For any practicing emergency physician, this scenario probably occurs several times a week: You admit an alcoholic patient with what looks like a cellulitis of a lower extremity. The patient has pain, erythema, swelling, and maybe a fever. Even the most experienced physician, however, would be surprised to discover the patient ended up in the ICU after a trip to the operating room and required pressor support. This “cellulitis” was obviously more than that; it was likely necrotizing fasciitis.
Necrotizing fasciitis is a deadly soft tissue infection that often looks like an average cellulitis, but can quickly turn fatal without a high level of suspicion and proper treatment. First described by Hippocrates in the 5th Century, B.C., and called many things through the years (putrid ulcer, phagedena, and hospital gangrene), the term necrotizing fasciitis was coined in 1952 by Wilson (Am Surg 1952;18:416).
More recently, the lay press has used the term “flesh-eating bacteria.” Classically, it is a soft tissue infection due to group A streptococci, but is more often polymicrobial and sometimes due to gram negative organisms alone. In 1999, the Centers for Disease Control and Prevention reported a total of 10,200 cases of invasive group A streptococcal illness with 1,200 deaths. Necrotizing fasciitis accounted for six percent of the total cases (MMWR April 6, 2001;48). Mortality rates, which have remained unchanged since the 1920s, are approximately 30 percent (Ann Surg 1995;221:558).
In 1999, necrotizing fasciitis accounted for 6% of cases of invasive group A streptococcal illness
Necrotizing fasciitis is an infection that develops at the superficial fascial layer, usually from a break in the skin. For example, an insect or animal bite, surgical incision, site of injection drug use or insulin injection, or even a superficial scratch. It most commonly occurs in the abdominal wall, extremities, and perineum, although it can occur anywhere on the body, including the head and neck. It also has been thought to occur from hematogenous spread, for example, from an antecedent sore throat, although this is rare (Am Surg 1994;60:490).
In one large review, 17 percent of patients with necrotizing fasciitis had no antecedent trauma (Am Surg 2002;68:109). In men, perineal necrotizing fasciitis is called Fournier's gangrene. Risk factors for developing the disease include advanced age, peripheral vascular disease, diabetes, alcoholism, and renal and liver disease. Children rarely develop it; it occurs only in immunosuppressed patients, and even more rarely with fetal scalp monitors or circumcision. It also has been observed that NSAID use in streptococcal disease leads to suppression of early signs and symptoms, which results in a delayed diagnosis and worse outcome (New Engl J Med 1996;334:240).
The microbiology of this disease is most often polymicrobial (70%) with single organism aerobic (10%) and anaerobic (20%) infections occurring more rarely. The usual aerobic organisms are group A streptococci, Staphylococcus aureus, and E. coli, and the usual anaerobes consist of Peptostreptococcus, Bacteroides, and Clostridium. Unusual organisms include Vibrio vulnificus and cholerae which are thought to cause this disease by entering through a wound caused by a marine animal or exposure to sea water. The tissue damage in group A streptococcal disease is thought to be due to the release of two exotoxins, which has treatment implications.
The usual sequence in the development of necrotizing fasciitis is as follows: approximately five to seven days after the inciting event, the affected area becomes red, hot, and swollen. Fever comes earlier than with a cellulitis, and the area initially is very painful. The skin then becomes tense, swollen, and shiny, and soon afterwards develops bullae. Because this is a superficial fasciitis, the muscle and deep fascial layers are often spared. It also usually involves the subcutaneous fat, which contains the neurovascular structures.
The loss of the neurovascular structures is a result of thrombosis of the dermal blood supply, leading to gangrene and anesthetic skin. Necrosis of this fatty layer produces a thin, watery foul-smelling liquid known as dishwater pus. This disease is often confused with cellulitis, especially early on, but there are some diagnostic clues to help differentiate between the two. In necrotizing fasciitis, the subcutaneous tissues have a hard, wooden feel, the superficial skin will often have bullae, and there will not be lymphangitis.
Another diagnostic clue is crepitus, although this is only seen approximately 30 percent of the time. Direct exploration, either with a digit or blunt probe will find the thin, brown dishwater pus, rather than the usual thick yellow-green fluid. Hypotension, acute renal failure, and ARDS also are seen more frequently than with cellulitis. With group A streptococcal disease, the bullae are often dark red or ecchymotic, and these areas will later turn gangrenous.
Radiology, Tissue Sampling
Diagnosis is initially based on clinical suspicion, but both radiology and tissue sampling can help confirm. Making the diagnosis based on physical findings alone is difficult early in the disease, when appropriate disposition and treatment can really make a difference. Certainly the septic patient with an area of skin that is necrotic and gangrenous is an easy diagnosis, but how does one differentiate between a run-of-the-mill cellulitis and life-threatening necrotizing fasciitis? Radiology, using plain films, CT, ultrasound, or MRI, has been shown useful. Plain radiographs, looking for subcutaneous air, is more reliable than palpation, and is easily obtained.
Although probably not easily obtained from the emergency department, biopsy with frozen sections has been shown to be very accurate, and will lead to earlier diagnosis and lower mortality than those patients diagnosed on clinical grounds alone (New Engl J Med 1984;310:1689).
Treatment is truly surgical, with debridement, fasciotomy, and drainage. Multiple studies show early surgical intervention led to better survival rates than delayed intervention (Br J Surg 1993;80:1190; Am J Surg 1983;145:784). Even after initial exploration, the patient may need several trips to the operating room until all affected fascial areas are explored. In Fournier's gangrene, urinary diversion is often required, as is a colostomy in perineal necrotizing fasciitis.
Hyperbaric oxygen therapy has been studied in the treatment of necrotizing fasciitis, and remains controversial. It has certainly been shown helpful in clostridial gas gangrene (Clev Clin J Med 1992;59:517), and is probably helpful in necrotizing fasciitis. Interestingly, therapy using maggots has been reported to be helpful in a case of head and neck disease (J Laryngol Otol 2002;116:70). Antibiotics, along with IV fluids and vasopressor support (if needed) should be instituted in the emergency department after blood and, if possible, wound cultures.
Antibiotic choices should be broad-spectrum, with penicillin g, clindamycin, and gentamycin a reasonable trio with ceftriaxone and clindamycin other good choices. Clindamycin has been shown to inactivate the toxins produced by group A streptococcal infections (J Clin Invest 1981;67:1249). After cultures and gram stain results have returned, antibiotic choices can be tailored to the organism found.
Subscriptions to EMN
If you are an emergency physician, you should already be receiving a subscription to EMN. If not, qualify for a complimentary monthly subscription by contacting Deborah Holmes at email@example.com or (212)886-1209 (fax).
If you are not an emergency physician and are interested in subscribing, contact Lippincott Williams & Wilkins, 16522 Hunters Green Parkway, Hagerstown, MD 21740; (800)638-3030; in Maryland, call collect, (301)824-7300.