Enteric Bacteriology: Stool Workup, Enterobacterales, and Curved Oxidase-Positive Organisms

Infectious diarrhea workup starts with the specimen that actually contains the pathogen, then a method that can still recover an isolate when public-health or susceptibility testing is needed. Culture-independent nucleic acid amplification test (NAAT) panels are now often first-line stool tests because they return a faster, more sensitive result than culture alone. Culture capability must remain available for isolates that need susceptibility testing and for outbreak submissions to public-health laboratories.^1,2

Stool collection, transport, and workup strategy

The preferred specimen is diarrheal stool that takes the shape of the container, collected in a clean, tightly lidded cup that is free of urine, barium, and antidiarrheal compounds. A rectal swab is less sensitive than stool for culture and is reserved for Neisseria gonorrhoeae, Chlamydia trachomatis, Shigella spp., and vancomycin-resistant Enterococcus (VRE) surveillance. Insert the swab past the anal sphincter until it is visibly stained, rotate it, and place it into Cary-Blair or the manufacturer’s molecular transport medium. Prompt delivery within 2 hours matters because falling pH as stool cools inhibits Shigella in particular. Use Cary-Blair if transport will be delayed.^1,2

Multiple stool specimens are rarely indicated for bacterial enteric pathogens. In adult studies of serial specimens, the first sample detected 87% to 94% of bacterial pathogens and a second sample raised detection to about 98%. In children, the first specimen already detects about 98% of bacterial enteric pathogens. A second specimen may be considered in selected adults when suspicion remains high after a negative first test. After 3 hospital days of diarrhea, routine community enteric-pathogen testing other than Clostridioides difficile is restricted unless the patient has eaten food brought in from outside the hospital.²

Place rectal swabs for N. gonorrhoeae and C. trachomatis in the appropriate transport medium and deliver them promptly or refrigerate them briefly. NAAT is more sensitive and specific than culture for these specimens. NAATs cleared for rectal and oropharyngeal specimens are available. Assays without that clearance require in-house validation for extragenital use.^1,2

Testing for toxin-mediated food poisoning is handled separately from culture-based diarrhea workup. Short-incubation illness from preformed Staphylococcus aureus or Bacillus cereus enterotoxin, and long-incubation illness from Clostridium perfringens, is usually referred to public-health laboratories. Suspected botulism testing is also referred for toxin or organism detection in feces because most clinical laboratories are not equipped for this work.¹

Selective and differential media

Selective media inhibit competing flora. Differential media display a color reaction that separates pathogens from commensals.

Most laboratories that still culture stool routinely recover Salmonella, Shigella, and Campylobacter. Shiga toxin-producing E. coli should be sought in the same specimens by toxin immunoassay or NAAT, with or without O157 culture. Isolation of Vibrio and Yersinia in the United States is usually a special request and requires additional media or incubation conditions.^2,3

Targeted stool pathogen workup

Serotyping of Escherichia coli. The E. coli serotype designation combines the O (somatic lipopolysaccharide) antigen with the H (flagellar) antigen and, for some strains, the K (capsular) antigen. O157:H7 is one example. Shiga toxin-producing E. coli (STEC) O157:H7 is the classic sorbitol-negative, MUG (4-methylumbelliferyl-β-D-glucuronide)-negative serotype. It fails to ferment sorbitol on sorbitol-MacConkey agar and lacks the β-glucuronidase activity that produces fluorescence in the MUG test; both reactions are atypical for E. coli generally. A presumptive sorbitol-negative colony is confirmed with O157 and H7 antisera by slide agglutination. Non-O157 STEC serogroups O26, O45, O103, O111, O121, and O145 now account for a substantial share of Shiga toxin-producing disease in the United States. The sorbitol screen misses them because most ferment sorbitol normally. Direct detection of Shiga toxin by enzyme immunoassay or of the stx1/stx2 genes by NAAT is therefore the primary strategy for detecting STEC of any serogroup in the clinical laboratory. State public health and reference laboratories perform serotyping with O:H antisera panels or whole-genome-sequence-based serotype prediction to confirm a Shiga toxin-positive isolate’s serogroup and link cases during outbreak investigation.^1,3

Enterobacterales

The Enterobacterales are facultatively anaerobic, non-spore-forming, oxidase-negative, gram-negative bacilli that ferment glucose with acid production and reduce nitrate to nitrite. Clinically important genera include Citrobacter, Enterobacter, Escherichia, Klebsiella, Morganella, Proteus, Providencia, Salmonella, Serratia, Shigella, and Yersinia. Plesiomonas was reclassified into this order and remains its only oxidase-positive member.¹

Endotoxin (lipopolysaccharide) drives much of the morbidity of gram-negative infection, including fever, hypotension, granulocytosis followed by consumption, thrombocytopenia, disseminated intravascular coagulation, and complement activation. These effects can culminate in septic shock. Organism-specific virulence factors include the K1 capsular antigen of neonatal-meningitis E. coli, the antiphagocytic capsule of K. pneumoniae, the Vi capsular antigen of Salmonella Typhi, which impedes intracellular killing, and fimbrial adhesins generally. E. coli diarrheal disease is classified by mechanism: enterohemorrhagic (Shiga toxin-producing), enterotoxigenic, enteroinvasive, enteropathogenic, and enteroaggregative. In enterotoxigenic disease, heat-labile toxin activates adenylate cyclase/cAMP and heat-stable toxin activates guanylate cyclase; both drive secretory watery diarrhea.

Presumptive colony clues include swarming Proteus spp. on blood agar; mucoid, lactose-positive Klebsiella spp.; red-pigmented Serratia marcescens; H₂S-producing Salmonella spp.; and indole-positive E. coli. MALDI-TOF identifies Enterobacterales reliably and quickly. The close genomic relationship between Shigella spp. and E. coli prevents reliable mass-spectrometry discrimination, so biochemical or serologic confirmation remains necessary.

Biochemical differentiation of representative organisms uses +, ≥90% positive; −, ≥90% negative; V, variable:

Susceptibility among these organisms is unpredictable enough that testing is warranted whenever therapy is planned. Uncomplicated Salmonella enteritis is an exception, because antimicrobial therapy can prolong carriage, as are mixed-flora infections in which a single-isolate result may not apply. First-generation cephalosporin resistance is expected in chromosomal-AmpC Enterobacterales, including Enterobacter cloacae complex and Klebsiella aerogenes (formerly Enterobacter aerogenes), Citrobacter freundii complex, Serratia marcescens, Providencia spp., Morganella morganii, and Hafnia alvei; the same intrinsic pattern occurs in Aeromonas spp. and Pseudomonas aeruginosa. Ampicillin susceptibility, once common in E. coli and P. mirabilis, has eroded substantially. Extended-spectrum β-lactamases (ESBLs) and acquired AmpC enzymes are recognized by resistance to extended-spectrum cephalosporins and/or cephamycins with retained carbapenem susceptibility. They are increasingly common in E. coli, K. pneumoniae, and P. mirabilis. Carbapenem-nonsusceptible isolates are sent to the carbapenemase workup. Carbapenemase-producing strains, initially those producing K. pneumoniae carbapenemase (KPC), are now a global infection-control priority. Phenotypic and genotypic detection methods are covered with antimicrobial susceptibility testing.^1,4

Curved and oxidase-positive enteric organisms

Gram-negative spiral rods that are oxidase-positive and hydrolyze sodium hippurate may be reported as C. jejuni without further testing, although some C. jejuni strains are weakly positive or negative and then need another method. C. coli is typically hippurate-negative. Biochemical and disk identification methods, including nalidixic-acid and cephalothin disks, are no longer recommended for species identification because of poor accuracy and common fluoroquinolone resistance. MALDI-TOF or species-specific PCR is preferred. Primary selective agar is incubated 48 to 72 hours at 41 to 42.5 °C under microaerophilic conditions (classically 5% O₂, 10% CO₂, 85% N₂). That temperature favors C. jejuni and C. coli and can miss C. fetus and other non-thermophilic species, which grow better at 35 to 37 °C, often with hydrogen enrichment.⁵

H. pylori is diagnosed noninvasively wherever possible. The urea breath test (detects urease activity via labeled CO₂ in expired breath) and monoclonal-antibody stool antigen EIA are both preferred over serology, which cannot distinguish active from past infection. Culture, when pursued, requires incubation for 10 or more days in a microaerobic, humid, CO₂-enriched atmosphere. Culture remains slow and is not first-line diagnosis. Phenotypic or molecular susceptibility testing is used when the next regimen is uncertain after prior treatment and antibiotic exposure, and before clarithromycin- or levofloxacin-containing regimens.⁶

V. cholerae and V. mimicus grow without added NaCl; the other medically important species require salt. A string test (0.5% sodium deoxycholate) is positive in most Vibrio isolates and usually negative in Aeromonas. Oxidase testing is performed from blood agar, not from TCBS. The CDC differential scheme uses growth in nutrient broth with 0% or 1% NaCl, oxidase, nitrate reduction, myo-inositol fermentation, arginine dihydrolase, lysine and ornithine decarboxylase, and sucrose fermentation on TCBS:^1,7

Most V. vulnificus isolates are sucrose-negative and green on TCBS, but a minority ferment sucrose and appear yellow. Colony color on TCBS does not exclude V. vulnificus. MALDI-TOF, species-specific PCR, or whole-genome sequencing confirms identification when biochemical results are inconclusive.⁷

Anaerobic bacteria in enteric specimens

Anaerobes are the predominant flora of the gastrointestinal tract and are common in mixed wound and abscess cultures after mucosal or tissue disruption. True (obligate) anaerobes cannot tolerate more than about 0.5% O₂; aerotolerant anaerobes show scant growth in room air with CO₂. General anaerobe specimen strategy belongs with primary processing; the organism-facing work here is identification of the enteric anaerobes recovered from stool or mixed gastrointestinal specimens.

Specimens are plated to enriched non-selective media (Brucella, CDC, or Columbia agar with sheep blood, vitamin K, and hemin), selective/differential agars, and a broth. Selective/differential options include kanamycin-vancomycin laked blood agar, Bacteroides bile esculin (BBE) agar, phenylethyl alcohol (PEA), and colistin-nalidixic acid (CNA) agar. Plates are incubated anaerobically at 35 °C in 80-90% N₂, 5% H₂, and 5-10% CO₂. Wait at least 48 hours before the first reading because transient oxygen exposure during premature examination kills slower-growing species.

C. difficile detection uses the GDH / toxin immunoassay / NAAT algorithm above. Testing is restricted to patients with ≥3 unformed stools in 24 hours who have not recently taken laxatives; formed-stool testing is discouraged because asymptomatic carriage is common. Infants 12 months of age or younger are not routinely tested because asymptomatic carriage of toxigenic C. difficile is high. The 2021 IDSA/SHEA focused update changed treatment; it did not change this diagnostic stewardship. On CCFA the colonies are white, with a “horse stable” odor and chartreuse fluorescence. Enterotoxin TcdA and cytotoxin TcdB damage the colonic epithelium and produce antibiotic-associated diarrhea and pseudomembranous colitis.^2,8

Clostridium perfringens, a major wound isolate, forms large gray colonies with a double zone of β-hemolysis; the rods are boxcar-shaped and usually nonmotile, lecithinase-positive, lipase-negative, and reverse CAMP-positive.

Anaerobic susceptibility testing is indicated for isolates from normally sterile sites, pure culture, and highly virulent or unpredictable genera (Clostridium, Bacteroides, Fusobacterium, Prevotella). Testing uses microbroth dilution in supplemented Brucella broth (hemin, vitamin K₁, 5% lysed horse blood). Antimicrobial susceptibility testing covers testing-menu selection and the CLSI reference method.