Antibiotic choice for the prophylaxis and empirical treatment of traveller's diarrhoea

• Fluoroquinolones are generally recommended for adults (except for pregnant or breastfeeding women and people travelling to locations with widespread quinolone resistance) due to the resistance of micro-organisms to other antibiotics [Hill et al, 2006; CDC, 2009].
• In countries where quinolone resistance is unknown, the use of quinolones (such as ciprofloxacin) is supported by the majority of CKS expert reviewers.
• Although other fluoroquinolones may be offered, ciprofloxacin is preferred because:
• It is less expensive than other fluoroquinolones and there is evidence to support its use for the prophylaxis and treatment of traveller's diarrhoea.
• Unlike other fluoroquinolones, ciprofloxacin is licensed for the treatment of gastrointestinal tract infection, including the empirical treatment of severe traveller's diarrhoea [ABPI Medicines Compendium, 2008c].
• Although rifaximin (a poorly absorbed antibiotic) is also recommended in some sources, it is not available in the UK.
• Doxycycline, taken for malaria prophylaxis, offers little or no protection against traveller's diarrhoea in most areas of the world.

Azithromycin as an alternative to ciprofloxacin for empirical treatment

• Although it is not licensed for the treatment of traveller's diarrhoea, azithromycin is recommended by the World Health Organization (WHO), and by experts, for the treatment of traveller's diarrhoea if resistance to fluoroquinolones is suspected or known [Al-Abri et al, 2005; Hill et al, 2006; WHO, 2009].
• It is also recommended as the first-line antibiotic by WHO, and by experts, for treating traveller's diarrhoea in children and pregnant women [Hill and Ryan, 2008; WHO, 2009].
• Ciprofloxacin and other quinolones should be avoided in children and adolescents, and in pregnant or breastfeeding women because of the increased risk of quinolone-associated arthropathy in the child or fetus [ABPI Medicines Compendium, 2008c].
• The manufacturer of ciprofloxacin recommends that ciprofloxacin treatment should be initiated in children only 'by physicians who are experienced in the treatment of cystic fibrosis and/or severe infections in children and adolescents' [ABPI Medicines Compendium, 2008c].
• Manufacturers of other quinolones warn that these antibiotics are contraindicated in children and growing adolescents [ABPI Medicines Compendium, 2009c; ABPI Medicines Compendium, 2009d; ABPI Medicines Compendium, 2009e].
• All the manufacturers of quinolones state they should not be used in pregnant or breastfeeding women.
• The off-label dosages provided are extrapolated from those recommended by the manufacturer of azithromycin [ABPI Medicines Compendium, 2008b].

Antibiotics that are not recommended for prophylaxis and empirical treatment

• Co-trimoxazole, doxycycline, and trimethoprim are no longer considered effective against the enteric bacterial pathogens involved in traveller's diarrhoea, due to widespread resistance in most areas of the world [Al-Abri et al, 2005; CDC, 2009].

Duration of antibiotic prophylaxis

• The recommendation to offer short-term antibiotic prophylaxis for up to 3 weeks is based on expert opinion and reflects the time period used in trials [Public Health Agency of Canada, 2001; Al-Abri et al, 2005; Hill et al, 2006; CDC, 2009; DuPont et al, 2009b]. Some experts discourage prophylaxis for longer than 3 weeks as there is a possibility that this might inhibit the development of natural immunity seen with medium- to long-term stays in the country [DuPont et al, 2009b].

Duration of empirical treatment

• The recommendation to offer a 3-day course is in line with WHO recommendations [WHO, 2007] and expert opinion [Ericsson, 2003; Al-Abri et al, 2005; Hill et al, 2006; Hill and Ryan, 2008; Dupont, 2009].
• A variety of antibiotic regimens (from a single dose to 5-day course) were found to be effective against traveller's diarrhoea in a Cochrane systematic review [de Bruyn et al, 2000].
• Consequently, most experts recommend that the individual should re-evaluate their symptoms after 24 hours of starting antibiotic treatment and to complete the 3-day course if they are still unwell, or to stop sooner if they are improved [Ericsson, 2003; Al-Abri et al, 2005; Hill et al, 2006; Hill and Ryan, 2008].

Antibiotic prophylaxis for children and adolescents, and women who are pregnant or breastfeeding

• CKS recommends seeking specialist advice because there are no data supporting antibiotic prophylaxis for these groups [Hill et al, 2006].
• In addition quinolones should be avoided in these people because of the increased risk of quinolone-associated arthropathy in the child or fetus [ABPI Medicines Compendium, 2008c].
• Although azithromycin is recommended as an alternative to quinolones for treating traveller's diarrhoea, no studies have been undertaken to determine an appropriate dose for azithromycin prophylaxis [Al-Abri et al, 2005; Hill et al, 2006].

Antibiotic prophylaxis for regions where quinolone resistance is widespread

• CKS recommends seeking specialist advice if the person is planning to travel to regions where resistance to fluoroquinolones is high (predominantly Campylobacter species), such as India, Thailand, and the Far East, because:
• There is currently no other antibiotic with demonstrated prophylactic efficacy against Campylobacter species [Hill et al, 2006; Hill and Ryan, 2008].
• No studies have been undertaken using azithromycin for prophylaxis.

Combination of antibiotic and loperamide

• The combination of an antibiotic and loperamide has been recommended to reduce the frequency of bowel movement for people who need immediate control of symptoms [Hill and Ryan, 2008; CDC, 2009; WHO, 2009].
• There is limited evidence from a systematic review that a combination of loperamide and antibiotic treatment is more effective than antibiotic alone in treating traveller's diarrhoea [Riddle et al, 2008]. More people were cured after 24 hours and 48 hours, but not after 72 hours, of starting treatment. However, in terms of the time to the last diarrhoeal stool for the two treatments, the differences were variable, ranging from 2–22 hours. This result should be interpreted with caution given the significant heterogeneity of the studies.
• The combined use of loperamide and antibiotics in the presence of fever or blood in the stool is controversial [Ericsson, 2003]. Some experts recommend this combination should only be used when there is no fever or blood in the stool [Hill et al, 2006; Hill and Ryan, 2008]. For people with fever or blood in the stool, there are concerns that loperamide can inhibit peristalsis, resulting in complications such ileus, megacolon, and toxic megacolon.
• Given that the above systematic review failed to assess adverse effects (people with fever or blood in the stool were excluded from the trials) and the limited evidence supporting its use, CKS advises that the combination should be avoided when there is fever or blood in the stool.