Bacterial Skin and Soft-Tissue Infections: A Systematic Approach to Diagnosis and Treatment

Part II: Complicated and Serious Infections

Author: Ramin R. Samadi, MD, FACEP, FACP, President and Chief Medical Officer, Tarrant Acute Care Physicians, PA, Fort Worth, TX; Medical Director, Trinity XPress Med Medical Center, Fort Worth, TX; and Assistant Clinical Professor, Division of Emergency Medicine, University of Texas Southwestern Medical School, Dallas.

Peer Reviewers: Larry B. Mellick, MD, MS, FAAP, FACEP, Vice Chairman for Academic Development and Research, Department of Emergency Medicine, Medical College of Georgia, Augusta; and Sandra M. Schneider, MD, Professor and Chair, Department of Emergency Medicine, University of Rochester School of Medicine, Rochester, NY.

This is the second article in a two-part series on bacterial skin and soft-tissue infections (SSTIs). Part I discussed infections that are common and frequently encountered in the general practice of medicine. Furthermore, these infections predominantly are benign in nature and, in majority of cases, do not result in significant morbidity or mortality.

Part II will cover diseases that have a virulent and serious nature. Some of these entities occur relatively frequently in the general population, and some others are quite unique and can only be found in a particular subset of the patient population. For most clinicians, the rarity of some of these infections poses a major diagnostic challenge. If left untreated, and occasionally even with proper and timely therapy, these infections result in significant morbidity or mortality. The Editor

Diabetic Foot Ulcer Infections

Foot ulcers in patients with diabetes mellitus (DM) are a growing and costly public health concern. Considering the on-going global rise in the prevalence of DM, the importance of related foot ulcers cannot be underestimated. DM foot ulcers occur at a rate of 2.5% each year.1,2 It is estimated that approximately 15% of all diabetics will be afflicted by foot ulcers during their lifetimes.3 All infected ulcers should be considered potentially serious and very difficult to treat. In DM patients, infected ulcers are the leading cause of admission to the hospital. Additionally, they are one of the major causes of limb amputation in the United States. Non-healing ulcers lead to amputation 85% of the time.4 Following the first amputation, as a result of the altered biomechanics, the patient will be at a greater risk for additional amputation(s). Up to 50% of patients with one amputation will require another amputation within five years of the first surgery.4 The annual cost of treating foot diseases in DM patients is estimated at $1 billion.4

DM foot ulcers are due to the accumulation of repetitive painless micro-trauma to pressure points of the feet and toes. Anatomical changes in the patient’s feet and toes are major contributors to development of these ulcers. Furthermore, DM-related peripheral sensory neuropathy, peripheral microvascular compromise and the resultant focal ischemia, and altered host defenses play a significant role in the genesis of these ulcers. Sensory neuropathies will make the patient unable to feel the pain that is otherwise present when trauma and pressure are applied to the pressure points. It has been estimated that 60% of DM patients with non-healing ulcers suffer from arterial insufficiencies. As a result, 50% of this group of patients will face the possibility of one or more amputations.5 Poorly controlled glucose hemostasis will result in altered host responses, manifested as impairment of polymorphonuclear leukocytes chemotaxis, adherence, phagocytosis, and intracellular bactericidal activities.6

Although not all DM foot ulcers are infected, most frequently they act as a port of entry for infectious pathogens. A very common adverse consequence of these ulcers is the occurrence of skin, soft-tissue (superficial and deep tissue), and bone infections.

Presentation and Classification

Not all DM foot ulcers are infected; however, the identification and diagnosis of an infected DM foot ulcer is not always a simple task. Any increase in the size of the ulceration, presence of surrounding erythema or underlying crepitation, or increase in drainage should alert the clinician. The infected ulcers are not necessarily painful. Fever is usually absent. Other findings may include referral pain in the leg or ankle. In an otherwise stable DM patient, presence of unexplained hyperglycemia should mandate complete exposure and a careful examination to rule out the presence of such an infection.

Osteomyelitis is one of the major complications of DM foot ulcers. It has been estimated that close to 15-20% of all DM foot ulcers will result in osteomyelitis. Infected ulcerations and secondary osteomyelitis are more prevalent in men, in patients with disrupted skin integrity due to either penetrating injuries or fungal infections, and in patients suffering from DM for more than 10 years. If, upon probing the patient’s wound, bone is encountered, the risk of the presence of osteomyelitis is 89%. In patients with osteomyelitis, the metatarsal bones are involved in 51.7% of cases. The second highest incident occurs in the toes. (See Table 1.)

Table 1. Diabetic Foot Ulcers

Infected DM foot ulcers are divided into the following two categories:

1. Minor foot infections. These superficial ulcers are surrounded by a non-advancing erythema (cellulitis) of fewer than 2 cm. Systemic signs, abscesses, osteomyelitis, or gangrene are absent. The majority of these foot ulcers are infected by gram-positive organisms, predominantly Staphylococcus aureus.

2. Serious foot infections. These ulcers are deep, penetrating and potentially full-thickness. The surrounding cellulitis spreads beyond the margins of the ulcer. Systemic signs (i.e., malaise, fever and night sweats) usually are present. These ulcers are potentially limb- and/or life-threatening. The infections in these ulcers are polymicrobial.

Diagnosis. A high rate of suspicion should assist the clinician in making the correct diagnosis. Excluding the minor foot infections, the majority of DM foot ulcer infections are polymicrobial and involve a wide range of aerobic and anaerobic pathologic organisms.7 All DM foot ulcers will be superficially colonized with a plethora of bacteria. Cultures should be obtained from the base of a surgically debrided wound. Another approach may involve obtaining culture material through a quantitative bone biopsy. Properly obtained cultures, both aerobic and anaerobic, offer the best opportunity for the identification of the causative organisms and have 95% sensitivity and 99% specificity. In most cases, an average of 2-6 organisms will grow.7

Radiological studies of the foot and soft tissue must be pursued in all cases. The presence of gas in soft tissue or along the fascial planes will be diagnostic for the presence of deep soft-tissue infection. In a DM patient, due to neuropathic arthropathies and healing fractures, radiological studies have low specificity for diagnosing osteomyelitis.

Nuclear bone scan is of limited value in the diagnostic work-up of these ulcers. Another option is the use of a radio-labeled leukocyte scan. Recent advancements in magnetic resonance imaging (MRI) studies have made this modality a promising tool for the diagnosis of osteomyelitis in the infected DM foot ulcer.

Assessment of adequacy of arterial perfusion of the involved limb must be performed through modalities such as Doppler waveforms of the involved pedal arteries, non-invasive studies of segmental pressures, duplex scanning, measurement of toe pressures, and transcutaneous oxygen monitoring.8,9 In DM patients, due to prevalent medial calcification, calculation of ankle-brachial index is not a reliable tool.

Treatment and Prognosis. The treatment of infected DM foot ulcers is best managed through a multidisciplinary approach. Surgical debridement, wound care management, and antibiotic therapy are the primary therapeutic modalities. If necessary, vascular surgery, endocrinology, dietetic, and podiatry consultations also should be obtained.

Drainage of the abscesses and debridement of the necrotic tissue are very important aspects of care. Thorough, sharp wound debridement down to the uninfected viable tissue should be performed and repeated as necessary. Due to associated tissue injury, use of disinfectants (e.g., Dakin’s solution, hydrogen peroxide, and povidone-iodine) is not recommended. Antibiotic therapy should be reserved for true cellulitis, presence of abscess, or purulent wound drainage. Non-inflammatory ulcers should be considered uninfected.

In the past decade, there has been growing clinical interest in a modified treatment of DM foot ulcer osteomyelitis with a less aggressive surgical ablation, favoring instead more reliance on conservative and limb-sparing surgery and long-term antimicrobial therapy. In a large series involving 223 patients undergoing therapy with a modified and conservative approach, authors demonstrated successful healing of either deep-tissue foot infections and/or osteomyelitis in 39% patients without any amputation.10 Higher cure rates, 81% and 75% respectively, with similar conservative approach has been reported in smaller studies.11

Optimal antibiotic therapy for infected DM foot ulcers is still a matter of clinical controversy. Considering the polymicrobial nature of the majority of these infected ulcers, the use of broad-spectrum antibiotics is highly recommended.

In minor foot infections, the patient can be treated as an outpatient and with oral antibiotic therapy. Patients must be instructed to follow up in 72 hours to assess the response to the prescribed antibiotic regiment. Appropriate antibiotics include amoxicillin-clavulanate, clindamycin, cephalexin, or dicloxacillin. Other very attractive alternatives are moxifloxacin, or ciprofloxacin plus either clindamycin or metronidazole.

Patients with serious foot infections must receive parenteral antibiotic therapy. After the completion of wound debridement, and based upon the severity of the infection, patient’s health, reliability and compliance, and socio-economic status, treatment can be provided either in a hospital setting or as an outpatient. The outpatient therapy is less costly and more patient-friendly. Antibiotic therapy should include coverage for aerobic and anaerobic gram-positive and gram-negative organisms. Appropriate antibiotics include ampicillin-sulbactam, ticarcillin-clavulanate, piperacillin-tazobactam, imipenem-cilastin, cefoxitin, meropenem plus an aminoglycoside or aztreonam +/- vancomycin. Another excellent choice is ciprofloxacin plus either clindamycin or metronidazole. Moxifloxacin is another attractive alternative. The therapy should continue for a minimum of 4-6 weeks. Parenteral antibiotics can be switched to an oral form after significant wound improvement has been accomplished.

Due to the recent emergence of community-acquired methicillin resistant Staphylococcus aureus (CAMRSA), provision of coverage for this organism also should be considered in the selection of antibiotic therapy. Antibiotic therapy for CAMRSA in patients without any significant risk factors includes trimethoprim-sulfamethoxazole, and clindamycin. Tetracyclines, although less sensitive, are another alternative.12 Additional information for CAMRSA can be found in Part I of this article.

In appropriate settings additional therapies may include vascular surgery or use of pentoxifylline (to improve micro-circulation); application of bioengineered tissue (Apligraf); or use of platelet-derived growth factor gel (Becaplermin); or hyperbaric oxygen therapy. These modalities may be able to enhance the treatment of refractory infection.13

Several mechanical interventions or adjustments also may be helpful. Limb elevation will reduce the dependent edema and use of a shoe prosthesis can redistribute the pressure away from the pressure points. Patients also should be advised to refrain from weight-bearing whenever possible.

The prognosis depends on multiple factors. Unfortunately, even with the best therapy, amputation may become a necessity. The rapidly progressive ulcers, presence of gas in the underlying tissues, and presence of gangrene in the overlying tissues are associated with poor prognosis. Other co-morbidities, such as renal impairment, cardiac disease, immunosuppression, malnutrition, obesity, and malignancy also contribute to a poor outcome. With aggressive, targeted antibiotic therapy and early debridement, the short-term outcome generally is good. However, the three-year recurrence rate is high. In this subset of patients, the rate of amputation exceeds 22%. Annually 50% of the 125,000 limb amputations in the United States are attributed directly to DM foot ulcers.14 Furthermore, the three-year mortality in this group of patients is 27%.

Mammalian Bites

Approximately 35% of American households own a dog. In 2001, it was estimated that an average of 68 million canines were kept as pets in the American households. The majority of the mammalian bites are due to dogs (90%), followed by cats (5%), and humans and rodents (2-3% each). It is postulated that, annually, about 1-2% of all the emergency department (ED) visits are due to a mammalian bite. CDC estimations reflect that 4.5 million bites occur annually.15,16 The annual incidence of dog and cat bites has been reported as 300/100,000 of the general population.17 In one large study, the incident of animal bites was reported as 83% dogs, 14% cats, 1.7% rats and mice, and 1.3% other animals. On average, 17% of all the dog bites require medical attention, 38% of which are evaluated in an ED. In 1992-1994, an annual average of 334,000 patients were treated in an ED due to a dog bite.18 In 2001, this number increased to 368,245. More than 50% of bites occurred in patients aged 14 years or younger. Young boys are more susceptible to dog bites than girls of the same age group.18 The majority of dog bites are inflicted by a family pet. During 1997-1998, 75% of fatal dog bites were inflicted on a family member or guest on the family’s property.18 In 2001, 8% of the dog bites in patients aged 16 years or older were work-related.18 The number of insurance claims from dog bites has been increasing 1-2% each year. In 1996, State Farm Insurance’s payout for dog bite claims increased by 25%.

At the onset, the majority of mammalian bites appear to be minor; however, these bites have the potential for causing serious, localized tissue damage, including tendon, nerve, joint, and bone damage, as well as systemic complications. Infection is the most common complication and may occur in up to 10% of the victims. On rare occasions it may even lead to sepsis and/or limb amputation. (See Table 2.)

Table 2. Mammalian Bite Wound Complications

Bacteriology. Regardless of the origin of the mammalian bite, penetration of the tooth into the human skin has the potential of inoculation of high concentrations of bacteria into the soft tissue. The mammalian mouth and gingiva contain more than 200 species of aerobic and anaerobic bacteria. Bacteria such as Staphylococcus aureus, beta-hemolytic streptococcus, Streptococcus viridans, Pasturella multocida, anaerobic streptococcus, Fusobacterium, spirochetes, Corynebacterium species, and Bacteroides species are included in this mixture. Polymicrobial mixed bacterial infections are fairly common. Anaerobic infections can be identified in 41% of all the infected wounds. P. multocida is present in up to 50% of infected dog bites and 75% of infected cat bites. The role of various pathogenic bacteria as the causative agent is somewhat different in infected wounds secondary to dogs and cats. (See Table 3.)

Table 3. Mammalian Bite Wounds: Infectious Organisms

In most cases, the results of the initial cultures that are obtained from a non-infected wound do not necessarily correlate with the results of the subsequent cultures that have been obtained from an infected mammalian bite. Hence, obtaining routine cultures is not considered standard of care. Wound cultures should be obtained from obviously infected wounds and in febrile or immunocompromised patients.19

Risk Stratification. (See Table 4.) Proper risk stratification should be made based on the following criteria:20

Table 4. High Risk Mammalian Bite Wounds

Anatomical Location. Bites on the hands and feet should be considered high risk. Due to the presence of thin membranous bones of the skull, the same should apply to the bites of skull and face in infants and young children.

Wound Depth. Involvement of deep structures (i.e., bone, joints, tendons, nerves, vessels, and viscera) classify the wound as high risk.

Puncture Wounds. Due to the inability of thorough irrigation and decontamination, puncture wounds are considered to be high risk.

Cat and Human Bites. These bites are usually of the puncture variety and should be considered as high risk.

Dog Bites. Despite the common occurrence of these bites, only a few related, controlled, small studies have ever been conducted. These studies suggest that a disproportionate number of the dog bites are inflicted by German shepherds, Rottweilers, Doberman pinschers, pit bulls, and Alaskan/malamutes.21 About 5% of patients with dog bites who are evaluated at an ED return to the department with a complication. About 1% of dog bites require hospitalization.22

Depending on the type of the attacking dog, injuries can present as minor contusions, tissue tearing and disruption, avulsions and lacerations, and/or crush injuries. A large dog can introduce 150-450 pounds per square inch of force to the afflicted tissue.23 Radiological studies may be necessary to explore the possibility of underlying bone or joint injury. More serious injuries may include significant blood loss, rib fracture, pneumothorax, airway compromise and even death.24

Regardless of age, approximately 75% of all the dog bites involve one or more of the extremities.25-27 The presence of numerous small compartments and the thin covering of the soft tissue over the bones and joints, compared to bites elsewhere, increase the likelihood of infection.

Severe facial injuries almost exclusively occur in children younger than 10 years of age. Their shorter stature and predisposition to inspect, kiss, or smell the animal puts children at a higher risk for bites of the face and scalp. The majority of these bites involve the lips, nose, cheeks, and parotid region.28,29 Dog bites of the face and neck may result in major injuries, including severe brain damage and even death. The resultant hemorrhage is usually pronounced and may involve the main vessels of neck. Cases of occult carotid artery injury and delayed cerebral infarction have been reported.

The majority of infected dog bite wounds are polymicrobial in nature. (See Table 3.) The involved pathogens in infected wounds secondary to dog bites include, but are not limited to, P. multocida, enterobacteria, Pseudomonas aeruginosa, S. aureus, Bacillus subtilus, and Streptococcal species (most commonly Strep. viridans).30,31 Wounds infected by P. multocida have a tendency to occur within 24 hours of the dog bite and more frequently in younger children (younger than 4 years) or adults older that 55 years of age. Dog bite wounds infected by P. multocida may present with a disproportionate amount of pain. These victims also may be febrile.

Capnocytophaga canimorsus (formerly CDC group DF-2) is a unique dog bite-related infectious pathogen. In 1977, infections caused by C. canimorsus were reported as a new disease in humans. Dogs are considered a reservoir of this organism. This facultative anaerobic gram-negative bacillus can be found in the normal oral flora of 16% of dogs.32,33 Human infections are fairly uncommon. C. canimorsus even can be transmitted through the licking of a wound or open sore by the dog. Subsequent to the infections of the animal bite wound, the bacteria can become fairly virulent. The onset of systemic manifestations of infection is delayed and usually occurs in about 7-14 days after the bite. Many of the infected wounds seem insignificant, and local signs of infection usually are absent.33 However, these infected wounds may result in systemic manifestations of fever, malaise, myalgia, vomiting, diarrhea, abdominal pain, dyspnea, confusion, headache, and skin rash.

A subset of patients, including those with immunocompromised status, asplenia, chronic alcoholism, trauma, Hodgkin’s disease, idiopathic thrombocytopenic purpura (ITP), chronic corticosteroid use, chronic lung disease, peptic ulcer disease, hypertension, and atherosclerotic heart disease are at a significantly high risk. In this subset, the bacteria can cause severe cases of gangrene, sepsis, purpura, disseminated intravascular coagulopathy (DIC), and death.31,34,35 The mortality rate can exceed 25%.31,35

In a febrile patient who previously has been been bitten by a dog or has been in contact with a dog, special attention should be drawn to this pathogen. The laboratory should be specifically instructed to culture C. canimorsus. Due to the bacteria’s slow growth rate, the timely yield of the wound culture is questionable. Conversely, this organism may be more easily identified in the peripheral blood smear.

CDC Nonoxidizer 1 Group (NO-1) Infections. NO-1 is a recently identified bacterium associated with dog and cat bites. From 1974-1998, 22 isolates of this unusual bacterium were sent to the CDC for identification. Although possessing different morphology and cellular fatty acid profile, this organism was found to have phenotypic characteristics that were similar to asaccharolytic strains of Acinetobacter.36 All isolates were recovered from infected wounds secondary to animal bites. Dog bites were responsible for 77% of the infected wounds, and 18% were due to cat bites. All the infections were identified in previously healthy patients with no underlying illness. At present, no surveillance system exists for reporting dog or cat bite wound-associated infections and, therefore, the incident of NO-1-related wound infections is unknown. Furthermore, the exact nature of the etiologic role and the pathogenicity of NO-1 are also unclear.

The most prominent clinical presentation of infected wounds with NO-1 was in the form of localized infections, i.e. cellulitis, purulent drainage, and/or abscess formation. In the reported case studies, 58% of the patients with NO-1 infected wounds required hospitalization. The median time between the occurrence of the animal bite and the manifestation of the infection was 17.5 hours (range, 3-78 hours). All the involved patients responded to antibiotic therapy and recovered in full.

The NO-1 organism is susceptible to aminoglycosides, beta-lactams, tetracyclines, quinolones, and sulfonamides. Fifty percent of the isolates are resistant to trimethoprim.

Cat Bites and Scratches. Cat bites, compared to dog bites, occur at a substantially lower rate. However, they are associated with a significantly higher rate of infection.37 The incidence of wound infection following a cat bite is significantly higher than in dog bites (28-80% vs. 3-18%). Similarly, in infected wounds following a cat bite, the time span between the bite and clinical manifestations of infection is significantly shorter than for dog bites (7-18 hours vs. 12-48 hours). It has been estimated that 29% of all cat bites or scratches will become infected.28,37,38 The sharp, pointed feline teeth act as a hypodermic injector, which can penetrate the tendon sheaths, tendon, ligaments, and joints. The resultant puncture wound(s) close easily and quite frequently appear benign. Additionally, these wounds are very difficult to irrigate and virtually impossible to debride. Subsequently, the inoculated bacteria may grow in the deep tissues and produce serious infection.

Compared to dog bites, cat bites or scratches occur more frequently in women and 25% of the time happen while playing with the cat.28,38 Children younger than 6 years of age have the highest incidence.39

The involved pathologic organisms include P. multocida, Strep. viridans, other strains of streptococcus, S. aureus, and strains of Bacteroides.

Treatment of Dog and Cat Bites. Multiple bite wounds are common, and the clinician must spend adequate time to investigate the less-than-obvious bites. The integrity of the neurovascular bundle also must be ascertained. Sharp mechanical debridement in a sterile environment followed by copious irrigation to remove contaminated material must be considered standard of care. The combination of debridement and irrigation with at least 250 cc of saline, can reduce the risk of wound infection by 30-fold.40 In the majority of cases, the use of additional antiseptic solutions is not recommended. Only in cases considered as being high risk for rabies transmission should the wound be irrigated with a 1% benzalkonium chloride solution.41 Appropriate tetanus and rabies prophylaxis also should be provided based on the patient’s and animal’s immunization history.

High-risk bites of face and neck in children or extensive multiple animal bite wounds often are best examined and managed in the operating room and under general anesthesia. Radiological studies are significantly beneficial and highly recommended in injuries involving the body cavities, face, head, joints, or bones.

Primary wound closure following a mammalian bite is controversial. The majority of wounds are considered minor and do not require any closure. However, close to 10% of bite-related wounds will require closure or surgical intervention.22 Large and potentially disfiguring low-risk wounds may be sutured. After appropriate wound care and irrigation, wounds of the face, scalp, trunk, or proximal extremities may be sutured at the time of the patient’s first visit to the ED.

High-risk wounds must be left open and laceration(s) should not be sutured.42 Hand, foot, and other high-risk dog or cat bites should be treated vigorously and aggressively. Wound repair can be managed with a delayed primary closure. Bite wounds must be loosely packed with fine mesh gauze soaked in saline. The patient’s wound should be reassessed in 48-72 hours after the first care of the wound. On the second visit, the wound will require additional irrigation and possible debridement. If the wound remains uninfected, it can be repaired around the fourth post-bite day. If the wound becomes infected, the sutures should be removed. Consideration should be given to admitting these patients to the hospital.

Antibiotic Therapy. There are no definitive databases upon which to make concrete recommendations for the antimicrobial therapy in animal bites. No single antibiotic agent is consistently active against all the numerous pathogens that potentially are involved in an infected bite wound. The selection of antimicrobial agent should be based on the animal species, the timing of rendered care, and the suspected pathogens. The initiation of empiric therapy should be broad-spectrum and aimed at aerobic and anaerobic bacteria. Most superficial contusions, abrasions, and wounds do not require antibiotic therapy. The only exception to this axiom is the puncture wounds from either a tooth or a claw.

The antibiotic therapy for either prophylaxis or treatment of dog or cat bite wounds must include coverage for P. multocida. This bacterium demonstrates a specific resistance pattern. In up to 50% cases, P. multocida is resistant to first-generation cephalosporins and the semi-synthetic penicillins. Amoxicillin-clavulanate has remained as the standard of care. P. multocida often is resistant to antibiotics such as cephalexin or erythromycin. Macrolides demonstrate very poor in-vitro activity against P. multocida and should not be used in animal bite wounds.43 With the ongoing increase of infections secondary to CAMRSA, utilization of other antibiotics may become a necessity. Other beta-lactams, including cephalosporins, may be another option. Fluoroquinolones, specifically moxifloxacin, also should be considered another valuable group of antibiotics. (See Table 5.)

The issue of prophylaxis antibiotic therapy in mammalian bites has remained controversial. In some studies, the prescription of prophylactic antibiotic, regardless of the nature of the wound, did not reduce the risk of wound infection. In other studies, the prescription of prophylactic antibiotics in high-risk wounds was found to be beneficial. In a recently published paper, the meta-analysis of eight related clinical trials revealed that the use of prophylactic antibiotics following either dog or cat bites was not associated with reduction of infections. The only exception was found to be related to bites of hands, where a statistically significant reduction in the rate of infection was found.44 However, the use of the term prophylactic antibiotic therapy often is not strictly correct. In most cases, empiric antibiotics are initiated before the overt presentation of a clinical infection is manifested. Furthermore, considering the potential for a catastrophic course of infections related to dog or cat bites, some authors have recommended the prescription of antibiotics as prophylaxis, even in minor bites. In another meta-analysis study, published in 1994, a risk reduction of approximately 1 patient avoiding infection for every 14 patients was identified in patients who received prophylactic antibiotic therapy.45

If a prophylactic antibiotic is not prescribed, the wound must be observed very closely. Within few days, even the most benign-appearing bite may turn into a severely infected wound. The clinician must consider the fact that following any bite, the incidence of wound infection will markedly increase if the patient is older that 50 years, has a puncture or hand wound, is immune-suppressed, or if the wound is sutured. The same concern should apply to the wounds that have not been tended to for more than 24 hours.

Cat Scratch Disease (CSD). CSD was first described in 1931; however, it was only in 1983 that an unclassified gram-negative bacterium was identified as the cause of the disease. The causative organism is Bartonella henselae. CSD is a feline-associated zoonotic disease. Although the CSD implies that cats are the primary vector, dogs may also cause the same condition. Kittens are the main source of the disease. The organism is inoculated into the skin by the scratching of the animal. It is most often seen in young patients. Approximately 80% of all the CSD are seen in patients younger than 21 years of age. Contrary to cat bites, males are slightly more afflicted. The incidence of CSD is estimated at 3.3/100,000 in the general population. The estimated annual rate is 22,000 cases. CSD predominantly occurs in fall and winter. Cat fleas are involved in the transmission of B. henselae among cats, but the role of fleas or other arthropods in the transmission of the disease to humans is unknown. Cats rarely demonstrate overt signs of infection.

The incubation period for CSD is 3-10 days. Most patients are not ill-appearing. CSD is manifested by a tender papule at the site of the scratch. Within two weeks, a large and impressive regional lymphadenopathy will develop. Lymph nodes may become pustular. CSD is one of the most common causes of chronic lymphadenopathy among children. In 25% of cases, at this stage of the disease, CSD also may have systemic manifestations, including fever, chills, malaise, and headache. The course of CSD may last a few weeks to months.46 Recovery frequently is spontaneous. Rare severe complications may include granulomatous conjunctivitis, neuroretinitis, atypical pneumonia, encephalitis, meningitis, oculuglandular syndrome, and lytic bone lesions. Although more prevalent in immunocompromised and HIV-positive patients, these complications may even occur in immunocompetent patients. CSD also may manifest itself as fever of unknown origin (FUO). In immunocompromised patients, B. henselae infection also may cause other potentially life-threatening disease manifestations, including bacillary angiomatosis (epithelioid vascular nodules, papules, and tumors on the skin and mucous membrane) and peliosis hepatis (vascular tumors of the liver).

Diagnosis usually is made based on the clinical findings and the history of contact with a cat or the presence of a scratch. Culture of the organism is quite difficult. However, staining of the pus from infected lymph nodes with Warthin-Starry silver stain maybe beneficial for the identification of the pathogen. Serologic testing is the standard method of diagnosis.47 A single elevated, indirect immunofluorescence assay titer or enzyme immunoassay for IgG or IgM antibodies generally is sufficient to confirm CSD. Initiation of humoral response usually precedes the onset of the symptoms. Other diagnostic modalities include polymerase chain reaction, cat-scratch skin test, negative studies for other causes of regional lymphadenopathy, and characteristic findings on the lymph node biopsy can confirm the diagnosis.48

The evidence for the use of antibiotics in CSD is controversial and conflicting. Treatment recommendations for B. henselae-associated diseases, CSD included, depend on the specific disease presentation. For most part, due to the generally self-limiting nature of the disease, the assessment of the efficacy of antibiotic therapy is difficult. Some studies have found no benefit in regard to the use of antibiotics. However, recent experience with azithromycin suggests that this antibiotic hastens the resolution of the lymphadenopathy of CSD.49 In more severe cases, other antibiotic therapy regimens may be beneficial. These combinations include azithromycin plus rifampin, doxycycline plus rifampin, ciprofloxacin, or rifampin alone.50 Doxycycline or erythromycin are the drugs of choice in bacillary angiomatosis and peliosis hepatis.51

Rare Infections Related to Mammalian Bites. Infectious endocarditis (IE). Rare cases of IE have been reported following dog bites. Only one case of acute IE secondary to a cat bite has been reported in the literature. The majority of the acute cases of IE are due to Staphylococcus aureus and occur in previously healthy heart valves. S. aureus commonly can be found on the skin and hair of dogs. Most cases of sub-acute IE are due to infections caused by alpha-hemolytic Streptococci or Enterobacter species. The majority of the sub-acute IE cases involve previously damaged heart valves.52 It is noteworthy to mention that the direct causal relationship between dog bites and IE has not been proven.

Iguana and Other Non-venomous Reptilian Bites. Pet reptiles have increased in popularity in United States. Currently 3% of American households contain a reptile.53 The most popular reptilian pet is the green iguana (Iguana iguana). As the number of these household pets increases, so will the risk of their non-venomous bites. Certain reptilian behavior, such as head-bobbing, nodding, or charging might warn of an impending bite. No information is available on the frequency of related facial or head injuries. Current literature does not identify any specific pathogen that is associated with iguana bites. Cases of Serratia marcescens have been reported. Furthermore, the issue of potential Salmonella infection must always be a consideration. More than 80% of captive iguana shed Salmonella in their feces. Considering the depth or severity of iguana bite wound, prophylactic antimicrobial therapy may be necessary. The prescribed antibiotic must provide coverage for Salmonella as well as human skin flora.54 Fluoroquinolones in adults and semi-synthetic penicillins in children, are good choices of therapy.

Patients who are at high risk for Salmonella infection-related morbidity and mortality, i.e., pregnant women, children younger than 5 years of age, asplenic, and/or immunocompromised patients, must avoid contact with reptiles.

Ferret Bites. With the ever growing popularity of ferrets as pets, not surprisingly, reported cases of severe bites related to these animals have been published. The majority of these bites are unprovoked. Attacks may occur while the child is sleeping and may result in severe facial damage.55

Human Bites. Human bites are very similar to other mammalian bites except for the following three significant differences: the location of the bite, the presence of Eikenella corrodens, and the human element.

In men, human bites occur more frequently on the hand and fingers and in women on the breast.

At the onset, most of the infected wounds following a human bite are due to S. aureus or Streptococcus. However, with the delay in the manifestation of the infection, the role of anaerobic bacteria, specifically E. corrodens, as the causative agent increases. (See Table 6.) E. corrodens is a slow-growing gram-negative bacillus that is a normal flora in human dental plaques and/or saliva. This organism is identified in 10-30% of infected human bites. It exhibits a synergistic behavior with Streptococcus, S. aureus, Bacteroides species, and other gram-negative organisms.56 E. corrodens has a peculiar antibiotic resistance pattern. It usually is sensitive to ampicillin, penicillin, and cephalosporins but is resistant to clindamycin and semi-synthetic penicillins, i.e., methicillin, oxacillin, or nafcillin.57

Table 6. Human Bite Wounds: Infectious Organisms

In human bites, the human element is related to the victim’s inability and/or lack of forthcoming about the mechanism of injury. Due to inebriation, the victim may not remember the fight that caused the injury or may be embarrassed or reluctant to report it. The time span to become lucid after a drinking binge also may be a factor. These issues will result in delays in provision of medical care and/or confusion in obtaining an accurate medical history. The delay in seeking care will change the pathogenic organisms from gram-positive flora to a more virulent anaerobic bacteria flora. As a result, the patient will be faced with a more prolonged treatment course and potentially severe complications.56

Clenched Fist Syndrome. This syndrome is unique to human beings and is a result of striking another person in the mouth with a closed fist. In the human hand, the dorsal expansion hoods do not cover the metacarpophalangeal (MCP) joints. As result, upon making contact, the opponent’s teeth easily can penetrate into the joint space or tendon sheath. Upon the opening of the assailant’s fist and extension of the fingers, the movement of the involved tendons will carry the opponent’s saliva and potentially other foreign bodies into the joint and/or tendon sheath. The involved trauma and violation of the skin potentially results in a multilayer infection of the skin, subcutaneous tissue, joint capsule, and extensor tendons. The small laceration on the MCP joint can easily be overlooked and mask the potential severity of the case. Obtaining a radiograph of the hand strongly is recommended and may identify a piece of broken tooth in the joint, the presence of a fracture, or evidence of osteomyelitis. If highly suspicious, osteomyelitis must be ruled out by means of a radionuclide bone scan.

Most of these cases occur during the warm seasons and in the dominant hand of men 25-35 years of age who are seeking medical care with a delay. The most prominent manifestation of this syndrome is a small draining wound on one or more MCP joints. The third and/or forth MCP joints of the patient’s dominant hand usually are involved. These infections are very difficult to treat. The progression of the infection to osteomyelitis, septic arthritis, and tenosynovitis is fairly common. Persistent infection may lead to amputation. Even with proper care, it is possible that the patient may not regain the full mobility of this proximal interphalangeal joint. Early surgical exploration is recommended to identify and treat possible joint injuries.58 (See Figure 1.)

Figure 1. Clenched Fist Syndrome

Treatment of Human Bite. In general, the care and treatment of human bite wounds are very similar to other mammalian bites. However, it must be mentioned that full-thickness human bites are highly susceptible to infections and the resultant infections are very difficult to treat. All human bite wounds of more than 24 hours will require surgical exploration and possibly drainage. This examination is best performed in the operating room. If the joint space or tendon sheath is involved, these patients may also need arthrotomy or drainage of tendon sheath or both. Some institutions admit all human bites to the hand to the hospital for intravenous antibiotic therapy. The choices of antibiotic therapy are similar to those for the infected wounds due to dog and cat bites. (See Table 5.) With the persistent drainage and other symptoms and signs, although rare, infection caused by CAMRSA must be ruled out.59 In recent clinical trials linezolid, a novel oxazolidine antibiotic available in intravenous form, has proven to be as efficacious as vancomycin in treatment of severe skin and soft-tissue infections caused by MRSA.60

In human bite wounds, the use of prophylactic antibiotic therapy is highly recommended. Amoxicillin-clavulanate for five days is the recommended antibiotic.

Necrotizing Fasciitis

Necrotizing fasciitis (NF) infections are highly destructive and potentially life-threatening infections of the soft tissues and fascia. Reported cases of these infections date back to the time of Hippocrates. These relatively rare and uncommon diseases are very aggressive and fulminant bacterial infections that involve the subcutaneous, soft tissue, and fascia and cause progressive necrosis and tissue damage. The subtle and non-specific onset of these diseases may result in an erroneous diagnosis of muscle strain, acute viral syndrome, or gastroenteritis. Despite their insidious onset, these diseases progress very rapidly. The majority of misdiagnosed patients will be hospitalized with a fulminant disease within 1-3 days of the initial visit. Furthermore, any delay in diagnosis and therapy will invariably result in significant morbidity or mortality. Mortality rates as high as 74% have been reported. The infections typically are initiated following a minor skin trauma. Varicella infections have been associated with NF infections in children. Other predisposing events include frost bite, chronic leg ulcers, or a surgical incision. Risk factors for NF infections include advanced cardiac, liver, pulmonary and renal disorders, C4 deficiency, chronic alcoholism, chemotherapy, chronic lymphocytic leukemia, chronic skin disorders, diabetes mellitus, hypertension, immunosuppression and immunocompromised states, intravenous drug abuse, malignancy, malnutrition, paraplegia, puerperium, radiotherapy, remote infections, and tuberculosis. Newer medications, such as infliximab and FK506 (a new immunosuppressant macrolide used in post transplant patients) also recently have been associated with NF infections.61 Other identified risk factors are the two extremes of age, obesity, lower socioeconomic status, history of burn, African-American and Native American heritage. Although still remaining controversial, there are also suggestions of a negative correlation between the use of NSAIDs and steroids and NF infections. However, in majority of patients a risk factor cannot be identified. Thirty percent of NF infections occur in previously healthy individuals. The overall incidence of NF infections has been reported at 0.4/100,000 in the general population and 0.08/100,000 in children. In the last decade, there has been a five-fold increase in the number of NF infections.

The frightening progression of the lesion may consume more than one inch of previously healthy flesh each hour. The infections are traditionally polymicrobial. Pathogenic bacteria include Clostridium, Peptostreptococcus, E. coli, Pseudomonas, Streptococcus pyogenes, Staphylococcus aureus, and Streptococcus marcescens. However, in the last decade, monobacterial infections with Group A beta-hemolytic streptococcus (GAS) is frequently identified in pathogenesis of NF.

Types of NF Infections. Different categories have been proposed for the classification of NF infections. Differentiation can be made based on the rapidity of the progression of clinical presentation or the type of pathogenic organism(s).

The classification based on the length, progression, and extent of the disease divides the NF into the following three separate groups: Fulminant, acute, and subacute. Fulminant cases develop extremely fast within hours, develop few blisters, and progress rapidly to shock. In the acute cases, disease progresses over a span of few days and involves large areas of skin. The subacute cases progress over a few weeks and only involve a small area of localized skin.

The classification based on the presence of pathogenic organisms divides the NF into two groups. Type 1 NF, also known as bacterial synergistic gangrene, is a synergistic polymicrobial infection consisting of aerobic and anaerobic bacteria (including Bacteroides and Clostridia species). Type 2 NF, also known as streptococcal gangrene, is a monobacterial infection caused by Group A beta-hemolytic Streptococcus (GAS) with or without Staphylococcus aureus co-infection. (See Table 7.)

Table 7. Pathologic Organisms Isolated in Necrotizing Fasciitis Type 1

The majority of the NF infections are of the Type 1 kind, where a polymicrobial group of aerobic and anaerobic organisms invade the soft tissue and fascia in a synergistic manner. Most of these infections involve a combination of beta-hemolytic streptococcus (90%), Staphylococcus aureus, E. coli, Klebsiella pneumoniae, Proteus mirabilis, Enterobacter species, Enterococcus, Pseudo-monas aeruginosa, Peptostreptococcus, Clostridial species, Fusobacterium, Corynebacterium, and Bacteroides species.

Fewer than 6% of patients with NF suffer from Type 2 disease, where the infectious process is monobacterial and predominantly involves Group A beta-hemolytic streptococcus.

Despite their rarity, Clostridial infections must be considered in all NF infections. The suspicious rate should increase specifically when NF is associated with muscle necrosis or central nervous system manifestations. Risk factors include soft-tissue penetrating injuries, deep lacerations, crush injuries, compound fractures, and burns. Clostridial species are quite abundant in the environment and can contaminate any wound or fracture. Risk of gas gangrene will increase if the contaminated wound has a vascular injury and/or if the wound is closed. In recent years Clostridial NF has been reported in patients using black tar heroin. Clostridial species isolated from these cases include C. perfringens, C. histolyticum, C. novyi, C. sordellii, and C. septicum. Patients with gastrointestinal malignancies and lesions are also prone to NF caused by Clostridia septicum. Clostridial infections can cause significant tissue damage without any significant manifestation of erythema or inflammatory process. Quite interestingly, clostridial infections may recur spontaneously after months to decades. It is suggested that after the initial NF infection, the spores of the organism may reside in the tissue for indefinite amount of time. These cases are known as nontraumatic gas gangrene. The reactivation usually occurs in neutropenic patients or following gastrointestinal malignancies, diverticular diseases, radiation therapy to the abdomen, or after a local trauma and hematoma. C. septicum predominantly has been linked with this phenomenon.

Group A Streptococcus (GAS)

GAS infections are not a new disease. They can range from mild and non-invasive infections, such as impetigo, pharyngitis, and scarlet fever, to very severe invasive infections, such as NF, septicemia, and toxic shock syndrome. Since 1987, GAS has been identified as a cause of monobacterial NF infection. Furthermore, in the past two decades, the incidence of these infections has been increasing. A close association with varicella skin lesions has been proposed. NF caused by GAS is the most rapidly progressive and devastating form of NF infections. These infections can appear in clusters and are also known as flesh-eating disease. Approximately 50% of these infections are associated with toxic shock and multiorgan failure.62,63 The majority of the infected patients are younger individuals. In 2001, there were approximately 9,930 cases of invasive GAS infections. The prevalence of these infections in the general population is around 3.5/100,000; 5.9% of invasive GAS infections resulted in Streptococcal Toxic Shock Syndrome (STSS) and 6.7% in NF infections. The reported mortality rates for invasive GAS have been reported in a range of 30-70%.63 The disease specific mortality rates are 50% for STSS and 20% for NF.

GAS enters the soft tissue either through normal appearing skin and mucosa or a surgical incision. Patients suffering from a GAS-induced NF infection and/or myonecrosis may proceed to development of a concomitant STSS. This syndrome is associated with multiorgan failure and disseminated intravascular coagulopathy (DIC). The most commonly affected organs include the kidneys, lungs, and liver. The virulence of the GAS is related to the surface M proteins of the bacteria, specifically M types 1 and 3. These proteins can be identified in more than 80% of all GAS isolates.

The bacteria can invade the soft tissue directly. Most probably, the M proteins and hyaluronic acid capsule play a significant role in the invasiveness of GAS by protecting it from phagocytosis.64 Furthermore, GAS is in possession of tissue destructive enzymes, such as streptolysins O and S, DNase, streptokinase, proteinase, collagenase, lipase, and hyaluronidase. These enzymes allow liquefaction of bloody and purulent exudates, digestion of the connective tissue, and facilitation of rapid bacterial growth along the tissue planes. In addition to direct bacterial invasion, production and release of streptococcal pyrogenic exotoxins (SPEs) by GAS, play a very significant role in the pathogenesis of NF infections and STSS. These toxins, better known as superantigens (SAgs) A, B, and C, are common extracellular bacterial proteins that can be found in more than 90% of GAS isolates. Other streptococcal SAgs include streptococcal mitogenic exotoxin (SMW Z-1) and streptococcal pyrogenic exotoxin J (SPE-J). SAgs can cause activation and outpouring of the immune cells that induce synthesis and massive release and outpouring of cytokines. Tumor necrosis factors alpha (TNF-alpha) and beta (TNF-beta), and interleukins 1b, 2, and 6 (IL-1b, IL-2, and IL-6) are the most specific cytokines that are related to pathogenesis of STSS.65-70 It has been postulated that TNF-alpha is one of the prime mediators of shock.

Rare incidences of NF have been reported due to unusual organisms including Streptococci Groups B, C, and G, Streptococcus agalactiae, Erysipelothrix rhusiopathiae (encounter with fish or fish products),71 and Aeromonas hydrophila (water moccasin snake bites).72

Pathophysiology. Regardless of the species of the pathogenic organism and or classification of NF, the introduction of the bacteria into the subcutaneous tissue will be followed by the secretion of various toxins, including pyrogenic exotoxin A. These toxins stimulate the production of cytokines, which in turn damage the endothelial lining. The addition of free oxygen radicals and nitrous oxide worsens the tissue injury and damage. The resultant extravasation and leakage of serum into the extravascular space will increase the interstitial pressure, reduce the regional blood flow, generate tissue hypoxemia, and cause subsequent tissue damage and death. Furthermore, a secondary inflammatory vasculitis and arteriolar thrombosis will cause additional tissue injury and necrosis. In untreated cases, gangrene becomes apparent in 4-5 days and tissue sloughing is manifested in 8-10 days. At the end of the second week, the release of complex infections and toxic material into the bloodstream will produce sepsis and profound shock. Disseminated intravascular coagulopathy (DIC), acute respiratory distress syndrome (ARDS), depression of myocardial function, and multisystem organ failure will follow. Death usually occurs within 1-4 days following this stage of disease.

Clinical Presentation. In adults, the majority of NF occurs in the extremities. Conversely, most of the lesions occur on the trunk in young children (younger than 1 month of age). At the onset, NF infections present with the onset of an erythematous dermal patch and can be misdiagnosed as cellulitis. The sudden onset of a localized, deep-seated, severe, and unrelenting pain should alert the clinician to seek an alternative diagnosis. Pain even may be resistant to narcotic analgesics. Other suggestive clues include the very rapid progression of the disease despite the presence of appropriate antibiotic therapy, the disproportionate amount of pain compared to the extent of dermal findings, and toxic appearance. Patients may develop fever, chills, tachypnea, tachycardia, malaise, nausea, vomiting, diarrhea, generalized myalgia, dizziness, and confusion. These symptoms should be considered as red flags. The localized skin usually is shinny and tense, lacks a distinct margin, and spreads quickly. Crepitation may be present in 50% of type 1 NF infections. Type 2 NF infections usually do not generate any gas. Furthermore, in more than 50% of patients suffering from type 2 NF, the focal skin manifestations are absent at the time of presentation to a medical facility and only become obvious after the patient has developed shock. The erythematous patches transform into ecchymotic vesicular lesions which, in turn, convert into large purple bullae. These lesions are filled with a thin, watery, and foul-smelling liquid known as dishwater pus. The large bullae also may become hemorrhagic. The extent of blood loss even may cause hypovolemia and anemia. Subsequently, with the development of superficial fascia and fat necrosis, these bullae progress to a rapidly expanding graying green slough and a deep blue, purple, or black areola. The rupture of bullae may result in a dry black eschar. Although less common, lymphadenopathy and lymphangitis also may be present. Metastatic abscesses may occur in various organs, including the remote skin areas, brain, spleen, lung, and pericardium. Following the necrosis of the involved nerves, the excruciating pain disappears and the involved skin becomes anesthetic. If left untreated, patients develop a change in mental status followed by shock and decreased level of consciousness. Following this stage of disease, death occurs in few days. The highest mortality is encountered in the NF infections of head and neck, chest, and abdomen.

STSS presentation includes a sudden onset of fever, generalized erythematous rash, hypotension, change in mental status or level of consciousness, and multiorgan failure manifested as renal, liver, or pulmonary failure, coagulopathy, or soft-tissue necrosis (NF or myositis). Contrary to sepsis, acute renal failure occurs fairly early in STSS.

Diagnosis. Diagnosis is mainly a clinical one. High index of suspicion will assist the clinician in reaching the proper diagnosis. Fundamental to the successful treatment of NF infections is early diagnosis. Associated laboratory findings include leukocytosis with a left shift, anemia, thrombocytopenia, hypoalbuminemia, hypocalcemia, acidosis, hyponatremia, elevated blood urea nitrogen (BUN) and creatinine, and hyperbilirubinemia.

Diagnosis of STSS is based on the proposed case definition criteria of isolation of GAS from a normally sterile site (blood, CSF, pleural effusion, peritoneal effusion, pericardial effusion, or joint effusion), hypotension, and two or more of the following: renal failure, coagulopathy, liver failure, ARDS, generalized erythematous macular rash, myositis or soft tissue necrosis (NF). (See Table 8.) Cultures also should be obtained from the patient’s throat, rectum, and vagina. Cultures of the tissue, obtained from a skin biopsy, also can identify the pathogenic organism(s).

Table 8. Diagnostic Criteria for STSS

In highly suspicious cases, the definite diagnosis should be made surgically. The characteristic ease of separation of the skin from the subcutaneous tissue and the underlying necrotic gray fascia, due to suppurative or liquefactive necrosis, is a highly sensitive diagnostic sign. This is considered a positive finger test. This finding should mandate performance of a tissue biopsy, gram staining, culture, and/or frozen section procedures. In both streptococcal and clostridial NF infections, tissue biopsies would reveal a massive amount of tissue destruction in conjunction with vessels thrombosis, and a paradoxically very small number of leukocytes. Due to a high incidence of sampling errors, obtaining punch biopsies is not a viable diagnostic option.

In cases where the level of suspicion is low, imaging of the involved tissue is another diagnostic option. Computed tomography (CT) and MRI are valuable diagnostic imaging tools. Either modality can identify pockets of gas in the involved soft tissue and along the fascial planes. Other suggestive findings include asymmetrical fascial thickening and fascial stranding.

Treatment. Timely aggressive surgical debridement is the main therapeutic modality. To promptly abort the progression of the disease, all the dead and infected tissue and fascia must be removed until healthy, clean, pearly-gray fascia is identified in all margins of the wound. Once the diagnosis is seriously entertained, empiric antibiotic therapy also should be initiated immediately.

In Type 1 NF infections, which generally are caused by a polymicrobial flora, the standard therapeutic regimens should include provision of coverage for clostridia, streptococci, and Peptostreptococcus. One or a combination of the following antimicrobial agents should be used: Penicillin, ampicillin, third-generation cephalosporins, penicillinase resistant penicillins, clindamycin, aminoglycosides, and metronidazole. Imipenem and meropenem have the ability to inhibit the production of endotoxins by the gram-negative bacilli. (See Table 9.)

In Type 2 NF infections caused by GAS, clindamycin, erythromycin, and ceftriaxone have shown significant efficacy. Although the related mechanism is not fully understood, this superiority may be due to these antimicrobial agents’ ability to suppress M protein synthesis, inhibit toxin production, or their greater affinity for GAS penicillin-binding proteins. Clindamycin also is indifferent to the growth stage of the bacteria and resists the inoculum effect. Although still considered as an accepted therapeutic agent for GAS infections, use of penicillin as single agent is somehow discouraged in severe cases of type 2 NF. It has been shown that penicillin as well as beta-lactams are less effective in fulminant infections.73-75 Furthermore, the use of penicillin and other beta-lactam antibiotics actually may cause the release of additional toxins and worsen the course of Streptococcal toxic shock syndrome. Although the reported rate of streptococcal resistance to clindamycin in the United States is very low, considering the gravity of NF, this issue cannot be overlooked.62 As a result, as a safety measure, an expanded spectrum penicillin can be added to the clindamycin therapy regimen. (See Table 9.)

Table 9. Antimicrobial Agents in Treatment of NF Infections

All people who have been in contact with the patient suffering from an invasive GAS soft-tissue infection must receive prophylactic antibiotic therapy. Contact with an invasive GAS-infected patient increases the risk of acquisition of an invasive GAS infection by a factor of 20-60. Choices of antimicrobial regimen include penicillin, erythromycin, and clindamycin. Furthermore, GAS is highly transmittable and, since GAS infections are transmitted through a person-to-person route, eradication of the primary source of GAS will require therapy with penicillin plus rifampin. As a public safety and infectious control measure, cultures must be obtained from the nares, throat, skin, vagina, and rectum of all the involved health care providers and hospital staff. Any staff member with a positive culture must refrain from work until receiving at least 24 hours of antibiotic therapy. After completion of treatment, all identified carriers must undergo periodic cultures for up to one year.

If highly suspicious of a Clostridial NF infection, high doses of penicillin must be administered immediately. In penicillin allergic patients, clindamycin or metronidazole are antimicrobial agents of choice. Limb amputation also must be seriously considered. (See Table 9.)

Tetanus is an occasional complication of high-risk skin and soft-tissue infections, specifically NF infections that are associated with gangrene. In high-risk wounds, combined active and passive tetanus prophylaxis is indicated in non-immunized patients and in those whose immunization is considered outdated. These measures can be accomplished with the use of absorbed tetanus toxoid and tetanus hyper-immune globulin.71

Due to the inherent vascular insufficiency, vasculitis, thrombosis, and obstruction, the delivery and subsequent tissue levels of the prescribed antibiotics may be altered and render the therapy unsuccessful.

Use of hyperbaric oxygen therapy (HBO) has been proposed as an adjunct therapeutic modality. A summary of multiple clinical series revealed a significant reduction of mortality in patients undergoing HBO (16-23%) in comparison with patients who did not receive HBO (35-66%).76 For a successful therapeutic outcome, 95-100% oxygen should be delivered at pressures greater than 1.0 atmosphere absolute. These values increase the amount of available oxygen to the tissues by 1500%. High oxygen tensions enhance the antimicrobial activities and promote killing of the anaerobic organisms. HBO inhibits overall bacterial growth and potentially halts or reduces the production of the endotoxins. Furthermore, it strengthens and enhances the leukocytes’ activity. HBO causes angiogenesis, promotes production of collagen tissue, and improves the neovascularization process and the repair of the damaged endothelial lining. The interstitial edema also is diminished secondary to the vasoconstriction which results from HBO. Use of HBO must be seriously considered in all Clostridial infections. Logistically, the provision of HBO to a seriously ill and hemodynamically unstable patient is a very difficult and problematic endeavor. Other related concerns include risks of barotrauma, pneumothorax, and oxygen toxicity. This toxicity may result in ocular changes and progressive myopia, and a remote risk of generalized seizures.

Recent published reports have demonstrated a high rate of success in the salvage of limb and function in NF infections involving the hand. In NF infections of hand without liquefactive necrosis, the extensively exposed tendons, bones, and joints can be preserved. Following extensive surgical debridement and HBO, the exposed tendons, bones, and joints should undergo a delayed free muscle flap coverage surgical procedure and a subsequent skin graft repair. Muscle flap coverage will cause revascularization of vital structures, improves the delivery of antibiotics, and enhances leukocyte activities.77 Most patients will regain a satisfactory functional ability to extend all digits.77

Use of polyspecific immunoglobulin (IVIG) therapy also has been advocated as an adjunct therapeutic modality in NF infections caused by GAS, specifically those associated with STSS. IVIG inhibits T cell proliferation. It contains many antibodies that bind to circulating exotoxins and superantigens. Furthermore, IVIG down regulates tissue necrosis factor alpha (TNF-alpha) and interleukin 6 (IL-6). Additionally, by inhibiting the binding of superantigens to T-cell receptors, IVIG neutralizes mitogenic and cytokines-inducing activities of GAS superantigens A, B, or C. The first dose of IVIG must be administered in the first 1-2 days after the establishment of diagnosis. The recommended IVIG dosage ranges from 0.4-1 g/kg. 78 Due to potentially a larger volume of circulating endotoxins, higher doses must be administered in more severe diseases.

Wound care also is a very important aspect of management of NF patients. Reported successful use of maggots in non-surgical areas, i.e., neck, can be found in the recent medical literature.

Other therapeutic measures include use of heparin to combat vasculitis and venous thrombosis. Aggressive hydration and nutritional support are a necessity in the treatment of these patients. Due to the presence of a catabolic state, the daily calorie requirement of patients suffering from NF infections may exceed 2-3 times their normal levels. Nutritional support must be initiated in the first 24 hours of hospitalization. Ideally it should be provided through a combined parenteral and enteric route. Enteral feeds should be initiated as soon as tolerated by the patient to combat the enteric bacterial overgrowth. For proper and expedient wound healing, patients recuperating from NF infections require high amounts of protein, iron, and vitamins C and E.

Necrotizing Gangrene of the Perineum and Genitalia

Like other NF, these fulminant and potentially life-threatening infectious processes are exemplified by their relative rarity, insidious onset, and rapidly progressive nature. The uncommon nature of these events cause a significant diagnostic challenge. Furthermore, any delay in diagnosis will result in significant morbidity and mortality.

In 1883 Fournier, the French venereologist, provided the first full description of a gangrenous infection of the genitalia in a previously healthy male.79 As a result, the NF of the genitalia is still referred to as Fournier’s gangrene. Although the majority of these infections are encountered in men, 14% of cases of NF of the genitalia and perineum occur in women.80 In modern era, the most frequent sources of necrotizing gangrene of genitalia are anorectal, genitourinary infections, and cutaneous injuries of the perineum. Currently, most of these NF infections stem from infectious processes that initially are originated in the gastrointestinal tract (30-50%). In this subset of patients, 70% of infections are secondary to a perianal, intrasphyncteric, or ischial abscess.81 Other related causes may include gastroenteritis malignancies, minor anorectal biopsies, or other surgical procedures, thrombosed hemorrhoids, fissures, appendicitis, or diverticulitis.82,83 The second most common cause of NF infections of the genitalia initiate from a genitourinary site (20-40%).80 The contributing factors include urethral and epididymal infections and strictures; minor trauma to the urethra such as urethral instrumentation, catheterization, and other surgical procedures; urinary extravasation; urethral calculi; urogenital malignancies; and prostate massage or instrumentation.78,82 The third most common cause of these infections results from either infections or injuries of the perineum (20%).81 Even minor cutaneous injuries, such as insect or human bites, may be the cause. Other etiologies may include genital dermatological infections, injections, circumcision, vasectomy, or prosthetic insertions.80,85 In women, gender-specific etiologic factors include coital injuries, infections of vulva and Bartholin’s gland, episiotomies, pudendal blocks, and septic abortions.86,87 The risk factors for these NF infections include comorbidities such as malnutrition, immunosuppression, malignancy, renal failure, diabetes mellitus, and alcoholism.87-89 Mortality rates have been reported from 0-80% range.

Pathogenesis. The infectious process starts in the subcutaneous tissues of the genitalia and perineum and spreads along the subcutaneous planes. Subsequently, and as a result of the infection of deeper tissues, the infectious process will progress and proceed along the superficial fascia. The resultant catastrophic tissue necrosis is the result of obliterative endarteritis and thrombosis of the subcutaneous vessels. The inevitable bacterial colonization and overgrowth will be responsible for a declining tissue oxygen tension, decreased blood flow, and propagation of a relative anaerobic environment. This will promote the growth of anaerobic pathogens. The synergistic nature of this polymicrobial flora accentuates the fulminant nature of the infectious process. The progression of necrosis will eventually involve the fascia and larger arteries and veins. Leukocytic infiltration and fibrinoid thrombosis and necrosis of the larger vessels will accentuate the tissue damage. Excluding the clostridial infections, the extension of the necrotic process beyond the fascia and the involvement of the deeper tissues or muscle (myonecrosis) is rare. The progression of the disease easily can be explained by the anatomy of the fascia of the perineum, abdominal wall, and external genitalia. In 4-5 days the evidence of gangrene will be visible and in 8-10 days suppuration and necrosis will separate the infected tissue from the underlying viable tissue.

Infections often are polymicrobial in nature and usually involve at least 2-4 bacteria. Pathogenic organisms include Bacteroides, Coliforms, Staphylcoccus, Streptococcus, Peptostreptococcus, and anaerobic bacteria. Monobacterial aerobic infections, clostridial and non-clostridial gas-forming infections, and combination aerobic-anaerobic infections play an important role in anorectal NF infections. In genitourinary infections, the prominent bacteria include gram-negative bacilli, staphylococci, and streptococci. Normal skin flora, such as Staphylococci, are the dominant pathogens in the genesis of the cutaneous infections. The multibacterial synergism can be a product of various factors. These factors include gram negative bacilli’s cell wall lipopoly-saccharide endotoxins, and streptokinase, hyaluronidase, streptodornase, coagulase extracellular enzimes of streptococci and staphylococci, and heparinase of anaerobic bacteria. The metabolism of microorganisms produces cell toxic chemicals such as nitrogen, hydrogen, nitrous oxide, and hydrogen sulfide. Furthermore, these chemicals generate subcutaneous emphysema. Although the presence of gas in the soft tissue traditionally is associated with clostridial infections, bacteria such as Klebsiella, E. coli, Bacteroides, and Peptostreptococcus also may cause soft-tissue crepitance.

Not all of genitalia and perineal gangrene have an infectious origin. Vasoocclusive disorders such as vasculitis secondary to IgE hypersensitivity vasculitis, polyarteritis nodosa, and pyoderma gangrenosum also can cause gangrene of the genitalia and perineum.

Clinical Presentation. Most patients complain from an insidious onset of scrotal discomfort. With the progress of the disease, patients develop fever (60-80%), chills, and malaise. This is followed by the erythema, massive edema (80-100%) and tenderness (100%) of the scrotal skin. With the development of necrosis and infective neuritis, the pain may subside. Crepitus may also be present at this stage (60-70%). If left untreated, the disease invariably proceeds to shock, altered level of consciousness, and eventually death.

This disease can be confused by more frequently encountered disorders such as scrotal cellulitis, incarcerated hernia, and scrotal abscesses.

Diagnosis. The diagnosis is primarily made clinically. A high index of suspicion and awareness should assist the clinician in pursuing the diagnosis for these relatively rare diseases. Obtaining a full history, including history of any recent ano-rectal and genitourinary infection, trauma, or surgical procedures is very important. Ancillary diagnostic imaging tests include plain radiography, ultrasonography, and CT.

Plain radiographs of the abdomen and pelvis can identify presence of gas in the soft tissue even when they are not recognizable during physical examination. Ultrasonography has even a higher sensitivity for the presence of gas in the soft tissue. It easily can be easily utilized for the examination of the scrotum and intrascrotal structures, perirectal area, and abdomen. The role of CT of the abdomen and pelvis has not been fully investigated for the diagnosis of NF of genitalia and perineum.

Treatment. The NF of genitalia and perineum is a surgical emergency. Aggressive surgical debridement can be life- and organ-saving. The timeliness and extent of correct debridement has a direct correlation with the final outcome of therapy. Medical therapy and antimicrobial agents are utilized as an adjunct therapeutic modality. Aggressive fluid resuscitation and vasopressor agents should be administered in shock and sepsis. Blood and/or fresh frozen plasma transfusions may be required. Empiric, broad spectrum combination antibiotic therapy regimens should include penicillin (coverage of streptococci, clostridia, and some of anaerobics), gentamycin (coverage of gram-negative bacilli), and clindamycin (coverage of Bacteroides and other anaerobics). Semi-synthetic penicillins and third-generation cephalosporins could be used as an alternative for aminoglycosides. (See Table 9.)

References

1. Mayfield JA, Reiber GE, Sanders LF, et al. Preventive foot care in people with diabetes. Diabetes Care 1998;21:2161-2177.

2. National Institute of Diabetes and Digestive and Kidney Diseases. Diabetic neuropathy: The nerve damage of diabetes. Washington, DC: US Dept. of Health and Human Services; 1995.

3. Ramsey SD, Newton K, Blough D, et al. Incidence, outcomes, and cost of foot ulcers in patients with diabetes. Diabetes Care 1999; 22:382-387.

4. National Institute of Diabetes and Digestive and Kidney Disease. Feet Can Last a Lifetime: A Health Care Provider’s Guide to Preventing Diabetes Foot Problems. Washington, DC: US Dept. of Health and Human Services; 1997.

5. Blumer JL, Lemon E, O’Horo J, et al. Changing therapy for skin and soft tissue infections in children: Have we come full circle? Pediatr Infect Dis J 1987;6:117-122.

6. Caputo ZGM, Joshi N, Weitekamp MR. Foot infections in patients with diabetes. Am Fam Physician 1997;56:195-202.

7. Lipsky BA, Baker PD, Landon GC, et al. Antibiotic therapy for diabetic foot infections: Comparison of two parenteral-to-oral regiments. Clin Infect Dis 1997;24:643-648.

8. American Diabetes Association. Consensus development conference on diabetic foot wound care. Diabetes Care 1999;22:1354-1360.

9. Laing P. The development and complications of diabetes foot ulcers. Am J Surg 1998;176 (suppl 2A):11S-19S.

10. Lavery LA, Armstrong DG, Harkless LB. Classification of diabetes foot wounds. Ostomy Wound Manage 1997;43:44-53.

11. Pettit D, Wyssa B, Herter-Clavel C, et al. Outcome of diabetic foot infections treated conservatively: A retrospective cohort study with long term follow up. Arch Intern Med 1999;159:851-856.

12. Fergie JE, Purcell K. Community-acquired methicillin-resistant Staphylococcus aureus infections in south Texas children. Pediatr Infect Dis J 2001;20:860-863.

13. Pollack RA, Edington H, Jensen JL, et al. A human dermal replacement for the treatment of diabetic foot ulcers. Wounds 1997;9:175-178.

14. Schachner L, Taplin D, Scott GB, et al. A therapeutic update of superficial skin infections. Pediatr Clin North Am 1983;30:397-403.

15. Sacks JJ, Lockwood R, Hornreich J, et al. Fatal dog attacks, 1989-1994. Pediatrics 1996;97:891-895.

16. Weiss HB, Freidman DI, Coben JH. Incidence of dog bite injuries treated in emergency departments. JAMA 1998;279:51-53.

17. Sacks JJ, Kresnow M, Houston B. Dog bites: How big a problem? Inj Prev 1996;2:52-54.

18. 2003 Centers for Disease Control and Prevention. MMWR Morbidity and Mortality Weekly Report 2003;52:605-610.

19. Sacks JJ, Lockwood R, Hornreich J, et al. Fatal dog attacks, 1989-1994. Pediatrics 1996;97:891-895.

20. Blackman JR. Animal Bites. In: Rakel RE, Pedersen D, eds. Saunders Manual of Medical Practice. Philadelphia:WB Saunders;1996.

21. Avner JR, Baker MD. Dog bites in urban children. Pediatrics 1991;88:55-57.

22. Callahan ML. Human and animal bites. Top Emerg Med 1982;4:1-15.

23. Ahrenholz DH. Necrotizing soft tissue infections. Surg Clin North Am 1988;68:199-214.

24. Sacks JJ, Sattin RW. Dog bite related fatalities from 1979 through 1988. JAMA 1989;262:1489-1492.

25. Miller SJ, Copass M, Johansen K, et al. Stroke following Rottweiler attack. Ann Emerg Med 1993;22:262-264.

26. Ruskin JD, Laney TJ, Wendt SV, et al. Treatment of mammalian bite wounds of the maxillofacial region. J Oral Maxillofac Surg 1993;51:174-176.

27. Snyder KB, Pentecost MJ. Clinical and angiographic findings in extremity arterial injuries secondary to dog bites. Ann Emerg Med 1990;19:983-986.

28. Kizer KW. Epidemiologic and clinical aspects of animal bite injuries. JACEP 1979;8:134-141.

29. Lackmann GM, Tollner U. More on dog bite injuries [letter]. Pediatrics 1991;122:356.

30. Ordog GJ. The bacteriology of dog bite wounds on initial presentation. Ann Emerg Med 1986;15:1324-1329.

31. Rosen RA. The use of antibiotics in initial management of recent dog bite wounds. Am J Emerg Med 1985;3:19-23.

32. Griego RD, Rosen T, Orengo IF, et al. Dog, cat , and human bites: A review. J Am Acad Dermatol 1995;33:1019-1029.

33. Forlenza SW. Capnocytophaga: An update. Clin Microbiol Newslett 1991; 13:89-91.

34. Pers C, Gahrn-Hansen B, Frederiksen W. Capnocytophaga canimorsus septicemia in Denmark, 1982-1995: Review 39 cases. Clin Infect Dis 1996;23: 71-75.

35. Linton DM, Potgieter PD, Rpditi D, et al. Fatal Capnocytophaga canimorsus (DF-2) septicemia. A case report. S Afr J Med 1995;84:857-860.

36. Zook EG, Miller M, Van Beek Al. Successful treatment protocol for canine fang injuries. J Trauma 1980;20:243-247.

37. Aghababian RV, Conte JE. Mammalian bite wounds. Ann Emerg Med 1980;9:79-83.

38. Dire DJ. Cat bite wounds: Risk factors for infection. Ann Emerg Med 1991; 20:973-979.

39. Boenning DA, Fleisher GR, Campos JM. Dog bites in children: Epidemiology, microbiology and penicillin prophylactic therapy. Am J Emerg Med 1983;1:17-21.

40. Newcomer VD, Young EM. Unique wounds and wound emergencies. Derm Clinics 1993;11:715-727.

41. Hopman L, Stewart CE. Rabies. Emerg Med Serv 1986;May:22G-22J.

42. Wounds and Injuries of the Soft Tissues. In: Emergency War Surgery First United States Revision of Emergency War Surgery NATO Handbook. Washington, DC: US Government Printing Office;1975.

43. Levin JM, Talan DA. Erythromycin failure with subsequent Pasteurella multocida meningitis and septic arthritis in a cat bite victim. Ann Emerg Med 1990;19:1458-1461.

44. Brown CG, Ashton JJ. Dog bites: The controversy continues [editorial]. Am J Emerg Med 1985;3:83-84.

45. Cummins P. Antibiotics to prevent infection in patients with dog bite wounds: A meta analysis of randomized trials. Ann Emerg Med 1994;24: 535-540.

46. Bass JW, Vincent JM, Person DA. The expanding spectrum of Bartonella Infection: II. Cat scratch disease. Pediatr Infect Dis J 1997;16:163-179.

47. Dalton MJ, Robinson LE, Cooper J, et al. Use of Bartonella antigens for serological diagnosis of cat scratch disease at a national referral center. Arch Intern Med 1995;155:1670-1676.

48. Margileth AM. Update on cat scratch disease. Hospital Med 1989;Dec.: 61- 81.

49. Bass JW, Freitas BC, Freitas AD, et al. Prospective randomized double blind placebo controlled evaluation of azithromycin for treatment of cat scratch disease. Pediatr Inf Dis J 1998;17:447-452.

50. Arisoy ES, Correa AG, Wagner ML. Hepatosplenic cat scratch disease in children: Selected clinical features and treatment. Clin Infect Dis 199;28: 778-784.

51. Regnery RL, Childs JE, Koehler J. Infections associated with Bartonella species in persons infected with human immunodeficiency virus. Clin Inf Dis 1995;21:S94-S8.

52. Quinlan KP, Sacks JJ. Hospitalizations for dog bite injuries. JAMA 1999; 281:232-233.

53. Mermin J, Hoar B, Angulo FJ. Iguanas and salmonella marina infections in children: A reflection of the increasing incidence of reptile associated salmonellosis in the United States. Pediatrics 1997;99:399-402.

54. Kelsey J, Ehrlich M, Henderson SO. Exotic reptile bites. Am J Emerg Med 1997;15:536-537.

55. Paisley JW, Lauer BA. Severe facial injury to infants due to unprovoked attacks by pet ferrets. Department of Pediatrics, Denver General Hospital. Vol. 259, No. 13, April 1, 1988.

56. Basadre JO, Parry SW. Indications for surgical debridement in 125 human bites to the hand. Arch Surg 1991;126:65-67.

57. File TM, Tan JS. Treatment of skin and soft tissue infections. Am J Surg 1995;169:26S-33S.

58. Phair IC, Quinton DN. Clenched fist human bite injuries. J Hand Surg[Br] 1989;14:86-87.

59. Berlet G, Richard RS, Roth JH. Clenched fist injury complicated by methicillin-resistant Staphylococcus aureus. Can J Surg 1997;40:313-314.

60. Zhming LJ, Willke RJ, Rittenhouse BE, et al. Effect of linezolid versus vancomycin on length of hospital stay in patients with complicated skin and soft tissue infections caused by known or suspected resistant Staphylococci. Surg Inf 2003;4:57-70.

61. Chan ATY, Cleeve V, Daymond TJ. Necrotizing fasciitis in a patient receiving infliximab for rheumatoid arthritis. Pstgrad Med 2002;78:47-48.

62. Kaul R, McGreer A, Low DE, et al. Population based surveillance for group A streptococcal necrotizing faciitis: Clinical features, prognostic indications, and microbiological analysis for seventy-seven cases. Am J Med 1997;103: 18-24.

63. Stevens DL, Tanner MH, Winship J, et al. Severe group A streptococcal infections associated with a toxic shock-like syndrome and scarlet fever toxin A. N Engl J Med 1989;321:1-7.

64. Diazgranados CA, Bisno AL. Clues to early diagnosis of Group A Streptococcal necrotizing fasciitis. Infect Med 2001;18:198-206.

65. Hackett SP, Stevens DL. Streptococcal toxic shock syndrome: Synthesis of tumor necrosis factor in interleukin exotoxin A and streptolysin O. J Infect Dis 1992;165:879-885.

66. Fast D, Schlievert P, Nelso R. Toxic shock syndrome associated with staphylococcal and streptococcal pyrogenic factor production. Infect Immun 1989; 57:291-294.

67. Bonverte P, Heeg H, Cullen C, et al. Toxicity of recombinant toxic shock syndrome toxin 1 mutant toxin: Infection model of toxic shock syndrome. Infect Immun 1993;61:793-799.

68. Miethke T, Walh C, Heeg K, et al. T cell-mediated lethal shock triggered in patient with enterotoxin B: Critical role of tumor necrosis factor. J Exp Med 1992;175:91-98.

69. Stevens DL, Bryant AE, Hackett SP, et al. Group A streptococcal bacteremia: The organ failure. J Infect Dis 1996;173:619-626.

70. Hsueh P, Wu J, Tsai P, et al. Invasive group A streptococcal disease in Taiwan is not associated with pyrogenic exotoxin genes. Clin Infect Dis 1998; 26:584-589.

71. Simionescu R, Grover S, Shekar R, et al. Necrotizing fasciitis caused by Erysipelothrix rhusiopathiae. South Med J 2003;96:937-939.

72. Angel MF, Zhang F, Jones M, et al. Necrotizing fasciitis of the upper extremity resulting from a water moccasin bite. South Med J 2002;95: 1090-1094.

73. Stevens DL, Bryant AE, Hackett S. Suppression on mononuclear cell synthesis of tumor necrosis factor by clindamycin. Program and abstracts of the European Conference of Toxic Shock Syndrome; Sept. 10-12, 1997; London. Abstract P23.

74. Majeski, JA, John JF. Necrotizing Soft Tissue Infections: A Guide to Early Diagnosis and Initial Therapy. South Med J 2003;96:900-905.

75. Stevens SL. Skin and soft tissue infections. Infect Med 2003;20:483-493

76. The Working Group on Severe Streptococcal Infections. Defining the Group A Streptococcal toxic shock syndrome: Rationale and consensus definition. JAMA 1993;269:390-391.

77. Yuen JC, Feng Z. Salvage of limb and function in necrotizing faciitis of the hand: Role of hyperbaric oxygen treatment and free muscle flap coverage. South Med J 2002;95:255-257.

78. Cowley MJ, Briggs BM, Haith LR, et al. Intravenous immunoglobulin as adjunctive treatment for streptococcal toxic shock syndrome associated necrotizing faciitis: Case report and review. Pharmacotherapy 1999;19: 1094-1098.

79. Fournier JA. Gangrene foudroyante de la verge. Medecin Pratique 1883;4:589.

80. Stephens BJ, Lathrop JC, Rice WT, et al. Fournier’s gangrene: Historic (1764-1978) versus contemporary (1979-1988) differences in etiology and clinical importance. Am Surg 1993;59:149.

81. Flanigan R, Kursh E, McDougal W, et al. Synergistic gangrene of the scrotum and penis secondary to colorectal disease. J Urol 1978;119:369.

82. Cunningham BL, Nivatvongs S, Shons AR. Fournier’s syndrome following anorectal examination and mucosal biopsy. Dis Colon Rectum 1979;22:51.

83. Paty R, Smith AD. Gangrene and Fournier’s gangrene. Urol Clin North Am 1992;19:149.

84. Walker L, Cassidy MT, Hutchinson AG, et al. Fournier’s gangrene and urethral problems. Br J Urol 1984;56:509.

85. Chantarasak ND, Basu PK. Fournier’s gangrene following vasectomy. Br J Urol 1989;61:538.

86. Ahrenholz D. Necrotizing soft tissue infections. Surg Clin North Am 1998; 68:199.

87. Roberts H, Hester L. Progressive synergistic bacterial gangrene arising from an abscess of the vulva and Bartholin’s gland duct. Am J Obstet Gynecol 1972;114:285.

88. Berg A, Armitage JO, Burns CP. Fournier’s gangrene complicating aggressive therapy for hematologic malignancy. Cancer 1986;57:2291.

89. McKay TC, Waters WB. Fournier’s gangrene as the presenting sign of an undiagnosed human immunodeficiency virus infection. J Urol 1994;152:1552.