view, Velpeau view, shoulder trauma, shoulder instability, shoulder dislocation, shoulder subluxation, shoulder fracture
INTRODUCTION
Shoulder pain is a common complaint at office visits and emergency departments. When a patient presents with suspected trauma, a standard series of radiographs comprised of anteroposterior (AP), Grashey (AP view of the glenoid), trans-scapular (Y), and/or axillary views is often ordered.5,6 The Grashey, Y, and axillary views all visualize the glenohumeral joint in the orthogonal plane and are consequently most useful in identifying glenohumeral instability (dislocation and subluxation).9,13 If a patient has only received an initial series comprised of AP, Grashey, and Y views, it can be orthopaedic dogma to request an additional axillary or Velpeau view.
Only a few studies have investigated the clinical utility of an additional axillary/Velpeau views after other orthogonal views have already been taken.1-3,7,14 Some studies have shown that these views are particularly instrumental in detecting albeit rare posterior glenohumeral dislocations, whereas other studies have demonstrated that they are not accurate and do not affect clinical decision making in the context of proximal humerus fractures, the most common diagnosis associated with acute shoulder trauma.1-3,14 Furthermore, one study found that axillary views did not significantly contribute to pre-treatment radiographic findings regarding anterior shoulder instability as compared to other views.7 In sum, the small amount of literature that exists on this topic suggests that additional axillary/Velpeau views may not be clinically useful.
Besides the uncertainty surrounding its clinical utility, the axillary view and even its modifications like the Velpeau view are also difficult to obtain and can be quite painful to patients, especially to the same patients with shoulder instability who could ostensibly benefit the most from the additional view.2,4,11 Furthermore, they cost the patient and health care system while potentially delaying the delivery of care.1 Some studies have concluded that orthopaedic surgeons and radiologists should discuss which views to order in evaluating shoulder trauma on an individual basis since the optimal series of images often varies from patient to patient.8,10 Thus, utilizing axillary or Velpeau views only when they are truly beneficial constitutes a potential opportunity to prevent patients from unnecessarily undergoing additional pain and radiation exposure while potentially reducing economic costs.
The aim of this study was to investigate whether additional axillary/Velpeau views of suspected shoulder trauma obtained after orthogonal Grashey and Y views would meaningfully change or otherwise affect diagnosis and/or treatment. Based on the authors’ clinical experiences and the available literature, we hypothesized that the additional views would not contribute to clinical decision making, except in patients with shoulder instability whose pathology would be optimally visualized by these views.
MATERIALS AND METHODS
Study Design
A hospital billing registry at one academic institution was queried for all patients between January 1, 2007 and December 31, 2016 who received an initial shoulder radiograph series followed by an additional shoulder radiograph within 24 hours. This search yielded 511 cases. Individual chart reviews were then performed by a shoulder fellowship-trained orthopaedic surgeon to determine which records fulfilled two inclusion criteria: patients had to be 12 years of age or older and must have received an additional axillary or Velpeau view radiograph following an initial shoulder trauma series that included AP, Grashey, and Y views. A total of 271 cases fulfilled these criteria. This study received approval from our Institutional Review Board.
Data Collection and Analysis
The search query yielded each patient’s medical record number, date of birth, gender, race and ethnicity, insurance status, date of encounter, and encounter setting (emergent, inpatient, and/or outpatient). Additional information garnered from the subsequent chart reviews included the radiologist impressions following the initial and secondary radiograph series, respectively, as well as whether the additional series included axillary/Velpeau views. Moreover, the clinical utility of the additional views, including the final diagnosis and treatment plan, was ascertained through examination of progress notes as well as consultation of radiograph images accessed through the Epic Systems (Verona, WI) electronic medical record. Cost information was provided by the institution’s billing office. Basic statistical analyses were performed within Microsoft Excel 2016 (Redmond, WA)
RESULTS
A total of 271 patients received additional axillary/Velpeau views within 24 hours of an initial shoulder trauma radiograph series. The patient demographics and characteristics are listed in Table 1. Of note, 242 (89.3%) of patients were seen in the emergency department. Furthermore, 128 (47.2%) of the additional views taken were axillary and 139 (51.3%) were Velpeau, whereas 4 (1.48%) patients received both of these views.
Table 1.
Characteristics of 271 patient cohort who received additional axillary/Velpeau views following an initial series of shoulder radiographs.
| Mean Age (SD) | 50.7 (25.2) |
|---|---|
| Sex, n (%) | |
| Female | 151 (55.7) |
| Male | 120 (44.3) |
| Encounter Setting, n (%) | |
| Emergency Department | 242 (89.3) |
| Inpatient | 6 (2.21) |
| Outpatient | 23 (8.49) |
| Insurance Status, n (%) | |
| Medicare | 96 (35.4) |
| Medicaid | 51 (18.8) |
| Private or Commercial | 90 (33.2) |
| Self Pay and Other | 34 (12.5) |
| Additional View Type, n (%) | |
| Axillary | 128 (47.2) |
| Velpeau | 139 (51.3) |
| Both Axillary and Velpeau | 4 (1.48) |
Out of the 271 total patients, 117 (43.2%) were ultimately diagnosed with proximal humerus fractures, 8 (2.95%) with proximal humerus fracture-dislocations, 44 (16.2%) with shoulder instability (dislocation or subluxation), 15 (5.54%) with clavicle or scapula fractures, 12 (4.43%) with AC joint widening or separation, and 75 (27.7%) with minimal to no pathology evident on imaging (including patients with osteoarthritis or other degenerative changes) (Table 2). Importantly, 3 of the 117 patients with proximal humerus fractures, 3 of the 8 patients with proximal humerus fracture-dislocations, and 29 of the 44 patients with shoulder instability received post-treatment axillary/Velpeau views to confirm successful glenohumeral reduction. Importantly, none of these patients received initial axillary/Velpeau views as part of their initial trauma series of radiographs. As a result, these 35 patients were not included in the final study cohort because – instead of contributing to the initial diagnostic work-up – the additional views were acquired after clinical intervention in order to confirm a successful therapeutic outcome. Thus, the final study cohort was comprised of 236 patients.
Table 2.
Final diagnoses of all 271 patients who received additional axillary/Velpeau views following an initial series of shoulder radiographs; 236 patients were included in the final study cohort.
| Diagnosis | Total Included Patients (%) |
Excluded Patients, n* |
|---|---|---|
| Proximal humerus fracture | 114 (48.3) | 3 |
| Proximal humerus fracture-dislocation | 5 (2.12) | – |
| Anterior dislocation | 3 (1.27) | 3 |
| Inferior dislocation | 2 (0.847) | – |
| Shoulder instability | 15 (6.36) | – |
| Anterior dislocation/subluxation | 6 (2.54) | 26 |
| Inferior dislocation/subluxation | 5 (2.12) | 1 |
| Posterior dislocation/subluxation | 4 (1.69) | 2 |
| Clavicle or scapula fracture | 15 (6.36) | – |
| AC joint widening/separation | 12 (5.08) | – |
| Minimal to no pathology (degenerative changes) | 75 (31.8) | – |
| Total (N) | 236 (100.0) | 35 |
Patients who received additional axillary and/or Velpeau views to confirm successful glenohumeral reduction were excluded from the final study cohort.
Additional axillary/Velpeau views did not change the treatment plan in 230 (97.5%) of these 236 cases. As such, there were 6 patients in which the additional view(s) ultimately provided utility in the treatment plan, all of whom carried the diagnosis of shoulder instability. These patients whose care benefitted from the additional view(s) accounted for 40% (6 of 15 patients) of the shoulder instability group (Table 3). Importantly, in 5 of these 6 cases, the additional axillary/Velpeau views were utilized to definitively diagnose possible dislocations that were not adequately visualized in the initial radiograph series. However, in only 1 case did the additional view change both the diagnosis and treatment plan (Case 2 in Table 3). Specifically, a Velpeau view demonstrated a posterior dislocation that was read as an inferior subluxation on initial radiographs, thereby altering the treatment plan by leading to a closed glenohumeral reduction procedure (Figures 1a - 1d).
Table 3.
Cases in which additional axillary/Velpeau views contributed to clinical decision making, accounting for 6 (40%) of 15 patients diagnosed with shoulder instability.
| Case | View Type | Initial Diagnosis | Post-Axillary and/or Velpeau Views Diagnosis |
Effect of Axillary and/or Velpeau View on Clinical Decision Making |
|---|---|---|---|---|
| 1 | Velpeau | Prosthetic hemi-arthroplasty in anatomical alignment (no pathology) | Anterior dislocation of prosthetic humeral head | The dislocation was recognized by the orthopaedic surgeon in the initial radiograph series, but the radiologist required an additional Velpeau view to visualize the pathology. This resulted in a glenohumeral reduction procedure. |
| 2 | Velpeau | Inferior subluxation | Posterior dislocation | The correct diagnosis of posterior dislocation was made with the Velpeau view, resulting in a glenohumeral reduction procedure. |
| 3 | Both (Axillary and Velpeau) | Anterior dislocation of prosthetic humeral head | No dislocation | The lack of pathology was recognized by the orthopaedic surgeon in the initial radiograph series, but the radiologist required additional axillary and Velpeau views to confirm this. |
| 4 | Axillary | Anterior subluxation (equivocal) | Anterior subluxation (confirmed) | The axillary view was useful in confirming the diagnosis, thereby resulting in a glenohumeral reduction procedure. |
| 5 | Axillary | Anterior dislocation or subluxation | No dislocation | The lack of pathology was recognized by the orthopaedic surgeon in the initial radiograph series, but the radiologist required an additional axillary view to confirm this. |
| 6 | Both (Axillary and Velpeau) | Posterior subluxation (equivocal) | Posterior subluxation (confirmed) | The axillary and Velpeau views were useful in confirming the diagnosis, thereby resulting in a glenohumeral reduction procedure. |
Figure 1a – 1d.
Anteroposterior (1a), Grashey (1b), trans-scapular (1c), and Velpeau (1d) radiographs of a patient with a posterior shoulder dislocation. Based on the first three images, the patient was diagnosed with an inferior subluxation. The subsequent Velpeau view changed the diagnosis and led to a closed glenohumeral reduction procedure.
DISCUSSION
In order to promote quality improvement and reduce health care spending, evidence-based medicine challenges clinicians to re-evaluate longstanding practices that may increase cost without improving care. To our knowledge, this is the first study to evaluate the clinical utility of obtaining an additional orthogonal view (specifically, axillary/Velpeau views) to evaluate suspected shoulder trauma. It is a commonplace practice of consulted orthopaedic surgeons to recommend such an additional view, but there is minimal literature available on whether this actually contributes to clinical decision making.13 Our aim was to therefore investigate whether it is beneficial for patients to undergo this imaging, or if they can instead potentially avoid further pain, radiation exposure, costs, and prolongation of their care. We found that additional axillary/Velpeau views taken after an initial trauma series in patients who had not received therapeutic intervention (such as reduction) did not affect diagnosis or treatment in 97.5% of cases. This supported our hypothesis that two orthogonal views (in this study, Grashey and Y views) are generally sufficient for the diagnosis and treatment of suspected shoulder trauma.
Notably, the only diagnostic category that benefitted from the additional views was shoulder instability (dislocation or subluxation). The care of patients diagnosed with other conditions – such as proximal humerus fractures and fracture-dislocations – did not benefit from additional views. This substantiated and further expanded upon the findings of a previous study which showed that axillary views did not contribute to clinical decision making in the context of proximal humerus fractures.1 As mentioned above, these findings should not be applied to patients who received post-reduction axillary view radiographs to confirm a successful therapeutic outcome, a cohort which was excluded from this study.
The results of this study are especially pertinent to instances in which orthopaedic surgeons and radiologists may disagree regarding the findings from the initial series: notably, half of the cases in which the additional views contributed to clinical decision making corresponded to the resolution of disagreements between radiologists and orthopaedic surgeons. In the future, a discussion between the radiologist and surgeon regarding more nuanced cases could prevent the ordering unnecessary imaging studies by allowing each provider to implement his or her expertise. Indeed, when the axillary/Velpeau views did contribute to clinical decision making, it was usually with regards to confirming an inadequately visualized finding on the initial series and/or resolving the abovementioned disagreements between team members.
In only one circumstance did a Velpeau view change both the diagnosis and treatment course of a patient. Interestingly, this patient had experienced a posterior dislocation that was initially read as an inferior subluxation. Posterior dislocations are rare, with one recent study finding that they account for 2.7% of all dislocations.14 A previous study did suggest that axillary views and/or their modifications like the Velpeau view were crucial in helping diagnose posterior dislocations and that these views should be ordered in patients with a history of seizures, electrocution, congenital conditions, or other causes of trauma that increase the likelihood of a posterior dislocation.3 Consequently, it is our recommendation that additional axillary or Velpeau views be obtained in patients in which a definitive diagnosis of stability or instability cannot be made or when a potential posterior dislocation is suspected.
One limitation of this study is that it only examined physician practices and patients at one institution, though we did include patients from over a decade to attempt to capture the variety of pathologies seen across other institutions. Another limitation of this study is that we did not delineate the adequacy of all radiographs. This was primarily due to the fact that reading radiographs can be subjective (as well as corresponding radiology reports), such that a clear finding to one orthopaedic surgeon or radiologist may be more ambiguous to another. However, it is likely that some additional axillary/Velpeau views were ordered because the initial series had suboptimal positioning. Moreover, it was evident during the chart review component of this study that some axillary/Velpeau views themselves were suboptimally captured, with poor visualization of the glenohumeral joint.
The paucity of literature regarding the utilization and clinical contribution of axillary and its modified views presents an opportunity for significant future research. For example, studies could examine institutional practices across the United States, including which views comprise initial and additional suspected shoulder trauma radiograph series across various institutions. Furthermore, an in-depth characterization of the radiographic visualization of shoulder instability could provide clearer insight regarding which particular types of shoulder instability – such as posterior dislocations – benefit the most from additional orthogonal views of the glenohumeral joint. Lastly, cost-utility analyses could project the potential savings of greatly reducing the number of additional shoulder radiographs taken to patients and the health care system as a whole. For example, an additional 1-view radiograph at our institution incurred a billing charge of $428 during the 2018 fiscal year, representing an opportunity for significant cost savings to our patients and institution.
In summary, we confirmed our hypothesis that additional axillary/Velpeau views of suspected shoulder trauma would rarely lead to a change in diagnosis or treatment plan, except in patients in which a definitive diagnosis of stability or instability could not be made. Specifically, among the shoulder instability group, 40% benefited from the addition of an axillary/Velpeau view radiograph. However, the patients with other diagnoses – who comprised the majority of our study cohort – did not benefit from these secondary views and instead were subjected to unnecessary pain, radiation exposure, and costs. A cost/benefit analysis is required to weigh the cost of additional radiographs with the benefit of capturing infrequent yet serious dislocations (usually posterior). As a result, we currently recommend ordering axillary and related views only when initial evaluation through physical exam and radiographic findings suggest the possibility of a shoulder instability diagnosis or when a posterior dislocation is suspected.
Acknowledgements:
The authors would like to thank the Clinical Research Data Warehouse (CRDW) maintained by the Center for Research Informatics (CRI) at University of Chicago for their assistance in querying our hospital billing registry.
Conflicts of Interest and Source of Funding: None of the authors declare conflicts of interest related to this study. This project was supported in part by the National Center for Advancing Translational Sciences of the National Institutes of Health through Grant Number UL1 TR000430. Data from this study were provided by the Clinical Research Data Warehouse (CRDW) maintained by the Center for Research Informatics (CRI) at University of Chicago. The Center for Research Informatics is funded by the Biological Sciences Division, the Institute for Translational Medicine/CTSA (NIH UL1 TR000430) at the University of Chicago.
Footnotes
Level of Evidence: Level III diagnostic study.
REFERENCES
- 1.Berkes MB, Dines JS, Birnbaum JF, et al. The axillary view typically does not contribute to decision making in care for proximal humeral fractures. HSS J. 2015;11(3):192–197. doi: 10.1007/s11420-015-9445-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bloom MH, Obata WG. Diagnosis of posterior dislocation of the shoulder with use of Velpeau axillary and angle-up roentgenographic views. J Bone Joint Surg Am. 1967;49(5):943–949. doi: 10.2106/00004623-196749050-00011 [DOI] [PubMed] [Google Scholar]
- 3.Clough TM, Bale RS. Bilateral posterior shoulder dislocation: the importance of the axillary radiographic view. Eur J Emerg Med. 2001;8(2):161. [DOI] [PubMed] [Google Scholar]
- 4.Geusens E, Pans S, Verhulst D, et al. The modified axillary view of the shoulder, a painless alternative. Emerg Radiol. 2006;12(5):227–230. doi: 10.1007/s10140-006-0484-x [DOI] [PubMed] [Google Scholar]
- 5.Greenspan A, Beltran J. Upper Limb I: Shoulder Girdle In: Orthopedic Imaging: A Practical Approach. 6th ed. Lippincott Williams & Wilkins; 2014. p. 107–161. [Google Scholar]
- 6.Moghadamian ES, Bosse MJ, MacKenzie EJ. Principles of mangled extremity management In: Rockwood and Green’s Fractures in Adults. Lippincott Williams & Wilkins; 2010. p. 429–451. [Google Scholar]
- 7.Pavlov H, Warren RF, Weiss CB, et al. The roentgenographic evaluation of anterior shoulder instability. Clin Orthop. 1985;(194):153–158. doi: 10.1097/00003086-198504000-00020 [DOI] [PubMed] [Google Scholar]
- 8.Prato N, Berri L, Bergamaschi G, et al. [Diagnostic reliability of radiography in shoulder trauma. Considerations on the use and choice of more radiographic projections]. Radiol Med (Torino). 1996;91(6):687–692. [PubMed] [Google Scholar]
- 9.Rowe CR, Zarins B. Chronic unreduced dislocations of the shoulder. J Bone Joint Surg Am. 1982;64(4):494–505. [PubMed] [Google Scholar]
- 10.Sandstrom CK, Kennedy SA, Gross JA. Acute Shoulder Trauma: What the Surgeon Wants to Know. RadioGraphics. 2015;35(2):475–492. doi: 10.1148/rg.352140113 [DOI] [PubMed] [Google Scholar]
- 11.Senna LF, Pires e Albuquerque R. Modified axillary radiograph of the shoulder: a new position. Rev Bras Ortop. 2016;52(1):115–118. doi: 10.1016/j.rboe.2016.12.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Shields DW, Jefferies JG, Brooksbank AJ, et al. Epidemiology of glenohumeral dislocation and subsequent instability in an urban population. J Shoulder Elbow Surg. 2018;27(2):189–195. doi: 10.1016/j.jse.2017.09.006 [DOI] [PubMed] [Google Scholar]
- 13.Silfverskiold JP, Straehley DJ, Jones WW. Roentgenographic evaluation of suspected shoulder dislocation: a prospective study comparing the axillary view and the scapular “Y” view. Orthopedics. 1990;13(1):63–69. [DOI] [PubMed] [Google Scholar]
- 14.Simon JA, Puopolo SM, Capla EL, et al. Accuracy of the axillary projection to determine fracture angulation of the proximal humerus. Orthopedics. 2004;27(2):205–207. [DOI] [PubMed] [Google Scholar]