Infectious Enteritis and Proctocolitis

Host Defense Factors

Gastric acidity is a crucial first-line host defense that ingested pathogenic bacteria and other pathogens must survive to infect the small or large intestine. In general, bacterial pathogens are highly susceptible to low pH with a pH below 4.0 being rapidly bactericidal. In contrast, achlorhydria facilitates bacterial survival during gastric passage. Consistent with the importance of gastric acid as a host defense, treatment with PPIs, and to a lesser degree the shorter-acting and less potent H2RAs, is a risk factor for bacterial gastroenteritis, including Salmonella , Campylobacter , and C. difficile infection, and viral gastroenteritis, including norovirus.

The intestinal epithelium provides multiple components that contribute to host protection against potential enteric pathogens. Commensal microbiota as well as potential enteric pathogens first encounter the mucus layer that coats the epithelium. The outer loosely organized luminal mucus layer serves as a habitat for commensal microbiota in the colon, whereas the inner gel-like mucus layer largely excludes direct bacterial-epithelial cell contact. Bacteria that do penetrate this layer are thought to be cleared rapidly by the host mucosal immune system.

Multiple cell types within the intestinal epithelium—including enterocytes, Paneth cells, goblet cells, and M cells—all help protect the host against enterocolitis. Mechanisms of protection include barrier formation and water efflux via the tight junctions of enterocytes, enterocyte and Paneth cell secretion of antimicrobial molecules, goblet cell secretion of mucins, and M-cell presentation of antigens to the mucosal immune system, initiating, in many instances, a protective mucosal immune response. Specific molecules on intestinal epithelial cells also contribute to resistance to pathogens. Of particular importance are pattern recognition receptors, which mediate the recognition of microbes and lead to the activation of innate and adaptive immune responses in the intestinal mucosa. The inflammatory responses that follow can serve to protect the host or, conversely, contribute to disease development. A key host protective mechanism is the production of secretory immunoglobulin A, which can be both nonspecific (e.g., through microbial agglutination) or specific, as part of a pathogen-specific adaptive immune response. Members of the commensal microbiota, as well as potential enteric pathogens, can initiate specific mucosal immune responses. Studies of these mucosal immune mechanisms have provided abundant and evolving insights into the complexity of host immune responses. Lastly, the enteric nervous system (ENS), which is involved with normal intestinal motility (see Chapters 99 and 100 ), also contributes to protect the host. Impaired intestinal motility in the setting of enteric infections, as well as hormones and molecules produced by the host, are known to influence disease severity.

The luminal resident microbiota population in the colon comprises 10 ¹³ to 10 ¹⁴ bacteria per gram of stool and has increasingly been recognized as a powerful contributor to resistance to colonization and disease development with enteric pathogens. This is arguably best illustrated by the marked clinical responses of recurrent C. difficile colitis to intestinal microbiota transplant therapy (see Chapter 112 ). Finally, emerging data from human and murine models are beginning to uncover ways in which host genetics and diet, functioning by microbiota-dependent and -independent mechanisms, change host susceptibility to intestinal infections (e.g., Card9 , a key innate immunity gene and IBD-associated risk gene controls virulence of Citrobacter rodentium in mouse models; human genetic variation in FUT2 is associated with a 9- to 26-fold increase risk of infection with some strains of norovirus and rotavirus ).

Bacterial Factors

Bacterial pathogens have evolved various virulence factors and mechanisms that enable them to overcome host defenses, including adherence factors, enterotoxin and cytotoxin elaboration, and mucosal invasion among others. The ability of bacteria to adhere to host mucosal cells is critical to the initial interaction of each enteric pathogen with the intestinal epithelium. Numerous adhesins that differ in morphologic features and receptor specificities have been identified and vary in their capacity to mediate colonization in human compared with animal hosts. The complexity of adherence mechanisms is enhanced by the observation that particular bacteria express and use more than 1 adhesin, a redundancy that likely enhances bacterial virulence. Many bacterial adhesins recognize oligosaccharide residues of glycoproteins or glycolipids displayed in the mucus or surface of intestinal epithelial cells.

Enteropathogenic Escherichia coli (EPEC) cause disease, in part, because of tight intestinal epithelial adherence and serve as a classic model of the potential stages of adherence for enteric pathogens and their mechanistic complexity. EPEC initially exhibit nonintimate attachment to intestinal epithelial cells. This initial attachment is mediated by a bundle-forming pilus associated with a large plasmid common to EPEC isolates. Next, EPEC induce signal transduction events in the intestinal epithelial cells that lead to cytoskeletal changes in the enterocyte. Ultimately, intimate attachment of EPEC to the host cell membrane results, which is mediated by an EPEC outer membrane protein called intimin that is encoded by the eaeA gene cluster on the EPEC chromosome. This classic ultrastructural change, known as an attaching-effacing lesion , leads to elongation and destruction of microvilli, with classic pedestal formation ( Fig. 110.1 ). The role of the eaeA gene as a virulence factor that causes diarrhea in human EPEC infection has been confirmed in volunteer challenge studies.

Nonintimate attachment using other adhesin molecules is typical of other noninvasive bacteria such as ETEC, and molecular studies have demonstrated that nonintimate attachment through surface protein antigens known as pili or fimbriae (also referred to as adherence- or colonization-factor antigens ) is required for a bacterium to colonize and be fully pathogenic. These adhesins bind to specific receptor sites on the surface of the intestinal cell via specific ligand-receptor interactions. Classic studies by Moon and coworkers, in which loss or gain of fimbriae by genetic manipulation resulted in the loss or gain of the ability to adhere to and colonize the intestine, identified that these colonization factors (CFs) are important to the pathogenesis of E. coli diarrheal disease in animals. Adherence not only permits colonization but also can facilitate delivery of enterotoxin to the epithelium.

Production of toxins, usually proteins, by enteric pathogens is a key mechanism that contributes to the expression of diarrhea after infection. Nonetheless, asymptomatic colonization by enteric pathogens known to produce toxins and other virulence factors is common. Enteric toxins can be classified by their functional effect on the intestinal epithelium or by their precise molecular mechanism of action. From a functional perspective, there are 2 major groups of enteric bacterial toxins: enterotoxins and cytotoxins. Classic enterotoxins such as cholera toxin or the enterotoxins of ETEC induce intestinal secretion largely without altering the morphology of the intestinal epithelium. The predominant site of action of most enterotoxins is thought to be the small intestine. In contrast, cytotoxins such as the C. difficile toxins or Shiga toxins (Stx) of Shigella dysenteriae and E. coli (e.g., E. coli O157:H7 among others) often act in the colon, where morphologic changes in the intestinal epithelium result through a variety of mechanisms such as by inducing epithelial cell injury or death, altering the cytoskeleton of the epithelial cells, or inducing an inflammatory response that results in injury to the epithelium. In addition, the actions of several enteric bacterial toxins, such as the C. difficile toxins, are known to be complemented by their ability to alter ENS activity, also likely contributing to disease pathogenesis.

Although the biologic activity of many potential enteric bacterial toxins has been abundantly identified, precise molecular mechanisms of action have been identified for relatively few. Classic mechanisms of action include alteration of intestinal epithelial cell cyclic nucleotide levels (e.g., cholera toxin, E. coli LT [heat-labile] or STa [heat-stable] toxins); inhibition of protein synthesis (e.g., Stx); modifications of actin cytoskeleton (e.g., C. difficile toxins [see Chapter 112 ]); and pore formation (e.g., Clostridium perfringens enterotoxin [see Chapter 111 ]). Other mechanisms likely contributing to induction of disease by enteric bacterial toxins include alteration of epithelial cell calcium signaling and changes in arachidonic acid metabolism among others. Ultimately, proof of the role of a particular enteric bacterial toxin in human disease is via volunteer studies and is infrequent. Examples of enteric bacteria and their toxins studied in humans include Vibrio cholerae , Shigella spp., and pathotypes of E. coli among others.

Disease induction by some enteric pathogens requires their ability to invade and multiply within intestinal epithelial cells, with resultant cell injury and possibly cell death that incites a mucosal inflammatory response. Shigella species, Salmonella species, Campylobacter jejuni , Yersinia enterocolitica , and some (enteroinvasive) strains of E. coli are classic examples of bacteria for which invasion contributes to disease pathogenesis. Although the colon is most often the primary site of pathology with invasive enteric bacteria, non- dysenteriae Shigella spp. also produce enterotoxins that stimulate small bowel secretion and contribute to the first phase of watery diarrhea seen in shigellosis. Subsequent colon colonization and cellular invasion by non dysenteriae Shigella spp. then results in colon inflammation with colitic symptoms and, occasionally, bloody diarrhea.

Evaluation

The Infectious Diseases Society of America (2017) and the American College of Gastroenterology (2016) have both published guidelines in recent years for the evaluation and management of acute diarrheal infections. Only about 10% of patients who develop diarrheal disease present for medical consultation. The initial step in the diagnostic evaluation of a patient with acute diarrhea is a thorough history and physical examination, the goals of which are to identify patients who may be at risk of severe illness or susceptible to complications, and to identify risk factors for infection and those who will benefit from specific therapy. Most patients simply need rehydration therapy. Consideration of the patient’s general health, severity and duration of illness, and the setting in which the illness was acquired should enable the clinician to determine who needs further evaluation ( Fig. 110.2 ).

Patients who are debilitated, malnourished, or immunocompromised, and those who have severe comorbid illnesses, are at increased risk for complications of diarrhea and infection. The morbidity and mortality of infectious diarrheal diseases are highest in children younger than 5 years of age (particularly severe in those <2 years old ) and older adults. These high-risk patient groups may require hospitalization for diagnosis and treatment. Other patients who also require a more aggressive approach include those with systemic signs and evidence of inflammatory diarrhea, illnesses lasting greater than 3 to 5 days, a history or physical examination suggesting specific pathogens that will benefit from specific therapy ( Table 110.2 ), and infection with certain specific organisms (e.g., V. cholerae , Salmonella Typhi ). Bloody diarrhea can be considered a medical emergency and often requires hospitalization for diagnosis and management ; for example, Stx (Shiga toxin)-producing E. coli (STEC) infection can present with bloody diarrhea, and early, aggressive fluid management may diminish complications (e.g., renal failure) of this infection.

A traditional approach to help guide clinical diagnostic considerations has been the division of acute, presumably infectious, diarrheal disease into 2 broad clinical syndromes: a watery, noninflammatory diarrheal syndrome and an inflammatory diarrheal syndrome ( Table 110.3 ); a subgroup of the latter is the proctitis diarrheal syndrome. Classically, patients with noninflammatory diarrhea present with watery stools without visible blood or pus, and sometimes complain of severe abdominal pain. These patients generally have few systemic signs or symptoms, and fever often is absent; abdominal cramping, nausea, and vomiting can occur. This syndrome is most frequent in individuals infected with enterotoxigenic pathogens or viruses (see later). Many pathogens that cause inflammatory disease, however, can mimic this syndrome, particularly in the early phases of disease development (see later Shigella section).

Classically, patients with inflammatory diarrhea present with numerous small-volume stools that may be visibly mucoid, grossly bloody, or both. Such patients may appear toxic and are more often febrile. Abdominal cramping may be severe. Because of the small stool volumes, these patients are less likely to be dehydrated than those with noninflammatory diarrhea. Organisms causing inflammatory diarrheas may vary in their clinical presentations ( Table 110.4 ), are typically invasive, and usually affect the colon (see Invasive Pathogens, later). The acute inflammatory diarrheal syndrome also can have a noninfectious etiology, such as UC, Crohn disease, radiation or ischemic colitis, diverticulitis, and idiosyncratic reactions to medications. Fig. 110.3 A-C shows the endoscopic appearance of selected inflammatory diarrheas.

Proctitis syndrome is characterized by frequent painful bowel movements that contain blood, pus, and mucus. Tenesmus, often with rectal pain, usually is prominent. When acute, the most likely cause is one or more infectious agents (see Sexually Transmitted Infectious Proctitis, later). When a careful history reveals persistent, intermittent symptoms, noninfectious causes such as IBD (see Chapter 115 ) become diagnostic considerations.

Our growth in understanding the pathophysiology of infectious diarrheal diseases has led us to appreciate that inflammation is common to all infectious enterocolitides, with some infections (e.g., S. dysenteriae type 1) inducing dramatic colonic inflammation, and others (e.g., V. cholerae and ETEC ) inducing more modest small bowel or colon inflammation. Thus, the classic division of enteric pathogens into those inducing noninflammatory or inflammatory diarrhea is imperfect, and inflammation in response to enteric pathogens represents a continuum. Critical to the clinician, however, is identifying the patient at risk for morbidity or mortality due to infectious diarrheal disease. Most cases of watery diarrhea result in self-limited illnesses that require only advice on hydration maintenance and possibly diet. Evaluation of such patients who have mild diarrhea for less than 3 to 5 days is typically unnecessary. By contrast, patients who often require diagnostic evaluation are those with risk factors such as extremes of age, one or more immunocompromising conditions, marked stool frequency, bloody stools, or dehydration. Thus, the clinician’s assessment of the degree of illness, combined with the patient’s history, are key contributors that guide clinical judgment regarding diagnostic evaluations in diarrheal disease. Clinical or fecal evidence of acute inflammatory diarrhea or bloody diarrhea always deserves a diagnostic evaluation, and diagnostic evaluations, whenever possible, should precede initiation of empiric antibiotic therapy of diarrheal disease.

Risk Factors

As already discussed, age is a primary determinant of diarrheal disease morbidity, with those at the extremes of age (i.e., <5 years and >65 years), particularly those living in long-term care facilities, at increased risk for poorer outcomes. Gastroenteritis outbreaks are common in nursing homes in high-income countries, with notable pathogens including norovirus, rotavirus, STEC, and C. difficile among others. Enhanced risk of diarrheal disease in older adults is due to numerous factors, such as comorbid medical conditions, altered immunity with senescence, intestinal dysmotility, exposure to medications including antibiotics, and drug-induced or acquired hypochlorhydria.

Immunocompromising conditions influence the risk of acquisition and disease severity for numerous enteric pathogens. Classic immunocompromising conditions include malignancy and its treatment, hemopoietic or solid organ transplantation, medications such as glucocorticoids or other immune modulators, and HIV infection. For example, lymphoproliferative diseases, glucocorticoid therapy, and HIV infection markedly increase the risk for salmonellosis. However, a number of other conditions, the immune effects of which are less well characterized, also enhance the risk of infection with enteric pathogens, including diabetes mellitus or liver disease (e.g., Vibrio vulnificus ), hemolysis (e.g., salmonellosis), and iron overload (e.g., yersiniosis).

Food exposures or ingestion of contaminated water also constitute major risk factors for acquisition of enteric pathogens (see Table 110.1 and Chapter 111 ). Ingestion of undercooked or raw foods (e.g., meats such as hamburger or chicken, oysters, clams, other seafood) and unpasteurized milk or juices (e.g., apple cider) is of particular concern. Travel, largely because of exposure to contaminated food or water, is a classic risk factor for acquiring infectious diarrheal disease (see “Travelers’ Diarrhea,” later). Animal exposures such as on farms or in petting zoos at county or state fairs (e.g., Campylobacter , Salmonella , STEC) or recreational exposures (e.g., camping or adventure tourism) also pose risk for enteric infections. Exposure to health care facilities, including hospitalization and treatment with antibiotics, increase the risk for C. difficile infection (see Chapter 112 ). Both gastroenteritis or asymptomatic carriage of enteric pathogens are common in children in day care centers and serve as a source for infections in adults caring for the children.

Pregnancy and the peripartum period enhance risk for transmission of enteric pathogens. A pregnant woman may transmit enteric pathogens carried asymptomatically to the newborn child, who may then develop clinical disease. Conversely, neonates or infants can transmit infection, particularly C. difficile , to their mothers or caretakers, because C. difficile infection is typically asymptomatic in children under the age of 1 year. Ingestion of delicatessen foods by pregnant women may lead to bacteremia, septicemia, meningoencephalitis, fetal loss, or neonatal infection from Listeria monocytogenes . Onset of these serious illnesses may occur up to 30 days after the implicated food exposure. Older adults and those with impaired immune systems are also at increased risk for invasive L. monocytogenes infection after contaminated food ingestion. In contrast, the immunocompetent host ingesting a food highly contaminated with L. monocytogenes develops abrupt but a short-lived diarrheal illness with fever.

Differentiating IBD and Infectious Diarrhea

Clinical distinction between infectious dysentery (diarrhea with blood and mucus in the stool) and IBD (particularly UC) can be difficult because diarrheal stools in both illnesses contain mucus and blood. Two features of infectious dysentery that distinguish it from UC are detection of a pathogen on diagnostic studies and a self-limited course that responds to antimicrobial therapy without relapse. Positive diagnostic tests for a pathogen in adults with diarrhea, however, are obtained in only about 50% of those studied.

Longitudinal studies in the 1990s provided some guidance in differentiating acute infectious colitis and early-onset IBD. Most patients with infectious colitis present early (within 1 week) and with fever to the physician; histopathology of the colon in infectious colitis also is different from that of chronic colitis in IBD (see later). In contrast, patients with IBD often have prior abdominal symptoms upon a careful history, less fever, and present later (>1 week after symptom onset). Travel may cause a patient to manifest either an acute exacerbation of IBD because chronic symptoms worsen as a result of travel-acquired enteric infection or de novo IBD that develops as a result of an altered intestinal microbiome superimposed on a genetic predisposition (see Chapter 115 ). Lastly, higher volume or frequency of diarrhea (especially >10 stools per day) is more commonly due to an infectious pathogen than IBD.

Histopathologic examination of colonic mucosa obtained by endoscopic biopsy can be helpful. Both the microbial form (dysentery) and various types of acute colitis typically show edema, neutrophils throughout the lamina propria, and superficial cryptitis with preservation of the normal tubular crypt pattern. Idiopathic UC, however, shows signs of chronicity, including lymphoplasmacytosis that typically involves the lower third of the lamina propria, and architectural crypt distortion (e.g., colonic glands with signs of regeneration such as branching). Resolving dysentery may have lymphocytes infiltrating the lamina propria, similar to UC, but crypt distortion and regeneration will not be present. Ultimately, there is no substitute for correlating clinical, endoscopic, and histopathologic features of disease to arrive at the proper diagnosis.

Laboratory Diagnosis

Laboratory diagnosis of acute diarrheal disease should be pursued in any patient judged as moderately to severely ill by the clinician (e.g., fever, toxicity); in patients with risk factors (see earlier) including exposure history (e.g., high-risk foods, travel, antibiotics); in patients whose symptoms suggest inflammatory diarrhea or who have bloody diarrhea; and in some patients with persistent diarrhea beyond 7 days. In general, most episodes of acute diarrheal illness in the USA are self-limited; diagnostic testing may be kept to a minimum, and treatment is aimed at preventing dehydration.

Recently published guidelines on the use of the microbiology laboratory for the diagnosis of infectious diseases provide a comprehensive resource on fecal testing to identify enteric pathogens including bacterial, viral, and parasitic agents). Crucial points in diagnostic testing are: (1) Only diarrheal stools (i.e., those that take the form of the container) should be tested; (2) Rectal swabs are inferior to stool culture for testing in adults (and rejected by many clinical microbiology laboratories) but perform similarly in children; (3) Communication with the clinical microbiology laboratory is essential because the diagnostic testing capabilities of clinical microbiology laboratories vary. In particular, if the clinician is concerned about a particular organism(s), this should be communicated to the laboratory; specific examples are STEC, Vibrio , and Yersinia testing, which may only be done on special request and may require the assistance of public health authorities; (4) With improving technology and increasing use of molecular, culture-independent tests, single stool specimens in adults detect approximately 90% of enteric pathogens and, in children, 98%. Thus, one sample for children and a second for selected adult patients is reasonable for diagnosis. In patients with the proctitis syndrome, 1 sample is typically adequate for diagnosis.

A range of tests are used to identify enteric pathogens, including bacterial and viral cultures, various pathogen-specific nucleic acid amplification tests (NAATs), and immunoassays (e.g., enzyme- or fluorescent-based; see later STEC section), ova and parasite examinations, and certain specialized stains. In recent years, multiplex molecular assays, testing for an array of bacterial, parasitic, and viral pathogens, have increasingly been used for diagnosis with greater than 97% specificity and greater than 98% sensitivity . Multiplex assays reduce cost and turnaround time for most pathogens compared with conventional methods, yet require selection of a pre-focused panel of pathogens and will detect viable and nonviable organisms. Of note, re-analysis of the Global Enteric Multicenter Study (GEMS) specimens using quantitative molecular diagnostics ascribed 89.3% of childhood diarrheal cases to a pathogen compared with 51.5% of cases using conventional diagnostics. These techniques changed the population-level characterization of disease incidence and are an important illustration of how historical and contemporary incidence data needs to be considered in the context of the assays used.

Conventional biomarkers to differentiate noninflammatory and inflammatory diarrhea are unreliable. Fecal examination for leukocytes is limited by the ability of most enteric pathogens to induce some inflammation and observations that even classic inflammatory pathogens (e.g., Salmonella , Shigella , C. difficile ) often do not induce a marked fecal leukocyte response ( Table 110.4 ); fecal calprotectin may be emerging as more useful , but is not elevated in all cases. Serum markers of inflammation, including ESR and CRP, are also insufficient for all cases.

More invasive investigations, including flexible sigmoidoscopy with biopsies and upper GI endoscopy with duodenal aspirate and biopsies, are reserved for special situations such as the immunocompromised host in whom stool examination has not yielded a diagnosis. Flexible sigmoidoscopy can be useful in evaluating patients with proctitis, tenesmus, or sexually transmitted diseases (STDs), or in identifying the pseudomembranes of C. difficile infection.

Vibrio cholerae

Cholera, the prototypical toxigenic diarrhea, can cause dehydration and death within a few hours of onset. Stool output can exceed 1 L/hr, with daily fecal outputs of 15 to 20 L if parenteral fluid replacement keeps up with losses. The acutely ill patient typically has marked signs of dehydration: poor skin turgor, “washerwoman’s” hands, absent pulses, reduced renal function, and hypovolemic shock.

More has been learned about pathophysiology—and normal intestinal function—from cholera than from any other intestinal disease.

Other Vibrio Species

Aeromonas Species

Aeromonas species are ubiquitous environmental organisms found principally in fresh and brackish water, especially in the summer months, with clinical manifestations of infection similar to those caused by Vibrio species. Aeromonas species are divided into 2 groups: psychrophilic (Greek: psychros , cold) aeromonads, which grow optimally at temperatures ranging from 22°C to 25°C, and mesophilic aeromonads, which grow best between 35°C and 37°C. Psychrophilic strains usually are isolated from environmental water sources and fish; Aeromonas salmonicida is the most common strain in this group. Based on their phenotypic features, the mesophilic aeromonads are further grouped into 3 complexes: Aeromonas hydrophila, Aeromonas caviae, and Aeromonas veronii . All 3 of these Aeromonas species have been associated with human infection. In order to cause disease, aeromonads must adhere to and colonize the intestinal epithelium. Aeromonas strains produce an array of toxins, including heat-labile enterotoxin, hemolysin, and cytotoxin. There also may be some degree of invasion of the epithelial cells, with resultant dysentery or colitis.

Plesiomonas shigelloides

Plesiomonas shigelloides is a motile gram-negative rod, member of the family Enterobacteriaceae, and ubiquitous freshwater organism . Most cases are associated with consumption of raw seafood, and the organism has been reported to cause outbreaks. P. shigelloides also causes travelers’ diarrhea and constitutes 1% to 3% of such cases in Latin America and Africa and about 5% of such cases in Asia. The pathogenesis of P. shigelloides infection is poorly understood. There is evidence that potential virulence factors include cytotoxic hemolysin, iron acquisition systems, and lipopolysaccharide. Diarrhea ranges from mild and watery to severe colitis with visible blood. Abdominal pain often is prominent, and fever and vomiting are common. Extraintestinal manifestations, usually sepsis or meningitis, are rare and are more common in children and immunosuppressed patients. The diarrhea is usually self-limiting, so antibiotics are likely of limited use. Chronic diarrhea and extraintestinal disease may benefit from antibiotic therapy. P. shigelloides is commonly resistant to aminopenicillins and tetracyclines. Sporadic resistance to other antibiotics may occur. Potential effective treatment regimens are likely similar to those for other microbiological causes of dysentery and involve a fluoroquinolone, third-generation cephalosporin, or even carbapenems in severe or resistant infections, but little information is available on the efficacy of treatment.

Escherichia coli Species

E. coli are common, but constitute a minority of the commensals of the intestinal microbiota in humans and animals. Although most strains are relatively innocuous in the bowel, others possess virulence factors that cause diarrheal disease. At least 6 types of E. coli intestinal pathovars have been recognized ( Table 110.7 ). Their virulence factors include toxin production, adherence to epithelial cells, and invasiveness, each of which is encoded by specific genetic elements (plasmids or chromosomal genes) that determine pathogenicity.