An audit of the accuracy of MRI shoulder arthrography reports

An audit to assess the accuracy of MRI shoulder arthrograms when compared to shoulder arthroscopy findings.

The shoulder has the greatest range of motion out of all the major joints in the body. As a consequence, it is prone to dislocations, and once dislocated, possible recurrent instability [1,2]. 

A number of different bone and soft tissue lesions may be encountered following shoulder dislocation and contribute to ongoing instability [3] - failure to address these intra-operatively may lead to recurrent instability.  Although shoulder arthroscopy is deemed the gold standard [3], MRI shoulder arthrography is a useful and less invasive technique to diagnose these lesions, the demonstration of which may help influence the decision to go for surgical intervention over conservative management.  Additionally, MRI shoulder arthrography plays a vital role in surgical planning prior to arthroscopic repair.

The standard: 

MRI shoulder arthrography should be reasonably accurate (in terms of sensitivity and specificity) in the diagnosis of the following lesions which may be encountered following shoulder dislocation/recurrent instability:

1. Antero-inferior labral tear *

2. Superior labral tear *

3. Posterior labral tear *

4. Bony Bankart lesion

5. Hill-Sachs lesion

6. Rotator cuff tear

7. Humeral avulsion of inferior glenohumeral ligament (HAGL) injury

* Differentiation between different locations of labral tears is important not only from a surgical planning perspective but also the reported sensitivities and specificities of labral injuries vary depending on the location.

Target: 

Based on published literature [4-9], the following targets should be reached:

1. Antero-inferior labral tear - sensitivity 92%, specificity 92%

2. Superior labral tear - sensitivity 80%, specificity 67%

3. Posterior labral tear - sensitivity 57%, specificity 96% 

4. Bony Bankart lesion - sensitivity 90%, specificity 90% 

5. Hill-Sachs lesion - sensitivity 96%, specificity 90% 

6. Rotator cuff tear - sensitivity 92%, specificity 97% 

7. Humeral avulsion of inferior glenohumeral ligament (HAGL) injury - sensitivity 88%, specificity 100%

Indicators: 

Sensitivity and specificity of:

1. Antero-inferior labral tear

2. Superior labral tear

3. Posterior labral tear

4. Bony Bankart lesion

5. Hill-Sachs lesion

6. Rotator cuff tear

7. Humeral avulsion of inferior glenohumeral ligament (HAGL) injury

Data items to be collected: 

1. List of patients who have had shoulder arthroscopic stabilization procedures performed.

2. Whether or not these patients have had MRI shoulder arthrograms performed.

3. Whether or not these patients have had previous surgery (if previous surgery has been performed, these should be excluded from the audit as the anatomical distortion can make interpretation difficult).

4. Presence / absence of the above listed lesions on the operation note.

5. Comparing the MRI shoulder arthrogram report with data from the operation note in order to determine the true positive, false positive, true negative and false negative findings of the above listed lesions.

6. Calculation of the sensitivities and specificities of the above listed lesions using data collected above (sensitivity = true positive / true positive + false negative;  specificity = true negative / true negative + false positive).

Suggested number: 

-In high volume specialist centre, all patients who had had arthroscopic shoulder stabilization procedures in the last 6 months.

-Otherwise, a 12 month period is suggested.

1. Discuss possible technical reasons for reduced accuracy amongst radiologists and radiographers. For example, if image quality was reduced by contrast extravasation, the arthrogram procedural technique should be examined to see if this could be improved upon. 

2. Additionally, discuss whether an abduction external rotation (ABER) view, if not already performed, would have helped improve the accuracy of the scan, and if so, consideration of implementing this routinely in the scan protocol if possible.  It is believed that an ABER view may improve the detection of certain labral lesions and rotator cuff tears [10] though it may not be possible to perform this in all cases due to patient intolerance and time constraints.

2. Retrospectively, with the intra-operative results known, radiologists should review the scans with the false positive and false negative findings in order to learn from any discrepancies.

Suggested resources:

1. Local orthopaedic surgical database (e.g. Bluespier) to search for previously operated cases and operation notes.

2. Local RIS/PACS for MRI reports.

1. Bencardino JT, Gyftopoulos S, Palmer WE. Imaging in anterior glenohumeral instability. Radiology 2013; 269(2): 323 337. 

2. Subhas N, Gyftopoulos S MR Imaging of the Shoulder published Feb 2020 DOI: https://doi.org/10.1016/j.mric.2020.02.001

3. Mohana-Borges AV, Chung CB, Resnick D. Superior labral anteroposterior tear: classification and diagnosis on MRI and MR arthrography. AJR Am J Roentgenol 2003; 181(6): 1449 1462.

4. Palmer WE, Caslowitz PL. Anterior shoulder instability: diagnostic criteria determined from prospective analysis of 121 MR arthrograms. Radiology 1995; 197(3): 819 825.

5. Sheridan K, Kreulen C, Kim S, Mak W, Lewis K, Marder R.  Accuracy of magnetic resonance imaging to diagnose superior labrum anterior-posterior tears. Knee Surg Sports Traumatol Arthrosc 2015 ;23(9): 2645 2650.

6. Saqib R, Harris J, Funk L. Comparison of magnetic resonance arthrography with arthroscopy for imaging of shoulder injuries: retrospective study.  Ann R Coll Surg Engl 2017; 99(4): 271 274.

7. Hayes ML, Collins MS, Morgan JA, Wenger DE, Dahm DL. Efficacy of diagnostic magnetic resonance imaging for articular cartilage lesions of the glenohumeral joint in patients with instability. Skeletal Radiol 2010; 39(12): 1199-1204.

8. de Jesus JO, Parker L, Frangos AJ, Nazarian LN. Accuracy of MRI, MR arthrography, and ultrasound in the diagnosis of rotator cuff tears: a meta-analysis.  AJR Am J Roentgenol 2009; 192(6): 1701 1707.

9. Chandnani VP, Gagliardi JA, Murnane TG, Bradley YC, DeBerardino TA, Spaeth J, Hansen MF. Glenohumeral ligaments and shoulder capsular mechanism: evaluation with MR arthrography. Radiology 1995; 196(1): 27 32.

10. Steinbach LS, Palmer WE, Schweitzer ME. Special focus session. MR arthrography. Radiographics. 2002;22(5): 1223 1246.

Dr Luthan Lam

Co-authors

Dr Khalid Ali