• 
• 

Three Part Question

In [ED patients presenting with bite wounds (dog, cat, human)] are [antibiotics] indicated in all patients to [prevent wound infection]?

Clinical Scenario

An 18-year-old man presents to the Emergency Department having been bitten by his neighbour's dog three hours previously. He has a simple but ragged wound without signs of infection or inflammation. He is normally fit and healthy with no regular medications or allergies and has been immunised in accordance with the National Immunisation Programme (including five doses of tetanus immunisation).

You wonder whether you should use simple wound care and irrigation alone or whether he should be discharged with prophylactic antibiotics, in conjunction with safety netting advice.

Search Strategy

MEDLINE PubMed (1950-present). The search was limited to human studies and studies in English.
Cinahl Plus Full Text (1937–present)
Cochrane Library including Central Register of Controlled Trials
Embase (1974‐present)
Emerald
ERIC (1960’s-present)
EBM Reviews (1991–present)
Scoops

Additional efforts to locate RCTs were identified from the following data sources:
1. References cited in identified RCTs
2. Google Scholar
3. Register of clinical trials (http://www.who.int/ictrp/en)
4. Textbooks
5. “Grey literature”:
a. SIGLE (system for information on grey literature in Europe)
b. ZETOC
A keyword search was performed using the following terms:
1) Emergency Department OR Accident and Emergency Department OR Accident and Emergency OR Emergency Room OR ED OR ER
2) Bites OR Dog Bites OR Cat Bites OR Human Bites
3) Wounds
4) Infection
5) Adults
6) Child

Search Outcome

Initial searches revealed 6786 articles. These were reviewed by title, abstract and paper and duplicates were removed, resulting in

1 SYSTEMATIC REVIEW
1 META-­ANALYSIS
10 RCTS

Relevant to the search question

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Quinn et al 2010 USA	94 randomised 48 to antibiotic group 46 to control (placebo) group Inclusion: All dog bites regardless of site Adults/children Age 31‐34 (mean) Exclusion: Wounds > 12 hours at presentation or already infected Immunosuppression Penicillin allergy Wounds with suspected neurovascular, tendon, joint or bony injury	Double blinded randomised controlled trial 3/7 of Amoxicillin-­clavunanic acid versus placebo prepared by pharmacy Follow up: Phone call at 14 days	Infection rates	2%: 0/48 in antibiotic group, 2/48 in placebo group	Under powered Compliance not assessed 42% randomised out of 230 recruited Infection was determined by the patient Included high risk wounds such as puncture and extremities without breakdown of wound site
Dire et al 1992 USA	185 randomised 89 received antibiotics, 99 control group Inclusion Dog bite Children Age 1-16 Exclusion Puncture wounds Hand/foot wounds Wounds >12 hours Immunosuppression Use of antibiotics in previous 7/7 Unable to obtain informed consent	Prospective randomised controlled trial 1/52 of oral dicloxacillin/cephalexin or erythromycin plus wound care versus wound care only Follow up: wound re-evaluation (timescale unclear) Standardised bite wound care sheet	Infection rates	1/89 with antibiotics (1.1%), 5/96 in control group (5.2%)	Underpowered, no blinding, compliance with treatment regime not assessed
Skurka et al 1986 USA	39 patients randomised 19 to antibiotics 20 to control group Inclusion Children aged 1-16 years Dog bites presenting within 24h Exclusion Obvious wound infection Penicillin allergy Antibiotics in preceding 3/7 Indications for hospital admission	Double blinded, randomised controlled trial 2/7 of liquid penicillin (100,000U/kg/day) versus placebo liquid plus local wound care Follow up at 48-72h then at 7-10/7 with enrolling clinician Defined protocol for wound care	Infection rates	2/19 for antibiotic group (1.1%), 1/20 for control group (5%)	Small sample Results given as percentage rather than numbers; no P value or confidence intervals calculated
Rosen 1985 USA	66 patients randomised 35 to antibiotic group 31 to control Inclusion: All ages Dog bites penetrating the dermis Wound <8hrs old Exclusion: Unable to swallow capsules Wounds involving bone, tendon, tendon sheath or major neuromuscular structures	Randomised, blinded controlled trial 5/7 of cloxacillin, dicloxacillin or erythromycin versus local wound care plus placebo Follow up: 48-72hrs Strict wound care protocol	Infection rates	2/35 in antibiotic group (5.7%), 3/31 in control group (9.7%)	Compliance with medication regime not assessed Preceding antibiotic use not excluded
Jones et al 1985 UK	113 randomised 55 to antibiotics 58 to control group Inclusion Dog bite wounds Age >3 years Exclusion Superficial abrasions Other condition requiring antibiotics Allergy to co-trimoxazole	Randomised, placebo-controlled, double-blind trial Co-trimoxazole x 5/7 versus placebo Follow-up: reassessed at 7/7	Infection rates	3/55 in antibiotic group (5.5%), 8/58 in control group (13.8%)	Underpowered Large number lost to follow up Randomisation process not outlined
Boenning et al 1983 USA	55 patients randomised 25 to antibiotics 30 to control Inclusion Children Wounds <24h old Wounds not requiring closure No history of penicillin allergy No concurrent antibiotic use Exclusion Facial wounds	Randomised controlled trial Phenoxymethylpenicillin x 5/7 plus local wound care versus wound care alone Follow up: day 2 and day 5 or as needed by patient	Infection rates	1/25 in antibiotic group (4%), 1/30 in control group (3.3%)	No blinding, poor randomisation, no placebo
Elenbaas 1982 USA	46 patients randomised 22 to antibiotics 24 to control Inclusion Wound <24 hours old Adults Full‐thickness (unequivocal penetration of the subcutaneous tissue or underlying muscle layer) Exclusion Requiring hospitalisation Violation of periosteum On antibiotics for another reason Penicillin allergy	Prospective randomised double-blinded placebo-controlled trial Oxacillin x 5/7 plus local wound care versus oral placebo and wound care Follow up: every 2 days for minimum of 5/7 until all wounds healed	Infection rates	2/22 in antibiotic group (9%), 0/24 in control group (0%)	Randomisation process not described therefore unsure regarding allocation concealment Small sample size (46) High drop out rate (17/63)
Elenbaas 1984 US	12 patients randomised 5 to antibiotics 6 to control 1 lost to follow up Inclusion Cat bites Wound <24 hours old Adults Full‐thickness (unequivocal penetration of the subcutaneous tissue or underlying muscle layer) Exclusion Infected wounds On antibiotics for another reason Penicillin allergy	Prospective randomised double-blinded placebo-controlled trial Oxacillin x 5/7 plus local wound care versus oral placebo and wound care Follow up: review at 24 hours then every 2 days for 5/7 until all wounds healed	Infection rates	0/5 in antibiotic group (0%), 4/6 in control group (66.7%)	Small sample size
Zubowicz 1991 USA	48 patients randomised 16 to oral antibiotics 17 to IV antibiotics 15 to control Inclusion Human bites Wound <24 hours old Bite not infected; no other concurrent infection Bite not penetrating joint capsule No tendon injury Exclusion [none given]	Prospective randomised controlled trial Oral cefaclor or IV cefazolin and penicillin G plus local wound care versus local wound care alone Follow up: daily wound review for 5/7	Infection rates	0/33 in antibiotic group (0%), 7/15 in control group (47%)	No blinding No power calculation Doesn’t say how long study lasted Compliance not assessed All patients were admitted Risk of bias No matching of baseline characteristics Unclear ownership of study: ED clinician or plastic surgeons?
Broder 2003 USA	127 patients randomised 63 to antibiotics 62 to control 2 lost to follow up Inclusion Human bites not involving hands or feet or overlying cartilaginous structures Exclusion Immunocompromised Age < 18 years Penicillin allergy Bite > 24 hours	Prospective double-blinded placebo-controlled trial Oral cephalexin/penicillin plus local wound care versus placebo and local wound care Follow up: wound review at 48 and 96 hours	Infection rates	0/63 in antibiotic group (0%), 1/62 in control group (1.6%)	Small sample size Study underpowered

Comment(s)

The Cochrane Systematic Review, which included 6 of these studies, concluded that the use of prophylactic antibiotics is not associated with a statistically significant reduction in infection. Nearly all studies included in this review emphasized the importance of local wound care to prevent infection. They also agreed that initial culture of wounds does not predict which wounds will subsequently develop a wound infection. Wounds including puncture wounds and hand wounds appear to have a higher risk of infection. In the majority of cat bites, the mechanism appears to be a puncture wound, making them potentially high risk. Unfortunately, all the studies appraised have a small number of subjects and are underpowered to detect a statistically significant difference. There is little evidence from the studies appraised on dog bites that prophylactic antibiotics reduce the risk of infection in non-hand/puncture wounds.

Editor Comment

BF

Clinical Bottom Line

Wound debridement and irrigation is an important component of bite wound management and infection prevention. Puncture and hand wounds appear to be high risk, therefore antibiotics are recommended. However, there is little evidence that low risk dog bites should be treated with antibiotics. There is insufficient evidence on cat and human bites. Further large multicentre RCTs are needed to answer the study question.

References

1. Quinn JV, McDermott D, Rossi J et al. Randomised controlled trial of prophylactic antibiotics for dog bites with refined cost model. West J Emerg Med. 2010 Dec;11(5):435-­‐41.
2. Dire DJ, Hogan DE, Walker JS. Prophyactic oral antibiotics for low‐risk dog bite wounds. Pediatr Emerg Care. 1992 Aug;8(4):194-­‐9.
3. Skurka J, Willert C, Yogev R. Wound infection following dog bite despite prophylactic penicillin. Infection. 1986 May-­Jun;14(3):134‐5.
4. Rosen RA. The use of antibiotics in the initial management of recent dog‐bite wounds. Am J Emerg Med. 1985 Jan;3(1):19-23.
5. Jones DA, Stanbridge TN. A clinical trial using co-­trimoxazole in an attempt to reduce wound infection rates in dog bite wounds. Postgrad Med J. 1985 Jul;61(717):593‐4.
6. Boenning DA, Fleisher GR, Campos JM. Dog bites in children: epidemiology, microbiology, and penicillin prophylactic therapy. Am J Emerg Med. 1983 Jul;1(1):17-21.
7. Elenbaas RM, McNabney WK, Robinson WA. Prophylactic oxacillin in dog bite wounds Ann Emerg Med 1982 May;11(5):248-­51.
8. Elenbaas RM, McNabney WK, Robinson WA. Evaluation of prophylactic oxacillin in cat bite wounds. Ann Emerg Med. 1984 Mar;13(3):155‐7.
9. Zubowicz VN, Gravier M Management of Early Human Bites of the Hand: A Prospective Randomised Study. Plastic and Reconstructive Surgery 1991 Jul;88(1):111-4.
10. Broder J, Jerrard D, Olshaker J, Witting M. Low risk of infection in selected human bites treated without antibiotics. Am J Emerg Med. 2004 Jan;22(1):10-3.