A 46-Year-Old Man With Excruciating Shoulder Pain*

A 46-year-old man with no significant medical history presented to his local emergency department complaining of excruciating right shoulder pain. The patient was in his usual state of excellent health until 4 days prior, when right shoulder pain developed while he was using a chainsaw to cut wood. The next day, flu-like symptoms developed with fevers, chills, and headache. An MRI revealed that the right pectoralis major was torn from its attachment to the acromion. His shoulder pain intensified despite treatment with hydrocodone and acetaminophen, and the flu-like symptoms progressed over the next 2 days. Finally, on the day of hospital admission, he was weak and unable to arise out of bed. He was taken by family members to the local emergency department. (CHEST 2005; 127:1039- 1044)

Key words: necrotizing fasciitis; streptococcal gangrenous myositis; streptococcal soft-tissue infections

Abbreviations: CK = creatine kinase; IVIG = IV Ig; NSAID = nonsteroidal antiinflammatory drug

The patient had no significant medical history and was receiving no medications. There were no medication allergies. He smoked one package of cigarettes per day for many years, but discontinued smoking 3 months earlier. The patient drank alcohol occasionally, and did not use IV or recreational drugs. Despite being separated from his wife, he saw his wife and two sons frequently. The patient was employed as a county maintenance employee, and recently had been cutting wood with a chainsaw frequently in boggy, marshy areas. The family history was noncontributory, with the exception that both of his sons had been treated for streptococcal pharyngitis in the prior 2 weeks.

Physical Examination

Physical examination in the local emergency department revealed a toxic-appearing man with a pulse rate of 170 beats/min, BP of 110/ 70 mm Hg, and temperature of 38.0C. The most notable finding was intense pain, with palpation, induration, swelling, and bruising over the right pectoral muscle, extending into the axilla. The WBC count was 6,000/mL, with 14% band forms. Arterial blood gas analysis on 100% fraction of inspired oxygen revealed a pH 7.28; PCO^sub 2^, 28 mm Hg; and PO^sub 2^, 215 mm Hg. The lactate level was 8.2 mmol/ L, and the creatine phosphokinase level was 2,031 U/L. CT scan of the chest revealed swelling and inflammation of the pectoral muscles as well as the soft tissues of the chest wall. No subcutaneous air or gas were noted. The inflammation extended into the right flank and abdomen. There were no pulmonary infiltrates. lie was intubated for airway protection after he became more lethargic and was then transferred to our institution for further management.

On arrival to the ICU, vital signs were as follows: temperature, 38.0C; pulse rate, 156 beats/min; respiration rate, 19 breaths/min with mechanical ventilation; and BP, 116/47 mm Hg. He was pharmacologically sedated. The pupils were equal and reactive. An oral endotracheal tube was in place. There were no oral lesions nor lymphadenopathy of the neck. The chest was clear to auscultation bilaterally. A cardiac examination revealed a rapid but regular tachycardia and no murmur. The skin overlying the right chest was indurated, erythematous, and edematous appearing. There were multiple 1-cm bullae noted as well. No puncture wounds nor skin breaks were observed. There was no crepitus. These findings extended to include the right flank, abdominal wall, and groin. There was no lower-extremity edema, but the skin of the lower extremities had a livedo reticularis pattern. The distal extremities, both hands and feet, were very cool to the touch. Pulses in the distal extremities were nonpalpable. His skin overlying the scrotum was cyanotic.

Laboratory and Radiographic Findings

Laboratory data at our institution (blood drawn immediately on arrival, approximately 6 to 8 h after arrival at the outside emergency department) revealed a WBC count of 5,800/mL with 14% neutrophils, 41% bands, 24% lymphocytes, 6% monocytes, and 15% myelocytes and metamyelocytes. The hemoglobin level was 15.8 g/dL, and the platelets were 212,000/mL. The international normalized ratio was 1.32, fibrinogen was 987 mg/dL, and d-dimer was elevated at 2.34. The sodium level was 135 mmol/L; potassium, 5.1 mmol/L; chloride, 105 mmol/L; bicarbonate, 17 mmol/L; BUN, 60 mg/dL; creatinine, 6.4 mg/dL; and ionized calcium, 0.84 mmol/L. Aspartate transaminase level was 281 U/L; alanine transaminase, 94 U/L, creatinine kinase (CK), 15,117 U/L; lactate, 6.8 mmol/L; and phosphorus, 12.7 mg/dL. A CT scan of the chest, abdomen, and pelvis revealed more extensive soft-tissue edema and stranding as well as soft-tissue swelling of the right chest and abdominal wall involving both muscles and subcutaneous tissue. No focal fluid collections or abnormal gas were identified (Fig 1, 2).

What is the diagnosis?

FIGURE 1. Extensive soft-tissue edema and stranding is noted as well as soft-tissue swelling in the right chest wall. No focal fluid or gas collections are observed.

FIGURE 2. The soft-tissue swelling with edema and stranding extends inferiorly into the right abdominal wall. Again, no focal fluid or gas collections are observed.

Diagnosis: Streptococcal gangrenous myositis

Group A streptococci are common human pathogens. Pharyngitis is the most common presentation of Streptococcal infections. However, streptococci are also common sources of skin and soft-tissue infections. These include erysipelas and cellulitis, as well as life- threatening illnesses such necrotizing fasciitis or gangrenous myositis.

Streptococcal gangrenous myositis is a rare softtissue infection caused by Streptococcus pyogenes. It has an extremely high mortality (in excess of 80%). A literature review from 1985 by Adams et al1 reported only 21 cases from 1900 to 1984. The disease does not appear to have a predilection for either sex, and most cases appear between the ages of 30 years and 60 years.2 Underlying conditions such as diabetes do not appear to predispose patients.2 However, there has been controversy regarding whether the use of nonsteroidal antiinflammatory drugs (NSAIDs) are associated with gangrenous myositis or necrotizing fasciitis.3 Clearly, NSAIDs reduce symptoms, so this may delay diagnosis. However, is it unclear whether there is an interaction between the NSAIDs and these soft-tissue infections or if NSAIDs are used as analgesics by patients when these painful infections are developing.

The clinical manifestations of gangrenous myositis are not unlike the presentation of other severe systemic infections. The hallmark symptom, though, is intense pain and tenderness over the involved muscle. The intensity of the pain often causes suspicion of a torn or ruptured muscle. There is frequently a prodrome of fever, malaise, and myalgias. On examination, there is board-like induration as well as swelling and edema in the affected area. The skin overlying may have petechiae or bullae or may simply be erythematous.4 It is notable that hypotension and renal failure may actually develop prior to the onset of skin changes.

In addition to the often dramatic physical findings, there are also certain laboratory values and radiographic abnormalities that point toward the diagnosis. An elevated serum CK level is a helpful clue, as this suggests muscle damage. An elevated WBC count with a marked bandemia is typically present. Lactic acidosis and rapidly progressive renal failure are also common. Radiographically, swelling and edema of the soft tissue as well as fat stranding are noted on MRI or CT. Gas formation is not seen radiographically or on examination, a finding that would be more consistent with a clostridial infection.

Confirmation of the diagnosis of Streptococcal myositis is twofold. Identifying S pyogenes when it is the offending organism is usually done easily. Blood culture results will typically be positive if drawn before antibiotics are initiated.2 Also, Gram stain of the necrotic tissue and debrided muscle will commonly reveal sheets of Gram-positive cocci in chains. Gangrenous myositis is a pathologic diagnosis, so the observation of necrotic skeletal muscle histopathologically is necessary to verify the diagnosis.

The pathogenesis of streptococcal gangrenous myositis is not clear. It is likely that recent or preexisting muscle injury is needed for the bacteria to infect the muscle. Overuse of muscles, blunt trauma, and skin abrasions all appear to predispose patients to this infection.2,4 The actual infection likely occurs either in the pharynx or skin, with the infecting organisms then reaching the muscle via hematogenous spread or direct extension. Hematogenous spread may be the more common mechanism, given the prevalence of positive blood culture results as well as the lack of an obvious portal of entry through the skin in many cases. There has also been a suggestion of a prodromal viral infection in some patients, suggesting a viral myositis may precede development of a streptococcal infection.2

The differential diagnosis of streptococcal myositis depends on the stage of the infection. Early in the course, cellulitis may be a consideration. Also, a torn or injured muscle is commonly a consideration, given the intensity of the pain. However, the level of discomfort and an elevated CK level, indicating muscle involvement, certainly raises suspicion that a more serio\us process is ongoing. Other infectious entities that involve fever and skin changes, such as toxic shock syndrome, clostridial myonecrosis, and meningococcemia may be included in the differential. A clostridial infection would involve subcutaneous gas, whereas the rashes of meningococcemia and toxic shock syndrome may help distinguish these illnesses. However, as the infection progresses, the primary disorder to differentiate streptococcal myositis from is necrotizing fasciitis. Necrotizing fasciitis is a deep infection of the subcutaneous tissue that leads to destruction of fascia and subcutaneous fat, whereas gangrenous myositis involves destruction of skeletal muscle. Gangrenous myositis often occurs at sites where no break in the skin has occurred. In contrast, fasciitis is typically preceded by major trauma, surgery, or skin injuries such as lacerations or puncture wounds. The local pain and erythema tend to be less severe in necrotizing fasciitis. Finally, the mortality rate of necrotizing fasciitis is approximately 20%, whereas gangrenous myositis has a mortality rate of at least 80%.1,2 These two entities certainly have numerous similarities, and differentiation between the two may be difficult. Recognition of both disorders is crucial since both are surgical emergencies.

Streptococcal gangrenous myositis is clearly a surgical emergency. This disease was associated with 100% mortality until Doebbeling and Wenzel5 described a case of patient survival following early surgical debridement and subsequent re-exploration at 36 h. Studies involving necrotizing fasciitis have shown early aggressive surgical debridement to be associated with improved outcome. For instance, Freischlag et al6 concluded that delayed intervention of > 24 h was the primary cause of death resulting in a 70% mortality rate, compared to a 36% mortality rate when debridement was performed < 24 h after recognition of the disease. Sudarsky et al7 found decreased survival and increased risk of amputation in patients in whom debridement was delayed for > 12 h. Mandatory re-exploration was then performed at 24 h with the average number of debridements being 3.2 1.1 ( SD). Havwood et al8 described eight patients with combined necrotizing myositis and fasciitis. A mean of three debridements per patient was performed, with a resulting mortality of 37.5%. Once debridement is complete, options for skin coverage include secondary closure, split thickness skin grafts, or muscle flaps.

Selection of the appropriate antibiotics in treating streptococcal myositis is also crucial. For a long time, penicillin was thought to be the antibiotic of choice. However, monotherapy with penicillin seemed to be associated with extremely high rates of morbidity and mortality. An animal model study published in 1988 by Stevens et al9 showed that when treating streptococcal myositis in mice, erythromycin and clindamycin had better efficacy than penicillin. It is unclear precisely why penicillin demonstrated reduced efficacy. One possibility is that an inoculum effect exists, whereby there is a reduction in efficacy when a larger inoculum of organisms is present. Further evidence for the reduced efficacy of penicillin resulted from a retrospective study by Zimbelman et al10 in 1999. This study compared treatment outcomes of invasive streptococcal infections when different antibiotic regimens were utilized. These investigators10 found that treating a deep streptococcal infection with an antibiotic regimen that included a protein synthesis inhibiting antibiotic (typically clindamycin), instead of monotherapy with a cell wall-inhibiting antibiotic (such as a β-lactam), had a significant impact on outcome. Clindamycin appears to be superior because the efficacy of the drug is not impacted by the size of the bacterial inoculum. Furthermore, clindamycin suppresses toxin synthesis and facilitates phagocytosis of S pyogenes by inhibiting production of the M-protein of the organism. Hence, the regimen of choice for streptococcal myositis definitely should include clindamycin. A regimen of penicillin G with clindamycin is commonly used currently.

Other potential therapies include hyperbaric oxygen and IV Ig (IVIG). Hyperbaric oxygen was considered helpful in a case of necrotizing fasciitis11 but at present does not have a clear role in therapy for streptococcal myositis. IVIG, however, may be useful. There have been case reports of patients with streptococcal soft- tissue infections, including myositis, who have shown significant improvement after therapy with IVIG.12 Furthermore, an observational study by Kaul et al13 found promising results: 30-day survival was significantly higher in streptococcal toxic shock syndrome patients treated with IVIG than those whose therapy did not include IVIG. The enhanced survival in the IVIG group appears related to the ability of IVIG to inhibit T-cell stimulation and subsequent cytokine release. Hence, IVIG is an option for treatment in streptococcal myositis

Hospital Course

This patient received a clinical diagnosis of streptococcal gangrenous myositis. He had the history of a torn pectoral muscle, as well as recent streptococcal infections in his two sons. His physical examination, laboratory data (specifically the impressive bandemia, lactic acid, and CK levels) and radiographic findings all supported the diagnosis. The plastic surgery team performed a radical debridement of the right lateral chest wall, axilla, right abdominal wall, and right groin. Frankly necrotic appearing muscle was found. The area of debridement extended from the sternum to the midline of the back, and from the axilla to the groin. Cultures of the necrotic tissue were sent for microbiologic analysis, which rapidly grew S pyogenes. Blood culture specimens drawn at the referring institution grew S pyogenes in the aerobic bottle from two separate sites. Pathologic analysis of the debrided tissue revealed necrotic skeletal muscle. The patient was treated with penicillin, clindamycin, and 5 days of IVIG. He remains hospitalized at this time (5 months after his initial hospital admission) on the general medical floor. His hospital course has included a stay of > 3 months in the ICU. He required a tracheostomy for chronic respiratory failure, which has now resolved. The acute renal failure, which required hemodialysis for > 1 month, has now resolved. Furthermore, numerous surgeries have been performed, initially for debridement and later for skin grafting.

REFERENCES

1 Adams EM, Gudmundsson S, Yocum DE, et al. Streptococcal myositis. Arch Intern Med 1985; 145:1020-1023

2 Yoder EL, Mendez J, Khatib R. Spontaneous gangrenons myositis induced by Streptococcus pyogenes: case report and review of the literature. Rev Infect Dis 1987; 9:382-385

3 Bisno AL, Stevens DL, Streptococcal infections of skin and soft tissues. N Engl J Med 1996; 334:240-245

4 Jahnson L, Berggren L, Bjorsell-Ostling E, et al. Streptococcal myositis. Scand J Infect Dis 1992; 24:661-665

5 Doebbeling BN, Wenzel RP. Spontaneous Streptococcal gangrenous myositis: survival with early debridement. South Med J 1989; 82:900- 902

6 Freishlag JA, Ajalat GA, Busuttil RW. Treatment of necrotizing soft tissue infections. Am J Surg 1985; 149:751-755

7 Sudarsky LA, Laschinger JC, Coppa GF, et al. Improved results from a standardized approach in treating patients with necrotizing fasciitis. Ann Surg 1987; 206:661-665

8 Haywood CT, McGeer A, Low DE. Clinical experience with 20 cases of group A Streptococcal necrotizing fasciitis and myonecrosis: 1995- 1997. Plastic Reconstr Surg 1999; 103:1567-1573

9 Stevens DL, Gibbons AE, Bergstrom R, et al. The Eagle effect revisited: efficacy of clindamycin, erythromycin and penicillin in the treatment of streptococcal myositis. J Infect Dis 1988; 158:23- 28

10 Zimbelman J, Palmer A, Todd J. Improved outcome of clindamycin compared with β-lactam antibiotic treatment for invasive Streptococcus pyogenes infection. Pediatr Infect Dis J 1999; 18:1096- 1100

11 Riseman JA, Zamboni WA, Curtis A, et al. Hyperbaric oxygen therapy for necrotizing fasciitis reduces mortality and the need for debridements. Surgery 1990; 108:847-850

12 Lamothe F, D’Amico P, Ghosn P, et al. Clinical usefulness of intravenous human immunoglobulins in invasive group A streptococcal infections: case report and review. Clin Infect Dis 1995; 21:1469- 1470

13 Kaul R, McGeer A, Norrby-Teglund A, et al. Intravenous immunoglobulin therapy for streptococcal toxic shock syndrome: a comparative observational study. Clin Infect Dis 1999; 28:800-807

Eric J. Olafsson, MD; Tallal Zeni, MD, and David S. Wilkes, MD, FCCP

* From the Division of Pulmonary and Critical Care Medicine (Drs. Olafsson and Wilkes) and the Department of Surgery (Dr. Zeni), Indiana University School of Medicine, Indianapolis, IN. Manuscript received April 26, 2004; revision accepted September 7, 2004.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@chestnet.org).

Correspondence to: David S. Wilkes, MD, FCCP, Division of Pulmonary Medicine and Critical Care Medicine, Indiana University School of Medicine, Van Nuys Medical Sciences Building MS224, 635 Barnhill Dr, Indianapolis, IN 46202-5120: e-mail: dwilkes@iupui.edu

Copyright American College of Chest Physicians Mar 2005