Update on Shiga Toxin-Producing Escherichia coli
Nearly every year, a well-publicized outbreak of disease is caused by the enteric bacterium, Escherichia coli O157:H7. Infection with this organism can cause hemorrhagic colitis, and may lead to hemolytic uremic syndrome (HUS), which is a life-threatening condition characterized by hemolytic anemia, thrombocytopenia, and kidney failure.1 Traditionally, associated risks include consumption of undercooked ground beef or unpasteurized milk, or animal contact. In addition, newly associated food risks include consumption of contaminated sprouts, spinach, or other leafy green vegetables. The pathogenicity of E coli O157:H7 relates to its ability to produce Shiga toxin; non-O157:H7 strains of E coli are increasingly recognized to produce Shiga toxin and have been associated with hemorrhagic diarrheal outbreaks. In response to these recent developments, as well as increasing confusion regarding testing for this pathogen, the Centers for Disease Control and Prevention (CDC) has published recommendations2 for diagnosis of Shiga toxin-producing E coli (including O157:H7 and non-O157:H7 strains). One key recommendation is to routinely test stools being cultured for enteric bacterial pathogens with an assay that detects Shiga toxin. Mayo Medical Laboratories has recently introduced a molecular test to detect Shiga toxin-producing E coli. This article reviews the diagnostic tools for detecting E coli O157:H7 and other Shiga toxin-producing E coli.
Pathogenic Escherichia coli
E coli includes a variety of related strains, most of which are part of the usual intestinal flora of mammals. Some strains commonly cause human disease, producing both diarrheal infections and extraintestinal infections (eg, urinary tract infection and meningitis). The diarrheagenic E coli strains are often grouped into at least 5 categories based on their pathogenic features. One is Shiga toxin-producing E coli (STEC) also referred to as enterohemorrhagic E coli (EHEC); others are referred to as enterotoxigenic E coli (ETEC), enteropathogenic E coli (EPEC), enteroaggregative E coli (EAggEC), and enteroinvasive E coli (EIEC). These groups of strains can be differentiated by the identification of virulence factors or the use of specific antisera. Typically, strains within a serotype cause similar disease. For example, one of the more well-known serotypes is E coli O157:H7, so named because it has the 157th somatic or O antigen ever identified and the 7th flagellar or H antigen.
The EHEC serotypes of E coli that cause hemorrhagic colitis produce 1 of 2 types of Shiga toxin (Stx). The 2 types of Shiga toxin (Stx1 and Stx2), are encoded by 2 separate genes, stx1 and stx2, one or both of which may be carried, independent of each other, in the genomes of various isolates. Serotypes of E coli that have stx1 or stx2 or both, encoded in their genomes are referred to as Shiga toxin-producing E coli, or STEC for short. Although STEC and EHEC are often used interchangeably, the definitions differ subtly. The EHEC group comprises strains based on a clinical definition, whereas the STEC group is based on a biochemical definition; nevertheless, the 2 groups have significant overlap. For the purpose of this discussion, henceforth, the STEC designation will be used, and serotypes will be referred to by only the somatic (O) antigen type.
STEC causes diarrhea, and often, hemorrhagic colitis. Unlike other agents of bacterial diarrheal disease (eg, Campylobacter jejuni, Salmonella enterica), infections by STEC can lead to HUS. Aside from O157 STEC, there are also less common serotypes of STEC, including O26, O111, and O103. These non-O157 STEC usually, though not always, cause less severe disease than that caused by O157 STEC. Analogously, it is generally the case that STEC carrying stx2 cause more severe disease than those carrying only stx1. It must be noted, however, that outbreaks, severe infections, and cases of HUS have been associated with non-O157 STEC and with isolates that only carry stx1. Although non-O157 STEC are not easily identified in a clinical laboratory, epidemiology studies have demonstrated that they may cause about half of STEC infections.3,4
About 73,000 people are known to be infected by O157 STEC each year, and if non-O157 STEC are additionally considered, the number surpasses 100,000 cases, resulting in 3,000 hospitalizations and 90 deaths. Depending on the specific outbreak and ages of the affected population, HUS can occur in 3% to 20% of cases (it is usually more common in children). Most outbreaks of STEC share a common feature, which is contact with some amount of fecal contamination by ruminants (typically cattle) that are colonized with and shed STEC in their feces. Some outbreaks, however, are due to person-to-person transmission in settings where close interaction and substandard hygiene results in transmission (eg, day care and institutionalized care settings). The low infectious dose of STEC, and O157 STEC in particular, makes outbreaks particularly likely in these settings. It has been estimated that only 10 to 100 O157 STEC organisms are needed to cause an infection. This is in contrast to Vibrio cholerae, the agent of cholera, which requires up to a million organisms to establish an infection. The infectious doses of most non-O157 STEC have not been well characterized.
The incubation period for STEC infections is typically 3 to 7 days from exposure to presentation, and patients with STEC usually report diarrhea, often bloody, with upper right quadrant pain. Other enteric pathogens in the differential diagnosis include bacteria such as Campylobacter, Salmonella, Shigella, and Yersinia species Patients normally do not have recent travel history, though STEC infections are found worldwide. Treatment for STEC is primarily supportive, with evidence that parenteral volume expansion is helpful, probably by diluting circulating Shiga toxin. Studies with O157 STEC suggest that antibiotics may exacerbate disease by providing a stress signal that may increase Shiga toxin production and release.5 For this reason, a rapid and reliable diagnostic assay for detection of STEC is necessary to assure prompt and appropriate treatment and limit use of potentially harmful antibiotics. Poor performance of a diagnostic assay could lead to a pseudo-outbreak in the case of false-positive results,6 which can lead to extensive waste of monetary and personnel resources. Similarly, false-negative results may lead to missed opportunities to contain outbreaks in the community. On the single patient level, consequences of delayed or missed diagnosis of STEC include severe disease and sequelae, with the potential for irreversible kidney damage and death.
The traditional method of detecting O157 strains of E coli in the clinical laboratory is based on the use of selective and differential culture methodology. Unlike most strains of E coli, O157 STEC typically lack the ability to ferment sorbitol, and so appear as colorless colonies on sorbitol MacConkey (SMAC) agar. Several commercial media are available that exploit this metabolic feature. For example, Chromagar O157 is analogous to SMAC, but the O157 STEC colonies appear mauve-colored, while normal flora appear as other colors, primarily dark blue (Figure 1). Culture is a common and useful technique, in part because the laboratory supplies needed are inexpensive and the results can be highly specific. A disadvantage is that non-O157 STEC (and sorbitol-fermenting O157 STEC) are not easily detected by culture in the clinical laboratory, and detection of these strains is primarily performed by public health laboratories. To identify non-O157 STEC (and sorbitol-fermenting O157 STEC), less selective media are used. Then, panels of antibodies for STEC O-antigens (eg, O157, O26, O45, O103, O111, O121, and O145) are used to differentiate potential STEC from other E coli. Alternatively, antigen or molecular techniques are used on the isolates to test for Shiga toxin genes or proteins.
Culture methods require well-trained technologists to review culture plates, and a relatively high per-sample hands-on time, the cost of which must be factored into the cost per specimen. Additionally, the large amounts of organism produced during culture combined with the low infectious dose of STEC can present a biosafety risk in the laboratory. The turnaround time for culture is typically at least 2 days, sometimes longer. Outside the clinical laboratory, culture is, and will remain, an essential public health tool for investigation of infection clusters and outbreaks.
Figure 1: Colonies of normal flora (A) and presumptive E coli O157:H7 (B) on Chromagar O157 medium
Shiga Toxin Antigen Assays
As an alternative to culture, Shiga toxin antigen detection methods theoretically enable detection of all STEC, not just O157 STEC. The methods utilize antibodies that recognize and bind to Stx1 and/or Stx2 resulting in agglutination of latex beads or the generation of a chemical signal that is visualized by a plate reader or on an immunochromatographic (IC) strip. Latex agglutination and IC strips are useful for small testing volumes, but suffer from inherent subjectivity. On the other hand, enzyme immunoassays (EIAs) are more objective and can often be automated, which can enable more rapid throughput in the laboratory. However, to rival the sensitivity and specificity of culture, the EIA should be performed on overnight broth cultures of fecal specimens. The necessity of this step means that most EIAs still take about 18 to 24 hours to report a result. Although EIAs have not improved turnaround time, they represent a major advance because they enable detection of both O157 and non-O157 STEC.
Over the past decade, real-time polymerase chain reaction (PCR) has become an important tool in the clinical diagnostic laboratory.7 With this technique, a target region of nucleic acid is amplified to high levels and the presence of the sequence is detected in real time. Because there is no need to open the tube containing the nucleic acid, the assay is considered a closed system. These advantages are significant improvements over traditional PCR with gel and blot detection systems, which are not necessarily faster than culture, and are subject to contamination (because they are open systems). Assays that utilize real-time PCR for pathogen detection are typically highly sensitive and specific, have a turnaround time measured in hours, and, as a closed system, are much less prone to nucleic acid contamination than open system molecular assays. The drawbacks of real-time PCR include the expense of instrumentation and expertise needed to establish and maintain testing, as well as the limitation of detecting just 1 or 2 gene targets. The specific target sequences enable increased specificity. Although no strain is isolated during real-time PCR, the impact on public health is essentially neutral; positive specimens, rather than isolated strains, can be submitted to public health and/or state laboratories, and they may be submitted 1 day faster than with culture or EIA. In this way, public health laboratories can still carry out their vital services of typing isolates toward the goal of identifying and investigating outbreaks, and any contact investigations pertaining to the patient may be initiated faster than with culture or EIA. See Table 1 for the advantages and disadvantages of testing methods.
|Culture||24 to 48 hours||
|EIA||24 to 48 hours||
|Real-time PCR||4 hours||
Table 1: Advantages and disadvantages to testing methods
New CDC Recommendations
Guidelines published by the CDC in 2009 recommend that stool specimens from all patients with community-acquired diarrhea be tested with an assay that targets Shiga toxin antigens or genetic determinants to detect non-O157 STEC.2 This recommendation is made without regard to the age of the patient, the season of the year, or the presence of blood in the stool. The broadness of the recommendation acknowledges that STEC disease is no longer limited to summertime outbreaks of bloody diarrhea or the pediatric population. It is unknown whether the apparent change of epidemiology reflects a true shift, or simply results from increased awareness and testing. However, there are reports that STEC are detected with frequencies that rival those of other common bacterial stool pathogens such as Campylobacter, Salmonella, and Shigella species.8
Unlike testing for parasites in stool, which requires submission of 2 to 3 stool samples to be submitted for an adequate investigation, 1 specimen is usually sufficient for STEC testing. As mentioned above, the CDC recommendations describe a nonselective testing strategy for community-acquired diarrhea. However, an instance where selective testing may be considered is the case of a patient who has been hospitalized for >3 days. Typically, diarrhea in this population is associated with Clostridium difficile. If there is an ongoing STEC outbreak in the community or in the hospital, however, STEC testing may be useful in the hospitalized patient. When testing by any method in any patient, it is important to submit a stool specimen rather than a rectal or perianal swab, since a swab may not contain adequate specimen for STEC or other enteric pathogen testing.
Mayo Medical Laboratories’ Real-Time PCR Assay for Shiga Toxin
The Shiga toxin assay used at Mayo Clinic and Mayo Medical Laboratories is a real-time PCR assay that targets 2 sequences approximately 200 base pairs long and is specific for the genes encoding Shiga toxin, stx1 and stx2. Unlike EIA assays and many other PCR assays targeting Shiga toxin, the real-time PCR assay done at Mayo can be performed directly on a nonenriched stool specimen. This means results of testing are then available the same day the specimen arrives in the laboratory—a significant benefit in turnaround time. The test begins with a small amount of stool, which is mixed with water. Total nucleic acids from this prepared sample are extracted using an automated instrument. The nucleic acids are then combined with a polymerase and specific oligonucleotide primers and probes and placed in a thermocycler. The thermocycler completes 45 cycles of temperature changes. During that time, either or both of the 2 target sequences, stx1 and stx2, are amplified to detectable levels. Detection of the amplified target sequence utilizes fluorescently labeled hybridization probes, which are specific for a region within the amplified target sequence. When the correct target is amplified, the probes hybridize to the target sequence and, in the presence of the correct excitation wavelength, produce a signal via Förster resonance energy transfer (FRET). This methodology mandates 2 levels of specificity before a signal can be produced—one imparted by the primer, and another by the probe sequences.
The performance of Mayo’s assay was evaluated using 2 sets of specimens. One set was a prospectively collected group of 204 consecutive stool specimens submitted to the clinical microbiology laboratory for testing using a routine, US Food and Drug Administration (FDA)-approved EIA (which uses broth enrichment) for detection of Shiga toxin. These specimens were also cultured on Chromagar O157 (BD BBL, Sparks, MD) for detection of presumptive O157 STEC. The second group of specimens was an archival set of 85 clinical specimens (both normal and diarrheal) that, after initial testing with culture for O157 STEC or EIA for Shiga toxin, were frozen at -70°C.
For both sets of specimens, a positive PCR result was considered concordant if either the EIA or culture tests were positive. A negative result by PCR was only concordant if both EIA and culture were negative. Discordant results were resolved by testing at the Minnesota Department of Health utilizing a traditional PCR assay that targeted the stx genes with different primers. The results of the study (Table 2) showed that, after resolution of all discordant results, Mayo’s real-time PCR assay was 100% sensitive and specific.9 In addition, 1 of 4 (25%) specimens positive in the prospective arm of the study was an O103 STEC, which was not detected by culture. This is an example of the advantage of nonculture methods of STEC detection. Conversely, the method evaluation did not demonstrate any apparent advantage gained by culture testing.
|Combined Reference Methods|
|Mayo Real-time PCR Assay||Positive||
Table 2: Method comparison data from the validation study of the
real-time PCR assay developed by Mayo Clinic9
a Positive by EIA and/or culture (n=43). Some specimens (n=3) were negative by EIA and/or culture, but positive by the Minnesota Department of Health PCR assay.
b Negative by EIA and/or culture (n=233). Specimens with discordant results (n=10) were all negative by Minnesota Department of Health PCR assay.
Clinical Experience with the Assay
After 12 months of use (May 2009 to April 2010), the initial clinical experience of the PCR assay was tabulated, and positive results were again compared to testing at Minnesota Department of Health. In select instances, the clinical presentation of positive cases was also reviewed. The PCR assay yielded 18 positive results from 3956 specimens tested, for a positivity rate of 0.5%.10 Sixteen of the 18 positive specimens were also tested by the Minnesota Department of Health where 14 were positive by PCR and 3 of those were also positive by culture for O157 STEC. Of the 2 specimens negative for both PCR and O157 STEC culture at Minnesota Department of Health, 1 was from a child with HUS, and the other was from an adult with an acute diarrheal illness.
Supporting the rationale for the CDC guidelines, only 4 of the 18 Shiga toxin-positive specimens were from young children (17 months to 10 years), and the average age of positive patients was 48.4 years old (range 17 months to 84 years). In the same time period that the 18 Shiga toxin-positive specimens were identified, culture for other pathogens yielded 17 Salmonella, 0 Shigella, 41 Campylobacter, and 8 Yersinia isolates, making STEC the second most prevalent bacterial stool pathogen identified, following Campylobacter species.
Interpretation of Results
The 2 possible results from the real-time PCR assay are: 1) Negative and 2) Positive. All positive results are accompanied by the following comment: “Indicates the likely presence of Shiga toxin-producing Escherichia coli”. It is important to realize that although the name Shiga toxin may seemingly imply an association with Shigella species, the association is of limited clinical significance. There are 4 species of Shigella, but only S dysenteriae type 1, carries stx, and disease caused by this strain has a reported annual incidence of 5 cases a year.11 Thus, a positive result should be interpreted as indicating that either STEC O157 or STEC non-O157 is likely present in the stool.
The Shiga toxin real-time PCR assay described herein provides rapid turnaround time for results, which in turn enables care providers to make timely and accurate patient care decisions. Furthermore, contact investigations and potential outbreaks may be pursued more rapidly than with culture or EIA. Real-time PCR technology provides superior sensitivity and specificity to either culture or EIA and uses a closed system that reduces potential contamination. Since the assay is designed to detect both O157 STEC and non-O157 STEC, infections by either type of STEC are identified quickly and accurately.
Authored by Dr. Thomas Grys and Dr. Robin Patel
- Tarr PI, Gordon CA, Chandler WL. Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 2005 Mar 19-25;365(9464):1073-1086
- Centers for Disease Control and Prevention. Recommendations for diagnosis of shiga toxin--producing Escherichia coli infections by clinical laboratories. MMWR Recomm Rep 2009 Oct 16;58(RR-12):1-14
- Hedican EB, Medus C, Besser JM, et al: Characteristics of O157 versus non-O157 Shiga toxin-producing Escherichia coli infections in Minnesota, 2000-2006. Clin Infect Dis 2009 Aug 1;49(3):358-364
- Johnson KE, Thorpe CM, Sears CL. The emerging clinical importance of non-O157 Shiga toxin-producing Escherichia coli. Clin Infect Dis 2006 Dec 15;43(12):1587-1595
- Wong CS, Jelacic S, Habeeb RL, et al: The risk of the hemolytic-uremic syndrome after antibiotic treatment of Escherichia coli O157:H7 infections. N Engl J Med 2000 Jun 29;342(26):1930-1936
- Centers for Disease Control and Prevention. Importance of culture confirmation of shiga toxin-producing Escherichia coli infection as illustrated by outbreaks of gastroenteritis--New York and North Carolina, 2005. MMWR Morb Mortal Wkly Rep 2006 Sep 29;55(38):1042-1045
- Espy MJ, Uhl JR, Sloan LM, et al: Real-time PCR in clinical microbiology: applications for routine laboratory testing. Clin Microbiol Rev 2006 Jan;19(1):165-256
- Klein EJ, Boster DR, Stapp JR, et al: Diarrhea etiology in a Children’s Hospital Emergency Department: a prospective cohort study. Clin Infect Dis 2006 Oct 1;43(7):807-813
- Grys TE, Sloan LM, Rosenblatt JE, Patel R. Rapid and sensitive detection of Shiga toxin-producing Escherichia coli from nonenriched stool specimens by real-time PCR in comparison to enzyme immunoassay and culture. J Clin Microbiol 2009 Jul;47(7):2008-2012
- Nyre LM, Kiemele DL, Zomok CD, et al: Clinical Experience with Rapid PCR for Detection of Shiga Toxin in Stool. Paper presented at: 110th General Meeting of the American Society of Microbiology, San Diego, CA; 2010 May
- Centers for Disease Control and Prevention. Shigella Surveillance: Annual Summary, 2005. Atlanta, Georgia: US Department of Health and Human Services, 2006 November
Frequently asked questions about the Shiga toxin real-time PCR assay:
When should I test for STEC?
The CDC recommends testing when any routine stool culture is ordered.
What specimen sources are best for this test?
Stool specimen in a transport container, or in a stool transport media, such as ParaPak C&S or Cary-Blair, can be used. Stool in buffered glycerol saline is also acceptable. Rectal or stool swabs are not acceptable.
Does the assay detect E coli O157:H7?
Yes, but it can also detect non-O157 STEC and sorbitol-fermenting O157 STEC, which is recommended in the new CDC guidelines.
Does a positive result mean my patient has Shigella?
It is highly unlikely. Although 1 of the 4 Shigella species may produce Shiga toxin, the most likely interpretation is that a positive result indicates that either an O157 STEC or non-O157 STEC was detected. Both groups of STEC have been associated with severe disease.
Will I know whether the patient has stx1 or stx2? Does it matter?
No, the type of Shiga toxin detected is not reported. Both types have been associated with severe disease.