DOI:10.2214/AJR.07.3473
AJR 2008; 191:W58-W61
© American Roentgen Ray Society
CT-Guided Shoulder Arthrography at the Rotator Cuff Interval
Michael E. Mulligan1
1 Kernan CT/MR Imaging Center, 2202 Kernan Dr., Baltimore, MD 21207.
Received November 29, 2007;
accepted after revision March 4, 2008.
Address correspondence to M. E. Mulligan.
WEB
This is a Web exclusive article.
Abstract
OBJECTIVE. The purpose of this study was to describe a CT-guided
approach to the glenohumeral joint at the rotator cuff interval and to compare
that approach to the more traditional approach, including technical
considerations.
CONCLUSION. CT-guided glenohumeral joint injections at the rotator
cuff interval are faster and easier to perform than those performed using the
more traditional approach.
Keywords: CT • glenohumeral joint • rotator cuff interval • shoulder arthrography • technique
Introduction
Tears of the glenoid labrum are a common result of shoulder trauma and
overhead throwing activities in young patients. MR arthrography is advocated
as the imaging examination of choice by many radiologists and orthopedic
surgeons [1,
2]. MRI with direct injection
of contrast material into the glenohumeral joint is reported to be one of the
best methods of investigation to determine the presence or absence of such
tears. Needle placement for glenohumeral arthrography is usually performed
with fluoroscopic guidance, although some practitioners use sonography or may
be able to access the joint without imaging guidance. However, outpatient
imaging centers may only have CT and MR scanners on site. In this situation,
CT can be used to guide needle placement followed by MRI for complete
evaluation of the labrum, long head of the biceps tendon, glenohumeral
ligaments, and rotator cuff.
Arthrography of the shoulder has been performed by several different
methods since the technique was first described. Radiologists currently in
practice in the United States are likely to have been taught an anterior
approach popularized by articles published in the 1970s and 1980s
[3,
4]. More recently, use of the
anterior rotator cuff interval approach with fluoroscopic guidance has been
advocated. This approach is said to require less fluoroscopic time and be less
painful for the patient. It is also reported to be easier to teach to
residents. It avoids damage to the anterior labrum that is theoretically
possible with the traditional anterior approach
[5]. I have found the rotator
interval approach to be more useful than the traditional approach when using
CT to guide needle placement at an outpatient imaging center that does not
have a fluoroscopy unit available. The purpose of this article is to describe
the CT-guided approach to the glenohumeral joint at the rotator cuff interval
and to compare it to the more traditional approach.
Materials and Methods
Eighteen consecutive patients with requests for direct shoulder MR
arthrography were seen over a 9-month period at a new freestanding outpatient
imaging center that is equipped with only one CT scanner (LightSpeed, GE
Healthcare) and one 1.5-T magnet (Signa, GE Healthcare). Informed written
consent was obtained from each patient. There were 15 men and three women
whose ages ranged from 17 to 56 years (average age, 33.5 years). One
musculoskeletal radiologist with more than 20 years of experience performed
the glenohumeral joint injection in each patient using either a traditional
approach (first eight patients, Table
1) or a rotator cuff interval approach (last 10 patients,
Table 2) as described later.
Patients were not randomized.
The technique was changed from the traditional approach to the rotator cuff
interval approach after the first eight patients because of difficulties
encountered with the traditional approach. For either approach, the procedure
time listed in the tables is measured from the time of the first CT image
(scout or axial slice image) showing the needle in the soft tissues to the
first CT image showing contrast material in the glenohumeral joint.
Preparation time (skin cleansing, local anesthesia, and drawing solutions into
syringes) was the same for each approach and is not in cluded in the procedure
time listed in the tables. Subjective assessments related to procedure
difficulty and patient tolerance were noted.
View larger version (90K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —33-year-old woman with left shoulder pain. CT scout image for
left shoulder arthrography shows two paper clip markers present for
localization. Using standard fluoroscopic technique, one would aim for skin
entry at lateral and lower margin of lateral paper clip (arrow).
|
|
View larger version (83K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —33-year-old woman with left shoulder pain. Axial CT image
shows initial needle position using aiming point determined from A.
Note that needle entry position is far lateral to desired position
(asterisk).
|
|
Traditional Approach
Patients were positioned supine on a CT-compatible backboard (transfer
board) with the hand fully supinated to achieve external rotation of the
humeral head. Skin markers (paper clips) were placed over the humeral head.
The CT scout function was used to obtain a radiograph to localize areas for
needle placement. Standard spinal needles (18- to 20-gauge) were used. A point
along the medial edge of the humeral head near the junction of the middle and
lower thirds of the articular surface was chosen as an aiming point as
previously described. CT scout images or axial CT slices were obtained to
confirm intraarticular contrast injection (Figs.
1A,
1B,
2A, and
2B).
View larger version (107K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —30-year-old man with left shoulder pain. CT scout image for
left shoulder arthrography shows two paper clip markers present for
localization. Using standard fluoroscopic technique, one would aim for skin
entry along lateral edge of lateral paper clip (arrow).
|
|
View larger version (66K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —30-year-old man with left shoulder pain. Axial CT image with
paper clips still in place shows projected initial needle entry position
(dotted line) using aiming point determined from A. Note that
scout image distorts needle entry position such that it would be lateral to
desired position (asterisk).
|
|
Rotator Cuff Interval Approach
Patients were positioned as for the traditional approach. Skin markers were
placed at the level of the coracoid. Axial sections were obtained
(3–5-mm slice thickness) through the level of the coracoid and the
humeral head with the skin markers in place. Using the axial sections, the
midpoint of the coracoid was determined and the CT laser slice location light
was used to mark the correct level on the skin. Then, with skin markers as a
guide, the needle was positioned into the glenohumeral joint capsule at the
margin of the humeral head through the anterior rotator cuff interval for
installation of the arthrographic contrast material, either iodinated for CT
arthrography or gadolinium–saline mixture for MR arthrography. Axial CT
slices were obtained to confirm intraarticular contrast injection (Figs.
3A,
3B, and
3C)
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —22-year-old man with right shoulder pain. CT scout image for
right shoulder arthrography shows two paper clip markers present for
localization. Using scout image as guide at level of rotator interval, one
would aim for skin entry at medial edge of lateral paper clip
(arrow).
|
|
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —22-year-old man with right shoulder pain. Axial CT image at
level of rotator interval with paper clips still in place shows that scout
image distorts location of desired skin entry position determined from
A. Axial CT image should be used instead of CT scout image to guide
initial needle placement between limbs of medial paper clip (dotted
line)—not medial edge of lateral paper clip as suggested by scout
image in A.
|
|
View larger version (73K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —22-year-old man with right shoulder pain. Axial CT image
shows successful needle placement along margin of humeral head articular
surface at level of rotator cuff interval with intraarticular contrast visible
between humeral head and glenoid.
|
|
Results
Traditional Approach
The traditional approach was used with the first eight of our 18 patients
(six men, two women; average age, 39 years; age range, 29–56 years).
Procedure time ranged from 3 to 26 minutes, with an average of 12.5 minutes
(Table 1). The two longest
procedure times (23 and 26 minutes) in this group were due to multiple
redirections of the needle required to enter the joint capsule.
Rotator Cuff Interval Approach
The rotator cuff interval approach was used with the remaining 10 of the 18
patients (nine men, one woman; average age, 28 years; age range, 17–43
years). Procedure time ranged from 2 to 12 minutes, with an average of 4.2
minutes (Table 2). The longest
procedure time in this patient group (12 minutes) was due to technical
problems with the CT scanner.
As is evident from a review of the data in Tables
1 and
2, the rotator cuff interval
approach resulted in an average procedure time that was 8.3 minutes faster
than the traditional approach. Subjectively, the rotator cuff interval
approach was easier to perform and was better tolerated by the patients.
Because of the small number of patients, a statistical analysis was not
performed.
Discussion
Radiologists who are used to performing fluoroscopy-guided shoulder
arthrography will need to make some modifications in their technique to
consistently be able to inject the glenohumeral joint with ease using CT
guidance as described. This CT method is not preferred over traditional
fluoroscopic guidance but is useful when fluoroscopic guidance is not
available. One drawback of most CT scanners is lack of instant feedback
regarding needle position—a feature of fluoroscopic guidance. When using
CT, one must leave the patient and the room to review images acquired for
further guidance. CT scanners with real-time fluoroscopic capabilities are
available and can reduce the procedure time, but they are not commonly found
at outpatient imaging centers. One could also use an MR-compatible needle and
perform the injection in the MR room with MRI guidance; however, few
radiologists are trained in this technique. Methods to inject the glenohumeral
joint for MR arthrography without any imaging guidance have been described
[6] but are not performed by
many radiologists in the United States.
Technical difficulties with the traditional approach were encountered in
almost all patients in the first group. The initial needle placement always
seemed to end up in a more lateral position than expected and led to numerous
repositionings (Figs. 1A,
1B,
2A, and
2B). An exact number of
repositionings was not recorded because small corrections of the needle
direction are routinely made. It was obvious that one factor contributing to
the undesirable initial needle position was distortion of the CT scout image.
CT scout image distortion may not be appreciated by many radiologists because
the scout image is used primarily by the CT technologist to plan slice
positions for the CT examination. Such image distortion has been reported and
quantitated. Chidiac et al. [7]
compared measurements on skull specimens to those obtained with computed
radiographs and CT. They found good correlation for angular measurements and
vertical distances, but transverse linear measurements on CT did not correlate
well with the actual measurements from the skull specimens. This transverse
measurement error (image distortion) is thought to explain the difficulty with
lateral offset of initial needle entry position when one tries to rely on the
CT scout image versus an axial CT slice for glenohumeral arthrography. This
image distortion directly contributed to longer procedure times using the
traditional approach as described. Any improvement in procedure technique that
leads to a decrease in overall proce dure time is of clear benefit to the
patient and to busy radiologists.
This report has limitations. It describes one radiologist's experience with
two technically different CT-guided approaches to the glenohumeral joint for
contrast injection before MR arthrography. It is not intended as a scientific
investigation of either technique. Other radiologists may have different
results. The small number of nonrandomized patients does not allow significant
statistical analysis. A large prospective randomized study might be useful to
validate the findings of this report.
In conclusion, CT-guided glenohumeral joint injections at the rotator cuff
interval are faster and easier to perform than those performed using the more
traditional approach.
Acknowledgments
The author wishes to acknowledge and thank all the technologists who helped
to refine the CT-guided approach at the rotator cuff interval, especially
Christopher Sullivan.
References
- Magee T, Williams D, Mani N. Shoulder MR arthrography: which
patient group benefits most? AJR 2004;183
: 969-974[Abstract/Free Full Text]
- Waldherr P, Snyder SJ. SLAP lesions of the shoulder [In German].
Orthopade 2003;32
: 632-636[CrossRef][Medline]
- Schneider R, Ghelman B, Kaye J. A simplified injection technique
for shoulder arthrography. Radiology1975; 114:738
-739[Abstract]
- Resnick D. Shoulder arthrography. Radiol Clin North
Am 1981; 19:243
-253[Medline]
- Dépelteau H, Bureau NJ, Cardinal E, Aubin B, Brassard P.
Arthrography of the shoulder: a simple fluoroscopically guided approach for
targeting the rotator cuff interval. AJR2004; 182:329
-332[Abstract/Free Full Text]
- Catalano O, Manfredi R, Vanzulli A, et al. MR arthrography of the
glenohumeral joint: modified posterior approach without imaging guidance.
Radiology 2007;242
: 550-554[Abstract/Free Full Text]
- Chidiac JJ, Shofer FS, Al-Kutoub A, Laster LL, Ghafari J.
Comparison of CT scanograms and cephalometric radiographs in craniofacial
imaging. Orthod Craniofac Res 2002;5
: 104-113[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
