University Of Miami and Jackson Memorial Hospital Musculoskeletal section, Diagnostic Radiology

P.O. Box 016960 (R-109) Miami, Florida 33101 305-549-7844

Injuries to muscle are common, accounting for up to 30% of all sports injuries [1,2]. Like the common cold, this pathologic condition has been neglected by medical science, until the recent interest in sports medicine [3].

Conventional radiography has been limited to the evaluation of end-stage muscle trauma, namely myositis ossificans. Ultrasound can detect a much wider spectrum of pathologic conditions, but operator experience is a prerequisite [2]. Recent articles in the radiologic literature have addressed MR utilization in evaluation of sports-related muscle pathology [4-7].

Our attempt here is to demonstrate and attempt an understanding of the wider spectrum of clinically imaged muscle trauma. As MRI is still considered prohibitively expensive by many clinicians, it is currently utilized mostly in the context of problem-solving and less often to depict well-established clinical diagnoses.

Fifty MR exams of traumatic muscle pathology were retrospectively reviewed. Cases were excluded if there was fracture, surgery, radiation, chemotherapy, pure tendon injury, or recent intramuscular injection. These cases were obtained from multiple institutions (mostly University of Miami) and were imaged on a variety of magnets (mostly high field strength). T1 and T2-weighted images were available in all cases.

The diagnosis was established as being correct by surgery in five cases (10%) and by a benign or non-progressive clinical course in forty-five cases (90%). The study patient population consisted of 38 males and 12 females with an average age of 35 (range 10 - 79).

Prior to imaging, the diagnosis of traumatic muscle pathology was suspected in only 17 patients (34%). In seven of these cases the MRI was for confirmation or staging of severity of injury. In the other ten cases, muscle injury was included in the differential diagnosis.
In 25 cases (50%) there was no clear history of trauma. The most common indication for MR examination was to rule out tumor in 12 patients (24%). Pain after trauma was present in only 10 patients (20%).

Multiple muscles were affected in ten (20%) of these cases, involving the same extremity in nine. Most frequent muscle location was the thigh in 20 cases, with involvement of the quadriceps muscle (n = 10), hamstrings (n =7), or the adductors (n = 3). Next most frequent site was the lower leg muscles in ten cases, most frequently the medial gastrocnemius (n =4). The remainder of cases occurred with decreasing frequency in the pelvis, shoulder, and arm; with the most frequently involved muscle at each of these sites gluteus maximus, deltoid, and biceps, respectively.

Five patterns of injury were noted in this series. Hematoma or cyst formation was present in 14 (28%). All were bright on T2-weighted images and half (7) were also bright on T1-weighted images (Figs 1-2). Five cases had a peripheral low signal rim, felt to represent either a hemosiderin rim or fibrous capsule. Less frequent features were central hypointensity (n = 2), hematocrit levels (n = 2), and peripheral scarring (n = 1). Half were in the muscle substance and half were in relation to peripheral fibers. Three of these patients required open surgery, as closed drainage was followed by hematoma reaccumulation.

Partial tears were next in frequency (n = 11), being defined as visible fiber disruption with edema, but no muscle retraction or separation to indicate a complete tear (Fig 3). Most of these injuries (n = 8, 73%) were close to the musculotendinous junction. Two partial tears were in relation to muscular insertion on bone. Four cases were noted on routine knee studies and an additional three cases were incidentally noted on other routine joint studies.

Muscle edema without a visible tear or discrete hematoma was present in 8 cases (Figs 4,5). As this edematous appearance can be non-specific, clinical information was essential to exclude cases of infection and iatrogenic insults such as injection or chemotherapy. Most frequently this was present in the mid muscle belly and only in one case was there perimuscular fluid. In one case there was a history of direct muscle trauma and in another there was a history of intermittent tender bulge, but in no case was there recent exercise which is well established as a cause of muscle edema [4,7-9].

Complete tears were present in six cases, with discontinuity equally divided between mid-muscle and musculotendinous junction (Fig 8). In these cases the symptoms were most dramatic, with the MR performed for the purpose of staging the tear. Two patients had surgical repair. One patient with infraspinatus tear, on the other hand, carried the diagnosis of myofascial syndrome prior to imaging.

Six patients had chronic injuries or tears (Fig 9). Half of these had plain film evidence supporting the diagnosis of myositis ossificans. Compensatory hypertrophy of muscles with similar function was noted in half of cases. Focal thinning or scarring of muscle was present in half. These cases were equally evident on T1 and T2-weighted images because there was no fluid or edema.

Complex or combined patterns were present in six patients (Fig 10). In four of these, there were asynchronous appearing lesions. Specifically, two patients with chronic anterior cruciate tears had acute partial tears of hip girdle muscles and another two patients had both chronic and acute tears in different muscle groups. This suggests that the older injury may have predisposed to the newer injury, by resulting in a muscular imbalance which places more stress on interrelated muscles.

There are two basic types of trauma which must be considered- direct (compression) and indirect (distraction) [2,3]. A crush injury is an example of direct trauma, and this would more likely result in hematoma formation than muscle tear. Indirect trauma related to contraction of the muscle against force would result in tear, less often resulting in hematoma. Physiologic studies have also shown that eccentric muscle contraction is more likely to result in tear than concentric muscle contraction [3,6-8,10]. The hamstring muscles, which function as decelerators of the leg (an eccentric function), are therefore predisposed to injury [3]. The relative inelasticity of the musculotendinous junction also predisposes this area to injury [3,11]. Fifteen (30%) of injuries in this series were at the musculotendinous junction.

The distribution of injuries in this series parallels that reported in the literature [1-3,12]. Thigh involvement is most frequent, the quadriceps muscle greater than the hamstrings. Muscles which cross two joints are especially prone to injury as the degree of elongation is greater in these muscles [3,10]. Thus, rectus femoris muscle is more frequently torn than the other quadriceps muscles as it is the only muscle in this group to cross two joints. Also, muscles which do the most work tend to be more proximally positioned in the extremities, so as not to interfere with mobility. These muscles have a greater proportion of type 2 fibers (fast twitch), which are also more prone to disruption [3,10].

Although much of the recent interest in muscle MRI and physiology is related to sports medicine and exercise physiology, it is evident from this survey that the majority of cases are not related to athletics. The entity of delayed-onset muscle soreness, about which much is written, was not seen at all in this series. A history of trauma, which would have suggested muscle injury to the referring clinician, was absent in the majority of our cases.

The single most frequent indication for imaging was to rule out tumor. A soft tissue swelling caused by hematoma, tear, or compensatory hypertrophy of muscle could be seen in all of these cases. The discontinuity of muscle fibers and characteristic location of these injuries is definitive in ruling out neoplasm. Neglected muscle tears have later presented as masses and when physical exam cannot determine this pathologic state, MRI can well demonstrate the anatomy [13]. On the other hand, it is also well established that neoplastic lesions can masquerade as sports related injuries and therefore be delayed in diagnosis [14]. MRI can solve these problems, being very accurate in differentiating tumor from trauma, based on mass, signal characteristics, and anatomic features.

Hematoma formation was the most frequently observed pattern of injury. We found that the signal paralleled the evolution of hemorrhage elsewhere in the body [15-18]. In this series approximately half of the hematomas were subacute and half were chronic. Extracellular methemoglobin causes the characteristic increased signal on both T1 and T2-weighted images. Older hematomas, in our experience, were similar to simple cysts in signal. Specific features such as a peripheral hemosiderin ring were noted in a few cases. We had no cases of traumatic hematoma which were indistinguishable from hemorrhagic tumor. In the latter we have always noted the presence of soft tissue components and more heterogeneity of signal.

Chronic tears resulted in deformity and were also predisposing to additional muscle injury in our experience. Atrophy, scar formation, "pseudotumor", and compensatory hypertrophy were the specific findings noted. Other authors have noted that strength imbalance of antagonistic muscles predisposes the weaker muscle to injury [3]. Unlike more recent muscle trauma, chronic injuries showed no evidence of increased signal on T2-weighted imaging. Comparison of muscle size and shape with a contralateral extremity was therefore very important in establishing these diagnoses, and has been emphasized in general by other authors [16].

Muscle edema is a non-specific muscle reaction with increase in fluid but no imaging evidence of muscle fiber discontinuity. Trauma is a common cause of this finding, and should be suspected after infection and iatrogenic etiologies are excluded. When there is a history of blunt trauma, the edema is due to contusion. In other cases, what is likely being visualized on MRI is the inflammatory response to microscopic muscle tears [8,9]. Although the pathophysiology of traumatic muscle edema in these cases is therefore similar to delayed onset muscle soreness, there was no history of recent exercise, excluding this diagnosis. The process is more likely a "first-degree" muscle strain. Muscle imbalance, fatigue, or other stress may predispose to these minor injuries.

We subdivided muscle tears of more recent onset into partial and complete categories, corresponding to what has been referred to as second and third-degree strains [16,17]. Direct visualization of muscle fiber discontinuity is possible with MRI and an essential feature for either of these diagnoses. Although partial tears may lead to complete tears, we found the location to differ in that partial tears were more frequent in the musculotendinous junction and complete tears were more frequent in mid muscle belly [3,17]. As the treatment of sports injuries approaches the intensity of interest in this area, there is likely to be more surgical intervention, especially in the instance of complete tear [19]. MRI should be critical in the staging and follow-up of these conditions.

1. Renstrom P. Swedish research in sports traumatology. Clin Orth Rel Res 1984;191:144-158

2. Holsbeeck MV, Introcaso JH. Musculoskeletal ultrasound. St. Louis: Mosby Year Book, 1991:13-56

3. Garrett WE. Injuries to the muscle-tendon unit. Instr Course Lect 1988;37:275-282

4. Fleckenstein JL, Canby RC, Parkey RW, Peshock RM. Acute effects of exercise on MR imaging of musculoskeletal muscle in normal volunteers. AJR 1988;151:231-237

5. Fleckenstein JL, Weatherall PT, Parkey RW, Payne JA, Peshock RM. Sports-related muscle injuries: evaluation with MR imaging. Radiology 1989;172:793-798

6. Shellock FG, Fukunaga T, Mink JH, Edgerton VR. Exertional muscle injury: evaluation of concentric versus eccentric actions with serial MR imaging. Radiology 1991;179:659-664.

7. Shellock FG, Fukunaga T, Mink JH, Edgerton VR. Acute effects of exercise on MR imaging of skeletal muscle: concentric vs eccentric actions. AJR 1991;156:765-768.

8. Newham DJ, McPhail G, Mills KR, Edwards RHT. Ultrastructural changes after concentric and eccentric contractions of muscle. J Neurol Sci 1983;61:109-122

9. Smith LL. Acute inflammation: the underlying mechanism in delayed onset muscle soreness? Med Sci Sports Ex 1991;23:542-551

10. Alexander RM, Ker RF. The architecture of leg muscles. In: Winter JM, Woo SL, eds. Muscle biomechanics. New York: Springer-Verlag, 1990:568-577

11. Miller WA. Rupture of the musculotendinous junction of the medial head of the gastrocnemius muscle. Am J Sports Med 1973;5:191-193

12. Conwell HE, Alldredge RH. Ruptures and tears of muscles and tendons. Am J Surg 1937;25:22-33

13. Peterson L, Stener B. Old rupture of the adductor longus muscle. Acta Orthop Scand 1976;47:653-657

14. Lewis MM, Reilly JF. Sports tumors. Am J Sports Med 1987;15:362-365

15. Gomori JM, Grossman RI. Mechanisms responsible for the MR appearance and evolution of intracranial hemorrhage. Radiographics 1988;8:427-435

16. Deutsch AL, Mink JH. Magnetic resonance imaging of musculoskeletal injuries. Rad Clin NA 1989;27:983-1002

17. Ehman RL, Berquist TH. Magnetic resonance imaging of musculoskeletal trauma. Rad Clin NA 1986;24:291-319

18. Dooms GC, Fisher MR, Hricak H, Higgins CB. MR imaging of intramuscular hemorrhage. JCAT 1985;9:908-913

19. Stanley RK. Review of quadriceps tendon ruptures. Ortho Trans 1991;15:171

1. SE (700/20) axial image of hip in 20 yo football player demonstrates hematoma with hematocrit level in superficial layer of tensor fascia lata muscle. Low signal border represented fibrous capsule at surgery.

2A. SE (500/20) axial image of proximal calf in 35 yo female with suspected tumor shows subacute hematoma in anterior aspect of medial gastrocnemius muscle, with peripheral hemosiderin ring.

2B. SE (2200/90) coronal image demonstrates its fusiform orientation between muscle fibers.

3. Sagittal SE (1500/80) image of elbow in 65 yo woman with mass of proximal forearm is shown to have a partial tear in the distal biceps longus muscle.

4. Coronal SE (1400/70) of 45 yo man with intermittent tender mass of left lateral thigh has a defect in fascia lata with focal edema of vastus lateralis related to its intermittent herniation.

5A. Coronal SE (500/20) image of calves in a 35 yo dialysis patient with swelling of the left calf.

5B. Axial SE (2000/80) shows diffuse muscle edema and perimuscular fluid involving superficial and deep posterior compartments. Cultures were negative and the patient gave history of direct trauma.

6A. Axial SE (700/20) image of mid thigh in 18 yo male with tight hamstrings on right side and mass of posterior left thigh. Note focal atrophy of semitendinosus muscle with hypertrophy of biceps femoris on the left.

6B. Coronal SE (2300/70) shows the focal tear in mid left semitendinosus with tendinous bridging. No fluid seen to indicate recent injury.

7A. Axial SE (3000/80) shows hematoma with scarring of right adductor muscles in this 10 yo who had run the marathon 1 year prior. Note also asymmetry of hamstring insertion indicating avulsion injury to the right hamstrings.

7B. Axial SE (3000/40) in mid thigh shows atrophy of right hamstrings due to old injury. There is focal edematous change in the left rectus femoris muscle, due to muscle imbalance and mild strain.