Chấn thương đụng dập cơ: Tùy chọn điều trị hiện tại docx

Vol 9, No 4, July/August 2001 227 Traumatic muscle contusion is a common cause of soft-tissue injury in virtually all contact sports. In fact, contusion and strain injuries make up approximately 90% of all sports-related injuries. 1,2 Other than strain injuries, contusion caused by impact with a blunt, nonpenetrating object is the most frequent muscle injury. 3 The symptoms of a contusion injury are often nonspecific, and include soreness, pain with active and passive motion, and limited range of motion. Without a straight- forward history of impact to the area, the diagnosis often becomes one of exclusion. Many contusion injuries go unreported and un- treated. Healing of these injuries is a complex phenomenon depending on multiple factors, both within and outside the control of the clinician. No universally accepted treatment modalities have been developed. Most treatments follow the “RICE” principle (rest, immobilization, cold, and elevation) at least in the short term, but clinicians differ as to the best long-term treatment. Common sites of contusion injuries include the anterior, poste- rior, and lateral aspects of the thigh and the upper arm in the region of the brachialis (causing “tackler’s exostosis”). Contusions in the area of the quadriceps and the lateral thigh may cause excessive hema- toma to accumulate due to the large potential space. 4 A frequent com- plication is ossification of the hema- toma in response to mechanisms that are as yet unclear. It is generally felt that injury sufficient to cause proliferative repair is essential to the development of myositis ossifi- cans. 5 At the level of the muscle fibers, capillary bleeding and edema can lead to hematoma formation and can cause compartment syn- drome in areas in which the volume is limited by the fascial envelope. There are a number of common types of muscle injuries (Table 1). Several excellent reviews of muscle strain injuries 2,6,7 and of exercise- induced muscle injury 8-10 have ap- peared in the recent literature, but there have been none that summarize the body of literature dedicated to muscle contusion injury. There are, however, a large number of studies, especially those reporting on ani- mal research, that detail the mecha- nisms of injury, the natural history, and the effects of various treatment modalities. The lessons learned in the laboratory can now begin to be translated to the care of the injured patient. Dr. Beiner is Resident, Department of Ortho- paedics and Rehabilitation, Yale University School of Medicine, New Haven, Conn. Dr. Jokl is Vice Chairman and Chief, Section of Sports Medicine, Department of Orthopaedics and Rehabilitation, Yale University School of Medicine. Reprint requests: Dr. Jokl, Yale University School of Medicine, Suite 600, 1 Long Wharf Drive, New Haven, CT 06511. Copyright 2001 by the American Academy of Orthopaedic Surgeons. Abstract Muscle contusion is second only to strain as the leading cause of morbidity from sports-related injuries. Severity depends on the site of impact, the activa- tion status of the muscles involved, the age of the patient, and the presence of fatigue. The diagnosis has traditionally been one of clinical judgment; however, newer modalities, including ultrasonography, magnetic resonance imaging, and spectroscopy, are becoming increasingly important in both identifying and delineating the extent of injury. Although controlled clinical studies are scarce, animal research into muscle contusions has allowed the description of the nat- ural healing process, which involves a complex balance between muscle repair, regeneration, and scar-tissue formation. Studies are being performed to evalu- ate the effects of anti-inflammatory medications, corticosteroids, operative repair, and exercise protocols. Prevention and treatment of complications such as myositis ossificans have also been stressed, but recognition may improve the outcome of these ubiquitous injuries. J Am Acad Orthop Surg 2001;9:227-237 Muscle Contusion Injuries: Current Treatment Options John M. Beiner, MD, and Peter Jokl, MD Mechanisms of Injury The clinical entity of a muscle con- tusion injury is most often seen after a direct blow to an extremity. In football, this frequently occurs in the anterior, medial, or lateral thigh in the area of the muscle belly of the quadriceps femoris. 11 The greatest number of quadriceps contusions in one study occurred in tackle foot- ball, although the percentage of injuries was higher in rugby, karate, and judo. 11 In soccer, after the widespread adoption of the use of shin guards, the thigh is now the most commonly injured area as well. However, these injuries have been reported in virtually all contact sports. The injury is associated with pain and swelling, a decreased range of motion of joints spanned by the injured muscles, and occasionally a permanent palpable mass. 11 In ani- mal studies, at a microstructural level, contusion injury usually causes a partial rupture of the muscle, cap- illary rupture, and infiltrative bleed- ing, leading to hematoma formation within the developing gap and around the intact muscle fibers, edema, and inflammation. 12 De- spite all these changes, some func- tional capacity usually remains in the affected muscle. 13,14 The archi- tecture of the damaged muscle bed is a mix of disrupted muscle cells and collagen connective tissue. The healing process is a delicate balance between the formation of scar tissue by fibroblasts and the regeneration of normal muscle by migrating myoblasts. Injury Severity Information regarding the structural, cellular, and biochemical events in contusion injury is essential to the rational application of sports therapy. Studying these injuries is difficult, however, because of the inherent variability in severity. In contrast, the research setting provides a means to control many of the confounding variables involved in muscle contu- sion research. Models of contusion that have been developed use spring- loaded hammers, crushing hemostat forceps, reflex hammers, and a vari- ety of other devices to cause single or multiple contusion injuries ranging from the mild to the severe in rodents and nonhuman primates. Only two, however, have been able to deliver a standardized crush injury. Järvinen and co-workers 15-17 developed a rat model of muscle contusion injury involving the use of a spring-loaded hammer and compared the effects of mobilization and immobilization on the healing process. They found that early mobilization increased the ten- sile strength of the muscle compared with similarly injured muscles immo- bilized in a plaster cast. 15-17 Stratton et al 18 used a drop-mass technique that delivers a single blow to muscle to study the effects of therapeutic ultrasound on the injury. A problem common to all of these models, however, is the in- ability to characterize the injury in terms of force, displacement, energy, and impulse of the impact actually experienced by the target muscle. Crisco et al 19 developed a model to record these variables in the pro- duction of a standard, reproducible muscle contusion injury to the rat gastrocnemius-soleus muscle com- plex. Others have used this same model to observe a standard contu- sion injury that causes hematoma formation, with disruption of indi- vidual muscle fibers but preserva- tion of others, a brisk inflammatory reaction, and marked interstitial edema. 20 The extrinsic factors that affect injury severity have not been well documented. The debate continues in the sports arena as to whether athletes should “tighten up” before impact during athletic contests in order to minimize injuries. In stud- ies of muscle strain injury, it has been shown that an activated or contracted muscle will absorb more energy and require a much higher force to failure than passively stretched muscle. 2,21 Crisco et al 22 showed that con- tracted muscle was able to absorb more energy during impact than relaxed muscle. The peak force recorded was less pronounced than that in passively impacted muscle. This is complicated, however, by the fact that the impacted legs were an in vivo composite of skin, mus- cle, fascia, and bone. Contraction simply stiffened the muscle relative to the bone, allowing protection from injury. Later experiments by Beiner 23 continued the work of Crisco et al 22 and found that the relaxed muscle- bone composite was significantly (P<0.05) stiffer than the contracted muscle-bone composite. This was Muscle Contusion Injuries Journal of the American Academy of Orthopaedic Surgeons 228 Table 1 Common Types and Causes of Muscle Contusion Exercise-induced injury (“delayed onset muscle soreness”) Strain Laceration Traumatic Surgical Vascular Tourniquet Traumatic vascular injury Infectious Bacterial Viral Neurologic Denervation Viral (central or peripheral) Traumatic (central or peripheral) Neuropathic Metabolic Viral Genetic Myopathies due to the fact that on impact some of the bulk of the relaxed muscle parted, concentrating the force of the impacting sphere on part of the muscle near the bone. In contrast, the contracted muscles were able to absorb energy by displacing less, distributing the force over the entire muscle belly, and avoiding severe damage to any one area. Energy absorbed was 10% more than in the relaxed muscle-bone composite (P<0.05). These concepts are illus- trated by Figure 1, showing that two peaks are present for impacts to relaxed muscle, one for initial impact on the muscle and the sec- ond as the impactor compresses the remaining muscle and hits the bone. Changing the shape of the impact- ing surface into a bar rather than a sphere changed the injury slightly, but did not seem to change the over- all force-generating capacity of the muscles following injury. To model the effect of constrain- ing hard or soft padding or taping on muscle injury, Beiner 23 analyzed the effects of muscle contraction with exterior constraint (by enclos- ing the entire leg in a narrow-walled chamber during impact), which lim- ited the extent of the lateral defor- mation available to the muscle as it absorbed impact. This seemed to cause a much more severe injury. When the muscle was externally constrained during impact, the force-displacement curves of the contracted and relaxed muscle-bone composites were comparable. The injury was 11% greater for con- strained muscles in subsequent con- tractile testing (P<0.05). Constrain- ing the muscle also caused the energy absorbed to increase by ap- proximately 11%, as occurs with contraction. It may be that the mus- cle could not deform while con- strained, resulting in more severe injury. John M. Beiner, MD, and Peter Jokl, MD Vol 9, No 4, July/August 2001 229 Figure 1 Force-displacement behavior of rat gastrocnemius-soleus muscle complex impacted in either the contracted or relaxed state with a drop-mass technique. The constrained muscles were held with walls on either side, limiting their lateral displacement. Constraining and contraction caused the peak forces to be distributed over a broader area, changing the impulse to the muscles. All impact stimulation was at 100 Hz and 70 V, with a 0.1-msec pulse duration and 1.5-sec train duration. Curves are mean ± SD (N = 27). 220 200 180 160 140 120 100 80 220 200 180 160 140 120 100 80 200 180 160 140 120 100 80 60 40 20 0 0 2 4 6 8 10 12 Displacement, mm 200 180 160 140 120 100 80 60 40 20 0 0 2 4 6 8 10 12 Displacement, mm 60 40 20 0 0 2 4 6 8 10 12 Force, N Displacement, mm 60 40 20 0 0 2 4 6 8 10 12 Force, N Displacement, mm 220 Force, N 220 Force, N Nonconstrained Constrained Contracted Relaxed Beiner 23 found that both the status of the activation of the muscle during impact (contracted versus relaxed) and the relative level of external con- straint of the muscle predicted the force the muscle could generate in contractile testing. Contracted mus- cle generated a 10% increased force relative to relaxed muscle (P<0.05), while constrained muscle was weaker by 11%. Clinical correlates to exter- nal constraint include design of pads; the relative volume of muscle that is protected by an enclosing hard plas- tic pad may affect how the muscle absorbs the energy of impact. More research is needed in this area before further recommendations can be made in the sports arena regarding equipment design and protective measures for impact. Fatigue has been shown to affect the ability of stretched muscle to withstand injury, 24 as has tempera- ture 25 ; no similar studies have been performed in the setting of contusion injury. Fatigue lessens the ability of a muscle to fully contract, and con- traction seems to protect the muscle from injury, but a direct causal rela- tionship has yet to be established. Physiologists have long known that muscles operate best within a certain temperature range. Warm-up before exertion thus has obvious benefits, but a direct relationship between overheating, fatigue, and injury has not been delineated. Muscles in young rats seem to undergo more intense inflammation, with more proliferation of fibro- blasts and production of collagen, than old muscles. 26 Young muscles also heal more rapidly and more completely, suggesting the greater power of young regenerating tissue to respond to injury. Diagnosis The clinical diagnosis of contusion injury is often fairly direct (Fig. 2). The patient experiences local swell- ing, tenderness, pain, and impaired athletic performance. The extent and type of soft-tissue injury, how- ever, are less readily established. Many researchers have attempted to demonstrate the usefulness of imaging in determining the extent and the healing of contusion injury. Ultrasound has been used success- fully to distinguish pervasive swell- ing and edema from a localized, cir- cumscribed hematoma. 27 It has also been advocated as a noninvasive aid in determining when to consider surgical evacuation of the hema- toma and when to choose the less aggressive compression and early mobilization. Magnetic resonance (MR) imaging has also been used to evaluate patients with the clinical signs and symptoms of contusion injury, but its role is currently limited to selected patients. It is most useful in the sub- acute setting when a definite history of trauma is lacking. 28 Although the clinical uses of MR imaging in fol- lowing contusion injury are less well defined, it has been shown to be more sensitive than computed to- mography (CT) for the detection of hemorrhage. 29 It may allow sequen- tial follow-up during healing, and the addition of contrast material may enhance injury recognition and eval- uation of the extent of injury. 30 Muscle Contusion Injuries Journal of the American Academy of Orthopaedic Surgeons 230 Consider operative repair Early mobilization with passive range of motion, stretching Pain-free passive range of motion Consider myositis ossificans Immobilize in neutral position (no tension on repair) Contusion with muscle tear (gap, fascial tear, or avulsion detected by physical examination or imaging) Contusion without muscle tear Immobilize muscle in stretched position for 24 hours, NSAIDs 24 to 48 hours, avoid steroids Assess severity of muscle contusion injury: • Physical examination (range of motion, palpable gap) • Ultrasound, magnetic resonance imaging Progress to concentric active range of motion and strengthening to tolerance Prolonged painful range of motion, swelling, erythema Functional rehabilitation with graded increased eccentric range of motion Figure 2 Algorithm for the evaluation and treatment of muscle contusion injuries. NSAIDs = nonsteroidal anti-inflammatory drugs. Standard MR imaging provides in- formation regarding the site and ex- tent of injury, but MR spectroscopy, in limited use for some years, can also be used to estimate the ratio of inorganic phosphate to phosphocrea- tine, which reflects the metabolic response to muscle injury. 31 The Healing Process Fisher et al 32 gave a detailed account of the ultrastructural events after muscle contusion injury to the rat gastrocnemius muscle. Figure 3 shows the histologic appearance of normal healing of contused muscle. Muscle consists primarily of tissue derived from cells of two separate and distinct lineages: fibroblasts and myoblasts. After injury, the damaged segments show gross tear- ing and degeneration. A large num- ber of mononuclear cells are drawn to the injured area, with an intense inflammatory response and intersti- tial edema. By 24 to 48 hours, there is an increase in the number of sar- colemmal nuclei, with activation and proliferation of the satellite myogenic cells lying between the basal lamina and the plasma mem- brane of the muscle fibers. By day 3, regenerating muscle cells display central nuclei and reorganizing sar- comeres. By day 6, focal interstitial collagen formation suggests mini- mal to mild scar formation. After 14 to 21 days, no residual evidence of the injury is apparent. Lehto and Järvinen 33 described the important role played by the basal lamina in the regeneration of muscle. If it is intact, it acts as a bar- rier to fibroblast infiltration and as a scaffold for myoblast proliferation. With more severe injuries, when the gap in the damaged muscle fibers is larger, the ruptured gap can be filled with proliferating granulation tissue and later by a connective tis- sue scar. 16,34 As described by Lehto and Järvinen, 33 healing of injuries is dependent on several factors: dam- age to the neural input, vascular ingrowth, oxygen supply, the rate John M. Beiner, MD, and Peter Jokl, MD Vol 9, No 4, July/August 2001 231 Figure 3 Histologic sections of muscle tissue after contu- sion injury (hematoxylin-eosin; original magnification ×200). A, At day 2, hematoma is evident, as well as a brisk inflam- matory reaction with marked interstitial edema. B, At day 7, there is evidence of removal of the necrotic tissue, disper- sal of the inflammatory cells, and infiltration. C, At day 14, the tissue looks very similar to normal muscle, with clearing of necrotic tissue, regeneration of fibers, and relatively nor- mal tissue architecture. (Reproduced with permission from Beiner JM, Jokl P, Cholewicki J, Panjabi MM: The effect of anabolic steroids and corticosteroids on healing of muscle contusion injury. Am J Sports Med 1999;27:2-9.) C A B and geography of myoblast fusion to myotubes, the collagen cross- linking, and the overall race be- tween regenerating myoblastic cell infiltration and granulation and scar formation. Some, but not total, re- modeling occurs later. Histologic staining with vimentin provides qualitative and quantitative markers for mesenchyma-derived cells. Trichrome staining tracks colla- gen. Crisco et al 19 used markers for protein and collagen formation to study the healing after contusion injury in a rat model. At day 0 after contusion injury, no vimentin was noted, but inflammatory cells were present. At day 2 of healing, an in- tense inflammatory response with phagocytosis of necrotic muscle fibers and supporting tissue was noted. The basement membranes were intact, and spindle-shaped fibroblasts were present in moderate numbers. Trichrome stains demonstrated the presence of collagenous material beginning to form in the area. Slight vimentin activity was noted at the periphery, indicating differentiation of myoblast precursor cells from satel- lite stem cells (Fig. 4, A). At day 7 of healing, trichrome staining of colla- gen showed increased scarring in the central areas where the muscle archi- tecture was destroyed. A marked increase in vimentin staining was noted, localized to the center as well as to the periphery at this time point (Fig. 4, B). By 24 days after injury, there was no difference between damaged muscles and control mus- cles with regard to the staining pat- terns. Some scar tissue was still evi- dent, however, in the most severely damaged muscle. Healing in the rat model is recog- nized as more accelerated than in humans, but just how much faster is a matter of controversy. Certainly there are phylogenetic differences between animals and humans, and healing in humans is usually shown to be slower and less complete than in an animal model. Muscle Contusion Injuries Journal of the American Academy of Orthopaedic Surgeons 232 Figure 4 A, Histologic sections at day 2 after injury (original magnification ×200). Top, Trichrome stain shows intense inflammatory response with phagocytosis. Intact basement membranes are seen as thin lines stained blue. Bottom, With vimentin stain, slight activity (red) is noted at the periphery of the injury adjacent to the intact fibers (IF). B, Histologic sections at day 7 after injury (original magnification ×200). Top, With trichrome staining, collagenous (blue) and proteinaceous (red) ground substance can be differentiated. Bottom, Intense vimentin activity (red) is noted at the periphery of the injury and extends centrally. (Reproduced with permission from Crisco JJ, Jokl P, Heinen GT, Connell MD, Panjabi MM: A muscle contusion injury model: Biomechanics, physiology, and histology. Am J Sports Med 1994;22:702-710.) A B IF Clinically, studies of the healing of contusion injuries are necessarily influenced by the type of treatment used, whether it be immobilization, activity ad libitum, or some other modality. Animal studies have been conducted in an attempt to define the clinical course of thigh contu- sions. In a sheep model, the injury caused extensive scarring, with periosteal bone formation and het- erotopic bone formation in 17% of the legs within 3 weeks to 3 months after trauma and replacement of muscle tissue by intramembranous ossification within scar tissue. 12 Several earlier studies reported no ossification, despite extensive ne- crosis, regeneration, and granula- tion tissue. Human studies of contusion in- juries are limited. The most impor- tant of these are the West Point studies of quadriceps femoris con- tusions. 4,11 The initial study deter- mined a rationale for treatment and therapy with an emphasis on achieving full extension, with im- mobilization in extension during rest. 4 Later, the researchers found that normal flexion was the vari- able that was slowest to return, and this lack of flexion prolonged dis- ability after pain resolved. 11 They subsequently modified their proto- col to immobilize the muscle in a stretched position, with early motion emphasizing flexion. They classified injuries by range of motion at 12 to 24 hours after in- jury. Mild injuries were defined as those after which range of motion greater than 90 degrees was possi- ble; moderate, 45 to 90 degrees; and severe, less than 45 degrees. Aver- age disability (defined as inability to participate in full cadet activi- ties) was 13 days for mild contu- sions, 19 days for moderate inju- ries, and 21 days for severe injuries. This contrasted with the much longer disability (up to 72 days) with the previous treatment pro- tocol. Myositis Ossificans Myositis ossificans has long been recognized as a leading complica- tion of muscle contusion injury. Although certain regions are more prone to the development of myosi- tis, such as the quadriceps and brachialis, the mechanisms have not been clearly established. Similar to the development of heterotopic ossi- fication after surgical dissection, the factors that make some patients prone to this complication are un- clear. Myositis ossificans was a complication of 9% of the contusion injuries in the West Point studies, and was found to be related to the initial grade of injury (based on range of motion). 4,11 Several different kinds of myosi- tis have been identified. In the stalk type, there is a thin stalk of bone connecting the ossified muscle to the underlying bone. In the periosteal type, there is a broad-based region of ossification in contact with the underlying bone. In the third type, the ossified muscle is not connected to the underlying bone at all, but rather seems to derive entirely from the affected muscle. Within 3 weeks after injury, os- teoblastic activity can be detected with bone scanning. To minimize the risk of recurrence, surgical removal should be delayed until the bone has matured (usually after 6 months to 1 year) and no longer shows increased uptake on a bone scan. Treatment A general approach to the treat- ment of muscle contusion injuries is shown in Figure 2. Operative Treatment Traditionally, muscle contusion injuries have been treated nonoper- atively. Many surgeons have reported their anecdotal sense that in the presence of hematoma and a palpable defect in the muscle belly, it is difficult to suture the muscle together, as there are frequently no fascial ends to close, and muscle fibers are poorly reapproximated. However, recent animal studies have provided increasing evidence that in the setting of a contusion injury that causes a spatial defect in the muscle belly, suturing with large absorbable sutures through the thick substance of the muscle does decrease the distance between the lacerated edges, allowing faster healing. 27 Following the healing of rat gastrocnemius muscles with MR imaging, Mellerowicz et al 30 found that “suture of the divided muscles resulted in more rapid healing with- out major defects.” In a mouse model, suturing of the cut ends of the muscle resulted in “better healing of the injured muscle and prevented the develop- ment of deep scar tissue in the lacer- ated muscle.” 35 The authors found that tetanic strength was 81% of that in control muscles for sutured mus- cles, 35% for untreated lacerated muscles, and 18% for immobilized muscles at 1 month after injury. They recommended repair with a modified Kessler stitch. Another study stressed the need for exercise after laceration of mus- cle. The authors found that the regenerating muscle-scar composite eventually regained almost com- plete (96%) resistance to stress, but the surrounding area of atrophied muscle made the muscle unit as a whole weaker when immobilized. They did not perform contractile testing. 36 Human studies in this area are lacking. Immobilization Versus Early Mobilization Immobilization was long used as part of the rehabilitation of muscle contusion injuries. The complica- tions of immobilization, even for short periods of time, including rerupture, muscle atrophy, joint stiff- John M. Beiner, MD, and Peter Jokl, MD Vol 9, No 4, July/August 2001 233 ness, and a high incidence of myosi- tis ossificans, prompted studies of early mobilization. In a study com- paring mobilization and immobiliza- tion after contusion injury in rats, the immobilized legs lost 30% of their weight, but no such atrophy was observed in the mobilized legs. In addition, delayed contraction and maturation of the fibrous scar were noted in the third week after injury. 16 These effects occurred even after only 2 to 5 days of immobilization. Studying load-deformation curves when pulling injured muscle to fail- ure after contusion injury, Järvinen 13 found that the muscles mobilized on a treadmill failed at a significantly greater force than the immobilized muscles. Immediately after injury, the muscles pulled to failure at approximately 20% of the force needed to cause contralateral nonin- jured muscles to fail. After 1 week of treatment, the decrease in tensile stiffness for immobilized muscles (compared with intact control mus- cles) averaged 33%. In contrast, mobilized muscles had healed to within 11% of the force to failure of control muscles. The mobilized muscles recovered tensile strength more quickly and more completely than the muscles treated with “no specific treatment” (i.e., cage activity ad libitum). After 3 weeks of re- training, these levels had not nor- malized to those of muscles mobi- lized immediately after injury. The authors concluded that early mobi- lization restored functional capacity of healing muscle earlier than im- mobilization. Lehto et al 34 found that immobi- lization after injury accelerated gran- ulation tissue production. However, they also found that if continued too long, it can “lead to contraction of the scar and to poor structural orga- nization of the components of regen- erating muscle and scar tissue.” These conclusions were based on the characteristics revealed by histo- chemical staining, measurement of tensile properties, and the gross ap- pearance of the muscles during heal- ing. The authors concluded that a certain period of immobilization (5 days for rats) is beneficial to allow subsequent mobilization without causing further trauma to the heal- ing tissue. In another study, Järvinen 13 eval- uated four exercise regimens imple- mented after contusion injury in rats, using the local concentrations of leukocytes, erythrocytes, and colla- gen fibers in the injured muscle as a way of measuring the rate of resolu- tion of the contusion. It was found that running immediately after injury is the regimen of choice, because of more rapid disappearance of the injury than with the delayed or no-exercise regimens. Running was also better than swimming. Capillary density after injury has been found to transiently decrease after immobilization of muscle. Similar trends have evolved in the treatment of humans. Jackson and Feagin 4 developed a treatment strategy for West Point cadets who suffered contusion injuries to the quadriceps muscle. They initially emphasized rest of the injured leg in extension and early restoration of full knee extension. With this treat- ment, the trainers and therapists noted that “normal flexion was the slowest parameter to return,” caus- ing prolonged disability. A later study 11 emphasized immobilization in muscle tension (flexion for quad- riceps contusion) for a short period of time (24 hours for mild injuries, 48 hours for severe injuries), fol- lowed by well-leg and gravity- assisted motion as soon as pain re- lief permits. Patients are advanced to functional rehabilitation when 120 degrees of pain-free active knee motion is achieved. These studies have led to the now-common clini- cal practice of immobilization only in the period immediately after injury to limit hematoma formation, followed by early mobilization. Cryotherapy The most common treatment of musculoskeletal injuries is the appli- cation of ice. One group tested the hypothesis that cryotherapy after contusion injury is effective because it reduces microvascular perfusion and subsequent edema formation. 37 The authors found that cryotherapy caused vasoconstriction and de- creased perfusion transiently, but found no long-term microvascular effects. Thus, the therapeutic win- dow of opportunity is relatively small for the effects of cryotherapy. Pharmacologic Treatment Inflammation is thought to be beneficial in attracting reparative cells as a part of muscle healing, allowing clearance of nonviable tis- sues and preventing scar formation. However, it is also thought by some to be the cause of continued pain and swelling that may limit mobility and prevent healing. Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly prescribed by physi- cians dealing with musculoskeletal injury. Once again, animal studies provide some information. Fisher et al 38 studied the effect of systemic inhibition of prostaglandin synthe- sis (by naproxen) on muscle protein balance after contusion injury in the rat. Their findings were similar to those in many of the early studies of muscle contusion injury, in that nor- mal muscle healing for the first 3 days was characterized by a marked catabolic response, followed by muscle protein repletion for several weeks. Inhibition of prostaglandin synthesis significantly (P<0.05) reduced the catabolic loss of muscle protein seen locally and peripheral to the injury site. 38 In another well-designed study, Järvinen et al 39 used their model to test the effects of two different NSAIDs as well as hydrocortisone on the healing of contusion injuries. Histologic, enzyme, and mechani- cal measurements were recorded. Muscle Contusion Injuries Journal of the American Academy of Orthopaedic Surgeons 234 They found that the drugs all sig- nificantly (P<0.05) decreased the acute inflammation, but also caused a slight decrease in tensile proper- ties in the longer term. They noted delayed elimination of hematoma and necrotic tissue and retardation of muscle regeneration in the hy- drocortisone group but not in the NSAID groups. Similar studies have been per- formed with the use of other muscle injury paradigms. In the study by Mishra et al, 40 rabbit muscles were subjected to a repetitive exercise program and treated with flurbipro- fen. The authors reported that the treatment group showed a “more complete functional recovery than the untreated controls at 3 and 7 days but had a deficit in torque and force generation at 28 days.” Nonsteroidal anti-inflammatory drugs have not been studied in rela- tion to healing of muscle contusion injuries in humans. However, the data on NSAIDs in strain injuries are conflicting, and there are no definitive conclusions as to their efficacy or long-term effects on mus- cle regeneration. Corticosteroids are also used by some in the treatment of muscle injuries. Using the contusion injury model, Beiner et al 20 studied the effect of systemic (depot intramus- cular) treatment with a corticoste- roid (methylprednisolone) versus that with an anabolic steroid (nan- drolone). With corticosteroid treat- ment, there was a marked lack of the initial inflammation at the con- tusion site, with increased force- generating capacity in those mus- cles during the early phases. Later, however, the corticosteroid-treated muscles demonstrated a retardation of the normal healing response, with delayed clearing of necrotic tissue and muscle regeneration. Although comparable to the doses used in other animal studies, the doses of corticosteroid were large, and may not simulate accepted doses in human studies. In con- trast, the muscles treated with the anabolic steroid demonstrated a robust initial inflammation but proved to have an increased force- generating capacity in the long run, relative to control muscles. Thus, it appears that corticosteroids may have a beneficial effect in the short term on muscle healing but may be detrimental over the longer term, inhibiting the normal muscle regen- eration cascade in this animal model. Studies in humans have had con- flicting results. In one trial, Levine et al 41 retrospectively reviewed a series of hamstring injuries in Na- tional Football League players and found no adverse effect of injection of corticosteroid directly into the area of hamstring injury. However, these injuries were strain injuries rather than contusions, and no con- trol group was used. Furthermore, the outcome measures were subjec- tive (e.g., pain control, time to re- turn to active status) rather than ob- jective (e.g., isometric strength, time to fatigue). More research is neces- sary to determine whether cortico- steroids have a role in treatment of contusion injuries. Other pharmacologic agents have also been studied in the setting of muscle contusion injury. Using the model of blunt contusion injury developed by Crisco et al, 19 one group studied eight growth factors and their effect on healing. They found that three growth factors— fibroblast growth factor (FGF)-beta, insulinlike growth factor-I, and nerve growth factor—enhanced myoblast proliferation and differen- tiation in vitro and improved the healing of the injured muscle in vivo. 42 Injection of the growth fac- tors also led to enhanced fast-twitch and tetanic strength of the contused muscles 15 days after injury. The study suggested that gene therapy, in the form of myoblast transplanta- tion into injured tissue, might be used to stimulate persistent expres- sion of growth factors capable of promoting the recovery of skeletal muscle after injury. Another group studied FGF-6 and its up-regulation after skeletal muscle injury in mice. 43 Strains of mice lacking the gene for FGF-6 show a severe regeneration defect following injury, with fibrosis and myotube degeneration. They con- cluded that FGF-6 is a “critical com- ponent of the muscle regeneration machinery in mammals, possibly by stimulating or activating satellite cells.” Summary Muscle contusion injuries are com- mon events in the athletic world. Various diagnostic modalities are becoming more commonly used to establish the nature and extent of the lesions. The factors influencing the severity of such injuries are becoming delineated, as are the microstructural events following injury. As more and more clinical re- search is done, several trends in treatment are evolving. Long-term immobilization is to be avoided in favor of a more rapid return to mo- tion and exercise. Nonsteroidal anti-inflammatory drugs, similar to corticosteroids, may have initial beneficial effects, but their long- term effects on muscle healing and regeneration remain to be estab- lished. Other medications, includ- ing growth factors and some ste- roids with anabolic effects, may prove beneficial to the healing process. Animal studies indicate that perhaps surgeons should give more thought to open repair of these muscle injuries, as it appears that, as is the case with nerve tissue, reapproximating the damaged ends may allow the balance between scar formation and tissue regeneration to shift toward a more useful repara- tive process. John M. Beiner, MD, and Peter Jokl, MD Vol 9, No 4, July/August 2001 235 References 1. Canale ST, Cantler ED Jr, Sisk TD, Freeman BL III: A chronicle of injuries of an American intercollegiate football team. Am J Sports Med 1981;9:384-389. 2. Garrett WE Jr: Muscle strain injuries: Clinical and basic aspects. Med Sci Sports Exerc 1990;22:436-443. 3. Holbrook TL, Grazier K, Kelsey JL, Stauffer RN (eds): The Frequency of Occurrence, Impact, and Cost of Selected Musculoskeletal Conditions in the United States. Chicago: American Academy of Orthopaedic Surgeons, 1984. 4. Jackson DW, Feagin JA: Quadriceps contusions in young athletes: Relation of severity of injury to treatment and prognosis. J Bone Joint Surg Am 1973; 55:95-105. 5. Cushner FD, Morwessel RM: Myositis ossificans traumatica. Orthop Rev 1992;21:1319-1326. 6. Garrett WE Jr: Muscle strain injuries. Am J Sports Med 1996;24(6 suppl):S2-S8. 7. Noonan TJ, Garrett WE Jr: Injuries at the myotendinous junction. Clin Sports Med 1992;11:783-806. 8. Fridén J, Lieber RL: Structural and mechanical basis of exercise-induced muscle injury. Med Sci Sports Exerc 1992;24:521-530. 9. Clarkson P: Exercise-induced muscle damage: Animal and human models. Med Sci Sports Exerc 1992;24:510-511. 10. Stone MH: Muscle conditioning and muscle injuries. Med Sci Sports Exerc 1990;22:457-462. 11. Ryan JB, Wheeler JH, Hopkinson WJ, Arciero RA, Kolakowski KR: Quadri- ceps contusions: West Point update. Am J Sports Med 1991;19:299-304. 12. Walton M, Rothwell AG: Reactions of thigh tissues of sheep to blunt trauma. Clin Orthop 1983;176:273-281. 13. Järvinen M: Healing of a crush injury in rat striated muscle: 4. Effect of early mobilization and immobilization on the tensile properties of gastrocnemius muscle. Acta Chir Scand 1976;142:47-56. 14. Lehto M: Collagen and Fibronectin in a Healing Skeletal Muscle Injury: An Experimental Study in Rats Under Variable States of Physical Activity. Turku, Finland: Turun Yliopisto, 1983. 15. Järvinen M, Sorvari T: Healing of a crush injury in rat striated muscle: 1. Description and testing of a new method of inducing a standard injury to the calf muscles. Acta Pathol Microbiol Scand [A] 1975;83:259-265. 16. Järvinen M: Healing of a crush injury in rat striated muscle: 2. A histological study of the effect of early mobiliza- tion and immobilization on the repair processes. Acta Pathol Microbiol Scand [A] 1975;83:269-282. 17. Järvinen MJ, Lehto MUK: The effects of early mobilisation and immobilisa- tion on the healing process following muscle injuries. Sports Med 1993;15: 78-89. 18. Stratton SA, Heckmann R, Francis RS: Therapeutic ultrasound: Its effects on the integrity of a nonpenetrating wound. J Orthop Sports Phys Ther 1984;5:278-281. 19. Crisco JJ, Jokl P, Heinen GT, Connell MD, Panjabi MM: A muscle contusion injury model: Biomechanics, physiolo- gy, and histology. Am J Sports Med 1994;22:702-710. 20. Beiner JM, Jokl P, Cholewicki J, Panjabi MM: The effect of anabolic steroids and corticosteroids on healing of muscle contusion injury. Am J Sports Med 1999;27:2-9. 21. Lieber RL, Fridén J: Muscle damage is not a function of muscle force but active muscle strain. J Appl Physiol 1993;74:520-526. 22. Crisco JJ, Hentel KD, Jackson WO, Goehner K, Jokl P: Maximal contrac- tion lessens impact response in a mus- cle contusion model. J Biomech 1996; 29:1291-1296. 23. Beiner JM: Muscle Contusion Injuries: Impact Biomechanics, Healing, and Treatment Aspects [thesis]. New Haven, Conn: Yale University School of Medi- cine, 1997. 24. Mair SD, Seaber AV, Glisson RR, Garrett WE Jr: The role of fatigue in susceptibility to acute muscle strain in- jury. Am J Sports Med 1996;24:137-143. 25. Noonan TJ, Best TM, Seaber AV, Garrett WE Jr: Thermal effects on skeletal mus- cle tensile behavior. Am J Sports Med 1993;21:517-522. 26. Järvinen M, Aho AJ, Lehto M, Toivonen H: Age dependent repair of muscle rupture: A histological and microangio- graphical study in rats. Acta Orthop Scand 1983;54:64-74. 27. Aspelin P, Ekberg O, Thorsson O, Wilhelmsson M, Westlin N: Ultra- sound examination of soft tissue injury of the lower limb in athletes. Am J Sports Med 1992;20:601-603. 28. De Smet AA: Magnetic resonance findings in skeletal muscle tears. Skeletal Radiol 1993;22:479-484. 29. Swensen SJ, Keller PL, Berquist TH, McLeod RA, Stephens DH: Magnetic resonance imaging of hemorrhage. AJR Am J Roentgenol 1985;145:921-927. 30. Mellerowicz H, Lubasch A, Dulce MC, Dulce K, Wagner S, Wolf KJ: Diag- nosis and follow-up of muscle injuries by means of plain and contrast- enhanced MRI: Experimental and clin- ical studies [German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1997;166:437-445. 31. McCully K, Shellock FG, Bank WJ, Posner JD: The use of nuclear magnet- ic resonance to evaluate muscle injury. Med Sci Sports Exerc 1992;24:537-542. 32. Fisher BD, Baracos VE, Shnitka TK, Mendryk SW, Reid DC: Ultrastruc- tural events following acute muscle trauma. Med Sci Sports Exerc 1990;22: 185-193. 33. Lehto MUK, Järvinen MJ: Muscle injuries, their healing process and treatment. Ann Chir Gynaecol 1991;80: 102-108. 34. Lehto M, Duance VC, Restall D: Collagen and fibronectin in a healing skeletal muscle injury: An immunohis- tological study of the effects of physi- cal activity on the repair of injured gastrocnemius muscle in the rat. J Bone Joint Surg Br 1985;67:820-828. 35. Menetrey J, Kasemkijwattana C, Fu FH, Moreland MS, Huard J: Suturing versus immobilization of a muscle lac- eration: A morphological and func- tional study in a mouse model. Am J Sports Med 1999;27:222-229. 36. Kääriäinen M, Kääriäinen J, Järvinen TL, Sievänen H, Kalimo H, Järvinen M: Correlation between biomechani- cal and structural changes during the regeneration of skeletal muscle after laceration injury. J Orthop Res 1998; 16:197-206. 37. Curl WW, Smith BP, Marr A, Rosen- crance E, Holden M, Smith TL: The effect of contusion and cryotherapy on skeletal muscle microcirculation. J Sports Med Phys Fitness 1997;37: 279-286. 38. Fisher BD, Baracos VE, Reid DC: Effect of systemic inhibition of prostaglandin synthesis on muscle protein balance after trauma in the rat. Can J Physiol Pharmacol 1991;69: 831-836. 39. Järvinen M, Lehto M, Sorvari T, Mikola A: Effect of some anti-inflam- matory agents on the healing of rup- tured muscle: An experimental study in rats. J Sports Traumatol Rel Res 1992; 14:19-28. Muscle Contusion Injuries Journal of the American Academy of Orthopaedic Surgeons 236