Bites

Bite wounds are common injuries caused by a wide variety of domestic and wild animals, as well as humans.¹ The American Pet Products Association 2017–2018 survey reported the US pet dog and cat ownership number at approximately 184 million.² Bites occur in 4.7 million Americans each year³ and account for 800,000 medical visits, including approximately 1% of all emergency department visits.⁴ There were 1610 animal-related fatalities in the United States during the period of 2008 to 2015. The majority of human fatalities were from farm animals (e.g., cattle and horses), insects (hornets, wasps, and bees), and dogs.⁵ Most dog bites (85%) are provoked attacks by either the victim's own pet or a dog known to the victim and occur during the warm weather months.⁶ Bite wounds are often to the extremities, especially the dominant hand. Facial bites are more frequent in children younger than 10 years of age and lead to 5 to 10 deaths per year, often because of exsanguination.⁷ Larger dogs can exert more than 450 lb/in³ of pressure with their jaws, which can lead to extensive crush injuries.

The bacteria associated with bite infections may come from the environment, the victim's skin flora, or most frequently, the oral flora of the biter, which can also be influenced by the microbiome of the biter's ingested prey and other food (Table 315.1). Patients who present early after an incident often do not have an established infection and are usually concerned about crush injuries, care of disfiguring wounds, or the need for rabies or tetanus immunization.⁶ These uninfected wounds are frequently contaminated with multiple strains of aerobic and anaerobic bacteria, similar to the spectrum found in documented bite infections. Between 2% and 30% of wounds will become infected and, rarely, may require hospitalization.^8–10,11,12 Patients presenting late, often longer than 8 hours after injury, usually have established infection.^6,8,10,11,13 Infection is manifested as localized cellulitis or abscess with gray and malodorous purulent discharge.¹⁴ Fever, regional adenopathy, and lymphangitis occur in the minority of patients. Bites involving bones and joints may result in tenosynovitis, septic arthritis, and osteomyelitis. Chronic pain in a joint with limited range of motion may be suggestive of infection within a joint or bony structure.

Rarely, overwhelming sepsis, endocarditis, meningitis, or brain abscesses may develop after a bite injury. Fatal infections caused by Capnocytophaga canimorsus in association with asplenia or liver disease have been noted.^15–17 This organism may be difficult to isolate and identify and may require up to 14 days of incubation to grow on blood cultures. In addition, it has the potential capacity to escape the host immune system by both passive and active mechanisms.¹⁸ It is generally susceptible to penicillin, cephalosporins, and fluoroquinolones but variably resistant to aztreonam and aminoglycosides (Table 315.2).¹⁹

Individuals with immunocompromising conditions, including chronic corticosteroid use, and those with preexisting edema of an extremity are more prone to severe infections and complications.

Dog and cat bite wound infections are considered to be predominantly related to the microbiology of their oral flora.^{6,9,13,14,20–22} Although most attention has focused on Pasteurella multocida, the spectrum of organisms associated with dog and cat bite wound infections is much greater. Holst and colleagues²³ noted the following distribution of 159 Pasteurella strains isolated over a period of 3 years from human infections, mostly from bite wounds: Pasteurella multocida subsp. multocida (60%), which was the isolate in all bacteremia cases; Pasteurella multocida subsp. septica (13%), which has a greater prevalence in cats than in dogs and may have a preferential affinity for the central nervous system; Pasteurella canis biotype 1 (18%), which was isolated exclusively from dog bite wound infections; Pasteurella stomatis (6%); and Pasteurella dagmatis (3%), which may cause systemic infections. A study of 107 infected dog and cat bite wound infections showed that 75% of cat bites grew Pasteurella species on culturing (P. multocida subsp. multocida, 54%), as did 50% of dog bites (P. canis, 26%; P. multocida subsp. multocida, 12%).¹⁴ In this study, other common aerobic organisms isolated from infected dog and cat bite wounds included Streptococcus, Staphylococcus species (including Staphylococcus aureus), and Neisseria species. Anaerobic organisms, when present, were almost always in mixed infections with aerobic organisms and commonly included Fusobacterium, Porphyromonas, and Prevotella species in dog bites and Fusobacterium, Porphyromonas, and Bacteroides species in cat bites.¹⁴

Table 315.1 lists common pathogens found in dog and cat bite wound infections. Staphylococcus intermedius is coagulase positive, can be mistaken for S. aureus, and is fourfold more common in canine flora^24,25 but possesses β-galactosidase activity, which differentiates it from S. aureus. It may masquerade as methicillin-resistant S. aureus (MRSA) owing to false-positive rapid penicillin-binding protein 2a latex tests,²⁶ although an increasing number (approximately 30%) of isolates may be resistant to oxacillin. MRSA has been cultured from a variety of companion animals, including cats, and has been documented to be transmitted from a healthy pet cat to humans, and the human strains and feline strains are indistinguishable.^27,28 MRSA may be a potential causative secondary invader, especially in patients who are not responding to initially administered antibiotics that often do not exhibit activity against MRSA and in those known to be colonized with MRSA.

Capnocytophaga canimorsus¹⁷ is difficult to grow on most routine solid media but can grow on chocolate agar and heart infusion agar with 5% rabbit blood when incubated in CO₂ and a variety of liquid media, including BACTEC aerobic medium (Becton, Dickinson and Company, Franklin Lakes, NJ).¹⁶ This species can be differentiated from other Capnocytophaga species by the presence of positive oxidase and catalase reactions.¹⁷ Centers for Disease Control and Prevention (CDC) group DF-2–like strains have been classified as Capnocytophaga cynodegmi. CDC group M-5 has been classified as Neisseria weaveri and has been associated with dog bites.²⁹ CDC group EF-4a is now called Neisseria animaloris, and EF-4b is Neisseria zoodegmatis. Haemophilus felis, initially identified as Aggregatibacter (Haemophilus) paraphrophilus, requires factor V and CO₂ for growth and is common in cat nasopharyngeal flora.³⁰

Bergeyella (previously designated as Weeksella) zoohelcum has been associated with bite cellulitis, sepsis, and meningitis.³¹ Other new aerobic species include Neisseria canis from a cat bite,³² Flavobacterium group IIb–like isolates from a pig bite,³³ Actinobacillus lignieresii and Actinobacillus equi–like bacterium from horse bites,³⁴ and CDC group NO1, a nonoxidative gram-negative rod³⁵ different from Acinetobacter species associated with dog and cat bites. Orf virus infection has been transmitted by a sheep bite.³⁶ When appropriate culturing techniques are used, anaerobes are isolated in up to 70% of animal bite wounds, almost always in mixed culture.^6,11,13 Approximately 50% to 60% of cat and dog bite wounds contain Bacteroides tectus, Prevotella heparinolytica, Prevotella zoogleoformans, Prevotella bivia, Porphyromonas gingivalis, Porphyromonas canoris, Fusobacterium nucleatum, and Peptostreptococcus anaerobius.^37–40 Fusobacterium canifelinum is an intrinsically fluoroquinolone-resistant species isolated from dog and cat bites.⁴¹

Little difference has been noted in the types of bacteria isolated from noninfected wounds seen early and infected wounds seen later.⁶ All moderate-to-severe bite wounds, except those not clinically infected and more than a few days old, should be considered contaminated with potential pathogens.

Wounds inflicted by cats are frequently scratches or tiny but somewhat deep punctures located on the extremities and are at higher risk of becoming infected.⁴² Deep puncture wounds over or near a joint, especially on the hands, may result in osteomyelitis and septic arthritis. Pasteurella multocida has been isolated from 50% to 70% of healthy cats and is a frequent pathogen in cat-associated wounds.^14,23,42 Erysipelothrix rhusiopathiae has also been isolated from cat bite wounds.¹⁴ Cougar, tiger, and other feline bites also yield P. multocida.³⁴ Tularemia has likewise been transmitted by cat bites.⁴³ People are also bitten by a variety of other animals, including unusual domestic pets, farm animals, wild animals, aquatic animals, and laboratory animals.^4,44–47 Monkey bites cause more swelling and infection than do many other animal bites.⁴⁸ Old World monkeys may transmit a potentially lethal subtype B virus (herpesvirus simiae; see Chapter 141).⁴⁹ Case series or reports of various animal bites—including terrestrial mammals (e.g., monkeys, bears, pigs, ferrets, horses, sheep, Tasmanian devils); reptiles (e.g., snakes, Komodo dragons, lizards, iguanas, alligators, crocodiles); rodents (e.g., rats, guinea pigs, hamsters, prairie dogs); swans; and sharks with unusual isolates—have been reported.^50–53

Management of Animal Bites

Table 315.3 notes the elements for the treatment of animal bite wounds. The most problematic elements of the management of wounds that are seen early include the following:

1. 1. The use of preemptive antibiotics in wounds that are seen early but are not infected. As a routine, antimicrobial therapy is not advocated in every uninfected animal bite injury. However, because 85% of such wounds harbor potential pathogens and one cannot reliably predict which wounds will become infected, selected wounds are best treated with oral antimicrobial therapy with agents active against common bite pathogens (see Table 315.2) for 3 to 5 days. Recommendations about patient selection vary and are based on limited evidence.⁵⁴ In general, 3 to 5 days of preemptive antimicrobial therapy is advocated for acute moderate-to-severe injuries, especially those to the hand or face, in the presence of preexisting edema, and in immunocompromised hosts.⁵⁵ Longer courses may be considered for injuries that are severe or those penetrating joints or bony structures.
2. 2. The decision to suture the wound. The role of primary wound closure in animal bites remains controversial. The literature on this topic has a number of limitations such as the time from injury to presentation, the extent and location of injuries, wound closure methods, and administration of antibiotics. For cosmetic and functional reasons, facial and neck wounds and extensive large wounds, especially those overlying the joints, are difficult to leave open and are usually sutured after irrigation coupled with antimicrobial prophylaxis. Limited studies are available to determine whether the risk of infection is increased with primary wound closure in animal bites.^56–58,59 In one study of 345 dog bite wounds, puncture wounds and wounds that were closed were more likely to become infected.⁵⁸ In a randomized controlled trial of primary closure versus nonclosure of dog bite wounds, the infection rate between the two groups was similar but primary suturing improved cosmetic appearance.⁵⁹ It is our experience and recommendation that, if cosmetically or functionally reasonable, small bite wounds can be left open and allowed to close by secondary intention. Larger wounds, especially those overlying the face, neck, and joints, may be sutured or loosely approximated and closed by delayed primary closure.

The most common causes of therapeutic failure are the following:

1. 1. Failure to elevate the affected edematous extremity.
2. 2. Inadequate antimicrobial therapy (see Table 315.2). Most fastidious animal pathogens are susceptible to penicillin or amoxicillin. Because of resistance of certain bacteria, including P. multocida, first-generation cephalosporins, dicloxacillin, erythromycin, clindamycin, and metronidazole should be avoided as a sole antimicrobial agent; however, they can be used in combination with antibiotics that have activity against Pasteurella species. In vitro data suggest that some fluoroquinolones (ciprofloxacin, levofloxacin, and moxifloxacin), trimethoprim-sulfamethoxazole, ertapenem, and second-generation oral cephalosporins (e.g., cefuroxime) are active against many bite isolates.^60,61–68 Ceftaroline, an anti-MRSA cephalosporin, has activity against Pasteurella species and other aerobic gram-positive isolates recovered from animal bite wounds.^69,70,71 Empirical regimens for marine- and freshwater-acquired infection should cover Vibrio and Aeromonas species, respectively, with agents such as third-generation cephalosporins (e.g., cefotaxime) and fluoroquinolones.
3. 3 Failure to recognize osteomyelitis and septic arthritis. Long-standing pain or continuous diminished range of motion of a joint near a bite wound may be an indication of complications such as osteomyelitis or septic arthritis. Pain out of proportion to the severity of the wound may be an indicator of potential periosteal or joint penetration.

Venomous snakes, usually vipers (rattlesnakes, copperheads, cottonmouths, or water moccasins), bite approximately 8000 people in the United States yearly, and 5 or 6 of these die, usually children or the elderly, who receive either no or delayed antivenom therapy.⁷² The majority of bites occur in young men in the southwestern United States between April and September.⁷² Envenomation can cause extensive tissue destruction and devitalization that predisposes to infection from the snake's normal oral flora. Sparse data exist on the incidence and bacteriology of snakebite infections. In rattlesnakes, the oral flora appears to be fecal in nature because the live prey usually defecates in the snake's mouth coincident with ingestion. Common oral isolates include Pseudomonas aeruginosa, coagulase-negative staphylococci, and Proteus and Clostridium species.^73,74 Other potential pathogens isolated from rattlesnakes’ mouths have included Bacteroides fragilis and Salmonella enterica subsp. arizonae and S. enterica subsp. diarizonae (Salmonella groups IIIa and IIIb, respectively). Crotalus rattlesnake venom has innate, broad activity against aerobic gram-positive and gram-negative bacteria but not against anaerobes.^75,76 The role of empirical antimicrobial therapy for noninfected wounds is not well defined.

Lip Wounds and Paronychia

Wounds of the lips and paronychia and infections of the structure surrounding the nails account for most self-inflicted bite wounds. Paronychia is more frequent in children who suck their fingers and results from direct inoculation of the oral flora into the fingers. Brook⁷⁷ took cultures from 33 children with paronychia. Aerobes and anaerobes were each found in pure culture in 27% of cases, and mixed infection was found in 46% of cases. The most frequent aerobic organisms isolated were viridans streptococci, group A streptococci, S. aureus, Haemophilus parainfluenzae, Klebsiella pneumoniae, and Eikenella corrodens. The most frequently isolated anaerobic bacteria were Bacteroides species, Fusobacterium species, and gram-positive cocci. Therapy should include drainage, antibiotics for moderate-to-severe infections, and avoidance of further bacterial contamination.

Human Bites

Human bites generally have higher complication and infection rates than do animal bites. In addition, transmission of human immunodeficiency virus, hepatitis B virus, or hepatitis C virus must be considered. Occlusional human bites may affect any part of the body but most often involve the distal phalanx of the long or index fingers of the dominant hand. About 10% to 20% of wounds are “love nips” to the breasts and genital areas.^20,78,79 Bites to the hand are often deep and more frequently become infected than do bites to other areas.⁸⁰ Bites may also be caused by, or be harbingers of, child abuse.⁸¹

Important prognostic factors for the development of infection include the extent of tissue damage, the depth of the wound and which compartments are entered, and the pathogenicity of the inoculated oral bacteria.^82–85 The typical patient is a young male who is assaulted by another young male. The first infectious symptoms occur approximately within the first 24 hours after injury, but patients often do not seek medical care until approximately a day or two after the incident.⁸⁵ Viridans streptococci, especially Streptococcus anginosus, are the most common wound isolates. Staphylococcus aureus infection occurs in 30% to 40% of wounds and is usually present in patients who present 3 to 4 days after the injury and have attempted self-débridement of the wound. There have been few reports of MRSA isolated from clenched-fist injuries.^86,87 Other reported isolates have included Haemophilus influenzae, H. parainfluenzae, Aggregatibacter (Haemophilus) aphrophilus, Aggregatibacter (Haemophilus) paraphrophilus, Klebsiella species, Enterobacter cloacae, Prevotella and Peptostreptococcus species, and F. nucleatum.^82,85 Up to 45% of the anaerobic gram-negative bacilli isolated from human bite wounds may be penicillin resistant and β-lactamase positive.^62,88 Candida species were found in 8% of patients in one study, although their pathogenicity was not determined.⁸⁵

Management of Human Bites

Aerobic and anaerobic cultures should be obtained for infected wounds. Wounds should be irrigated with water and, if necessary, surgically débrided.⁸⁹ As much as possible, immobilization of the affected area, including splinting if necessary, and elevation should be instituted.

Initial antimicrobial therapy should provide coverage for Staphylococcus, Streptococcus, Eikenella, and anaerobic organisms such as Prevotella, Fusobacterium, and Veillonella species. Patients who present early with uninfected wounds may also be considered candidates for shorter duration of antimicrobial therapy. Amoxicillin-clavulanic acid as monotherapy or combination therapy with metronidazole plus trimethoprim-sulfamethoxazole can be used for the treatment of human bites. First-generation cephalosporins, dicloxacillin, erythromycin, clindamycin, and metronidazole are not effective as monotherapy because of inactivity of these agents against E. corrodens. In addition, first-generation cephalosporins, dicloxacillin, and erythromycin do not provide adequate activity against culprit anaerobic organisms.

Many patients (32% in one study⁸⁵), especially those with extensive deep hand injuries, require hospitalization. Plain radiographs can be considered for wounds in close proximity to joints and bones. Similar to animal bite wounds, we do not recommend primary closure of small wounds, even for uninfected human bite wounds, especially those on the hands. If possible, approximation of the wound margins with adhesive tape or delayed primary closure (3–5 days) can be alternatives to primary closure in certain circumstances. Facial wounds may present a special situation because of the possibility of scarring and disfigurement and may require primary closure.

Clenched-Fist Injuries

“Clenched-fist” injuries are traumatic lacerations that occur when one person strikes another in the mouth with a clenched fist. These injuries are most common over the third and fourth metacarpophalangeal joints of the dominant hand, but they may also occur over the proximal interphalangeal joints. Complications of this type of injury often include septic arthritis or osteomyelitis. These lacerations are commonly only 12 to 14 mm long but, despite their innocuous appearance, frequently lead to serious complications because of the proximity of the skin over the knuckles to the joint capsule and the potential spread of infection into subcutaneous tissues and web spaces.

Typically, patients who sustain a clenched-fist injury do not take good care of their wound, ignore it until a day or more after the injury, and then awaken with a painful, throbbing, and swollen hand. The swelling usually spreads proximally but not distally and results in decreased range of motion. A purulent discharge is often present. Lymphangitis, adenopathy, fever, or other signs of systemic infection are infrequent.

The bacteriology of clenched-fist injuries is similar to that of human bites and usually consists of the normal human oral flora.^10,82,90 Viridans streptococci, especially S. anginosus, are the most frequent isolates, but S. aureus may be present in 20% to 40% of cases. Anaerobic bacteria can be recovered in more than 55% of clenched-fist injuries, including Prevotella species, F. nucleatum, and Peptostreptococcus (including Finegoldia magna). Eikenella corrodens is an often overlooked but especially important pathogen in clenched-fist injury infections.^90–93 It has a prevalence rate of 59% in human gingival plaque⁷⁹ and may be isolated in 25% of clenched-fist injuries.⁹¹ It can act synergistically with viridans streptococci and is a common cause of osteomyelitis. Although E. corrodens is susceptible to penicillin, it is resistant to penicillinase-resistant penicillins, clindamycin, and metronidazole and is variably resistant to cephalosporins.^62–66

Initial physical examination is often limited because the majority of patients are in severe pain and unable to fully comply with the examination. Management should include irrigation and débridement when necessary. Elevation of the injured extremity is helpful in reducing associated swelling. Plain radiographs can be obtained to rule out fracture and to serve as a baseline for assessing osteomyelitis. Tetanus immunization should be administered when indicated. Secondary débridement to remove necrotic tissue or to drain abscesses may be required.

Empirical antimicrobial therapy should include coverage for Staphylococcus, Streptococcus, Eikenella, and anaerobes, including Prevotella, Fusobacterium, and Veillonella species. Failure of first-generation cephalosporins and penicillinase-resistant penicillins, when used alone, has been reported and is often due to their inactivity against E. corrodens.^91–96 If resistant gram-negative rods are isolated, therapy should be altered according to the results of culture. What role β-lactamase–positive Prevotella and Porphyromonas species will have in the selection of antimicrobial therapy has not been specifically determined.