Traumatic injuries of thigh and calf muscles in athletes: role and clinical relevance of MR imaging and ultrasound

This group of muscles occupies the posterior compartment of the thigh and consists of the long and the short heads of the biceps femoris laterally and the semimembranosus and the semitendinosus muscles medially. Hamstrings cross both hip and knee joints, and integrate extension at the hip with flexion at the knee. The greatest musculotendinous stretch is incurred by the biceps femoris [16], which may contribute to its tendency to be the most commonly injured muscle in the hamstring muscle group, especially during high-speed running [17]. The semitendinosus muscle is a thin, band-like muscle with a long tendon distally, which may predispose the muscle to rupture [18]. The semimembranosus muscle arises from the superolateral part of the ischial tuberosity. It is important not to mistake this muscle for the semitendinosus muscle, since the proximal tendon of the latter may not always form a distinct structure.

Hip adductors comprise, from lateral to medial, the pectineus, the adductors (longus, brevis, and magnus) and the gracilis. Of this group, the adductor longus is the most commonly injured [19]. This muscle is thought to be most susceptible to overstretching during such movements as lunging for a ball or sidestepping an opponent during a football or a rugby match. Injury may also be caused by sudden resistance to a strong adduction force, such as an opponent blocking forceful hip adduction in a block tackle [19].

The quadriceps muscles occupy the anterior compartment of the thigh and comprise the rectus femoris, vastus medialis, vastus lateralis and vastus intermedius. Proximally, the straight head of the rectus femoris originates from the anterior inferior iliac spine, while the reflected head originates from the groove just above the acetabulum. Distally, these four muscles converge to insert on the superior pole of the patella as the trilaminar quadriceps tendon [20]. Of the quadriceps muscles, the rectus femoris muscle is the most commonly injured [9, 21]. This is thought to be because it crosses both the hip and knee joints, contains a high percentage of type II fibres, and has a complex musculotendinous architecture [22‐24].

The calf muscle complex occupies the posterior compartment of the distal lower limb and includes the medial and the lateral heads of the gastrocnemius, and the soleus, which are collectively known as the triceps surae. The two heads (medial and lateral) of the gastrocnemius unite into a broad aponeurosis, which eventually unites with the deep tendon of the soleus to form the Achilles tendon [25]. The gastrocnemius is considered at high risk for injury because it crosses the knee and ankle joints and has a high density of type II fibres [21, 26]. The soleus is considered low risk for injury because it only crosses the ankle joint and is largely comprised of type I fibres [27]. The plantaris muscle crosses the knee and ankle joints before it joins the Achilles tendon insertion on the posterior surface of calcaneus. However, the plantaris is thought to be largely vestigial and is rarely involved in calf injuries [26, 27], and thus is not included in this review.

Musculotendinous strains and tears may be caused by a single traumatic event from excessive stretching on musculotendinous fiber (e.g., in high-speed runners) [17], from movements involving excessive range over sequential joints (e.g., in dancers) [32], or as a result of eccentric contractions (e.g., in football players) [33]. In some elite sports, such as track and field, football and rugby, the micro-damage from mild eccentric exercise may progress to more major tears [33]. The lesion is located at the musculotendinous junction, commonly in the superficial muscle layers, but the location may vary depending on the mechanism of injury. Musculotendinous strains can be clinically classified as grade 1, grade 2, and grade 3 based on absent, mild, or complete loss of muscle function, respectively [3]. These injury grades can be used to estimate the convalescent period and to design an appropriate rehabilitation program [34]. Muscle strains typically affect the muscles that extend across two joints, have a high proportion of fast-contracting type II fibres and fusiform shape, and undergo eccentric contractions [3, 35]. In the lower extremities, the hamstrings, rectus femoris and gastrocnemius muscles are commonly involved.

In musculotendinous strain without a tear, some fibre disruptions are seen as a result of a stretch injury, but muscle functions are maintained and treatment is conservative. On MR images, interstitial oedema and haemorrhage are present at the musculotendinous junction and extend into the adjacent muscle fascicles, producing a feathery appearance (i.e., hyperintensity) on fluid-sensitive sequences [28, 36] (Figs. 1, 2 and 6a). However, up to 45 % of clinically diagnosed grade 1 hamstring injuries may have a normal appearance on MR imaging according to one study [8]. On ultrasound, the lesion may be depicted as hyperechogenicity [37], or hypoechogenicity, or may appear normal [37].

In the presence of partial tears of fibres without retraction, there is a mild loss of muscle function. On MR images, in addition to interstitial oedema and haemorrhage, haematoma at the musculotendinous junction and perifascial fluid collection appear as hyperintensity on fluid-sensitive sequences (Figs. 3, 4 and 6b). On ultrasound, these pathologic features are depicted as hypoechogenicity (Figs. 3 and 5). Disruption of muscle fibres will be depicted as notable echo inhomogeneity (Fig. 8). Treatment of partial tears is also conservative.

Complete musculotendinous rupture is commonly accompanied by a haematoma (Fig. 7). The diagnosis is usually made on clinical grounds, i.e., complete loss of muscle function, with palpable gap and muscle fiber retraction. Surgical repair is an option, depending on the location of the rupture [38], and both MR imaging and ultrasound may be useful for preoperative assessment of the extent of retraction [3]. Extensive acute oedema and haemorrhage may limit accurate evaluation of the injured muscle. If the tears are left untreated, the ends may become rounded and tether to adjacent muscles or fascia [39].

Muscle contusions result from direct trauma, usually by a blunt object [35]. The injury consists of a well-defined sequence of events involving microscopic rupture and damage to muscle cells, macroscopic defects in muscle bellies, infiltrative bleeding, and inflammation. As a complication, myositis ossificans traumatica may develop [40]. Unlike strains, these traumas usually occur deep in the muscle belly and tend to be less symptomatic than strains. Severity depends on the site of impact, the activation status of the muscles involved, the age of the patient, and the presence of fatigue [41].

On ultrasound, contusion is characterised by discontinuity of normal muscle architecture, with ill-defined hyperechogenicity that may cross fascial boundaries [37]. MR imaging varies according to severity of injury, but typically there is a feathery appearance of diffuse muscle oedema on STIR and FS T2-w images [1] (Fig. 8). Increased muscle girth can be observed but there are no other architectural changes, such as fibre discontinuity or laxity. In case of severe trauma with muscle fiber disruptions, deep intramuscular haematoma is seen [3] (Fig. 9). Signal intensity within the haematoma is influenced by the concentration of protein, methaemoglobin, magnetic susceptibility at high field strength, and tissue clearance [42]. Acute haematomas (<48 h) are typically isointense on T1-w images, and subacute haematomas (<30 days) appear hyperintense relative to muscle on both T1-w and fluid-sensitive sequences secondary to methaemoglobin accumulation [28]. As the haematoma evolves, a wide range of MR signal intensity can be seen within the collection, depending on the age of degradation products. Chronic haematoma characteristically shows a rim of hypointensity on all pulse sequences due to haemosiderin. As blood degradation products get reabsorbed over a course of 6–8 weeks, the size of haematoma will decrease [30].

Acute avulsion injuries result from extreme, unbalanced and often eccentric muscular contractions, and patients with such injuries present with severe pain and loss of function [43]. Adolescents are particularly vulnerable to avulsion injuries because of the inherent weakness of the apophyses. The many apophyses in the pelvis and hip are common sites of avulsion injuries (Fig. 10). The single most common site of apophyseal avulsion is at the ischial tuberosity [44]. Cheerleaders, sprinters, gymnasts, track athletes, American football players, and baseball players are commonly affected [44]. Treatment for avulsion injury is generally conservative and the prognosis is good, but non-union may occur.

In acute avulsion injury, periosteal stripping with haematoma at a tendon attachment site can be depicted by MR imaging. A wavy appearance and retraction of the torn end of the tendon with fragments of bone/cartilage is characteristic. The redundant tendon edge may be lying in a large fluid collection/haematoma. Ultrasound evaluation is also useful, but may be difficult due to the presence of a mixed echogenicity haematoma which has echogenicity similar to the avulsed tendon [37].

Imaging features of chronic musculotendinous injury include muscle or tendon retraction or compensatory hypertrophy, muscle atrophy and formation of scar tissue (fibrosis) [45]. In chronic injuries, T1-w images may be normal in low-grade injuries, but the fluid-sensitive sequences are helpful for detection of symptomatic old tears which are depicted as abnormal hyperintensity [30]. There may be associated surrounding oedema and haemorrhage due to re-injury at the site (Fig. 4) [30]. Scar tissue may be observed as early as 6 weeks after initial injury [13]. On MR imaging, it appears as hypointensity on all pulse sequences and, on ultrasound, areas of scar tissue have irregular morphological features and show heterogeneous echogenicity [31]. It is important to identify the scar tissue because recurrent injuries can occur in the close proximity due likely to elasticity differences and altered contractility [30]. Imaging features of all the types of injuries discussed above are summarised in Table 1.

At follow-up, grade I injury may be manifest as a region of hyperechogenicity in up to 50 % of cases on ultrasound [46]. In such cases, normal healing is typically evident as a decrease in size or resolution of the area of hypoechogenicity, together with return of normal muscle architecture and echotexture [29]. More severe injuries may be characterised by the presence of hypoechoic regions indicative of fluid adjacent to muscle fibrils or the epimysium. Resolution or notable reduction in the amount of fluid is expected during the normal healing process. Any haematoma or fluid collection should decrease in size, and macroscopic muscle tears may show echogenicity of the margins of the tear as healing occurs. As the healing process progresses, small tears may fill with echogenic material, which is presumed to be scar tissue [29]. Scar tissue formation at the site of injury can be seen at 6 weeks [30]. However, ultrasound is less sensitive than MR imaging to residual muscle injury during follow-up [13].

MR imaging of healing muscle injuries typically show gradual resolution of fluid between muscle fascicles and in relation to the epimysium, together with gradual reduction in the extent and intensity of T2 signal within muscle. The degree of resolution of T2 hyperintensity resolution is variable depending on the severity of the initial injury, but in many cases MR signal abnormality has not resolved by 6 weeks despite return to competition, especially when the initial injury was severe. Persistent high T2 signal suggests ongoing healing and resolution of injury in keeping with the fact that the ultrastructural healing process continues for weeks to months, even after the time when athletes usually return to competition [29]. Hypointensity may also be seen during muscle healing, and reflects the formation of scar tissue and/or haemosiderin deposition following haemorrhage. These changes may contribute to the susceptibility artifacts seen on T2-w images during follow-up [30]. During the first few weeks of healing, there may be thickening and T2 hyperintensity of the central intramuscular tendon at the site of injury. As maturation of the scar occurs, T2 hypointensity replaces the hyperintensity. A recent study of athletes with grade I and II hamstring injuries, follow-up MRI was performed 5–23 months post-injury [45]. Hypointensity representing scar tissue was seen along the musculotendinous junction in 79 % (11/14) of subjects. Muscle volume reduction following the injury of long head of biceps was observed, but fatty infiltration was infrequently seen on these follow-up MR imaging [45].

Research efforts are ongoing to develop MR imaging techniques for assessment of skeletal muscle ultrastructure. Diffusion tensor imaging (DTI) can provide detailed information on muscle structural changes [5]. A recent study showed that the water diffusivity parameters of the thigh muscles were not influenced by the age or gender of healthy subjects. However, the hamstrings and the quadriceps did show differences in such parameters, which, the authors speculate, may reflect differences in hydration and muscular architecture [47]. Muscle fibre tracking with DTI has been used successfully to assess muscle damage [48] and to measure pennation angle [49]. Muscle is fairly uniform in bulk directionality and exhibits orderly arrangements on DTI. Following muscle injury, disturbance of the normal arrangement will be observed [5]. DTI and fiber tracking packages are sold by all the major MR imaging vendors and can be performed using routinely available clinical scanners. Thus, DTI may become a useful tool in our clinical practice in the future.

Additionally, the use of MR spectroscopy (MRS) using phosphorus [50] or sodium [51] to assess skeletal muscle disorders has been explored. Phosphorus (³¹P) MRS measures oxidative metabolism and sodium (²³Na) MRS measures tissue sodium levels in muscle during exercise and recovery. These techniques may be useful as a noninvasive tool in exercise physiology or sports medicine to better understand the dynamics of an exercising muscle [5], but their applicability and relevance to evaluation of muscle injury are yet to be determined.

There are high demands on sports medicine physicians to quantify the prognosis and to predict when the athlete can return to training and competition. Imaging may assist in the prognostication of the healing process and in predicting the risk of recurrence, but the decision on return to play cannot be dependent on the imaging findings alone and must be balanced against the clinical situation [15, 52]. It has been shown that athletes with a normal MR imaging study in the presence of clinically suspected muscle injury require a shorter convalescence interval (1–2 weeks) [8] and have a lower recurrence rate [29] than those with an abnormal MR study. However, when there is an abnormality on MR or ultrasound, there is no conclusive evidence that the extent of the abnormality can predict the risk of recurrent injury, whether the images are acquired shortly after the injury or just prior to returning to competition [29].

Imaging parameters used to estimate the extent of muscle injury include the percentage of the cross-sectional area of the affected muscle, the craniocaudal length of the muscle lesion adjacent to the musculotendinous junction, and the approximate volume of muscle injury [29]. These parameters are associated with the duration of absence from competition and thus may guide clinicians in managing muscle injuries [9, 13, 53, 54]. Studies have shown that complete tear of the hamstring, as well as hamstring injuries involving >50 % of muscle cross-sectional area, haemorrhage, fluid collections and distal myotendinous involvement, were associated with a longer recovery time [53, 54]. A study of sprinters demonstrated that time to return to pre-injury level following hamstring injury involving the proximal tendon was significantly longer (approximately 35 weeks) than when the proximal tendon was not involved (less than 15 weeks) [17]. In regard to the quadriceps, sprinters with acute injury involving the central tendon of the rectus femoris was shown to have a mean recovery time of 26.85 days, which was significantly longer than those who sustained injury to the peripheral tendon of the rectus femoris (9.17 days) and vasti muscles (4.42 days). Other studies also showed that the larger the size of the injury on MR imaging the greater the risk of recurrence [12, 55]. Follow-up imaging may thus provide additional information to support clinical progress through a rehabilitation program [45], although many athletes will return to activity before the MR imaging findings are resolved [11] (Figs. 7, 8 and 9).

High quality imaging is of great clinical relevance in planning and guiding athlete rehabilitation. It is well established that MR imaging is the best technique for monitoring the muscle healing process, although this is tempered by the need to let the clinical evaluation guide return-to-play decisions [13]. No official guidelines on the role of MR imaging or ultrasound in evaluating muscle injuries in high-level athletes have been published. An algorithm that integrates clinical and imaging information into an explicit management plan would be very useful and needs to be developed [15, 29].