Escherichia coli strains that cause diarrhea are classified into four major categories: enterotoxigenic, enteroinvasive, enteropathogenic, and enterohemorrhagic E. coli. This chapter describes these four groups of E. coli and discusses their transmission from food and water. It explores a newer technique to identify these E. coli strains in food and water. The proportion of diarrhea that is food- and waterborne may be lower in developing than in developed countries because of the more varied and frequent opportunities for other modes of transmission. From volunteer studies in adults the infective dose of Enteropathogenic E. coli (EPEC) is highsuggesting that EPEC is most likely transmitted by food or water. EPEC outbreaks of gastroenteritis in adults that were either water- or foodborne have been reported from Romania, Sweden, and the United States. As reviewed by F. L. Bryan, E. coli is commonly isolated from hands of workers in the food industry in developed countries.

Although numerous studies have shown that diarrhea is the most common illness occurring during the first few weeks of travel, systematic studies of the incidence of diarrhea during long-term residence in developing countries have not been performed. We conducted a cohort study of the incidence and etiology of diarrhea among 77 expatriate adults who had lived in Nepal for <2 years. Persons were followed prospectively for up to 1 year (mean, 9 months). The incidence of diarrhea during the surveillance period was 3.3 episodes of diarrhea per person per year, or 0.27 episodes per person per month. The annual attack rate of specific pathogens was 42% for enterotoxigenic Escherichia coli, 32% for Cyclospora species, 16% for Giardia lamblia, 16% for Shigella species, 10% for Campylobacter species, > or =10% for rotavirus, and 6% for Entamoeba histolytica. This study suggests that adult persons from developed countries who move to developing countries such as Nepal remain at high risk for diarrhea during their first 2 years of residence.

Diarrheal disease research is focused on new interventions, new and improved vaccines, and new diagnostic methods. Micronutrient intervention using zinc and vitamin A has been very important in its impact on childhood diarrheal disease in developing countries. Vaccines for rotavirus and bacterial pathogens such as enterotoxigenic Escherichia coli, Shigella and Salmonella now seem feasible. New diagnostic methods have been useful in defining new viral causes of diarrhea.

The etiology of gastroenteritis was determined in children and adults with diarrhea seen at a district hospital and three government health clinics in Suan Phung, western Thailand, in 1991. Enteric viruses (rotavirus and astrovirus) were identified in 40%, shigellae in 18%, attaching and effacing Escherichia coli in 13%, Campylobacter jejuni in 9%, and enterotoxigenic E. coli in 7% of children <5 years old with diarrhea seen at the hospital. Enteric viruses were detected in 15% (24/156) of patients with diarrhea greater than or equal to 5 years old and were the only enteric pathogens identified in 12 patients ages 7-79 years (2 astrovirus, 10 rotavirus infections). Attaching and effacing E. coli, rotavirus, and astrovirus were potential causes of diarrhea in children and adults in this population.

Diarrhoea in developing countries is caused by an increasingly long list of bacterial, viral, and parasitic pathogens with rotavirus, enterotoxigenic Escherichia coli, Campylobacter, Shigella, and Salmonella heading the list. Using methods to detect most of the known enteropathogens, one or more enteropathogen(s) is isolated in two-thirds of diarrhoeal illnesses in the developing world. Many of these enteropathogens are also frequently isolated from children without diarrhoea. An aetiologic agent is more frequently isolated from cases of invasive diarrhoea than from those with secretory diarrhoea. Deoxyribonucleic acid probes have proved very useful in detecting pathogens such as enterotoxigenic (ETEC), enteroinvasive (EIEC), and enteropathogenic E. coli (EPEC), but have not yet proved to be particularly rapid or less expensive. Molecular biology has proved useful in epidemiological studies as a means of strain identification. Plasmids were initially used as convenient markers and proved useful in identifying epidemic strains of bacteria. Other molecular markers, such as ribotyping, are accurate enough to be used as taxonomic tools.

We performed a case-control study of diarrhoea to determine its causes in children less than 1 year old in Guangzhou, People's Republic of China, in April to September 1989. Stools were cultured for Salmonella, Shigella, Campylobacter and vibrios by standard techniques; rotavirus (RV) was identified by polyacrylamide gel electrophoresis; and specific deoxyribonucleic acid probes were used to identify Escherichia coli containing genes coding for Shiga-like toxin I and II, enteropathogenic E. coli adherence factor, and enteroinvasive E. coli (EIEC). E. coli isolates were tested for heat-labile toxin (LT) and heat-stable toxin (ST) production and mannose-resistant adherence to HeLa cells. Rotavirus was identified in 13 of 174 children with diarrhoea (cases) and in 2% of 174 age-matched children without diarrhoea (controls), P less than 0.001. C. jejuni was identified in 10% of cases and 2% of controls, P = 0.003. Giardia lamblia was identified in 4 cases, LT and ST enterotoxigenic E. coli (ETEC) in 2, and S. flexneri in 1 case; they were not found in controls. ETEC that produced LT only was isolated from 5 cases and 3 controls, P = 0.721; E. coli that adhered to HeLa cells in a diffuse pattern was isolated from 30 cases and 40 controls, P = 0.229; and E. coli that adhered in an aggregative pattern was isolated from 20 cases and 18 controls, P = 0.863. EIEC was not isolated from cases or controls. Nine cases (5%) developed persistent diarrhoea (greater than 14 d duration). C. jejuni and aggregative E. coli were isolated from different children with persistent diarrhoea.(ABSTRACT TRUNCATED AT 250 WORDS)

Venom and oropharyngeal swabs from freshly captured Malayan pit vipers (Calloselasma rhodostoma) in southern Thailand and captive specimens in England were cultured aerobically and anaerobically to identify the bacterial flora which might contaminate wounds inflicted by bites of this species. The snakes' mouths contained a wider range of organisms than their venoms, especially gut-related Gramnegative rods sucfi as Enterobacter and Pseudomonas species and some staphylococci and clostridia. There were fewer positive cultures from captive snakes. C. rhodostoma venom inhibited the growth of group A streptococci and, to a lesser extent, that of Staphylococcus aureus and Clostridium perfringens but not that of 2 Gram-negative organisms. Secondary bacterial infection is an important complication of snake bite, especially of necrotic wounds. A combination of gentamicin with benzyl penicillin would have prevented infection with, or treated, most of the bacteria isolated from snake venoms and mouths in Thailand.

A severe illness characterised by bloody diarrhoea and intestinal dysfunction was recognised at an evacuation site on the Thai-Kampuchean border. From June, 1985, to July, 1986, the illness occurred in 62 Khmer children aged 10 months to 10 years (mean 4 years); it was characterised by bloody diarrhoea (94%), fever (90%), and abdominal pain (78%). The overall mortality rate was 58%. Among 16 children who died and underwent necropsy, small-intestinal necrosis of varying severity was found; in 5 of these children small-intestinal lesions with areas of full-thickness necrosis were seen that histologically resembled those in cases of enteritis necroticans (pigbel) in Papua New Guinea. Beta-toxin-producing Clostridium perfringens type C was isolated from 2 of 23 children from whom specimens for anaerobic cultures were collected, and antibodies to beta toxin were detected in 5 of 9 survivors but not in 10 healthy, age-matched control children. These cases show that enteritis necroticans can cause substantial morbidity and mortality outside Papua New Guinea.

Of 35 US Peace Corps volunteers in Thailand, 20 (57%) had a total of 30 episodes of diarrhoea during their first 6 weeks in the country. Enteric pathogens were associated with 90% of the episodes. A single pathogen was identified in 17 (57%) episodes, 2-4 pathogens were identified in 10 (33%) episodes, and there were 15 symptomless infections. Enterotoxigenic Escherichia coli (ETEC) was identified in 37% of these episodes, and various salmonella serotypes were isolated in 33%. Infections with 9 other enteric pathogens were also identified: Campylobacter jejuni (17%), Plesiomonas shigelloides (13%), Aeromonas hydrophila (10%), Blastocystis hominis (7%), Norwalk virus (7%), Vibrio parahaemolyticus (3%), non-O1 Vibrio cholerae (3%), Vibrio fluvialis (3%), and rotavirus (3%). In total, 56 enteric infections were documented in 35 volunteers.

Forty-two strains of Escherichia coli that agglutinated in pools of antisera used to identify "enteropathogenic" serotypes were tested for heat-labile and heat-stable toxin production and for their ability to invade intestinal mucosa. None of the strains tested were enterotoxigenic or enteroinvasive as determined by the adrenal cell (heat-labile toxin), the suckling mouse (heat-stable toxin), or guinea pig eye (invasive capacity) assays. Our observations suggest that serotyping of E. coli is an unreliable method to identify isolates that are capable of causing gastroenteritis, at least as determined by available in vitro techniques.

A HAZARD of suboptimal antimicrobial therapy is the potential of creating an environment conducive to the growth of resistant organisms. This risk is potentially present in infants with meningitis caused by enteric gram-negative bacteria because of the poor accumulation of aminoglycoside antibiotics in the cerebrospinal fluid.1 Enteric gram-negative bacteria have previously been reported to acquire resistance to gentamicin during therapy of neonatal meningitis,2 , 3 but the phenomenon was incompletely documented and mechanism of acquisition not characterized.4 We have observed an Escherichia coli causing meningitis in a newborn infant that appeared to acquire resistance to several aminoglycoside antibiotics during therapy.Case Report . . .

(52%) with 01 Ogawa, and 10 (3%) with both serogroups. During the same time period the previous year, 319 culture-confirmed cholera cases occurred, all serogroup 01. Questionnaires were obtained from 105 patients with O139 Bengal and 103 with 01 infections; for each case patient, two asymptomatic age- and sex-matched control persons were selected. Of the patients with 0139 Bengal infections, 93% were adults (>15 years) compared with 92% of patients with 01 infections. Risk factors for cholera identified by case-control comparisons were similar for the two serogroups and included consumption of untreated water, uncooked seafood, and food served at group gatherings. V. cholerae 0139 Bengal has emerged in Thailand as a cause of endemic cholera, with epidemiologic features and incidence similar to those of the preexisting 01 strain. Am J Epidemiol 1996; 143:263-8.