Travellers' diarrhoea

Travellers’ diarrhoea ROBERTO HERBERT

C. ARDUINO L. DUPONT

INTRODUCTION Travellers’ diarrhoea (TD) is defined as three or more unformed stools per day plus one or more of the following signs or symptoms: abdominal pain or cramps, nausea, vomiting, dysentery (passage of bloody stools), faecal urgency, and tenesmus, occurring in a resident of an industrialized country during a temporary stay in a developing nation. The illness characteristically occurs within 2 weeks after arrival in the foreign country and most often within the first week. Although TD is an acute and self-limited illness that rarely lasts more than 5 days, approximately 30-50% of the estimated 16 000 000 people who travel from industrialized countries and spend at least 3 weeks in less developed areas each year will experience diarrhoea. Moreover, among those people with diarrhoeal illness, 30% will be confined to bed and another 40% will have to modify their travel plans (Gorbach, 1982). Epidemiology Attack rates of TD are similar in men and women. Younger travellers are more frequently affected. The lack of acquired immunity, increased faecaloral contamination, more adventurous travel styles and different eating habits may explain this finding. TD occurs worldwide; a background rate of diarrhoea exists in all countries which varies with the population and region studied from a low of 2-5 % to a high of 50%. The risk of developing TD seems to be highly dependent on the geographical area visited. Data from various studies have allowed a division of the world into three regions according to the risk that persons from industrialized countries have of acquiring diarrhoea during travel in the region. The low-risk areas (2-5% diarrhoea rates) include the USA, Canada, northern Europe, South Africa, New Zealand and Australia. High-risk areas (with average diarrhoea rates of approximately 40%) are Latin America, Africa and southern and south-eastern Asia. Intermediate-risk areas (10-20% rate of illness or unknown risk) include most southern European countries, China, the Commonwealth of Independent States (formerly the Soviet Union) and some Caribbean islands. 365

Baillitre’s Clinical GastroenterologyVol. 7, No. 2, June 1993 ISBNO-702&1749-3

Copyright01993, All rightsof reproduction

byBailli6reTindall in any form reserved

366

R.

C. ARDUINO

AND

H.

L.

DUPONT

There is epidemiological information that persons travelling from one high-risk area to a different high-risk area experience lower rates of diarrhoeal illness compared to persons from low-risk areas (Kean, 1969; Lowenstein et al, 1973), indicating the ubiquity of the common enteric pathogens in each of these high-risk areas and the development of immunity through prior exposure. Studies of travellers from high-risk or low-risk countries travelling to industrialized countries demonstrated low attack rates (Ryder et al, 1977). Diarrhoea has been reported in people from high-risk areas like Mexico or Asia travelling to low-risk areas such as the USA (Macdonald and Kean, 1962; Dandoy, 1966) and in people from low-risk areas travelling to another low-risk area (Steffen, 1986). However, the incidence is low and the diarrhoea is generally mild and clinically unimportant. These changes in the bowel habits are probably common among travellers, regardless of country of origin, and they are most likely related to altered dietary and alcohol consumption patterns, exposure to normally non-pathogenic bacteria in foods, ingestion of non-absorbable salts and other substances found in local drinking water, or emotional factors. Economic aspects The economic impact of TD is profound. The overall cost to the traveller can be great since a high percentage of those who develop troublesome diarrhoea are required to modify their itinerary, nearly everyone who acquires illness purchases drugs to treat the disorder, some seek medical attention, and many lose time from pleasure or employment activities. Diarrhoea is a serious and recurrent problem for business persons and governmental employees travelling abroad. Fifty per cent of the US troops during the Desert Shield operation in the Persian Gulf were affected on at least one occasion and 20% were temporarily incapacitated at some time during the campaign (Hyams et al, 1991). The economic implications for developing nations are of even greater importance. For 1980, the World Tourism Organization and other authorities estimated that approximately $100 billion were spent to support foreign travel. Since 20% of these dollars can be expected to support travel to developing areas, any potential loss could represent a major impact on the countries where tourism is economically so important. Risks of developing diarrhoea Gastric acidity, intestinal mucosal integrity and intestinal motility are protective barriers directed against enteric pathogens that may reach the intestines through the mouth. TD is acquired through the ingestion of faecally contaminated food and/or water, and the potentially pathogenic bacteria are susceptible to low pH. Therefore, drugs that alter these barriers, such as antacids or histamine (HZ) blockers, or gastric hypoacidity as well as gastrectomy may increase the susceptibility to bacterial and parasitic infections (Howden and Hunt, 1987). Gastric hypoacidity has been

TRAVELLERS’

DIARRHOEA

367

reported to predispose to Salmonella typhi and non-typhi Salmonella infections, shigellosis, cholera and giardiasis (Buchwald and Blaser, 1984). HIV-infected travellers are at special risk, since shigellosis and salmonellosis may be protracted, severe and bacteraemic in these patients, despite relatively minor gastrointestinal symptoms. Campylobacter infection has been associated with bacteraemia and cholecystitis in HIV-infected persons (Wilson et al, 1991). Cryptosporidiosis has also resulted in more severe outcome in HIV-infected subjects where the organism is associated with chronic infection leading to malnutrition, wasting, and complications with other infections (Navin and Juranek, 1984). Gastric secretory failure and impaired mucosal immunity due to depletion of CD4 lymphocytes may explain these more severe and probably more frequent enteric infections in HIV-infected travellers. Diarrhoea rates of 40% per month were observed in newly arrived US summer students in Mexico compared to 20% for those who had been attending classes at the school for a year or longer (DuPont et al, 1977). In the same study, the diarrhoea rates and severity of illness among Latin American students were the lowest. Prolonged or repeated exposure to enterotoxigenic Escherichia coli (ETEC) probably explains the natural immunity as those persons live in a high-risk area. In a population living in a high-risk area, a clear correlation was found between the serum antibody titres to heat-labile E. coli enterotoxin (LT) and the susceptibility to ETEC infection (Evans et al, 1977). Among those who developed illness, there was a rise in the geometric mean serum LT antibody titre during convalescence. It was of interest that the lowest base-line antibody titres were seen in individuals who were more susceptible to ETEC diarrhoeal illness. AETIOLOGICAL

AGENTS

A wide variety of pathogens, including bacteria, viruses and protozoa have been reported to cause TD. There is no doubt, however, that bacteria are the most common agents causing diarrhoea in travellers and that ETEC is the single most frequently isolated species, identified in 40-70% of the cases, depending on the geographical location, season, and techniques used. Other bacterial pathogens commonly isolated are Shigella, Salmonella, Campylobatter, Aeromonas, Plesiomonas, and Vibrio. Table 1 shows the relative frequency of the various enteric pathogens for different geographical areas. Seasonal variation in the incidence and aetiology of TD has been reported. Diarrhoeal illness was studied in tourists who went to Morocco from Finland during winter and autumn (Mattila et al, 1992) and a lower attack rate was observed during the winter months compared to the autumn months (15% versus 31%) respectively). ETEC were isolated in 32% of subjects during the autumn but in only 8% of those studied during winter months. Salmonella enteritidis was also more frequently isolated during autumn than winter (25% versus 16%). On the other hand, Campylobacter strains were cultured in 28% of diarrhoeal stools during winter compared to 7% in the autumn. Rotavirus

368

R. C. ARDUINO AND H. L. DUPONT Table 1. Proportion of pathogens that commonly cause travellers’ diarrhoea in Latin America, Asia and Africa. Organism Enterotoxigenic E. coli* Enteroadherent E. coli Enteroinvasive E. coli Shigella species Salmonella species

Latin America (%I 40-70 ?-12

Asia (%I

Africa (%)

20-34 ?

36 ?-33

6

3

2

Rotavirus

2-30 O-16 l-7 2 l-4 4-36

Entamoeba histolytica Giardia lamblia

o-9 o-9

2-13 11-15 2-15 1-57 1-16 1-8 2-6 O-6 33-53

2-15 O-4 1-28 l-8 O-4 O-6 O-2 0 15-53

Campylobacter jejunit Aeromonas hydrophila Vibrio non-01

Undiagnosed

20-30

* Highest rates in rainy summertime. + Highest rates in dry wintertime. References: Speelman et al (1983), Gracey et al (1984), Mathewson et al (1985), Taylor et al (1985), Mathewson et al (1987), Wanger et al (1988), Black (1990), Haberberger et al (1991), Hyams et al (1991).

was more commonly identified in the study during the winter months. This seasonal variation has implications for the selection of the antimicrobial therapy and selected prophylaxis. A quinolone is therefore preferred over trimethoprim-sulphamethoxazole (TMP-SMX) during the winter season in those areas where Cumpylobacter strains are more common, due to the invariable resistance of Campylobacter to TMP. The isolation of more than one pathogen has been reported to occur in lO-20% of episodes of diarrhoea in travellers. This probably relates to the background rate at which various enteropathogens contaminate foods and beverages rather than to a true association of the prevalent enteropathogens. Approximately 20% of individuals with TD after careful microbiological study have an unclear aetiology. Non-infectious diarrhoea, inadequate laboratory techniques or pathogens yet to be recognized may help to explain this finding. However, the remarkable effectiveness of antibacterial therapy in the prevention and treatment of TD suggests that bacterial pathogens may be involved in the aetiology of these cases with non-recognizable enteropathogens. Bacteria Enterotoxigenic

E. coli (ETEC)

ETEC possess virulence properties that relate to production of toxins and to specific binding to the intestinal mucosa. Prior to the development of illness, ETEC strains must attach to the upper gut via the cell surface fimbriae of the bacteria, known as colonization factor antigens or CFAs, which interact with specific membrane receptors in the intestinal mucosa. Once the ETEC

TRAVELLERS’

DIARRHOEA

369

have adhered to the mucosa, enterotoxins produce changes in the electrolyte flux across the enteric cells. ETEC produce a heat-stable enterotoxin (ST), which is a low molecular weight non-antigenic molecule; heat-labile enterotoxin (LT), which is closely related to cholera enterotoxin; or both. The exact incidence ofETEC varies with the geographical area studied, but overall they are the most common pathogens isolated. These organisms have been isolated in approximately 42%) 34% and 20% of travellers with diarrhoea in studies carried out in Latin America, Africa and Asia, respectively (Black, 1990). Enteroadherent

E. coli (EAEC)

EAEC strains adhere to HEp-2 cells like classical enteropathogenic E. coli (EPEC) but, in contrast, they do not belong to traditional enteropathogenic serotypes. These organisms have been associated with approximately 12% of episodes of diarrhoea in travellers to Mexico (Mathewson et al, 1985) and up to 33% in travellers to Egypt (Haberberger et al, 1991). EAEC have also been implicated as important causes of both acute and persistent diarrhoea (L 14 days’ duration) among infants in developing countries (Mathewson et al, 1987; Bhan et al, 1989). Resistance of EAEC to multiple antimicrobial agents was recently reported (Arduino et al, 1991; Haberberger et al, 1991; Yamamoto et al, 1992). Enteroinvasive

E. coli (EIEC)

EIEC strains invade the colonic mucosa and spread laterally in the epithelium, causing inflammation, ulcer production and clinical dysentery indistinguishable from that produced by Shigella spp. This invasive property is plasmid-mediated, as seen for strains of Shigella. EIEC have been reported as being causative of diarrhoea in 6% of travellers to Mexico (Wanger et al, 1988) and 3% of US troops in the Middle East (Hyams et al, 1991). Shigella Shigella species cause disease by invasion of the mucosa, a plasmid-encoded property, and in the case of at least some strains, by production of toxin. Initially, shigellosis causes watery diarrhoea that is characteristic of the cholera-like syndromes. After l-3 days, this phase is followed by the passage of bloody mucoid stools (classic dysentery), a finding that reflects colonic involvement. Due to the small inoculum necessary to produce diarrhoea (as shown by volunteer studies, 101-lo2 viable cells), Shigella spp. may be transmitted by direct contact and secondary spread decreases with improved hygiene. In the developing tropical world food-borne infection is common. The geographical distribution is variable, the organism having been isolated from O-30% of travellers with diarrhoea in Latin America, 2-13% in Asia and O-15% in Africa (Black, 1990).

370

R. C. ARDUINO

AND

H.

L.

DUPONT

Salmonella

Non-typhi Salmonella are responsible for an important number of cases of TD. These organisms are isolated from O-30% of the travellers with diarrhoea, depending on the geographical area. Median isolation rates of 1% in Latin America and 4% in Asia have been reported (Black, 1990). Typhoid fever is not a major problem in the traveller since its reported frequency is 1 per 25 000 travellers to developing countries (Steffen, 1986). Campylobacter Campylobacter jejuni causes disease through invasion of the mucosa and possible toxin production. The clinical manifestations are watery diarrhoea and dysentery indistinguishable from that of EIEC and shigellosis. The relative incidence of isolation from travellers with diarrhoea is variable, but it has been recognized as a cause of TD in some countries. Campylobacter jejuni was associated with TD in 17% of episodes in Thailand (Taylor et al, 1985)) in 41% of US military troops exercising in central Thailand (Petruccelli et al, 1992), in 15% of travellers to Bangladesh (Speelman et al, 1983), and in 28% of travellers to Morocco during the wintertime (Mattila et al, 1992). Campylobacter jejuni is an uncommon cause of summertime TD in Mexico, where these strains have been implicated in 1% of episodes (Mathewson et al, 1985). Vibrios

Toxin-producing strains of Vibrio cholerae that agglutinate in 01-antiserum are the agents of epidemic cholera. V. cholerae 01 is a very rare finding as a cause of diarrhoea in travellers to Latin America and has been reported in less than 1% of travellers to Asia. It has been estimated that the incidence of cholera is 1 per 500000 per journey to an endemic area (Snyder and Blake, 1982; Morger et al, 1983). In these rare instances, cholera has been epidemiologically related to eating raw or undercooked shellfish, such as crabs and shrimps, and drinking contaminated water. South America was spared from cholera epidemics until January 1991, when an outbreak of cholera was reported in Peru and rapidly spread to neighbouring countries (Centers for Disease Control, 1991). Sporadic V. cholerae 01 transmission to the USA has occurred continuously since the early 1970s along the Gulf Coast states of Louisiana and Texas, and spread has occurred as a result of crab consumption during a flight from South America. Strains of V. cholerae non-01 have been isolated from stool specimens from tourists to coastal areas during warmer months. V. parahemolyticus may account for up to 70% of episodes in Japan; it has also been found in several Asian studies and in the Mexican Gulf. Poorly cooked crabs and raw shellfish consumption, and improperly handled raw seafood, have been reported as vehicles of the organism. Aeromonas Aeromona.s,

a water-borne

organism, has been associated with diarrhoeal

TRAVELLERS’

DIARRHOEA

371

illness in Australian travellers to South East Asia (Gracey et al, 1984) and US travellers to Thailand (Pitarangsi et al, 1982). This organism has been reported as a cause of TD in 1% of tourists to Bangladesh (Speelman et al, 1983) and in 2% of those travelling to Mexico from the USA (Mathewson et al, 1985). Plesiomonas Plesiomonas shigelloides, which can cause TD through the consumption of uncooked shellfish and, usually, raw oysters, has been occasionally isolated from travellers to South East Asia (Peace Corps volunteers in Thailand), as well as in travellers to Mexico or the Caribbean (Holmberg et al, 1986). Viruses

Although viruses are common causes of diarrhoea in adults and children worldwide, they are less common causes of diarrhoea among travellers than bacterial agents. Studies have reported the identification of rotaviruses in between 10 and 26% of travellers to Mexico with diarrhoeal illness (Bolivar et al, 1978; Ryder et al, 1981). In up to half of these individuals, other enteric pathogens may be identified and the viruses may be found in asymptomatic controls (Bolivar et al, 1978)) casting doubt on the aetiological role the agent plays in these patients. Norwalk virus has rarely been systematically sought in travellers with diarrhoea. This small round virus has been implicated as a causative agent in cases of TD acquired in Mexico (Keswick et al, 1982; Johnson et al, 1990). Protozoa Entamoeba histolytica, Giardia lamblia, Cryptosporidium parvum, Isospora belli and Blastocystis hominis cause only a small percentage of TD.

However, they play an important role in the aetiology of diarrhoea in returning or long-term travellers because of the long period required for them to become symptomatic and because of the prolonged illness that may be produced. Entamoeba histolytica is responsible for approximately l-5% of the TD in studies done in Asia and in Mexico, and most of the cases are seen in long-term travellers and expatriates (Speelman et al, 1983; Mathewson et al, 1990). The infection is transmitted by cysts from other human cases, particularly as a result of excretion by food handlers. Cysts of Giardia lamblia can be found in sewage effluents, in surface waters, and in tap water, even in developed countries; cysts have also been detected in salad and in fruit (Casemore, 1990). These findings and the very low infectious dose reported (5 10 cysts) explain the transmission of this parasite, which is recovered in approximately 5% of travellers with diarrhoea, particularly to St. Petersburg (Brodsky et al, 1974; Jokipii and Jokipii, 1974) or to the Rocky Mountain area of the USA (Wanner et al, 1963; Schultz, 1975).

372

R.

C. ARDUINO

AND

H.

L.

DUPONT

Cryptosporidiumparvum is acquired through the ingestion of oocysts contaminating water or food; however, direct person-to-person infection has also been reported (Koch et al, 1985). Cryptosporidium plays a role in diarrhoea developing in certain regions where drinking water is the major source. Studies in Finnish tourists have shown the importance of cryptosporidiosis among those travelling to St Petersburg (Jokipii and Jokipii, 1974). Isospora belli and Blastomyces hominis are infrequent causes of diarrhoea in travellers. Cyanobacteria-like

organisms

Spherical bodies resembling one type of organelle of blue-green algae cyanobacterium-like organisms) have been observed in diarrhoeal stool from travellers to the Caribbean, Mexico, South America, India and South East Asia, and in patients with the acquired immunodeficiency syndrome (Long et al, 1990). Recent findings, however, provide evidence that the organism may be a new coccidian species of the genus cyclospora (Ortega et al, 1993). The morphology of this new parasite and the clinical presentation in humans is similar to that of cryptosporidia. Interesting, the acid-fast property and the association with HIV positive patients with diarrhoea can lead to a misidentification as Cryptosporodium cysts. Patients infected with the organism typically have presented with nausea, vomiting, anorexia, weight loss, and explosive watery diarrhoea which cleared within two weeks although some patients have experienced more protracted illness. The exact geographic distribution, epidemiology, pathogenesis, and life cycle of this new protozoan parasite are under investigation, but cyclospora along with the other parasites should be considered in those travellers with prolonged and unexplained diarrhoea. CLINICAL

ASPECTS

TD frequently occurs within the first 2 weeks of arrival to the foreign country. The diarrhoea is a self-limiting illness that lasts an average of 5 days when untreated, but approximately a quarter of the travellers affected will have a second episode if they remain in the same area for more than a month. The average number of stools passed over a 5-day period is approximately 12. Abdominal pain and cramps are the most common associated symptoms, and are present in 2060% of episodes; half of the patients complain of nausea, in 15% of patients bloody stools are passed, and 10% of patients have fever. The variable clinical manifestations probably reflect the diversity of the aetiological agents, microbial virulence traits, size of the initial inoculum, and thevariable host response. In an attempt to correlate the clinical presentation of diarrhoea with the most likely enteric pathogen to guide the initial diagnosis and to delineate the subpopulation to treat with antimicrobial agents, enteric illness can be designated as belonging to one of four syndromes: non-inflammatory or secretory cholera-like watery diarrhoea, inflammatory or dysenteric diarrhoea, vomiting out of proportion to diarrhoea and persistent diarrhoea, lasting longer than 2 weeks.

TRAVELLERS’

DIARRHOEA

373

The non-inflammatory or secretory illness presents as a watery diarrhoea which is rarely severe enough to be life-threatening, but which can be worrisome in the very young, in the elderly or in those patients taking diuretics, mainly because of important loss of fluid and electrolytes. There is generally an absence of urgency, tenesmus, gross blood or faecal leukocytes in this syndrome. This type of diarrhoea is usually mediated by enterotoxins that alter fluid and electrolyte transport. These bacterial products act on the epithelial cells of the small intestine, causing increased fluid secretion without damage to the mucosa. Enterotoxigenic E. coli and Vibrio cholerae 01 and non-01 are the prototypes of secretory diarrhoea but other pathogens may produce toxins that resemble those of these organisms, such as Salmonella spp., Campylobacter jejuni and Shigella spp. The dysenteric syndromes are caused by an inflammatory or invasive process that involves the colon and occasionally the distal small intestine. The incidence of dysenteric illnesses in travellers varies from 11% reported from Tunisia to 5% in the Far East. The presence of numerous leukocytes in the faeces is an indicator of the presence of diffuse colonic inflammation. In these cases, there is cellular damage or invasion of the intestinal mucosa. Clinical manifestations include fever, passage of bloody stools (dysentery), chills, low abdominal cramping, and tenesmus. The most common causes of inflammatory diarrhoea in travellers are Shigella spp., Campylobacter jejuni, non-typhi Salmonella spp., Entamoeba histolytica, Vibrioparahemolyticus, enteroinvasive E. coli, and Aeromonas spp. Although clinical signs and symptoms of TD are sometimes associated with the aetiological agent, non-dysenteric shigellosis has been difficult to differentiate from other causes of TD, including ETEC (Ericsson et al, 1987b); faecal leukocytes may not be present in early shigellosis or in some cases of salmonellosis. Furthermore, the number of stools and the presence of nausea, cramps or fever have a low positive predictive value in distinguishing inflammatory from non-inflammatory diarrhoea. When vomiting is the major clinical feature of an enteric illness in a traveller, enterotoxin-mediatedfoodpoisoning due to Staphylococcusaureus or Bacillus cereus or viral gastroenteritis (Norwalk virus, other small round viruses or rotavirus) should be considered. When diarrhoea persists longer than 2 weeks, other aetiologies are usually responsible, including: the conventional bacterial agents; Giardia, Entamoeba and Cryptosporidium; small bowel bacterial overgrowth; lactase deficiency with lactose malabsorption; and non-specific enteritis or colitis resembling Brainerd diarrhoea (chronic diarrhoea of unknown but presumed infectious aetiology following consumption of unpasteurized milk or untreated water) (Blaser, 1986). Diarrhoea may also occur after the traveller has returned to his or her country of origin. Those patients who develop acute diarrhoea within a week after returning from a developing country should be considered and empirically treated as having TD. Although ETEC is responsible for almost 50% of cases of acute diarrhoea in travellers to developing countries, Campylobacter and Salmonella are more common in the returning traveller. Chronic diarrhoea or remitting symptoms are an indication for stool cultures for Shigella, Salmonella, and Campylobacter, in addition to a careful search

374

R. C.

ARDUINO

AND

H.

L.

DUPONT

for stool parasites, since Giardia lamblia is the most common parasite acquired by travellers worldwide. Giardiasis can present with abdominal cramps and distension, the passage of many soft stools, and frequent passage of foul-smelling gas; when the infection is chronic, malabsorption and weight loss are characteristic. Cryptosporidium and Entamoeba histolytica are other protozoan parasites that cause prolonged diarrhoea. Clostridium difficile antibiotic-associated diarrhoea should be suspected in those patients with prior exposure to antibiotics before or concurrent with the onset of the diarrhoea. PREVENTION There are four ways that travellers may be able to reduce the likelihood of developing diarrhoeal illness while travelling from industrialized regions to developing tropical countries: careful food and beverage selection; use of prophylactic bismuth subsalicylate; use of prophylactic antimicrobial agents; and, finally, immunizations (an area of research). Dietary prevention The major area for the prevention of illness is to decrease exposure to the infectious agents. The available evidence indicates that contaminated food and water are the most important vehicles of diarrhoeal illness transmission in developing areas. This is particularly the case for ETEC diarrhoea, shigellosis and salmonellosis. Direct person-to-person spread may occur with some of the enteric pathogens, but it is not a frequent mode of transmission in the traveller. Public restaurants, cafeterias and street vendors often fail to meet adequate food hygiene standards, and food served may be contaminated with faecal flora, including Shigella, Campylobacter, Salmonella, Aeromonas and ETEC (Tjoa et al, 1977; Wood et al, 1983). Escherichia coli strains have been demonstrated in more than 40% of food items in restaurants and street vendors in Mexico, and most commonly in dairy products such as cheese, milk and butter, and from cooked food. Hamburgers and ice cubes have been reported as a source of Shigella. Vegetables may be grown in soil containing human excreta, they may be improperly washed, foods may be put back into their original contaminated containers after processing or cooking, adequate refrigeration may be lacking, and, finally, food preparers are not encouraged to wash their hands thoroughly nor are they often prevented from working while convalescing from a bout of diarrhoea. Studies of TD have been performed among US students who were living temporarily in Mexico while taking classes in a rural or urban setting, and a clear association between the location of food consumption and development of diarrhoea was demonstrated (Tjoa et al, 1977; Ericsson et al, 1980; Wood et al, 1983). Students who prepared more than half of their meals in their own apartments showed the lowest rates of diarrhoeal illness as well as isolation of LT-producing ETEC and Shigella strains; those who consumed

TRAVELLERS’

DIARRHOEA

375

most of their meals in the homes of Mexican families had the highest rate, while eating in public restaurants appeared to be of intermediate risk. Therefore, the single most important thing that can be done to minimize the potential for eating contaminated food is to select a safe restaurant. Advice from local persons or tour operators responsible for previous groups is extremely important because it is not possible to predict the degree of bacterial contamination of food by examining the appearance and outward cleanliness of public eating establishments. Cooked foods are generally safe if they are brought to the table steaming hot. Peeled fruits are safe. Leafy vegetables should be considered contaminated unless adequately cleaned at home, and hot sauces served in public restaurants are often contaminated. Food that sits on counters and is then rewarmed is dangerous unless brought to near-boiling temperatures. Rewarming temperatures are not ordinarily high enough to kill the bacterial enteropathogens. Unpasteurized milk and milk products should be avoided. Attention to careful food and beverage selection, as mentioned above, can decrease the risk of developing diarrhoea, but travellers frequently find it difficult to comply with these precautions, particularly those who are vacationing and who have an appetite for the local food. A follow-up study performed to investigate the influence of dietary indiscretion on the incidence of diarrhoeal illness among travellers to a developing area found that more than one-half of the trav.ellers had consumed salads, uncooked vegetables, unpeeled fruits or ice cubes, and the incidence of diarrhoea was significantly associated with the number of mistakes (Koziki et al, 1985). In developing countries, water often meets acceptable standards as it leaves the chlorination plant, but tap water may be contaminated because of breaks in water pipes permitting cross-contamination with sewage or surface water. Water-borne infections due to Giardia lamblia, E. coli, Shigella, Salmonella spp. and Norwalk virus have been reported. Bottled water or personally boiled water should be used for drinking and for brushing teeth. It is likely that tap-dispensed heated water, it if is too hot to touch, is safer than the cold water, and it may be used for brushing teeth when bottled water is unavailable. However, a temperature of 59°C or above is needed to eliminate the common enteric bacterial pathogens, and this temperature was commonly not reached in the tap water of a majority of hotels tested in tropical or subtropical areas (Bandres et al, 1988). Other sources of infection are ice cubes made with tap water and swimming pools without adequate chlorination. Carbonated bottled drinks are generally safe, because they have an acid pH. Methods to disinfect potentially contaminated water include heat, halogenation, water purification and filtration. Boiling water will kill almost all the enteropathogens but subsequent contamination must be avoided, specifically when storing the water in a container. Many commercial preparations containing iodine or chlorine are available to treat the water and make it potable. Although the susceptibility of enteropathogens to the halogen shows some differences, two drops of 5% chlorine bleach in a litre of water is enough to kill most of them. Treatment of the water with iodine is not recommended for pregnant women or for

376

R.

C. ARDUINO

AND

H.

L.

DUPONT

patients with thyroid disease. Water purifiers contain microfilters to remove soil particles and protozoan cysts, and release resin-bound iodine to eliminate bacteria and viruses. The exchange resin contained in these units is combined with triiodine, which is released when in contact with bacteria and/or viruses, resulting in a very low iodine residue in the drinking water and no change in taste. Finally, commercial filters have been shown to be useful for removing bacteria and parasites. Problems with these filters are their inability to remove viruses and frequent rupture. Chemoprophylaxis Controlled studies have demonstrated that bismuth subsalicylate (BSS) is effective in both the treatment and prophylaxis of TD. The mechanism of action of BSS, when employed as a preventive measure in TD, has not been completely elucidated. BSS, composed of approximately 60% bismuth and 40% salicylate, has antisecretory properties, important in its use, that are probably related to the salicylate moiety. The drug and the intestinal reaction products, bismuth oxychloride and bismuth hydroxides, are bactericidal (Sox and Olson, 1989), which possibly explains the most important property of the drug when employed for prophylaxis. Initial studies using the liquid presentation of BSS at a dose of 60 ml four times a day, which represents 4.2 g per day of the active agent, demonstrated a protective effect of 62% (DuPont et al, 1980). Two studies to determine the minimal protective dose indicate that 2.1 g per day given in tablet form (two tablets four times a day with meals and at bedtime) is not only effective but is the most practical formulation to use (Steffen et al, 1986; DuPont et al, 1987). It appears that the four-times-per-day dosing schedule is important, probably because of recurrent exposure to the infecting organism during meals. This dosage can be administered safely for periods of up to 3 weeks. For travellers or others expecting to reside in the high-risk area for more than 3 weeks, chemoprophylaxis is probably not an appropriate means of reducing the threat of illness. The most frequent side-effects are harmless darkening of stools and tongue, and mild transient tinnitus. When BSS is taken at a dose of 2.1 g per day, the amount of salicylate ingested (1.6 g) is considerably less than the recommended dose of aspirin (4.0 g). Nevertheless, those travellers already taking salicylate, anticoagulants or hyperuricaemic drugs, and patients with gout or those allergic to salicylate should avoid BSS. Antimicrobial prophylaxis has been highly effective in preventing episodes of TD. The protection, however, lasts only as long as the drug is being taken. Doxycycline has been studied in several clinical trials as a prophylactic agent for TD because its activity against most of the entericpathogens, long half-life which permits once-a-day administration, enterohepatic circulation with excretion primarily (90%) in the faeces, and a low incidence of side-effects. However, resistance to the tetracyclines has become increasingly common among enteric pathogens isolated from different parts of the world, which limits the use of these antibiotics in the prophylaxis and treatment of diarrhoeal illness (Murray, 1986).

TRAVELLERS’

DIARRHOEA

377

Doxycycline has been shown to confer a protection of approximately 85% when used at a dose of 100 mg daily. One study showed that when administered at a dose of 100mg twice a week, doxycycline was not protective (Santosham et al, 1981). Photosensitivity, Candida overgrowth and gastrointestinal symptoms, such as nausea and vomiting, have been reported in less than 1% of the adults treated. Tetracyclines are not recommended for use in patients younger than 12 years of age and pregnant women because they cause brown discolouration of the teeth and may retard growth of bone in the human fetus and in children. Trimethoprim-sulphamethoxazole (TMP-SMX) is highly effective for the prevention of TD. TMP-SMX conferred 71% protection over a period of 3 weeks when 160 mgl800 mg twice daily was administered to US students visiting Mexico. When the same dose of drug was administered once daily, 95% protection over a period of 2 weeks was observed (DuPont et al, 1983). To a lesser extent, trimethoprim alone was also effective (50% protection) in preventing TD for a period of 2 weeks when taken once a day (DuPont et al, 1983). Common problems resulting from using TMP-SMX are skin rash in l-5% of cases, and the development of resistance of the gut flora during drug administration. In addition, the drug is inactive against Cumpylobacter. More serious side-effects of TMP-SMX usage may be seen, although rarely, including the Stevens-Johnson syndrome and antibiotic-associated colitis. TMP-SMX treatment had to be discontinued because of cutaneous eruptions in 14% of travellers when used at a dose of 160 mgB00 mg twice daily, while only 1% of those receiving 160mg/800mg once daily had to discontinue the prophylaxis (DuPont et al, 1983). The quinolones have also been studied in the prophylaxis of TD because of their broad spectrum of antimicrobial activity against most enteric bacterial pathogens, long serum half-life, high concentration in the intestinal tract, and general safety. Norfloxacin and ciprofloxacin have been demonstrated to decrease the chance of infection by more than 90% when taken orally, in doses of 400 mg per day and 500 mgper day, respectively (WistrGm et al, 1987; Rademaker et al, 1989; Scott et al, 1990). No significant side-effects have been reported in placebo-controlled trials. Another advantage that these agents may have over other prophylactic agents such as TMP-SMX or doxycycline is the apparent low rate of development of resistance during the treatment, probably because plasmid-mediated resistance to quinolones has not yet been found in clinical specimens. The quinolones are not approved for use in children or in pregnant or nursing women because of the reported damage to the developing cartilage in young animals. In considering how and when antimicrobial agents should be employed in preventing TD, several points should be kept in mind. While efficacy of antimicrobial agents has been proven to be high, there is a measurable risk of causing severe side-effects that, infrequently, are life-threatening. Furthermore, development of antibiotic resistance of the gut flora during drug administration may also occur and, indeed, is occurring worldwide. Finally, antimicrobial agents have been proven to be so effective in treating the disorder that early antimicrobial therapy of illness represents an

378

R.

C. ARDUINO

AND

H.

L.

DUPONT

excellent alternative approach to prophylaxis for most travellers. Currently, prophylaxis with antimicrobial agents is recommended only: for travellers with concomitant serious medical illness in whom an episode of diarrhoea might be poorly tolerated, such as inflammatory bowel disease, acquired immunodeficiency syndrome, insulin-dependent diabetes mellitus, and chronic renal failure; for those whose underlying condition might predispose to complications of dehydration, such as patients taking digoxin or diuretics; for those travellers with decreased gastric acidity; or, finally, for those travellers on critically important travel in whom temporary incapacity is unaffordable. Immunization Cholera vaccination is required for entry to some countries, but it is a minor threat for regular travellers, the estimated risk being 1: 500 000. Although a parenteral vaccine for Vibrio cholerue 01 is available, this vaccine reduces the incidence of diarrhoea in the vaccinated persons in only X1-70% of instances, and it does not prevent asymptomatic infection. The vaccine is given in two doses, separated by 3-4 weeks, and is associated with local and systemic side-effects, such as fever and malaise (Centers for Disease Control, 1983). This preparation is not recommended for anyone unless required by law. A new oral-killed whole-cell and B subunit-binding cholera toxin vaccine has been shown to give short-term protection against severe diarrhoea due to ETEC in endemic areas. It has been shown to induce a protection rate of 86% for 3 months against severe ETEC-associated diarrhoea when given to Bangladeshi mothers and children (Clemens et al, 1988). Because ETEC is the most common single cause of TD, a prospective double-blind study was conducted among Finnish tourists travelling to Morocco to evaluate protection against the illness. The overall reduction of diarrhoeal illness among those travellers vaccinated was 23%. The protection was 52% when ETEC strains had been isolated from diarrhoeal stools and increased to 60% when LT ETEC was the causative agent (Peltola et al, 1991). This crossprotection can be explained by the structural and immunological similarities between the heat-labile E. coli enterotoxin and cholera toxin. Other vaccines currently under investigation include killed whole-cell vibrios in combination with ETEC toxins and CFAs. TREATMENT Fluid therapy The most important aspect of treatment of acute watery diarrhoea is replacement of fluids and electrolytes. Enterotoxigenic organisms can cause severe fluid loss, which is the most serious consequence of diarrhoeal illness. The degree of dehydration, the presence of systemic toxicity, the underlying disease and the clinical findings are useful in determining the need for

TRAVELLERS’

DIARRHOEA

379

hospitalization and the need for antimicrobial therapy in all patients with a diarrhoeal illness, but severe dehydration does not occur in most previously healthy travellers with a diarrhoeal illness. In patients with mild to moderate dehydration, oral glucose-electrolyte solutions are recommended. Glucose is needed because sodium and water are absorbed as a result of coupled glucose absorption, a mechanism that is intact even in the setting of severe diarrhoea. Glucose-electrolyte fluids and anhydrous packets of salt, such as those recommended by the World Health Organization, are commercially available. These solutions, however, have a high concentration of sodium and they are primarily recommended in the more severe forms of diarrhoea. Those adults with mild or no dehydration can be treated with a diet consisting of simple fluids with a high content of electrolytes, such as fruit juices (potassium supplementation) and salty soups (sodium chloride supplementation), without milk and other dairy products, to avoid the osmotic load in the intestine in the face of disaccharidase depletion due to the intestinal mucosal damage. Non-antibiotic

therapy

The use of antimotility agents in TD is controversial. Despite some studies that have shown that dysentery may be prolonged by antimotility drugs, they continue to be prescribed frequently for travellers. Of the available nonantibiotic agents, BSS and loperamide hydrochloride confer significant relief to the traveller with diarrhoea. Loperamide hydrochloride, however, at a daily dosage of 8-16 mg for treatment of acute non-specific diarrhoea, provides faster and more effective relief than BSS (DuPont et al, 1990). The use of these agents seems to be reasonable in patients with mild to moderate diarrhoea without fever or dysentery. Loperamide hydrochloride is well tolerated without addiction potential and has an excellent safety record. BSS taken orally at a dosage of 30ml every half hour for eight doses significantly decreased the mean number of unformed stools passed by a group of US students travelling to Mexico when compared with placebo (Ericsson et al, 1986). BSS was notably effective in a subgroup of patients with diarrhoea due to ETEC, but the efficacy was not significant among those travellers with shigellosis. Combination therapy with TMP-SMX and loperamide hydrochloride was highly effective for relieving symptoms in travellers with diarrhoeal illness. Active diarrhoea was abated, on average, within 1 h after beginning treatment, compared to 59h in the placebo group, and the combination regimen was associated with a low rate of treatment failure (Ericsson et al, 1990). This combination therapy seemed to be more effective than loperamide hydrochloride alone and superior to TMP-SMX given for 3 days. On the other hand, one study done in military personnel travelling to Cairo, where ETEC was the predominant cause of diarrhoea in the traveller, showed that the combination of ciprofloxacin and loperamide hydrochloride was not significantly better than treatment with ciprofloxacin alone for the therapy of diarrhoeal illness (Taylor et al, 1991). A rational approach might be to instruct patients to use an antimicrobial

380

R.

C. ARDUINO

AND

H.

L.

DUPONT

agent and loperamide hydrochloride at the first sign of distressing diarrhoea, as long as visible blood in the stool and fever are absent. In these cases, the antimicrobial agent without loperamide would be the preferred treatment. Antimicrobial

treatment

TD has been successfully treated with various antimicrobial agents in several trials, These trials have shown that the duration of post-treatment diarrhoea, which usually last 3-5 days, can be shortened to approximately 1 day by the use of effective antimicrobial agents. Patients with mild to moderate diarrhoea (i.e. fewer than five unformed stools per day, without fever and/or cramps, pain, nausea and vomiting) may be treated without antimicrobial drugs in light of the self-limiting nature of the diarrhoeal illness. Antimicrobial agents are strongly recommended as empirical therapy for those travellers with moderate to severe diarrhoea, especially when bloody stools, fever and/or symptoms of systemic toxicity occur, or for those travellers with moderate diarrhoea who fail to respond adequately 24-48 h after the initiation of symptomatic therapy. The empirical selection of an antibiotic is based on the susceptibility of enteric pathogens prevalent in the area where the diarrhoea was acquired and the clinical presentation. Resistance to ampicillin among enteric pathogens is currently widespread, and resistance to trimethoprim and TMP-SMX is becoming more common in areas of the world where TD occurs (Murray, 1986). Severe watery diarrhoea may be associated with ETEC infection. In these cases, TMP-SMX, one tablet twice daily for 3-5 days, is the traditional treatment and is still active against most enteric pathogens from many regions of the world while resistance is occurring in other areas. Trimethoprim alone was also demonstrated to be effective in the treatment of acute diarrhoea caused by ETEC strains or Shigella strains, and in diarrhoeal illness not associated with identifiable organisms in travellers to Mexico (DuPont et al, 1982). Antimicrobial therapy is not indicated in patients with mild to moderate enterocolitis produced by Salmonella except in elderly patients, young infants (< 1 year), toxic patients, those patients receiving immunosuppressive therapy or those who have a malignancy, a haemoglobinopathy, chronic gastrointestinal disease or HIV infection. This recommendation relates to the associated bacteraemia in 6-S% of patients with Salmonella gastroenteritis and, consequently, increased incidence of complications in these population with underlying diseases or at the extremes of age. Dysentery may be caused by bacterial enteropathogens in over half of the cases, and Shigella and Campylobacter jejuni are frequently responsible. The quinolones are agents with a broad spectrum of activity against the invasive gastrointestinal pathogens which cause diarrhoea in the adult traveller, including Salmonella, Shigella, Campylobacter, invasive E. coli, Yersinia, Plesiomonas, Aeromonas and non-cholera vibrios. For adult patients with dysentery, a quinolone, such as norfloxacin 400mg, ciprofloxacin 500mg or ofloxacin 300mg, taken twice daily for 3-5 days, is

TRAVELLERS’

DIARRHOEA

381

clinically effective (Ericsson et al, 1987a; Wistrom et al, 1989; DuPont et al, 1992b). A recent study done on US military personnel in central Thailand showed that duration of the diarrhoeal illness was the same whether ciprofloxacin was given as a single 750-mg dose or as 500 mg twice daily for 3 days (Petruccelli et. al, 1992). The same study further indicated that the addition of loperamide to ciprofloxacin gave little overall additional benefit. Trimethoprim alone or combined with sulphamethoxazole is not effective as therapy for diarrhoeal disease due to Campylobacter jejuni, a common cause of wintertime TD, or to resistant enteric bacteria. In areas of the world where campylobacteriosis is more prevalent or where resistance to TMPSMX among other enteropathogens is high, the use of quinolones is recommended for adults. There is evidence that erythromycin reduces Campylobacter excretion and improves dysentery when administered early in the course of the disease. The recommended oral dose is 500mg four times a day for 7 days. Oral aztreonam, which is a poorly absorbed monobactam antibiotic with a high degree of activity in vitro against bacterial enteropathogens was demonstrated to be effective therapy for bacterial diarrhoea in travellers to Mexico. Aztreonam was administered in a dose of 100 mg three times a day for 5 days (DuPont et al, 1992a). SUMMARY

Although TD is usually a mild and self-limited illness, 30-50% of travellers from industrialized to less developed countries are affected. Enterotoxigenic E. co/i (ETEC) remain the most frequent cause, being identified in 40-70% of cases. TD frequently occurs within the first 2 weeks of arrival in the foreign country. The clinical manifestation is variable, but watery diarrhoea is the most common clinical presentation. Chronic diarrhoea or remitting symptoms after empirical therapy in the returning traveller are indications for a stool culture and a careful search for stool parasites. Since the major precaution against TD is to avoid exposure to the infectious agents, careful selection of food and beverage is crucial. Bismuth subsalicylate has been proven to be safe and effective in the treatment and prophylaxis of TD. The tablet form has removed the inconvenience of previously required luggage space. Doxycycline, trimethoprimisulphamethoxazole, trimethoprim and the quinolones have been shown to be effective for prevention of diarrhoea. However, side-effects, superinfection, development of antibiotic resistance and easy-to-treat illness may limit the use of these antimicrobial agents to those travellers with concomitant serious medical conditions that would be adversely affected by diarrhoea, or travellers with unaffordable temporary incapacity. A new oral-killed whole-cell and B-subunit cholera toxin vaccine was demonstrated to induce protection against severe ETEC-associated diarrhoea. This is a promising field under investigation.

R. C. ARDUINO AND H. L. DUPONT

382

Finally, fluid replacement is the most important aspect of treatment. Patients with moderate to severe TD can be treated with one of the abovementioned antimicrobial agents for 3-5 days. Selection of the antimicrobial agent is based on the pattern of resistance and the enteric organism prevalent in the geographical area. While TMP-SMX remains active against the strains prevalent in Mexico during summertime, the quinolones represent the choice for the therapy of diarrhoea acquired in the high-risk areas of South America, Africa and Asia. REFERENCES Arduino RC, Mathewson JJ, Smith MA & DuPont HL (1991) Antimicrobial resistance of Enteroadherent E. coli (EAEC). Program and Abstracts of the Thirty-first Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago abstiah 112: p 117. Bandres JC. Mathewson JJ & DuPont HL (1988) Heat suscentibilitv of bacterial enteropathogens. Implications for the prevention of travelers’ diarrhea: Archives of Internal Medicine

148: 2261-2263.

Bhan MK, Raj P, Levine MM et al (1989) Enteroaggregative Escherichia coli associated with persistent diarrhea in a cohort of rural children in India. Journal of Infectious Diseases 159: 1061-1064. Black RE (1990) Epidemiology of travelers’ diarrhea and relative importance of various pathogens. Reviews of Infectious Diseases 12 (supplement 1): S7LLS79. Blaser MJ (1986) Brainerd diarrhoea: a newly recognized raw milk-associated enteropathy. The Journal of the American Medical Association 256: 510-511. Bolivar R, Conklin RH, Vollet JJ et al (1978) Rotavirus in travelers’ diarrhea: study of an adult student population in Mexico. Journal of Infectious Diseases 137: 324-327. Brodsky RE, Spencer HC Jr & Schultz MG (1974) Giardiasis in American travelers to the Soviet Union. Journal of Infectious Diseases 130: 319-323. Buchwald DS & Blaser MJ (1984) A review of human salmonellosis: II. Duration of excretion following infection with nontyphi Salmonella. Reviews of Infectious Diseases 6: 345-356. Casemore DP (1990) Foodborne illness. Lancer 336: 1427-1432. Centers for Disease Control (1983) Health information for international travelers 1983. Morbidity

and Mortality

Weekly’Report

32 (supplement)

7@71.

Centers for Disease Control (1991) Cholera-New York. 1991. Morbiditv and Mortality Weekly Report 40: 516-516. ’ Clemens JD, Harris JR, Sack DA et al (1988) Field trial of cholera vaccine in Bangladesh: results of one year of follow-up. Journal of Infectious Diseases 158: 60-69. Dandoy S (1966) The diarrhea of travelers: incidence in foreign students in the United States. California

Medicine

104: 458-462.

DuPont HL, Haynes GA, Pickering LK et al (1977) Diarrhea of travelers to Mexico. American Journal

of Epidemiology

105: 37-41.

DuPont HL, Sullivan P, Evans DG et al (1980) Prevention of travelers’ diarrhea (emporiatric enteritis). Prophylactic administration of subsalicylate bismuth. Journal of the American Medical

Association

243: 237-241.

DuPont HL, Reves RR, Galindo E et al (1982) Treatment of travelers’ diarrhea with trimethoprimisulfamethoxazole and trimethoprim alone. New England Journal of Medicine 307: 130-135. DuPont HL, Galindo E, Evans DG et al (1983) Prevention of travelers’ diarrhea with trimethoprimsulfamethoxazole and trimethoprim alone. Gastroenterology 84: 75-80. DuPont HL. Ericsson CD, Johnson PC & Cabada FJ (1986) Antimicrobial agents in the prevention of travelers’ diarrhea. Reviews of Infectious Diseases 8 (supplemek 2): S167s171. DuPont HL, Ericsson CD, Johnson PC et al (1987) Prevention of travelers’ diarrhea by the tablet formulation of bismuth subsalicylate. Journal of the American Medical Association 257: 1347-1350.

TRAVELLERS’

383

DIARRHOEA

DuPont HL, Sanchez JF, Ericsson CD et al (1990) Comparative efficacy of loperamide hydrochloride and bismuth subsalicylate in the management of acute diarrhea. American Journal of Medicine 88 (supplement 6A): 15s19% DuPont HL, Ericsson CD, Mathewson JJ et al (1992a) Oral aztreonam, a poorly absorbed yet effective therapy for bacteria1 diarrhea in U.S. travelers to Mexico. Journal of fhe American Medical Association 267: 1932-1935. DuPont HL, Ericsson CD, Mathewson JJ & DuPont MW (1992b) Five versus three days of ofloxacin therapy for travelers’ diarrhea: a placebo-controlled study. AntimicrobialAgents and Chemotherapy

36: 87-91.

Ericsson CD, Pickering LK, Sullivan P & DuPont HL (1980) The role of location of food consumption in the prevention of travelers’ diarrhea in Mexico. Gastroenterology 79: 812-816. Ericsson CD, DuPont HL & Johnson PC (1986) Non-antibiotic therapy for travelers’ diarrhea. Reviews

of Infectious

Diseases

8 (supplement

2): S202-S206.

Ericsson CD, Johnson PC, DuPont HL et al (1987a) Ciprofloxacin or trimethoprimsulfamethoxazole as initial therapy for travelers’ diarrhea. Annals of Internal Medicine 106: 216220.

Ericsson CD Patterson TF & DuPont HL (1987b) Clinical presentation as a guide to therapy for travelers’ diarrhea. American Journal of the Medical Sciences 30: 91-96. Ericsson CD, DuPont HL, Mathewson JJ et al (1990) Treatment of travelers’ diarrhea with sulfamethoxazole and trimethoprim and loperamide. Journal of the American Medical Association

263: 257-261.

Evans DJ Jr, Ruiz-Palacios G, Evans DG et al (1977) Humoral immune response to the heat-labile enterotoxin of Escherichia coli in naturally acquired diarrhea and antitoxin determination by passive immune hemolysis. Infection and Immunity 16: 781-788. Gorbach SL (1982) Travelers’ diarrhea. New England Journal of Medicine 307: 881-883. Gracey M, Burke V, Robinson J et al (1984) Aeromonas spp. in travellers’ diarrhoea. British Medical

Journal

289: 658.

Haberberger RL Jr, Mikhail IA & Ismail TF (1991) Enteritis due to multiresistant enteroadherent Escherichia coli. Lancet 337: 235-236. Holmberg SD, Kaye Wachsmuth I, Hickman-Brenner FW et al (1986) Plesiomonas enteric infections in the United States. Annuls of Internal Medicine 105: 69&694. Howden CW & Hunt RH (1987) Relationship between gastric secretion and infection. Gut 28: 96-107. Hyams KC, Bourgeois AL, Merrel BR et al (1991) Diarrhea1 disease during operation Desert Shield. New Enpland Journal of Medicine 325: 1423-1428. Johnson PC, Hoy J, Mathewson JJ, Ericsson CD &DuPont HL (1990) Occurrence of Norwalk virus infections among adults in Mexico. Journal of Infectious Diseases 162: 389-393. Jokipii L & Jokipii AM (1974) Giardiasis in travelers: a prospective study. Journalof Infectious Diseases

130: 295-299.

Kean BH (1969) Turista in Teheran. Travellers’ diarrhea at the Eighth International Congress of Tropical Medicine and Malaria. Lancet ii: 583-584. Keswick BH, Blacklow NR, Cukor GC et al (1982) N orwalk virus and rotavirus in travellers’ diarrhoea in Mexico. Lancet i: 109-110 (letter). Koch KL, Phillips DJ & Aber RC (1985) Cryptosporidiosis in hospital personnel. Evidence for person to person transmission. Annals of Internal Medicine 102: 593-596. Kozicki M, Steffen R & Schar M (1985) ‘Boil it, cook it, peel it or forget it.’ Does this rule prevent travellers’ diarrhoea? International Journal of Epidemiology 14: 169-172. Long EG, Ebrahimzadeh A, White EH et al (1990) Alga associated with diarrhea in patients with acquired immunodeficiency syndrome and in travelers. Journal of Clinical Microbiology 28: 1101-1104. Lowenstein MS, Balows A & Gangarosa EJ (1973) Turista at an international congress in Mexico. Lancet i: 529-531. Macdonald I & Kean BH (1962) Diarrhea of travelers. Journal of the Americnn Medical Association 181: 804 (letter). Mathewson JJ, Johnson PC, DuPont HL et al (1985) A newly recognized cause of travelers diarrhea: enteroadherent Escherichia coli. Journal of Infectious Diseases 151: 471-475. Mathewson JJ, Oberhelman RA, DuPont HL et al (1987) Enteroadherent Escherichia coli as a cause of diarrhea among children in Mexico. Journal of Clinical Microbiology 25: 1917-1919.

384

R. C. ARDUINO

AND

H.

L.

DUPONT

Mathewson JJ, Flores JF, DuPont HL et al (1990) Risk of acquisition of Entamoeba histolytica infection during long-term travel to Mexico. Travel Medicine International 82: 65-68. Mattila L, Sitonen A, Kyronseppa H et al (1992) Seasonal variation in etiology of travelers’ diarrhea. Journal of Infectious Diseases 165: 385-388. Morger H, Steffen R & Schar M (1983) Epidemiology of cholera in travellers, and conclusions for vaccination recommendations. British Medical Journal 286: 184-186. Murray BE (1986) Resistance of Shigella, Salmonella, and other enteric pathogens to antimicrobial agents. Reviews of Infectious Diseases 8 (supplement 2): S172-S181. Navin TR & Juranek DD (1984) Cryptosporidiosis: clinical, epidemiologic, and parasitologic review. Reviews of Infectious Diseases 6: 313-327. Ortega YR, Sterling CR: Gilman RR et al (1993) Cyclospora species-A new protozoan vathogen of humans. The New England Journal of Medicine 328: 130&1312. Pelt&a H,?Zitonen A, Kyronseppa H it al (1991) Prevention of travellers’ diarrhoea by oral B-subunit/whole-cell cholera vaccine. Lancet 338: 1285-1289. Petruccelli BP, Murphy GS, Sanchez JL et al (1992) Treatment of travelers’ diarrhea with ciprofloxacin and loperamide. Journal of Infectious Diseases 165: 557-560. Pitarangsi C, Echeverria P, Whitmire R et al (1982) Enteropathogenicity of Aeromonas hydrophila and Plesiomonas shigelloides: prevalence among individuals with and without diarrhea in Thailand. Infection and Immunity 35: 666673.Rademaker CMA. Hoevelman IM. Wolfhaaen MJHM et al (1989) Results of a double-blind placebo-controlled ‘study using ciprol?oxacin for prevention of travelers’ diarrhea. European Journal of Clinical Microbiology and Infectious Diseases 8: 69G694. Ryder RW, Weels JG & Gangarosa EJ (1977) A study of travelers’ diarrhea in foreign visitors to the United States. Journal of Infectious Diseases 136: 605-607. Ryder RW, Oquist CA, Greenberg H et al (1981) Travelers’ diarrhea in Panamanian tourists in Mexico. Journal of infectious Diseases 144: 442-448. Santosham M, Sack RB, Froehlich JL et al (1981) Biweekly prophylactic doxycycline for travelers’ diarrhea. Journal of Infectious Diseases 143: 598-602. Schultz MG (1975) Giardiasis. Journal of the American Medical Association 233: 1383-1384. Scott DA, Haberberger RL, Thornton SA & Hyams KC (1990) Norfloxacin for the prophylaxis of travelers’ diarrhea in US military personnel. American Journal of Tropical Medicine and Hygiene 42: 16&164. Snyder JD & Blake PA (1982) Is cholera a problem for U.S. travelers? Journal of the American Medical Association 247: 2268-2269. Sox TE & Olson CA (1989) Binding and killing of bacteria by bismuth subsalicylate. Antimicrobial Agents and Chemotherapy 33: 2075-2082. Speelman P, Struelens MJ, Sanyal SC & Glass RI (1983) Detection of Campylobacterjejuni and other potential pathogens in travellers’ diarrhoea in Bangladesh. Scandinavian Journal of Gastroenterology 18 (supplement 84): 19-23. Steffen R (1986) Epidemiologicstudies of travelers’ diarrhea, severe gastrointestinal infections, and cholera. Reviews of Infectious Diseases S (supplement 2): S122-S130. Steffen R, DuPont HL, Heusser R et al (1986) Prevention of travellers’ diarrhoea by the tablet form of bismuth subsalicylate. Antimicrobial Agents and Chemotherapy 29: 625-627. Taylor DN, Echeverria I, Blaser MJ et al (1985) Polymicrobial aetiology of travellers’ diarrhoea. Lancet i: 381-383. Taylor DN, Sanchez JL, Candler W et al (1991) Treatment of travelers’ diarrhea: ciprofloxacin plus loperamide compared with ciprofloxacin alone. Annals of Internal Medicine 114: 731-734. Tjoa WS, DuPont HL, Sullivan P et al (1977) Location of food consumption and travelers’ diarrhea. American Journal of Epidemiology 106: 61-66. Wanger A, Murray BE, Echeverria P et al (1988) Enteroinvasive Escherichia coli in travelers’ diarrhea. Journal of Infectious Diseases 158: 64&642. Wanner RG, Atchley FO & Wasley MA (1963) Association of diarrhea with Giardia lamblia in families observed weekly for occurrence of enteric infections. American Journal of Tropical Medicine and Hygiene 12: 851-853. Wilson ME, von Reyn F & Fineberg HV (1991) Infections in HIV-infected travelers: risk and prevention. Annals of Internal Medicine 114: 582-592. Wistrom J, Ragnar Norrby S, Burman LG et al (1987) Norfloxacin versus placebo for prophylaxis against travellers’ diarrhoea. Journal of Antimicrobial Chemotherapy 20: 563-574.

TRAVELLERS’

DIARRHOEA

385

WistrCim J, Jertborne M, Hedstrom SA et al (1989) Short-term self-treatment of travellers’ diarrhoea with norfloxacin: a placebo-controlled study. Journal of Antimicrobial Chemotherapy

23: 905-913.

Wood LV, Ferguson LE, Hogan Pet al (1983) Incidence of bacterial enteropathogens in foods from Mexico. Applied Environmental Microbiology 46: 328-332. Yamamoto T, Echeverria P & Yokota T (1992) Drug resistance and adherence to human intestines of enteroaggregative Escherichia coli. Journal of Infectious Diseases 165: 744749.