Use of co-amoxiclav for the treatment of dog bites

Learning Outcomes

After completing this module, the paramedic will be able to:

If you would like to send feedback, please email jpp@markallengroup.com

Physiology of human skin

Skin acts as a physical barrier to the outside world, protecting the body against shearing forces, friction, infection, chemicals and ultraviolet radiation. It prevents excessive absorption or loss of water, and is involved in the synthesis of vitamin D from ultraviolet light. It is also involved in temperature regulation and wound healing (Wilkinson et al, 2017). The anatomy of the skin is shown in Figure 1.

The approximate surface area of an adult's skin is 1.5–2.0 m². The skin consists of three layers: the epidermis, dermis and subcutaneous layer (Phillips, 2008), and varies in thickness depending on anatomical location, age and sex (Bardia, 2017). Skin is thicker in men than women in all anatomical areas; children's skin is thin and progressively thickens over time until the fourth and fifth decade when it starts to thin as elastic fibres are lost (Bardia, 2017). Although large numbers of microorganisms live on the skin, these cannot breach the barrier created by intact, healthy skin (Phillips, 2008).

The epidermis is the protective, waterproof, outermost layer; it consists of four or five layers of cells and has no nerve endings or blood vessels (College et al, 2010). It is thicker on areas exposed to friction and constant wear and tear such as the hands and feet, which have a fifth layer of cells called the stratum lucidum, and has thinner layers on the eyelids and postauricular region (Bardia, 2017).

The epidermis is mainly (approximately 90%) made of structural cells called keratinocytes, which originate from the basal layer as they travel outwards to the surface, where they are shed (College et al, 2010). The keratinocytes are live cells that are capable of dividing to produce new cells to form the epidermis. This migration of keratinocytes from the base cells to the surface takes approximately 30 days in normal skin (Kumar and Clark, 2017). The remaining 10% of the epidermal cells are Langerhans cells, melanocytes and merkel cells (College et al, 2010).

Langerhans cells are located in the suprabasal layer, originate from bone marrow and are concerned with assisting immunoregulation and antigen presentation (Kumar and Clark, 2017). Melanocytes and merkel cells can be found in the basal layer. Merkel cells assist in sensation within the oral cavity/fingertips, with melanocytes synthesising the pigment melanin, which provides protection against ultraviolet radiation by transferring into surrounding keratinocytes (Seeley et al, 2016).

The epidermis is fixed to the medial layer of skin, the dermis, by a mesh of proteins linking keratin filaments of basal keratinocytes to collagen fibres in the superficial dermis (Drake et al, 2015).

The dermis lies beneath the epidermis. It gives the skin its strength, as it is a matrix of loosely packed collagen and elastin fibres, which make it elastic and robust enough to provide a protective function (Phillips, 2008). It contains reticular fibres and specialised cells—sebaceous glands, hair follicles, sensory nerve fibres, sweat glands and blood capillaries that supply oxygen and nutrients (Glynn and Drake, 2012).

The plentiful blood supply in the superficial and deep plexuses consists of arterial and venous capillaries, arterioles and venules, which—combined with hair, sweat glands and arrectores pilorum muscles—are integral to thermoregulation (College et al, 2010). Immune system cells are also present, ready to respond to any damage with fibroblasts; these create and maintain the extracellular matrix by producing collagen that is continuously being remodelled and moves around the dermal matrix to maintain a steady state of homeostasis (Open University, 2006).

The subcutaneous layer is made of adipose tissue and also contains blood vessels and nerves, which are separated into lobules by fibrous septa (Phillips, 2008). This provides protection against trauma, provides insulation, and contains approximately 80% of all body fat (Kumar and Clark, 2017).

Pathophysiology of human skin following a dog bite

A dog bite is inflicted by the mouth and teeth causing bruising and deep anatomical structure disruption, as the protective layers of the epidermis and dermis are broken. This allows infectious agents to enter, with microorganisms deposited into the wound that originate from the microflora in dog's mouth, as well as from the outside of the skin, clothing or environment of the victim (McPhee and Hammer, 2010). As dogs can exert more than 400 lb/in² of pressure with their jaws, they can, in addition to inflicting puncture wounds and laceration, cause crush injury, avulsion and tissue devitalisation (Morgan and Palmer, 2008).

Trauma to the skin causes an inflammatory response, increasing blood flow and transporting white blood cells and macrophages to the site of the bite to repair the tissue and fight infection, which is characterised by swelling, pain, pyrexia and heat. Blood from damaged blood vessels in the skin fills the wound area and starts to clot, forming a mesh of fibres that acts like a temporary extracellular matrix (Underwood, 2007). These are formed by fibrin instead of the normal extracellular matrix proteins such as collagen. These fibrin clots act as a temporary repair until fibroblasts arrive to make new extracellular matrix proteins such as collagen (Phillips, 2008).

Spencer and Banerjee (2019) state that many wounds sustained from dog bites become infected with an incidence of approximately 3–18% within 24 hours of injury, with the infections likely to be polymicrobial (a mix of aerobes and anaerobes) with a median of five isolates per wound. The Infectious Diseases Society of America (IDSA) (2014) notes that discharging (purulent) dog bite wounds are more likely to yield a mix of aerobes and anaerobes (polymicrobial) than non-purulent wounds, which are more likely to produce Staphylococci and Streptococci. In addition to this, Pasteurella are commonly isolated from both purulent and non-purulent wounds, although non-purulent wound infections may also be polymicrobial (IDSA, 2014).

Morgan and Palmer (2008) suggest that the Pasteurella are the most infectious pathogens in a dog bite as they are aggressive Gram-negative pathogens and can cause an intense early inflammatory response in a significant amount of tissue; it is likely to cause metastatic infection with a 30% mortality rate if septicaemia develops as a result. Pasteurella are present in more than 50% of dog bites and are the most likely infective source in wounds that present within 12 hours of the bite occurring (Morgan and Palmer, 2008).

Pharmacology

Co-amoxiclav (Augmentin, GlaxoSmithKline UK) contains two active ingredients: amoxicillin and clavulanic acid. Amoxicillin is part of the penicillin pharmacotherapeutic group, which are effective against a wide range of bacteria; they are sometimes referred to as broad-spectrum antibiotics and are used when the exact strain of the infection cannot be identified (Young and Pitcher, 2016).

Amoxicillin kills bacteria by inhibiting the synthesis of the peptidoglycan wall. A key part of its molecular structure is the beta-lactam ring, which is essential for antimicrobial activity (Neal, 2015). The beta-lactam action of amoxicillin prevents crosslinks from forming within the cell wall by inhibiting the formation of covalent bonds with penicillin-binding proteins that have carboxypeptidase and transpeptidase activities (Rang et al, 2016). This inhibits the final phase of transpeptidation by inhibiting enzymes that create crosslinks between the peptide chains that are attached to the peptidoglycan wall; this prevents growth and division of the bacteria, as well as weakening the cell wall, causing cell lysis and death (Dale and Haylett, 2008).

Amoxicillin is more hydrophilic than other penicillins such as benzylpenicillin and has the ability to pass through the outer phospholipid membrane so is active against some Gram-negative bacteria (e.g. Pasteurella) (Neal, 2015; GlaxoSmithKline UK, 2018). It is also effective against some Gram-positive aerobes and anaerobes (e.g. Staphylococci and Streptococci) (Medicines and Healthcare products Regulatory Agency (MHRA), 2009; GlaxoSmithKline UK, 2018). However, this action can be inhibited by beta-lactamase-producing bacteria strains, causing resistance as it disrupts the beta-lactam ring (Dale and Haylett, 2008; MHRA, 2009; Rang et al, 2016).

Clavunic acid is a beta-lactamase molecule that is produced by Streptomyces clavuigerus, a species of Gram-positive bacterium; clavunic acid has a high affinity for beta-lactamase (MHRA, 2009).

Clavulanate molecules bind irreversibly to bacterial beta-lactamase after penetrating the bacterial cell (Downie et al, 2007). When clavulanic acid is combined with amoxicillin (co-amoxiclav), the clavulanic acid protects amoxicillin from inactivation, allowing it to be effective against penicillinase-producing organisms (Neal, 2015).

Co-amoxiclav can be given orally (in tablets or a suspension) or intravenously; both components are well absorbed in the gastrointestinal tract (GlaxoSmithKline UK, 2018), with absorption unaffected by the presence of food in the stomach (Joint Formulary Committee, 2019).

Both amoxicillin and clavulanic acid have a peak plasma concentration at approximately 1 hour with a similar half-life of 1 hour also. However, bioavailability differs between them, with amoxicillin's at 72–94% and clavulanic acid's at approximately 60% (MHRA, 2009; GlaxoSmithKline UK, 2018).

Co-amoxiclav is widely distributed throughout the body into body fluids, joints, pericardial area, pleural cavities, saliva and bile. It does not, however, enter mammalian cells, cerebrospinal fluid or cross the blood-brain barrier (Rang et al, 2016; GlaxoSmithKline UK, 2018).

Both amoxicillin and clavulanic acid can enter breast milk and can cross the placental barrier (GlaxoSmithKline UK, 2018), although the Joint Formulary Committee (2019) states that co-amoxiclav is appropriate to use in pregnancy and when breastfeeding, as it is not known to be harmful. It also states that caution and close monitoring are needed in patients with any hepatic impairment, as both components are metabolised by the liver (Joint Formulary Committee, 2019).

Clavulanic acid is excreted both by renal and non-renal means, with amoxicillin being excreted via the kidneys so a dose adjustment is needed in renal impairment if the estimated glomerular filtration rate is <30 ml/minute/1.73 m² (GlaxoSmithKline UK, 2018; Joint Formulary Committee, 2019).

The main adverse drug reaction is allergy, which ranges from a rash to life-threatening anaphylaxis (Young and Pitcher, 2016). In addition, side effects such as gastrointestinal disturbances can occur following oral consumption as the bacterial flora of the gut is altered, which can lead to Clostridium difficile infection (National Institute for Health and Care Excellence (NICE), 2015) and antibiotic-associated colitis (Joint Formulary Committee, 2019).

Co-amoxiclav is relatively cost-effective to use orally in both tablet and suspension form (£1.77–£6.97 per course of treatment) with the powder solution for injection being more expensive (£10.60–£29.70 per 10 vials); its shelf life in all these forms is 24 months (Joint Formulary Committee, 2019).

Summary of evidence

The use of co-amoxiclav orally for 7 days, including for prophylaxis treatment following a dog bite, is recommended by NICE (2018) and the Joint Formulary Committee (2019). NICE (2018) states that this choice of antibiotic is based on expert opinion in Public Health England's (PHE) (2019)Summary of Antimicrobial Prescribing Guidance—Managing Common Infections. This publication is supported by IDSA's (2014) guidelines on the diagnosis and management of skin and soft tissue infection. The PHE (2019) publication uses essentially three papers as its evidence base: a Cochrane review (Medeiros and Saconato, 2001); the IDSA (2014) guidelines; and a review of systematic reviews/clinical evidence/NHS guidelines (Morgan and Palmer, 2007). These three appear to form the main evidence for the prophylactic antibiotic treatment of animal bites including dog bites in the UK.

Cochrane reviews of randomised controlled trials (RCTs) are widely regarded as evidence that provides accurate, up-to-date guidance for interventions and treatments (Aveyard et al, 2016). These reviews sit at the very top of the hierarchy of evidence (Bettany-Saltikov and McSherry, 2016), as they are highly regarded as the most trustworthy and reliable evidence-based reviews (Aveyard, 2007).

Medeiros and Saconato's (2001) Cochrane review now appears out of date as it was based on six studies of dog bites ranging from the 1980s to the 1990s. The numbers of participants were also low, ranging from fewer than 40 to just above 100, which Polit and Beck (2006) state affects the credibility of the studies as the numbers are not large enough to demonstrate any substantial findings. Within all six studies, the participants who were bitten by dogs were split into two groups: antibiotic and placebo groups. One group received prophylactic antibiotics after sustaining a dog bite to the hand, while the other had a placebo.

As the studies were all RCTs, these groups were formed by randomisation. The ethical principles of beneficence and non-maleficence (minimising harm while promoting good) (Parahoo, 2006) potentially may have been overlooked as there is no mention of any ethics committee involvement within the studies, which would have ensured that ethical principles were adhered to and upheld. However, as ethics committees were not formalised until 1991, this lack of reference to ethics committee involvement is most likely down to the age of the studies rather because ethical principles were overlooked (Rolfe, 1998).

The antibiotics used in the treatment groups were narrow spectrum, beta-lactam antibiotics: phenoxymethylpenicillin, first-generation cephalosporins and co-trimoxazole (sulfamethoxazole/trimethoprim). This makes for an interesting point of discussion as this Cochrane review forms part of up-to-date, evidence-based guidelines for the use of co-amoxiclav (a beta lactamase-inhibiting antibiotic) for the treatment of dog bites, although it was not used in any of the six studies in this review. The results from these studies in this Cochrane review have positive findings, showing a reduction in the rate of infection when prophylactic antibiotics used, but limited data affects their significance, which suggests further trials are needed (Medeiros and Saconato, 2001).

A more recent review of the IDSA 2005 guidelines included in the IDSA (2014) update for the management of skin and soft-tissue infections form part of the PHE (2019) publication for the summary of antimicrobial prescribing guidance. IDSA is an American society that represents health professionals involved in infectious diseases and regularly produce guidelines (British Infection Association, 2019). Evidence from eight RCTs of dog bite wounds were analysed for these guidelines.

There were, however, some serious limitations within these, including the low number of patients involved (between 12 and 190 per study), preventing sufficient amounts of data from being gathered. In addition, four of the eight studies did not specify the method of randomisation employed, which Katz (2006) suggests could be linked to a form of bias as the authors could have ‘cherry picked’ the participants, thereby unfairly influencing the results.

Even though there are limitations with the studies, the Cochrane study (Medeiros and Saconato, 2001) authors concluded that positive early prophylactic antibiotic intervention was effective, especially with immunocompromised/asplenic patients or those with advanced liver disease, oedema following the bite to the affected area or a dog bite to the face or hands. They advocated the use of oral co-amoxiclav as it is effective in treating polymicrobial bacteria (aerobes and anaerobes), Staphylococci, Streptococci and Pasteurella, which are commonly associated with dog bites (IDSA, 2014), stating that macrolides should not be used as they have variable activity against Pasteurella.

This view is echoed by Morgan and Palmer (2007), who reviewed the 2001 Cochrane review and a meta-analysis of another study undertaken in 1994. They added that erythromycin or flucloxacillin should never be used prophylactically as Pasteurella infection is usually resistant to them. Within these studies, again, the numbers of participants are limited, with eight trials being reviewed, six of which were common to both studies. Only one trial of 172 people involved giving co-amoxiclav prophylactically following a dog bite to 84 of those patients, with 88 being given a placebo. Of the patients who were given co-amoxiclav, 33% became infected compared with 60% who received the placebo, which supports the effectiveness of the prophylactic use of co-amoxiclav.

Conclusion

The use of co-amoxiclav (including prophylactically) following a dog bite has been shown to be the most effective antibiotic treatment, as it covers the most likely polymicrobial aerobic and anaerobic organisms that infect dog bite wounds (Staphylococci, Streptococci and Pasteurella) (IDSA, 2014). However, the evidence base used by NICE (2018), PHE (2019), IDSA (2014) and local health board guidelines (Rx Guidelines, 2019) for the treatment of dog bites is dated, which demonstrates that further research is required involving larger numbers of participants to provide some significant data regarding the use of co-amoxiclav for the treatment of dog bites.

How this could be achieved ethically in any future trial would need to be carefully planned with the inclusion of patient randomisation methods to minimise bias. Currently, the use of co-amoxiclav is closely monitored by health boards as gastrointestinal disturbances can occur, leading to Clostridium difficile infection (NICE, 2015; All Wales Medicines Strategy Group, 2018), so considering beneficence and non-maleficence is essential when prescribing.

Co-amoxiclav is cost-effective to prescribe and simple for patients to use as there is no need to take additional medication, as would be the case if the patient had a penicillin allergy and had to take two antibiotics concurrently (e.g. metronidazole and doxycycline) (Rx Guidelines, 2019); having only one medication for a patient to take improves compliance (IDSA, 2014; NICE, 2018).

The importance of a person-centred, holistic and responsible approach to prescribing is vital for the effective use of co-amoxiclav to prevent infection with careful management and knowledge of the potential for adverse effects (Royal Pharmaceutical Society, 2016), although it has proven to be effective as a first-line, prophylactic oral antibiotic treatment following a dog bite.