Radiography (2002) 8, 235240

doi:10.1016/radi.2002.0382
Evaluation of the symptomatic supraspinatus
tendona comparison of ultrasound and
arthroscopy
K. M. Venu, FRCS, Specialist Registrar in Orthopaedics*
D. C. Howlett, MRCP, FRCR, Consultant Radiologist
R. Garikipati, FRCS, SHO in Orthopaedics*
H. J. Anderson, FRCR, Consultant Radiologist
and A. V. Bonnici, FRCS, Orthopaedic Consultant*
*Department of Orthopaedics and Department of Radiology, Eastbourne General Hospital,
Kings Drive, Eastbourne, BN21 2UD, U.K.
Purpose: A prospective study was undertaken to determine the accuracy of
ultrasound compared with arthroscopy in the evaluation of the symptomatic supra-
spinatus tendon, and to identify whether ultrasound diagnosis was helpful in
pre-operative planning.
Methods: A total of 276 consecutive patients with shoulder impingement symp-
toms underwent ultrasound examination of the supraspinatus tendon. Of these
patients, 41 proceeded to open or arthroscopic surgical procedure on clinical
grounds, and in this group direct comparison with ultrasound ndings was made.
Results: There was full correlation between ultrasound and arthroscopy in the
diagnosis of a normal supraspinatus tendon, full-thickness tear, tendinopathy and
tendon rupture. There was some discrepancy between the two techniques. Two
patients with partial thickness tear on ultrasound had a full thickness tear at
arthroscopy. Ultrasound was able to identify intra-substance partial thickness tears in
three patients with supraspinatus tendon said to be normal at arthroscopy. Ultra-
sound helped plan the surgical approach and operative time needed in cases of
supraspinatus tendon rupture and full thickness tear.
Conclusion: In this study ultrasound was effective in the evaluation of the
symptomatic supraspinatus tendon, and was also able to diagnose intra-tendinous
lesions not visible at arthroscopy.
2002 The College of Radiographers. Published by Elsevier Science Ltd. All rights reserved.
(Received 3 May 2001;
revised 28 May 2002;
accepted 16 July 2002)
INTRODUCTION
Rotator cuff tears involving the supraspinatus
tendon (SST) are commonly seen in clinical practice
and are often associated with sub-acromial
impingement type pain. Imaging plays an import-
ant role in the diagnosis of such lesions, as it
is often not possible to clinically differentiate a
full-thickness tear (FTT) from a partial thickness
tear (PTT) or tendinopathy. Accurate pre-
operative diagnosis also helps to plan operative
treatment. A variety of modalities have been used
to image the rotator cuff, including computerized
tomography (CT), magnetic resonance imaging
(MRI) and arthrographic techniques.
Ultrasound (US) evaluation of rotator cuff has
been shown to be accurate for the diagnosis of
rotator cuff tears [17]. As a technique it is widely
Correspondence should be addressed to: Mr K. M. Venu,
47 Speranza Street, Plumstead, London SE18 1NX, U.K. Tel:
(00)44 (0) 20 88548416; E-mail: vmlkavarthapu@hotmail.com
10788174/02/040235+06 $30.00/0 2002 The College of Radiographers. Published by Elsevier Science Ltd. All rights reserved.
available, non-invasive and well tolerated. In this
study, we have compared the ability of US to
identify and differentiate SST pathologies, when
compared with the ndings at arthroscopy.
MATERIALS AND METHODS
A total of 276 consecutive patients with clinical
supraspinatus impingement syndrome were evalu-
ated by US over a two-year period, from June 1997
to June 1999. Of these, 41 proceeded to arthro-
scopic decompression, debridement or open repair
of the tendon on clinical grounds. In this group,
24 were male, 17 female and the mean age was
57 years (range 3479).
Two radiologists experienced in shoulder ultra-
sound performed the US examinations in all
patients using a linear array 510 MHz transducer,
with the shoulder internally rotated to visualize the
SST. Clinical data and plain radiographs were avail-
able at the time of performing the US examination.
Conventional longitudinal and transverse views of
the SST were obtained. In our institution, the biceps
tendon is routinely scanned as part of a shoulder
examination, although these ndings were not
recorded for the purpose of this study. The other
rotator cuff tendons are not routinely examined.
For the purpose of the study, the appearances of
the SST on ultrasound were classied into 5 cat-
egories (ultrasound and equivalent arthroscopic
images are included):
Normal tendon (Fig. 1a & 1b)
Tendinopathy (Fig. 2a & 2b)
Partial thickness tear (Fig. 3)
Full thickness tear
Rupture (Fig. 4a & 4b)
Diagnostic criteria for the classication:
Tendinopathy may be diagnosed if the tendon is
seen to be thickened and of patchy and often
decreased echogenicity.
A PTT was diagnosed in the presence of a
hypo-or hyperechoic tendon defect, not involv-
ing the full thickness of the tendon. This can be
an intra-substance partial-thickness tear not
involving the tendon surface.
A FTT was diagnosed if the hypo or hyperechoic
tendon defect involved the full thickness of the
tendon.
Rupture was diagnosed if the tendon was absent
with often only the retracted proximal tendon
visualized.
Initially all patients received a course of
shoulder physiotherapy and a variable number
of sub-acromial steroid injections as rst line of
management. The decision to proceed to arthro-
scopy was made on clinical grounds. The indication
for surgery was failure of clinical improvement with
conservative measures within one year of the onset
of symptoms. The surgeon knew the US diagnosis
prior to surgery and accordingly, the pre-operative
planning was made. All patients underwent
shoulder joint arthroscopy and bursoscopy. All
Figure 1 (a) Longitudinal US of a normal left SST.
Hypoechoic humeral head cartilage lies inferiorly (arrows)
with a line of increased echogenicity superiorly represent-
ing deltoid muscle fascia (curved arrow). (b) Bursoscopic
image of a normal SST (arrow).
236 VENU ET AL.
patients underwent an arthroscopic subacromial
decompression or rotator cuff debridement. Those
patients with a full thickness tear or tendon rupture
were treated with an open repair following arthro-
scopic decompression. In this group, pre-operative
sonographic diagnosis helped plan the surgical
procedure and allocation of appropriate theatre
time.
A single surgeon performed all the operations.
The average time to surgery from the time of US
was 6 months. The state of the SST was noted at
the time of surgery and comparison was then made
with the US ndings.
RESULTS
The results are summarized in Table 1. It can be
seen that there is a discrepancy between the US
and arthroscopic ndings in ve patients in the PTT
group. In three of these patients where US diag-
nosed intra-substance partial tear, arthroscopy was
normal. Two patients with PTTs at ultrasound had
FTTs at the time of surgery.
DISCUSSION
Rotator cuff disorders are commonly associated
with impingement syndrome, which can result from
outlet and non-outlet factors. Outlet factors such
as the shape and slope of the acromion and promi-
nence of the acromio-clavicular joint are common
causes of impingement. Non-outlet factors are less
frequently seen and include prominence of the
greater tuberosity, loss of humeral head depres-
sors, loss of gleno-humeral fulcrum, lesions of the
acromion and thickened sub-acromial bursa or cuff
[8]. The rotator cuff may impinge against the
coraco-acromial arch leading to a microtraumatic
process, giving rise to tears commonly affecting
the supraspinatus tendon [9]. Infraspinatus, sub-
scapularis, and teres minor tendons are much less
commonly involved [10, 11]. Tears can range
from partial or intra-substance tears to complete
Figure 2 (a) Longitudinal US of a right SST, which
is thickened and of diffusely abnormal echotexture
containing multiple poorly dened hypoechoic foci.
The appearances are consistent with tendinopathy.
(b) Bursoscopic image showing irregular and frayed
appearance on the bursal surface of the SST consistent with
tendinopathy (arrow).
Figure 3 Longitudinal US of a left SST, which contains a
focal hypoechoic intra-tendinous lesion (callipers) not
involving the bursal or articular aspects. This is consistent
with an intra-substance partial tear and was not visualized at
arthroscopy.
EVALUATION OF THE SYMPTOMATIC SSTA COMPARISON OF US AND ARTHROSCOPY 237
rupture of the tendon. They can involve the bursal,
articular or both surfaces.
Accurate diagnosis is necessary to plan the man-
agement of rotator cuff tears. Surgical treatment is
advocated if conservative measures fail. In our unit,
the gleno-humeral joint and the subacromial space
are initially assessed arthroscopically and either a
subacromial decompression or open tendon
repair performed depending on type of cuff lesion.
Most surgeons prefer to treat SST tendinopathy,
intra-substance tear and PTT by arthroscopic
sub-acromial decompression to control impinge-
ment, with or without debridement of the
tendon. Some authors prefer to convert a PTT to
FTT and repair the tendon directly [12]. Open
surgery is carried out for FTT, if the tendon is
thought to be repairable without tension. Accurate
pre-operative diagnosis helps plan the surgical
approach and allows allocation of appropriate
theatre time.
Shoulder arthrography is a useful diagnostic
procedure and can reliably detect full-thickness
tears and, at times, partial thickness tears on the
inferior (joint) surface of the cuff [1315]. It cannot
detect intra-substance tears, tendinopathy or
partial-thickness tears affecting the bursal surface
of the cuff. Diagnostic accuracy is increased with
CT or MR arthrography. Arthrography however
is an invasive procedure and is associated with
complications [13, 14].
Noninvasive imaging techniques for the evalu-
ation of rotator cuff pathology have considerably
improved over the past decade. Plain radiographic
ndings include narrowing of the acromio-humeral
space, decalcication of the greater tuberosity and
reversal of the normal convexity of the inferior
surface of the acromion, but these appearances are
not specic in the diagnosis of rotator cuff tears
[16]. MR imaging of the shoulder has been found to
be accurate in diagnosing FTTs, but the results are
variable with PTTs [1, 17, 18]. MR examination has
advantages over plain arthrography or US in that
the former technique provides information of the
whole rotator cuff, the glenohumeral and acromio-
clavicular joints. It also gives information about the
presence of signicant fatty degeneration of the
rotator cuff muscles, which may preclude a major
rotator cuff tear repair. It has been shown in some
early studies the limitations of the clinical utility of
MR due to the frequent occurrence of abnormal
Figure 4 (a) Longitudinal US demonstrates rupture of
right SST with distal tendon remnant demonstrated
(arrow). The proximal tendon has retracted out of eld of
view. (b) Arthroscopic image of a ruptured SST with
retraction. The retracted segment is held with forceps
(arrow).
Table 1 Summary of results
US ndings
Correlation at
arthroscopy
Normal 13 13
Tendinopathy 11 11
PTT 7 2
FTT 7 7
Rupture 3 3
TOTAL 41 36
238 VENU ET AL.
signal in the rotator cuff in asymptomatic individuals
[19]. However, in a recent study by Wright et al., it
was suggested that either fast spin echo inversion
recovery (FSEIR) images or fast-spin echo (FSE)
fat-saturated images with TE greater than 66 be
used to facilitate the differentiation of uid signal
from intermediate increased signal intensity in rota-
tor cuff imaging in order to rene the normal
variability of the rotator cuffs signal [20]. The cost
factor and limited availability mean that MRI is not
the rst line of investigation in many centres.
Ultrasound as a technique is attractive. It is safe,
widely available and well tolerated. Since Seltzer
et al. described the sonographic appearances of
tears of the rotator cuff in 1979; there have been
several other studies supporting the usefulness of
this investigation [21]. It has been shown that
operator skills and equipment specications play an
important role in obtaining accurate sonographic
results [22]. US has been found to be accurate in
diagnosing FTTs, but not reliable in detecting PTTs
[57]. Dynamic US studies have also been per-
formed to diagnose sub-acromial impingement
[6, 7]. Sonographic changes in the biceps tendon,
which accompany a lesion of the cuff, are well
understood [5]. Some authors have described direct
correlation between the diameter of the biceps
tendon and the extent of tears [5, 23]. Ultrasound
can reliably detect biceps tendon abnormalities
including dislocation, rupture, and tendinitis.
Our results showed that US was very accurate in
diagnosing the normal tendon, tendinopathy, FTTs
and ruptures of the SST. However only two out of
seven cases in the PTT group were correlated at
surgery. Three of these patients had a focal intra-
substance tear diagnosed at US, but the supra-
spinatus tendon appeared normal at arthroscopy.
Intra-substance tears are often not visible at
arthroscopy but are readily diagnosed by ultra-
sound. The diagnosis of intra-substance tears is
important, as they may be responsible for continu-
ing symptoms following normal arthroscopy. In the
remaining two patients said to have PTT at US, FTT
was found at surgery. It is likely that during 6
months from the time of US to surgery, the PTT
may have progressed to FTT.
CONCLUSION
Our study demonstrated that US is effective in
evaluating the symptomatic SST. Ultrasound is
accurate in identifying the normal tendon,
tendinopathy, FTT and SST rupture and is also able
to diagnose intra-substance tears of the SST not
visualized during arthroscopy. US is a useful
method of pre-operative assessment in sympto-
matic SST to plan the operative approach and for
allocation of theatre time.
ACKNOWLEDGEMENT
The authors would like to thank Nick Taylor for his
help to prepare the illustrations.
REFERENCES
1. Seibold CJ, Mallisee TA, Erickson SJ, Boynton MD,
Raasch WG, Timins ME. Rotator Cuff: Evaluation
with US and MR Imaging. Radiographics 1999; 19:
685705.
2. Brenneke SL, Morgan CJ. Evaluation of ultrasonog-
raphy as a diagnostic technique in the assessment of
rotator cuff tendon tears. Am J Sports Med 1992; 20:
2879.
3. Hodler J, Fretz CJ, Terrier F, Gerber C. Rotator
cuff tears: correlation of sonographic and surgical
ndings. Radiology 1988; 169: 7914.
4. Crass JR, Craig EV, Feinberg SB. Ultrasonography of
rotator cuff tears: a review of 500 diagnostic studies.
J Clin Ultrasound 1988; 16(5): 31327.
5. Middleton WD, Reinus WR, Totty WG, Melson GL,
Murphy WA. Ultrasound evaluation of the rotator
cuff and biceps tendon. J Bone Joint Surg Am 1986;
68-A: 44050.
6. Read JW, Perko M. Shoulder ultrasound: diagnostic
accuracy for impingement syndrome, rotator cuff
tear, and biceps tendon pathology. J Shoulder Elbow
Surg 1998 MayJun; 7(3): 2647.
7. Sonnabend DH, Hughes JS, Giuffre BM, Farrell R.
The clinical role of shoulder ultrasound. Aust N Z J
Surg 1997; 67(9): 6303.
8. Neer II CS. Cuff Tears, Biceps Lesions and
Impairment. Shoulder reconstruction, chapter 2,
4454.
9. Coeld RH. Current concepts review: rotator cuff
disease of the shoulder. J Bone Joint Surg 1985; 67-A:
974.
10. Gartsman GM. Arthroscopic acromioplasty for
lesions of the rotator cuff. J Bone Joint Surg Am 1990;
72: 16980.
11. Gartsman GM, Milne JC. Articular surface partial-
thickness rotator cuff tears. J Shoulder Elbow Surg
1995; 4: 40915.
EVALUATION OF THE SYMPTOMATIC SSTA COMPARISON OF US AND ARTHROSCOPY 239
12. Fukuda H, Hamada K, Nakajima T, Yamada N,
Tomonaga A, Goto M. Partial-thickness tears of the
rotator cuffA clinicopathological review based on
66 surgically veried cases. Int Orthop 1996; 20(4):
25765.
13. Goldman AB, Ghelman B. The double-contrast
shoulder arthrogram: A review of 158 studies.
Radiology 1987; 127: 655.
14. Hall FM, Rosenthal DI, Goldberg RP, Wyshak G.
Morbidity from shoulder arthrography: Etiology,
incidence, and prevention. AJR 1981; 136: 59.
15. Killoran PJ, Marcove RC, Freiberger RH. Shoulder
arthrography. Am J Roentgenol 1968; 103: 658.
16. Nixon JE, DiStefano V. Ruptures of the rotator cuff.
Orthop Clin North Am 1975; 6: 42347.
17. Bachmann GF, Melzer C, Heinrichs CM, Mohring B,
Rominger MB. Diagnosis of rotator cuff lesions:
comparison of US and MRI on 38 joint specimens.
Eur Radiol 1997; 7(2): 1927.
18. Shellock FG, Bert JM, Fritts HM, Gundry CR, Easton
R, Crues JV. Evaluation of the rotator cuff and
glenoid labrum using a 0.2-Tesla extremity magnetic
resonance (MR) system: MR results compared to
surgical ndings. J Magn Reson Imaging 2001; 14(6):
76370.
19. Sher JS, Uribe JW, Pasoda A, Murphy BJ, Zlatkin MB.
Abnormal ndings on magnetic resonance images of
asymptomatic shoulders. J Bone Joint Surg Am 1995;
77: 1015.
20. Wright T, Yoon C, Schmit BP. Shoulder MRI rene-
ments: differentiation of rotator cuff tear from
artifacts and tendonosis, and reassessment of nor-
mal ndings. Semin Ultrasound CT MR 2001; 22(4):
38395.
21. Seltzer SE, Finberg HJ, Weissman BN, Kido DK,
Collier BD. Arthrosonography: Grey-scale ultra-
sound evaluation of the shoulder. Radiology 1979;
132: 4678.
22. Weiner SN, Seitz WH. Sonography of the shoulder
in patients with tears of the rotator cuff: accuracy
and value for selecting surgical options. Am J
Roentgenol 1993; 160: 1037.
23. Wallny T, Wagner UA, Prange S, Schmitt O, Reich
H. Evaluation of chronic tears of the rotator cuff
by ultrasound. J Bone Joint Surg Br 1999; 81-B:
6758.
240 VENU ET AL.