An extremely heterogeneous group of micro-organisms, Escherichia coli encompasses almost all features of possible interactions between intestinal microflora and the host, ranging from a role of mere harmless presence to that of a highly pathogenic organism. In fact, the E. coli species is made up of many strains that profoundly differ from each other in terms of biological characteristics and virulence properties.29
Escherichia coli is a Gram-negative, lactose-fermenting motile bacillus of the family Enterobacteriaceae. Currently, 171 somatic (O) and 56 flagellar (H) antigens are recognized. Six distinct categories of E. coli are currently considered enteric pathogens (based on either outbreak data or volunteer studies) (Table 10.2). The diagnosis of diarrheagenic E. coli relies on isolation from stool and subsequent differentiation from commensal E. coli either by using genetic probes or by phenotypic assays. With the exception of E. coli O157:H7, assays for detection are not routinely available in clinical laboratories.
This was the first group of E. coli serotype shown to be pathogens for humans and has been responsible for devastating outbreaks of nosocomial neonatal diarrhea and infant diarrhea in virtually every corner of the globe. Strains of enteropatho-genic E. coli (EPEC) are distinguished from other E. coli strains by their ability to induce a characteristic attaching and effacing lesion in the small-intestinal enterocytes and by their inability to produce Shiga toxins. Between the 1940s and the 1960s, EPEC was associated with infant diarrhea in summertime and nursery outbreaks of diarrhea in the USA and other industrialized countries. Since then, it has become extremely uncommon in industrialized countries, although it is occasionally reported in child-care settings.30
Enteropathogenic E. coli (EPEC) Enterotoxigenic E. coli (ETEC) Enteroinvasive E. coli (EIEC) Enterohemorrhagic E. coli (EHEC) Diffusely adherent E. coli (DAEC) Enteroaggregative E. coli (EAggEC)
However, EPEC persists as an important cause of infantile diarrhea in many developing countries.31 In nursery outbreaks, transmission was thought to occur via the hands of caregivers and via fomites. In less developed countries, contaminated formula and weaning foods have been incriminated. Volunteer studies and epidemiological observations suggest that the infective dose for EPEC is high (approximately 109 colony-forming units (CFU)).32 EPEC causes a self-limited watery diarrhea with a short incubation period (6-48h). There may be associated fever, abdominal cramps and vomiting, and EPEC is a leading cause of persistent diarrhea (lasting 14 days or longer) in children in developing countries.33 Although few data exist to guide antibiotic therapy of EPEC diarrhea, administration of appropriate antibiotics seems to diminish morbidity and mortality. A 3-day course of oral, non-absorbable antibiotics such as colistin or gentamicin (if available) has been shown to be effective.34 Some clinicians also advocate the use of oral neomycin; however, this drug causes diarrhea in about 20% of people. In a placebo-controlled trial among Ethiopian infants with severe EPEC diarrhea, TMP-SMX and mecillinam resulted in significant clinical and bacteriological cure rates by the third day, as compared with placebo.35 Strategies for the prevention of EPEC infection include efforts to improve social and economic conditions in developing countries, efforts to encourage breast feeding and prevention of nosocomial infections.
Strains of enterotoxigenic E. coli (ETEC) are an important cause of diarrheal disease in humans and animals worldwide. The clinical importance of these micro-organsims was first outlined in the 1970s by epidemiological studies in India that identified them as a major cause of endemic diar-rhea.36 Their pathogenicity is related to the elaboration of one or more enterotoxins that are either heat stable (ST) or heat labile (LT) (see Pathogenesis section) without invading or damaging intestinal epithelial cells. Together with rotavirus, ETEC is the leading cause of dehydrating diarrheal disease among weaning infants in the developing world. These children experience 2-3 episodes of ETEC diarrhea in each of the first 2 years of life. This represents over 25% of all diarrheal illness37 and results in an estimated 700 000 deaths each year.38 In industrialized countries, ETEC does not contribute to endemic disease, but is notorious for being the leading agent of travelers' diarrhea, accounting for about half of all episodes.39 Transmission occurs by ingestion of contaminated food and water, with peaks during the warm, wet season. Like EPEC, ETEC requires a relatively high inoculum40 and a short incubation period (14-30h). The cardinal symptom is watery diarrhea, sometimes with associated fever, abdominal cramps and vomiting. In its most severe form, ETEC can cause choleralike purging, even in adults. The illness is typically self-limited, lasting for less than 5 days and with few cases persisting beyond 3 weeks. Infection with ETEC has also been associated with short-and long-term adverse nutritional consequences in infants and children.
Most diarrheal illnesses due to ETEC are self-limited and do not require specific antimicrobial therapy. Empirical therapy is reserved for those whose diarrhea is moderate to severe despite re-hydration and supportive measures. Antibiotic regimens that have been efficacious in clinical trials, shortening the duration of illness by 1-2 days, include doxycycline, TMP-SMX, cipro-floxacin, quinolones, and furazolidone.41 In the past, the drug of choice for children has been TMP-SMX; however, except in Central Mexico,42 a large proportion of ETEC is now resistant. An alternative regimen for children is furazolidone.
Prevention of ETEC infection is based on avoiding contaminated vehicles. Although antibiotics are effective as prophylactic agents, their use is not recommended. Some experts advocate the use of bismuth subsalicylate to diminish the risk of travelers' diarrhea.43 The development of vaccines against ETEC has received a great deal of attention because of its disease burden. Oral vaccines for ETEC are being developed by five different strategies, including killed whole cells, toxoids, purified fimbriae, living attenuated strains and live carrier strains elaborating ETEC antigens. A killed whole-cell Vibrio cholerae vaccine given with cholera toxin B (CTB) provided 67% protection against LT-producing ETEC diarrhea for 3 months.44 A formalin-inactivated whole-cell oral vaccine consisting of ETEC strains bearing colonization factor antigens (CFAs) in combination with CTB has entered field trial.45
This group consists of invasive E. coli strains that are genetically, biochemically and clinically nearly identical to Shigella. This section will serve only to highlight relevant characteristics that distinguish this pathogen. Strains of enteroinvasive E. coli (EIEC) are endemic in developing countries, where they exhibit similar epidemiology to Shigella and cause an estimated 1-5% of diarrheal episodes among patients visiting treatment centers.46 The occurrence of EIEC in industrialized countries is limited to rare food-borne outbreaks.47 From volunteer studies, it appears that the infectious inoculum contains more organisms than that required to cause shigellosis.48 Like Shigella, EIEC can produce dysentery, but watery diarrhea is more common.49 The rare episodes for which treatment is desired are treated with antibiotics recommended for shigellosis. The same general preventive measures used for Shigella infections apply to EIEC-associated diarrhea.
These E. coli strains produce either one or both phage-encoded potent cytotoxins termed Shiga-like toxin I (SLT I) (which is neutralized by antisera to Shiga toxin produced by S. dysenteriae type 1) or Shiga-like toxin II (SLT II) (which is not neutralized) and can cause diarrhea or HUS. E. coli O157:H7 is the prototypic (but not the exclusive) enterohemorrhagic E. coli (EHEC) serotype, since it is the predominant SLT-producing E. coli, the one most commonly associated with HUS in North
America and the type most readily identified in stool specimens.50 In 1982, a multistate outbreak of hemorrhagic colitis that was linked to the consumption of hamburgers at the same fast-food restaurant led to the identification of EHEC.51 The causative organism was E. coli O157:H7, a serotype not previously recognized as a human pathogen. Soon after, Canadian investigators uncovered an association between O157:H7 and other SLT-producing strains of E. coli and HUS.52 EHEC is now recognized as a global health problem; in 1996, an outbreak in Japan linked to eating radish sprouts affected over 6000 persons.53 One the most severe EHEC outbreaks in the USA took place in New York State in 1999, with more than 1000 ascertained cases, two HUS-related casualties, and eight children in dialysis because of renal failure. Most of the infected individuals attended a fair whose underground water supply was contaminated by cow manure from a nearby cattle barn. The predominant mode of transmission is ingestion of contaminated, undercooked ground beef. However, the spectrum of vehicles is widening to include raw fruits (including apple juice) and vegetables,54,55 raw milk,56 processed meats,57 and drinking58 or swimming59 in contaminated water. The uncooked food vehicles are usually contaminated with manure from infected animals during growth or processing. Person-to-person transmission is the mode of spread in day-care outbreaks, where secondary transmission rates of 22% have been reported.60
EHEC also causes sporadic diarrhea. Isolation from stools of unselected patients is low (< 1%), but isolation from stools of patients with bloody diarrhea may be as high as 20-30%.61 A USA national laboratory-based study demonstrated that infection was more frequent in northern states and that it peaked from June to September.61 The highest age-specific isolation rates are in patients 5-9 and 50-59 years of age. A population-based incidence rate based on stool samples submitted to a large health maintenance organization laboratory in the state of Washington was eight per 100 000 person-years.62 Illness with EHEC follows 3-9 days after ingestion of as few as 100 organisms.63 Crampy abdominal pain and non-bloody diarrhea are the first symptoms, sometimes associated with vomiting. By the 2nd or 3rd day of illness, diarrhea becomes bloody in about 90% of cases, and abdominal pain worsens.64 Bloody diarrhea lasts for 1-22 days (median 4 days). Unlike other infectious causes of bloody diarrhea, fever is usually absent or remains low-grade. Younger children appear to excrete the organisms for longer (median 3 weeks) than older children and adults.65
In outbreaks, approximately 25% of patients are hospitalized, 5-10% develop HUS and 1% die.66,67 Intestinal complications include rectal prolapse, appendicitis, intussusception, and pseudomem-branous colitis.68,69 Extraintestinal complications are rare.
The most frightening complication of EHEC infection is HUS. It is usually diagnosed 2-14 days after the onset of diarrhea.52 Risk factors include young and old age, bloody diarrhea, fever, an elevated leukocyte count and treatment with antimotility agents.68,70 Two-thirds of patients who develop HUS are no longer excreting the organism at presentation.71 The most widely accepted indication for seeking a clinical diagnosis of E. coli O157:H7 infection is a patient with bloody diarrhea, in whom an accurate diagnosis may avoid unnecessary medical procedures because a surgical abdomen (such as appendicitis or intussusception) is suspected. A multicenter study found that, when the presence of fecal blood was used as the sole criterion for culturing O157 strains, only 3% of stools would be cultured to detect 63% of infections.70,71 Diagnosis may also be helpful in patients with HUS or with any type of diarrhea in a patient in contact with HUS. E. coli O157:H7 is not detected by routine stool culture. A relatively inexpensive method exploits the inability of E. coli O157 rapidly to ferment sorbitol after 24 h of incubation on sorbitol-MacConkey agar, in contrast to about 90% of commensal E. coli. The 'sorbitol-negative' colonies can then be screened for the presence of the O157 antigen, using commercially available antisera. These strains should be considered pathogenic pending the determination of the H type in a reference laboratory. Although data are not available from prospective randomized double-blind trials, there is considerable evidence to suggest that patients who receive antibiotics to which the offending E. coli O157:H7 is sensitive have either the same or a poorer outcome when compared with untreated patients.68,72,73 Therefore, antibiotic therapy is not recommended for EHEC infection. As mentioned above, antimotility agents have been identified as a risk factor for the development of HUS and should be avoided.
Prevention of E. coli 0157:H7 is a complex process. From a public health standpoint, control measures at the level of farms, slaughterhouses and processing plants can decrease the risk of colonization of cattle and contamination of beef. Since these procedures are unlikely to achieve complete success, regulations governing proper processing and cooking of contaminated foods are also required. Advice to consumers should include recommending complete avoidance of raw foods of animal origin. Hamburger should be cooked until no pink remains and until the juices are clear.
Because of the severity of disease, there has been a recent focus on vaccine development for EHEC infection. Efforts have concentrated on three approaches: parenteral toxoids and live oral carrier strains elaborating the B subunit of Shiga toxin;74 vaccines expressing the adhesin intimin, designed to prevent intestinal colonization;75 and a parenteral 0157 polysaccharide protein conjugate.76
Until recently, diffusely adhering E. coli (DAEC) was considered a non-pathogenic E. coli, since early studies failed to find an association between this micro-organism and diarrheal disease.77-79 However, more recent studies have demonstrated such an association, particularly in children older than 2 years of age. A community-based case-control study in southern Mexico revealed that DAEC was significantly associated with diarrhea in children less than 6 years of age.80 Prospective cohort studies in Chile81 and Bangladesh81 also demonstrated a diarrheagenic role for DAEC that peaked in the 48-60-month age group.81 This micro-organism was more frequently isolated from cases of prolonged diarrhea,82 and it showed a seasonal pattern similar to that of ETEC, occurring more frequently in the warm season.81 The gastrointestinal symptoms that characterize DAEC infection are practically indistinguishable from those caused by ETEC, with self-limiting watery diarrhea rarely associated with vomiting and abdominal pain. The diagnosis is mainly based on the DNA probe technique and on the pattern of adherence of the micro-organism to HE-2 cells. Given the technical challenge of both assays, their use is limited to epidemiological surveys rather than the diagnosis of single individuals.
Enteroaggregative E. coli (EAggEC) are diar-rheagenic E. coli defined by a characteristic aggregating pattern of adherence to HEp-2 cells and the intestinal mucosa. They have been particularly associated with cases of persistent diarrhea in the developing world. It has been hypothesized that the aggregating pattern of adherence may be a result of non-specific, possibly hydrophobic interaction, and therefore, not all organisms meeting the definition of EAggEC may be pathogenic in humans. Moreover, since epidemiological studies have not uniformly implicated EAggEC as pathogenic, some investigators have questioned the virulence of all EAggEC isolates. Volunteer studies performed to address both of these questions82 confirmed that at least some EAggEC strains are genuine human pathogens but that virulence is not uniform among isolates. More recently, EAggEC pathogenicity has also been proven in several outbreaks.
From the earliest epidemiological reports, EAggEC was most prominently associated with persistent cases of pediatric diarrhea (i.e. lasting > 14 days),84 a condition that represents a disproportionate share of diarrheal mortality. On the Indian subcontinent, several independent studies have demonstrated the importance of EAggEC in pedi-atric diarrhea.85 These studies include hospitalized patients with persistent diarrhea,78 outpatients visiting health clinics,85 and cases of sporadic diarrhea detected by household surveil-lance.77 In Fortaleza, Brazil, Fang et al have demonstrated a consistent association between EAggEC and persistent diarrhea;86 in this area, EAggEC accounts for more cases of persistent diarrhea than all other causes combined.86 EaggEC have been implicated as a cause of sporadic diarrhea in other developing countries (including Mexico, Chile, Bangladesh, Congo and Iran) as well as in industrialized countries such as Germany and England.87 Besides being responsible for sporadic cases of diarrhea, EAggEC has also been associated with outbreaks in India,88 Serbia,89 Japan90 and the UK.69
The clinical features of EAggEC diarrhea are becoming increasingly well defined in outbreaks, in sporadic cases and in the volunteer model. Typically, illness is manifested by a watery, mucoid, secretory diarrheal illness with low-grade fever and little or no vomiting.77,91 However, in epidemiological studies, grossly bloody stools have been reported in up to one-third of patients with EAggEC diarrhea.92 This phenomenon may well be strain-dependent. In volunteers infected with EAggEC strain 042, diarrhea was mucoid, of low volume, and notably, without occult blood or fecal leukocytes; all patients remained afebrile. In such volunteers, the incubation period of the illness ranged from 8 to 18 h.82
Perhaps even more significant than the association of EAggEC with diarrhea are the recent data from Brazil that link EAggEC with growth retardation in infants.92 In this study, the isolation of EAggEC from the stools of infants was associated with a low z-score for height and/or weight, irrespective of the presence of diarrheal symptoms. Given the high prevalence of asymptomatic EAggEC excretion in many areas,84,93 such an observation may imply that the contribution of EAggEC to the human disease burden is significantly greater than is currently appreciated. Colonization of EAggEC is detected by the isolation of E. coli from the stools of patients and the demonstration of the aggregative pattern in the HEp-2 assay. Implication of EAggEC as the cause of the patient's disease must be cautious, given the high rate of asymptomatic colonization in many populations.84,93 If no other organism is implicated in the patient's illness and EAggEC is isolated repeatedly, then EAggEC should be considered a potential cause of the patient's illness. A DNA-fragment probe has proven highly specific in the detection of EAggEC strains. A polymerase chain reaction (PCR) assay using primers derived from the aggregative probe sequence shows similar sensitivity and speci-ficity.94
The optimal management of EAggEC infection has not been studied. Acute diarrhea is apparently self-limiting; however, more persistent cases may benefit from antibiotic and/or nutritional therapy. Given the high rate of antibiotic resistance among
EAggEC,95 susceptibility testing is recommended when available.
Was this article helpful?