Sports-Related Muscle Injury
in the Lower Extremity
Derek R. Armfield, MDa,b,*, David Hyun-Min Kim, MDc,
Jeffrey D. Towers, MDa, James P. Bradley, MDd,
Douglas D. Robertson, MD, PhDa,b
aDepartment of Radiology, University of Pittsburgh Medical Center, 200 Lothrop Street,
Pittsburgh, PA 15213, USA
bJefferson Regional Medical Center, 565 Coal Velley Road, Pittsburgh, PA 15236, USA
cUniversity of Southern California, Department of Radiology, 1500 San Pablo Street,
Los Angeles, CA 90033, USA
dBurke and Bradley Orthopaedics and Department of Orthopaedic Surgery, University
of Pittsburgh Medical Center, 200 Lothrop Street, Pittsburgh, PA 15213, USA
30% of injuries with the quadriceps (32%), hamstring (28%), adductor (19%),
and gastrocnemius (12%) muscle injuries being the most common . Treat-
ment typically consists of rest, ice, compression, elevation, and stretching and
rehabilitation. Programs are designed to treat and prevent, as those with prior
injuries are prone to recurrence. This problem is particularly important in elite
athletes where decisions regarding return to play and player performance can
have significant financial or strategic consequences for the player and team.
This paper reviews the basic imaging techniques and the pertinent findings
associated with common muscle injuries of the lower extremity, and emphasizes
the imaging features, whichcan help guide treatmentandofferprognosis. A gen-
eral overview of muscle injury and imaging modalities is provided followed by
a more detailed analysis of injuries to specific muscle groups of the lower ex-
tremity including the hamstrings, quadriceps, adductors, and lower leg muscles.
uscle injuries are common and often occur during sport or training
with over 90% caused by excessive strain or contusion . A 5-year
study of European soccer players showed muscle strain represented
TYPES OF MUSCLE INJURY
There are many different sizes and shapes of muscle. Some are long like the
biceps femoris with tendon insertions on bone at both ends and cross two joints
(biarticular). Others are short with a single tendon insertion spanning a single
*Corresponding author. Department of Radiology, University of Pittsburgh Medical Center,
200 Lothrop Street, Pittsburgh, PA 15213.
E-mail address: email@example.com (D.R. Armfield).
0278-5919/06/$ – see front matter
ª 2006 Elsevier Inc. All rights reserved.
Clin Sports Med 25 (2006) 803–842
CLINICS IN SPORTS MEDICINE
joint like the popliteus. Some have long muscle bellies (sartorius) whereas others
have long tendons (plantaris). Some have muscle fibers aligned with tendons in
a colinear fashion (unipennate) whereas others have muscle fibers inserting at
an angle on an intramuscular tendon (bipennate), which increases muscle fiber
density and ultimately contractile forces (ie, hamstrings, rectus femoris).
What they have in common is a propensity for injury at the interface of two
different materials namely muscle and tendon typically referred to as the my-
otendinous junction [3,4]. The term musculotendinous junction has also been
used synonymously. It is important to note that the typical connotation of a my-
otendinous junction is that of a specific focal point at either the proximal or dis-
tal aspect of a muscle before tapering to the tendon insertion. However the
hamstring and quadriceps tendons have large intramuscular or central tendons
and injury often occurs along this interface [5,6].
Most sports-related muscle injuries involve strains, contusions, and uncom-
monly lacerations. Muscle strains or tears often affect muscles with primarily
fast-twitch type-2 muscle fibers, span two joints (biarticular), and undergo ec-
centric contraction . As mentioned earlier, strain injuries typically occur at
the myotendinous junction. However, strains have also been described involv-
ing the periphery of a muscle (instead of the myotendinous junction) extending
to the epimysium seen with ultrasound and MRI [8,9]. Based on the patient’s
age and the underlying condition of the tendon itself, injury can occur any-
where along the muscle-tendon-bone chain . For example those with degen-
erated tendon because of aging or chronic use may develop a tear of the tendon
itself. Those with strong tendons may experience an avulsion of the tendon
from the bone or myotendinous strain. In skeletally immature patients, an
apophyseal avulsion may occur, as this junction biomechanically represents
the weakest interface.
Strains are often diagnosed clinically on a three-point scale: 1 ¼ mild, 2 ¼
partial tear, 3 ¼ complete . Mild injuries have no discernable loss of
strength or motion restriction. Partial tears demonstrate some loss of strength
and motion that is not complete, unlike type 3 injuries . Strain injury is as-
sociated with inflammation, edema, and sometimes hemorrhage with prolifer-
ation of inflammatory cells and fibroblastic activity in the first 24 to 48 hours
. Histological animal models of muscle stretch injury have shown that my-
otendinous injury results in inflammation, bleeding, and muscle fiber necrosis
initially. This destructive phase is followed by a concomitant repair and remod-
eling phase involving recruitment of progenitor cells, scar formation, and re-
modeling of organized tissue .
Because of the common nature of these injuries, many muscle strains are
treated clinically. However the clinical scenario may be unclear and grading
of injury may be difficult. Imaging may help delineate the presence and extent
of muscle injury. The main modalities used for evaluation almost exclusively
include MRI and ultrasound.
804ARMFIELD, KIM, TOWERS, ET AL
Radiographs are useful for evaluating bony avulsion injuries in adolescents
particularly of the pelvis that can be missed with MR and ultrasound. Subtle
areas of soft tissue swelling and unexpected bone-related problems (tumor,
stress fracture, and so forth) might be detected with plain films. While cross-sec-
tional imaging findings of muscle strain were originally described with com-
puted tomography (CT), currently CT has little role for evaluating acute
muscle injury because of its relative lack of tissue contrast as compared with
MR . It is useful to evaluate osseous structures associated with avulsion in-
juries and complications like myositis ossificans.
At some institutions ultrasound may be the preferred primary modality for
evaluation of muscle injury because of its portability, ease of use, and decreased
cost. While ultrasound does have excellent spatial resolution, the contrast res-
olution is not as good as MR particularly in the subacute or chronic phases
when injury-related edema begins to resolve. Also, because sound waves dissi-
pate and do not reflect over long distances, evaluation of deep structures in ath-
letes with bulky musculature may be difficult. Evaluation of more superficial
structures such as the patellar tendon is easier with ultrasound. Another rela-
tive disadvantage is the significant reliance on operator skill and expertise
that can only be achieved with dedication and practice.
At our institution we prefer evaluation of muscle injuries with MR because
of its superior soft tissue contrast, excellent spatial resolution, and reproducibil-
ity. Our typical protocol uses a combination of T1- and T2-weighted sequences
to emphasize anatomy and pathologic edema. Fatty structures appear bright on
T1-weighted images (and some T2-weighted images, ie, fast spin echo) and
muscle has intermediate signal intensity allowing for excellent anatomic detail
of fat planes. In general, fluid-sensitive or T2-weighted images, allow easy visu-
alization of mobile water protons, which means that pathologic processes in-
volving edema, like muscle strains, are easily detected. Contrast resolution is
increased when fat signal is nullified on fast spin echo T2-weighted images
with specific chemical fat-saturation pulse (ie, fat saturation). Alternatively,
fluid sensitivity may be achieved when a more diffuse nullifying signal is em-
ployed that limits non-water signal (ie, inversion recovery [IR] or STIR se-
quences). Either sequence is considered fluid sensitive and essential for the
evaluation of muscle strain injury.
Anatomic coverage includes long and short axis imaging of the region or
muscle of interest. We generally use a body coil to include both thighs and
lower legs depending on the area of concern to allow for comparative analysis
of anatomy in the symptomatic and asymptomatic extremity. Others prefer
dedicated unilateral imaging of the injured extremity. Studies have shown
that hamstring injuries can occur at multiple sites and involve multiple muscles
and therefore thorough evaluation along the course of the muscle group is
needed not just the area of pain [9,13,14].
For the screening protocol of the thigh or lower leg we include coronal T1,
coronal IR, axial T1, and axial T2 fat-saturated images. Depending on the clin-
ical scenario we may add additional sagittal T1 or fluid-sensitive sequences
805LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY
perhaps in the case of an ischial tuberosity avulsion. The important concept is
to include short and long axis imaging of the structures of interests with T1 and
fluid-sensitive sequences for each.
Intravenous gadolinium contrast is used very sparingly for routine cases of
clinically suspected muscle injury. Some have suggested that low-grade injuries
that appear normal on fluid-sensitive sequences may be seen with postintra-
venous contrast imaging although this report was only a case series of four
athletes with high clinical suspicions of injury . Others have found intrave-
nous contrast imaging useful for evaluating symptomatic proximal adductor
insertional injuries. These contrast-enhanced images revealed enhancing teno-
periosteal granulation tissue associated with symptoms and partial healing .
Contrast should be used when cases of infection, tumor, or myositis are within
the differential (Fig. 1).
MR appearance of myotendinous injury has been well described [10,17–20].
Type 1 injuries demonstrate bright signal on fluid-sensitive sequences repre-
senting fluid and hemorrhage around the myotendinous unit extending into
the adjacent muscle creating a feathery appearance. The myotendinous junc-
tion usually appears normal and there is typically less than 5% involvement
of muscle fibers (Fig. 2A). Type 2 injuries of the myotendinous junction are
more severe and may show a thin or irregular appearance of the myotendinous
junction itself along with edema and hemorrhage (increased T2 signal inten-
sity) that often tracks along the fascial plane. However, increased T2 signal in-
tensity changes in strain injury may not necessarily be related to hemorrhage.
One recent study evaluated hamstring strain injuries and included gradient se-
quences, which are highly sensitive for detecting blood products, and found
only 1 case of 37 had the typical blooming artifact associated with blood prod-
ucts . Another article has characterized hematoma as a pathognomonic find-
ing of type 2 injury  (Fig. 2B). Type 3 injuries reveal complete disruption
Fig. 1. (A) Enhanced axial T1-weighted image with fat saturation of the calf showing enhanc-
ing muscle with areas of nonenhancement compatible with necrosis in this patient found unre-
sponsive. (B) Peripheral enhancement of the calf muscles in a patient with dermatomysositis.
806ARMFIELD, KIM, TOWERS, ET AL
and discontinuity of muscle typically at the myotendinous junction with
complete replacement of organized collagen with fluid signal on fluid sensitive
sequences. There is often an associated wavy tendon morphology and retrac-
tion. Surrounding edema or hemorrhage is usually extensive (Fig. 2C). MR
findings usually correlate with the clinical grading scheme and can help differ-
entiate mild injury from partial tears and referred pain in clinically indetermi-
nate cases .
Epimyseal or peripheral injury not associated with myotendinous injury has
also been described in the hamstring and quadriceps muscles and manifests as
peripheral edema in the muscle extending to and around the epimysium [8,9].
Contusions of muscle are a result from direct trauma (ie, football helmet), and
may predispose to hematoma formation. Infiltrative focal edema is a typical
finding on fluid-sensitive sequences and may resemble muscle strain. MR ap-
pearance of contusion is typically that of increased size with intact muscle fibers
Fig. 2. Coronal fluid sensitive images of posterior thighs demonstrating (A) Type 1 muscle
strain injury with mild feathery edema along the intramuscular myotendinous junction of biceps
femoris in a professional football wide receiver; (B) Type 2 injury of the proximal myotendinous
junction of biceps femoris with intramuscular hematoma formation; and (C) Type 3 injury prox-
imal biceps femoris with retraction of the tendon (arrow) in a professional football cornerback.
807LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY
and increased fluid signal that is diffuse or geographic with feathery margins
 (Fig. 3).
Hematoma may result from direct trauma associated with contusion or re-
lated to myotendinous injury and subsequent bleeding. MRI and ultrasound
helps assess size and location and determine if it is intermuscular or intramus-
cular in nature. Large hematomas may result in compartment syndrome or sig-
nificant pain and aspiration may be needed.
The MR appearance of hematomas can be variable depending on age and
magnetic field strength and T1- and T2-weighted images can help determine
the age and relative oxidative state of hemoglobin [22–24]. Acute hematomas
are usually isointense to muscle on T1-weighted images. T2-weighted images
show increased signal intensity possibly with central decreased signal related
to deoxyhemoglobin (Fig. 4A). Subacute hematomas (>48 hours) have in-
creased amounts of methemoglobin, which has increased T1 signal. Chronic
hematomas may have a peripheral dark rim related to hemosiderin. A seroma
may ultimately develop with resorption of blood products (Fig. 4B).
HAMSTRING MUSCLE COMPLEX
The hamstring complex is composed of three major muscles: biceps femoris
and semimembranosus and semitendinosus muscles. The biceps femoris is
composed of a long and short head. The long head arises on the medial aspect
of the posterior ischial tuberosity with a common tendon insertion with the
semitendinosus called the conjoined tendon  (Fig. 5A–D). Distally it inserts
on the fibular head. Depending on leg positioning and relationship to the
ground it can serve as a hip extensor, knee flexor, and external rotator of
the hip and knee (Fig. 5E,F).
The short head of the biceps tendon is not biarticular but has a proximal at-
tachment on the lateral aspect of the linea aspera below the gluteal tuberosity
and inserts distally on the fibular head . The short head of the biceps can be
absent, and unlike the long head that receives innervation via a tibial portion of
Fig. 3. Axial fluid sensitive image mid thigh shows increased fluid signal in rectus femoris con-
sistent with contusion. Note enlargement and diffuse edema in this soccer player that sustained
a direct blow to the thigh.
808 ARMFIELD, KIM, TOWERS, ET AL
the sciatic nerve, the short head receives innervation from the common pero-
neal nerve. This dual innervation has been hypothesized a source of potentially
discordant contraction which can lead to injury .
The semitendinosus is another biarticular muscle with a common origin of
the long head of the biceps femoris via the conjoined tendon (Fig. 5). Distally
it has a long tendon, which inserts on the proximal medial tibia posterior to the
sartorius. Its function is similar to that of the long head of biceps femoris al-
though because of its medial sided insertion distally it functions as an internal
rotator of the hip and knee. It has been classified as a digastric muscle owing to
a central raphe where the proximal fibers insert .
Semimembranosus is the third major muscle of the hamstring complex with
a proximal attachment on the ischial tuberosity anterior the conjoined tendon
(Fig. 5). The distal insertion is primarily on the medial posterior aspect of the
tibial plateau but has multiple slips extending to surrounding structures such as
the medial collateral ligament, and popliteus muscle . Its function is similar
to the semitendinosus.
The ischial tuberosity also has insertion sites of the sacrotuberous ligament
posteromedially in close proximity to the conjoined tendon insertion. The
Fig. 4. (A) Prominent acute intramuscular medial gastrocnemius hematoma. Note mixed in-
creased and decreased signal probably related to deoxyhemoglobin. (B) Intermuscular fluid
collection presumable a seroma from a resorbed gastrocnemius hematoma. Note dark rim
compatible with hemosiderin (arrow).
809LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY
posterior head of the adductor magnus arises from the anterior inferior aspect
of the ischial tuberosity. This portion inserts distally on the adductor tubercle
and functions as a hip extensor, as well, but is not typically categorized with the
hamstring muscle complex.
Fig. 5. (A) Axial T1-weighted images of proximal thighs. Note how T1-weighted images al-
low good depiction of muscle fat planes. This image is proximal to the ischial tuberosity
and shows the sacrotuberous ligament (white arrow) insertion on the tuberosity. (B) Mid tuber-
osity level shows the anterior semimembranosus insertion (black arrow) and the posterior con-
joined tendon of biceps femoris and semitendinosus (white arrow). (C) Inferior aspect of ischial
tuberosity shows semimembranosus (black arrow) and conjoined tendon separating (white
arrow). Note origin of adductor magnus anteriorly (open arrow). (D) Continued separation
of the three tendons. (E) Distally the semitendinosus has a long tendon (arrow) and lies poste-
rior to the semimembranosus. sm, semimembranosus; s, sartorius; g, gracilis; bf, biceps femo-
ris. (F) Tendons of the posterior knee: semimembranosus (arrowhead), semitendinosus (white
arrow), gracilis (open arrow), biceps femoris (black arrow). lg, lateral gastrocnemius; mg,
810ARMFIELD, KIM, TOWERS, ET AL
Location of Injury and Imaging Prognosis
Intrinsic and extrinsic factors associated with recurrent hamstring injuries in-
clude inadequate warm-up, muscle fatigue, inadequate preseason training, mus-
cle strength imbalances, decreased flexibility, increasing age, and history of
prior injury . Some suggest that an injured muscle may heal with scarring
resulting in suboptimal muscle length that predisposes recurrent injury .
Because of the high risk of recurrent injury and variable convalescence pe-
riod, imaging may have a prognostic role in evaluating hamstring injuries, par-
ticular for the elite athlete where strategic and financial stakes can be high.
The first MR study that described findings with poor prognosis of muscle
injury evaluated 14 patients and found that muscle rupture and retraction,
hemorrhage, ganglion-like fluid collections, and greater than 50% cross-sec-
tional involvement were associated with convalescent periods of more than 6
A more recent study of 30 MRI-proven hamstring injuries in Australian
Rules football players showed high correlation with volume of involvement
(range 0.04 cm3to 175.6 cm3, median 16.8 cm3) and maximum cross-sectional
percentage (8 to 100, median 46%) with time lost from competition (13 to 48
days, median 27) . Linear fluid signal representing the length inter- and in-
tramuscular fluid and edema showed strong correlation but was not statistically
significant in this study. More injuries occurred distally (19 versus 11 defined as
above or below origin of biceps femoris short head) but there was no correla-
tion with location of injury and missed competition (Fig. 6).
One important concept to reiterate is the myotendinous junction was in-
volved in 28 of 30 cases with 24 of these cases involving the intramuscular ten-
don of the muscle and only 4 cases involving the conventional proximal or
distal myotendinous junctions. Five cases involved the intramuscular tendon
and then extended to the conventional myotendinous junction.
In terms of predicting recurrence and length of convalescence another study
imaged 31 Australian Rules footballers with clinical grade 1 injury. Forty-five
percent had a negative MR exam and returned to full team training in 6.6
days versus 20.2 days for the MR-positive group. In this study the length of
the injury had a stronger correlation coefficient than the cross-sectional areas
with the rehabilitation interval unlike the previously mentioned study. Six of
17 MR-positive cases developed recurrent strains with no correlation between
length or cross-sectional area as a predictor for recurrence .
Verrall and colleagues  also compared the clinical finding of posterior
thigh injury with MRI findings of hamstring strain. Again, not all clinically sus-
pected hamstring injuries had findings on MR for muscle strain. Of the 83
patients imaged, 68 (82%) had typical hyperintense signal on fluid-sensitive
images as interpreted by a musculoskeletal radiologist, compatible muscle
strain, whereas 12 (14%) had no signal change at all. The remaining three pa-
tients had MR evidence of muscle injury outside of the hamstring muscle com-
plex (lower gluteus maximus, vastus lateralis, and adductor magnus). Those
with MRI-detectable signal changes had more pain (5/10 versus 2/10), were
811LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY
more likely to have acute onset, and missed more days from practice (27 versus
16 days) as compared with the group without MR findings of muscle strain.
The authors hypothesized that those without MR findings have a referred
pain syndrome or neuromeningeal cause of posterior thigh pain. Thus, MR
helps accurately define the extent and location of injury and helps define causes
of referred pain and types of injury that might heal more quickly.
fessional Americanfootballplayerexperience issomewhatdifferent asonestudy
showed that the majority of cases result in no loss from game competition .
The reported 13-year National Football League (NFL) experience from 1985
to 1998 found 431 hamstring injuries with 324 first-degree type and 107 second
and third-degree type injuries. The first-degree type injuries had no loss of prac-
tice or game time. Some of the more advanced cases with a focal palpable abnor-
mality(58 cases) atthe expected locationof the proximalmyotendinous junction
underwent intramuscular steroid injection within 72 hours. Average time loss
Fig. 6. Grade 1 strain of the distal semitendinosus muscle (A). Coronal T1 showing partial
tear of distal biceps femoris tendon in this professional football defensive back (B).
812 ARMFIELD, KIM, TOWERS, ET AL
until full practice was 7.6 days and the averagetrainingroom treatment time was
missed one game and one player missed two games. Those with more serious
injuries were not treated with injections.
Distribution of Injuries and Ultrasound versus MR
A review of 179 cases of injury to the hamstring muscle complex (HMC) using
ultrasound (102 cases) and MR (97 cases) showed there were 21 injuries in-
volving the proximal insertion on the ischial tuberosity with 16 tendon avul-
sions; 154 injuries of the muscle belly, and only 4 injuries of the distal
tendon or bone insertion site . Approximately 80% (124 cases) of injuries
involved the biceps femoris (54 proximal, 48 mid, and 22 distal); 61% involved
the myotendinous junction and 35% were considered epimyseal or involving
the periphery of the muscle. Multiple muscle involvement was only seen in
5% cases for these authors, others have shown using MR primarily that multi-
ple muscle injury occurs nearly 30% to 40% of the time [13,14,21].
MR correctly identified all of the proximal hamstring avulsion injuries (16/
16), whereas those patients who also underwent ultrasound evaluation had
the avulsion injury detected in slightly more than half of the patients (7/12)
(Fig. 7). The authors did find ultrasound useful for detecting distal superficial
injuries (fairly uncommon) involving the distal semitendinosus and semimem-
branosus tendons. Operator dependence and skill were noted to be a factor for
successful interpretation of muscle injury using ultrasound.
A more recent longitudinal study of hamstring muscle injures compares so-
nography with MR in 60 professional Australian Rules football players . All
players were imaged within 3 days, at 2 weeks, and 6 weeks with both modal-
ities. Sonography detected 45, 25, and 10 cases of injuries over the three time
frames and MR detected 42, 29, and 15 injuries respectively. All injuries ap-
peared larger (length and cross-section) on MRI at all time points. The length
of the tear measured on coronal images and the cross-sectional area on MRI
Fig. 7. Partial chronic tear of hamstring insertion seen on MRI coronal fluid sensitive images in
a former world-class female marathon runner.
813 LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY
was thebestpredictorfortimeto returntocompetition. Tearsshowed decreased
cross-sectional involvement over time with both modalities. Ultrasonography
was found to be more useful for evaluating epimyseal injuries and MR better
for intramuscular tendon abnormalities. Distribution of injuries confirmed bi-
ceps femoris as being most commonly injured typically along the intramuscular
versus myotendinous junction injuries. This study also showed a relative infre-
quent association with multiple muscle injuries (about 5%). While ultrasound
was good, bulky musculature in athletes limited its use and overall, the authors
felt MR was the preferred modality for the elite athlete when there is concern for
optimizing rehabilitation and a need for follow-up imaging.
The quadriceps muscle group is composed of the rectus femoris and vastus
muscles (intermedius, lateralis and medialis) (Fig. 8). The primary mechanism
of action is knee flexion. Only the rectus femoris is biarticular. Proximally the
rectus femoris has a direct head insertion on the anterior inferior iliac spine and
an indirect head extending slightly laterally blending with the lateral aspect of
acetabulum and hip capsule . The direct head forms the anterior fascia of
the proximal third of the muscle whereas the indirect head continues centrally
located within the muscle and terminating at the distal aspect of the muscle
Proximal attachments of the vastus lateralis are multiple and include the in-
tertrochanteric line, anterior and inferior border of greater trochanter, lateral
gluteal tuberosity, upper linea aspera, and lateral intermuscular septum. Dis-
tally it inserts on the lateral border of the patella and patellar tendon.
Fig. 8. Axial T1-weighted imaging of the mid thigh showing muscle of the quadriceps group: l,
vastus lateralis; m, vastus medialis; i, vastus intermedius; q, rectus femoris.
814 ARMFIELD, KIM, TOWERS, ET AL
The proximal attachment of the vastus intermedius is the upper two thirds of
anterolateral surface of femur and distally attaches to the upper border of pa-
tella and patellar tendon.
Vastus medialis proximal attachments include the entire length of linea as-
pera and medial condyloid ridge with the distal attachment primarily involving
the medial half upper border of patella and the patellar tendon. The most distal
fibers referred as the vastus medialis oblique (VMO) due to its oblique orien-
tation of muscle fibers that are important of patellar stability in last 10 to 20
degrees of knee extension .
Intramuscular/Central Tendon Injury
The largest study of the imaging appearance of quadriceps muscle strains fol-
lowed 40 professional Australian Rules football players for 3 years and com-
pared the rehabilitation interval (time to return to full-time training) after
Fig. 9. Axial T2 fat-saturated images from unilateral right hip MR arthrogram (higher resolu-
tion technique) shows direct and indirect tendons of the rectus femoris at the myotendinous
junction (A), tendon (B), and tendon insertion (C) levels. The white arrow represents the indirect
head, which forms the central tendon. The direct head (black arrow) inserts on the anterior in-
ferior iliac spine.
815 LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY
having completed a predefined rehabilitation regimen . Fifteen cases in-
volved the rectus femoris, six vastus intermedius, one vastus lateralis, and three
had normal MRI exam.
This study found central injuries around the central tendon had a statistically
significantly longer time to rehabilitation as compared with peripheral injuries or
epimyseal injuries that did not involve the central tendon (26.8 versus 9.2 days).
The vastus tears had an average rehabilitation interval of 4.4 days. The MR-
negative group had rehabilitation interval of 5.7 days. Injuries involving the
as well (16.2 versus 10.8 days) (Fig. 10). No distal injuries occurred in this study.
Thus, the most significant injuries were rectus femoris central tendon injures
greater than 13 cm in length or greater than 15% cross-sectional area resulting
in rehabilitation intervals of 32.7 to 35.3 days. These injuries were termed acute
bullseye lesions because of MR appearance. Peripheral injures less than 15% of
cross-sectional area had the smallest rehabilitation interval of zero days in three
cases. One hypothesis for longer rehabilitation times for healing central tendon
injuries is that scar tissue from the healing process predisposes to discordant
contraction of deep and superficial fibers resulting in chronic irritation and pro-
This central tendon injury pattern of the rectus femoris has also been de-
scribed with ultrasound with good MR correlation but the experienced authors
of this study suggested that low-grade injuries may be difficult to detect and
may be overlooked .
Proximal nonapophyseal avulsions of the rectus femoris tendon have been de-
scribed but considered rare, although with increasing awareness of intra-artic-
ular hip pathology and imaging of unilateral hips, recognition may increase. A
recent case report describes injury in two professional football kickers in the
NFL . Each had MR findings of retraction of the direct head (1 and 3.5
cm). Both were treated conservatively with the patient with the 1-cm retraction
injury ultimately returning to competition. Another case report describes a sur-
gically corrected chronic rupture of proximal myotendinous junction of the rec-
tus femoris in a soccer player with good clinical outcome  (Fig. 11).
Distally the quadriceps tendons merge together before inserting on the patella.
MR appearance of the quadriceps tendon is that of a layered structure usually
trilaminar (56%), although occasionally one (6%), two (30%), or four (8%)
layers are seen. The superficial layer represents the rectus femoris, the deep
layer the vastus intermedius, and the middle layer consists of variable contribu-
tions of the vastus lateralis and medialis  (Fig. 12).
Distal injury to the quadriceps is an unusual injury most commonly occur-
ring in individuals over 40 . Injury may occur as result of direct trauma
but usually related to forced eccentric contraction in a mildly flexed position
often in effort to regain balance during falls . Spontaneous ruptures and
816 ARMFIELD, KIM, TOWERS, ET AL
bilateral ruptures have been described in those with systemic metabolic disease Download full-text
and anabolic steroid use [40–42]. Because of the large forces required to disrupt
the tendon proper, most injury involves the myotendinous junction or under-
lying weakened tendon .
Fig. 10. (A) Focal edema around the central tendon of the rectus femoris on axial fluid sen-
sitive image. The central location of injury suggests longer rehabilitation time. (B) Coronal IR
images in a different patient with a long segment (>13 cm) injury of the central tendon. (C)
Focal peripheral injury of the rectus femoris involving a large cross-sectional area of the mus-
cle. (D) Chronic central tendon lesion of rectus femoris that has healed. Note fibrous prolifer-
ative scar tissue and lack of adjacent edema (arrow).
817LOWER EXTREMITY SPORTS-RELATED MUSCLE INJURY