Background

Many bacteria produce β-lactamase enzymes that break down, and confer resistance to, β-lactam antibiotics such as amoxycillin. Co-amoxiclav contains a mixture of amoxycillin plus a β-lactamase-inhibitor, clavulanic acid. In vitro, the addition of clavulanic acid extends the spectrum of amoxycillin to include some otherwise resistant β-lactamase-producing bacteria such as some strains of Escherichia coli, Haemophilus influenzae and Moraxella catarrhalis, penicillin-resistant (but not methicillin-resistant) Staphylococcus aureus and β-lactamase-producing strains of Neisseria gonorrhoeae and anaerobic bacteria such as Bacteroides fragilis. Co-amoxiclav is not active against penicillin-resistant pneumococci or enterococci, whose resistance is not β-lactamase-mediated.

Co-amoxiclav preparations

Co-amoxiclav is available as standard and dispersible tablets, as a paediatric suspension and as a powder for reconstitution to be given by i.v. injection or infusion. These formulations are available in various strengths and are recommended for administration three times daily. The standard 375mg tablet contains amoxycillin 250mg plus clavulanic acid 125mg. In the 625mg tablet and the standard paediatric suspension the ratio of amoxycillin to clavulanic acid is 4:1; for the i.v. preparations it is 5:1. Recently, co-amoxiclav became available as a suspension licensed for twice-daily administration (Augmentin-Duo) and recommended for use in children aged 2 months and over. Each 5ml dose contains amoxycillin 400mg plus clavulanic acid 57mg (7:1). The recommended dosage gives a higher daily dose of amoxycillin and a slightly lower daily dose of clavulanic acid than does the standard paediatric suspension. In one study, Augmentin-Duo given twice daily was as effective as the standard paediatric suspension given three times daily for the treatment of children with mild-to-moderate lower respiratory tract infection.2 In this study, however, microbiological information was only available for a minority of patients.

Depending on the formulation used, the cost of co-amoxiclav is about 2-11 times that of the equivalent dose of generic amoxycillin alone.

Clinical studies

A review of 415 clinical trials published between 1979 and 1992 suggests co-amoxiclav cured or improved about 90% of various infections treated.3 The efficacy of co-amoxiclav seems to have changed little over the years.3

Respiratory infections

Community-acquired pneumonia - Streptococcus pneumoniae is the commonest bacterial cause, being isolated in 50-70% of cases. Other non-viral causes include H. influenzae particularly in patients with chronic lung disease, Mycoplasma pneumoniae, Chlamydia pneumoniae, Chlamydia psittaci or Legionella pneumoniae.

S. pneumoniae is generally sensitive to amoxycillin (and penicillin, with about 4% of isolates being resistant4) but there may be local variation associated with higher resistance rates.5 Amoxycillin remains the treatment of first choice for adults with uncomplicated community-acquired pneumonia.6 If treatment with amoxycillin results in no improvement within 24-48 hours, a macrolide should be added to cover atypical organisms or β-lactamase-producing H. influenzae. During influenza epidemics treatment may need to include flucloxacillin because of the likelihood of secondary pneumonia caused by Staph. aureus.

Acute exacerbations of chronic lung disease - H. influenzae is the commonest causative organism. Amoxycillin is the drug of choice. Patients who do not respond to amoxycillin should be suspected of infection with resistant H. influenzae (the prevalence of which is about 10%7), M. catarrhalis or S. pneumoniae. Resistance to S. pneumoniae is not β-lactamase-mediated. Therefore, where initial treatment with amoxycillin has failed, a second-line drug which has activity against both resistant S. pneumoniae and H. influenzae, such as a macrolide, should be used. A macrolide should be prescribed for a patient with penicillin allergy.

The commonest use of co-amoxiclav in the UK is for the treatment of community-acquired respiratory tract infections. However, the use of co-amoxiclav may be inappropriate and usually offers no advantage over established first- and second-line treatments.

Ear, nose and throat infections

Otitis media - The usual bacterial causes are S. pneumoniae, H. influenzae and M. catarrhalis.8 About 80-90% of M. catarrhalis isolates in the UK are β-lactamase producers.8 Amoxycillin is the usual first-line therapy for otitis media in children. Co-amoxiclav is a second-line treatment for infections caused by amoxycillin-resistant H. influenzae or M. catarrhalis in those who have failed to respond to amoxycillin. Co-amoxiclav is as effective as cephalosporins such as ▼ceftibuten9 and cefaclor.10

Sinusitis - Co-amoxiclav appears no more effective than amoxycillin in treating acute sinusitis in children.11 Amoxycillin therefore remains first-line treatment for this condition.

Surgical prophylaxis

A meta-analysis of 21 randomised, comparative trials (involving a total of 4905 patients) of antibiotic prophylaxis for surgery (chiefly abdominal and gynaecological) showed i.v. co-amoxiclav to be as effective as other antibiotics (typically metronidazole plus either gentamicin or a cephalosporin) in preventing wound infections.12 On this evidence therefore, co-amoxiclav is a reasonable first-choice antibiotic prior to abdominal or gynaecological surgery for preventing wound infections, particularly since it allows use of a single product rather than two separately administered preparations.

Urinary tract infections

In the UK, about 45% of urinary pathogens are resistant in vitro to amoxycillin.13 Resistance of urinary pathogens to trimethoprim is about 25% and is increasing.13 Resistance to nitrofurantoin is still low, possibly because it is used less than other antibiotics. About 6% of urinary pathogens in the UK are resistant to co-amoxiclav.14 In other European countries co-amoxiclav resistance rates are higher: 10-25% of E. coli isolates in France15 and about 25% of urinary pathogens in Italy.16

In a randomised, double-blind study involving 67 patients over 69 years old, co-amoxiclav cured 80% of those with amoxycillin-resistant organisms while amoxycillin cured only 10%.17 There are no such published studies comparing co-amoxiclav with trimethoprim, nitrofurantoin, ciprofloxacin or cephalosporins. For empirical therapy prior to the availability of urine culture results, trimethoprim remains the treatment of choice. Co-amoxiclav is one of a choice of second-line drugs for infections caused by bacteria shown to be resistant to initial trimethoprim therapy.

Animal and human bites

Bacteria that commonly infect dog bite wounds include Pasteurella multocida, Staph. aureus, Staphylococcus intermedius, α-haemolytic streptococci, Capnocytophaga canimorsus (DF-2) and anaerobic bacteria. Those infecting human bites include α-haemolytic streptococci, Eikenella corrodens, Haemophilus species and anaerobic bacteria. Co-amoxiclav has a broader spectrum of activity than penicillin against these organisms.18 In one study, oral co-amoxiclav was found to be more effective than placebo in preventing infection after animal bites which were 9-24 hours old.19 Co-amoxiclav is a reasonable first-line empirical treatment for bites and should be used with wound cleansing and, if indicated, tetanus prophylaxis.

Dentistry

Oral co-amoxiclav 375mg three times daily was compared with penicillin V 250mg four times daily in a randomised trial of the treatment of acute dento-alveolar abscess in 92 patients.20 All patients improved but symptoms resolved more rapidly in patients given co-amoxiclav. Penicillin V therapy is still appropriate for minor dental infections but co-amoxiclav should be considered for more severe infections as it has activity against anaerobes.

Postpartum infection

Co-amoxiclav (500mg orally or 1.2g i.v. every 8 hours) was compared with ampicillin (500mg orally or 2g i.v. every 8 hours) plus metronidazole (500mg orally or i.v. every 8 hours) with or without an aminoglycoside (type not reported) (1.5mg/kg i.m. every 12 hours) in the treatment of 101 women with suspected postpartum endometritis. Co-amoxiclav and the combination were equally effective in relieving clinical signs of endometritis including fever.21 Co-amoxiclav allows the use of single (rather than multiple) preparation treatment and has activity against anaerobes. Co-amoxiclav is therefore a reasonable first-choice empirical treatment for postpartum infection.

Pelvic inflammatory disease

In a study of 60 women with pelvic inflammatory disease (PID), co-amoxiclav was as effective as triple therapy (ampicillin plus gentamicin plus metronidazole).22 Co-amoxiclav is not effective against Chlamydia (which is commonly implicated in PID in the UK) and therefore inappropriate for use on its own for the treatment of PID.

Gonorrhoea

Co-amoxiclav can be an effective treatment for gonorrhoeal infection caused by penicillinase-producing N. gonorrhoeae (PPNG).23 However, in one study 15% of isolates were PPNG and only 39% of non-PPNG isolates were fully susceptible to penicillin, the resistance being mediated by mechanisms other than penicillinase production.24 The value of co-amoxiclav in gonorrheal infection is therefore limited. Treatment choice should be guided by local resistance patterns.

Skin and soft tissue infections

These infections are most commonly caused by Staph. aureus. Co-amoxiclav has not been compared with flucloxacillin, the usual first-choice antibiotic. For mixed infections, particularly those involving anaerobes, co-amoxiclav can be useful where a combination of drugs might otherwise be needed.

Unwanted effects

Serious unwanted effects are rarely seen with co-amoxiclav, although clavulanate-associated cholestatic jaundice has been reported, usually in patients aged over 60 years. The reaction may occur up to 6 weeks after completion of the antibiotic course and is usually reversible.25 Data collated by the manufacturer suggest unwanted effects in those taking oral co-amoxiclav include diarrhoea (about 4%), nausea (3%), vomiting (2%), indigestion (2%), rash (1%) and Candida superinfection (1%).26 However, in one study involving children, the incidence of gastrointestinal upset was as high as 52%.27

Conclusion

Co-amoxiclav (Augmentin or Augmentin-Duo) should rarely be seen as a first-line drug for the treatment of common bacterial infections in general practice. Although commonly used in respiratory infections, co-amoxiclav is not an appropriate alternative to empirical amoxycillin in most cases of community-acquired pneumonia or acute exacerbations of chronic lung disease. In urinary infections, co-amoxiclav should only be considered as a second-line treatment if organisms are shown to be resistant to trimethoprim. In childhood otitis media and acute sinusitis, amoxycillin remains the treatment of choice, with co-amoxiclav used as a second-line treatment for those who do not respond to amoxycillin. Co-amoxiclav does, however, have a place as antibiotic prophylaxis prior to abdominal and gynaecological surgery for prevention of wound infections and is simpler than other commonly used therapy employing two drugs. It also has a place in the treatment of pelvic inflammatory disease, in conjunction with a drug effective against chlamydia, and for skin and soft tissue infections particularly where there is a mixed infection. Co-amoxiclav is a first-line treatment for severe dental infections, postpartum infections and human and animal bites.