Rationale:

• Mammalian bite wound infections usually present within 24 hours; therefore, bite wounds presenting late without clinical evidence of infection are less likely to benefit from prophylaxis.
• Cat bites become infected in up to 80% of patients due to the predominance of puncture wounds, which cause deep inoculation of bacteria.
• Human hand bites and puncture wounds carry a high risk of infection and can result in serious complications and loss of function; therefore, they may benefit from antibiotic prophylaxis.
• Carefully selected, low-risk human bite injuries carry a low risk of infection and may not benefit from antibiotic prophylaxis.
• Dog bites carry a lower risk of infection (2% to 20%) and the efficacy of antibiotic prophylaxis is less established than for cat and human bites. Only one of seven randomized double-blind studies involving dog bites has shown a statistically significant reduction in bite wound infection with treatment.
• Amoxicillin-clavulanate has excellent coverage against the most common organisms isolated in dog, cat, and human bite infections: P. multocida, Staphylococcus aureus, Streptococcus species, Eikenella corrodens, and various anaerobic bacteria.

Evidence:

• A randomized, double-blind, placebo-controlled trial of prophylactic oxacillin for uninfected cat bites led to a significant reduction of infection from 67% (4/6) to 0% (0/5), although the study comprised only 11 patients (47).
• A randomized, placebo-controlled trial of antibiotic prophylaxis (oral cefaclor or intravenous cephazolin) for patients with human hand bites presenting within 24 hours revealed a significant reduction of infection from 47% to 0% with either oral or intravenous treatment (22).
• A meta-analysis combining three studies involving hand bites (2 dog, 1 human) revealed a significant reduction of infection with antibiotic treatment from 28% to 2% (39).
• The rate of puncture wound infection was nonsignificantly reduced by prophylactic antibiotics from 31% (5/16) to 7% (1/14) in a meta-analysis of mammalian bites (39).
• A prospective, randomized, double-blind, placebo-controlled trial of antibiotic prophylaxis for low-risk human bite injuries (defined as nonpuncture wounds to the body, excluding the hands and feet, without penetration through the epidermis, and occurring in immunocompetent individuals presenting within 24 hours of the bite) showed a low rate of infection in both the placebo (1/62) and antibiotic (0/63) arms (48).
• A randomized study of prophylactic penicillin for nonfacial dog bite wounds did not reveal a significant reduction of wound infection, with a treatment infection rate of 4% (1/25) and a placebo rate of 3.3% (1/30) (49).
• Prophylactic amoxicillin-clavulanate was shown to significantly lower the risk of mammalian bite infection in patients presenting 9 to 24 hours after a bite from 73% (8/11) to 13% (1/8) (50).
• A randomized double-blind placebo-controlled study of oral dicloxacillin, cephalexin, or erythromycin in 191 adults presenting within 12 hours of a dog bite showed a nonsignificant reduction of infection from 5.1% (5/96) to 1.1% (1/89). Full-thickness bites showed a stronger nonsignificant reduction in wound infection rate from 13.5% (5/37) to 2.8% (1/36) (51).
• A randomized, double-blind, placebo-controlled trial of oxacillin for dog bites in 63 patients did not reveal a significant reduction in infection, with 0/56 placebo treated patients developing infection (52).
• Prophylactic co-trimoxazole failed to significantly decrease (13.8% to 5.5%) the wound infection rate in 113 patients treated in a randomized, double-blind, placebo-controlled trial. Subgroup analysis did reveal a possible benefit of prophylaxis in hand bites, with a decrease in wound infection rate from 16.7% to 0% (53).
• A randomized, double-blind, placebo-controlled study of prophylactic dicloxacillin or cloxacillin for dog bite wounds failed to show a significant reduction of infection rate, with a decrease from 11% (2/18) to 6.7% (1/15) (54).
• A randomized, double-blind, placebo-controlled study of prophylactic penicillin for dog bites in children ages 1 to 16 failed to show a significant reduction of bite wound infection, with a decrease from 10.5% (2/19) to 5% (1/20) (55).
• A meta-analysis combining 6 randomized, placebo-controlled studies involving antibiotic prophylaxis for dog bites failed to show a significant reduction in bite wound infection (39).
• Another meta-analysis that combined eight studies (1 not randomized; 1 involved animals other than dogs) of prophylactic antibiotics for dog bites did reveal a significant reduction in bite wound infection with treatment, with a relative risk of 0.56 (56).

Comments:

• All antibiotic prophylaxis trials involve small numbers of patients and are underpowered to determine significant differences. It has been estimated that 740 patients would need to be randomly selected to lower the probability of type II error to 10% (53).