ACP medicine, 3rd Edition

Infectious Disease

Infections Due to Escherichia Coli and Other Enteric Gram-Negative Bacilli

Michael S. Donnenberg MD1

1Professor of Medicine, Professor of Microbiology and Immunology, and Head, Division of Infectious Diseases, University of Maryland School of Medicine

The author has no commercial relationships with manufacturers of products or providers of services discussed in this chapter.

December 2005

This chapter describes infections caused by Escherichia coli and related members of the family Enterobacteriaceae, excluding genera that principally cause enteric infections. Infections caused by Salmonella, Shigella, and Yersinia are described elsewhere [see 7:IX Infections Due to the Enteric Pathogens Campylobacter, Salmonella, Shigella, Yersinia, Vibrio, and Helicobacter]. The family Enterobacteriaceae comprises facultative anaerobic gram-negative bacilli that ferment sugars. These organisms are often motile because of the presence of peritrichous flagella (i.e., flagella that are distributed around the entire cell). Many Enterobacteriaceae species reside principally in the gastrointestinal tract of vertebrates, though some are found primarily in the environment. They are oxidase negative and catalase positive and are capable of reducing nitrate to nitrite. They cause a variety of diseases, including diarrhea, urinary tract infections (UTIs), and nosocomial infections.

Escherichia coli Infections

  1. coli, the most common facultative anaerobe in the human intestine, is distinguished from other members of the family Enterobacteriaceae primarily on the basis of E. coli'sability to ferment lactose and to produce indole and its inability to hydrolyze urea. Several virulence factors are shared by most members of the species; these include the ability to produce a highly reactive lipopolysaccharide in the cell wall, the ability to produce type 1 mannose-binding fimbriae, and, in many strains, the ability to produce an antiphagocytic capsule and to sequester iron.
  2. coliand Salmonelladiverged from a common ancestor about 100 million years ago.1 Ample time for diversifying selection and a prodigious capacity for genetic exchange have fostered a tremendous degree of genetic diversity in E. coli strains. Thus, it is not surprising that this species has the ability to cause a diverse array of infectious diseases. In fact, the organisms that are commonly referred to asShigella are actually members of the species E. coli,2 but because of historical and clinical considerations, they are often discussed separately [see 7:IX Infections Due to the Enteric Pathogens Campylobacter, Salmonella, Shigella, Yersinia, Vibrio, and Helicobacter]. The ability of E. coli to cause a variety of clinical syndromes by a plethora of mechanisms is entirely dependent upon unique virulence attributes that are encoded by distinct sets of virulence genes. Each group of E. coli that causes a particular clinical syndrome by a recognized pathogenic mechanism may be referred to as a pathotype [see Table 1]. With the advent of genomic sequencing, it has become evident that different strains of E. coli, which are so alike in most characteristics that they could easily be regarded as identical in clinical microbiology laboratories, are remarkably different in genetic content. For example, a strain isolated from a woman with pyelonephritis, a strain isolated from a child with hemorrhagic colitis, and a laboratory strain originally isolated from an asymptomatic volunteer share only 39% of their genes; in contrast, 47% of the total number of genes in the three strains is unique to one strain only.3,4 It is not known what roles the overwhelming majority of these genes play in the biology of the organism or in the pathogenesis of infections.

Table 1 Clinical, Epidemiologic, Pathogenetic, and Therapeutic Aspects of Infection with Various Pathotypes of E. coli


Clinical Features

Epidemiologic Features

Virulence Factors



Watery diarrhea

Childhood diarrhea in developing countries; traveler's diarrhea

Pili, heat-labile and heat-stable enterotoxins

Fluid replacement; fluoroquinolone or rifaximin can be used alone or in combination with loperamide


Watery diarrhea, vomiting

Infants in developing countries

Bundle-forming pilus, attaching and effacing effect

Fluid replacement


Watery diarrhea, hemorrhagic colitis, hemolytic-uremic syndrome

Food-borne and water-borne outbreaks in developed countries

Shiga toxins, attaching and effacing effect

Fluid replacement, supportive care; antibiotics and antimotility agents are contraindicated


Diarrhea with mucus

Childhood diarrhea, traveler's diarrhea

Pili, cytotoxins

Fluid replacement; antibiotic treatment for patients with AIDS?



Older children?


Fluid replacement


UTI, neonatal meningitis, nosocomial infections

Sexually active women, neonates, hospitalized patients

Fimbriae of types 1, P, and others; hemolysin; capsule; iron-acquisition systems

Antibiotics for symptomatic patients and in selected patients with asymptomatic UTI; antibiotic therapy guided by susceptibility testing for other infections

UTI—urinary tract infection

Because each pathotype of E. coli produces disease that has more or less distinct epidemiologic, pathogenetic, and clinical features, the discussion of specific pathotypes is organized on the basis of whether the infections caused by those pathotypes occur within or outside of the GI tract.


Six pathotypes of E. coli cause diarrhea [see Table 1].5 Of these, the epidemiologic, pathogenetic, and clinical features of enteroinvasive E. coli are the same as those of Shigella, and they are described elsewhere [see 7:IX Infections Due to the Enteric Pathogens Campylobacter, Salmonella, Shigella, Yersinia, Vibrio, and Helicobacter].

Diarrhea Due to Enterotoxigenic E. coli


Enterotoxigenic E. coli (ETEC) is a common cause of watery diarrhea in children in developing nations and in people of all ages who visit these countries. ETEC infections are spread through ingestion of contaminated food and water in regions where sanitation is inadequate. A relatively large inoculum is required to produce illness; therefore, person-to-person transmission is not significant for this organism.


ETEC produces either a heat-labile toxin (LT), a heat-stable toxin (ST), or both. LT has the same three-dimensional structure, receptor, and mechanism of action as cholera toxin [see 7:IX Infections Due to the Enteric Pathogens Campylobacter, Salmonella, Shigella, Yersinia, Vibrio, and Helicobacter]. Like cholera toxin, LT toxin induces a chain of events that leads to an elevation in the intracellular concentration of cyclic adenosine monophosphate (cAMP) and the opening of the cystic fibrosis transmembrane conductance regulator [see Figure 1]. The efflux of chloride ions into the intestinal lumen is accompanied by sodium ions and water, and diarrhea ensues. ST is a small disulfide-rich peptide that resembles the endogenous hormone guanylin. Like its homologue, ST causes increases in the levels of cyclic guanosine monophosphate (cGMP), which also lead to electrolyte efflux and fluid efflux through the cystic fibrosis transmembrane conductance regulator. For these toxins to reach their targets in sufficient quantity and for a sufficient duration to induce symptoms, the organisms must adhere to the intestinal mucosa.6 This binding is accomplished through an antigenically and morphologically diverse array of pilus and nonpilus adhesins known as colonization factor antigens.5


Figure 1. The pathogenesis of enterotoxigenic Escherichia coli (ETEC) infections. ETEC organisms adhere to intestinal epithelial cells by one of several adhesins known as colonization factor antigens. No damage is caused to the epithelial cells. ETEC produces either a heat-labile toxin (LT), a heat-stable toxin (ST), or both. LT is composed of a pentamer of receptor-binding B subunits that are noncovalently attached to a single enzymatically active A subunit. After binding its glycolipid receptor and entering the enterocyte, LT is transported retrograde in membrane-bound vesicles to the Golgi apparatus. The A subunit reaches its target in the basolateral cell membrane, which is the alpha subunit of the stimulatory heterotrimeric G protein receptor (GA), and catalyzes the adenosine diphosphate (ADP) ribosylation of the target. The modified alpha subunit is locked in its guanosine triphosphate (GTP)-bound active form, which forces it to constitutively activate adenylyl cyclase. The elevated levels of cyclic adenosine monophosphate (cAMP) that result lead to activation of protein kinase A (PKA) and to phosphorylation and opening of the cystic fibrosis transmembrane conductance regulator (CFTR). ST activates a distinct but convergent pathway. A disulfide-rich peptide, ST resembles the endogenous peptide hormone guanylin. Like its homologue, ST binds to and activates membrane-bound guanylyl cyclase-C, causing elevations in the levels of cyclic guanosine monophosphate (cGMP). As is the case with LT, this leads to electrolyte and fluid efflux through the CFTR.5,29 (Cl-—chloride; GDP—guanosine diphosphate; GTP—guanosine triphosphate; NAD—nicotinamide-adenine dinucleotide).


After ingestion of food or water that is contaminated with large numbers of ETEC organisms, there is an incubation period of 0 to 2 days before the onset of symptoms. The disease begins abruptly with the onset of watery diarrhea without blood or mucus. Nausea, vomiting, abdominal cramps, and fever are rarely prominent. The disease is self-limited, lasting 3 to 5 days in travelers.5 The endemic disease is sometimes more severe, occasionally inducing cholera-like purging in children residing in developing countries. Immunity appears to be strain specific and may reflect protective responses against particular colonization-factor antigens and toxins. The diagnosis of ETEC infection is often suspected on the basis of epidemiology and the clinical presentation but is rarely confirmed. Microbiologic diagnosis requires the use of bioassays to identify LT and ST toxins or the use of DNA probing techniques or polymerase chain reaction to detect the genes that encode these toxins; these tests are performed on E. coli organisms recovered from the stool.


As with all diarrheal disease, the initial management of ETEC infection involves ensuring adequate fluid repletion [see 4:III Diarrheal Diseases]. Several well-conducted clinical trials have demonstrated that traveler's diarrhea that is partly caused by ETEC responds rapidly to treatment with any of several regimens containing antibiotics, with or without antimotility agents.7,8 Ciprofloxacin, taken every 12 hours for 3 days in combination with loperamide, is a particularly effective regimen.9 Rifaximin, a nonabsorbable agent, is approved by the Food and Drug Administration for traveler's diarrhea caused by noninvasive strains of E. coli. Studies have found rifaximin to be comparable in efficacy to ciprofloxacin.10 Azithromycin is another effective alternative.11 Rifaximin can be used as a single agent or in combination with loperamide. ETEC organisms are generally susceptible to fluoroquinolones; however, should resistance to fluoroquinolones increase, rifaximin may prove to be an important alternative to ciprofloxacin for the treatment of traveler's diarrhea. For adults traveling to countries where ETEC is endemic, these medications can be provided before they travel, for prompt use if symptoms occur. Travelers should also be counseled to seek medical care if diarrhea is accompanied by blood or fever or if it persists despite treatment. Travelers can reduce the risk of acquiring ETEC by fastidiously avoiding all food that is not served steaming hot, fruit that they have not peeled, and beverages that are not bottled [see CE:VII Health Advice for International Travelers]. Bismuth subsalicylate tablets, taken four times a day, also provide some protection, but difficulty with compliance limits the use of this therapy.7 The prospects for developing an ETEC vaccine are limited by the antigenic heterogeneity of the organisms.

Diarrhea Due to Enteropathogenic E. coli

Enteropathogenic E. coli (EPEC) strains are characterized by the ability of these organisms to bind intimately to the apical surface of enterocytes, where they destroy microvilli and induce the formation of cellular pedestals, which then embrace the bacteria [see Figure 2]. This histopathologic appearance is known as the attaching and effacing effect.12 EPEC strains are distinguished from enterohemorrhagic E. coli (EHEC) strains by the inability of the former to produce Shiga toxin (see below). Although rare outbreaks involving contaminated food or water may strike individuals of any age,13,14,15 most EPEC infections occur in infants in developing countries.16 However, atypical EPEC strains lacking the bundle-forming pilus found in typical strains have recently been identified as a cause of diarrhea in children in developed countries as well.17,18


Figure 2. The pathogenesis of enteropathogenic E. coli (EPEC) infections. Typical EPEC strains produce a plasmid-encoded bundle-forming pilus that is required for reversible aggregation of the bacteria and for full virulence.63 The bacteria are able to inject into host cells a number of proteins, including Tir and EspF, through a specialized chromosome-encoded type III secretion system. Tir forms a dimer in the host cell membrane, where it serves as a receptor for the bacterial outer membrane protein adhesin intimin, a protein of multiple domains that is also essential for virulence. Binding of intimin triggers Tir to activate the cellular machinery for actin polymerization, leading to the formation of the characteristic attaching and effacing pedestals to which the organisms adhere. EspF causes loss of intestinal barrier function through disruption of tight junctions and induces host cell death through apoptosis.12

Diarrhea Due to Enterohemorrhagic and Other Shiga Toxin-Producing E. coli


The Shiga toxin-producing E. coli strains constitute a heterogeneous group of organisms, of which EHEC is a subset. Like EPEC organisms, EHEC organisms induce the attaching and effacing effect on epithelial cells. Of the many Shiga toxin-producing E. coli strains that have been described in the literature, EHEC of serotype O157:H7 is the most important, having caused both the largest number of outbreaks and the outbreaks involving the greatest number of patients.19,20,21 There are approximately 0.9 cases of EHEC O157:H7 infections per 100,000 persons annually in the United States.22 The reservoir for EHEC is infected cattle, but the organism can be found in a variety of other ruminants. The disease often appears in outbreaks associated with the consumption of contaminated food. Undercooked ground beef has been associated with many outbreaks and is the leading risk factor for sporadic cases, but a variety of other foods and beverages, including other beef products, lettuce, sprouts, fruit, fruit juices, and milk, have been implicated. The disease has a low inoculum and can spread from person to person, particularly in day care centers or within the families of young children. An outbreak of EHEC infection resulting from airborne dispersal of bacteria after a country fair emphasizes the highly infectious nature of this pathogen.23 It can also be spread through contamination of the water supply or through contamination of swimming pools, lakes, or water parks, and it can be contracted directly from infected animals at farms and petting zoos.19,24,25,26


Animal models have demonstrated that the diarrhea caused by EHEC infection depends on the attaching and effacing effect.27 However, the severe systemic complications of EHEC infection are caused by the expression of Shiga toxins and are independent of the attaching and effacing effect. Shiga toxins are encoded by bacteriophages related to the classic lambda phage; these bacteriophages stably infect the bacteria.28 The three-dimensional structures of these toxins resemble those of LT toxin and cholera toxin, but the receptors, enzymatic activities, and targets are entirely different. Shiga toxins catalyze the depurination of ribosomal RNA, leading to a cessation of protein synthesis and to cell death [see Figure 3].29 Microvascular endothelial cells, such as those of the kidney, are particularly sensitive to the toxin.


Figure 3. The pathogenesis of enterohemorrhagic E. coli (EHEC) infections. Like EPEC, EHEC strains express intimin and Tir and employ the type III secretion system, which cause the attaching and effacing effect on cells. However, EHEC also produces Shiga toxins, which are encoded by bacteriophages that stably infect the bacteria. Shiga toxins are similar in quaternary structure to the heat-labile toxin of ETEC; they are composed of a pentamer of receptor-binding B subunits that are noncovalently bound to a single enzymatic A subunit. Shiga toxins are able to cross intestinal epithelial monolayers by an unknown mechanism, after which they presumably are spread systemically through the bloodstream. Endothelial cells appear to be particularly important target cells for Shiga toxins. After binding its glycolipid receptor and entering the host cell, the toxin is transported retrograde in membrane-bound vesicles to the endoplasmic reticulum. The A subunit induces depuration of a specific adenine residue in ribosomal RNA; this leads to cessation of protein synthesis and death of the cell. The altered surface of the intoxicated endothelial cell serves as a nidus for activation of the coagulation cascade, which leads to the formation of microthrombi and causes distal ischemic necrosis, platelet consumption, and red cell fragmentation—the hallmarks of the hemolytic-uremic syndrome.12


The classic presentation of EHEC infection begins with watery diarrhea and severe cramps and progresses to bloody diarrhea. Fever is low grade or absent. However, symptomatic disease may range from heme-negative watery diarrhea to frank hematochesia. The severe cramps, the absence of fever, or the presence of blood can lead clinicians to confuse EHEC infections with a variety of noninfectious illnesses, including intussusception, inflammatory bowel disease, and bowel ischemia; this in turn can lead to unnecessary diagnostic procedures or surgical intervention.30 The dreaded complication of EHEC infection is the hemolytic-uremic syndrome (HUS), which sometimes manifests itself as thrombotic thrombocytopenic purpura. HUS ensues in approximately 5% to 10% of those patients with EHEC O157:H7 infection several days after the onset of diarrhea. It carries a high risk of death or permanent renal impairment. Various studies have demonstrated that children, the elderly, those who have consumed antimotility agents, and those with high white cell counts have an increased risk of HUS.31,32 Antibiotic therapy may also be a risk factor for HUS. Studies on animal models have shown that many antibiotics induce the production of Shiga toxins33; clinical studies have reported an association between antibiotic therapy and an increased risk of HUS,34although some studies have found no such association.35 It is possible that some antibiotics increase the risk of HUS, whereas others decrease the risk. However, prospective, randomized studies will be required to test this hypothesis.

HUS is characterized by microangiopathic hemolytic anemia, which results from Shiga toxin-induced damage to endothelial cells, leading to activation of coagulation in the microvasculature. The kidney is particularly susceptible, but ischemic necrosis of the bowel, brain, eye, or virtually any organ can occur.

The diagnosis of EHEC infection is extremely important, both for the patient (to prevent unnecessary interventions and provide appropriate care) and for public health, because any patient with EHEC could represent the index case of a large outbreak—the detection and interruption of which depend critically upon timely diagnosis and reporting. Fortunately, unlike most strains of E. coli, O157:H7 ferments sorbitol slowly, enabling the identification of these organisms on specific indicator plates. However, not all microbiology laboratories offer this test, and many of those that perform it do so only on request. Therefore, the recognition of sporadic cases and outbreaks of EHEC infection depends largely on the acumen of the astute clinician, who must suspect the diagnosis and request the test.

The diagnosis of E. coli infection caused by Shiga toxin-producing organisms other than O157:H7 is more difficult, because these organisms do not have a distinguishing phenotype that can be assayed in most laboratories. Colonies cultured from the stool of patients with suspected non-O157:H7 Shiga toxin-producing E. coli infection should be sent to referral laboratories, state health laboratories, or the Centers for Disease Control and Prevention for assays to detect Shiga toxins or the genes that encode them.


The treatment of EHEC infection is supportive and includes oral rehydration, observation, and, if necessary, hospital admission for the management of complications. Until proved safe and effective, antibiotics should be regarded as contraindicated in patients with known or suspected EHEC infection. Antimotility agents are also contraindicated in these patients, because their use is associated with a possible risk of HUS31 and with a theoretical risk of increasing the duration of toxin exposure.

The risk of EHEC infection can be reduced by routine adherence to good hygienic practice in food preparation. Ground beef should always be cooked to an internal temperature of 68.3° C (well done or until the juices run clear).21 A broad array of efforts to reduce EHEC infections is under development. These include measures to reduce colonization in cattle; vaccine development; and improved food-safety measures.

Diarrhea Due to Enteroaggregative E. coli and Diffuse-Adhering E. coli

Enteroaggregative E. coli (EAEC) and diffuse-adhering E. coli (DAEC) are distinguished by characteristic patterns of adherence to tissue culture cells on in vitro assays. It is likely that these pathotypes contain a heterogeneous mixture of strains that do not entirely share pathogenetic or clinical features.5,36,37 Our understanding of the epidemiology, pathogenesis, and management of these infections is rudimentary, but several interesting features have been defined.38


EAEC organisms are associated with acute and chronic diarrhea in children in developing and developed countries17,39,40; they have also been isolated from patients with AIDS.41,42 In addition, recent studies have found that EAEC is a frequent cause of travelers's diarrhea, rivaling the incidence of ETEC.43 Interestingly, carriage of EAEC, even in the absence of overt symptoms, is associated with evidence of growth retardation in children.44 DAEC has been associated with diarrhea in older children in developing countries.45,46


EAEC strains produce a variety of pili and toxins, but further study is needed to clarify the role that these factors play in causing disease. The pathogenetic mechanisms of DAEC remain obscure.


Diarrhea caused by EAEC may contain mucus or blood and may be accompanied by cramps. In some cases, particularly in patients with AIDS, the diarrhea can be protracted. The clinical features of DAEC infection have not been well described. The diagnosis of these infections requires tissue culture assays that are performed only in the research setting.


Case reports and studies in small series of patients have suggested that AIDS patients who are infected with EAEC may respond to antibiotic treatment.47


  1. colican infect a variety of extraintestinal sites, including the urinary tract, the meninges (in neonates), the lungs, the peritoneum, the gallbladder, and the biliary tree; E. coliinfections can also occur in association with intravascular and prosthetic devices [see Table 1]. E. coli can also be involved in polymicrobial infections such as intra-abdominal abscesses and skin and soft tissue infections. Bacteremia can complicate extraintestinal E. coli infections. Although the virulence factors responsible for all of these diseases have not been well studied, it is clear that the strains that cause extraintestinal infections are not a random sample of E. coli from the host intestine. On the other hand, these strains lack most of the virulence factors associated with those that cause diarrhea; rather, the strains that cause extraintestinal infections are more likely to possess the genes for hemolysin to produce an antiphagocytic capsule and a variety of fimbriae, and to sequester iron than are strains isolated from the GI tract. The similarity in the repertoire of putative virulence factors in strains isolated from UTIs and other extraintestinal infections has led to the conclusion that extraintestinal pathogenic E. coli (ExPEC) should be considered a single pathotype.48
  2. coliis by far the leading cause of UTIs in otherwise healthy persons, and it is an important cause of UTIs in those patients whose urinary tracts are compromised by anatomic abnormalities, foreign bodies, or host immune defects. The bacteria that cause UTIs are more likely than fecal E. coliisolates to possess the genes for a variety of factors under study as potential virulence determinants. Of these factors, P fimbriae and hemolysin are prominent.49 Interestingly, however, the ubiquitous type 1 fimbriae that are produced by virtually all strains ofE. coli and related organisms play a critical role in the pathogenesis of E. coli-associated UTI. However, there is much more to the pathogenesis of UTI than the production of type 1 fimbriae, because the presence of these organelles does not distinguish uropathogenic E. coli from other strains. The genome of a prototype strain of extraintestinal pathogenic E. coli confirms that these strains possess many genes that are absent from E. coli K-12.4 More detailed information on the pathogenesis, clinical features, and treatment of UTI is presented elsewhere [see 7:XXIII Infections of the Urinary Tract].

The diagnosis of these infections is confirmed by isolating the organism in culture from appropriate clinical specimens. Treatment is guided by susceptibility testing. The rising prevalence of resistance to trimethoprim-sulfamethoxazole in E. coli isolates from the community and the alarming rise in the prevalence of strains that produce extended-spectrum β-lactamases are narrowing therapeutic options.50,51,52

Proteus Infections

Named after a character in Homer's Odyssey who could change shape, organisms belonging to the genus Proteus are noted for their ability to take two forms: (1) typical bacillary swimmer cells, which express a variety of surface fimbriae as well as flagella, and (2) highly elongated swarm cells, which express hundreds of flagella and few other surface structures. Swarm cells provide a challenge to the clinical microbiologist because the organisms frequently swarm over culture plates, which interferes with the isolation of individual colonies. By far, the most commonly isolated Proteus species is P. mirabilis.


Proteus rapidly hydrolyzes urea to form carbon dioxide and ammonium hydroxide. In the urinary tract, this reaction results in an increase in pH. This reaction also alters the solubility of polyvalent ions and leads to the formation of struvite calculi. These stones can then obstruct urinary catheters, leading to bacterial persistence and urosepsis. In a murine model of ascending UTI, a urease-negative mutant of P. mirabilis was found to be severely limited in its ability to colonize the urinary tract, and this mutant strain failed to form stones.53 Studies in animal models have confirmed the ability of P. mirabilis to produce several fimbriae that play a role in infection.54


Although P. mirabilis occasionally causes UTI in otherwise healthy persons, it is more commonly isolated from patients with abnormalities of the urinary tract, particularly those with indwelling urethral catheters and those with stones. Notably, Proteus has a predilection for causing upper urinary tract infections.55 Whereas P. mirabilis rarely causes infections outside the urinary tract, other members of the genus and closely related organisms such as Morganella morganii may be isolated from skin infections and soft tissue infections, especially skin ulcers in patients with diabetes mellitus.


The diagnosis of a urinary tract infection caused by Proteus organisms can be suspected in patients with compatible symptoms and signs who have complicated urinary tracts, including those patients with indwelling urinary catheters and functional or anatomic abnormalities. Evidence of upper urinary tract involvement (e.g., fever, flank pain, and sepsis), catheter obstruction, or known renal calculi strongly supports the diagnosis, which can be confirmed by culture.


Most strains of P. mirabilis are susceptible to many antibiotics, but some Proteus species are highly resistant. In the treatment of UTIs caused by Proteus organisms, the underlying urinary abnormality and the clinical appearance of the patient must be taken into account. Treatment of asymptomatic bacteriuria in patients with urinary calculi or anatomic or functional abnormalities provides no long-term benefit to the patient and merely increases the potential for antimicrobial resistance [see 7:XXIII Infections of the Urinary Tract].

Klebsiella Infections

Klebsiella pneumoniae and, to a lesser extent, K. oxytoca are responsible for the vast majority of human Klebsiella infections. These organisms are found in the environment and in the GI tract.


Although Klebsiella species occasionally cause UTI in otherwise healthy persons, the majority of Klebsiella infections occur in compromised hosts. K. pneumoniae can cause community-acquired pneumonia, classically in patients with alcoholism, with dramatic findings such as currant jelly sputum and a bulging fissure sign (Friedländer pneumonia). In practice, pulmonary infections caused by Klebsiella are difficult to distinguish on clinical grounds from those caused by other organisms.56 Most Klebsiella infections occur in hospitals, where these bacteria cause UTI, pneumonia, wound infections, biliary infections, and primary bacteremia.57 Klebsiella are often second only to E. coli as a cause of nosocomial gram-negative bacteremia.57,58 These infections can occur as epidemics or sporadically.59 Particularly problematic are outbreaks of nosocomial infections caused by strains of Klebsiella that produce extended-spectrum β-lactamases.51,52 These organisms are frequently resistant to multiple antibiotics, owing to the presence of plasmids that encode multiple antibi-otic-resistance genes.


The primary virulence factor for Klebsiella is a luxuriant polysaccharide capsule, which has antiphagocytic properties. Klebsiella organisms also produce type 1 fimbriae; like ExPEC organisms, they also produce other fimbriae and have iron-acquisition systems and a reactive lipopolysaccharide.60


The clinical features of community-acquired pneumonia caused by K. pneumoniae are not sufficiently distinctive to differentiate patients with infections caused by this organism from those who have infections caused by other agents. Similarly, nosocomial infections and UTIs caused by Klebsiella cannot be diagnosed on the basis of clinical features.

In the microbiology laboratory, a presumptive identification of Klebsiella infection is often made when a highly mucoid, lactose-fermenting, ampicillin-resistant, gram-negative bacillus is isolated and is confirmed by further testing.


Treatment of Klebsiella infections requires consultation with a clinical microbiology laboratory or specialists in infectious diseases to ensure selection of an antibiotic to which the organism is susceptible. Klebisella isolates that do not produce extended-spectrum β-lactamases are usually resistant to ampicillin but may be sensitive to cephalosporins, trimethoprim-sulfamethoxazole, aminoglycosides, combinations of penicillins and β-lactamase inhibitors, and fluoroquinolones, any of which may be used to treat infections caused by susceptible organisms.

Enterobacter and Serratia Infections

Epidemiology and Etiology

Enterobacter and Serratia are closely related to Klebsiella; like Klebsiella, they are principally opportunistic pathogens that cause a variety of nosocomial infections. In addition, S. marcescens has been isolated from the blood or other clinical specimens of patients who use injection drugs.61 Although several species of Enterobacter—including E. cloacae and E. aerogenes, as well as the closely related speciesPantoea agglomerans—cause infections, S. marcescens is the only Serratia species that has been isolated from patients with appreciable frequency.


Enterobacter and Serratia organisms share with Klebsiella organisms the ability to produce antiphagocytic capsules. One important clinical difference is that Enterobacter species have an inducible chromosomal cephalosporinase. This enzyme renders the organisms resistant not only to first-generation cephalosporins but also to many second- and third-generation cephalosporins after high-level induction of cephalosporinase in the presence of these antibiotics. Such resistance may not be detected in the microbiology laboratory on initial testing and may become apparent only when the organism is again isolated from patients who fail to respond to therapy.


As with infections caused by Klebsiella, infections caused by Enterobacter and Serratia do not have distinguishing clinical features and are diagnosed by culture of the organism from a normally sterile site or from a nonsterile site in a patient with associated signs or symptoms.


Second- and third-generation cephalosporins should be used cautiously in patients with serious Enterobacter infections, even if the isolate appears to be susceptible on initial testing.62 When the organism is susceptible to alternative agents such as extended-spectrum penicillins, carbapenems, or fluoroquinolones, their use may be advisable.


Figures 1 through 3 Seward Hung.


  1. Whittam TS: Genetic variation and evolutionary processes in natural populations of Escherichia coli. Escherichia coliand Salmonella. Cellular and Molecular Biology. Neidhardt FC, Ed. ASM Press, Washington, DC, 1996, p 2708
  2. Pupo GM, Karaolis DK, Lan R, et al: Evolutionary relationships among pathogenic and nonpathogenic Escherichia colistrains inferred from multilocus enzyme electrophoresis and mdh sequence studies. Infect Immun 65:2685, 1997
  3. Perna NT, Plunkett G 3rd, Burland V, et al: Genome sequence of enterohaemorrhagic Escherichia coliO157:H7. Nature 409:529, 2001
  4. Welch RA, Burland V, Plunkett C 3rd, et al: Extensive mosaic structure revealed by the complete genome sequence of uropathogenicEscherichia coli. Proc Natl Acad Sci USA 99:17020, 2002
  5. Nataro JP, Kaper JB: Diarrheagenic Escherichia coli. Clin Microbiol Rev 11:142, 1998
  6. Zafriri D, Oron Y, Eisenstein BI, et al: Growth advantage and enhanced toxicity of Escherichia coliadherent to tissue culture cells due to restricted diffusion of products secreted by the cells. J Clin Invest 79:1210, 1987
  7. DuPont HL, Ericsson CD: Prevention and treatment of traveler's diarrhea. N Engl J Med 328:1821, 1993
  8. Guerrant RL, Van Gilder T, Steiner TS, et al: Practice guidelines for the management of infectious diarrhea. Clin Infect Dis 32:331, 2001
  9. Petruccelli BP, Murphy GS, Sanchez JL, et al: Treatment of traveler's diarrhea with ciprofloxacin and loperamide. J Infect Dis 165:557, 1992
  10. DuPont HL, Jiang ZD, Ericsson CD, et al: Rifaximin versus ciprofloxacin for the treatment of traveler's diarrhea: a randomized, double-blind clinical trial. Clin Infect Dis 33:1807, 2001
  11. Adachi JA, Ericsson CD, Jiang ZD, et al: Azithromycin found to be comparable to levofloxacin for the treatment of travelers with acute diarrhea acquired in Mexico. Clin Infect Dis 37:1165, 2003
  12. Donnenberg MS, Whittam TS: Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J Clin Invest 107:539, 2001
  13. Schroeder SA, Caldwell JR, Vernon TM, et al: A waterborne outbreak of gastroenteritis in adults associated with Escherichia coli. Lancet 1:737, 1968
  14. Viljanen MK, Peltola T, Junnila SY, et al: Outbreak of diarrhoea due to Escherichia coliO111:B4 in schoolchildren and adults: association of Vi antigen-like reactivity. Lancet 336:831, 1990
  15. Hedberg CW, Savarino SJ, Besser JM, et al: An outbreak of foodborne illness caused by Escherichia coliO39:NM, an agent not fitting into the existing scheme for classifying diarrheogenic E. coli. J Infect Dis 176:1625, 1997
  16. Donnenberg MS: Enteropathogenic Escherichia coli. Infections of the Gastrointestinal Tract. Blaser MJ, Smith PD, Ravdin JI, et al, Eds. Raven Press, New York, 1995, p 709
  17. Robins-Browne RM, Bordun AM, Tauschek M, et al: Escherichia coliand community-acquired gastroenteritis, Melbourne, Australia. Emerg Infect Dis 10:1797, 2004
  18. Cohen MB, Nataro JP, Bernstein DI, et al: Prevalence of diarrheagenic Escherichia coliin acute childhood enteritis: a prospective controlled study. J Pediatr 146:54, 2005
  19. Tarr PI: Escherichia coliO157:H7: clinical, diagnostic, and epidemiological aspects of human infection. Clin Infect Dis 20:1, 1995
  20. Michino H, Araki K, Minami S, et al: Massive outbreak of Escherichia coliO157:H7 infection in schoolchildren in Sakai City, Japan, associated with consumption of white radish sprouts. Am J Epidemiol 150:787, 1999
  21. Bell BP, Goldoft M, Griffin PM, et al: A multistate outbreak of Escherichia coliO157:H7-associated bloody diarrhea and hemolytic uremic syndrome from hamburgers: the Washington experience. JAMA 272:1349, 1994
  22. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food—10 sites, United States, 2004. MMWR Morb Mortal Wkly Rep 54:352, 2005
  23. Varma JK, Greene KD, Reller ME, et al: An outbreak of Escherichia coliO157 infection following exposure to a contaminated building. JAMA 290:2709, 2003
  24. Outbreaks of Escherichia coliO157:H7 infections among children associated with farm visits—Pennsylvania and Washington, 2000. MMWR Morb Mortal Wkly Rep 50:293, 2001
  25. Hilborn ED, Mermin JH, Mshar PA, et al: A multistate outbreak of Escherichia coliO157:H7 infections associated with consumption of mesclun lettuce. Arch Intern Med 159:1758, 1999
  26. Boyce TG, Swerdlow DL, Griffin PM: Escherichia coliO157:H7 and the hemolytic-uremic syndrome. N Engl J Med 333:364, 1995
  27. Tzipori S, Gunzer F, Donnenberg MS, et al: The role of the eaeAgene in diarrhea and neurological complications in a gnotobiotic piglet model of enterohemorrhagic Escherichia coli infection. Infect Immun 63:3621, 1995
  28. O'Brien AD, Newland JW, Miller SF, et al: Shiga-like toxin-converting phages from Escherichia colistrains that cause hemorrhagic colitis or infantile diarrhea. Science 226:694, 1984
  29. Sears CL, Kaper JB: Enteric bacterial toxins: mechanisms of action and linkage to intestinal secretion. Microbiol Rev 60:167, 1996
  30. Griffin PM, Ostroff SM, Tauxe RV, et al: Illnesses associated with Escherichia coliO157:H7 infections: a broad clinical spectrum. Ann Intern Med 109:705, 1988
  31. Bell BP, Griffin PM, Lozano P, et al: Predictors of hemolytic uremic syndrome in children during a large outbreak of Escherichia coliO157:H7 infections. Pediatrics 100:E12, 1997
  32. Siegler RL, Pavia AT, Christofferson RD, et al: A 20-year population-based study of postdiarrheal hemolytic uremic syndrome in Utah. Pediatrics 94:35, 1994
  33. Zhang X, McDaniel AD, Wolf LE, et al: Quinolone antibiotics induce Shiga toxin-encoding bacteriophages, toxin production, and death in mice. J Infect Dis 181:664, 2000
  34. Wong CS, Jelacic S, Habeeb RL, et al: The risk of the hemolytic-uremic syndrome after antibiotic treatment of Escherichia coliO157:H7 infections. N Engl J Med 342:1930, 2000
  35. Safdar N, Said A, Gangnon RE, et al: Risk of hemolytic uremic syndrome after antibiotic treatment of Escherichia coliO157:H7 enteritis: a meta-analysis. JAMA 288:996, 2002
  36. Okeke IN, Lamikanra A, Czeczulin J, et al: Heterogeneous virulence of enteroaggregative Escherichia colistrains isolated from children in Southwest Nigeria. J Infect Dis 181:252, 2000
  37. Czeczulin JR, Whittam TS, Henderson IR, et al: Phylogenetic analysis of enteroaggregative and diffusely adherent Escherichia coli. Infect Immun 67:2692, 1999
  38. Nataro JP: Enteroaggregative Escherichia colipathogenesis. Curr Opin Gastroenterol 21:4, 2005
  39. Bhan MK, Raj P, Levine MM, et al: Enteroaggregative Escherichia coliassociated with persistent diarrhea in a cohort of rural children in India. J Infect Dis 159:1061, 1989
  40. Huppertz HI, Rutkowski S, Aleksic S, et al: Acute and chronic diarrhoea and abdominal colic associated with enteroaggregativeEscherichia coliin young children living in western Europe. Lancet 349:1660, 1997
  41. Polotsky Y, Nataro JP, Kotler D, et al: HEp-2 cell adherence patterns, serotyping, and DNA analysis of Escherichia coliisolates from eight patients with AIDS and chronic diarrhea. J Clin Microbiol 35:1952, 1997
  42. Wanke CA, Mayer H, Weber R, et al: Enteroaggregative Escherichia colias a potential cause of diarrheal disease in adults infected with human immunodeficiency virus. J Infect Dis 178:185, 1998
  43. Adachi JA, Jiang ZD, Mathewson JJ, et al: Enteroaggregative Escherichia colias a major etiologic agent in traveler's diarrhea in 3 regions of the world. Clin Infect Dis 32:1706, 2001
  44. Steiner TS, Lima AA, Nataro JP, et al: Enteroaggregative Escherichia coliproduce intestinal inflammation and growth impairment and cause interleukin-8 release from intestinal epithelial cells. J Infect Dis 177:88, 1998
  45. Giron JA, Jones T, Millan-Velasco F, et al: Diffuse-adhering Escherichia coli(DAEC) as a putative cause of diarrhea in Mayan children in Mexico. J Infect Dis 163:507, 1991
  46. Gunzburg ST, Chang BJ, Elliott SJ, et al: Diffuse and enteroaggregative patterns of adherence of enteric Escherichia coliisolated from aboriginal children from the Kimberley region of Western Australia. J Infect Dis 167:755, 1993
  47. Wanke CA, Gerrior J, Blais V, et al: Successful treatment of diarrheal disease associated with enteroaggregative Escherichia coliin adults infected with human immunodeficiency virus. J Infect Dis 178:1369, 1998
  48. Russo TA, Johnson JR: Proposal for a new inclusive designation for extraintestinal pathogenic isolates of Escherichia coli: ExPEC. J Infect Dis 181:1753, 2000
  49. Stamm WE, Hooton TM, Johnson JR, et al: Urinary tract infections: from pathogenesis to treatment. J Infect Dis 159:400, 1989
  50. Talan DA, Stamm WE, Hooton TM, et al: Comparison of ciprofloxacin (7 days) and trimethoprim-sulfamethoxazole (14 days) for acute uncomplicated pyelonephritis in women: a randomized trial. JAMA 283:1583, 2000
  51. Wiener J, Quinn JP, Bradford PA, et al: Multiple antibiotic-resistant Klebsiellaand Escherichia coli in nursing homes. JAMA 281:517, 1999
  52. Lautenbach E, Patel JB, Bilker WB, et al: Extended-spectrum beta-lactamase-producing Escherichia coliand Klebsiella pneumoniae: risk factors for infection and impact of resistance on outcomes. Clin Infect Dis 32:1162, 2001
  53. Johnson DE, Russell RG, Lockatell CV, et al: Contribution of Proteus mirabilisurease to persistence, urolithiasis, and acute pyelonephritis in a mouse model of ascending urinary tract infection. Infect Immun 61:2748, 1993
  54. Mobley HLT: Virulence of Proteus mirabilis. Urinary Tract Infections: Molecular Pathogenesis and Clinical Management. Mobley HLT, Warren JW, Eds. ASM Press, Washington, DC, 1996, p 245
  55. Fairley KF, Carson NE, Gutch RC, et al: Site of infection in acute urinary-tract infection in general practice. Lancet 2:615, 1971
  56. Sahly H, Podschun R, Ullmann U: Klebsiellainfections in the immunocompromised host. Adv Exp Med Biol 479:237, 2000
  57. Carpenter JL: Klebsiellapulmonary infections: occurrence at one medical center and review. Rev Infect Dis 12:672, 1990
  58. Wisplinghoff H, Bischoff T, Tallent SM, et al: Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 39:309, 2004
  59. Geerdes HF, Ziegler D, Lode H, et al: Septicemia in 980 patients at a university hospital in Berlin: prospective studies during 4 selected years between 1979 and 1989. Clin Infect Dis 15:991, 1992
  60. Jarvis WR, Munn VP, Highsmith AK, et al: The epidemiology of nosocomial infections caused by Klebsiella pneumoniae. Infect Control 6:68, 1985
  61. Mills J, Drew D: Serratia marcescensendocarditis: a regional illness associated with intravenous drug abuse. Ann Intern Med 84:29, 1976
  62. Chow JW, Fine MJ, Shlaes DM, et al: Enterobacterbacteremia: clinical features and emergence of antibiotic resistance during therapy. Ann Intern Med 115:585, 1991
  63. Bieber D, Ramer SW, Wu CY, et al: Type IV pili, transient bacterial aggregates, and virulence of enteropathogenic Escherichia coli. Science 280:2114, 1998

Editors: Dale, David C.; Federman, Daniel D.