The Journal of Foot & Ankle Surgery xxx (2014) 1–5

Contents lists available at ScienceDirect

The Journal of Foot & Ankle Surgery

journal homepage: www.jfas.org

Original Research

Examining the Relation of Osteochondral Lesions of the Talus

to Ligamentous and Lateral Ankle Tendinous Pathologic Features:
A Comprehensive MRI Review in an Asymptomatic
Lateral Ankle Population
Melissa M. Galli, DPM, MHA, AACFAS 1, Nicole M. Protzman, MS 2, Eiran M. Mandelker, MD 3,
Amit D. Malhotra, MD 3, Edward Schwartz, DPM, FACFAS 4,
Stephen A. Brigido, DPM, FACFAS 5
1
Fellow, Department of Foot and Ankle, Coordinated Health, Bethlehem, PA
2
Research Associate, Department of Clinical Education and Research, Coordinated Health, Bethlehem, PA
3
Musculoskeletal Radiologist, Department of Imaging, Coordinated Health, Bethlehem, PA
4
Attending Physician, Department of Foot and Ankle, Coordinated Health, Bethlehem, PA
5
Fellowship Director, Department of Foot and Ankle, Coordinated Health, Bethlehem, PA

a r t i c l e i n f o a b s t r a c t

Level of Clinical Evidence: 3 Given the frequency and burden of ankle sprains, the pathologic features identified on magnetic resonance
imaging (MRI) scans are widely known in the symptomatic population. Ankle MRI pathologic features in the
Keywords:
cartilage asymptomatic population, however, are poorly understood. Such examinations are rarely undertaken unless
injury an ankle has been injured or is painful. We report the systematic MRI findings from the reports of 108
magnetic resonance image consecutive asymptomatic lateral ankles (104 patients). Our purpose was to (1) report the prevalence of
osteochondral defect osteochondral lesions of the talus (OLTs) and pathologic features of the medial and lateral ligaments, peroneal
surgery tendons, and superior peroneal retinaculum (SPR); (2) correlate the presence of OLTs with the pathologic
tibia features of the medial and lateral ligaments, peroneal tendons, and SPR; and (3) correlate ligamentous
discontinuity with the peroneal pathologic features, OLTs, and SPR pathologic features. A total of 16 OLTs
(14.81%) were present (13 medial and 3 lateral). Of the 16 patients with OLTs, 8 (50.00%) had concomitant
peroneal pathologic findings. Healthy medial and lateral ligaments were noted in 41 patients (37.96%), and
ligamentous discontinuity was grade I in 25 (23.15%), II in 32 (29.63%), III in 5 (4.63%), and grade IV in 5
patients (4.63%). A weak positive correlation was found between attenuation or tears of the superficial deltoid
and medial OLTs (phi coefficient ¼ 0.23, p ¼ .0191) and a moderate positive correlation between tears of the
posterior talofibular ligament and lateral OLTs (phi coefficient ¼ 0.30, p ¼ .0017). Additionally, a moderate
positive correlation between ligamentous discontinuity and tendinopathy of the peroneus brevis was noted
[Spearman’s coefficient(106) ¼ 0.29, p ¼ .0024]. These findings add to the evidence of concomitant pathologic
features in the asymptomatic population. To definitively assess causation and evaluate the clinical evolution of
radiologic findings, future, prospective, longitudinal cohort studies are necessary.
Ó 2014 by the American College of Foot and Ankle Surgeons. All rights reserved.

After ankle sprain, the prevalence and interrelatedness of mag-

netic resonance imaging (MRI)-noted ankle pathologic findings have
been well documented. However, denotation in the nontraumatic,
Financial Disclosure: None reported. asymptomatic population is scarce. Ankle sprains are one of the most
Conflict of Interest: S. A. Brigido has served on the surgery advisory board for common musculoskeletal injuries, affecting more than 30,000 in-
Alliqua and Bacterin International and as a consultant for Stryker; Alliqua, Bacterin dividuals everyday (1). Depending on the mechanism of injury,
International, and Stryker had no knowledge or influence in the study design, protocol, additional damage can occur, resulting in osteochondral lesions of the
or data collection.
Address correspondence to: Stephen A. Brigido, DPM, FACFAS, Department of Foot
talus (OLTs), peroneal tendon pathologic features, and loose bodies
and Ankle, Coordinated Health, 2775 Schoenersville Road, Bethlehem, PA 18017. (2–4). When left untreated or undertreated, repeat ankle sprains can
E-mail address: drsbrigido@mac.com (S.A. Brigido). lead to the development of chronic ankle instability. In patients with

1067-2516/$ - see front matter Ó 2014 by the American College of Foot and Ankle Surgeons. All rights reserved.
http://dx.doi.org/10.1053/j.jfas.2014.03.014
2 M.M. Galli et al. / The Journal of Foot & Ankle Surgery xxx (2014) 1–5

chronic ankle instability, isolated lateral ligament injuries will be Table 1

extremely rare and could become associated with tendinous and Inclusion and exclusion criteria for MRI selection

chondral damage. Numerous post-traumatic reports have described Inclusion criteria

the implications of ligamentous, tendinous, and cartilaginous ankle Asymptomatic lateral ankle
damage. No documented history of ankle trauma
Initial MRI scan read by musculoskeletal radiologist
DiGiovanni et al (2) demonstrated a high frequency of peroneal Exclusion criteria
tendon and retinacular pathologic findings as well as anterolateral History of ankle trauma
impingement lesions in patients with chronic lateral ankle instability. History of acute or chronic ankle dislocation
In a subsequent MRI evaluation, Stasko et al (5) presented the Lateral ankle pain
Previous lateral ankle surgery
concomitant incidence of OLTs and pathologic features of the peroneal
Comparison examination
tendons. Of the 76 patients identified with an OLT, 65% had MRI obtained at an outside institution
concomitant pathologic findings of the peroneal tendons. Collectively, MRI performed on a scanner with a <1.0 Tesla magnet
these findings suggest an association between injury to the lateral MRI procured with the use of gadolinium enhancement
ankle ligaments and pathologic features of the peroneal tendons, Abbreviation: MRI, magnetic resonance imaging.
retinaculum, and talus. However, the prevalence of these disorders in
the asymptomatic population is largely unknown. To our knowledge, scans and the clinic records that immediately preceded the ordering of the MRI to
no systematic, quantitative data regarding the prevalence of lateral determine whether the inclusion and exclusion criteria had been met (Table 1). For
inclusion, the medical record review had to indicate that the patient had an asymp-
ankle ligament injuries and the associated pathologic findings in
tomatic lateral ankle and no documented history of ankle trauma, and the initial MRI
asymptomatic lateral ankles are available. Subclinical information has scan had to have been read by a musculoskeletal radiologist at our institution (Table 1).
also been lacking for the medial ankle and, in particular, the rela- A history of any of the following items related to the ankle under observation served as
tionship between OLTs and the presence of deltoid injuries. specific exclusion criteria: any previous ankle trauma, acute or chronic ankle disloca-
tion, lateral ankle pain, and previous lateral ankle surgery. MRI examinations ordered to
The purpose of the present study was to review ankle MRI scans of
compare the pathologic extremity with the uninvolved, contralateral side (comparison
patients without a history of ankle trauma or lateral ankle pathologic examination) were also excluded, because these scans had not been read in a sys-
features and determine the prevalence of OLTs and the prevalence of tematic fashion. Furthermore, MRI scans that had been obtained at an outside insti-
pathologic features relating to the medial and lateral ankle ligaments, tution, performed on a scanner with a less than 1.0 Tesla magnet, or procured with
peroneal tendons, and superior peroneal retinaculum (SPR). Addi- gadolinium enhancement were excluded (Table 1). Data were recorded into a
password-protected, secure database. The confidentiality and privacy of the individuals
tionally, we assessed the relationship between the presence of OLTs
was ensured and maintained. According to the Coordinated Health institutional review
and pathologic features of the medial and lateral ankle ligaments, board, the present research project met the conditions for exemption. The protocol was
peroneal tendons, and SPR. We hypothesized that the presence of approved, and the requirement for informed consent was waived.
OLTs would correlate positively with the presence of medial and
lateral ankle, peroneal tendon, and SPR pathologic findings. Further- Endpoints
more, we anticipated that ligamentous discontinuity would correlate
positively with the presence of OLTs, peroneal tendon pathologic Two musculoskeletal radiologists re-assessed the MRI scans that met the inclusion
features, and SPR pathologic findings. criteria. According to previously published MRI radiographic techniques (6,7), the
medial ankle was inspected for superficial and deep deltoid attenuation and/or tears.
Patients and Methods Similarly, within the lateral ankle, the anterior talofibular ligament (ATFL), calcaneo-
fibular ligament (CFL), posterior talofibular ligament (PTFL), and syndesmosis were
Aims inspected for signs of attenuation and/or tears. Furthermore, the peroneal tendons
were evaluated for tendinopathy and tears. Tendinopathy was graded as mild, mod-
To better understand the pathologic findings within an asymptomatic lateral ankle erate, or severe, and tears were categorized as longitudinal split, partial, or full. A
patient population, the primary aim of the present retrospective study was to report the longitudinal split was suspected when a tendon had a “multipartite” appearance in the
prevalence of OLTs, medial and lateral ankle ligamentous injuries, peroneal tendon peroneal tunnel or when the peroneus brevis tendon had an “arrowhead” or “flame-
pathologic findings, and SPR pathologic features. We secondarily aimed to correlate the shaped” appearance below the tip of the fibula (8). OLTs were evidenced by a defect in
presence of OLTs with medial and lateral ankle ligament, peroneal tendon, and SPR the cartilage of the talar dome with underlying bony deformity and/or signal changes,
pathologic findings. Our tertiary goal was to correlate ligamentous discontinuity with which manifested as a low signal on the T1-weighted images and an intermediate or a
the peroneal pathologic features, OLTs, and SPR pathologic findings. high signal on the fluid-sensitive images (9). The presence or absence of SPR pathologic
findings was also noted and, if present, was classified as acute or chronic. Acute injuries
Assessors of the SPR demonstrated disrupted retinacular fibers and extensive swelling of the
surrounding tissue. Chronic injuries of the SPR demonstrated a thickened and het-
Using our electronic imaging system, a single fellowship-trained surgeon (M.M.G.) erogeneous SPR with little or no swelling of the surrounding soft tissue. Finally, the
reviewed consecutive ankle MRI scans ordered within Coordinated Health during the information realized from inspection of the ankle’s medial and lateral ligaments was
observation period, from December 27, 2011 to April 9, 2013, their ordering criteria, and used to determine the grade of ligamentous discontinuity, which was categorized as
the associated radiology reports to determine whether the inclusion and exclusion follows: attenuation and/or tears of the ATFL alone were classified as grade I;
criteria had been met (Table 1). The MRI scans were filtered by body part, using the concomitant attenuation and/or tears of the ATFL and CFL were classified as grade II;
electronic search term “ankle.” Selected patient demographics were recorded, concomitant attenuation and/or tears of the ATFL, CFL, and PTFL were classified as grade
including patient age (years), gender (male or female), and imaged side (right or left). III; and concomitant attenuation and/or tears of the ATFL, CFL, PTFL, and deltoid were
The MRI scans that met the inclusion and exclusion criteria were simultaneously classified as grade IV.
re-evaluated by 2 musculoskeletal trained radiologists with 19 years of combined
experience (A.D.M., E.M.M.). When a disagreement occurred, the 2 musculoskeletal
Table 2
radiologists reviewed the images simultaneously by telephone. Consensus on the
Statistical description of patients with asymptomatic lateral ankles (N ¼ 108 reports)
presence and nature of the pathologic findings was reached and recorded. Statistical
analyses were performed by 1 of us (N.M.P.), who also serves as a research associate at Demographic Variable Value
our institution.
Age (yr) 41.9  20.42
Gender
Study Population
Female 71 (65.74)
Male 37 (34.26)
A database search was conducted in the Department of Radiology at Coordinated
Imaged ankle (side)
Health for all ankle MRI examinations ordered by attending orthopedic and podiatric
Left 54 (50)
physicians from December 27, 2011 to April 9, 2013. The MRI scans were filtered by
Right 54 (50)
body part, using the electronic search term “ankle.” The search returned 500 consec-
utive ankle MRI examinations. One fellowship-trained surgeon examined both the MRI Data presented as mean  standard deviation or n (%).
 M.M. Galli et al. / The Journal of Foot & Ankle Surgery xxx (2014) 1–5 3

Table 3 Table 5
MRI ascertainment of ligamentous pathologic features observed in asymptomatic MRI ascertainment of peroneal tendon pathologic features observed in asymptomatic
lateral ankles (N ¼ 108 MRI scans) lateral ankles (N ¼ 108 MRI scans)

Ankle Ligamentous Outcome Prevalence Tendon Pathologic Finding Outcome Prevalence

Structure Involved
Intact Attenuated Peroneus Brevis Peroneus Longus
and/or Torn Tear
Medial ankle ligaments None 104 (96.3) 108 (100)
Total 150 (69.44) 66 (30.56) Longitudinal split 3 (2.78) 0
Superficial deltoid 79 (73.15) 29 (26.85) Partial transverse 0 0
Deep deltoid 71 (65.74) 37 (34.26) Complete transverse 1 (0.93) 0
Lateral ankle ligaments Tendinopathy
Total 306 (70.83) 126 (29.17) None 74 (68.52) 67 (62.04)
Anterior talofibular 41 (37.96) 67 (62.04) Mild 32 (29.63) 37 (34.26)
Calcaneofibular 66 (61.11) 42 (38.89) Moderate 2 (1.85) 4 (3.7)
Posterior talofibular 96 (88.89) 12 (11.11) Severe 0 0
Tibiofibular syndesmosis 103 (95.37) 5 (4.63)
Abbreviation: MRI, magnetic resonance imaging.
Abbreviation: MRI, magnetic resonance imaging. Data presented as n (%).
Data presented as n (%).

included in the present study, 79 (73.15%) had no MRI pathologic

Statistical Analysis findings localized to the SPR, and 29 (26.85%) demonstrated pathologic
findings consistent with a chronic injury of the SPR (Table 6). Also, in
Statistical analyses were conducted using IBMÒ SPSSÒ Statistics, version 20 (IBM,
Armonk, NY). The sample size was determined for correlations based on 90% power to this group of asymptomatic lateral ankle MRI scans, 16 (14.81%) OLTs
detect a significant difference, an a of 0.05, and a correlation coefficient of 0.30. A were observed, and of these,13 (81.25% of talar lesions and 12.04% of all
correlation coefficient of 0.30 was selected because 0.30 generally represents a mod- asymptomatic lateral ankle MRI scans) were localized to the medial
erate correlation between 2 variables. A minimum of 92 MRI examinations were portion of the talar dome and 3 (18.75% of talar lesions and 2.78% of all
required. The data were considered in terms of type and distribution, described in
asymptomatic lateral ankle MRI scans) were localized to the lateral
statistical terms, and reported as the mean  standard deviation. Correlation analyses
were performed to investigate the relationship between OLTs and pathologic features portion of the talar dome (Table 6). Additional evaluation revealed that
of the medial and lateral ankle ligaments, peroneal tendons, and SPR, and the rela- 8 OLTs (7.41% of the 108 MRI scans or 50% of the 16 talar lesions) dis-
tionship between ligamentous discontinuity and peroneal tendon pathologic features, played evidence of tendinopathy localized to the peroneus brevis and
OLTs, and SPR pathologic findings. For the purpose of data analysis, the presence or
longus both, although none of these MRI scans demonstrated a pero-
absence of pathologic findings was coded as 1 or 0, respectively. Spearman’s rank order
correlations (rs) were used to determine the strength and direction of the relationship
neal tendon tear (Table 6).
between the ordinal and dichotomous variables, and Pearson’s product-moment cor- Correlations between the presence of OLTs and MRI evidence of
relations were used to determine the strength and direction of the relationship be- pathologic features localized to the ankle ligaments, peroneal tendons,
tween 2 dichotomous variables. When measuring the association between 2 and SPR were undertaken to determine the strength and direction of the
dichotomous variables, Pearson’s product moment correlation returns the phi coeffi-
association between the presence of an OLT and damage to these
cient (rf). Statistical significance for all tests was set at the 5% (p  .05) level.
structures (Table 7). Our results partially supported our hypotheses. A
weak, positive correlation was found between attenuation and/or tears
Results of the superficial deltoid and medial OLTs (rf ¼ 0.23, n ¼ 108), which was
statistically significant (p ¼ .0191). Also, a moderate positive correlation
A total of 500 consecutive MRI reports were reviewed. Of these, 108 was found between tears of the PTFL and lateral OLTs (rf ¼ 0.30, n ¼ 108),
examinations (21.60%) in 104 patients met the inclusion criteria and which was statistically significant (p ¼ .0017). In contrast to our ex-
were included in the present report. During the observation period, 2 pectations, no statistically significant correlations were observed be-
patients had repeat MRI examinations, and 2 patients had bilateral MRI tween the presence of medial OLTs and pathologic findings localized to
examinations. Patient information pertaining to each of the 108 MRI the deep deltoid (p ¼ .1147), ATFL (p ¼ .5730), CFL (p ¼ .2161), PTFL
scans was used to compute the descriptive information (Table 2). The (p ¼ .6792), syndesmosis (p ¼ .4017), tendinopathy of the peroneus
mean patient age was 41.9  20.42, females outnumbered males brevis (p ¼ .0651), tears of the peroneus brevis (p ¼ .4556), tendinopathy
(71 females [65.74%] and 37 males [34.26%]), and the distribution of of the peroneus longus (p ¼ .2120), or chronic pathologic findings of the
right (n ¼ 54) and left (n ¼ 54) ankles was equal. From the MRI reports, SPR (p ¼ .7462). Furthermore, no statistically significant correlations
the prevalence of ligamentous pathologic features, localized to the were observed between lateral OLTs and pathologic findings localized to
medial and lateral ankle, was determined (Table 3). Using the liga- the superficial deltoid (p ¼ .1167), deep deltoid (p ¼ .2342), ATFL
mentous pathologic data, the grade of ligamentous discontinuity was (p ¼ .1725), CFL (p ¼ .3214), syndesmosis (p ¼ .7020), tendinopathy of the
ascertained and the prevalence of each grade was determined
(Table 4). Similarly, the peroneal tendons were evaluated for MRI evi-
dence of tendinopathy and/or tearing (Table 5). Of the 108 MRI scans Table 6
MRI ascertainment of SPR defects and OLTs observed in asymptomatic lateral ankles
(N ¼ 108 MRI scans)
Table 4
MRI ascertainment of grade of ligamentous discontinuity observed in asymptomatic MRI Diagnosis Outcome Prevalence
lateral ankles (N ¼ 108 MRI scans) No SPR injury 79 (73.15)
Chronic SPR injury 29 (26.85)
Ankle Ligamentous Discontinuity Grade Outcome Prevalence OLTs
None 41 (37.96) Total 16 (14.85)
I 25 (23.15) Medial 13 (12.04)
II 32 (29.63) Lateral 3 (2.78)
III 5 (4.63) OLT and peroneus brevis and longus tendinopathy (no tears) 8 (7.41)
IV 5 (4.63)
Abbreviations: MRI, magnetic resonance imaging; OLTs, osteochondral lesions of the
Abbreviation: MRI, magnetic resonance imaging. talus; SPR, superior peroneal retinaculum.
Data presented as n (%). Data presented as n (%).
4 M.M. Galli et al. / The Journal of Foot & Ankle Surgery xxx (2014) 1–5

Table 7 Discussion
Correlation* of MRI ascertainment of asymptomatic lateral ankle pathologic findings
with OLTs (N ¼ 108 MRI scans)
The aftermath of ligamentous ankle sprains on cartilaginous and
Ankle Structure and MRI Pathologic Findings OLTs (n ¼ 16) tendinous pathologic entities has been well documented throughout
Medial (n ¼ 13) Lateral (n ¼ 3) clinical, surgical, and radiologic studies. Data on the prevalence of
Medial ankle ligamentous discontinuity, OLTs, peroneal pathologic features, and
Superficial deltoid attenuated and/or torn 0.23y 0.15 pathologic findings localized to the SPR are scant in the asymp-
Deep deltoid attenuated and/or torn 0.15 0.12 tomatic lateral ankle population and data on their interrelatedness
Lateral ankle
are absent. This is understandably because symptoms typically guide
Anterior talofibular attenuated and/or torn 0.06 0.13
Calcaneofibular attenuated and/or torn 0.12 0.1
the decision to obtain diagnostic imaging. It is well understood that
Posterior talofibular torn 0.04 0.3y large, traumatically induced OLTs often require attention, because
Tibiofibular syndesmosis torn 0.08 0.04 these defects can be a source of chronic pain, recurrent synovitis,
Peroneal tendons and/or intra-articular bodies (10). However, clinical symptoms
Peroneus brevis tendinopathy 0.18 0.01
caused by smaller lesions will not always be encountered. Our clin-
Peroneus brevis tear 0.07 0.03
Peroneus longus tendinopathy 0.12 0.02 ical experience has led us to believe that subclinical pathologic fea-
Chronic SPR injury 0.03 0.15 tures have the potential to progress and, therefore, should be
Abbreviations: MRI, magnetic resonance imaging; OLTs, osteochondral lesions of the
monitored closely and treated if any symptoms develop. Based on
talus; SPR, superior peroneal retinaculum. our present findings, the pathologic features noted on MRI should be
* Phi correlation coefficient for dichotomous outcomes. discussed openly with patients who have an asymptomatic lateral
y
Correlation statistically significant at p  .05. ankle, and they should be closely monitored thereafter. For example,
noting the OLT size and discussing lifestyle modifications could
prevent additional degradation of native anatomic structures and,
peroneus brevis (p ¼ .9448), tears of the peroneus brevis (p ¼ .7334),
consequently, prevent subchondral collapse. If related pathologic
tendinopathy of the peroneus longus (p ¼ .8684), or chronic pathologic
features are noted on MRI examination, we advise practitioners to
features of the SPR (p ¼ .1167).
continually monitor the linked pathologic features in an effort to
From our MRI findings, correlations were also computed to
prevent additional progression. With the increased success of
determine the strength and direction of the associations between
comprehensive lateral ankle rehabilitation programs and the high
ligamentous discontinuity (grade I to IV) and the presence of patho-
percentage of asymptomatic patients displaying abnormal ligamen-
logic features localized to the peroneal tendons, the presence of an
tous and tendinous pathologic features, we have followed the rec-
OLT, and the presence of SPR pathologic findings (Table 8). We hy-
ommendations of Saxena et al (11), who stated that the presence of
pothesized that increasing severity of the ankle’s ligamentous con-
pathologic findings on MRI alone should not be used as an indication
dition would correlate with the presence of peroneal tendon
for surgery and should only be discussed after exhaustion of con-
pathologic features. That hypothesis was partially confirmed. A
servative treatment.
moderate, positive correlation was found between ligamentous
Roughly one third of our patients demonstrated tendinopathy of
discontinuity (grade I to IV) and tendinopathy of the peroneus brevis
the peroneal tendons (31.48% peroneus brevis, 37.96% peroneus lon-
[rs(106) ¼ 0.29], which was statistically significant (p ¼ .0024).
gus; Table 5). The prevalence of tendinous pathologic features in our
However, no significant correlations were found between the grade of
asymptomatic population appeared to mirror the findings of Saxena
ligamentous discontinuity and tears of the peroneus brevis (p ¼ .1547)
et al (11), who also investigated the MRI findings in an asymptomatic
or between the grade of ligamentous discontinuity and tendinopathy
cohort. In their study (11), patients with lateral ankle pain, an ankle
of the peroneus longus (p ¼ .0714). None of the MRI scans displayed
injury within the preceding 10 years, and any previous lateral ankle
evidence of a tear of the peroneus longus; therefore, we were unable
surgery were excluded. Overall, 34% of their patients demonstrated
to correlate any pathologic findings with tears of the peroneus longus.
peroneal tendinopathy. Despite the limitations associated with his-
No statistically significant correlations were found between the grade
torical control comparisons, 68.52% of our patients and 66% of their
of ligamentous discontinuity and medial OLTs (p ¼ .6016) or the grade
patients demonstrated a peroneus brevis tendon without tendinop-
of ligamentous discontinuity and lateral OLTs (p ¼ .1471). Finally, no
athy, and 62.04% of our patients and 67% of their patients demon-
statistically significant correlation was found between the grade of
strated a peroneus longus tendon without tendinopathy (11). Because
ligamentous discontinuity and chronic pathologic features of the SPR
the exact definitions of these categories were not stated, for the
(p ¼ .1769).
purpose of comparison we assumed that intact referred to patients
devoid of abnormalities. With these assumptions, a similar percent-
Table 8 age of our patients exhibited tendons without evidence of tendin-
Correlation* of MRI ascertainment of asymptomatic lateral ankle pathologic findings opathy. Collectively, these data sets suggest that approximately one
with ankle ligamentous discontinuity (N ¼ 108 MRI scans) third of asymptomatic lateral ankles will have peronei that exhibit
Ankle Structure and MRI Pathologic Findings Ligamentous some degree of tendinopathy.
Discontinuity (grade I–IV) Moreover, we recorded the MRI findings localized to the lateral
OLTs ligamentous structures as torn and/or attenuated versus intact without
Medial 0.05 pathologic features. In our study, 37.96% of patients demonstrated a
Lateral 0.14
healthy ATFL. In contrast, in the study by Saxena et al (11), 71% of pa-
Peroneal tendons
Peroneus brevis tendinopathy 0.29y tients demonstrated an intact ATFL. For the CFL, pathologic findings
Peroneus brevis tear 0.14 were absent in 61.11% of our patients versus 89% in the study by Saxena
Peroneus longus tendinopathy 0.17 et al (11). Although it was unclear whether attenuation would have
Chronic SPR injury 0.13 been classified as abnormal and, as such, been included or excluded in
Abbreviations: MRI, magnetic resonance imaging; OLTs, osteochondral lesions of the the intact classification by Saxena et al (11), we believe that these in-
talus; SPR, superior peroneal retinaculum. terpretations could, in part, explain the discrepancy between the 2
* Spearman’s rank correlation coefficient.
y reports regarding the status of the ATFL and CFL.
Correlation statistically significant at p  .05.
 M.M. Galli et al. / The Journal of Foot & Ankle Surgery xxx (2014) 1–5 5

It was also interesting to note the similarities between the per- patients included in the present study had asymptomatic lateral an-
centage (50%) of concomitant OLTs and peroneal pathologic features kles. Also, although our key inclusion criterion was the absence of
observed in our asymptomatic lateral ankle population and the per- symptoms localized to the lateral aspect of the ankle, we did not
centage (65.3%) observed by Stasko et al (5), whose MRI sample was analyze (beyond the presence of hindfoot or nonlateral ankle symp-
composed primarily of symptomatic patients. Presumably, this high toms) the rationale for obtaining the ankle MRI scan. To best identify
prevalence (w50% to 65%) would suggest that concomitant peroneal the relations in an asymptomatic population, ideally, we would have
and OLT pathologic features are present in both symptomatic and included truly asymptomatic ankles for examination instead of pa-
asymptomatic patients in the Pennsylvania region and beyond. tients with pathologic features located elsewhere in the hindfoot and
In our investigation, we found statistically significant weak, but ankle. Compensation for 1 symptomatic aspect of the ankle could have
positive, correlations between attenuation and/or tears of the super- resulted in pathologic features elsewhere in the areas studied. More-
ficial deltoid and medial OLTs (rf ¼ 0.24, n ¼ 108, p ¼ .0191) and be- over, with MRI interpretation, there is room for error, in particular,
tween tears of the PTFL and lateral OLTs (rf ¼ 0.30, n ¼ 108, p ¼ .0017). given the quantity and concentration of structures along the lateral
In contrast to our hypothesis, statistically significant findings were not ankle (17). To minimize this source of error, we elected 2 musculo-
observed for the correlation between ligamentous discontinuity and skeletal radiologists with 19 years of combined experience to review
the presence of medial OLTs [rs(106) ¼ 0.05, p ¼ .6016] or lateral OLTs our studies.
[rs(106) ¼ 0.14, p ¼ .1417]. We did, however, find a weak positive cor- Although our study described correlations between MRI-noted
relation between ligamentous discontinuity and peroneus brevis pathologic findings in ankles that were asymptomatic laterally, we
tendinopathy [rs(106) ¼ 0.29, p ¼ .0024] using the MRI ligamentous appreciate that correlation analyses are sample-dependent, and our
discontinuity scale (which mimics the gradation of ankle sprains) in results could have been different if MRI scans from another group of
our asymptomatic lateral ankle MRI scans. However, no statistically patients had been analyzed. We also understand that causal re-
significant correlations were noted between ligamentous disconti- lationships between these findings could not be established, and our
nuity and peroneal longus tendinopathy or peroneus brevis tears. results were limited by the ability to identify actual pathologic fea-
Owing to the known findings from patients with symptomatic lateral tures using MRI. Statistically, we did not undertake a sensitivity
ankles, we believe our findings demonstrating a relationship between analysis to estimate the influence that unmeasured variables could
ligamentous discontinuity and asymptomatic peroneus brevis ten- have had on our results. Despite these limitations, we believe that the
dinopathy are rational, in part, because of the anatomic intimacy of the results of the present investigation are a valuable addition to the
involved structures. published data and could be used in the development of future,
The lateral collateral ligaments are the primary stabilizers of the multicenter, prospective cohort studies to identify patients at high
ankle against supination, adduction, and inversion stress (12). risk of developing cartilaginous damage.
Anatomic studies have provided information on the interplay of the
lateral ligaments and retinacula of the peroneal tendons. The SPR is References
composed of 2 fascial bands that span from the posterior ridge of the
fibula to the Achilles tendon sheath and to the posterior lateral aspect 1. DiGiovanni CW, Brodsky A. Current concepts: lateral ankle instability. Foot Ankle
Int 27:854–866, 2006.
of the calcaneus. The inferior band, with its calcaneal insertion, par- 2. DiGiovanni BF, Fraga CJ, Cohen BE, Shereff MJ. Associated injuries found in chronic
allels the course of the CFL, making it susceptible to inversion ankle lateral ankle instability. Foot Ankle Int 21:809–815, 2000.
injuries (13,14). Tearing of the lateral collateral ligaments and tearing 3. Ferkel RD, Chams RN. Chronic lateral instability: arthroscopic findings and long-
term results. Foot Ankle Int 28:24–31, 2007.
of the SPR occurs in 78% of traumatic ankle injuries and places the 4. Hintermann B, Boss A, Scha €fer D. Arthroscopic findings in patients with chronic
peroneal tendons at risk of abnormal motion and tendinopathy (15). ankle instability. Am J Sports Med 30:402–409, 2002.
The correlation between peroneal brevis tendinopathy and ligamen- 5. Stasko PM, McSpadden CK, Jung R, Mendicino RW, Catanzariti AR. Incidence of
talar dome lesions with concomitant peroneal tendon pathologic features: a
tous discontinuity confirmed our hypothesis, further emphasizing the magnetic resonance imaging evaluation. J Foot Ankle Surg 51:579–582, 2012.
anatomic relatedness of these pathologic entities, even in an 6. Rosenberg ZS, Beltran J, Bencardino JT. From the RSNA Refresher Courses. Radio-
asymptomatic population. Despite our expectations, no statistically logical Society of North America. MR imaging of the ankle and foot. Radiographics
20:S153–S179, 2000.
significant correlation was found between ligamentous discontinuity
7. Kier R, McCarthy S, Dietz MJ, Rudicel S. MR appearance of painful conditions of the
and chronic injuries of the SPR [rs(106) ¼ 0.13, p ¼ .1769]. ankle. Radiographics 11:401–414, 1991.
A comparative study (16) was conducted to evaluate the MRI 8. Yao L, Tong DJ, Cracchiolo A, Seeger LL. MR findings in peroneal tendonopathy. J
reading sensitivity of radiologists and orthopedic surgeons with Comput Assist Tomogr 19:460–464, 1995.
9. Mintz DN, Tashjian GS, Connell DA, Deland JT, O’Malley M, Potter HG. Osteo-
intraoperative confirmation of findings. The report highlighted the chondral lesions of the talus: a new magnetic resonance grading system with
challenges of detecting lesions on preoperative MRI and provided arthroscopic correlation. Arthroscopy 19:353–359, 2003.
recommendations to improve detection of OLTs. The authors noted 10. Hangody L, Kish G, Mo  dis L, Szerb I, Ga
spa
r L, Dio
szegi Z, Kendik Z. Mosaicplasty
for the treatment of osteochondritis dissecans of the talus: two to seven year
the reading consistency of specialized musculoskeletal radiologists, results in 36 patients. Foot Ankle Int 22:552–558, 2001.
the need for similar protocols, and the importance of obtaining 11. Saxena A, Luhadiya A, Ewen B, Goumas C. Magnetic resonance imaging and inci-
studies using comparable magnets. A notable strength of our report dental findings of lateral ankle pathologic features with asymptomatic ankles. J
Foot Ankle Surg 50:413–415, 2011.
was the adherence of our protocols to these recommendations. We 12. Chrisman OD, Snook GA. Reconstruction of lateral ligament tears of the ankle: an
hope that our findings will strengthen discussions with patients experimental study and clinical evaluation of seven patients treated by a new
regarding the prevalence of pathologic features encountered on MRI modification of the Elmslie procedure. J Bone Joint Surg Am 51:904–912, 1969.
13. Davis WH, Sobel M, Deland J, Bohne WH, Patel MB. The superior peroneal reti-
in asymptomatic persons. naculum: an anatomic study. Foot Ankle Int 15:271–275, 1994.
The present study had numerous limitations that could threaten the 14. Geppert MJ, Sobel M, Bohne WH. Lateral ankle instability as a cause of superior
validity of our conclusions. The most substantial methodologic short- peroneal retinacular laxity: an anatomic and biomechanical study of cadaveric
feet. Foot Ankle 14:330–334, 1993.
coming of the present investigation was its retrospective design. As
15. Demondion X, Canella C, Moraux A, Cohen M, Bry R, Cotten A. Retinacular disor-
such, our primary inclusion criteria were dependent on multiple ders of the ankle and foot. Semin Musculoskelet Radiol 14:281–291, 2010.
radiology technologists and clinical personnel inputting accurate 16. O’Neill PJ, Van Aman SE, Guyton GP. Is MRI adequate to detect lesions in patients
indication criteria and inclusive dictations. We recognize the inherent with ankle instability? Clin Orthop Relat Res 468:1115–1119, 2010.
17. Erickson SJ, Smith JW, Ruiz ME, Fitzgerald SW, Kneeland JB, Johnson JE, Shereff MJ,
biases and limitations associated with producing and maintaining Carrera GF. MR imaging of the lateral collateral ligament of the ankle. AJR Am J
medical records and, therefore, could not definitively state that all Roentgenol 156:131–136, 1991.