Mammalian bites are fairly common presentations in the emergency department (ED). In fact, half of all Americans will be bitten by an animal or another human during their lifetime.1,2 Not surprisingly, domestic cats and dogs are responsible for most bites. That said, emergency clinicians must be prepared to address bites from all types of animals, including exotic and wild species. Bite injuries can cause morbidity from pain, cosmetic disfigurement, poor functional outcome, and infectious complications. The emotional impact to patients and their families must also be managed. Closure techniques and the treatment of clinically apparent infections are relatively straightforward. The major controversies surrounding mammalian bites involve their initial management. Should patients with bite wounds be given antibiotics prophylactically at presentation? Should bite wounds be closed primarily, or does this procedure increase the risk of infection? Although the examples of mammalian bites given in the vignettes may appear extreme, emergency clinicians will likely be confronted with these types of treatment decisions multiple times during their career.
The emergency department is in chaos, and your chart rack is full. You pick up the next chart and are relieved to see the chief complaint: "Scratches on hand and leg." However, your relief turns to dismay when you see a 15-year-old boy in the room with his very concerned mother and you hear the following history: The patient was carrying the family cat into the house when the pet became frightened by an unleashed dog. The cat bit the patient on the left hand before running into the bushes. The dog ran after the cat, with the patient in hot pursuit. As the patient retrieved his cat, the dog bit him on the right calf. The patient quickly sought out the dog's owner, a local homeless man. When approached by the patient, the man rudely denied responsibility for the dog's actions. Still a novice at conflict resolution skills, the patient argued with the dog's owner, eventually punching him in the face and sustaining a wound to his third MCP joint.
The cat is healthy and has received all of its shots. The histories of the dog and the dog's owner, however, are unknown. The patient's mother just wants "some antibiotics and a tetanus shot." How should each of the bites be treated? Is there concern for transmission of HIV from a potential human bite wound? Should the patient receive antibiotic prophylaxis, viral prophylaxis, or both?
Your next patient is a healthy but upset 22-month-old child accompanied by his mother, who is clearly worried. The patient was playing with the family's new kitten, which became agitated and bit the toddler on the left forearm. The patient started running to his mother in the next room but tripped on the carpet and fell. As he landed, he accidentally bit himself on his right arm. How should each of these bites be treated?
Ovid MEDLINE®, CINAHL®, and PubMed databases were searched for all articles on mammalian bites published between 1995 and 2008. Search terms included mammalian bites, animal bites, bite wounds, cat bites, bites, pediatric wounds, and bites and stings. A search of the Cochrane Database of Systematic Reviews produced a single review of 8 studies on mammalian bites that was updated in 2001.3 In addition, the clinical report and policy statement databases of the American College of Emergency Physicians and the American Academy of Pediatrics were reviewed, and one relevant report was found. A search of the Agency for Healthcare Research and Quality's online National Guideline Clearinghouse™ also yielded one relevant report. These articles served as the basis for further literature searches, which were performed manually. In all, approximately 500 articles were reviewed, and 122 are included here.
Literature on the management of pediatric mammalian bites largely consists of case reports, retrospective reviews, and extrapolations from reviews of the adult population. Few reviews have focused on children. Studies containing adult data were included in this article if no comparable pediatric data were available or if the studies contained information on some pediatric patients. Case reports highlighting uncommon but clinically important complications were also included. The nature of this topic does not lend itself to randomized placebo-controlled clinical trials in the pediatric population.
The true incidence of mammalian bites is unknown because many victims do not seek treatment for minor injuries and mandatory reporting is not in effect in all states. However, it is estimated that each year nearly 2 million bites result in medical treatment in the United States, with nearly half of these presenting in the ED.4 In all, the evaluation and treatment of bite wounds account for about 1% of all visits to EDs.2 This number is expected to rise as animals and humans increasingly share space.
In most cases, animal bites are attributed to house pets.2 In one study on animal aggression, 89% of attacks were caused by dogs, 8% by cats, and 3% by other species.5 The study indicated that most injuries were single bites and that children younger than 15 years were 1.4 times more likely than other age groups to be attacked. Reports from the U.S. Department of Health and Human Services also indicate that male children are at greater risk of being bitten than female children.6 In addition, children are more likely than adults to sustain head and neck wounds because of their smaller stature. The majority of fatalities due to bite wounds occur in children.7 Traumatic encounters with mammals result in substantial economic impacts as well. Estimated costs of homeowners liability claims related to bites approach $1 billion annually in the United States, while ED care for dog and cat bites costs more than $850 million.2,8 These figures do not include lost wages or added expenditures for worker's compensation claims.
A review of the major categories of mammalian bites follows.
The relationship between man and dog goes back 12,000 years.9 According to the American Pet Product Association, Americans owned 73.9 million dogs in 2005-2006.2 Given this number, it is not surprising that up to 4.5 million dog bites occur annually in the United States.9 Nevertheless, only 1 in 20 dogs will bite a human.1 The majority of dog bites are minor and do not lead to hospital visits, and those patients who do seek medical care usually have only minor injuries.10 However, dog bites still account for 80% to 90% of all bite wounds in patients presenting for medical care, and some can be severe, with more than a dozen fatal dog bites reported annually in the United States.11
Children warrant special consideration in the setting of dog bites. Research demonstrates that they are particularly susceptible to this type of injury, with the highest rate of dog bites occurring in children aged 5 to 9 years.2 In a 2001 Centers for Disease Control and Prevention (CDC) report, the age group at highest risk was 0 to 9 years.14The incidence was especially high in young boys. The overall rate of ED presentation for dog bite wounds was 12.9 per 10,000 people. However, the rate for boys aged 5 to 9 years was nearly 5 times higher. Data from the CDC indicated that 369,245 bites were reported to U.S. EDs in 2001. The incidence was 293.2 bites per 100,000 boys versus 216.7 per 100,000 girls. However, this sex-related difference disappeared after age 14 years.
The type of injury caused by dogs is related to the animal's anatomy. Canine teeth are not sharp and often cause tears and avulsions, although punctures can occur. Some large-breed dogs can inflict severe injury because of the tremendous crushing force of their bite. When biting, some dogs can exert between 200 and 450 psi, enough power to perforate sheet metal.1 Pit bull bites can exceed 450 psi, and trained police dogs can exert a biting force of 1500 to 2000 psi.12 Dog bites can break bones, including the skull, and penetrate body cavities.13 Therefore, significant dog bites should be initially managed as multisystem trauma, especially when the patient is a small child.
Because of the substantial biting force exerted by trained police dogs, injuries associated with these bites differ significantly from those observed in civilian dog bites. In a study of 705 dog bites sustained by jail ward patients in Los Angeles between 1988 and 1995, 19.3% involved complications. These included fractures and cortical penetration in 4%, vascular injuries in 7%, nerve injury in 1.9%, and infections in 5%, as well as tendon injuries, and joint disruption.12 ED clinicians should be aware of the increased risk of these injuries in patients brought in with police dog bites.
The anatomic location of dog bites in children is of particular concern. Whereas adults sustain bites to the extremities, children are more likely to sustain bites to the head or neck, which have greater cosmetic consequences.9,15 Weiss et al reported in JAMA in 1998 that 73% of dog bites in patients aged 0 to 9 years were to the face versus only 30% of dog bites in the population as a whole.9 Differences in bite location among age groups were also apparent in other reports.1,14 According to the CDC, in the overall population, 71% of dog bites in 2001 were to an extremity, whereas 22% were to the head or neck.14 However, for those 4 years and younger, 65% of dog bites were to the head or neck. In those 15 years and older, 86.2% were to an upper extremity.14 (See Table 1.)
As mentioned previously, dog attacks are occasionally fatal, and children younger than 10 years are grossly overrepresented in these cases. Sacks et al reported in Pediatrics that 109 dog bite-related deaths occurred in the United States between 1989 and 1994, and 56.9% were younger than 10.7 The authors examined the breeds most frequently implicated in the fatalities. Pit bulls were responsible for 28.6% of fatal bites in which a breed could be identified. Rottweilers and German Shepherds are also associated with a high risk of fatal injuries. (See Table 2.) Notably, there was only one fatal bite from a police or guard dog during that time period. Prevention of these attacks is not simply an issue of keeping children away from strange dogs. In fact, the bite victim is familiar with the offending animal in 82% of cases.4 Stray dogs are rarely implicated in fatal attacks. According to Sacks et al, among 82 bite cases where the location of the fatal attack was known, 63 bites occurred on the dog owner's property.7 Among the 41 cases where the sex of the attacking dog was noted, 25 were males. Of the 20 dogs in the study whose neuter status was known, only 1 castrated male dog was responsible for a fatal bite.7 Given these data, people should be encouraged to neuter dogs that are not used for breeding. As with bites from humans and other types of animal, dog bites carry the risk of infection. The average dog's mouth harbors more than 64 species of bacteria.6,16 Not surprisingly, infections related to dog bites tend to be polymicrobial.4,17 The reported rate of infection from dog bites varies from 1.4% to 30.0%,compared with an infection rate of 0.53% to 47.0% for all types of bites combined. 2,18 Of all bites, dog bites tend to represent the lowest infection risk, with rates similar to those of nonbite lacerations (ie, 3%-20%).4
It is estimated that 5% to 15% of all bites are inflicted by cats. Assuming the incidence of all animal bites is 800 cases per 100,000 people per year, up to 300,000 cat bites occur annually in the United States.19 Feline attacks may be underreported because many people consider injuries from cats to be less serious than those caused by dogs and may not seek medical attention. In addition, patients may not be aware of the risks associated with cat bites and may not present until complications have already developed. This theory is supported by results from a recent Belgian study, which found that although 75% of dog-bite victims presented within 24 hours of the attack, all of the patients injured by cats presented relatively late, only after complications had developed.21
Despite popular perception, cat bites result in clinical infection in 30% to 80% of cases, more than double the rate for dog bites.1,2,20 The higher rate of infection may be due to the puncture injuries resulting from cats' fanglike teeth, which are able to deliver inoculum deep into the tissue. The entry point seals off very quickly, hindering drainage and irrigation efforts. Even when these wounds are explored and drained early, they can reseal quickly, resulting in infection. These puncture wounds are present in 57% to 86% of cat bite injuries but in only 13% to 34% of dog bite wounds, which are more likely to entail longitudinal lacerations, crushed tissue, and avulsion of skin.22 However, cats tend to exert less biting force than dogs, so significant trauma to the soft tissues is less common.
The injury patterns observed with cat bites are similar in adult and pediatric populations, with most bites to an upper extremity rather than to the face. Up to 90% of cat bites are to the hand and arm.23 Female patients are more likely to present with cat-related injuries because of the higher number of female cat owners.24 Siamese cats are most often implicated in aggressive episodes, accounting for 43% of attacks in one study.5 Most injuries occur when the victim attempts to intervene in a fight between animals or rescue or care for a sick or injured cat.2
Big cat attacks, although infrequent, involve children more often than adults.25 Children's smaller size makes them more attractive prey, particularly when no adults are present. Big cats such as cougars preferentially attack from behind and kill by shaking their victim. Children injured in this way are at risk for hyperextension injuries to the cervical spine in addition to the soft-tissue trauma incurred from the bite. A comprehensive evaluation of the cervical spine should always be included in the management of these types of injuries. The long front incisors of large cats can also easily inoculate the subarachnoid space, leading to central nervous system infection or abscess formation.
The number of big cat attacks in the western United States has risen, particularly in California, which initiated a moratorium on the killing of cougars in 1971.26 The resultant rise in the cougar population and the increasing popularity of outdoor sports are bringing people in more frequent contact with these animals. Because cougars rarely attack groups of adults, adults should always supervise children in areas where the cats have been sighted.25
Human bites are thought to be rare, accounting for less than 0.01% of all visits to EDs. However, the true incidence of human bites is unknown, as patients suffering from this type of injury do not always seek medical attention.27-30 Human bites can be categorized according to the circumstances precipitating them, as outlined by data from the New York City Department of Health in the late 1970s. About 75% of human bites result from aggression such as fighting and mugging.31 These bites tend to occur during the summer months, when people are outside and wearing less clothing. Nearly 25% of human bites occur accidentally during sports, especially basketball and football, or roughhousing.31 A small proportion of human bites are "love nips" (ie, acquired during sexual activities) or self-inflicted (eg, sustained during nail biting or thumb sucking).32,33
The peak incidence of human bites occur in two age groups - toddlers and early teenagers.34 A retrospective analysis of visits to 3 urban pediatric EDs found that only 0.04% were for evaluation of human bites. Bites in the adolescent population were due to altercations and play in equal proportions.35 However, in children aged 13 to 30 months, biting is associated with play; these toddlers lack the verbal skills to negotiate conflicts in a nonviolent way and may resort to biting to achieve their goals and get their point across. The majority of these wounds are minor and do not result in medical evaluation, according to one review of daycare center records.36
The pathophysiology of human bites varies depending on whether the wound results from an occlusive bite or a fight bite. An occlusive bite is similar to an animal bite, with the human teeth closing over the skin with sufficient force to breach the tissue.37 Human occlusive bites are usually elliptical or ovoid patterns of ecchymosis, abrasions, or lacerations. Ecchymosis is caused by positive pressure from the closing of teeth that disrupts small blood vessels and by negative pressure from suction and tongue thrusting.38 Due to the morphology of the human incisors and canines, human bites generally compress tissue but rarely avulse it like animal bites.34,38 In contrast, a fight bite occurs when a fist hits a tooth. Often dismissed by patients, fight bites can cause significant morbidity from infection. The injured tendon in the hand (most commonly over the third metacarpophalangeal joint) retracts as the fist is opened. The tendon retraction inoculates bacteria into the deep structures of the hand and conceals the penetration depth of the wound.39,40 Patients with a human bite injury may seek treat ment due to concerns about wound care, tetanus immunization, infection, or permanent disfigurement; however, nearly 25% of them wait more than 12 hours to do so.37,41 Late presenters are more likely to suffer from a fight bite and more likely to already have an active infection. These patients are also less likely to be honest about the circumstances of the injury because of embarrassment, concern about police intervention, and the involvement of alcohol.28,37,43
Victims of human bites often sustain more than one wound, so a full skin examination is required in these patients. In most studies, more than 50% of human bites occur in the upper extremities.41 The head and neck area (especially the ears, cheeks, and lips) is involved only slightly more often than the abdomen and thorax, except in children. Given the distribution of surface area in a child, human bites are more common on the head and neck.35 Human bites on the lower extremities are uncommon.29,44
The most common organism isolated in the oral flora of cats and dogs (both domestic and wild) is Pasteurella, a facultative anaerobic coccobacillus.42 A single animal can carry multiple strains of Pasteurella. Although the pathogenicity of these strains varies somewhat, there is no clinically significant difference in their antimicrobial susceptibility.45 Infections involving Pasteurella, especially Pasteurella multocida, are associated with a rapid development and progression.17 They can present as early as 12 to 24 hours after a bite, whereas infections due to other organisms typically do not present for 2 to 3 days.17,46 Signs of Pasteurella infection include tenderness and erythema at the bite site along with local lymphadenopathy, fevers, and chills. Severe infections such as septic arthritis, osteomyelitis, meningitis, or sepsis can be seen in very young or immunocompromised patients.47
A prospective case series of 107 patients presenting with strictly defined infected cat or dog bite wounds was conducted to delineate the organisms responsible for the infection.17 Combined anaerobes and aerobes were isolated in 56% of the infected wounds, with an average of 5 organisms cultured per bite wound. The most common isolates from both cat and dog bites were Pasteurella species, with Pasteurella canis predominating in dog bite wounds and P multocida more common in cat bite wounds. A common misconception is that Pasteurella is an unusual pathogen in dog bites. However, in this study, the organism was present in nearly 50% of cases. Streptococci, staphylococci, Moraxella, and Neisseria were common aerobic isolates. Anaerobic isolates, which included Fusobacterium, Bacteroides, Porphyromonas, and Prevotella, were more often isolated from abscesses and were more often part of a mixed infection. Purely anaerobic organisms were present in only 1% of cases.17
In a smaller study of 39 mammalian bite wounds in pediatric patients, 74% of infections were attributed to mixed aerobic and anaerobic pathogens.48 P multocida and Pseudomonas fluorescens were present only in animal bites. Group A Streptococcus was found only in the human bite wounds. Cultures of infected human bite wounds grow anaerobic organisms in 50% of cases.49,50 The most common organisms found in human bites are also frequently found in cat and dog bites, with the exception of viridans streptococci and Eikenella corrodens. The latter may spread hematogenously from a human bite and result in endocarditis or bacteremia with sepsis.51 (See Table 3.)
Although usually self-limited, scratches or bites from cats (and rarely dogs or monkeys) can result in cat-scratch disease. This infection is due to Bartonella henselae, a gram-negative rod, and is characterized by an erythematous papule at the site of inoculation, with the development of regional adenitis and fever several days later. Many studies have found a high prevalence of infection in stray cats or adopted cats of stray origin, especially in younger animals.52
Capnocytophaga canimorsus is found frequently in wound cultures, but it seldom leads to clinical infection because of its low virulence. The bacterium can lead to sepsis and death in certain patients, however. Risk factors for severe infection include asplenia, alcohol abuse, corticosteroid therapy, or other immunocompromised states.4,53 Fortunately, the organism is susceptible to most antibiotics used for prophylaxis.4,54 ED clinicians should be aware of this organism, as special culture techniques may be needed to diagnosis the infection.54 Although very rare, 5 cases of C canimorsus septicemia resulting from cat bites have been reported.53
The chief concerns in the differential diagnosis of bite wounds are those related to wound severity or potential exposure to infectious disease. Cases of human bites also bring the possibility of child abuse to the clinical situation. (See Table 4.)
Prehospital care providers must be aware that dog bites have the rare but serious potential to threaten life or limb and that patients may require triage to a trauma center. Although no data exist to support one method of prehospital care over another, it has been suggested that if ED care is likely to be delayed for more than 1 hour, wounds should be cleaned at the scene and covered with sterile dressings. Hands and feet should be immobilized if injured.6
A thorough patient history should be obtained, including the following information about the bite wound:
Bites vary in severity from minor to limb or life threatening. The emergency clinician should perform a complete trauma evaluation on a child who is bitten if the situation calls for it. All children who are victims of an animal attack should be evaluated from head to toe to identify signs of both penetrating and blunt trauma.
Wounds should be assessed for location, depth of penetration, amount of devitalized tissue, and the presence of any foreign material. Assessments of neurovascular function and tendon and joint function are also required, especially for bite wounds to the digits. Obvious foreign material and the involvement of deep structures such as tendons and joints in a severe wound increase the chances of future infection.
Evaluation of hand wounds complicated by infection should also include an assessment for flexor tenosynovitis. The four signs of flexor tenosynovitis as described by Kanavel are fusiform swelling of the digit, the digit being held in a partially flexed posture, pain with passive extension of the involved finger, and tenderness along the tendon sheath.
The physical examination is the most important aspect of the initial evaluation of bite wounds. Few diagnostic studies will affect initial treatment. Unless clinically indicated for excessive blood loss, a complete blood cell count is rarely useful. Even in the presence of infection, the white blood cell count is rarely elevated.55 Some controversy exists regarding the necessity of obtaining routine wound cultures on fresh, noninfected bite wounds because the results do not reliably predict which wounds will become infected.56 However, cultures in grossly infected wounds may guide therapy. Both aerobic and anaerobic cultures should be obtained, and the microbiologist should be informed of the animal responsible for the wound so that a diagnosis of Pasteurella infection can be considered if needed. Some pathogens are slow growing, so cultures should be held for 7 to 10 days. Erythrocyte sedimentation rate and C-reactive protein level are elevated in the majority of clinically evident infections. In the case of possible infection with C canimorsus, diagnosis can be confirmed with a peripheral blood film, which will demonstrate intracellular gram-negative rods.46 Radiography should be used liberally, especially for dog bites in small children. The force produced by the bite of a large dog is more than sufficient to fracture bone. Since open fractures require specialty consultation and possible operative management, these injuries should be identified early. In addition, bite wounds over joints, such as fight bites, necessitate radiographs to detect the presence of air, which may suggest joint penetration.57 Up to 70% of radiographs taken to evaluate fight bites show foreign bodies, fractures or bony fragments, narrowing of joint space, and air. However, despite negative results on physical examination and initial plain radiographs (including arthrograms), 16.5% of fight bites still have joint penetration.40 Radiographs may be useful if foreign bodies are suspected. Diagnostic imaging, including radiographs and ultrasound, has a more important role in the evaluation of late complications of a bite wound, such as osteomyelitis, septic arthritis, and tenosynovitis.
As mentioned previously, most bites are not life threatening. However, these injuries can be a source of stress for the emergency clinician, particularly when they occur in children. Some controversy surrounds the management of simple bites as well as life-threatening wounds. Treatment of bites should include consideration of analgesia, local wound management, infection prevention, and management of complications (as described in the Complications section).
Analgesia is often achieved locally during acute treatment, but patients may require systemic medications depending on the severity of the injury. Younger children in particular may require procedural sedation for the management of wounds. As with other traumatic injuries, immobilization and elevation of the bite wound promote analgesia.
Local wound management consists of wound exploration, thorough irrigation, debridement of necrotic tissue, and closure (if appropriate; see discussion that follows). Infection prevention in bite wounds requires attention to both viral and bacterial agents. Viral prophylaxis depends on the species of the assailant. B virus, or Cercopithecine herpesvirus, transmission from a monkey to a human can cause fatal encephalomyelitis. Animal bites also raise concerns of rabies and tetanus, whereas human bites require consideration of the transmission of viruses such as hepatitis and HIV. If the human bite is deemed to be high risk, laboratory studies such as rapid plasma regain (RPR), hepatitis B surface antigen, hepatitis C antibody, and HIV-1 and HIV-2 tests are indicated, with appropriate follow-up for the results and future testing.37
One decision the emergency clinician must make is whether to prescribe antibiotics to bite victims without obvious infection on initial presentation. This question was addressed in numerous small studies.58-62 Cummings reviewed 8 randomized trials on this topic in a 1994 meta-analysis reported in the Annals of Emergency Medicine.63 The author reported a significant decrease in infection rates when antibiotics were given prophylactically to patients with dog bites, and the study demonstrated a relative risk of infection of 0.56.18,63 (Although the term prophylactically is used widely in the literature, the administration of antibiotics to prevent an infection after a bite wound is actually therapeutic because injury and inoculation have already occurred.) However, an analysis of the use of prophylactic antibiotics for bite wounds by Turner did not support the routine use of antibiotics for dog or cat bites to areas other than the hand, although the author cautioned that the sample size was small.64 In addition, the studies analyzed by Turner used inappropriate antibiotics, as the microbiological analysis published in 1999 by Talan et al was not yet well-known. Turner did report a reduction of infections in human bites and bites to the hand.64 Based on this analysis, empiric coverage against Staphylococcus, Streptococcus, and Pasteurella is indicated in hand wounds, especially. Other wounds requiring prophylaxis include deep puncture wounds, wounds with extensive tissue trauma requiring debridement, and wounds in elderly patients, the immunocompromised, or those with poor circulation to the involved area.54 (See Table 5.)
Amoxicillin-clavulanate is commonly recommended as a first-line agent for the prevention and treatment of infection after animal bites because it offers excellent coverage against all common bite wound pathogens.13,18 The recommended dosage for outpatient prophylactic therapy with this agent is 45 mg/kg per day divided into 2 doses or 40 mg/ kg per day divided into 3 doses for 3 to 5 days. For inpatient therapy of infected wounds, the parenteral form of this combination drug should be given. The recommended dosage of ampicillin-sulbactam is 100 to 150 mg/kg per day in divided doses every 6 hours, given for 10 to 14 days depending on the severity of the infection. (See Table 6.) For patients who are allergic to penicillin and the cephalosporins, clindamycin plus trimethoprim-sulfamethoxazole can be used.13,54
On the basis of in vitro data and case reports, penicillin is still the drug of choice for treating solitary Pasteurella infection. This research also showed positive results for tetracycline, fluoroquinolones, and chloramphenicol, all of which can be used in the penicillin-allergic adult patient but are not recommended for use in children.47,65,66 However, because culture results for Pasteurella are not available initially and polymicrobial infections are so common, starting penicillin as monotherapy is not recommended. High rates of resistance were found with clindamycin and erythromycin. In addition, monotherapy with these agents is not recommended for the treatment of cat bites.66 Azithromycin was also studied in vitro with positive results.67 Although no clinical trials have been conducted, there is one case report of azithromycin being used successfully in combination with clindamycin in a patient who had multiple drug allergies.68 No human trials exist to support monotherapy with azithromycin for the treatment of Pasteurella infections. A report on the use of ceftriaxone for infected cat and dog bites determined that this agent was a convenient alternative to inpatient intravenous (IV) therapy.69 However, this study was based on only 23 patients, and cotherapy with other antimicrobial agents was not standardized. In addition, the protocol described in this study required daily follow-up in a cellulitis clinic, which is not available to all emergency clinicians. Thus, outpatient monotherapy with ceftriaxone cannot be recommended at this time.
Given the high infection rate with human bite wounds, antibiotic prophylaxis for an uninfected wound is usually considered but remains controversial. The evidence for antibiotic prophylaxis is poor. For example, Broder et al compared the use of placebo for the treatment of low-risk human bite wounds (defined as penetration through the epidermis only and not involving hands, feet, overlying joints, or cartilaginous structures) versus use of a cephalexin-penicillin combination.34 As only 1of 62 patients receiving placebo developed an infection (compared with 0 of 63 patients receiving cephalexin-penicillin), the statistical significance of the study must be questioned. Despite the lack of clear evidence supporting prophylactic antibiotics in the management of human bite wounds, their use is encouraged because these injuries carry both aerobic and anaerobic organisms.50,70
The antibiotic choice for human bite wound prophylaxis and treatment of wound infection can be difficult but must account for both the pathogenic organisms (see Table 3) and the characteristics of the antibiotic. With four isolates per wound, the organisms may be unique to human mouth flora.50 For instance, the slow-growing gram-negative E corrodens originates from gingival plaque and is significant due to its resistance to penicillin.42 The most common oral antibiotic used for human bite prophylaxis is amoxicillin-clavulanate. Goldstein et al advocated the use of amoxicillin-clavulanate for human bites in the 1980s despite criticism of their study.71 (The criticism included sponsorship by a drug company, poor protocol, and unclear superiority of amoxicillin-clavulanate.72) An alternative to amoxicillin-clavulanate for human bite prophylaxis is penicillin and dicloxacillin; however, both drugs are required, as penicillin has poor Staphylococcus aureus coverage and dicloxacillin has poor E corrodens coverage.71,73 Both prophylactic options are safe for children, who are more likely to receive antibiotics.35
The utility of viral prophylaxis after a human bite remains unclear due to the lack of evidence. Only a few case reports of hepatitis B, hepatitis C, and HIV transmission through human bites have been reported.74-77 Key factors when considering viral prophylaxis include the assailant's viral status (if known) and the integrity and gross contamination of the wound.37,74 A bite sustained in a work environment such as health care or law enforcement presents an added risk for viral transmission.37 In addition to baseline blood work (RPR, hepatitis B surface antigen, hepatitis C antibody, and HIV-1 and HIV-2 tests) and follow-up blood work, the patient may be offered hepatitis B immunoglobulin, hepatitis B vaccination, and antiviral medications to prevent HIV. Treatment for fight bites is more aggressive because of the difficulty in determining deep structure involvement, the increased risk of infection, and the morbidity from established infection. As a result, the decision to treat these wounds in the ED versus the operating room (OR) is sometimes a matter of debate. For example, fight bites often involve underlying damage not recognized on ED evaluation. In a study by Phair and Quinton of 29 fight bites, nearly 60% had deep structure involvement.78 Damage to the tendon also indicated likely damage to the bone and capsule. In addition, the bacterial inoculation of deep hand structures observed in many fight bites increases the risk of infection. According to the results of one randomized placebo-controlled study, local wound care alone is not adequate for managing these injuries.79 In addition to irrigation of the wound in the ED or the OR, depending on the extent of the injury, antibiotics must be used to either prevent or treat an active infection. For the recent, uninfected fight bite, prophylactic antibiotics (administered orally or intravenously) are also indicated.79 Moreover, fight bites are more likely than other human bites to harbor an active infection because of the age of the wound at presentation,39 increasing the risk of morbidity.
The goal of wound management is to achieve optimal cosmesis and maintain optimal function while preventing secondary infection. The decision to suture a wound is based on several considerations including its location, host immunity, time of injury, presence of contaminants, and amount of devitalized, crushed tissue, all of which contribute to the risk of infection. Copious irrigation with saline or sterile water prior to wound closure is recommended.54,80 Similar to the treatment of lacerations, treatment of bite wounds with pressure irrigation is thought to decrease the amount of potentially pathogenic microbes that may colonize the wound and ultimately result in infection. Tap water is also a consideration, but a 2002 study comparing sterile saline with tap water wound preparations notably excluded bite wounds.81 Debridement is recommended to remove any devitalized tissue, which contributes to the rate of infection and poor cosmetic outcome. In one study, debridement of dog bite wounds resulted in a 30-fold reduction in the rate of infection.82
The location of the bite may predict the risk of wound infection. In general, facial wounds have a low risk of infection due to the excellent vascularity of the facial tissues. A small study by Donkor and Bankas involving 30 patients who underwent primary closure of a human bite wound to the face confirmed the finding that primary repair of facial bites is safe, but wound closure did not elucidate clear superiority in preventing infections.83 In contrast, injuries to the hands and feet result in more frequent infections. Numerous small compartments in the hand and the absence of significant soft tissue separating the skin from deeper structures such as bone and joints make it more prone to infection. In one study of bite wounds to the hand, 18.8% of wounds became infected. If these wounds were closed primarily, the infection rate rose to 25%. In addition, the bite wounds of patients who presented more than 2 hours after injury were more likely to become infected.84 Infections of the hand can result in permanent loss of function, so these wounds should be treated aggressively with irrigation and debridement if needed. The bite wounds should be left open with the hand immobilized.85
Some large dog bites have cosmetic significance, and the patient may expect these wounds to be sutured. However, this may not always be the case. It has been reported that just 12% of dog bites in the ED are closed primarily.18,86 In 1988, a prospective controlled trial was conducted among 96 patients with 169 lacerations related to dog bites.87 Of these,92 were sutured and 77 were left open. Thirteen wound infections were subsequently noted, with an overall infection rate of 7.7%. The difference in rate of infection between wounds that were sutured and those left open was not significant.
The issue of primarily closing dog bite wounds was again discussed in a 2000 article presented in Academic Emergency Medicine.88 In this study, 88 of the 145 patients included had sustained dog bites. These wounds were closed primarily. Five of the 88 dog bites became infected, representing an infection rate of 5.7%. The infection rate for nonbite lacerations at the same hospital at that time was 3.4%. Although 81 of the 145 patients received antibiotics, their use was not controlled for in the study, clouding the picture. Nevertheless, the results indicate that dog bites may be closed, particularly in the setting of cosmetically important wounds. Other authors have concluded that suturing of puncture wounds should be avoided, that high-risk wounds to the hand should be closed only under special consideration, and that the risks and benefits of wound closure should be discussed with the patient prior to closure.13 Physicians may also choose delayed primary closure, in which wounds are left open by the ED clinician and then repaired a few days later after careful cleansing.13
Human bites are more likely than animal bites to become infected, although the infection rate varies. Lindsey et al described the natural history of 434 human bite wounds sustained by residents in a long-term care facility, using 803 lacerations for comparison.30 The 434 human bite wounds had a 17.7% infection rate without any serious complications. Factors that increase the possibility of infection include location of the bite (hand, foot, and joint), depth of the bite (bone or joint involvement and puncture or crush injury), vascular supply to the site (peripheral vascular disease, peripheral edema), immune status of the patient (age greater than 50 years, immunosuppression, diabetes mellitus, chronic alcohol use), and timing of treatment.41 (See Table 6) Thus, prevention of infection drives decisions in the treatment of human bite wounds.
On the basis of the data presented previously, primary repair of bite wounds is not recommended in the following cases because of the high risk of infection: bite wounds to the hand, human bites and cat bites to areas other than the face, and injuries in immunocompromised patients. If there has been a delay of greater than 6 hours in presentation for facial and scalp bites or greater than 12 hours for bites in other locations, the risks of infection versus the cosmetic benefits of primary wound closure should be weighed when deciding on management, although the cutoff of 12 hours has not been well studied. Wounds that are not sutured should be irrigated and dressed. Injured extremities should be immobilized, and all patients should be instructed to have their wounds reevaluated within 24 to 48 hours.
Bite wounds present a high risk of transmission of Clostridium tetani, which is present in the teeth and saliva of animals. For any breach in skin integrity, tetanus immunization is required. If the wound is clean with little devitalized or crushed tissue, tetanus immunization within the past 10 years is adequate. Patients with wounds that are contaminated or that involve crushed tissue or punctures require immunization within the past 5 years. If immunization status is unknown, both tetanus immune globulin and the diphtheria and tetanus toxoids booster vaccine should be administered.
More than 85% of reported rabies cases result from injuries caused by wild animals,89 with raccoons, skunks, bats, and foxes currently the main source of human rabies infection. However, oral vaccineimpregnated bait is being used in an attempt to decrease rabies in the wild animal reservoir, particularly among raccoons.93
Only 7.9% of all rabid animals reported in the United States are domesticated,90 with cats the most commonly reported domestic animal with rabies in the last decade.91 According to the CDC, although there were 79 cases of rabies in U.S. dogs in 2006, only one case of human rabies infection was attributed to a dog bite.92 (This occurred in the Philippines, with the infected individual later returning to the United States.) The rate of transmission from domestic animals remains low because of current aggressive rabies vaccination programs89 that have greatly reduced the incidence of canine-variant rabies.
The CDC still recommends rabies treatment in the setting of a suspicious dog bite. Bites that pose a risk of rabies should be washed with copious amounts of a virucidal solution such as povidoneiodine, as this agent has been shown to decrease the rate of rabies infection.89 However, the agency recommends no additional treatment as long as the dog appears healthy and can be observed for a 10-day period.92
Lagomorphs are thought to present a low risk of rabies transmission. Despite case reports of rabies in guinea pigs and rabbits, no reports of human rabies have resulted from these exposures.94 Therefore, routine postexposure rabies prophylaxis for children injured by these animals is not indicated.
Animal bite prevention is important in the pediatric group. Certain predictable behaviors observed in both humans and animals typically result in injury. Bites can result when a child tries to remove a pet from a fight with another animal. Although avoidance of this rescue attempt is most prudent, if a child insists on intervening in an animal fight, he or she should grab the pet's hind legs and pull it away rather than trying to pry the animals apart at the head and mouth.2 Interacting with an animal during feeding times can also be potentially dangerous, as the animal's natural instinct is to protect its meal. Education on how to interact with dogs may be helpful in preventing dog bites. An Australian study indicates that after only 30 minutes of instruction, children were much less likely than members of a control group to exhibit potentially dangerous behavior around a dog.42 The study involved 7- to 8-year-old children who were taught how to approach dogs, how to recognize angry or frightened dogs, and how to respond when pushed down or approached by a dog.
Parents should also be advised on preventive techniques. If the animal responsible for a bite is a family pet, a discussion should take place on the risks of keeping the pet in the home. An audit of dog bite referrals in one plastic surgery department found that of 17 pet dogs that attacked a child, 12 remained in the home afterward.95 Six of the 17 children had been bitten before, 3 by the same dog. The general public should also be taught the importance of seeking prompt medical care after injury by any animal, as this precaution may reduce the frequency of negative outcomes.
Complications from bite wounds vary but center around infection. In one retrospective review, the most common serious development from animal bites was osteomyelitis, which was seen in 10% of patients.96 Insufficient data were presented to determine which types of wounds were associated with this complication.
Some bites in children may be trivial and missed on initial examination, but these have the potential for negative outcomes. There are at least 3 reports of brain abscesses with neurologic compromise as a result of dog-inflicted head wounds that were missed or presumed to be minor. Initial assessment of bites to the head in children, especially when the child is attacked by a large animal, should involve meticulous wound care and an assessment to determine whether the skull has been punctured.97-100 Noninfectious complications such as arthritis or disfigurement may require outpatient treatment by a specialist.
Endocarditis is a rare complication of mammalian bites in children with congenital heart disease who have undergone surgical correction.101 When patients in this subset present with fever after a bite, multiple blood cultures may be useful. Even if the wound is identified as the source of the fever, bacteremia and subsequent seeding of the valvular tissue may have occurred.
Possible emotional effects in children following a dog bite deserve mention as well. One study found evidence of posttraumatic stress disorder (PTSD) in 12 of 22 pediatric patients 2 to 9 months after a bite. However, these children were usually bitten by dogs unknown to them, and those diagnosed with full PTSD had suffered severe or multiple bites.14 This possibility should be considered when discussing follow-up with patients and their parents. If a child is suffering from nightmares or is exhibiting excessive fear of animals after the event, psychological counseling should be recommended.102
The occurrence of bites to the genitalia is underreported, possibly because of self-treatment and embarrassment. Injuries to this area of the body are prone to complications because of delayed presentation. In addition, the nature and location of the genital tissues are conducive to the development of infection. In particular, the copious loose subcutaneous tissue of the penis allows bacteria to spread, causing infections such as E corrodens, Fournier gangrene, or a penile abscess.46,103-105 The literature on the management of these injuries is limited. One case series indicated that the majority of human bites to the genitalia presented as ulcers, while dog bite wounds had presentations ranging from small abrasions to amputation wounds.106
A complete discussion of wild animal bites is beyond the scope of this article. With the expansion of urban living areas into the wilderness, wild animal attacks on people are becoming more frequent. Episodes most often involve children attacked by cougars or coyotes.107 Bite wounds by large cat species are almost always deep puncture wounds because of the animal's substantial front incisors and their propensity for biting into the neck and abdomen, which have no bony protective covering. Pasteurella is among the normal oral flora in these larger cats, just as it is in their smaller, domestic counterparts. Mouth cultures of black bears have yielded results similar to those from cultures of domestic dogs.108 Treatment of coyote and bear bites is similar to the management of domestic dog bites.
Although the management of envenomation resulting from a bite is beyond the scope of this article, it should be remembered that the bites of certain mammals such as the platypus, slow loris, and short-tailed shrew involve venom transfer.
The oral flora of rodents is similar to that of other species. Rat bites occur most often in children younger than 5 years and in those from lower socioeconomic levels. Depending on the age of the child (ie, the ability to flee or fend off attacks), some bites may be fatal. Reports of hypovolemic shock in infants with multiple bites have been published.109 Rodent bites present a low risk of rabies transmission because the virus is not secreted in the saliva of infected animals. There have been case reports of hamster bite peritonitis in patients on peritoneal dialysis whose catheters were chewed by their pet rodents, resulting in inoculation and eventual infection of the peritoneal dialysate. In these cases, Pasteurella species were isolated.110 Rarely, animal care workers and pet owners with repeated low-level exposure to rodent allergens may have allergic reactions to saliva when bitten by a rodent.111 Case reports of hamster bites have also been noted. Patients with household dust mite allergies are more susceptible to this type of reaction because of some cross-reactivity between dust mite allergens and hamster allergens.112
A potential contaminant involved in rodent bite wounds is Streptobacillus moniliformis, which can cause rat-bite, or Haverhill, fever. Infection with this agent presents weeks after the initial bite with a viral-like illness, regional lymphadenopathy, arthralgia, and a maculopapular, urticarial, or vasculitic rash.113 Penicillin G can be used to treat this infection. The recommended dosage is 25,000 to 50,000 U/kg daily, divided into 4 doses, for 7 days.114 Alternatives for the penicillin-allergic patient include streptomycin (20-40 mg/kg, up to a maximum of 1g daily), erythromycin (30-50 mg/kg per day divided into 2-4 doses), and tetracycline (only for children older than 8 years; 25-50 mg/kg per day divided into 2-4 doses).
In some cases of mammalian bites, forensic evidence may be required by law enforcement officials. If a fatality has been attributed to an animal attack, all wounds should be carefully documented. Photographs may also be helpful, especially when the injuries are extensive. An attempt should be made to correlate the bite with the species implicated. Humans have 4 incisors and short canines, and human bite wounds are typically elliptical or ovoid.115,116 Dogs have 6 incisors and 2 very large canines, which can result in deep punctures with torn tissue. Dog bite wounds are not oval.
Although the majority of human bites to younger children result from frustration , the emergency clinician should be aware that human bites may involve potential child abuse. If the intercanine distance (the linear distance between the central points of the cuspid tips of the wound) is greater than 3 cm, the bite was likely inflicted by an adult. A forensic odontologist, a forensic pathologist, or a health care worker experienced in patterns of child abuse should be consulted for the collection of evidence such as photographs, dental casts, and saliva samples.38 In one case series, the forensic dentist was able to positively identify the adult perpetrator in 46% of cases.117 If a bite has resulted in significant disfigurement or if there is a risk of litigation, photodocumentation may also be warranted.
Monkey bites are rare, with only about 130 reported per year in the U.S.. Most of these bites are to pet owners or researchers.118 Monkey bites are similar to human bites, with the important exception of Cercopithecine herpesvirus, or the B virus. Native to wild Old World monkeys of North Africa and Asia (eg, rhesus and other macaque monkeys), the B virus causes little morbidity in the natural host.118 However, untreated B virus in humans has a 70% fatality rate from rapidly ascending encephalomyelitis. Although still uncommon, the B virus is gaining momentum as captive monkeys harboring latent infections are increasingly used for research. Standard precautions when dealing with monkeys may prevent B virus transmission. Once a worker is exposed to the B virus, research facilities follow detailed protocols including copious irrigation and scrubbing of the wound, evaluation of the monkey, contact with a veterinarian or the CDC, and possible empiric treatment with acyclovir.119 After this aggressive care, a monkey bite may still present to the ED for further treatment or with subsequent complications.
Avoid treating fight bites like other bites. Fight bites require careful evaluation for deep structure involvement as well as copious irrigation, antibiotics, and possibly a consultation with a hand surgeon.
Failure to recognize the presence of a foreign body is the most common reason for wound care-related lawsuits.121,122 The small pointed teeth of some mammals can break off and enter wounds.
Failure to appropriately determine the presence of joint space invasion is another common cause of wound care-related litigation.121,122 Joint space involvement is of particular concern in the setting of police dog bites.
Unrecognized nerve and tendon injuries are also a source of wound care-related lawsuits.123,124 A careful neurovascular examination helps to catch these injuries.
Disposition of a human bite wound depends on the location and depth of the bite, the active signs of infection, and any comorbid conditions. Although most human bite wounds can be managed in the ED with local wound care and oral antibiotics, the admission rate after these injuries varies greatly in the literature. Because human bite wound studies are very diverse, admission rates vary from 10% to 50%.29,37,50,120 Admission of children, however, is rare.35 Risk factors that increase the likelihood of admission include a bite sustained during an altercation, a bite to the hands or fingers (including fight bites), presentation for medical attention more than 24 hours after the bite, and active infection.120 Most fight bites do require admission, but about 10% of patients leave against medical advice.39
Indications for hospitalization include the following factors:1,82
Mammalian bites in children may result in wound infection, poor cosmetic outcome, and emotional stress. A meticulous physical examination should be performed on all pediatric patients in order to identify wounds that may later result in serious infections. Appropriate selection of wounds for primary closure or treatment with prophylactic antibiotics should lower the risk of wound infection. Preventive measures should be aimed at teaching parents and children appropriate behavior around animals. This may include limiting contact with pets while they are eating, avoidance of strange animals, and proper supervision of children around animals. In addition, children with certain medical conditions, such as congenital heart disease or an immunocompromised state, should be brought for immediate evaluation after any injury by an animal.
The patient is uncomfortable and is given analgesics. The cat and dog bites are superficial and require only exploration and irrigation. A small open wound without any bleeding is visible on the patient's right third MCP joint, but the finger's range of motion is significantly limited by pain. A radiograph of the hand reveals a foreign body in the wound, so the hand surgeon takes the patient to the OR for exploration and irrigation of the wound. Part of a tooth and a few small, loose cartilaginous bone fragments are extracted from the fight bite wound, and it is then primarily repaired. The patient receives 4 days of IV ampicillin-sulbactam and is discharged with 7 days of amoxicillin-clavulanate. The patient's mother declines HIV prophylaxis, but the patient does receive vaccines for hepatitis B, tetanus, and rabies. The patient's initial and follow-up tests are negative for syphilis, hepatitis B and C, and HIV. Animal control is contacted and watches the dog for 10 days. The patient recovers fully, with mild arthritis around his third MCP joint.
For the second patient, topical LET (lidocaine-epinephrine-tetracaine) is placed on the bites. After the patient calms down and anesthesia is adequate, the wounds are irrigated and explored. An examination of the cat bite on the left forearm reveals only puncture marks without any bleeding or surrounding erythema. The human bite on the right forearm is a superficial laceration requiring two Steri-Strips™ (3M, St Paul, Minnesota). After a long discussion with the patient's mother about the risk of infection and potential consequences, antibiotics are declined, but follow-up with the patient's pediatrician is scheduled in 2 days. The kitten is a house pet and has not shown any erratic, aggressive behavior. The animal has received appropriate immunization and will be monitored at home for 10 days. The patient leaves happily with a Popsicle®.
Evidence-based medicine requires a critical appraisal of the literature based upon study methodology and number of subjects. Not all references are equally robust. The findings of a large, prospective, randomized, and blinded trial should carry more weight than a case report.
To help the reader judge the strength of each reference, pertinent information about the study, such as the type of study and the number of patients in the study, will be included in bold type followingthe reference, where available.
infection in primary care. Cleve Clin J Med. 1997;64(3):137-141. (Review)
bitten, why, and where? An updated assessment of human bites. Ann Emerg Med. 2002:40(4, pt 2):S41. (Retrospective; 172 patients)
Brian Rempe; Kara Iskyan; Mara Aloi
September 1, 2009