Case Presentation

Necrotizing Fasciitis:

Resident Grand Rounds

December 1, 1998

HPI: B.T. is a 48 year old white female with multiple medical problems (see PMH) who had been hospitalized two weeks previously for cardiac catheterization with uncomplicated angioplasty of her LAD. She presented to the Emergency Department at Baptist Hospital on 8/4/97 complaining of a "painful boil on her bottom". She stated that the pain started three days earlier in the region of her right groin catheterization site and progressively worsened to the point of becoming unbearable. She subjectively reports mild fever and chills; but denies nausea, vomiting, diarrhea, dysuria, shortness of breath, chest pain, or vaginal discharge.

PMH:

1. End stage renal disease- on peritoneal dialysis
2. Diabetes mellitus
3. Hypertension
4. Coronary artery disease
5. Hyperlipidemia

Allergies: "cholesterol medication" (causes nausea and vomiting)

Medications:

Prilosec 20mg qd, Nu-Iron 150 mg bid, Atenolol 50mg qd, Folate 1mg qd, Imdur 30mg qd, Aspirin 325 mg qd, Insulin NPH 18 units q am, Nitroglycerin 0.4mg SL prn

Vital Signs: T 96.6, P 56, BP 86/42, R 20

General: obese white female in no acute distress

CV/Lungs: within normal limits

Abd: soft, obese, slightly tender in right lower quadrant, normal bowel sounds. A large hematoma is present in the right groin with surrounding erythema. No purulence was noted; but on deep palpation of the site, soft tissue crepitus was found.

GU: significant for a grossly swollen right labium, which also demonstrated crepitus on palpation.

Rectal: normal sphincter tone, heme positive stool.

WBC 28.3 with 94% segs, 2% bands, 2% lymphs, Hgb 10.3, Hct 29.2, platelets 324.

Na 134, K 3.3, Chloride 89, Bicarb 19, BUN 35, Cr 9.3, Glucose 383

ED Course: The patient was started on one gram IV Rocephin and KVO normal saline. The medicine service was consulted for possible infected Bartholin’s cyst.

The medicine consult recommended a plain film of the pelvis. Plain film revealed a large amount of air within the soft tissues extending from the groin into the labium. Given her crepitation on exam, increased white blood cell count, and diabetes, necrotizing fasciitis was high on the differential. A General Surgery consult was obtained. Surgery incised an area of skin over the groin hematoma, revealing foul-smelling, nonviable subcutaneous tissue.

Necrotizing fasciitis is a rare and life-threatening soft-tissue infection that is usually caused by toxin-producing bacteria. Its clinical course is characterized by widespread fascial necrosis with relative sparing of skin and underlying muscle. The infection can be associated with severe systemic toxicity and may rapidly progress to death unless promptly recognized and treated.

A large spectrum of skin and soft-tissue infections exist, which can be categorized anatomically. The superficial pyodermas do not extend beyond the skin (epidermis and dermis) and include erysipelas, impetigo, folliculitis, ecthyma, furunculosis, and carbunculosis. Cellulitis is a deeper skin infection than erysipelas, but does not extend to the deep fascia. Necrotizing fasciitis involves the superficial fascia, subcutaneous fat, nerves, arteries, veins, and the deep fascia. Myonecrosis results in rapid necrosis of muscle, with delayed involvement of overlying soft tissue and skin. (1)

The earliest report of necrotizing fasciitis dates back to Hippocrates’ description of a complication of erysipelas in the 5^th century B.C. "...the erysipelas would quickly spread widely in all directions. Flesh, sinews and bones fell away in large quantities...Fever was sometimes present and sometimes absent...There were many deaths. The course of the disease was the same to whatever part of the body it spread." (2) It was first described in the U.S. in 1871 by a Civil War surgeon who described cases of hospital gangrene. The term necrotizing fasciitis was first used in 1952 by Wilson, and describes the most consistent feature of the infection, fascial necrosis. (1)

Necrotizing fasciitis is rare. The Center for Disease Control estimates 500-1,500 new cases are reported in the United States each year. There has been recent speculation that the number of cases is increasing; however, there is little evidence to support that assertion. The actual incidence of necrotizing fasciitis is complicated by the use of multiple terms to describe the same clinical entity (acute non-clostridial crepitant cellulitis, non-clostridial gas gangrene, synergistic necrotizing cellulitis, necrotizing cellulitis, bacterial synergistic gangrene, gangrenous or necrotizing erysipelas, hemolytic streptococcal gangrene, and Fournier’s gangrene). Some reports try to distinguish between these different terms based on bacteriology, gross and microscopic findings, clinical presentation, and/or anatomic location. It may be more useful to view these different conditions as part of a clinical spectrum that share common traits and have a similar clinical course. There does not appear to be an age or sex predilection for the disease. (3)

In necrotizing fasciitis, approximately 10% of cases are due to aerobes, 20% due to anaerobes, and 70% of cases due to mixed flora. A single organism is isolated in <10% of cases. Usual aerobic organisms include Group A Streptococcus, Staphylococcus aureus, E. coli, and Enterobacteriaceae. Common anaerobic organisms include Peptostreptococcus, Prevotella, Bacteroides, Clostridium, Fusobacterium, and Porphyromonas. Unusual pathogens which have been documented include non-group A streptococcus, H. influenzae, Pseudomonas, Vibrio vulnificus, Mucor, and Rhizopus. (4)

Some authors have classified necrotizing fasciitis into two types:

· Type 1: Mixed anaerobic, aerobic, and facultative bacteria (ex. E. coli and Bacteroides)

· Type 2: Group A Streptococcus only (or occasionally GAS with Staphylococcus)

Type 1 accounts for approximately 90% of all cases, while Type 2 usually occurs 10% of the time.

Overall, streptococcus is the most common causative organism. (5) Most cases of necrotizing fasciitis caused by Group A Streptococcus are M-type 1,3,12, or 28 and produce exotoxin A or B. The M protein impedes phagocytosis and, along with the exotoxins, act as superantigens, which leads to liberation of cytokines and a toxic shock-like syndrome.

Known risk factors for necrotizing fasciitis that have been reported are:

· Diabetes Mellitus

· Peripheral Vascular Disease

· IV drug use

· Alcoholism

· Immunosupressed patients

· ?Old age/obesity/malnutrition

Necrotizing fasciitis can affect any part of the body. Eighty percent of cases are caused by bacteria that extend from contaminated disruptions of the skin or localized skin infections. Common etiologic events along with their sites are listed below.

· General- skin disruption (cut, abrasion, burn, laceration, contusion, bite, injection, or surgical incision), blunt/penetrating trauma, post-op complications, cutaneous infections/ulcers, or illicit/SQ drug injections.

· Abdominal Wall- post-op complication of abdominal surgery.

· Extremity- trauma, illicit drug use, insect bite, scratch, or wound.

· Perineum- post-op complications, pilonidal abscess, neglected ischiorectal/perineal abscess.

· Vulva- Bartholin’s gland duct abscess, vulvar abscess, pudendal nerve block; post-op wound infection from C-section, episiotomy, hysterectomy, etc.

· Fournier’s gangrene (necrotizing fasciitis of the male genital organs)- GU infections (UTI superimposed on urethral stricture), traumatic instrumentation, urethral   calculus, neoplasm, surgery, coital injury

· Head and Neck- Scalp/Periorbital- trauma, eyelid infection/pruritus

· Face/Neck- progressive dental infections, peritonsillar abscess, salivary gland infections, cervical adenitis, otologic sources

· Complication of percutaneous catheter placement- chest tube, PEG, and percutaneous drain of abdominal abscess

Approximately twenty percent of cases occur without an obvious portal of entry, and are referred to as idiopathic necrotzing fasciitis. Idiopathic necrotizing fasciitis is more likely to occur in patients with few comorbid medical problems, is typically caused by a single organism (Streptococcus pyogenes), and commonly involves the lower extremities. This condition is thought to be the result of infection from unrecognized breaks in the skin or from hematogenous spread. (1)

Patients initially present with nonspecific symptoms such as pain and high fever. Local signs can include edema and erythema. The presence of severe pain out of proportion to local findings and systemic toxicity should suggest the possibility of necrotizing fasciitis. Necrotizing fasciitis is often acute, but can follow a subacute progressive course.

Initial signs are an erythematous, exquisitely tender, swollen, warm area of cellulitis. As the infection proceeds, the skin often becomes smooth, tensely swollen, and shiny. Distinct margins and induration are notably absent in necrotizing fasciitis; rather the affected area gradually infiltrates into normal skin. As the disease process progresses, the skin darkens to a dusky blue, as early as thirty-six hours after onset. Within three to five days, blisters and bullae develop which may eventually become hemorrhagic. By this time, the involved areas are no longer painful, having become anesthetized secondary to destruction of superficial nerves and thrombosis of small blood vessels. Fascial necrosis is present with widespread undermining of the skin. Involved fascia and subcutaneous fat are edematous, dull grey, and necrotic with serosanguineous exudate.

Patients usually appear toxic, with tachycardia, hypotension, and tachypnea. Subcutaneous crepitation may occur. Leukocytosis with a leftward shift can also be seen. Other abnormal laboratory values that have been seen include acidosis, anemia, impaired renal function, DIC, and hypocalcemia. (5)

The precise mechanism resulting in the fascial necrosis is not known. Bacterial enzymes, including lipases and hyaluronidase, which degrade fat and fascia, are thought to be a cause. Normal appearing skin is undermined and the extent of fascial necrosis is more widespread than changes in the overlying skin. Thrombosis of the skin’s arteries can occur, resulting in focal areas of necrosis. The subcutaneous nerves are destroyed by the infectious process. As organisms and toxins are delivered into the bloodstream, signs of sepsis syndrome occur. (1)

Due to its potentially fatal course, prompt diagnosis of necrotizing fasciitis is the key to its treatment. In fact, one of the most important prognostic indicators is the time to diagnosis. The diagnosis of necrotizing fasciitis can be very difficult to distinguish from acute cellulitis. A study by Lille et al. suggests that a delay in the diagnosis of necrotizing fasciitis by a matter of hours significantly increases mortality. A ten-year retrospective analysis of twenty-nine patients with confirmed necrotizing soft tissue infections (by intraoperative findings) was performed. Patients were categorized into either an early group (surgery <24 hours of presentation; n=17) or a delayed group (surgery> 24 hours from presentation; n=12), based on time from presentation to initial operative debridement. There were no significant differences between these groups based on age, gender, location of infection, or premorbid medical conditions. Mortality in the early group was 6% versus 25% in the delayed group. Advanced clinical signs (shock, skin necrosis, crepitus, bullae, and cutaneous numbness) were present in 76% of early group versus 25% of the delayed group. Fine needle aspiration was positive (for pus or bacteria by Gram stain) in 80% of the early group versus 25% in the delayed group. Radiographic studies identified soft tissue gas in 58% of the early group versus 17% of the delayed group. 71% of the twenty-four patients who were admitted to a surgical service underwent early surgery. Five patients were admitted to a nonsurgical service and all had delayed surgery.

This study suggests prompt diagnosis followed by appropriate treatment decreases mortality rates in patients with necrotizing fasciitis. Factors contributing to diagnostic delay included a lack of clinical symptoms and signs, negative FNA and radiographic changes, and admission to a nonsurgical service. (6) For these reasons, it is important for clinicians to be aware of the clinical features suggesting deep soft tissue infection and the diagnostic modalities that are useful in establishing this diagnosis.

What is the best diagnostic modality available to establish the diagnosis of necrotizing fasciitis?

Clinical signs (crepitus, changes in skin appearance, etc.), fine needle aspiration, frozen section biopsy, and direct inspection of the fascia are tools that have been used to make the diagnosis of necrotizing fasciitis. Imaging studies are often used to provide further information regarding the type and extent of infection. Radiographic modalities employed in the diagnosis of necrotizing fasciitis include plain film, computed tomography scan, and magnetic resonance imaging.

A review of the literature was performed on the following diagnostic modalities:

· frozen tissue biopsy

· plain film

· computed tomography

· magnetic resonance imaging

The majority of the literature on necrotizing fasciitis is case reports and retrospective reviews. It is difficult to perform prospective, randomized, controlled trials because of the rarity of necrotizing fasciitis and the difficulty in suspecting and recognizing this disease. For these reasons, most of the studies are retrospective, quite small, and do not provide an independent, blind comparison with a gold standard diagnostic method. It is possible that uncontrolled clinical factors influenced the decision as to when or if the test under study was performed. Also, the outcome of the diagnostic test being studied sometimes influenced the decision to perform an intervention or obtain the gold standard test. These factors limit the validity of the results and should be considered when drawing conclusions.

Stamenkovic and Lew studied the use of frozen-section biopsy for diagnosis of the disease in its early stages. Retrospective review over a thirteen year period of nineteen cases of histologically proven (necrosis of superficial fascia, polymorphonuclear infiltration of deep dermis and fascia, fibrinous thrombi of arteries and veins passing through fascia, angiitis with fibrinoid necrosis of arterial and venous walls, presence of microorganisms on Gram’s stain, absence of muscle involvement) necrotizing fasciitis showed a decreased mortality rate (12.5%) in the frozen-section biopsy group compared to the mortality rate (72.7%) of the group of patients who did not undergo frozen-section biopsy, but had the diagnosis made later on clinical grounds or surgical resection. The main difference between the two groups was the time between the onset of symptoms and treatment with surgical excision. In the frozen-section biopsy group, the average time between symptoms and treatment was 21 hours; in the non-biopsied group, six days. The biopsies excluded the diagnosis in six cases, avoiding unnecessary surgical exploration. This study did show that performing frozen-section soft-tissue biopsy early on a suspicious lesion can provide a definitive diagnosis and decrease mortality. (7) This study had several drawbacks. The study size was small and retrospective. Frozen- section tissue biopsy is an invasive procedure that requires immediate histopathologic evaluation. When evaluated by inexperienced hands, the extent of infection may be underestimated.

Majeski and Majeski also retrospectively studied the use of tissue biopsy in diagnosing necrotizing fasciitis. Forty-three patients were evaluated for possible necrotizing infections with frozen section biopsy. All patients had a similar variety of clinical manifestations. Twelve out of forty-three patients had a frozen section diagnosis of necrotizing fasciitis. Twenty had cellulitis and eleven had abscess without advancing necrotizing infection. All the patients with necrotizing fasciitis survived. The average time between initial evaluation and surgical debridement was four hours. This study showed that frozen section biopsy provides a reliable diagnosis of necrotizing fasciitis.(8) Drawbacks of this study were similar to the previous study, as well as not having a control group with which results could be compared.

Fisher et al. examined the role of roentgenographic studies for soft tissue gas in diagnosing necrotizing fasciitis. They performed a retrospective review over a nine-year period of twenty-six cases of necrotizing fasciitis. The diagnosis had to satisfy six criteria: extensive necrosis of superficial fascia with widespread undermining of surrounding tissue, moderate to severe systemic toxic reaction, absence of muscle involvement, failure to demonstrate clostridia in wound/blood cultures, absence of major vascular occlusion; and pathological exam of debrided tissue showing leukocytic infiltration, focal necrosis of fascia and surrounding tissues, and thrombosis of microvasculature. X-rays were initially requested to exclude osteomyelitis. Soft-tissue gas was fortuitously found and subsequent plain films were requested to detect the presence and extent of soft-tissue gas. Gas was found in the soft tissues of nineteen patients (73%), seventeen of whom were diabetic. Crepitation was found by physical exam in only five patients (26%). With plain film exposed for maximum soft-tissue visualization of the involved site, gas was detected in every patient who had soft-tissue gas found at surgery (100%). Gas was seen as bubbles mixed with thin, foul-smelling discharge that arose from necrotic gray fascia and fat. Mortality in diabetics was 4/21 (19%); similar to nondiabetic mortality, 1/5 (20%).

This study showed that the physical exam finding of crepitation was not a reliable tool to detect soft-tissue gas as opposed to roentgenographic studies, which detected gas reliably and simply. (9) Some of the drawbacks to this study were the small study size. No discussion was made about the seven patients with necrotizing fasciitis in whom soft-tissue gas was not found. Also, plain films do not usually show abnormalities until a necrotizing process is well-advanced. X-rays are specific, but will miss those patients who have not yet developed soft-tissue gas. (A normal x-ray does not rule out necrotizing fasciitis.) Also, there is no data on the timing of symptoms, when the diagnosis was made, and how this affected mortality.

Wysoki et al. studied computed tomography (CT) characteristics of necrotizing fasciitis. Twenty preop CT scans of patients with pathologically proven necrotizing fasciitis over a four year period were retrospectively reviewed. The scans were reviewed for fascial thickening, fat infiltration, focal fluid collections, soft-tissue gas, muscle involvement, and intra-abdominal extension. Findings were correlated with clinical factors and outcome. Studies were interpreted by two readers on a consensus basis. Asymmetric fascial thickening with fat stranding was seen in 16 out of 20 patients (80%). Gas tracking along fascial planes was seen in 11 out of 20 patients (55%). Focal fluid collections were seen in seven patients (35%). The study concluded that asymmetric fascial thickening and gas on CT scan are valuable in the assessment of suspected necrotizing fasciitis. CT scans can also delineate the true extent of disease, including deep abscesses and thereby aid in treatment. Patients with necrotizing fasciitis who do not have soft-tissue gas create more of a diagnostic challenge as fascial thickening and stranding are more subtle on these CT images. (10) Drawbacks to this study include small study size. The fact that more subtle changes might be missed on CT scans was not mentioned. The authors also did not discuss the 20% of patients with necrotizing fasciitis who did not have CT changes.

Schmid et al. evaluated the diagnostic value of MRI in differentiating necrotizing fasciitis from cellulitis. MRI exams performed over four years on 17 patients with clinically suspected necrotizing fasciitis were retrospectively reviewed. Final diagnoses were necrotizing fasciitis in eleven out of seventeen patients (proven surgically or at autopsy) and cellulitis in six patients. Patients underwent MRI within 2-7 days after the onset of symptoms but less than 24 hours before surgery when necessary. MRIs were analyzed by two radiologists unaware of surgical/clinical findings. Cellulitis was diagnosed by thickening of subcutaneous tissues and contrast enhancement (with or without fluid collections) within subcutaneous tissues and superficial fascia and deeper structures were normal. Necrotizing fasciitis was diagnosed in all cases where deep fascia was involved (high signal intensity on T2-weighted images and contrast enhanced T1-weighted images). Results of the MRI were then compared with the surgical findings/clinical outcomes. Eleven cases of necrotizing fasciitis and five out of six cases of cellulitis were correctly identified. A single case of cellulitis was overstaged as necrotizing fasciitis yielding 100% sensitivity, 86% specificity. The high sensitivity is mainly secondary to the ability of MRI to detect fluid collections along deep fascial sheaths. When MRI images were compared with surgical findings, the extent of infection was usually overestimated in MRI. This was thought to be secondary to noninfectious edema of neighboring infected, necrotic fascia. MRI has high sensitivity in the diagnosis of necrotizing fasciitis, with its characteristic findings of thickening and fluid collections along deep fascial sheaths. However, it is limited secondary to its suboptimal specificity. Several other disorders (myositis, rheumatic diseases, lymphedema, or neoplastic disease) may have an increased water content of the soft-tissue; these entities should be distinguished from necrotizing fasciitis by their specific history and clinical findings. (11)

Top: cellulitis of the left forearm

Bottom: necrotizing fasciitis of the left leg

Loh et al. studied the specificity of hyperintense T2-weighted signal abnormalities within deep fascial planes by MRI for necrotizing soft-tissue infections. Retrospective review was performed on twenty-two MRIs selected for the presence of hyperintense T2-weighted signal within deep fascial planes without knowledge of clinical or pathologic diagnosis. Two reviewers evaluated the films in consensus for abnormalities in superficial and deep soft tissues. They determined the presence or absence of a necrotizing soft-tissue infectious process using T2-weighted deep fascia signal abnormalities. Deep fascial hyperintense T2-weighted signal was present in all twenty-two patients. All the cases were interpreted as necrotizing soft-tissue infection, including the twenty-one cases who had nonnecrotizing conditions (abscess, nonnecrotizing cellulitis, nonnecrotizing cellulitis with lymphatic obstruction, cellulitis with vascular thrombosis, recent radiation therapy, cellulitis with associated osteomyelitis, soft-tissue inflammation associated with metastatic disease, pyarthrosis, recent surgery, ruptured popliteal cyst, noninfected passive edema). The differentiation of necrotizing fasciitis from other abnormal soft-tissue conditions based on the presence of hyperintense T2-weighted signal in deep fascia is unreliable. Further investigation is needed to help define more accurate MRI criteria for diagnosing this condition. Several conditions can cause edema in the deep fascia layers. The clinical scenario should be correlated with the MRI findings to help establish the diagnosis. (12)

Rahmouni et al. studied the value of MRI for the diagnosis of severe necrotizing cellulitis. Thirty-six patients, over a three year period, referred for acute infectious cellulitis underwent MRI 1-13 days after the onset of symptoms. When an abscess was suspected on T2-weighted images, contrast enhanced T1-weighted images were also obtained. MRIs were analyzed by two radiologists who were blinded to clinical data. Two dermatologists assessed the clinical/biologic data and divided patients into two groups (Group 1- nonsurgical cellulitis (n=23), Group 2- signs of surgical cellulitis and necrotizing soft-tissue infections (n=13)). In Group 1, signal intensity abnormality was seen within deep fascia/muscles in three patients. These three underwent surgery and had confirmed necrosis by surgical or pathologic findings. All patients in Group 2 underwent surgery. Thirteen MRIs performed in eleven patients showed linear hyperintense signals within both superficial and deep fascia on T2-weighted images. MRI in the other two patients showed no signal abnormality within deep fascia or muscles. No necrosis or fascial edema was found during surgery.

In this study, MRI revealed clinical unsuspected necrotizing soft-tissue infections. Conversely, signal intensity abnormalities of deep soft tissues were present in all but two patients with signs of surgical cellulitis- no visible gross abnormalities were present during surgery in these two patients; however, one did have fascial necrosis on histologic exam. Given these findings, sensitivity was calculated to be 93% and specificity was 100%. Results confirm that T2-weighted images are reliable in determining depth of soft-tissue involvement in acute cellulitis and may help select patients with necrotizing cellulitis in whom surgical debridement may be indicated. (13) The strength of this study is its prospective type. However, MRI may be limited in depicting small foci of necrosis.

Saiag et al. evaluated the value of MRI in adults presenting with severe acute infectious cellulitis. Twenty-six patients admitted for clinical signs of severe infectious cellulitis over a three year period were included in the study. Inclusion criteria were clinical features of acute infectious cellulitis, the presence of at least one criterion of severity (purpura, bullae, post-bullous ulceration, skin necrosis, cutaneous hypoesthesia, fluctuation, confusion, agitation, oliguria, or hypovolemia) and lack of favorable progression of the above signs after the first twenty-four hours of antibiotic treatment. IV antibiotics were immediately started. Patients were examined twice a day and MRI was considered if there was lack of improvement in the clinical signs after 24 hours of antibiotics. In those patients who experienced little or no improvement with antibiotic treatment after the first MRI, a second MRI was performed. Patients suspected of having necrotizing fasciitis, subcutaneous abscess, or pyomyositis were surgically explored. Criteria for uncomplicated infectious cellulitis were 1) surgical exploration negative for necrotizing fasciitis, pyomyositis, or subcutaneous abscess or 2) complete recovery without intervention. MRIs were interpreted by two radiologists who were unaware of the diagnoses for each patient.

In patients with myositis (n=5), MRIs showed spindle-shaped or round, well-defined, homogeneous areas of high-signal intensity within the involved muscles. In patients with cellulitis with subcutaneous abscess (n=7), images on T2-weighted films showed round or spindle-shaped, well-defined, homogeneous areas of high-signal intensity in the hypodermis. In patients with necrotizing fasciitis (n=3), MRI showed numerous (>2), homogeneous, well-defined, dome-shaped, areas with very high signal abnormalities in the thickened hypodermis (areas were >10 cm high and >1.4 cm thick). Patients with infectious cellulitis without abscess formation (n=11) also had dome-shaped areas of high-signal intensity on T2-weighted images; but these were less numerous (1 or 2), heterogeneous, ill-defined, thinner(<1.4 cm), and smaller (<10cm) than those described in the patients with necrotizing fasciitis.

In this study, characteristic features were found on MRIs in patients with pyomyosistis, necrotizing fasciitis, and infectious cellulitis with or without subcutaneous abscess. The dome-shaped areas of high signal intensity on T2-weighted sequences, found in the hypodermis close to the muscles, in patients with necrotizing fasciitis, pyomyositis, or infectious cellulitis are more difficult to analyze. They may correspond to watery exudates as in infectious cellulitis or necrotic tissue as in necrotizing fasciitis. In this study, the high number of these areas, their homogeneity, and their greater thickness and height were suggestive of necrotizing fasciitis. Further studies are needed to determine the MRI aspects of necrotizing fasciitis as well as confirm that MRI provides the early and correct diagnosis of necrotizing fasciitis. Limitations of this study are the small number of patients with necrotizing fasciitis and the patient population being restricted to those patients who represented a diagnostic dilemma and required different treatment. (14)

Left: cellulitis of the leg

Right: necrotizing fasciitis of the thighs

MRI does provide good anatomic correlation and can help to plan surgical intervention or to perform closed drainage of poorly localized abscesses. However, the drawbacks of MRI include its contraindication for any patients with implanted magnetic devices (pacemakers, recent stents, etc.) and the need to lie completely still, which is often hard for agitated patients. MRI cannot visualize soft-tissue gas formation, a highly specific sign of anaerobic cellulitis. It is expensive and not widely available. Most importantly, no MRI sign alone is diagnostic of any particular type of infection.

Cost of Studies

Technical Professional

Treatment

Early diagnosis is the key to treatment. Immediate and aggressive surgical debridement is essential. Incision to the deep fascia should be performed and all necrotic skin and nonviable fascia should be debrided. Re-exploration should be done in twenty-four hours and as needed for evidence of disease extension. Skin grafting and amputation may be required.

Antibiotics should be stared as soon as possible, although they have not been shown to alter mortality. The culture and gram stain should guide the antibiotic therapy. While waiting for these, initial therapy should include broad-spectrum antibiotics aimed at streptococci, anaerobes, enteric gram-negative organisms, and staphylococci. Frequently used agents include ampicillin or cephalosporin, aminoglycoside, and clindamycin or metronidazole. Beta-lactamase inhibitor combinations have also been used.

Supportive care is also important. Aggressive fluid resuscitation and nutritional support has been shown to decrease mortality. Hyperbaric oxygen may or may not be benficial; there are no randomized, prospective, controlled trials demonstrating its efficacy. Its use should not preclude early surgical debridement. (5)

Mortality rates range from 9% to 74% (mean 42%). In the United States, the Center for Disease Control reported a 28% case fatality rate from 1989 through 1991. Delay in diagnosis and operative treatment (greater than twenty-four hours after presentation) has been shown to increase mortality. Other risk factors that have been shown to correlate with increased mortality include advanced age (>50 years), diabetes mellitus, arteriosclerosis, chronic renal disease, malnutrition, and anatomic site of infection involving the trunk. The cause of death in necrotizing fasciitis is usually overwhelming sepsis or multiple organ system failure. (1)

Case Follow-up: B.T. was taken emergently to the operating room that night for wide debridement. Extensive debridement from the right groin extending over the labia majora and onto the medial aspect of right thigh revealed necrotic subcutaneous tissue and fascia. Cultures were obtained and grew gram-positive cocci and gram-positive rods in significant quantities. The patient remained stable throughout the debridement procedure, but developed bradycardia (HR 30s) and hypotension (SBP 50s) while in the PACU. She was then started on broad-spectrum antibiotics (Zosyn and Gentamicin) and admitted to the surgical ICU. She had a complicated hospital course; including persistent temperature spikes, multiple antibiotic regimens, catheter sepsis, non-Q-wave myocardial infarction, massive anasarca, respiratory failure, and hemodynamic instability requiring CVVHD. On September 22, seven weeks after her admission, the patient had V tach/fib arrest and was unable to be resuscitated.

Summary Table of Studies

Necrotizing fasciitis is a rare but devastating infectious disease of the soft tissues. The indolent nature of early necrotizing fasciitis and its nonspecific clinical signs can considerably delay the diagnosis. The primary care physician should be aware of this disease and always consider it in the differential of acute cellulitis. In the dilemma of differentiating simple cellulitis from occult soft tissue infection, clinical suspicion alone warrants early surgical evaluation.

Several tests can be utilized to aid in the diagnosis of necrotizing fasciitis: plain films, FNA, CT scan, MRI, and frozen-section tissue biopsy.

• Plain films are highly specific for soft-tissue gas, a late finding in necrotizing fasciitis, but can miss those patients who have yet to develop this sign.
• Compared with MRI, CT is not as accurate in differentiating between cellulitis and necrotizing fasciitis due to its decreased sensitivity in detecting deep fascial fluid. CT does detect the presence of soft tissue gas more reliably than MRI.
• MRI detects deep fascial signal abnormalities, which have been shown to have a high sensitivity in the diagnosis of necrotizing fasciitis, although one study has reported low specificity. The clinical condition should be correlated with findings of the MRI to help rule out noninfectious causes of positive findings.

If a clinician suspects necrotizing fasciitis, he should check for crepitus on physical exam and for soft-tissue gas on plain film. These are quick, noninvasive, and inexpensive tools. However, if these are negative and the clinical suspicion is present, the investigation should be continued. Much controversy exists over the next step in diagnosis. Plain film and CT are reliable in detecting soft tissue gas, but it is unclear if CT provides any additional information over plain film. If soft tissue gas is not present, MRI might aid in making an early diagnosis because it excels at detecting the deep fascial involvement that is characteristic of necrotizing fasciitis. The logistic constraints of obtaining a MRI in patients with suspected necrotizing fasciitis should not delay any other diagnostic tests or appropriate treatment. If CT or MRI do not exclude its presence, further tests, such as frozen-section biopsy, should be performed. All available methods, including early surgical consultation, should be utilized to rule out necrotizing fasciitis if there is high clinical suspicion.

Further studies are necessary to prospectively investigate the specificity and sensitivity of these diagnostic tests in differentiating between necrotizing fasciitis and severe forms of nonsurgical cellulitis.

1. Green RJ, Dafoe DC, Raffin TA: Necrotizing fasciitis. Chest 110: 219-229, 1996.

2. Descamps V, Aitken J, Lee MG: Hippocrates on necrotising fasciitis. Lancet 344: 556, 1994.

3. Dellinger EP: Severe necrotizing soft-tissue infections: multiple disease entities requiring a common approach. JAMA 246 (15): 1717-1721, 1981.

4. Stone DR, Gorbach SL: Necrotizing fasciitis: the changing spectrum. Dermatology Clinics 5 (2): 213-220, 1997.

5. Chapnick EK, Abter EI: Necrotizing soft-tissue infections. Infectious Disease Clinics of North America 10 (4): 835-843, 1996.

6. Lille ST, Sato TT, Engrav LH, et al. Necrotizing soft tissue infections: obstacles in diagnosis. Journal of the American College of Surgeons 182:7-11, 1996.

7. Stamenkovic I, Lew PD: Early recognition of potentially fatal necrotizing fasciitis: The use of frozen-section biopsy. New England Journal of Medicine 310 (26): 1689-1693, 1984.

8. Majeski J, Majeski E: Necrotizing fasciitis: improved survival with early recognition by tissue biopsy and aggressive surgical treatment. Southern Medical Journal 90 (11): 1065-1068, 1997.

9. Fisher JR, Conway MJ, Takeshita RT, et al. Necrotizing fasciitis: importance of roentgenographic studies for soft-tissue gas. JAMA 241 (8): 803-806, 1979.

10. Wysoki MG, Santora TA, Shah RM, et al. Necrotizing Fasciitis: CT characteristics. Radiology 203 (3): 859-863, 1997.

11. Schmid MR, Kossmann T, Duewell S: Differentiation of necrotizing fasciitis and cellulitis using MR imaging. AJR 170: 615-620, 1998.

12. Loh NN, Chen IY, Cheung LP, et al. Deep fascial hyperintensity in soft-tissue abnormalities as revealed by T2-weighted MR imaging. AJR 168: 1301-1304, 1997.

13. Rahmouni A, Chosidow O, Mathieu D, et al. MR imaging in acute infectious cellulitis. Radiology 192: 493-496, 1994.

14. Saiag P, Le Breton C, Pavlovic M, et al. Magnetic resonance imaging in adults presenting with severe acute infectious cellulitis. Archives of Dermatology 130: 1150-1158, 1994.