Data Loading...

MR Imaging of the Post-operative Shoulder Flipbook PDF

Introduction Interpreting MR Images of the post-operative shoulder can be a daunting task for the radiologist – there ar


121 Views
113 Downloads
FLIP PDF 2.04MB

DOWNLOAD FLIP

REPORT DMCA

MR Imaging of the Post-operative Shoulder William B. Morrison, M.D. RSNA 2005

William B. Morrison, M.D. Associate Professor of Radiology Chief, Division of Musculoskeletal Imaging Thomas Jefferson University Hospital 111 S. 11th St. Suite 3390 Philadelphia, PA 19107 Phone: 215-955-6226 Fax: 215-923-1562 Email: [email protected]

Introduction

Interpreting MR Images of the post-operative shoulder can be a daunting task for the radiologist – there are numerous surgical procedures, all with particular approaches and complications. Surgical artifact often makes the study even harder to evaluate. Nevertheless, MRI can provide a great deal of information to the referring clinician regarding patients with recurrent or new symptoms after surgery. The goal of this chapter is to provide the radiologist with rules of thumb for protocoling and interpreting MR imaging exams in this population, and to discuss the role of CT and ultrasound.

MR Imaging Protocol: Most shoulder surgeries fall into one of two categories: 1) rotator cuff / impingement surgery, and 2) labral / capsular surgery in patients with instability or pain from labral tear. It is very helpful to know the surgical history when protocoling and reviewing the exam. Most articles written about MR imaging of the post-operative shoulder utilize non-contrast MRI, which has been shown to be accurate for diagnosing recurrent pathology [1]. However, experience at our institution and others [2] suggests that for both types of surgery direct MR arthrography (MRA) can provide improved diagnostic capability over noncontrast MRI; MRA also improves reader confidence and reduces the radiologist’s learning curve.

Dealing with Post-operative Artifact on MR Imaging MR imaging protocol depends on the scanner used, but some rules of thumb apply (Table 1): first, in the postoperative shoulder, gradient echo sequences should be avoided due to the severe magnetic susceptibility artifact that results. Refocusing pulses used in spin echo imaging help reduce this effect. Artifact is also influenced by TE, bandwidth, field of view / matrix, and method of fat suppression. To reduce artifact, TE should remain relatively low, bandwidth should be increased, and field of view and matrix should be increased. Fast spin echo (FSE) technique results in decreased artifact compared to spin

echo imaging, so a T1w FSE sequence can be performed (keeping the echo train length low, e.g., 2, to minimize blur). Fat suppression using selective fat presaturation is prone to artifact in the setting of field heterogeneity, so if there is significant metal artifact, fat suppression should be removed from subsequent sequences. STIR, normally a good method for obtaining fluid information with fat suppression in a postoperative setting, is limited for direct MRA since gadolinium tends to be dark. All this being said, however, it is not usually necessary to employ all of these techniques unless the artifact is severe; for most cuff repairs and labral repairs, artifact, although present, does not limit diagnostic capability of standard protocols. If there has been burring of the bone (e.g. acromioplasty or tendon to bone repair), some elements of the metal suppression protocol may be needed.

Direct MR Arthrography The advantages of direct MRA are distension of the joint and improved signal to noise ratio (SNR). By injecting contrast into the joint, the capsular structures are separated from the labrum. Fluid is forced through/around pathologic labrum and cuff lesions. Because T1 weighted imaging is used, SNR is higher than an exam relying on relatively noisy T2 weighted or STIR images to diagnose pathology. This provides improved contrast resolution, which helps define small structures. MR arthrographic technique is slightly different depending on the type of surgery performed. For labral/capsular repair, excess distension should be avoided to reduce the chance of extracapsular leakage. Therefore, volume of injection is kept low; 12cc is adequate. For cuff repair/impingement surgery, it is beneficial to distend the joint to its endpoint to force contrast through cuff defects (although post-operative scarring of the subacromial / subdeltoid bursa can prevent contrast extension). Injection is performed until capsular resistance is reached (unless pain prevents full distension), which is typically after injection of 16cc if there is no cuff defect. A 23 mmol solution of gadolinium contrast is optimal; with most gadolinium preparations this amounts to a 1: 200 dilution in saline, or 1cc gadolinium in 200cc saline. If there is a delay expected before imaging, the syringe can be rinsed with 1:1000 epinephrine prior to the procedure before the gadolinium is

drawn up, in order to constrict synovial vessels and reduce rate of joint fluid resorption.

Indirect MR Arthrography Indirect MRA is useful for evaluation of the postoperative shoulder, but the technique is subject to some limitations. Indirect MRA is performed by administering a standard IV dose of gadolinium (0.1 mmol/kg) followed by delayed MR imaging (after approximately 30 minutes). The IV contrast is taken up by the joint fluid, providing an arthrographic effect on T1-weighted images. However, there is no joint distension unless the patient already had a joint effusion. Lack of joint distension may reduce sensitivity for detection of cuff or labral pathology. Also, signal to noise is not as great with indirect MRA, and postoperative granulation tissue may enhance. However, it can be useful if the patient refuses an intraarticular injection or if logistical considerations prevent performance of direct MRA.

Impingement / Rotator Cuff Surgery Acromioplasty Acromioplasty is a common procedure performed for primary impingement. The procedure can be performed with an open approach (typically anterolateral, via deltoid split) or arthroscopically via a posterior approach. The undersurface of the acromion is smoothed with a burr. Most surgeons try to preserve or reattach the coracoacromial ligament. Often this is combined with other procedures, including Mumford (resection of the distal clavicle) and/or rotator cuff repair. Because of the bone burring, acromioplasty usually results in significant MR artifact. Although this can make visualization of acromial morphology difficult, it does not usually preclude rotator cuff evaluation. Fluid in the subacromial / subdeltoid bursa is common and is not necessarily indicative of symptomatic bursitis. This fluid can be seen years after surgery and may enhance on indirect MRA. However, if the finding is prominent, correlate for symptoms or recurrent impingement, subacromial pain, or crepitus on range of motion.

In patients with postoperative symptoms, recurrent subacromial impingement is important to evaluate. This can occur in delayed fashion with reformation of spurs. It can also occur due to incomplete or overzealous acromioplasty. Therefore, the acromial morphology should be evaluated. Surgeons burr off the subacromial spur and attempt to make the undersurface flat; if there is a residual lateral point or angle at the midportion (Figure 1,2), recurrent impingement may occur. Conversely, an excessive acromioplasty can lead to instability. If the burr artifact is too severe, CT maybe more useful to evaluated the acromial morphology. Unrecognized os acromiale can also lead to recurrent impingement. The coracoacromial ligament is often lysed during acromioplasty, but currently surgeons try to preserve this ligament or re-attach it after acromioplasty. On sagittal MR images comment should be made regarding its status, intact vs. lysed, and whether it is thickened. A Mumford procedure (resection of distal clavicle) is usually reserved for patients with symptomatic acromioclavicular osteoarthritis, but it may be performed in conjunction with acromioplasty. Usually there is little artifact, and the appearance may be mistaken for an old acromioclavicular joint separation. Axial images offer the best plane for evaluation (Figure 3).

Rotator Cuff Repair Surgery of the rotator cuff may involve simple debridement of granulation tissue at a partial thickness tear, or tendon-to-tendon suturing, or tendon to bone repair of a full thickness tear. It helps to know the type of surgery performed, although to a certain extent, location of the suture artifact can indicate type of repair; tendon substance (tendon-to-tendon) vs. greater tuberosity (tendonbone). Usually the artifact is not very prominent and MR imaging is diagnostic. The “normal” appearance of a repaired tendon (Figure 4) is low signal on T2, similar to normal tendon, though there may be a residual vague increase in signal from tendinosis, or curvilinear high signal related to suture artifact. The repaired tendon may appear thickened or thinned, and an irregular-appearing margin is not necessary abnormal. There may be scar tissue in the subacromial / subdeltoid bursa, and this can blend with cuff signal.

One important concept is that surgeons debride granulation tissue and reattach torn tendons, but they do not always make the rotator cuff watertight. Therefore, high T2 signal within the repaired cuff tendon, or even communication from the glenohumeral joint to the subacromial / subdeltoid bursa may be an expected finding after the rotator cuff repair [3-5]. Therefore, MR evaluation of the postoperative rotator cuff requires alteration of the diagnostic criteria for rotator cuff tear. If there is fluid signal within the rotator cuff tendons in the region of repair, it should be commented on, but with the caveat that it may not be clinically significant. Some findings that indicate a clinically significant postoperative finding are (Figure 5) [3,6-8]: ƒ

Large (e.g. > 1cm) communication: describe size and location.

ƒ

Change from prior examination, if available.

ƒ

Large amount of fluid in subacromial/subdeltoid bursa.

ƒ

Displaced/broken sutures.

ƒ

Retraction of rotator cuff tendons.

ƒ

Muscle atrophy.

ƒ

Recurrent/residual subacromial spur.

ƒ

Glenohumeral osteoarthritis

ƒ

Superior migration of the humeral head

Direct MR arthrography is useful, not to demonstrate communication through the rotator cuff, but to distend the joint recesses and improve contrast resolution to better evaluate the anastomosis. On indirect MRA, findings can be misleading, since granulation tissue or scar at the intact anastomosis may enhance. CT arthrography and ultrasound also have potential value.

Deltoid Dehiscence A rare complication of rotator cuff surgery is deltoid dehiscence, in which the deltoid muscle (used as a surgical approach) becomes devitalized and detaches from the acromion. On MR images fluid will be seen extending through the deltoid muscle, usually near the acromial attachment (Figure 6).

Instability Surgery

Patients with symptomatic labral tears may undergo debridement or repair (Figure 7). Repair typically involves suturing of the labrum to the glenoid using suture anchors, which may be seen on MR images as a series of circleshaped artifacts at the glenoid rim. Suture anchors, in particular bioabsorbable anchors, do not usually create much artifact (in fact, it may be difficult to tell that surgery was performed), so standard post-contrast protocols with fat suppression can be employed. However, gradient echo sequences should still be avoided since any artifact will be accentuated. Anteroinferior labral lesions often involve capsular stripping or tear; therefore, re-attachment of the anterior capsule to the glenoid is often performed in conjunction with anteroinferior labral repair (together referred to as the Bankart procedure). Anterior capsular laxity associated with multidirectional instability is usually treated with rehabilitation, but if conservative therapy fails, an 'inferior capsular shift' may be performed, in which the anteroinferior capsule is tightened and tacked down to the proximal humerus or to the glenoid. Capsular laxity is occasionally treated with thermal shrinkage, a controversial therapy (some believe it may weaken the capsule and predispose to re-tear). Isolated capsular injuries (such as the HAGL lesion) are typically sutured. The repaired capsule may appear thickened on MR images, but should be watertight, without discontinuity. If a direct MR arthrogram is performed using an anterior approach, contrast may leak from the capsule normally, but the leakage is medial along the glenoid, under the subscapularis muscle. If contrast leaks anteriorly or inferiorly into the axilla, capsular re-tear should be considered. Superior labral tears (SLAP lesions) are generally debrided or reattached to the glenoid rim with sutures. The appearance is similar to labral repair in other locations. The repaired labrum, regardless of location, is evaluated in similar fashion (Figure 8,9). There is evidence that on noncontrast MRI and MR arthrography the postoperative labrum can be evaluated using the same criteria as the virgin labrum: if fluid or contrast extends into or under the labrum, it is compatible with a re-tear [2,9]. Indirect MRA may be limited in this regard, since the postoperative granulation tissue at the surgical site may enhance after IV injection, simulating a re-tear. Also, it should be recognized that the postoperative labrum might appear diminutive or truncated due to debridement of the tear.

Biceps Tear / Tenodesis Tear of the proximal portion of the tendon of the long head of the biceps may occur in association with rotator cuff tear or superior labral tear. If symptomatic, the tendon is often resected off the superior glenoid, and released, or fixated to the bicipital groove at the proximal humeral shaft (tenodesis). The biceps will be absent at the glenoid and in the proximal bicipital groove. To differentiate this from a tear, look for surgical artifact at the anterior humeral shaft, as the site of first visualization of the distal tendon (Figure 10).

Bioabsorbable Sutures / Anchors Bioabsorbable components are being used with increasing frequency. They are not visible on radiographs or CT, but are seen as signal void on MRI. A characteristic C-shape of some anchors can indicate their presence (Figure 11). These bioabsorbable materials, as their name suggests, dissolve over time. They may be released into the joint as they absorb [10]. If exposed to the joint, they may induce a reactive synovitis, which can be quite dramatic and can even simulate infection in rare circumstances. Release of these components into the joint is not necessarily abnormal, but if seen in the joint recesses, two points should be addressed: 1) Could the patient’s present symptoms be explained by synovitis induced by the bioabsorbable material? 2) Did the suture material get released before healing of the repair? If there is fluid or contrast in or under the site of labral repair, the suture likely released too early.

SUMMARY Evaluation of the postoperative shoulder can be a challenge, but application of basic principles of artifact reduction, a few rules of thumb in interpretation, and basic knowledge of surgeries performed and materials used for repair can make evaluation fairly straightforward.

REFERENCES

1. Magee TH, Gaenslen ES, Seitz R, Hinson GA, Wetzel LH. MR imaging of the shoulder after surgery. AJR 1997; 168:925-928.

2. Rand T, Trattnig S, Breitenseher M, Freilinger W, Cochole M, Imhof H. MR arthrography of the shoulder joint in a postoperative patient sample. Radiologe 1996; 36:966-970.

3. Zanetti M, Jost B, Hodler J, Gerber C. MR imaging after rotator cuff repair: fullthickness defects and bursitis-like subacromial abnormalities in asymptomatic subjects. Skeletal Radiol 2000; 29:314-319.

4. Spielmann AL, Forster BB, Kokan P, Hawkins RH, Janzen DL. Shoulder after rotator cuff repair: MR imaging findings in asymptomatic individuals – initial experience. Radiology 1999; 213:705-708.

5. Knudsen HB, Gelineck J, Sojbjerg JO, Olsen BS, Johannsen HV, Sneppen O. Functional and magnetic resonance imaging evaluation after single-tendon rotator cuff reconstruction. J Shoulder Elbow Surg 1999; 8:242-246.

6. Jost B, Pfirrmann CW, Gerber C, Switzerland Z. Clinical outcome after structural failure of rotator cuff repairs. J Bone Joint Surg Am 2000; 82:304-314.

7. Mansat P, Cofield RH, Kersten TE, Rowland CM. Complications of rotator cuff repair. Orthop Clin North Am 1997; 28:205-213.

8. Thomazeau H, Boukobza E, Morcet N, Chaperon J, Langlais F. Prediction of rotator cuff repair results by magnetic resonance imaging. Clin Orthop 1997; 344:275-283.

9. Wagner SC, Schweitzer ME, Morrison WB, Fenlin JM Jr, Bartolozzi AR. Shoulder instability: accuracy of MR imaging performed after surgery in depicting recurrent injury--initial findings. Radiology 2002; 222:196-203.

10. Majors NM, Banks MC. MR imaging of complications of loose surgical tacks in the shoulder. AJR 2003; 180:377-380.

TABLE 1: Methods for artifact reduction on MR images of the postoperative shoulder. - Remove fat suppression - Avoid use of gradient echo sequences - Add contrast (e.g., direct or indirect MR arthrography) - Use STIR instead of fat suppressed T2w sequence (unless Gd arthrogram; Gd may be dark on STIR) - Use FSE technique to acquire T1w images instead of SE (keep echo train length low, e.g., 2, to avoid blur artifact) - Increase bandwidth - Use a lower TE - Increase FOV and matrix - Swap phase/frequency to shift artifact in a different direction

FIGURES

Figure 1. Diagrammatic representation of acromial morphology before and after acromioplasty. A) Subacromial spur causing lateral impingement. B) Optimal acromioplasty with flat acromial undersurface. C) Suboptimal acromioplasty leaving a lateral ridge that may lead to recurrent impingement. D) Suboptimal acromioplasty in which the lateral spur has been removed, leaving a medial ridge that results in more proximal impingement.

A

B

C Figure 2. Acromioplasty on MR Imaging. A) T2w FSE fat suppressed oblique coronal image showing recent acromioplasty with residual postoperative marrow edema (arrow) and subacromial / subdeltoid bursal fluid (arrowheads). Marrow edema generally resolves within six weeks. Bursal fluid is usually asymptomatic and may persist for years after surgery. B) Oblique sagittal T2w FSE image showing postoperative defect in the anteroinferior acromion (arrow). Note thickened coracoacromial ligament (arrowheads). C) Recurrent impingement and rotator cuff re-tear. Oblique coronal T1w SE image with fat suppression after IV Gadolinium injection (indirect MR arthrogram) demonstrates a residual medial ridge (long arrow) adjacent to the acromioplasty site that abuts the humeral head. Contrast (short arrows) extends through the supraspinatus tendon, the end of which is retracted (arrowhead).

Figure 3. Mumford procedure. Axial fat suppressed T1w SE image through the acromioclavicular region shows apparent widening of the AC joint related to prior resection of the distal clavicle (arrow).

A

C

B

D

Figure 4. Normal MR appearance of rotator cuff repair. A-B) Tendon-to-tendon suturing of the supraspinatus tendon: appearance on direct MR arthrography. Coronal oblique T2w FSE fat suppressed image (A) shows edema at the acromioplasty site (short arrow) as well as at the tendon anastomosis (long arrow). Coronal oblique T1w fat suppressed SE image demonstrates lack of contrast extension from the joint through the tendon anastomosis (arrowhead), representing a watertight repair. Not all rotator cuff repairs are watertight. C) Tendon-to-bone repair on a coronal oblique T2w FSE fat suppressed image with normal thinning of the distal supraspinatus tendon (arrowhead) and fluid in the bursa (arrow). D) Tendon-to-bone repair and acromioplasty on a coronal oblique T2w FSE fat suppressed image with normal thickening of the distal supraspinatus tendon (arrow) and scarring of the subacromial/subdeltoid bursa (arrowheads).

A

B

C Figure 5. Rotator cuff re-tear. A-B) Direct MR arthrogram demonstrates a large communication (arrow) through the repaired supraspinatus tendon with retraction (arrowhead) on a coronal oblique T2w FSE fat suppressed image (A) and coronal oblique T1w fat suppressed SE image (B). Because of scarring of the bursa, there is limited distribution of the fluid and contrast that extended through the cuff. C) Coronal oblique T1w fat suppressed SE image from a direct MR arthrogram shows retraction of the sutured supraspinatus tendon (long arrow) from its insertion site on the greater tuberosity (arrowhead). Contrast extends from the joint into the bursa (short arrows).

Figure 6. Deltoid dehiscence. Oblique coronal T1w fat suppressed SE image from an indirect MR arthrogram shows enhancing fluid extending through a defect in the deltoid muscle (long arrow). Note also thinning of the supraspinatus tendon (arrowheads) and a distal full thickness communication (short arrow).

A

B

Figure 7. Normal MR appearance of labral repair. A) Coronal oblique T2w FSE image acquired shortly after dislocation shows a Hill-Sachs defect (arrow) and an anterior labral periosteal sleeve avulsion (ALPSA, arrowheads). B) Subsequent coronal oblique T2w FSE image after repair shows artifact from suture anchors at the glenoid rim (arrowheads). The capsule (arrow) is thickened. The repaired labrum-capsular complex is watertight, and the postoperative labrum can be interpreted using the same criteria as a non-operated labrum.

A

B

C Figure 8. Labral-capsular re-tear. A) Axial T1w fat suppressed SE indirect arthrographic image after labral repair shows contrast undermining the anterior labrum (arrow) consistent with a re-tear. Note suture anchor in the glenoid rim (arrowhead). B) Axial T1w fat suppressed SE direct arthrographic image shows anterior labral re-tear (arrow) as well as a redundant, irregular anterior capsule (arrowheads) consistent with breakdown of capsulorrhaphy. C) Oblique coronal T1w fat suppressed SE direct arthrographic image demonstrates contrast extending through a re-tear of the anterior capsule (arrow).

A

B

C Figure 9. SLAP repair and re-tear. A-B) Oblique sagittal T1w SE direct arthrographic image through the glenoid (A) shows suture anchors (arrowheads) from superior labral repair. Oblique coronal T2w FSE fat suppressed image (B) of the same patient shows intact labral repair (arrowhead). Note acromioplasty (arrow). C) Oblique sagittal T1w SE direct arthrographic image of a different patient after SLAP repair shows contrast extending under and into the superior labrum (arrow) consistent with re-tear.

Figure 10. Biceps tenodesis. Oblique coronal T2w FSE fat suppressed image shows absence of the long head of the biceps at its normal origin at the superior glenoid (long arrow). Artifact is seen at the proximal humeral shaft (arrowhead) related to tenodesis; the residual tendon (short arrows) extends distally from this site.

Figure 11. Displaced bioabsorbable material. Axial gradient echo indirect arthrographic image shows ‘C’ shaped bioabsorbable suture anchors (arrows) within the joint.