Efficacy of ultrasonography-guided shoulder MR arthrography using a ...

Skeletal Radiol (2010) 39:575–579 DOI 10.1007/s00256-009-0793-8

TECHNICAL REPORT

Efficacy of ultrasonography-guided shoulder MR arthrography using a posterior approach Gokhan Gokalp & Abdurrahim Dusak & Zeynep Yazici

Received: 19 June 2009 / Revised: 9 August 2009 / Accepted: 20 August 2009 / Published online: 26 September 2009 # ISS 2009

Abstract Aim Shoulder MR arthrography has an important role in the assessment of rotator cuff lesions, labral tears, glenohumeral ligaments, rotator interval lesions, and postoperative shoulder status. Injection in direct MR arthrography can be performed with palpation, fluoroscopy, ultrasonography (US), or MRI. Recently, the posterior approach is the preferred method due to the presence of fewer stabilizers, absence of important articular structures and less extravasation, has been advocated. Our aim was to assess the efficacy of US-guided MR arthrography via a posterior approach on the glenohumeral joint. Materials and Methods Thirty MR arthrographies were performed on 29 patients. Ultrasonography (Xario, Toshiba) examinations were conducted by a wide-band 5–12 Mhz linear array transducer set to muscle-skeleton. Diluted contrast medium (1 ml gadolinium chelate and 100 ml saline, approximately 15 ml) was delivered into the glenohumeral joint space from between the humeral head and posterior labrum with a 20-gauge spinal needle. MRI examination was conducted by a 1.5 T scanner. Fat-saturated T1-weighted spin echo was applied on coronal, axial, and sagittal planes within the first 30 min after contrast material injection.

G. Gokalp (*) : A. Dusak : Z. Yazici Department of Radiology, Uludag University Medical Faculty, Görükle Kampusu, 16059 Nilüfer, Bursa, Turkey e-mail: [email protected] A. Dusak e-mail: [email protected] Z. Yazici e-mail: [email protected]

Results One (3.3%) arthrography was not successful due to technical reasons associated with obesity. Contrast extravasation around the infraspinatus and teres minör muscles was depicted in twelve examinations. One (3.3%) patient developed vasovagal collapse. Conclusion Ultrasonography-guided posterior approach is an easy, reliable, fast, and comfortable method in experienced hands. It may be an alternative for fluoroscopyguided shoulder MR arthrography. Keywords Shoulder MR arthrography . Ultrasonography-guided . Posterior

Introduction Shoulder magnetic resonance (MR) arthrography is a more sensitive method in detection of intraarticular pathologies compared with the conventional technique. While it can be conducted via direct injection of diluted gadolinium contrast medium into the joint, it may also be applied indirectly. Indirect MR arthrography is generally not preferred due to use of higher amounts of contrast medium and inadequate distinction of intraarticular structures as a result of failure to fill the joint proficiently [1]. Injection in direct MR arthrography can be performed with palpation, fluoroscopy, ultrasonography (US), or MR imaging (MRI). Shoulder arthroscopy and medications such as steroid therapy can be administered by palpation without resorting to an imaging modality [1]. Contrast injection is usually performed with fluoroscopy-guided anterior approach. However, deployment of ionizing radiation is a disadvantage. Interventions guided with US are evaluated to be valuable because the needle can be followed real-time during the procedure and no radiation is involved [2]. The

576

Skeletal Radiol (2010) 39:575–579

posterior approach is a useful technique, the value of which is enhanced by encountering fewer stabilizers, not invading the anterior structures where pathology is commonly encountered and by decreased contrast extravasation [2, 3]. Our aim was to assess the efficiency of US-guided MR arthrography with the posterior approach on glenohumeral joint.

Materials and methods MR arthrography was performed via US guidance by a posterior approach on 30 shoulders in 29 patients. The efficacy of the technique was evaluated relative to the success or failure to administer contrast medium into the glenohumeral joint and the number of attempts required to achieve that success. The study group was comprised of 10 males and 19 females, the age range was 23–66 (mean 51). Clinical indications for MR imaging were the anterior instability conditions (rotator cuff pathology, instability, labral lesions, glenohumeral ligament defect, capsular laxity). US (Xario, Toshiba) examinations were conducted by a wide-band 5–12 Mhz linear array transducer set to muscle-skeleton on patients in a lateral recumbency position. In order to make patients comfortable, the contralateral arm was placed under the head. The upper arm, the shoulder of which was to be used for the injection, was left to extend along the lateral side of the body in a neutral position. In order to prevent patients from moving anteriorly, a cushion was employed to support the anterior portion of the abdomen. The transducer was placed onto the humeral head. By following the humeral head posteriorly with the transducer, the glenoid rim and triangular hyperechoic labrum were found. The direction of the needle was determined by mildly angulating the transducer with the labrum presenting the best angle. Joint space between the humeral head and the labrum was targeted, the point of entry was determined from the side of the humeral head. Needle path was positioned as tangential to the humeral head curvature by targeting the free margin of the labrum. Point of entry was marked on the skin as parallel to the long axis of the transducer. The needle was inserted lateral to the transducer pointing medially toward the joint space. Transducer angle was set to approximately 30–40° in the axial plane (Fig. 1). A chelate of 1 ml gadolinium was diluted with 100 ml saline and ∼15 ml of the mixture was drawn with the injector. Following antisepsis of the skin, anesthesia was established with 2–5 ml 2% lidocaine. Diluted contrast medium was delivered into the glenohumeral joint space from between the humeral head and posterior labrum with a 20-gauge spinal needle. The echogenic needle could be observed real-time within the long axis plane of the transducer until insertion into the joint space. After seeing the tip of the needle within the joint space

Fig. 1 MR arthrography of a 55-year-old patient was ordered for rotator cuff tear. a. Position of the patient is seen in the photograph. By keeping the affected shoulder high, the patient is helped to lie down in a semiprone position. b Transducer is held parallel to the long axis of infraspinatus muscle-tendon junction at a level below spine of scapula, and insertion point of the needle is determined

and agreeing that we were in the right place, ∼15 ml of diluted contrast medium was delivered (Fig. 2). The signs indicating that we successfully inserted the needle into the joint space were absence of any difficulty during the injection and sensation of passing the joint capsule judged from the capsular tension. In case of any resistance during the injection, due to possibility of being in the cartilage or fibrocartilage, pulling back the needle a little with soft maneuvers helped a lot. However, if the resistance persisted, the exact location of the needle tip was checked with real-time US. In the absence of any resistance, injection is continued and the joint space was observed to expand in real-time. In some instances, this dilatation may not be well marked as a result of the anterior joint recess being the dependent area. In the current study, preparation of the patient and equipment along with performing the injection, took a mean duration of 10 min.


577

Fig. 2 MR arthrography of a 46-year-old patient was ordered for rotator cuff tear. a Anatomic structures are visualized for intraarticular injection: humeral head (thick arrow), glenoid labrum as a hyperechoic triangle (thin arrow), and glenohumeral joint space (target of

the needle tip) (arrowhead). b Needle tip is observed in the joint space (arrowhead). The intra-articular needle (thick arrow) and extraarticular needle (arrows) was shown. c Distension of the joint space following injection (arrow)

Fat-saturated T1-weighted spin echo (TR/TE, 704/12; field of view, 17 cm; matrix size, 256×128; slice thickness 3 mm, acquisition 1, with a 1-mm gap) was applied on coronal, axial, and sagittal planes within the first 30 min in patients who demonstrated no complications after the procedure. MRI examination was conducted by a 1.5 T scanner (Magneton Vision, Siemens, Erlangen, Germany). All patients were examined in the supine position by centering the glenohumeral joint with a phase-array surface coil.

sations did not affect the quality of the examination. Gas bubbles were not observed in any of the patients. One (3.3%) patient developed vasovagal collapse. The symptoms of the patient were resolved after resting for a while in a Trendelenburg position. Complications such as pain, skin sensitivity, swelling, hemorrhage, or reduced movement ability were not evident in any of the patients.

Discussion Results Contrast medium delivery into the glenohumeral joint space was successfully achieved on first attempt in 29 of 30 interventions (96.7%). Two or more attempts were not needed. One (3.3%) arthrography examination failed due to obesity and technical reasons (thick subcutaneous soft tissues and difficulty in good visualization of deep tissues with a superficial probe). A total of six (21%) cases were normal, three (10%) had tendinosis, ten (34.5 %) had partial rupture, and ten (34.5%) had complete rupture (Fig. 3a). Süperior labrum anterior posterior (SLAP) one and two ruptures were observed in three cases. One case demonstrated complete rupture of subscapsular tendon (Fig. 3b). Complete rupture in supraspinatus tendon was present in cases with SLAP and subscapular tears. Twelve arthrographies (40%) showed contrast medium extravasation from the glenohumeral joint to around the infraspinatus and teres minör muscles (Fig. 3b). Extrava-

MR arthrography of the glenohumeral joint is a valuable imaging method for the shoulder. By filling the joint space with contrast medium, subsequent capsular distension and discrimination of intraarticular structures lead to easier and more accurate evaluation of pathological changes compared with non-contrast shoulder MRI. It has an important role for assessment of rotator cuff lesions, labral tears, glenohumeral ligaments (GHL), rotator interval lesions, and postoperative shoulder status. Injection is usually performed via the anterior approach. In many patients, shoulder instability originates from pathologies of anterior articular structures. Delivery of contrast medium by the anterior approach reduces extravasation of the agent [1, 4, 5]. It may however lead to injury of the inferior glenohumeral ligament and labrum by the needle. Redondo et al. [6] advocate the anterior approach due to its fast, easy, and more tolerable characteristics. However, Shumacher reported the anterior approach as ineffective and recommended use of the posterior approach [7].

578


Fig. 3 T1-weighted fat-suppressed coronal and axial MR arthrography images. a Complete tear in the supraspinatus tendon (arrow). b Complete tear in the subscapular tendon (black arrow). Contrast extravasation around the infraspinatus muscle (thick white arrow)

Recently, the posterior approach is the preferred method due to the presence of less stabilizers, absence of important articular structures, and lesser impact of contrast medium over the physician’s comment in the posterior portion [2, 3]. Isolated infraspinatus tendon tears in the posterior portion are rare events while contrast extravasations within and around the infraspinatus can be more easily identified and differentiated morphologically from a tear [8]. The study of Catalano et al. [4] showed extravasation along the teres minor muscle and infraspinatus muscle in 71% and along the teres minor muscle in 29%. In the current study, we observed 40% contrast medium extravasation around the infraspinatus and teres minör muscles. However, since those extravasations occurred because of shoulder instability, anterior joint pathologies and have no effect on imaging interpretation they can be deemed as unimportant. Fluoroscopy-guided interventions expose patients and personnel to ionizing radiation. In some instances, when the follow-up is prolonged, this consequence becomes more evident. US-guided posterior MR arthrography has no such disadvantages. The procedure is relatively simple and causes no patient discomfort. Catalano et al. [4] needed a second injection in 13% of posterior injections performed with palpation. In experienced hands, success rate of fluoroscopy-guided injection is almost 100% [1]. Cicac et al. [9] managed to obtain 100% success by US-guided arthrography. In the current study, all procedures were successful at the first attempt, whereas only 1 of the 30 MR arthrography procedures failed because of obesity. Difficulties are encountered in cases with high body mass index. The posterior glenoid labrum is easily found by shifting the probe to the medial with mild rotatory and back-and-

forth movements after pinpointing the humeral head. When difficulty to locate is experienced in obese patients, a lowfrequency probe may be used. The posterior glenoid labrum is observed as a triangular hyperechoic structure. It is easier to see the area between labrum and humeral head when there is a certain amount of fluid. Some studies report that the suprascapular nerve and circumflex scapular vascular structures can be damaged as a result of the posterior approach. Moreover, inserting the probe from a point inferior to the teres minor muscle, is noted to be a potential risk for axillary nerve and posterior humeral circumflex artery injuries [10, 11]. Zwar et al. [2] preferred to insert the needle through the infraspinatus muscle transversely in an oblique direction extending from lateral to medial and reported no injury in those structures due to distance of the suprascapular and circumflex structures from this route. Farmer et al. [1] conducted 132 posterior injections which demonstrated no complications, which they explained as being a result of the small-sized needle used. In the current study, no complication associated with the aforementioned structures was seen. Vasovagal collapse may be seen in shoulder injections in cases where there is fear or emotional trauma [8]. One of our cases exhibited vasovagal collapse. A low number of cases was a limitation of our study. In conclusion, US-guided posterior approach is an easy, reliable, fast, and comfortable method in experienced hands. Generally, it reveals the causes of anterior stability more easily. Prevention of labral punctures and exclusion of ionizing radiation along with real-time visiualization of the glenoid labrum, make it a more reliable method and it may be an alternative for fluoroscopy-guided shoulder MR arthrography.


References 1. Farmer KD, Hughes PM. MR arthrography of the shoulder: fluoroscopically guided technique using a posterior approach. AJR. 2002;178:433–4. 2. Zwar RB, Read JW, Noakes JB. Sonographically guided glenohumeral joint injection. AJR. 2004;183:48–50. 3. de Villiers R. Anterior approach v. posterior approach: ultrasound-guided shoulder arthrogram injection. SA J Radiol. 2008;12:60–2. 4. Catalano OA, Manfredi R, Vanzulli A, Tomei E, Napolitano M, Esposito A, et al. MR arthrography of the glenohumeral joint: modified posterior approach without imaging guidance. Radiology. 2007;242:550–4. 5. Chung CB, Dwek JR, Feng S, Resnick D. MR arthrography of the glenohumeral joint: a tailored approach. AJR. 2001;177:217–9.

579 6. Redondo MV, Berná-Serna JD, et al. MR arthrography of the shoulder using an anterior approach: optimal injection site. AJR. 2008;191:1397–400. 7. Schumacher HR. Aspiration and injection therapies for joints. Arthritis Rheum. 2003;49:413–20. 8. Koivikko MP, Mustonen AO. Shoulder magnetic resonance arthrography: a prospective randomized study of anterior and posterior ultrasonography-guided contrast injections. Acta Radiol. 2008;49:912–7. 9. Cicak N, Matasovi T, Bajraktarevi T. Ultrasonographic guidance of needle placement for shoulder arthrography. J Ultrasound Med. 1992;11:135–7. 10. Vierola H. Ultrasonography-guided contrast media injection to shoulder joint using a posterior approach: a technique worth trying. Acta Radiol. 2004;45:616–7. 11. Valls R, Melloni P. Sonographic guidance of needle position for MR arthrography of the shoulder. AJR. 1997;169:845–7.