| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Orthopaedic Surgery, The Sports Clinic, 23961 Calle de la
Magdalena, Ste. 229, Laguna Hills, CA 92653.
2 Lakes Regional Healthcare, South Spirit Lake, IA.
3 Laguna Niguel MRI, Laguna Niguel, CA.
Received November 27, 2007;
accepted after revision March 27, 2008.
Address correspondence to S. Porat
(hamlet9634{at}aol.com).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One hundred consecutive community-referred patients at our MRI facility underwent non-imaging-guided glenohumeral contrast injection targeting the rotator interval, with no patients excluded on the basis of expected diagnosis. The studies were then retrospectively reviewed for accuracy of injection based on patient factors and diagnosis.
RESULTS. This method was 99% accurate in our study, regardless of diagnosis.
CONCLUSION. The relative ease, efficiency, reproducibility, and accuracy of this method of non-imaging-guided anterior glenohumeral injection make it the method of choice at our institution, and we believe this technique merits consideration for more widespread utilization.
Keywords: accuracy • arthrogram • imaging guidance • injection • MR arthrography • rotator cuff • shoulder
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The method of contrast injection was as follows: All patients underwent unenhanced MRI before injection during the same visit as the planned arthrography. These images were reviewed for anomalies, pathology, and general anatomy of the coracoid relative to the rotator interval by the radiologist performing the procedure. The patient was then positioned on a standard gurney with the shoulder externally rotated and the arm at the side. The field was then prepared in a sterile manner with fenestrated drape over the patient's anterior shoulder. The sulcus between the lateral tip of the coracoid and the humeral head was palpated (Fig. 1). A 21-gauge x 1.5-inch needle on a 10-mL syringe with 1% lidocaine was inserted in a slightly cephalad direction at the anterior-lateral tip of the coracoid. The needle was slowly advanced while infiltrating the local anesthetic until resistance was lost, indicating intraarticular position. A 20-mL syringe filled with a solution of Magnevist (gadopentetate dimeglumine, Bayer HealthCare) and saline at a dilution of 1:200 was exchanged in a sterile manner onto the needle, and the joint was filled with 15–20 mL according to patient comfort and resistance to injection. The volume of injection was occasionally decreased in cases of altered joint volume, such as in frozen shoulder. The patient was then transferred to the MRI suite for contrast-enhanced imaging.
|
|
All postinjection MRI sequences were reviewed by the radiologist and independently by a sports medicine orthopedic fellow for accuracy of intraarticular placement. An injection was judged to be intraarticular and diagnostically purposeful using the following criteria: all of the injected contrast material is confined by the normal joint capsule; most contrast material is in the joint, however, some is extraarticular but related to spontaneous decompression into locations unrelated to injection path and commonly encountered, even in imaging-guided procedures, that is, the subscapularis fossa and distally through the biceps tendon sheath; contrast material is present in both intraarticular and in extraarticular locations explained by pathologic lesions such as rotator cuff tears or capsular defects; and most contrast material is intraarticular although a tiny amount is visible along the injection needle track (i.e., within the anterior deltoid muscle). Contrast material in any extraarticular space was considered extraarticular, or a "miss," unless explained by a communicating defect, excluding small amounts along the needle track.
The following data were collected: patient age and sex, side injected, MRI evidence of prior surgery, major MRI diagnosis, and status of contrast injection (in or out) based on the listed criteria.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An anterior approach was chosen because it is the method most often taught to radiologists during training and is accepted as both safe and reliable, even in imaging-guided techniques [10]. The concept of obtaining unenhanced MR images has been previously described by Zlatkin [12]. Of note, anterior portals in arthroscopy using 8-mm cannulas through the rotator interval are commonly used, usually without attempts at rotator interval closure. The rarity of complications from this approach supports the notion that the passage of a 21-gauge needle through this same path is safe and not likely to confound interpretation.
Concerning logistics and efficiency, patients were removed from the MRI scanner to an adjacent examination room for contrast injection, avoiding the potentially long distances between MRI and fluoroscopy resources often encountered in large centers. Patients could have been injected while still on the MRI table, although we did not do this. Fluoroscopically guided injections are inherently more time consuming, no matter how proficient the operator [10].
Adding epinephrine to the contrast solution is another strategy commonly used to decrease the rate of intraarticular re absorption and may partially offset the negative effects of prolonged waiting intervals between injection and imaging. This is important to maintain contrast mate rial in the joint in cases of prolonged waiting times between contrast injection and contrast-enhanced imaging. It has been suggested that epin ephrine adversely affects morbidity after arthrography either by direct synovial ir ritation or by prolonging synovial contact time with contrast medium irritant, and some have even proposed increased contrast con centrations to compensate for extended postinjection waiting periods [11, 13]. The addition of epinephrine is not necessary with our technique because waiting time between unenhanced and contrast-enhanced MRI sequences is minimal.
Sethi et al. [8] reported on the accuracy of anterior glenohumeral injection in 41 consecutive patients. Using a single needle pass and spot fluoroscopy to determine accuracy, 26.8% of injections were deemed intraarticular. They do not recommend their technique without radiologic guidance.
We believe that two important differences in methodology may explain the disparity between results. First, the ability to review the unenhanced MRI sequences before the injection attempt enhances the appreciation for anatomic variability in the coracoid process, soft-tissue depth, glenoid version, and resting position of the humeral head in the position of injection. This is proposed as a potential benefit of unenhanced imaging, not as a justification, because no patient was excluded on the basis of unenhanced sequences. Second, the operator must be willing to reposition the needle based on tactile feedback. It is our experience that despite being awake, patient tolerance for subtle needle redirecting is quite high. This remains true despite often contacting the humeral head and using only a few milliliters of local anesthetic. The method of Sethi et al. [8] uses "only a single pass of the needle...." Rigid adherence to this technique, without minor readjustments to needle position during the pass, represents a difference from our method. Our technique uses a single pass through the superficial soft tissues, however, occasionally with small adjustments made to needle position based on tactile feedback from deeper structures until resistance to anesthetic infiltration is lost. The syringe containing contrast material is not attached to the needle until the MR arthrographer is confident that the needle is within the glenohumeral joint.
Our method of contrast injection is an anterior approach, targeting the rotator interval (Fig. 3), although this technique differs somewhat from that of Dépelteau et al. [5]. First, there is no ionizing radiation in our technique. The elimination of radiation exposure is a major advantage to non-imaging-guided injections because exposure is cumulative [10]. Second, the depth of needle penetration is limited by the decrease in resistance to anesthetic infiltration, signifying intraarticular position. This is in contrast to the alternative of impaling the needle tip into the humeral head and retracting.
|
There are potential limitations to this study. Patient tolerance indexes were neither acquired nor compared with an imaging-guided technique. Examination times were not measured or strictly compared quanti tatively with image-guided procedures. All injections in this study were performed by a single experienced musculoskeletal radiol ogist trained in the procedure. We accept that there is likely to be a modest learning curve with this technique among radiologists who are less experienced with intraarticular injections and that some level of training will be required to gain competence. With careful attention to the selection of a suitable needle entry site by palpation of appropriate land marks, the sudden reduction in resistance to a lidocaine test injection is an extremely good indicator of a true intraarticular needle position. Granted, certain landmarks, such as the coracoid process or even the sulcus between the coracoid and humeral head may be difficult to palpate, especially in very muscular or obese individuals. In this circum stance, a slightly more vigorous manual compression during the localization procedure and subsequent needle insertion will enhance the likelihood of a successful outcome, even with a relatively short needle such as used in this study.
The acromioclavicular joint tends to be easily palpable, even in troublesome cases, and tends to lie along a vertical line joining the coracohumeral sulcus and can be used as a rough guideline in the localization process. Reflux of fluid into the needle hub during the lidocaine–contrast syringe exchange is, by experience, a very good secondary sign of successful needle positioning. A report of pain during the initial phase of the contrast injection must be interpreted with suspicion and repositioning of the needle is strongly encouraged. It is rare for a patient to report discomfort during an intraarticular injection of gadolinium contrast.
The relative ease, efficiency, repro ducibility, and accuracy of this method of non-imaging-guided anterior glenohumeral injection, as outlined by this review, make it the method of choice at our institution and we believe this technique merits consideration for more widespread use.
Acknowledgments
We thank William R. Barfield for his contribution to this work.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
