Systematic review: the epidemiology and clinical features of travellers’ diarrhoea


  • H. L. DUPONT

    1. The University of Texas School of Public Health, St. Luke’s Episcopal Hospital, Baylor College of Medicine; and Kelsey Research Foundation, Houston, TX, USA
    Search for more papers by this author

Dr H. L. DuPont, 1200 Herman Pressler, Suite 733, Houston, TX 77030, USA.


Background  Travellers’ diarrhoea is the most common medical complaint among persons venturing into developing areas from industrialized regions.

Aim  To review recent developments dealing with microbiological, clinical, pathophysiological and therapeutic aspects of travellers’ diarrhoea.

Methods  The author’s extensive file plus a review of publications listed in PubMed on January 22, 2009 on the topic of travellers’ diarrhoea were reviewed.

Results  Travellers’ diarrhoea is largely caused by detectable and undetected bacterial enteropathogens, explaining the remarkable effectiveness of antibacterial agents in prophylaxis and therapy of the illness. A number of host genetic polymorphisms have been recently linked with susceptibility to travellers’ diarrhoea. Novel antisecretory agents are being developed for treatment considering their physiological effects in acute diarrhoea. All travellers should be armed with one of three antibacterial drugs, ciprofloxacin, rifaximin or azithromycin, before their trips to use in self therapy should diarrhoea occur during travel. Loperamide may treat milder forms of travellers’ diarrhoea and can be employed with antibacterial drugs.

Conclusions  Diarrhoea will continue to plague international travellers to high-risk regions. More studies of the incidence rate, relative important of the various pathogens by geographical region of the world, host risk factors and optimal therapeutic approach are needed.


The most common cause of disability among international travellers to developing tropical and semitropical regions of the developing world is diarrhoea (passage of ≥3 unformed stools plus a sign or symptom of enteric infection such as abdominal cramps or pain) (TD).1 The range in frequency of TD in these areas is between 10% and 60% with the highest rates seen in Latin America, Africa and the Indian subcontinent. The lowest rates of TD (<4%) are seen when people travel between two low risk regions, including the US, Canada, Western Europe, Japan, Australia and New Zealand. Intermediate rates of TD (8–15%) are seen for travellers to China, Russia, the Middle East and southeastern Asia (if Thailand is representative of the region). All travellers to the high-risk regions should exercise care in disease prevention by careful selection of safer foods and in certain situations through use of rifaximin chemoprophylaxis.2 All persons planning a trip to a high-risk region should take with them an antibiotic for self-treatment of diarrhoea that may occur.3

In this systematic review, we will consider the aetiology of TD, the important pathophysiological mechanisms of disease, host genetic risk factors, various clinical syndromes produced and recommended treatment.

Materials and methods

References published since 1954 listed as ‘Travellers Diarrhoea’ in PubMed were reviewed. PubMed listed 662 publications to consider on January 22, 2009. Specific references dealing with aetiology, pathophysiology, clinical features and therapy were reviewed. In addition, the author’s extensive file on TD was included in the review. Prevention of TD was not considered as this was the subject of a recent systematic review published in APT.2


Aetiology of TD

In the 1950s, studies demonstrated that antibiotics were effective in preventing diarrhoea among persons venturing into high-risk regions from industrialized regions4,5 providing the first evidence that bacterial enteropathogens were the major causes of TD. Also, it was shown in the early history that conventional enteropathogens were not important causes of the disease as studied in Mexico.6 Evidence that strains of diarrhoea-producing Escherichia coli produced the disease was provided by Rowe et al.7 studying British troops in the Middle East. It is now known that the most important group of pathogens causing TD are enterotoxigenic E. coli (ETEC),8 enteroaggregative E. coli (EAEC)9 and probably diffusely adherent E. coli (DAEC).10 While DEC cause a majority of TD cases in the high-risk regions, invasive enteropathogens, including Campylobacter jejuni, Shigella spp. and Salmonella spp. more commonly are encountered in south Asia than in Africa and Latin America (See Figure 1 for a summary of published studies on aetiology).

Figure 1.

 Major causes of TD according to World region in which illness occurs adapted from Shah N et al.84

A sizable proportion of TD cases remain without cause after comprehensive microbiological evaluation (∼20–40%). It appears that most of these undiagnosed cases represent undetected bacterial enteropathogens in view of the effectiveness of antibacterial drugs in successful treatment11–14 and prevention15,16 of the undiagnosable proportion of disease. Use of PCR methods,17,18 looking for new diarrhoea-producing E. coli,10,19 and studying additional E. coli colonies for possession of virulence properties18,20 have helped to show that an increased proportion of TD is caused by diarrhoeagenic E. coli. Using direct and indirect information, bacterial enteropathogens appear to explain up to 80% of TD cases. Strains of Campylobacter jejuni explain an important percentage of cases occurring in southern Asia. Noroviruses are the most commonly implicated nonbacterial cause of TD, explaining up to 15% of the disease.21 Parasites can be found in only about 2–10% of acute TD cases.8 This group of organisms is more important in the persons who present with persistent and chronic TD after returning home.22


While enteric pathogens differ in their mechanisms of virulence, the clinical features of the illness show important similarity regardless of aetiology. This relates to the limited ability of the gut to react to microbial threats. Most of acute diarrhoea cases experience frequent passage of unformed stools because of movement of fluid and salt across the mucosa from the vascular serosal side of the gut to the mucosal surface (secretion). For the aetiological agents responsible for acute diarrhoea, secretion and intestinal inflammation are the two most important pathophysiological alterations.23 Both lead to salt and water loss from the gut (Table 1).

Table 1.   Pathogenic Mechanisms in travellers’ diarrhoea by aetiology
Aetiological agentVirulence factorsSecretionInflammationBowel siteOther
  1. ETEC, enterotoxigenic Escherichia coli; EAEC, enteroaggregative E. coli.

ETECColonization factor fimbriae, heat labile enterotoxin (LT) and heat stable enterotoxin (ST) production++++Small bowel (SB) 
EAECEpithelial cell adherence+++SB & possibly colon 
ShigellaMucosal invasion++++++Colon 
SalmonellaMucosal invasion+++++SB & Colon 
CampylobacterMucosal invasion+++++Colon 
AeromonasEnterotoxin, cytotoxin, mucosal invasion++++Colon 
VibriosHeat labile enterotoxin (V. cholerae)+++++SB 
Mucosal invasion (non-cholera strains)?+++Colon 
NorovirusesMucosal invasion++SBDelayed gastric emptying, associated with vomiting, lactase deficiency, malabsorption
GiardiaEpithelial cell adherence by sucking disks+++SB 
Entamoeba histolyticaMucosal invasion & ulcer formation, cytotoxin production+++Colon 
CryptosporidiumMucosal attachment+++SB 

A majority of the enteropathogens causing TD lead to intestinal fluid and electrolyte losses from secretory processes. The most important microbial cause of TD, ETEC, produces small bowel secretory changes through release of one or two toxins working through mucosal cyclic nucleotide pathways.24–26 The invasive bacterial enteropathogens, Shigella, Campylobacter and Salmonella also produce secretory alterations.27–31 The resultant inflammatory colitis seen in these cases is associated with release of cytokines and other inflammatory mediators and mast cell degranulation contributing to altered gut motility and active secretion. Colonic surface and crypt cells both contribute to the secretory state seen.32 Parasites including Giardia also may produce enterotoxins and secretory alterations of the gut.33 In Table 1, the interplay of secretory events and inflammation in the intestine along with the intestinal site of the process is seen for the various enteropathogens. The most important mechanism operating in acute diarrhoea caused by a large number of bacterial enteropathogens is intestinal secretion, which has encouraged the development of a broad array of antisecretory drugs with potential value in treating the disease.34

Some enteropathogens contribute to enteric disease symptoms by nonsecretory mechanisms. A variety of small bowel pathogens induce lactase deficiency and produce malabsorption of other dietary substances.35 In giardiasis enterocyte injury leads to malabsorption.36 The degree of intestinal damage found may not correlate with extent of malabsorption found.37 The enteric viruses, rotavirus38 and noroviruses,39 inhibit gastric emptying explaining the importance of vomiting as a primary feature of the resultant illness.

Host genetics

People often give the history that they show a different susceptibility to TD from that of friends or spouse. While behaviour during high-risk travel, including food selection patterns, differs among travellers, host genetics may play the greatest role in variability of host susceptibility and resistance to TD. The three general host genetic targets showing a relationship with diarrhoea susceptibility include histo-blood groups and secretor status, inflammatory cytokines and lactoferrin and receptors of microbial attachment. Individuals with type O blood are more susceptible to severe cholera40 and secretor status has been associated with susceptibility to norovirus gastroenteritis.41–43 TD caused by a variety of enteropathogens is seen with increased frequency in subjects with a polymorphism in the lactoferrin gene44 or the osteoprotegerin gene.45 Lactoferrin is involved with polymorphonucelar leucocyte inflammation and osteoprotegerin is an immunoregulatory gene encoding for tumour necrosis factor (TNF) receptor and intestinal inflammation. ETEC diarrhoea appears to occur more commonly in people with a polymorphism in the IL-10 promoter gene and they show higher levels of IL-10 in stool.46 Polymorphism in the IL-8 promoter is associated with higher levels of the inflammatory chemochime, IL-8, and is associated with TD due to the inflammatory pathogen, enteroaggregative E. coli (EAEC).47 In animal studies, presence or absence of receptors of attachment of ETEC explains susceptibility to infection and diarrhoea48 and the finding of lack of serum antibodies to norovirus in subjects most resistant to viral gastroenteritis suggests that these subjects lack receptors for virus attachment.49 More work is needed in this area to understand better the interplay between genetic markers and susceptibility to TD.

Clinical features

The various clinical syndromes seen in travellers to high-risk are seen in Table 2. There are four clinical syndromes seen in enteric infection of travellers. They are described in this section.

Table 2.   Clinical presentation of diarrhoea acquired during International Travel to Developing Countries
Clinical syndromeFrequencyAetiological agents to consider
  1. *1–3% in Latin America and Africa50–52 and for travellers to diverse areas of the world53 and up to 9% for parts of Asia.54

Acute gastroenteritis (vomiting predominates as a clinical finding)10%Noroviruses, preformed toxins of Staphylococcus aureus or Bacillus cereus
Acute watery diarrhoea80%All agents, including ETEC, EAEC, Shigella, Salmonella, noroviruses
Bloody diarrhoea1–9%*Shigella, Campylobacter, less commonly Salmonella spp, non-cholera Vibrios and Aeromonas spp
Persistent diarrhoea (≥14 days)2–10%Giarda, Cryptosporidium, E. histolytica, Microsporidium, Cyclospora, Dientamoeba fragilis, Shigella, enteroaggregative E. coli, Brainerd diarrhoea, IBS or IBD
Post infectious IBS5–10%Acute diarrhoea due to invasive/inflammatory bacteria are the most important inciting event in a genetically susceptible host

Acute gastroenteritis

Approximately 10% of subjects with enteric disease during international travel present with vomiting as the primary feature of the disease. Diarrhoea may complicate the illness, but vomiting predominates. The two major causes of this illness are viral gastroenteritis (often caused by a norovirus) vs. ingestion of a preformed toxin of Staphylococcus aureus or Bacillus cereus from contaminated food. In viral gastroenteritis, the incubation period exceeds 14 h while for the intoxications, the incubation period is 2–7 h, often less than 4 h.

Acute TD

The most frequently encountered clinical syndrome in TD is acute watery diarrhoea with abdominal pain and cramps seen in approximately 80% of cases. All aetiological agents in TD including bacterial, viral and parasitic enteropathogens cause this syndrome. Thus, it is not possible to suspect a specific aetiology of TD in subjects with acute diarrhoea.

Dystenteric diarrhoea

Passage of bloody stools, often with associated fever, is seen in approximately 1–3% of TD cases among persons visiting Latin America or Africa50–52 or diverse regions of the developing world,53 although one report for TD occurring in India found a dysenteric rate of 9%.54 The major causes of dysenteric TD are listed in Table 2.

Persistent TD and post-infectious irritable bowel syndrome (PI-IBS)

The frequency of persistent TD, lasting two weeks or longer, is 2%. This syndrome occurs more commonly in persons living close to the locals including Peace Corps volunteers and medical missionaries. In these cases, the differential diagnosis includes protozoal parasites or bacterial enteropathogens22 and chronic gastrointestinal disease including irritable bowel syndrome (IBS), inflammatory bowel disease (IBD)55 and coeliac sprue.56 The invasive bacterial pathogens can cause persistent diarrhoea,57 while non-invasive diarrhoea-causing E. coli do not.58 In the case of IBD and coeliac sprue, an enteric infection appears to unmask an underlying propensity or worsen the pre-existent inflammatory disorder.59 In between 5% and 10% of TD cases, a chronic gastrointestinal illness develops where abdominal pain and discomfort are associated with change in stool form resembling irritable bowel syndrome.60,61 This condition is referred to as post-infectious (PI) IBS. Risk factors for PI-IBS include severity of bout of acute diarrhoea, virulence properties of the infecting organism, age <60 years and female gender.62


The mainstay of treatment of TD are antibiotics with known activity against prevalent enteropathogens in regions visited.63 There are advantages and disadvantages of the agents currently effective in the treatment of TD (Table 3). When trimethoprim/sulfamethoxazole-resistance became widespread, the fluoroquinolones became the important forms of therapy.63 Ciprofloxacin is often employed for empirical treatment by travel medicine experts as the drug has become less expensive due to its generic status in many areas of the world. While it is given twice daily for three days, single dose therapy is effective in most cases.64 Problems with the fluoroquinolones have emerged dampening enthusiasm for use of this class in travel medicine. On July 8, 2008, the U.S. FDA issued a Black Box warning for fluoroquinolones due to the occurrence of tendinitis and tendon rupture. In addition, fluoroquinolones including ciprofloxacin may deplete colonic flora65 and predispose persons treated with these drugs to Clostridium difficile colitis.66 Finally, Campylobacter strains from diverse regions have shown an increased resistance to the fluoroquinolone class of drugs.67

Table 3.   Appropriate treatment options for travellers’ diarrhoea (In order of development for this use)
  1. *Single dose therapy may be effective.64

Ciprofloxacin500 mg b.d. × 3 days*Generic drug, should be less expensiveNot always less expensive, has potentially important side effects (tendonitis and rupture, Clostridium difficile colitis, Campylobacter strains are often resistant
Levofloxacin500 mg q.d.s. × 3 daysAlso treats bacterial respiratory tract infectionSame side effect and susceptibility patterns as ciprofloxacin
Rifaximin200 mg t.d.s. × 3 daysExcellent safety profileNot effective against invasive forms of TD, especially associated with passage of bloody stools or presence of fever
Azithromycin1000 mg in Single doseBroad activity against all bacterial forms of TDCauses short lasting nausea, most effective drug against febrile or dysenteric TD

Rifaximin represents the safest compound to use to manage TD with efficacy against the broad range of the disease equivalent to fluoroquinolones.68,69 A subgroup of TD subjects with invasive disease will respond to non-absorbed (<0.4%) rifaximin more slowly than to the fluoroquinolones.69

Azithromycin has the broadest activity against TD bacterial eneropathogens and is the treatment of choice for TD associated with fever or passage of grossly bloody stools, potentially caused by fluoroquinolone-resistant Campylobacter70 and is the treatment of choice for rescue medication in the case of breakthrough diarrhoea during rifaximin chemoprophylaxis.71 All travellers should be armed with at least one of the effective drugs (Table 3) for trips to high-risk regions and advised to employ self treatment at disease onset.3

The most rapid response of TD is seen by combining a symptomatic treatment with rapid onset of symptom improvement with antibacterial drugs that have curative antibacterial effects. The studies have shown an advantage of adding loperamide to fluoroquinolone,72 rifaximin13 and azithromycin73 treatment of TD. The combination of loperamide and antibacterial therapy, decreases median time until passage of the last unformed stool, decreases the mean number of unformed stools passed and more rapidly leads to cure of diarrhoeal illness according to subjects than is the cases when either drug is given alone or subjects receive inactive placebo.13

In view of the importance of secretory pathways in TD of diverse cause (Table 2), anti-secretory drugs are being developed to treat all forms of acute diarrhoea including TD. Studies are needed to evaluate the additive effect of antisecretory drugs when combined with antibacterial drugs in the treatment of TD. Two antisecretory drugs have been shown to reduce effectively the frequency of diarrhoea stool passage in treated subjects with TD, crofelemer, a chloride channel blocker74 and zaldaride, a calmodulin inhibitor.75 Racecadotril, an enkephalinase inhibitor that decreases the degradation of endogenous opiates, has been evaluated in acute diarrhoea of children76 and adults with endemic diarrhoea treated with the drug vs. loperamide.77,78 Therapeutic trials looking at racecadotril in the therapy of TD have not been performed. None of the antisecretory drugs has been combined with antibacterial drugs in looking for additive effects in the management of TD.

Future research priorities (Table 4)

Table 4.   Future research priorities in the area of travellers’ diarrhoea
Area of studyFirst priorityLonger range question
  1. *Arcobacter spp, non-enteroxigenic and non-aggregative diarrhoea-producing E. coli, enterotoxigenic Bacteroides fragilis, Microsporidium, Cyclospora, Blastocystis.

Aetiology of TDEstablish the importance of conventional enteropathogens by various geographic areasDetermine the role of new and less well appreciated pathogens* as aetiological agents
Pathogenesis of post-infectious IBSDetermination of the importance of PI-IBS in subjects with and without TD, evaluation of host risk factors including genetic factors in disease occurrenceDevelopment of strategies for disease prevention (chemo- and immune-prophylaxis) of PI-IBS
TherapyDetermine the value of antisecretory agents in therapyDetermine additive or synergistic effect of combination therapy
TD PreventionEvaluate safety and effectiveness of various drugs (antimicrobial agents and probiotics) and immunoprophylactic (anti-ETEC) vaccines in the prevention of TDDetermine cost effectiveness of the various prevention strategies
Paediatric TDAetiologyResponse to treatment

The important regions of the developing world with known high rates of TD need to be better studied for incidence of diarrhoea and for importance of established and less well studied enteropathogens. In addition to the various diarrhoea-producing E. coli,10,18 enterotoxigenic Bacteroides fragilis,79Arcobacter spp80,81 and microsporidia82 should be sought for a causative role in TD.

Additional study is needed to determine variability of host susceptibility to diarrhoea focusing on definable and new genetic factors. Determining how the various genetic factors interrelate and defining people at special risk in a program designed to reduce illness fit with the modern concept of personalized medicine.

Novel approaches to self-treatment of TD are needed. It is known that beginning treatment once TD is established shortens the duration of illness, but does not prevent PI-IBS.60 Studies are needed to determine if PI-IBS can be prevented by starting therapy with the first sign or symptom of TD, possibly with passage of the first unformed stool. Antisecretory drugs which make physiological sense when considering the importance of secretory mechanisms in TD need to be evaluated in TD therapy, as single agents and in combination with antibacterial drugs.

Nearly all studies of TD have been carried out with adult subjects. While it is logical to think that children will respond similar to adults, little data are available on the microbiology, clinical features, natural history and response to treatment of TD in children.83 Children should be systematically studied for unique clinical and microbiological features of TD.


Declaration of personal interests: Dr DuPont has received honoraria for speaking for the following companies: Salix Pharmaceuticals; McNeil Consumer Healthcare; Merck Vaccine Division and IOMAI Corporation and has received grants through the University of Texas to support research from: Salix Pharmaceuticals; IOMAI Corporation; Intercell Corporation and Optimer Pharmaceuticals. Declaration of funding interests: Discretionary funds from the Kelsey Research Foundation and the University of Texas – Houston School of Public Health were used to support this review. The review was supported in part by grant DK 56338, which funds the Texas Gulf Coast Digestive Diseases Center.