Skeletal Radiol (2010) 39:481–488
DOI 10.1007/s00256-010-0907-3
TECHNICAL REPORT
MR arthrographic assessment of suspected posteroinferior
labral lesions using flexion, adduction, and internal rotation
positioning of the arm: preliminary experience
Mary M. Chiavaras & Srinivasan Harish & Janet Burr
Received: 5 December 2009 / Revised: 31 January 2010 / Accepted: 4 February 2010 / Published online: 26 February 2010
# ISS 2010
Abstract
Objective Imaging the shoulder in the position of flexion,
adduction, and internal rotation (FADIR) may be useful in
characterizing lesions of the posteroinferior labrum. The
purpose of this preliminary study is to illustrate the
diagnostic utility of FADIR positioning in the assessment
and characterization of posteroinferior labral tears.
Materials and Methods In the FADIR position, the arm is
placed across the chest, with the hand on the contralateral
shoulder and palm facing outwards. FADIR positioning
was performed if there was a subtle or equivocal
abnormality of the posteroinferior labrum on conventional
MR arthrography sequences. A retrospective review of the
charts of 9 people who were imaged using FADIR
positioning in addition to routine MR arthrographic
sequences of the shoulder was performed. The review
included the indication for the study, documentation of
presence of clinical posterior instability, and surgical
correlation, where available.
Results In all 9 patients, FADIR positioning helped
confirm, exclude, or better characterize a posteroinferior
labral abnormality by increasing the diagnostic confidence.
Conclusion Flexion, adduction, and internal rotation posi-
tioning appears to be a useful adjunct in evaluating patients
with equivocal or subtle posteroinferior labral abnormalities
on conventional MR arthrography sequences.
M. M. Chiavaras : S. Harish
Department of Radiology, McMaster University,
Hamilton, ON, Canada
S. Harish (*) : J. Burr
Department of Diagnostic Imaging, St. Joseph’s Healthcare,
Hamilton, ON L8N 2A6, Canada
e-mail: sriniharish@gmail.com
Keywords MRI . Shoulder . Posterior labrum .
Arthrography . Instability
Introduction
Posteroinferior labral tears can result in significant morbidity,
particularly in athletes engaging in overhead or contact sports,
but may go clinically unrecognized because of their relative
infrequency. Compared with anteroinferior labral pathology,
the posterior counterparts are much less common with
posterior shoulder instability (PSI) comprising only 5% of
all cases of instability [1, 2]. PSI often presents with non-
specific symptoms of poorly localized shoulder pain or
fatigue with activity rather than actual symptoms of
instability [1]. The majority of cases of PSI are initially
treated with physiotherapy and rehabilitation and, only with
failure to respond to conservative therapy, is surgical
intervention considered [2–6]. Surgical management has a
better outcome if a lesion is identified at the level of the
posteroinferior labrum [7, 8]. Hence, optimal radiological
assessment is helpful to diagnose these lesions for adequate
pre-surgical planning.
Magnetic resonance arthrography is the imaging modality
of choice in assessing abnormalities of the glenoid labrum
[9–14]. Anteroinferior labral lesions that are occult on
conventional MR arthrographic sequences are often better
visualized when positioning the arm in abduction and
external rotation (ABER), the position of provocation for
anterior instability [15]. MR arthrography in the adducted,
internally rotated position has been noted to be highly
accurate in diagnosing anterior labral periosteal sleeve
avulsion (ALPSA) lesions of the anteroinferior labrum
[16]. We hypothesized that, in cases of suspected poster-
oinferior labral tears, imaging the shoulder in the provocative
482 Skeletal Radiol (2010) 39:481–488

18]. In this preliminary report, we present our experience

with 9 cases and illustrate the potential impact of FADIR
positioning in patients with posterior instability and poster-
oinferior labral tears in which the conventional MR
sequences are equivocal.

Materials and methods

Research Ethics Board approval was obtained for this study.

Fig. 1 Flexion, adduction, and internal rotation (FADIR) position a
without and b with coil placement. In this position, the shoulder is
forward flexed/adducted and the elbow flexed with hand on the
contralateral shoulder and the palm facing outward to achieve internal
rotation
position of flexion, adduction, and internal rotation (FADIR)
may be useful in uncovering and characterizing lesions.
There are a couple of reports in the literature that
mention imaging the shoulder in the FADIR position [17,
Over an 18-month period, we identified 9 patients out of 420
MR arthrogram examinations of the shoulder that had
imaging using FADIR positioning. We analyzed the clinical
charts and MR arthrograms of these 9 patients. The review
included the indication for the study as well as the
documentation of presence of clinical PSI and surgical
correlation, where available. In the FADIR position, the arm
is placed across the chest, with the hand on the contralateral
shoulder and palm facing outwards (Fig. 1). This provocative
position places traction on the posteroinferior labrum,
causing displacement, and increasing the conspicuity of a
labral tear, if present (Fig. 2). FADIR positioning was
performed if there was a subtle or equivocal abnormality of
the posteroinferior labrum on conventional MR arthrography
sequences.
MR technique
All patients underwent intra-articular injection, using an
anterior approach, of 10 cc of a 0.1% gadolinium solution
followed by conventional imaging on either a GE Signa HDx
1.5 T MRI (n=7) or a Philips Achieva 3 T TX MRI (n=2).
On the GE scanner, the US Shoulder PA coil was used for

Fig. 2 Schematic diagram of

the arm in both a the neutral
and b the FADIR positions. a
The neutral position demon-
strates an equivocal tear of the
posteroinferior labrum (dashed
black arrow). b The chondrola-
bral separation is clearly evident
when the arm is placed in
flexion, adduction, and internal
rotation (black arrow). White
asterisk=bicipital groove
Skeletal Radiol (2010) 39:481–488
Fig. 3 Localizer sequences in the FADIR position showing the planning of the axial sequence, which was carried out on the coronal localizer
Table 1 Summary of 9 patients
Patient Age/sex Clinical/surgical correlation
1 23/male “Anterior instability” on the
MR request. PSI and a
posteroinferior labral tear
proven arthroscopically
2 37/male PSI
3 24/male PSI professional elite
football player
4 23/male PSI
5 47/male Suspected SLAP
No PSI
Posterior shoulder pain with
throwing
6 49/male Suspected SLAP
No PSI
Shoulder pain post-MVA
7 40/male Suspected SLAP
No PSI
Shoulder pain post-MVA
8 20/male Previous anterior dislocation
No PSI
Conservative treatment
9 36/male PSI, but no labral tear on
arthroscopy
FADIR: flexion, adduction, and internal rotation; PSI: posterior shoulder instability; MVA: motor vehicle accident; SLAP: superior labral anterior
posterior
Conventional sequences
No anteroinferior labral tear.
Subtle posteroinferior
labral tear
Hyperintense fluid (not as
bright as gadolinium)
tracking between
posteroinferior glenoid
and labrum
Subtle undercutting of
gadolinium at posterior
chondrolabral junction
Posterior cystic intralabral
focus, raises possibility of a
Kim’s lesion
Small cystic lesion within
posteroinferior labrum, but
no definite chondrolabral
separation
SLAP lesion, but also
questionable posteroinferior
labral tear
Questionable irregularity
Posteroinferior labrum
Suspected tear, with small
paralabral cyst
Bankart lesion, but also
irregular posteroinferior
labrum with suspected tear
Subtle high signal at
8’o clock position
FADIR positioning
More clearly demonstrates
unsuspected posteroinferior
labral tear
Demonstrates unequivocal
gadolinium signal and labral
detachment confirming
unstable labral tear
Reveals chondrolabral
separation to better
advantage
Fails to reveal chondro-labral
separation confirming
diagnosis of Kim’s lesion
Demonstrates both the
chondrolabral detachment
and paralabral cyst filling
with gadolinium
Reveals chondrolabral
separation confirming
posteroinferior labral tear
Demonstrates high signal
(not as bright as
gadolinium) in keeping with
granulation tissue/healed
tear
No evidence of displacement
or contrast extension
No evidence of displacement
or contrast extension
483
Impact of FADIR
Increases diagnostic certainty
Increases diagnostic certainty
Increases diagnostic certainty
Further characterizes
posteroinferior labral
abnormality
Detects clinically
unsuspected posteroinferior
labral tear equivocal on
routine sequences
Detects clinically
unsuspected posteroinferior
labral tear equivocal on
routine sequences
Fails to reveal posteroinferior
labral tear suspected on
routine sequences
Fails to reveal clear
posteroinferior labral tear
Further characterizes
apparent posteroinferior
labral abnormality
484
Fig. 4 Posteroinferior labral tear. A 23-year-old man, whose history at
the time of the arthrogram was anterior instability. a Double oblique
axial T1-weighted fat-saturated image demonstrates a subtle, thin
streak of contrast at the posteroinferior chondrolabral junction (white
all routine sequences and the cardiac coil for FADIR
sequences. On the Philips scanner, the Philips Sense
Shoulder 4 coil was used for both the routine and FADIR
sequences. Standard MR sequences were all performed
with fat saturation and included sagittal T1-, axial proton
density-, double oblique axial T1- and coronal T1-, and
coronal T2-weighted sequences. The following MR param-
eters were used: FOV 14 cm (all sequences), matrix 256×
256–512×512, slice thickness 3.0 mm, and gap of 0.3 mm.
In our department, the MR arthrogram images are
routinely checked for adequacy, completion, and need for
extra sequences by a musculoskeletal radiologist and/or
fellow. If, at that time, the radiologist notes equivocal or
subtle abnormalities of the posteroinferior labrum, the
patients are repositioned with their arm in the FADIR
position (Fig. 2). In 8 patients, T1-weighted fat-saturated
sequences in the FADIR position were obtained in the axial
plane planned primarily on the coronal localizer (Fig. 3). In
Fig. 5 Posterior glenoid labral articular disruption. A 49-year-old man
with shoulder pain after a motor vehicle accident and clinically
suspected superior labral anterior posterior (SLAP) tear. a The patient
had a SLAP tear (not shown) with an additional thin streak of high
signal at the posteroinferior chondrolabral junction (white arrow) on a
Skeletal Radiol (2010) 39:481–488
arrow) raising the possibility of a tear. b FADIR axial T1-weighted
fat-saturated sequence reveals a convincing posteroinferior labral tear
(white arrow) and a loose fragment (black arrow). This was
arthroscopically proven
one of our initial cases, an axial PD fat-saturated FADIR
sequence was performed instead of a T1-weighted fat-
saturated sequence. The acquisition time for the FADIR
sequence including the localizer sequence was approxi-
mately 9 min. The acquisition time to perform the entire
study including the conventional sequences and the FADIR
was between 35 and 40 min. For the FADIR sequence, no
saturation bands were used. Sponges were used to maintain
rotation, provide support, and the arm was secured by
strapping to ensure immobility and patient comfort. No
significant respiratory or motion-related artifact was seen in
any of the cases. Graphic saturation bands were applied to
reduce breathing artifacts for the conventional sequences.
Evaluation
The 9 cases were reviewed by a musculoskeletal radiologist
and a musculoskeletal imaging fellow in consensus.
double oblique axial T1-weighted fat-saturated image. b FADIR axial
T1–weighted fat-saturated sequence shows an unequivocal labral
detachment (white arrow) with posterior displacement of the labral
fragment and accompanying chondral flap
Skeletal Radiol (2010) 39:481–488
Fig. 6 Posteroinferior labral
tear. A 37-year-old man with
clinical posterior shoulder insta-
bility (PSI). a Double oblique
axial T1-weighted fat-saturated
image demonstrates mild hyper-
intense signal tracking between
the glenoid and the posteroinfe-
rior labrum (white arrow), but
not as bright as gadolinium. b
The chondrolabral detachment is
more definitively demonstrated
on the FADIR axial T1-
weighted fat-saturated sequence
(white arrow)
Evaluation was subjective and included comparison of the
conventional sequences and FADIR sequence on a PACS
workstation. The evaluation was primarily to determine if
the FADIR sequence added value to the conventional
images. Surgical correlation was made, where available.
Results
A summary of the results is provided in Table 1. In 2 cases,
FADIR positioning increased conspicuity of posteroinferior
labral tears, which were clinically unsuspected (Figs. 4 and 5);
one of these was arthroscopically proven. In 2 patients with
posterior instability, the FADIR sequence increased diagnostic
certainty by revealing chondrolabral separation to better
advantage (Figs. 6 and 7). In 2 patients with posterior
instability, the FADIR sequence increased diagnostic certainty
in characterizing the posteroinferior labral abnormality (Figs. 8
and 9); one of these was arthroscopically proven. In 2 cases
without documented posterior instability, where a poster-
oinferior labral abnormality was suspected on the conven-
tional sequences, the FADIR sequence failed to reveal a
Fig. 7 Posteroinferior labral tear. A 24-year-old male professional
elite football player with clinical PSI. a Double axial oblique T1-
weighted fat-saturated image demonstrates no convincing gadolinium
at interface between labrum and posteroinferior glenoid at the
485
convincing chondrolabral detachment. In 1 patient with
posterior shoulder pain, but with no clinical suspicion of
PSI documented, the FADIR sequence increased diagnostic
certainty by revealing chondrolabral separation to better
advantage than the conventional images (Fig. 10).
Discussion
Posterior shoulder instability is a potentially debilitating
condition, which is generally seen in young adults in three
clinical settings: following posterior dislocation, after direct
impact to the shoulder without frank dislocation, and, most
commonly, as the sequela of repetitive microtrauma associated
with various sports [1, 2, 11, 12]. These include American
football, which employs defensive blocking techniques with
arms in flexed and adducted positions, weightlifting with
strenuous axial and posterior loading, and overhead sports,
such as swimming and baseball [1, 2, 12, 19].
The clinical presentation of posterior instability is
variable with some patients describing a sense of joint
instability or looseness while the majority report poorly
8 o’clock position (black arrow), although here there is some high
signal medial to the labrum (black arrowhead). b FADIR axial T1–
weighted fat-saturated sequence more clearly demonstrates the
chondrolabral separation (white arrows)
486
Fig. 8 Concealed posteroinferior labral tear and noncommunicating
sub-labral cyst consistent with Kim’s lesion. A 23-year-old man with
clinical PSI. a Axial proton density (PD)-weighted fat-saturated MR
arthrographic image demonstrates a cystic lesion in intimate contact
localized posterior shoulder pain [1, 2]. In both instances,
the symptoms are exacerbated with the arm in the flexed,
adducted, and internally rotated position [1, 2, 7].
Posterior instability is associated with multiple structural
abnormalities of the glenohumeral joint, including osseous,
muscular, and labroligamentous pathologies [17, 20, 21].
Although less common than anteroinferior labral pathology,
tears of the posteroinferior labrum are not an infrequent
cause of posterior shoulder pain and instability and are
often seen following traumatic posterior dislocation [22].
Many variants have been described such as the reverse
Bankart, the posterior labrocapsular periosteal sleeve
avulsion (POLPSA), Kim’s lesion, and posterior glenola-
bral articular disruption (GLAD) [17, 18, 23–25]. Famil-
iarity with these lesions and optimal imaging techniques are
necessary for their identification and to guide timely,
lesion-specific intervention.
In this preliminary study, we have demonstrated the
utility of the FADIR position in addition to conventional
MR sequences in evaluating patients with equivocal or
subtle posteroinferior labral abnormalities on standard MR
arthrography. In this provocative position, the posterior
band of the inferior glenohumeral ligament creates tension
Fig. 9 No posteroinferior labral
tear on arthroscopy. A 36-year-
old man with posterior instabil-
ity. a Axial T1-weighted
fat-saturated image demonstrates
a small amount of subtle in-
creased signal (white arrow) at
the posteroinferior labrum. b
FADIR axial T1-weighted fat-
saturated image reveals no
change with the absence of a
clear chondrolabral detachment
(white arrowhead). Patient had
posterior capsular laxity, which
was corrected arthroscopically
Skeletal Radiol (2010) 39:481–488
with the deep aspect of the labrum (white arrow). b FADIR axial T1–
weighted fat-saturated sequence fails to reveal a chondrolabral
separation (white arrow)
on the posteroinferior chondrolabral junction, better delin-
eating chondrolabral separation. In addition, lack of
displacement may serve to confirm the absence of a definite
posteroinferior labral tear or the presence of an incomplete
or concealed tear of the posteroinferior labrum. Finally, it
may assist in distinguishing a displaceable tear from a
healed tear filled with granulation tissue. Recently, a
posterior labral cleft has been described as a normal variant
in the posteroinferior labrum at the 8’o clock position,
which was proven to be a normal finding at arthroscopy
[26]. FADIR positioning may potentially help in differen-
tiation of a true tear from this cleft. Another feature we
noted whilst reviewing the images was that the contrast
medium appeared to move preferentially into the anterior
recess on the FADIR images compared with the standard
axial images. Therefore, the technique might be potentially
useful to assess the undersurface of the subscapularis
tendon and the contents of the biceps tendon sheath
including loose bodies and synovial fold thickening/
septations. Song et al. [16] described an adducted,
internally rotated sequence for demonstrating ALPSA
lesions. However, the shoulder was extended in the patients
in that study [16].
Skeletal Radiol (2010) 39:481–488
Fig. 10 Posteroinferior labral tear and communicating paralabral cyst.
A 47-year-old-man with posterior shoulder pain. a Axial PD-weighted
fat-saturated MR arthrographic image demonstrates a small cystic
lesion at the deep aspect of the posteroinferior labrum (white arrow). b
Double oblique axial T1-weighted fat-saturated image demonstrates
There are limitations to this study. There were only 9
cases; hence, this is a preliminary study. Surgical correla-
tion was present in only 2 patients. Many patients with
posterior instability are treated conservatively, presenting an
impediment to more comprehensive surgical correlation.
Other limitations include the lack of objective selection
criteria, given that the decision to perform FADIR was
based on finding the abnormality at the time of the scan.
The lack of FADIR sequences in patients without any clinical
posterior instability or posteroinferior labral abnormality on
conventional sequences leaves a lack of a standard for this
sequence with which to make comparisons. However, we
only perform FADIR if there is any suspicion of an
abnormality on the conventional sequences and to do so
with a completely normal posteroinferior labrum on the
conventional sequences might be unethical. Finally, as with
ABER, patients may have difficulty performing FADIR as
this is the provocative position for their symptoms.
In conclusion, we have shown that FADIR positioning is
a useful adjunct in the evaluation of suspected poster-
oinferior labral pathology. A prospective study with a larger
number of patients and utilizing well-defined selection
criteria and surgical correlation is necessary to fully
evaluate the utility of this technique.
487
no definite chondrolabral separation or paralabral cyst filling with
gadolinium (white arrow). c FADIR axial T1–weighted fat-saturated
sequence, however, clearly demonstrates both the chondrolabral
detachment and communication with the paralabral cyst (white
arrows)
Acknowledgements We extend thanks to all MR technologists at
St. Joseph’s Healthcare for their help with this study. We would also like
to thank Glen Oomen for preparation of the illustrations, Dr. Nick Plaskos
for his help.
References
1. Bradley JP, Forsythe B, Mascarenhas R. Arthroscopic manage-
ment of posterior shoulder instability: diagnosis, indications, and
technique. Clin Sports Med. 2008;27:649–70.
2. Robinson CM, Aderinto J. Recurrent posterior shoulder instabil-
ity. J Bone Jt Surg Am. 2005;87:883–92.
3. Millett PJ, Clavert P, Hatch III GFR, Warner JJP. Recurrent
posterior shoulder instability. J Am Acad Orthop Surg.
2006;8:464–76.
4. Tibone JE, Bradley JP. The treatment of posterior subluxation in
athletes. Clin Orthop Relat Res. 1993;(291):124–37.
5. Lamar DS, Williams GR, Iannotti JP, Ramsey ML. Posterior
instability of the glenohumeral joint: diagnosis and management.
Univ Pa Orthop J. 2001;14:5–14.
6. Antoniou J, Harryman II DT. Posterior instability. Orthop Clin
North Am. 2001;32:463–73.
7. Vidal LB, Bradley JP. Management of posterior shoulder
instability in the athlete. Curr Opin Orthop. 2006;17:164–71.
8. Papendick LW, Savoie III FH. Anatomy-specific repair techniques
for posterior shoulder instability. J South Orthop Assoc.
1995;4:169–76.
488
9. Jbara M, Chen Q, Marten P, Morcos M, Beltran J. Shoulder MR
arthrography: how, why, when. Radiol Clin North Am.
2005;43:683–92.
10. Hottya GA, Tirman PFJ, Bost FW, et al. Tear of the posterior
stabilizers after posterior dislocation: MR imaging and MR
arthrographic findings with arthroscopic correlation. AJR Am J
Roentgenol. 1998;171:763–8.
11. Beltran J, Rosenberg ZS, Chandnani VP, et al. Glenohumeral
instability: evaluation with MR arthrography. Radiographics.
1997;17:657–73.
12. Beltran J, Kim DH-M. MR imaging of shoulder instability injuries
in the athlete. Mag Reson Imaging Clin N Am. 2003;11:221–38.
13. Tirman PF, Palmer WE, Feller JF. MR arthrography of the
shoulder. Magn Reson Imaging Clin N Am. 1997;5:811–39.
14. Chandnani VP, Yeager TD, DeBerardino T, et al. Glenoid labral
tears: prospective evaluation with MRI imaging, MR arthography,
and CT arthrography. AJR Am J Roentgenol. 1993;161:1229–35.
15. Cvitanic O, Tirman PF, Feller JF, et al. Using abduction and
external rotation of the shoulder to increase the sensitivity of MR
arthrography in revealing tears of the anterior glenoid labrum.
AJR Am J Roentgenol. 1997;167:837–44.
16. Song HT, Huh YM, Kim S, et al. Anterior-inferior labral lesions
of recurrent shoulder dislocation evaluated by MR arthrography in
an adduction internal rotation (ADIR) position. J Magn Reson
Imaging. 2006;23:29–35.
17. Harish S, Nagar A, Moro J, et al. Imaging findings in posterior
instability of the shoulder. Skeletal Radiol. 2008;37:693–707.
Skeletal Radiol (2010) 39:481–488
18. Resnick D, Kang HS, Pretterklieber ML. Shoulder. In: Resnick D,
Kang HS, Pretterklieber ML, editors. Internal derangement of
joints, vol 1. 2nd ed. Philadelphia: Saunders; 2007. p. 713–1122.
19. Kaplan LD, Flanigan DC, Norwig, et al. Prevalence and variance
of shoulder injuries in elite collegiate football players. Am J
Sports Med. 2005;33:1142–6.
20. Shah N, Tung GA. Imaging signs of posterior glenohumeral
instability. AJR Am J Roentgenol. 2009;192:730–5.
21. Tung GA, Hou DD. MR arthrography of the posterior labrocap-
sular complex: relationship with glenohumeral joint alignment and
clinicalposteriorinstability.AJRAmJRoentgenol.2003;180:369–75.
22. Saupe N, White LM, Bleakney R, et al. Acute traumatic posterior
shoulder dislocation: MR findings. Radiology. 2008;248:185–93.
23. Simons P, Joekes E, Nelissen RG, Bloem JL. Posterior labrocap-
sular periosteal sleeve avulsion complicating locked posterior
shoulder dislocation. Skeletal Radiol. 1998;27:588–90.
24. Yu JS, Ashman CJ, Jones G. The POLPSA lesion: MR imaging
findings with arthroscopic correlation in patients with posterior
instability. Skeletal Radiol. 2002;31:396–9.
25. Kim SH, Ha KI, Yoo JC, Noh KC. Kim’s lesion: an incomplete
and concealed avulsion of the posteroinferior labrum in posterior
or multidirectional posteroinferior instability of the shoulder.
Arthroscopy. 2004;20:712–20.
26. Lee GY, Choi JA, Oh JH, Choi JY, Hong SH, Kang HS.
Posteroinferior labral cleft at direct CT arthrography of the
shoulder by using multidetector CT: is this a normal variant?
Radiology. 2009;253(3):765–70.