Received: 15 February 2018 Revised: 3 September 2018 Accepted: 27 September 2018

DOI: 10.1002/jcu.22661

RESEARCH ARTICLE

Accuracy and efficacy of ultrasound-guided pes anserinus

bursa injection
Jong H. Lee M.D. | Jae U. Lee M.D. | Seung W. Yoo M.D.

Department of Physical Medicine and

Rehabilitation, Dong-A University College of
Medicine, Busan, Republic of Korea
Correspondence
Jong H. Lee, Department of Physical Medicine
and Rehabilitation, Dong-A University College
of Medicine, DaesinGongwon-Ro 26, Seo Gu,
Busan, Republic of Korea, 602-715
Email: jhlee08@dau.ac.kr
Funding information
Dong-A University
Abstract
Purpose: To compare the accuracy and efficacy of ultrasound (US)-guided versus blind pes
anserinus bursa (PAB) injection in patients with pes anserinus tendinobursitis (PATB).
Methods: Forty-seven patients with clinically diagnosed PATB were randomly assigned to a US-
guided group or a blind group of steroid injection. In the US-guided group, the injectate was
delivered under sonographic visualization. In the blind group, the conventional technique was
used without any visual guidance. After the PAB injection, the injectate location was identified
using US in both groups. Treatment effects were assessed using the visual analogue scale (VAS)
of knee tenderness. Outcomes were measured before, 1 week and 4 weeks after the injection.
Results: Both groups showed pain relieving at 1 week and 4 weeks after the injection. The injectate
in the US guided group were found to be accurately at the PAB in all subjects, whereas blind group
were found to be just in 4 of 22 subjects. The US-guided group showed significant improvement of
both of VAS scores compared to the blind group at 1 week and 4 weeks after the injection (P < .05).
Conclusion: Our results suggest that US-guided PAB injection is more accurate and effective
than blind injection in patients with PATB.

KEYWORDS

bursa injection, pes anserinus tendinobursitis, ultrasound

1 | I N T RO D UC T I O N The prevalence of PATB has been reported at varying levels,

between 2.5% and 70%, in patients with knee pain10. In a retrospec-
The pes anserinus (PA) is the conjoint tendon of the sartorius, gracilis, tive review of 509 MRIs of the knees of 488 patients with suspected
and semitendinosus muscles, inserting onto the proximal anteromedial internal derangement, a 2.5% prevalence of PATB was detected at an
tibia approximately 5 cm distal to the medial tibial joint line.1–4 Deep orthopedic outpatient clinic.11 In a prospective study, a total of
to the PA tendon the PA bursa (PAB) is located, which serves to
170 knees of 85 patients with knee OA were examined with US, and
reduce friction between the superficially located PA tendon and the
the incidence of bursitis was 20% (34 of 170).12 They suggested that
structures that lie deep to it, including the medial tibia and the medial
PATB was observed in one of every five symptomatic OA patients
collateral ligament (MCL).3
and was more common in females and at advanced ages.
Changes in this region were first described in 1937, when
To date, the diagnosis of PATB is based on symptoms, including
Moschcowitz reported knee pain more frequently in women, who
pain in the medial aspect of the knee when going up or down stairs,
complained of pain when going up/down stairs or upon rising from a
sensitivity to palpation (digital pressure) on the area of insertion, and
chair, or with referred difficulty when flexing the knees.5 PA bursitis
occasional local edema.11 Disappearance of the pain after an injection
or tendinitis, also called PA tendinobursitis (PATB), a condition caused
by repetitive friction over the bursa or by direct trauma, occurs of local anesthetic can contribute to the diagnosis.12

frequently with osteoarthritis (OA), rheumatoid arthritis, diabetes The accuracy and efficacy of US-guided injections have been

mellitus, obesity, and valgus knee deformity. 6–8

PATB is common in
symptomatic OA patients. The distinction between PA bursitis and
tendinitis is clinically difficult due to the proximity of the tissues.
9
In addition, its pathology remains unknown and is controversial.
investigated in many studies, with comparisons to blind injections. Gil-
liand et al.13 published a systemic review about the accuracy and effi-
cacy of US-guided injections in the small joints of the knee, shoulder,
foot, ankle, wrist, and hand. Berkoff et al.14 reviewed the accuracy of

J Clin Ultrasound. 2019;47:77–82. wileyonlinelibrary.com/journal/jcu © 2018 Wiley Periodicals, Inc. 77

78
US-guided shoulder and knee injections in comparison with blind
injections, on the basis of specific data.
However, PAB injections, which are performed at a position
closer to the superficial layer than in shoulder or knee injections, has
not been studied sufficiently with respect to US guidance; they are
typically performed by blind injections in actual clinical settings.
10
Finnoff et al. compared the accuracy of US-guided PAB injections
and blind PAB injections using colored liquid latex in 24 lower extrem-
ities of 12 cadavers. In that study, the accuracy of the two groups was
very different, 92% versus 17%, despite the high proficiency of the
researchers and the superficial location of the bursa. However, as
cadavers were used, only the accuracy of US-guided injections could
be investigated, as the therapeutic effect of the accuracy in an actual
clinical setting could not be identified.
In addition to the implementation of US-guided injections, the
exact pathology of PATB which has been studied for more than
80 years is not completely known. Therefore, the injection location that
will have the optimum effect has not been sufficiently investigated.
The purpose of this study was to evaluate the accuracy of US-
guided PAB injections, and to assess the clinical outcomes of the effects
of injection accuracy by comparing them to blind injections. In addition,
the study assessed whether the location of injectate within the PAB or
superficial to the PA tendon could suggest the main source of pain.
2 | MATERIALS AND METHODS
2.1 | Subjects
This study included outpatients who visited our hospital from June
2016 to October 2017 with a clinical diagnosis of PATB, defined as
medial knee pain when climbing or going up/down stairs, tenderness
at the area of insertion, and occasional local swelling.1 They were
FIGURE 1 Patient flowchart
LEE ET AL.
randomly divided into two groups: the US-guided injection group and
the blind injection group. The randomization procedure was carried
out by an independent person. The allocation sequence was gener-
ated using Microsoft Excel. Seven patients were lost to follow-up, and
a total of 47 patients were analyzed (Figure 1). The inclusion criteria
were grade I-III Kellgren-Lawrence knee OA, medial knee pain, and
symptoms lasting for at least 3 months. These patients demonstrated
maximal tenderness over the PAB, not on the joint line. The exclusion
criteria were traumatic knee injury, surgical intervention at the knee,
systemic inflammatory disease, concomitant severe rheumatic disease,
or a diagnosis of fibromyalgia.
Approval for the study was obtained from the institutional review
board of our hospital, and written consent was received from all
patients.
2.2 | US-guided injections
Before the injection, the subjects were placed in the supine position
with the knee flexed at a 5-10 angle. The point of maximal tender-
ness was palpated and marked, and the skin was prepped with the
usual products. US imaging was then performed on the medial part of
the knee with an ACUSON X700 (SIEMENS, Sungnam, Korea, Inc.)
sonography device. We used a linear array probe with a 6-12 MHz
frequency and penetration depths of 2.5-4 cm according to the
patients' habitus. The sonographic study was performed by an expert
physician working in the department of rehabilitation medicine with
over 10 years of musculoskeletal US experience in diagnosis and man-
agement on each subject's knee. The needle was inserted at the point
where the PA crossed over the anterior fibers of the MCL, and the
transducer was positioned in a longitudinal orientation relative to the
anterior fibers of the MCL, with an oblique transverse orientation rela-
tive to the PA. While maintaining this image, a 25-gauge 38-mm stain-
less steel needle was introduced through the skin proximal to the
LEE ET AL.
FIGURE 2 A, US-guided injection, the transducer was positioned in a longitudinal orientation relative to the anterior fibers of the medial
collateral ligament, with an oblique transverse orientation relative to the pes anserinus. B, Blind injection, inserting the needle perpendicular to
the skin, into the point of maximal tenderness
transducer and advanced in a proximal to distal direction in a longitu-
dinal axis relative to the transducer, and advanced into the tissue
plane between the PA and the MCL. When the needle tip was visual-
ized between the middle of the PA and the MCL, the contents of the
syringe were injected under direct sonographic visualization
(Figure 2A and 3).
2.3 | Blind injections
The blind injection technique used in this study was similar to the
technique described previously.15 The same physician who performed
US guided injections provide the blind injections. The subjects were
placed in the same position as described above for the US-guided
injections. The point of maximal tenderness was palpated and marked,
and the skin was prepped with the usual products. The needle was
inserted perpendicular to the skin into the point of maximal tender-
ness until the physician felt the bone, then was withdrawn slightly
(2-3 mm) to avoid injecting directly into the conjoined tendon. The
contents of the syringe were then injected, and were expected to flow
easily (Figure 2B).
2.4 | Injection material
A mixture of 20 mg of triamcinolone acetonide and 1 mL of 1% lido-
caine was injected at the point of maximal tenderness in the PAB area.
After the PA injections in both groups, sonographic scanning was per-
formed to identify the location of the injection material. A second
operator working in the department of rehabilitation medicine with
FIGURE 3 US-guided injection, longitudinal ultrasound image of a
needle (white arrow) in the pes anserinus bursa (asterisk) between the
medial collateral ligament (MCL) and the pes anserinus (PA) tendon
79
over 6 years of musculoskeletal US experience in diagnosis and
management blinded to the injection assessed the injectate location.
2.5 | Evaluation methods
Response to injection was evaluated with a pain visual analog scale
(VAS), on which the patient was asked to mark the degree of pain
intensity ranging from 0 (absence of pain) to 10 (the worst pain ever).
As all of the patients who participated in this study had degenera-
tive arthritis of the knees, it was important to distinguish only the
effects of the PAB injections. Therefore, the VAS scores were mea-
sured while constant force was applied to the PA by an examiner with
a pressure algometer (FPK-5Wagner Instruments). This measurement
was taken before, 1 week and 4 weeks after the injection.
We did not prescribe any drugs, including nonsteroidal anti-
inflammatory drugs that could have an effect on pain during the study
period. Four weeks after the injection, treatment complications were
assessed via individual interviews.
2.6 | Postinjection management
All patients were warned about discomfort or pain after injection. To
avoid possible tendon rupture, the patients were advised to rest for
2-3 days with the application of ice three times daily, and to do mini-
mal squatting, kneeling, and bending of the knee for 2 weeks.
2.7 | Statistical analysis
The Windows SPSS 18.0 was employed. The Wilcoxon signed-rank
test was used to evaluate outcome measurements before and after
treatment in each group. For comparison between the two groups,
statistical processing was conducted with the Mann-Whitney U test.
A statistical significance level was set at P < .05.
3 | RE SU LT S
3.1 | General characteristics of the subjects
A total of 47 subjects with clinically diagnosed PATB and knee OA
were recruited. The US-guided group included 25 patients (2 men and
23 women) with a mean age of 65.36  7.12 years. The blind group
included 22 patients (2 men and 20 women) with a mean age of
80 LEE ET AL.

TABLE 1 Baseline characteristics

US-guided group (n = 25) Blind group (n = 22) P-value

Age (y) 65.36  7.12 63.40  6.20 .412
Sex (male/female) 2/23 2/20
Right/left 11/14 12/10
Knee OA duration (y) 7.21  6.33 6.35  6.24 .137
Knee OA K-L grade 2.20  0.56 2.15  0.57 .652
Baseline pain score
VAS (knee pain) 5.24  1.20 4.80  1.45 .255
VAS (PA tenderness) 6.82  1.45 6.45  1.12 .320

Values are presented as mean  SD.

Abbreviations: US, ultrasound; OA, osteoarthritis; K-L grade, Kellgren-Lawrence grade; VAS, visual analogue scale; PA, pes anserinus.

63.40  6.20 years. No statistically significant differences were any of the subjects in either group. No complications were noted in
observed between the groups with regard to general characteristics both groups.
(Table 1).

3.2 | Intragroup analysis
In the US-guided injection group, changes on the VAS for PA
tenderness revealed statistically significant pain reduction, from
6.82  1.45 at baseline to 2.28  1.08 1 week after the injection and
3.27  0.94 4 weeks after the injection (P < .05) (Table 2). In the
blind injection group, changes on the VAS for PA tenderness also
showed statistically significant pain reduction, from 6.45  1.12 at
baseline to 3.95  1.23 1 week after the injection and 4.60  0.95
4 weeks after the injection (P < .05) (Table 2).
3.3 | Intergroup analysis
Table 3 shows the comparison between the US-guided injection group
and the blind injection group in VAS score changes for PA tenderness.
The US-guided injection group showed significantly greater pain
reduction than the blind injection group with regard to ΔVAS scores
for PA tenderness at 1 week after injection (4.54  0.98 vs. 2.50
 1.03) and 4 weeks after injection (3.55  1.14 vs. 1.85  0.84).
3.4 | Placement of injectate
When injectates were located with US following the PAB injections,
they were found to be accurately located at the PAB, between the PA
tendon and the tibia or the MCL, in all 25 subjects of the US-guided
group (Figure 4A). In contrast, the injectates in the blind injection
group were found to be accurately located at the PAB in 4 of 22 sub-
jects, deep to the MCL in 2, and superficial to the PA tendon in
16 (Figure 4B). Injection into the PA tendon was not performed on
TABLE 2 Changes of VAS of PA tenderness comparing baseline
VAS US-guided group Blind group
Baseline 6.82  1.45 6.45  1.12
1 wk after injection 2.28  1.08* 3.95  1.23*
4 wk after injection 3.27  0.94* 4.60  0.95*
Values are presented as mean  SD.
Abbreviations: VAS, visual analogue scale; PA, pes anserinus; US,
ultrasound.
*P < .05 by Wilcoxon signed-rank test.
4 | DI SCU SSION
Although the majority of authors call the condition “anserine bursitis”,
the structure responsible for the symptoms is not identified in most
cases.
In our study, the US-guided injection group showed significantly
greater pain reduction than the blind injection group after the proce-
dure. In the US-guided injections, all injectates were located in the PAB,
whereas only 18% (4 of 22) of the blind injections' injectates were
properly located. This suggests that the pain source might relate to the
bursa, so a diagnosis of bursitis is more appropriate for this condition.
US is widely used as a diagnostic tool to improve the assessment
of joints and soft tissues. Several studies have demonstrated certain
morphologic US findings of the PA tendon and bursa in patients with
clinically diagnosed PATB. Uson et al.16 examined US features of the
PA and the subcutaneous medial knee fat in patients with clinically
diagnosed PATB syndrome. Of 37 patients, anserine bursitis was clini-
cally diagnosed in only three knees and PA tendinitis in one.
In that study, the authors did not find US evidence of tendinitis or
bursitis in patients diagnosed with PATB. Unlu et al.17 observed US evi-
dence suggestive of PA tendinitis in only 8.3% of 48 patients with type
2 diabetes mellitus. These patients with clinically diagnosed PA tendini-
tis or bursitis syndrome have less frequent morphologic US changes of
the PA tendons. In 2005, Yoon et al.2 evaluated the US examinations of
26 patients who were clinically diagnosed with PATB, and only
2 patients (7.7%) showed US evidence of PATB. One patient had anser-
ine tendinitis (both thickening and loss of normal fibrillar echotexture)
and the other had bursitis (circumscribed anechoic fluid collection of
2 mm or greater). A more recent study by Uysal et al.12 found a positive
TABLE 3 Changes of VAS of PA tenderness between two groups
US-guided group Blind group P-value
ΔVAS
1 week after injection 4.54  0.98 2.50  1.03 .000*
4 weeks after injection 3.55  1.14 1.85  0.84 .003*
Values are presented as mean  SD.
Abbreviations: VAS, visual analogue scale; US, ultrasound.
*P < .05 by Mann-Whitney U-test.
LEE ET AL.
FIGURE 4 Ultrasound image of the injectate location after injection A, Intra-pes anserinus bursa injection B, Extra-pes anserinus bursa injection;
the injection was superficial to the tendon
correlation between OA grade and PAB size and area. They evaluated
the US examinations of a total of 170 knees from 85 patients with knee
OA, and 20% (34/170) of the knees showed anserine bursitis. Toktas
et al.18 reported that the mean thickness of PA in knees with OA with/
without PATB was significantly greater than the controls in 183 PATB
clinical diagnoses among the 314 knees with OA.
From these numerous studies, patients with clinically diagnosed
PATB less frequently have morphologic US changes of the PA tendon
or bursa. For this reason, although the efficacy of US examination in
the diagnosis of PATB is controversial, the value of US-guided injec-
tion treatment is sufficiently appreciated with respect to accuracy.
As known from previous studies, US-guided injections have
greater accuracy than blind injections. Various recent reviews of the
clinical utility of US-guided injections in locations such as the shoulder
girdle19, hip joint20, and knee joint14 all reported that US guidance
correlated to greater accuracy compared to blind injections in all loca-
tions. Overall, the accuracy of US-guided injections is about 90%-
100%, while that of blind injections is 40%-80%. For most of these
injections, the enhanced accuracy achieved with US guidance directly
improves the clinical outcomes.
In the present study, consistent with previous studies, we
observed that US guided injection was more accurate than blind injec-
tion, and intra-bursal injection had a better clinical outcome in
treating PATB.
There were some limitations to our study. The long-term clinical
effects could not be determined because of the short term follow-up
period. In addition, the morphologic US findings of the PA tendon or
bursa were not evaluated, which might help to determine the source
of pain in PATB. And potential differences caused by activity level
were not considered despite their potential impact on the study
results. Further investigations are needed to resolve these limitations.
5 | CO NC LUSIO N
Both US-guided and blind PAB injections significantly reduced short-
term pain intensity. Accuracy was greater in the US-guided injection
group than in the blind injection group, and clinical outcomes were
more greatly affected in the US group.
After the US-guided injections, all injectates were located in the
PAB, while only a small amount of the blind injections' injectates were
located in the PAB. This suggests that the main pain source might
relate to the bursa.
81
ACKNOWLEDGMENTS
This work was supported by the Dong-A University research fund.
CONFLIC T OF INT E RE ST
No conflict of interest was declared by the authors.
ORCID
Jong H. Lee https://orcid.org/0000-0003-2489-358X
RE FE RE NC ES
1. Helfenstein M Jr, Kuromoto J. Anserine syndrome. Rev Bras Reumatol.
2010;50:313-327.
2. Yoon HS, Kim SE, Suh YR, Seo YI, Kim HA. Correlation between ultra-
sonographic findings and the response to corticosteroid injection in
pes anserinus tendinobursitis syndrome in knee osteoarthritis patients.
J Korean Med Sci. 2005;20:109-112.
3. Marra MD, Crema MD, Chung M, et al. MRI features of cystic lesions
around the knee. Knee. 2008;15:423-438.
4. Handy JR. Anserine bursitis: a brief review. South Med J. 1997;90:
376-377.
5. Moschcowitz E. Bursitis of sartorius bursa, an undescribed malady
simulating chronic arthritis. JAMA. 1937;109:1362.
6. Kang I, Han SW. Anserine bursitis in patients with osteoarthritis of the
knee. South Med J. 2000;93:207-209.
7. Brookler MI, Mongan ES. Anserine bursitis, a treatable cause of knee
pain in patients with degenerative arthritis. Calif Med. 1973;119:8.
8. Cohen SE, Mahul O, Meir R, et al. Anserine bursitis and noninsulin
dependent diabetes mellitus. J Rheumatol. 1997;24:2162.
9. Klippel JH. Primer on the Rheumatic Diseases. Musculoskeletal Signs and
Symptoms - Regional Rheumatic Pain Syndromes - Disorders of the Knee
Region. 12th ed. Atlanta: Arthritis Foundation; 2001:182.
10. Finnoff JT, Nutz DJ, Henning PT, Hollman JH, Smith J. Accuracy of
ultrasound-guided versus unguided pes anserinus bursa injections. PM
R. 2010;2:732-739.
11. Rennie WJ, Saifuddin A. Pes anserine bursitis: incidence in
symptomatic knees and clinical presentation. Skeletal Radiol. 2005;34:
395-398.
12. Uysal F, Akbal A, Gökmen F, Adam G, Reş orlu M. Prevalence of pes
anserine bursitis in symptomatic osteoarthritis patients: an ultrasono-
graphic prospective study. Clin Rheumatol. 2015;34:529-533.
13. Gilliland CA, Salazar LD, Borchers JR. Ultrasound versus anatomic
guidance for intra-articular and periarticular injection: a systematic
review. Phys Sportsmed. 2011;39:121-131.
14. Berkoff DJ, Miller LE, Block JE. Clinical utility of ultrasound guidance
for intra-articular knee injections: a review. Clin Interv Aging. 2012;
7:89.
15. Brock G, Gurekas V. The occasional pes anserinus bursitis injection.
Can J Rural Med. 2014;19:71.
82 LEE ET AL.

16. Uson J, Aguado P, Bernad M, et al. Pes anserinus tendino-bursitis: 20. Hoeber S, Aly AR, Ashworth N, Rajasekaran S. Ultrasound-guided hip
what are we talking about? Scand J Rheumatol. 2000;29:184. joint injections are more accurate than landmark-guided injections:
17. Unlu Z, Ozmen B, Tarhan S, et al. Ultrasonographic evaluation of pes a systematic review and meta-analysis. Br J Sports Med. 2016;50:
anserinus tendino-bursitis in patients with type 2 diabetes mellitus. 392-396.
J Rheumatol. 2003;30:352.
18. Toktas H, Dundar U, Adar S, Solak O, Ulasli AM. Ultrasonographic
assessment of pes anserinus tendon and pes anserinus tendinitis
bursitis syndrome in patients with knee osteoarthritis. Mod Rheumatol. How to cite this article: Lee JH, Lee JU, Yoo SW. Accuracy
2015;25:128-133. and efficacy of ultrasound-guided pes anserinus bursa injec-
19. Aly AR, Rajasekaran S, Ashworth N. Ultrasound-guided shoulder girdle
tion. J Clin Ultrasound. 2019;47:77–82. https://doi.org/10.
injections are more accurate and more effective than landmark-guided
injections: a systematic review and meta-analysis. Br J Sports Med. 1002/jcu.22661
2015;49:1042-1049.