It's well known that a dog's mouth is a much less filthy place than that of a human's. This is evidenced by the fact that prohylactic antibiotics don't seem to reduce rates of infection after dog bites, but most certainly do after human ones [1].
There is a very good break down of the evidence for and against empiric antibiotic prophylaxis in the case of dog bites provided at First10EM which I won't bother regurgitating here for various reasons, pointlessness and plagiarism being the two main ones. What I will say is that the decision to use antibiotic prophylaxis is a more defensible one when in a remote environment. This is both because the likelihood of infection is higher (poorer ability to debride wounds) as are the consequences of said infection (termination of the expedition or requirement for resource intensive retrieval).
Ignoring rabies for a second, the pathogen seemingly most of concern is that of Pasteurella multocida - the gram negative facultative anaerobe susceptible to amoxicillin clavulanate (henceforth referred to by its less lengthy ex-brand name "Augmentin"). It is the bacteria most commonly seen as being synonymous with mammalian bites, and with some good reason - it's found in about half of the infected bites that get swabbed. But so are lots of bugs - dog bites are very polymicrobial: a mixture of canine oral flora (nasty anaerobes like Prevotella, Fusobacterium, and Bacteroides) as well as all that juicy human skin flora (nasty aerobes like Streptococci and Staphylococci). The treatment of infected bites may have less to do with it's specific role in the antibiotic susceptibility patterns of Pasteurella, and more to do with the fact that it's a convenient broad spectrum antibiotic that covers lots of things.* In the case of penicillin allergy, a suitable alternative would be doxycycline, given that it too has activity against aerobes and anaerobes, and is likely to be readily available, especially in tropical countries.
However, the bitten party should always have a thorough history taken to ascertain their risk of systemic infection. This extends beyond whether they are diabetic or currently receiving chemotherapy - the last of which shouldn't be dismissed out of hand, as patients with terminal conditions are perhaps more likely to want to engage in bucket-list activities than their healthy counterparts [3]. Important considerations include whether they are a heavy consumer of alcohol, on long term corticosteroids, and whether there is asplenia (either surgical or functional). These are the three major risk factors for developing systemic infection with Capnocytophaga canimorsus, an encapsulated gram negative anaerobe that frequently results in fulminant gram negative sepsis. These patients should definitely be given their empiric Augmentin prophylaxis, as Capnocytophagia is also susceptible.
Systemic unwellness usually presents around three days after the index bite, but can occur at any point in the following two weeks - long after the memory of a more minor bite may have faded. It features non-specific constitutional symptoms like fever and malaise, sometimes with headache, vomiting, diarrhoea, and abdominal pain. This is important, as it can lead to misdiagnosis as a febrile gastroenteritis. Only a third will have cutaneous features present at the site of injury, which when present will appear as localised purpura that can progress to gangrene [4]. Overwhelming infection and disseminated intravascular coagulation can occur, as well as seeding to distant sites that results in complications like meningitis and endocarditis. Textbooks and UpToDate both suggest a mortality rate of 30% but case reports have shown slightly more variable numbers. Beauruelle et al (n=44) found 11% died, though not all infections were due to bites [5], van Dam et al (n=32) found 13%, and Pers et al (n=39) found a mortality rate of 31% [7]. A rough-and-ready-napkin-math meta-analysis of these data would suggest a combined mortality rate of about 17%.
Capnocytophaga species have been shown to produce a whole host of different broad-spectrum beta lactamases, meaning cephalosporins are ineffective (up to and including third generation ones like ceftriaxone)[8]. Treatment with a combination of Augmentin with or without clindamycin have been shown to be consistently effective in vitro [9]. Given both the severity and undifferentiated nature of infection, in the remote environment we would recommend both.
In summary:
Patients presenting with a febrile illness or gastroenteritis in the 1-2 week period after dog bite should have infection by Capnocytophaga strongly considered, especially if immunocompromised, asplenic, or alcoholic, though it can also affect immunocompetent individuals [10]. Strongly consider aggressive broad spectrum antibiotic therapy and evacuation for intensive care level support and debridement in the case of local gangrene.
Notes:
*It should be said - sporadic case reports have arisen where Augmentin-resistant beta-lactamase-producing strains of Pasteurella has been isolated, though these were still susceptible to fluoroquinolones and tetracyclines [11].
Credit:
DR1
References:
[1] Medeiros I, Saconato H. Antibiotic prophylaxis for mammalian bites. The Cochrane Database of Systematic Reviews. 2001:(2):CD001738.
doi: 10.1002/14651858.CD001738.
[2] Dog bite management: the evidence. First10EM. 2019. Available at: Dog Bite Management: The Evidence - First10EM
[3] British cancer patient reaches Everest summit. The Guardian. 2017. Available at: British cancer patient reaches Everest summit | Sheffield | The Guardian.
[4] Walls RM, Hockberger RS, Gausche-Hill M, et al. Rosen's Emergency Medicine. 9th Ed. Elsevier. 2018.
[5] Beauruelle C, Plouzeau C, Grillon A, et al. Capnocytophaga zoonotic infections: a 10-year retrospective study (the French CANCAN study). Eur J Clin Microbiol Infect Dis. 2022;41(4):581.
[6] van Dam AP, Jansz A. Capnocytophaga canimorsus infections in The Netherlands: a nationwide survey. Clin Microbiol Infect. 2011;17(2):312-5.
[7] Pers C, Gahrn-Hansen B, Frederiksen W. Capnocytophagia canimorsus septicemia in Denmark, 1982-1995: review of 39 cases. Clin Infect Dis. 1996;23(1):71.
[8] Jolivet-Gougeon A, Tamanai-Shacoori Z, Desbordes L, et al. Genetic analysis of an Ambler class A extended-spectrum beta-lactamase from Capnocytophaga ochracea. J Clin Microbiol. 2004;42(2):888-90.
[9] Jolivet-Gougeon A, Sixou JL, Tamanai-Shacoori Z, et al. Antimicrobial treatment of capnocytophaga infections. Int J Antimicrob Agents. 2007;29(4):367-73.
[10] Nakayama R, Miyamoto S, Tawara T, et al. Capnocytophaga canimorsus infection led to progressively fatal septic shock in an immunocompetent patient. Acute Med Surg. 2022;9(1):e738.
[11] Wei A, Dhaduk N, Taha B. Wrist abscess due to drug-resistant Pasteurella multocida. IDCases. 2021;26:e01277.