AJR 2003; 181:1227-1231
© American Roentgen Ray Society
Complications of Rotator Cuff Surgery in Which Bioabsorbable Anchors Are Used
Thomas Magee1,
Marc Shapiro,
Galen Hewell and
David Williams
1 All authors: Department of Radiology, Neuroimaging Institute, 27 E Hibiscus
Blvd., Melbourne, FL 32901.
Received March 25, 2003;
accepted after revision May 1, 2003.
Address correspondence to T. Magee
(tmageerad{at}cfl.rr.com).
Abstract
OBJECTIVE. This study assessed the utility of MRI in patients with
new or persistent pain after surgery with bioabsorbable rotator cuff
anchors.
SUBJECTS AND METHODS. Three musculoskeletal radiologists
prospectively reviewed MRIs of 30 patients with pain after rotator cuff repair
with fixation by rotator cuff anchors. Each radiologist described the location
of the rotator cuff anchors and whether the supraspinatus tendon was intact or
not. MRI findings were correlated with second-look arthroscopy. Consensus MRI
interpretations by the three radiologists were obtained retrospectively.
RESULTS. Of the 30 patients, nine had dislodgement of the rotator
cuff anchors from the humeral head along with a full-thickness supraspinatus
tendon retear. The dislodged rotator cuff anchor position could be determined
on coronal and sagittal MRIs, providing the orthopedic surgeon a preoperative
map for tendon reattachment and retrieval of the dislodged rotator cuff
anchor. Four patients had loose rotator cuff anchors but intact supraspinatus
tendons. Three patients had supraspinatus tendon retears, but the rotator cuff
anchors were intact. In all 16 patients, arthroscopic findings confirmed MRI
findings. Fourteen patients had intact rotator cuff anchors and intact
supraspinatus tendons on MRI. Of these 14 patients, five had second-look
arthroscopy confirming MRI findings.
CONCLUSION. MRI is useful in the assessment of patients with
persistent or new-onset pain after supraspinatus tendon repair with rotator
cuff anchors. MRI provides a presurgical map for second-look arthroscopy to
assess retear of the supraspinatus tendon and also aids in retrieval of
dislodged rotator cuff anchors.
Introduction
Rotator cuff anchors are implants used for soft-tissue attachment to the
bone. These anchors are increasingly being used for rotator cuff repairs
because they greatly facilitate the mini-open deltoid-splitting technique for
rotator cuff repair. The rotator cuff anchors obviate creating drill holes and
passing sutures through a bony trough. This feature greatly eases performing a
rotator cuff repair with limited open exposure as occurs with the mini-open
deltoid-splitting technique. Most of these anchors are produced by Mitek
(Ethicon, Piscataway, NJ) or Arthrex (Athrex, Naples, FL). The anchors measure
approximately 1 cm in length and 5–6 mm in width. These bioabsorbable
anchors are not visualized on conventional radiographs and are used to attach
a torn supraspinatus tendon to the humeral head. On occasion these anchors can
dislodge and must be retrieved because they function as loose bodies in the
joint [1,
2].
Detection of loose bodies in a joint can be difficult. Patients typically
present with pain, recurrent effusions, locking, or decreased motion.
Radiography can detect calcified or osseous bodies but not bioabsorbable
screws. MRI and MR arthrography are superior to CT and CT arthrography for
detection of loose bodies [3,
4].
The purpose of this study was to determine the utility of MRI in showing a
supraspinatus tendon retear in patients with these anchors and to determine
whether MRI is useful in showing the exact position of dislodged anchors for
presurgical planning.
Subjects and Methods
Three musculoskeletal radiologists prospectively reviewed MRI shoulder
examinations performed between January 2001 and August 2002 of 30 patients
with pain after rotator cuff repair and fixation by rotator cuff anchors. All
patients underwent preoperative MRI of the shoulder in the coronal, sagittal,
and axial planes performed on a 1.5-T Symphony scanner (Siemens Medical
Solutions, Erlangen, Germany). Oblique coronal fast spin-echo T2-weighted
(TR/TE, 4,000/54), oblique sagittal fast spin-echo T2-weighted (4,000/72),
oblique coronal spin-echo T1-weighted (507/12), and axial proton-density fast
spin-echo images (4,000/18) were obtained in all patients. All sequences were
obtained with a 12-cm field of view and a 4-mm slice thickness with a 10%
interslice gap. A shoulder array coil was used, the echo-train length on fast
spin-echo images was 7, and the matrix was 192 x 256.
At the request of the referring physician, 12 patients also underwent MR
arthrography, performed with approximately 25 mL of a diluted gadopentetate
dimeglumine–saline mixture (Magnevist, Berlex Laboratories, Wayne, NJ)
with a concentration of 0.15 mL of Magnevist per 20 mL of normal saline. A
22-gauge needle was placed in the glenohumeral joint via an anterior approach
with the assistance of fluoroscopy. One of three musculoskeletal radiologists
performed the injection. After injection of this mixture into the shoulder
joint, the shoulder was exercised and T1-weighted fat-saturated oblique
coronal, oblique sagittal, and axial images (677/12) were obtained.
T1-weighted fat-saturated oblique coronal images (677/12) were obtained before
MR arthrography for direct comparison with images obtained after MR
arthrography. All MR arthrograms were obtained with a 12-cm field of view and
a 4-mm slice thickness with a 10% interslice gap. A shoulder array coil was
used.
All MRIs were prospectively interpreted by one of three musculoskeletal
radiologists. The age range of the 30 patients was 36–74 years (mean, 57
years). The timing of postoperative MRI after rotator cuff surgery ranged from
2 months to 2 years 6 months after surgery. Each radiologist described the
location of the rotator cuff anchor and whether the supraspinatus tendon was
intact. A full-thickness supraspinatus tendon retear was defined as an area of
fluid signal traversing the entire thickness of the supraspinatus tendon on
T2-weighted images with or without supraspinatus tendon retraction. An intact
supraspinatus tendon was defined as a tendon clearly shown to attach to the
humeral head. Dislodged rotator cuff anchors were defined as anchors seen as
visibly detached from the humeral head and the supraspinatus tendon on MRI.
Intact rotator cuff anchors were defined as anchors or anchor tracts (for
those that had been adsorbed) visualized in the humeral head attachment site
of the supraspinatus tendon. MRI findings were correlated with second-look
arthroscopy. All MRIs were then retrospectively reviewed by three
musculoskeletal radiologists by consensus.
Results
Of these 30 patients, nine (30%) had dislodgement of rotator cuff anchors
from the humeral head with an associated supraspinatus tendon retear. All nine
patients had three rotator cuff anchors placed by the orthopedic surgeon
during the original surgery. The exact number and position of the dislodged
rotator cuff anchors were provided to the surgeon on the MRI report. In five
of these nine patients, two of the three rotator cuff anchors were dislodged.
In four of the nine patients, all three of the rotator cuff anchors were
dislodged (Figs. 1A,
1B,
1C,
1D,
2A,
2B,
2C,
3A,
3B,
3C). Three of the 30 patients
had one loose rotator cuff anchor shown on MRI, but the supraspinatus tendon
was intact by MRI criteria (Fig.
4A,
4B). One patient had two loose
rotator cuff anchors on MRI with the supraspinatus tendon intact by MRI
criteria (Fig. 5A,
5B,
5C). Three patients had
supraspinatus tendon retears, but the rotator cuff anchors were intact
(Fig. 6). In none of the
patients was the dislodged bioabsorbable screw visualized on radiographs. In
all these cases, second-look arthroscopic findings confirmed MRI findings.
Second-look arthroscopy was performed within 1 week of MRI interpretations in
all cases. The number and position of the rotator cuff anchors were the same
on second-look arthroscopy as on preoperative MRI interpretations.
View larger version (120K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —52-year-old man with shoulder pain after surgery. Oblique
coronal T2-weighted image (4,000/72) shows dislodged rotator cuff anchors
(straight arrows). Screw tract of intact anchor is seen in humeral
head (curved arrow).
|
|
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C. —52-year-old man with shoulder pain after surgery. Oblique
sagittal T2-weighted image (4,000/72) shows dislodged rotator cuff anchor
(straight arrow) and full-thickness supraspinatus tendon tear
(curved arrow).
|
|
View larger version (99K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A. —62-year-old man with shoulder pain after surgery.
Fat-saturated oblique sagittal T1-weighted MR arthrogram (TR/TE, 507/12) shows
dislodged rotator cuff anchor (arrow). In this patient, other images
(not shown) revealed that two anchors were loose, and supraspinatus tendon was
torn.
|
|
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —49-year-old woman with pain after shoulder surgery.
T2-weighted oblique sagittal image (TR/TE, 4,000/72) shows dislodged rotator
cuff anchor (straight arrow) and full-thickness supraspinatus tendon
tear (curved arrow). Second dislodged anchor was also present (not
shown).
|
|
View larger version (98K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —53-year-old man with pain after surgery. T1-weighted
fat-saturated oblique coronal MR arthrogram (TR/TE, 507/12) shows dislodged
rotator cuff anchor (straight arrow) without evidence of
full-thickness supraspinatus tendon retear (curved arrow).
|
|
View larger version (119K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A. —47-year-old man with shoulder pain after surgery. T1-weighted
oblique coronal image (507/12) shows two dislodged rotator cuff anchors
(straight arrows) and screw tract of intact anchor (curved
arrow).
|
|
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B. —47-year-old man with shoulder pain after surgery. T1-weighted
fat-saturated oblique coronal MR arthrogram (507/12) shows dislodged rotator
cuff anchors (straight arrows) and screw tract of intact rotator cuff
anchor (curved arrow). Despite dislodged anchors, supraspinatus
tendon is intact.
|
|
View larger version (104K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C. —47-year-old man with shoulder pain after surgery. T2-weighted
oblique sagittal image (4,000/72) shows screw tracts in humeral head (thin
arrows). Supraspinatus tendon (thick arrow) is intact.
|
|
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6. —58-year-old man with shoulder pain after surgery. T2-weighted
oblique coronal image (TR/TE, 4,000/72) shows full-thickness supraspinatus
tendon tear (thick arrow) and screw tract of intact anchors in
humeral head (thin arrows).
|
|
In 14 patients, intact rotator cuff anchors and intact supraspinatus
tendons were seen on MRI. Five patients underwent second-look arthroscopy
because of persistent pain. Second-look arthroscopy on these five patients
confirmed MRI findings. All MRIs were then retrospectively reviewed by all
three musculoskeletal radiologists by consensus. Prospective and retrospective
readings were the same with regard to findings of full-thickness supraspinatus
tendon tears and loose or intact rotator cuff anchors.
Discussion
Bioabsorbable anchors are commonly used in rotator cuff repair surgery
because drill holes are unnecessary and sutures are not passed through a bony
trough. Performing a rotator cuff repair with limited open exposure is thus
eased.
The anchors must have adequate strength to attach soft tissue to bone and
to allow healing tissue to regain mechanical integrity, and they also need a
slow adsorption rate so that breakage and migration do not occur
[5]. Usually three rotator or,
in rare cases of large supraspinatus tendon tears, four rotator cuff anchors
are placed by the surgeon to secure the supraspinatus tendon. All 30 patients
in our series had three rotator cuff anchors placed.
The anchors undergo five stages in bioabsorption. In stage 1, water is
absorbed into the rotator cuff anchor from the surrounding environment. In
stage 2, the polymer in the rotator cuff anchor undergoes hydrolysis resulting
in decreased holding strength. In stage 3, the rotator cuff anchor fragments
and begins to be adsorbed with resultant decrease in fixation strength. In
stages 4 and 5, the implant fragments are phagocytized, and the products enter
the Krebs cycle and are eliminated through respiration. The entire process of
bioabsorption takes from 5 months to 2 years
[5].
MRI is useful in the assessment of patients with new or persistent pain
after supraspinatus tendon repair with rotator cuff anchors. The bioabsorbable
rotator cuff anchor position is not seen on radiography. MRI allows clear
delineation of whether the anchor is dislodged and, if so, the exact position
of the anchor. This provides a presurgical map for second-look arthroscopy to
assess a retear of the supraspinatus tendon and also aids retrieval of
dislodged rotator cuff anchors.
The nine patients with detached rotator cuff anchors and supraspinatus
tendon retears all presented with symptoms within 6 months of the original
surgery. The four patients with one or two loose rotator cuff anchors but an
intact supraspinatus tendon all also presented with symptoms within 6 months
of surgery. In all these patients, the dislodged anchors were clearly visible
on MRI and had not been bioabsorbed at the time of imaging or at second-look
surgery. All these patients presented with symptoms of pain, recurrent
effusions, and locking or decreased mobility of the shoulder. These symptoms
are typical for patients with loose bodies in the joint space
[4,
6,
7].
The three patients with supraspinatus tendon retears but intact rotator
cuff anchors presented with symptoms of pain and weakness more than 1 year
after the original surgery. These symptoms are more typical of supraspinatus
tendon tears than of loose bodies. In these patients, screw tracts could be
visualized, but the screws seemed to be in the process of bioabsorption.
In the group of 14 patients with normal-appearing findings on MRIs after
surgery, five underwent second-look arthroscopy confirming a normal-appearing
postoperative shoulder. In this set of patients, none of the second-look
arthroscopies showed loose rotator cuff anchors.
Why the rotator cuff anchors occasionally fail is not clear. In this study,
rotator cuff anchor failure occurred within 6 months of surgery, before the
anchor was bioabsorbed. Loose rotator cuff anchors that are not partially
bioabsorbed are seen on MRI.
The presurgical knowledge of the number and position of the rotator cuff
anchors may make second-look arthroscopy more efficient. In two patients, one
of the anchors was positioned in the subdeltoid bursa. The presurgical
knowledge of the exact number and position of dislodged anchors allows the
surgeon to be confident of complete retrieval of the anchors. The dislodged
anchors must be retrieved because otherwise they function as foreign
bodies.
In this study, all patients underwent second-look arthroscopy within 1 week
of abnormal findings on preoperative MRI. Surgery was expedited in these
patients to decrease the chance of change in position of the dislodged rotator
cuff anchors from the time of MRI interpretation to time of surgery. Also,
performing surgery soon after the MRI interpretation decreased the chance of
additional anchors being dislodged in the interim from MRI interpretation to
the time of surgery.
In conclusion, MRI is useful in showing supraspinatus tendon retears and
the exact number and position of dislodged anchors in patients with failed
rotator cuff anchor repairs of the supraspinatus tendon. This presurgical
information, provided to the surgeon before second-look arthroscopy, may speed
second-look surgery, may allow for more efficient retrieval of the anchors,
and may allow the surgeon to be confident that all dislodged anchors have been
retrieved. Performing surgery soon after MRI interpretation ensures that no
interim change has occurred in number or position of dislodged rotator cuff
anchors.
References
- Mitek Web site. Ethicon, Piscataway, NJ. Available at:
www.jnjgateway.com.
Accessed February 2, 2003
- Arthrex Web site. Arthrex, Naples, FL. Available at:
www.arthrex.com.
Accessed February 2, 2003
- Brossmann J, Preidler K-W, Daener B, et al. Imaging of osseous and
cartilaginous intra-articular bodies in the knee: comparison of MR imaging and
MR arthrography with CT and CT arthrography in cadavers.
Radiology1996; 200:509
–517[Abstract/Free Full Text]
- Rand JA, Berquist TH. The knee. In: Berquist TH, ed.
Imaging of orthopedic trauma, 2nd ed. New York: Raven,1992
: 333–432
- Major NM, Banks MC. MR imaging of complications of loose surgical
tacks in the shoulder. AJR2003; 180:377
–380[Abstract/Free Full Text]
- Sartoris DJ, Kursunoglu S, Pineda C, et al. Detection of
intra-articular osteochondral bodies in the knee using computed
arthrotomography. Radiology1985; 155:447
–450[Abstract/Free Full Text]
- Berquist TH. Shoulder and arm. In: Berquist TH, ed. MRI
of the musculoskeletal system, 4th ed. Philadelphia: Lippincott,2001
: 578–680
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
