Which Dog Bite Wounds to Close and When?

Dog bites lead to about 1% of all emergency room visits. Scant evidence supports interventions to reduce their high risk of infections and scarring. According to the most recent Cochrane review on management of mammalian bites,¹ best evidence-based practice is aggressive cleansing and debridement followed by a 2-day to 5-day course of systemic antibiotics and delayed primary or secondary wound closure. Randomized controlled trials on dog bites to the hand cautiously support use of prophylactic antibiotics to reduce their especially high risk of infection.¹ Some evidence suggests suturing facial dog bites to minimize scarring, but  insufficient evidence informs clinical decisions about techniques and timing of closure of dog bites at other body sites, details about optimal antibiotic usage, cleansing, debridement, wound dressings or other parameters of wound care, or risk factors  that affect dog bite outcomes such as healing, infection, or scarring. Two studies reviewed here strengthen the evidence base for the safety of early primary closure for dog bites in several locations² and the cosmetic benefits resulting from immediate primary closure of wounds on the face.³ 

Laura Bolton, PhD
Department of Surgery, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ 

Reference: Paschos NK, Makris EA, Gantsos A, Georgoulis AD. Primary closure versus non-closure of dog bite wounds. a randomised controlled trial. Injury. 2014;45(1):237-240.

Rationale: Dog bites have a high risk of infection. Stronger evidence is needed to inform decisions about risk factors associated with their immediate primary suturing.  

Objective: Conduct a randomized controlled trial (RCT) comparing infection and cosmetic outcomes of dog bites managed with immediate primary closure or nonclosure. 

Methods: A consecutive cohort of patients at least 16 years of age with a full-thickness dog bite not involving muscle or bone were included in the study. Patients were treated in a University of California hospital-based emergency room within 48 hours, the injury was surgically debrided and cleansed with up to 500 mL of povidone iodine. Each subject was randomly assigned for their wound to be left open (n = 92) or for the wound to be immediately closed (n = 90) under local anesthesia with 2% lidocaine. Depending on the wound site, 3-0 or 4-0 nylon sutures were used to close the injury using simple interrupted sutures. All subjects received prophylactic antibiotics and amoxicillin/clavulanic acid 500/125 mg every 12 hours for 5 days, plus tetanus toxoid or vaccination as appropriate. All wounds were dressed with gauze and all patients were advised to keep the wound dry for 48 hours after debridement, but not immobilized.  Sutures were removed at 7 days for wounds on the head, face or neck, 10 days for wounds on the upper extremities, and 14 days for wounds on the lower extremities or trunk. Infection was diagnosed by systemic fever, local abscess or lymphangitis, or relative increase in erythema, edema, drainage, temperature, or tenderness at the wound edges. Cosmetic appearance of the scar was rated using the Vancouver Scar Scale 4 weeks after injury. Both outcomes were evaluated by a surgeon blinded to treatment. Effects of timing of suturing, wound location, patient age, and wound area were also explored on both outcomes.

Results: Of the 182 patients enrolled, 82 in the primary suturing group and 86 in the nonsutured group completed the study sufficiently for evaluation and analysis of outcomes. Wounds were similar on all demographic variables and distributed similarly over body sites in the 2 groups, mainly on the hand or arm, followed by the head or neck, then the lower limbs. No head wounds became infected, significantly fewer than the 10.8% experienced in the upper extremities. The only significant effect of primary closure was to improve the cosmetic appearance of sutured facial injuries (P < 0.001). Patients reporting to the emergency room within 8 hours of their injury were less likely to develop an infection (P < 0.0025). Noninfected wounds had better cosmetic results than infected wounds, regardless of location. Larger wounds had poorer cosmetic results, but no different infection rates compared to smaller wounds. All infections subsided in 10 days or less after patients were admitted to the hospital and given appropriate antibiotics based on culture results, in addition to wound debridement and irrigation.  

Authors’ Conclusions: Pending replication by other large RCTs, one may cautiously conclude that early primary suturing of full-thickness dog bite wounds not involving muscle or bone, when performed after irrigation with povidone iodine and debridement and along with appropriate intravenous antibiotic therapy, improves scar cosmesis without increasing the likelihood of infection.  

Reference: Rui-feng C, Li-song H, Ji-bo Z, Li-qiu W. Emergency treatment on facial laceration of dog bite wounds with immediate primary closure: a prospective randomized trial study. BMC Emerg Med. 2013;13(Suppl 1):S2.

Rationale: Ten percent of all dog bites receiving emergency treatment in China are facial, with a risk of inducing intracranial infection. Prospective research is needed to explore safety and effects of their immediate primary closure in the Chinese population. 

Objective: Conduct a RCT exploring effects of primary closure compared to nonclosure on facial dog bites with systemic antibiotics given only if the dog bite became infected. 

Methods: Patients of all ages requiring surgical management at the Beijing Rabies Prophylaxis Clinic within 8 hours after experiencing a noninfected facial dog bite over 2 cm in length were randomized to have their facial wound closed  (n = 300) or not closed (n = 300) using 5/0 or 6/0 stylolite. Appropriate rabies and tetanus prophylaxis was given to all patients. Empiric systemic antibiotic prophylaxis was used only if the patient showed 3 major signs of infection (ie, body temperature > 38ºC wound abscess or lymphangitis) or 4 of 5 minor signs (ie, erythema > 3 cm from the wound edge, tenderness, swelling, purulent drainage, or white blood cell count > 12,000/mL). Drains were placed in all dog bites and removed as drainage declined, usually after 24-48 hours. Sterile gauze dressings were changed 24-48 hours after debridement and cleansing. Stitches were removed from sutured wounds 5-7 days after closure. Antibiotic use for the dog bites that became infected was empiric based on prior literature, because infected wounds were not cultured. Outcomes compared for the 2 groups at P < 0.05 for significance were percent of bites infected, infection time (hours from being bitten to infection diagnosis) and recovery time (days from being bitten to clinical healing).

Results:  Most subjects (53%) were younger than 10 years of age. Study wounds were 3.15 cm in length or 3.61 cm² in area on average. The 2 groups did not differ significantly on infection rate or time to infection diagnosis, but recovery time was faster for closed wounds whether they were infected or not (P < 0.05). Patients with closed wounds had 6.3% infections acquired in a mean of 25 hours, with a recovery time of 6.6. days compared to 8.3% infections acquired in 26 hours, with a recovery time of 9.1 days for wounds left open.  Noninfected closed wounds healed in an average of 6.6 days or 9.1 days for those left open (P < 0.05). Infection delayed healing, but closure shortened that delay to: 10.7 days compared to 14.2 days for open wounds with similar treatment for infection (P < 0.05).

Authors’ Conclusions:, Early primary closure after cleansing and debridement did not increase risk of facial dog bite infection and reduced healing time, whether the wounds become infected or not. 

The 2 studies reviewed here^2,3 on 782 wounds suggest that it may be time to re-examine the recommendation to delay closure¹ of uninfected dog bites involving full-thickness skin managed within 8 hours after the bite.^2,3 The earlier the bite is closed, the less likely it will be infected. As the authors of both studies note, it is important to clean and debride the wounds before closure, but they differed in details of these techniques as well as enrollment criteria. The Beijing study³ included more serious wounds mainly on children 10 years of age or younger, did not use prophylactic antibiotics on admission, and used wound drains during the first 24-48 hours after debridement. Any of these variables or risk factors may have led to its observed 7% incidence of facial dog bite infections, compared to no facial wound infections in the US-based study, which enrolled subjects at least 16 years of age and used prophylactic antibiotics but no wound drains. Further research is needed to determine which of these variables caused the different infection rates in facial dog bites between the 2 studies. This could provide important clues to improving clinical outcomes of all dog bite wounds. Other techniques reported to reduce the likelihood of infection, such as moist wound healing⁴ or procedures to manage excess subcutaneous fluid,⁵ also merit further study. These rigorous studies^2,3 are important early steps on the path to identifying best practice to improve dog bite outcomes. Slowly, clinical science is replacing dogma with evidence to inform management of these challenging wounds.

This article was not subject to the WOUNDS peer-review process.