Vibrio and Related Organisms

TAXONOMY Back to top

The family Vibrionaceae is presently composed of six genera (Vibrio, Photobacterium,

Salinivibrio, Enterovibrio, Grimontia, and Aliivibrio) and over 110 species with standing in

bacterial nomenclature (http://www.bacterio.cict.fr/). Vibrio is the type genus for the family,

and Vibrio cholerae, the causative agent of pandemic cholera, is the type species (25).

Pathogenic species for humans can be found in three genera, including Vibrio (10

species), Photobacterium, and Grimontia (one species each). Phylogenetic investigations

indicate that multiple clades (separate or distinct groups in a phylogenetic sense) exist within

this genus, indicating that many Vibrio species may eventually be reclassified into different

genera. Photobacterium damselae is currently the accepted taxon for Vibrio

damsela. Although definitive DNA relatedness studies are lacking, phylogenetic investigations

employing 16S rDNA, gyrB, rpoA, recA, and pyrH gene sequencing indicate that P.

damselae clusters within the genus Photobacterium, albeit at the extreme periphery to the

type species,P. phosphoreum (35, 83). P. damselae strains also possess defining characters

associated with members of the genus Photobacterium, including accumulation of poly-β-

hydroxybutyrate and the absence of a flagellar sheath (79). Most scientific and medical

publications now report infections associated with this taxon as P. damselae.When further

genetic data become available, the classification of P. damselae may need to be reassessed.

However, P. damselae is clearly not a member of the genus Vibrio. One of us (J. J. Farmer

III) strongly disagrees with the proposed reclassification of Vibrio damsela as Photobacterium

damselae and will propose (unpublished data) a revised classification in which Vibrio

damsela will be classified in a new genus rather than in the genus Photobacterium (25).

DESCRIPTION OF THE VIBRIONACEAE Back to top

The Vibrionaceae involved in clinical specimens are gram-negative, facultatively anaerobic,

straight, curved, or comma-shaped rods, 0.5 to 0.8 μm in width and 1.4 to 2.6 μm in length,

that are catalase and oxidase positive (except Vibrio metschnikovii) (25). Most species are

motile by means of sheathed monotrichous or multitrichous polar flagella when grown in

liquid media. Strains of some species, such as V. parahaemolyticusand V.

alginolyticus, swarm on solid media by production of numerous lateral flagella (25, 52).

All Vibrionaceaerequire Na+ for growth, with the minimal concentration for optimum growth

ranging from 0.029 to 4.1% NaCl (25). They also ferment D-glucose but rarely produce gas,

reduce nitrate to nitrite (except Vibrio metschnikovii), and grow on thiosulfate-citrate-bile

salts-sucrose (TCBS) medium. The G+C content of the DNA is 38 to 51 mol% (25). Key

properties or characteristics useful in separating clinically significant Vibrionaceae from

phylogenetically or phenotypically related species are listed in Table 1.

EPIDEMIOLOGY, TRANSMISSION, AND CLINICAL

SIGNIFICANCE Back to top

The genera covered in this chapter are primarily isolated from marine environments. Vibrios

such as V. cholerae and V. mimicus that require minimal amounts of Na+ for growth can be

found in freshwater rivers and lakes as well as estuarine and marine

environments. Vibrionaceae are commonly isolated from a variety of bivalves and

crustaceans, and like other genera found in marine environments, their concentrations peak

during the warmer months of the year. In aquatic environments, vibrios can persist in a freeliving

state or in association with phytoplankton and zooplankton (28, 48, 61, 78). In the

environment vibrios may enter a state referred to as viable but nonculturable, in which cells

retain basic metabolic processes even though they fail to grow on standard laboratory media

(10). Because the viability of these cells is still in question, the term “active but

nonculturable” has been proposed (61). Recent studies have shown that in mixed

populations of nonculturable and culturable cells of V. cholerae, the latter appear to be the

main contributors to human infections (61).

The Vibrionaceae can be isolated from a wide variety of intestinal and extraintestinal human

illnesses. These illnesses include diarrhea, soft tissue disease (cellulitis and necrotizing

fasciitis), septicemia, and eye and ear infections (25). In some cases of gastroenteritis and

extraintestinal infections it may be difficult to determine if a positive vibrio culture represents

true infection or merely colonization because of its widespread occurrence in marine and

estuarine waters. Ten Vibrio species and one species each

of Photobacterium and Grimontiaoccur in clinical specimens and are listed in Table

2 (25, 56). Extraintestinal Vibrio infections are frequently associated with traumatic injuries

or inapparent exposure to estuarine or marine waters. Primary septicemia may occur after

ingestion of raw seafood (oysters) or as a secondary bacteremia subsequent to a wound

infection.

V. cholerae

V. cholerae is the only species that causes endemic, epidemic, and pandemic cholera

(20, 68). It is divided into three major subgroups: V. cholerae O1, V. cholerae O139, and V.

cholerae non-O1. Vibrio cholerae O1 strains may be further subtyped. The three serovars,

Inaba, Ogawa, and Hikojima, and the two biotypes, El Tor and classical, can occur in any

combination.

V. cholerae O1

In 2007, the WHO reported over 177,000 cases of cholera worldwide, with more than 4,000

deaths (86). In Africa, the current focus of epidemic cholera, case fatality rates approach 5%

(54). More than 70,000 infections were reported in Zimbabwe alone between November

2008 and February 2009. The majority of persons ingesting toxigenic V. cholerae O1 have

asymptomatic infections. Although the ratio of asymptomatic to symptomatic infections is

presumed to range from 3 to 100, recent models of infection indicate that the number of

asymptomatic infections is much higher and that these undetected infections serve as an

ongoing reservoir for cholera in regions of the world where cholera is highly endemic (40).

Classic cholera typically results in copious amounts of watery diarrhea passed painlessly,

with fluid loss reaching 200 ml/kg of body weight/day. If untreated, the patient becomes

prostrate with symptoms of severe dehydration, electrolyte imbalance, painful muscle

cramps, watery eyes, loss of skin elasticity, and anuria. Dehydration subsequently leads to

hypovolemic shock, acidosis, circulatory collapse, and death, even in previously healthy

adults (37). In the United States, occasional cases of classic cholera are seen in travelers

returning from regions of the world where cholera is highly endemic. Traditional therapy

consists of fluid replacement by oral rehydration and/or intravenous fluids. The unique ability

of V. cholerae serogroup O1 to cause this fulminant form of diarrhea is due to the presence

of virulence cassette regions and pathogenicity islands on the bacterial chromosome. These

regions contain a number of key determinants, including the cholera enterotoxin

gene, ctx, which is responsible for the large excretion of fluids and electrolytes into the

lumen, and a toxin-coregulated pilus gene, tcpA, responsible for attachment to the

gastrointestinal epithelium (26). Traditional (classical) cholera can be produced by the two

biotypes of V. cholerae O1, designated classical and El Tor. The first six pandemics were

thought to be due to the classical biotype, while the ongoing seventh pandemic, which began

in 1961, is caused by the El Tor biotype, which was first isolated in 1905 (68). Recently,

genetically evolving hybrid strains of V. cholerae O1 have emerged in eastern Africa and

Bangladesh (86). These hybrid strains may be more virulent based upon projected case

fatality rates. Hybrid strains differ from traditional El Tor or classical isolates in that they

carry different combinations of genes from their expected biotype (e.g., El Tor strain with

classical ctxB on the chromosome) (62). These hybrid strains appear to have evolved via

lateral gene transfer and recombination events (81).

V. cholerae O139

In 1992, cholera cases that were attributed to a then new serotype of V. cholerae, O139

(synonym: V. cholerae O139 Bengal), emerged, particularly in India, Bangladesh, and

throughout Asia (3). This serogroup probably resulted from the lateral transfer of a novel

somatic antigen and capsule from an unknown bacterium to an El Tor strain (3). O139 and

O1 strains carry similar virulence factors, including the ctx and tcpA genes (56). Clinical

diseases due to O1 and O139 V. cholerae are also strikingly similar, except that adults are

more frequently affected with O139 since previous infection with O1 cholerae is not

protective (27). In 2002, O139 reemerged in Bangladesh, causing an estimated 30,000

cases of cholera, primarily in older patients (27). According to the latest WHO report, 41% of

165 cholera cases reported from China were laboratory confirmed as O139 (86). In contrast,

less than 0.5% of >1,400 cholera cases reported by Thailand were O139. To date, no cases

of V. cholerae O139 infection have been identified in Africa (86).

V. cholerae Non-O1

V. cholerae non-O1 strains (non-O1, non-O139) are the third most commonly isolated vibrios

in clinical laboratories in the United States, following V. parahaemolyticus and V.

vulnificus. Unlike O1 strains, non-O1 isolates rarely produce cholera toxin. While the diarrhea

produced is watery and severe disease is reported, infections are usually milder than typical

cholera. Non-O1 strains of V. cholerae that do harbor the cholera toxin gene, such as

serogroups O75 and O141, have caused sporadic cases of cholera-like disease in the United

States and elsewhere (22, 80). Risk factors for infection appear to be similar to those for

other vibrios and include consumption of oysters and other seafoods. Non-O1 vibrios, in

contrast to O1 V. cholerae, are associated with extraintestinal infections such as septicemia

and epidural brain abscess (7, 65). Persons at increased risk of developing non-O1

bacteremia include those with liver disease/ cirrhosis or hematologic malignancies (65). The

case fatality rate in these patients ranges from 24 to 65% (42, 65). Strains have also been

isolated from ears, wounds, the respiratory tract, and urine (57).

V. mimicus

V. mimicus, a nonhalophilic species, is biochemically similar to V. cholerae except that it is

sucrose negative. Human infections are uncommon, but it is recovered from patients with

diarrhea in which it may or may not have a causal role and is usually associated with

consumption of uncooked seafood, particularly raw oysters. Rare strains carry the ctx gene

and can produce cholera-like symptoms. Symptoms generally include abundant watery

diarrhea, vomiting, and severe dehydration. Most descriptions of V. mimicus gastroenteritis

involve individual case reports, but a large-scale foodborne outbreak of gastroenteritis was

reported in Thailand in 2004 (16). In that outbreak, over 300 persons were ill, and rectal

swabs collected from 24 patients all yieldedV. mimicus. Presumptive causes of this outbreak

included freshwater fish, seafood, and seafood soup. An earlier outbreak was reported from

Costa Rica from 1991 to 1993. It involved 33 persons with V. mimicus-associated diarrhea,

and raw turtle eggs were the implicated vehicle of infection (11).

V. parahaemolyticus

V. parahaemolyticus is the leading cause of bacterial foodborne intestinal infections in Asia

and is almost invariably associated with the consumption of raw fish or shellfish (75). In

Japan, 50 to 70% of the cases of foodborne diarrhea are due to V. parahaemolyticus. In the

United States, it is the Vibrio species most frequently isolated from clinical specimens and is

primarily associated with watery diarrhea. Symptoms of V. parahaemolyticus-associated

gastroenteritis often include nausea, vomiting, abdominal cramps, low-grade fever, and

chills. Fatalities are extremely rare but can occur in cases of severe dehydration. Rehydration

is usually the only treatment needed, but antimicrobial therapy may be beneficial in some

instances. A recent outbreak of gastroenteritis involving 22 passengers aboard a cruise ship

was linked to Alaskan oysters (53). A now widely dispersed clone of V.

parahaemolyticus, serotype O3:K6, emerged worldwide in 1997 (59). Strains of this serotype

caused an unusually high proportion of V. parahaemolyticus foodborne disease outbreaks in

Taiwan from 1996 to 1999, which suggests that there is something unusual regarding its

ecology, epidemiology, or virulence. This clone, with several of its serovariants, O4:K68,

O1:K25, and O4:K12, has continued to spread throughout Asia and to the United States,

Canada, Mexico, Russia, France, Italy, Brazil, Chile, Peru, and Mozambique (59). The

serologic variants that have arisen since O3:K6 do not appear to have the same capacity to

spread or a propensity for causing as severe infections with hospitalization (59).

V. vulnificus

V. vulnificus causes primary septicemia and wound infection and is responsible for more than

90% of deaths due to vibrios in the United States yearly. Primary septicemia has a fatality

rate exceeding 50%, even with hospitalization, and occurs predominantly in men over 50

years old (14, 34). Patients usually have predisposing conditions such as liver disease,

immunosuppression, increased serum iron, or other chronic diseases (8, 34). CDC data

indicate that >95% of patients consumed raw oysters within 7 days of their infection.

Patients typically present with a sudden onset of fever and chills, vomiting, diarrhea, and

abdominal pain. Secondary skin lesions often appear, progressing to bulla formation and

necrosis. Endotoxic shock often occurs and can rapidly lead to death. Both blood cultures and

biopsy samples (scrapings) from skin lesions are usually positive. V. vulnificus also causes

severe wound infections, usually after trauma and exposure to marine animals or the marine

environment (34). Wound infections may progress to cellulitis with extensive necrosis (often

requiring surgical debridement), myositis, and necrotizing fasciitis that may mimic gas

gangrene, and to secondary septicemia. The fatality rate for wound infections ranges from

20 to 30%. Three biogroups have now been defined for V. vulnificus (34). Most infections in

the United States are due to biogroup 1; biogroup 2 has been principally isolated from

diseased eels and also isolated from one human wound infection. V. vulnificus biogroup 3

was described in 1999 and has been limited to wound and blood-borne disease in Israeli

patients exposed to live tilapia grown in aquaculture.

V. alginolyticus

V. alginolyticus is very common in the marine ecosystem and is the fourth most commonly

isolated Vibriospecies in the United States. It is most frequently isolated from ear and wound

infections following seawater exposure. A 2006 European surveillance report detected three

cases in which V. alginolyticus was isolated (synovial fluid, hand wound, and otitis) in

persons who swam in an inlet of the North Sea (70). V. alginolyticushas also been isolated

from ocular infections and from infrequent cases of monomicrobial or polymicrobial

bacteremia, mostly in immunocompromised persons (15, 45). It is occasionally isolated from

diarrheal stool, but there is no evidence that it actually causes diarrhea (82).

Photobacterium damselae (V. damsela)

P. damselae is an aggressive marine pathogen causing serious life-threatening illnesses.

Fatality rates for P. damselae are unknown, but many reports in the literature describe fatal

infections, suggesting a fairly high attributable mortality rate (89). Disease syndromes

associated with this bacterium include soft tissue infections (cellulitis and necrotizing

fasciitis) and bacteremia. Most wound infections develop as indolent processes that progress

to more severe disease within a matter of hours, and vibrios are often not suspected as part

of the initial diagnosis. Medical intervention, in addition to antibiotics, is often required,

including irrigation, fasciotomy, debridement, and sometimes amputation.

Typically, P. damselae wound infections occur in fishermen and result from penetrating

traumas caused by fish fins, fish hooks, or harpoons (29, 60, 89). More recently, however,

other sources of infection have been reported for this organism. These include a case of

cellulitis in a healthy teenage surfer who sustained a laceration to his hand from his

surfboard, a 30-month-old child with sickle cell anemia who developed bacteremia after

handling fish and then scratching an open wound on her buttock, and a urinary tract

infection in a pregnant female with increased frequency of urination and dysuria who had

sexual intercourse in the Caribbean Sea 1 week prior to infection (2, 6, 41).

V. fluvialis and V. furnissii

V. fluvialis appears to cause sporadic cases of diarrhea worldwide, with severe cases of

gastroenteritis sometimes linked to bacteremia or associated with cholera-like symptoms

(5, 44). Based upon published reports, there appears to be a small but increasing incidence

of extraintestinal V. fluvialis infections. These include acute infectious and continuous

ambulatory peritoneal dialysis-associated peritonitis, soft tissue infections (cellulitis)

associated with cerebritis, and bacteremia (33, 44, 46, 67). For many of these systemic

infections, some of which have poor outcomes, vibrios are, again, not initially suspected as a

cause. They are eventually associated with seafood consumption or exposure to seawater. V.

furnissii is rarely isolated from human clinical specimens, but when it is recovered, it is

invariably from fecal specimens of patients with diarrhea (23). There is no convincing

evidence that it causes diarrhea.

Miscellaneous Vibrios and Vibrio-Like Organisms

V. harveyi (V. carchariae) is an important pathogen of marine fish and invertebrate species.

To date, there are only two confirmed cases of human infection attributed to V. harveyi. The

first report was of a wound infection resulting from a shark bite (66). The second report

involved a 9-year-old boy with anaplastic large cell lymphoma and central-line sepsis who,

after completing chemotherapy and autologous stem cell transplantation, developed a febrile

episode after swimming in the Mediterranean Sea (85). There is an anecdotal report of two

isolates from blood and gallbladder (histories unavailable) that were retrospectively identified

by rpoB sequencing (77). Grimontia hollisae is a halophilic, vibrio-like species primarily

associated with moderate to severe cases of diarrhea, sometimes involving hypovolemic

shock (31). Most recorded cases of infection involve a history of consumption of seafood,

such as oysters. V. metschnikovii is frequently isolated from freshwater and brackish and

marine waters. It was first reported to cause peritonitis and bacteremia in a patient with an

inflamed gallbladder. Subsequently, it has been isolated from additional patients with

bacteremia and, rarely, from wound infections; it has also been reported from cases of

cholecystitis, diarrhea, and pneumonia (47, 84). V. cincinnatiensis was first reported from a

patient with bacteremia and meningitis. Subsequent isolates have been from the stool of a

person with diarrhea, from aborted bovine fetuses, and from mussels (87).

COLLECTION, TRANSPORT, AND STORAGE OF

SPECIMENS Back to top

Pertinent clinical history (when known) should accompany specimens to alert the laboratory

to include appropriate isolation media for the Vibrionaceae in their stool workup; this is

especially important in areas where the isolation of vibrios is infrequent (50). Helpful

information includes history of travel, consumption of seafood, activity associated with

marine or brackish water or wounds associated with such exposure, and hobbies associated

with aquaria.

Vibrionaceae, like other enteric organisms, are particularly susceptible to desiccation, so

stool specimens that cannot be inoculated onto plating media within 2 to 4 h should be

placed in a transport medium. Cary-Blair or any noninhibitory transport medium that does

not contain glycerol is acceptable for vibrios; buffered glycerol in saline is unacceptable

because some lots of glycerol may be toxic to vibrios. For specimens collected in the field, if

necessary, liquid stool may be placed on strips of blotting paper or gauze, then inserted in

airtight plastic bags with a few drops of saline to maintain moisture, and then sent to the

laboratory. Detailed information on the collection and transport of specimens for vibrio

isolation is available elsewhere (12). Special methods for the collection and processing of

extraintestinal specimens (blood, wounds, etc.) for vibrio isolation are not required, as

vibrios are, as a rule, isolated in pure culture from these sites and the concentration of salt in

primary plating media is usually sufficient for their recovery. Upon isolation, however, salt

may need to be added to subsequent media to attain growth of salt-requiring vibrios.

Vibrionaceae may die within weeks in vitro, even in moist environments at room

temperature, and should be maintained at −70°C as directed in chapter 9.

DIRECT EXAMINATION Back to top

Direct microscopic detection of vibrios in stool is not routinely recommended, since it may

not be possible to distinguish pathogenic vibrios from other members of the enteric

microbiota.

Direct Detection of V. cholerae O1 in Stool

Direct detection of V. cholerae from stool requires experience to correctly interpret results

and is typically done only in laboratories where cholera is common or in field situations

where laboratory services are unavailable and rapid diagnosis is required. One of the oldest

assays, the microscopic immobilization test, detects loss of motility of V. cholerae O1

organisms by the addition of O1 antibody and can be used to detect V. cholerae O139 by

using O139 antibody. A direct fluorescent-antibody test, Cholera and Bengal SMART, and a

membrane antigen rapid test, SMART Cholera O1, are available from New Horizons

Diagnostics Corp., Columbia, MD (FDA approval pending), and a V. cholerae O1 latex

agglutination assay (Denka Seiken, Tokyo, Japan; not FDA approved for human clinical

specimens) is commercially available.

Molecular Detection in Clinical Specimens

Publications on molecular methodologies for detection of V. cholerae, V.

parahaemolyticus, and V. vulnificus in stool can be found in the literature starting in the

1990s and are too numerous to cover here. Most of the advantages of PCR-based assays

over culture methods apply to vibrios and include the ability to freeze stools for

epidemiological studies for delayed testing. Also, in areas where vibrios are rarely isolated

and laboratory experience or resources are limited, molecular methods may be more reliable

and may provide a shorter turnaround time. Inhibitors in stool may affect the analytical

sensitivity of an assay of PCR but can be overcome by dilution of the sample. Current PCR

assays for V. cholerae use a multiplex PCR that includes primers for the ctx gene, rfb genes

(O antigens, O1, and O139, allowing differentiation of serotypes), and tcpAgenes (specific for

the El Tor and classical biotypes, useful for epidemiological purposes) (32, 38). For clinical

strains of V. parahaemolyticus, the thermostable direct hemolysin, a major virulence

determinant, is a common target for detection, but it is not universally present in strains

isolated from human infections (9, 72). In reality, other than for surveys, in areas where V.

cholerae is seldom isolated or in noncoastal areas where V. parahaemolyticus is an

infrequent pathogen, the use of PCR is impractical and costly by comparison to alkaline

peptone water and TCBS medium.

ISOLATION PROCEDURES Back to top

Vibrionaceae associated with human disease can be isolated from routine enteric media, but

recovery is enhanced when specific media are used. On MacConkey or salmonella- shigella

agar, vibrios present as colorless colonies (with the exception of the lactose-fermenting

species, V. vulnificus). On sucrose-containing media such as Hektoen and xylose-lysinedeoxycholate,

sucrose-positive vibrios associated with human disease such as V. cholerae, V.

fluvialis, V. alginolyticus, and some strains of V. vulnificus cannot be differentiated from

normal enteric biotas that ferment sucrose. The addition of a blood agar plate allows colonies

to be screened for oxidase, which may improve recovery of vibrios as well as colonies

of Aeromonasspp. and Plesiomonas shigelloides. G. hollisae may grow poorly or not at all on

any enteric isolation medium, including TCBS; it is probably most reliably isolated from blood

agar.

TCBS agar is formulated specifically for the isolation of vibrios (25). Both powdered

formulations and prepared plates are readily available from a number of commercial sources.

Autoclaving is not required, so powdered media may be kept available in the laboratory and

easily prepared by boiling as needed. Inclusion of sucrose allows for preliminary

differentiation of Vibrio species, with V. cholerae, V. fluvialis, and V. alginolyticus producing

yellow colonies while V. parahaemolyticus, V. mimicus, and most strains of V.

vulnificus produce green colonies (sucrose not fermented). It should be noted that yellow

colonies may convert to green if plates are examined after more than 24 h or are

refrigerated after incubation. Oxidase testing is unreliable when performed directly on

colonies growing on this medium. Growth from a non-sugar-containing medium such as

nutrient agar should be used for oxidase testing.

A chromogenic agar, CHROMagar Vibrio (CHROMagar Microbiology, Paris, France), has been

developed primarily for the recovery of V. parahaemolyticus from seafood and supports the

growth of other vibrios as well (30). Vibrio colonies on this medium range in color from milk

white to pale blue to violet. Other members of the enteric biota usually do not grow, with the

exception of Proteus mirabilis and Providencia rettgeri, which also produce milk white

colonies. Marine agar (BD Biosciences, Sparks, MD), which does not contain any inhibitory or

selective ingredients, may be more appropriate for isolation of vibrios from the environment,

especially salt-requiring vibrios, because of its high salt content.

It is common for pure cultures of vibrios to produce multiple colony morphologies (as many

as five) on any medium, but this phenomenon is most readily noticeable on nonselective

media such as blood or heart infusion agars. Variations in morphology include smooth,

rough, convex, flat, spreading, and compact in various combinations. Occasionally V.

cholerae produces rugose (extremely wrinkled) colonies on non-carbohydrate-containing

media (4). Like their smooth counterparts, they are fully virulent for humans (90). Rugosity,

which is due to production of a unique extracellular polysaccharide, confers biofilm formation

and resistance to chlorine, acid pH, and serum killing (4, 90). Although it is believed to

enhance survival in aquatic environments, to date it has been demonstrated only in vitro.

Classical biotypes of V. cholerae also possess the vps (vibrio polysaccharide synthesis) gene

cluster that encodes this phenotype, but rugose variants have been demonstrated only in El

Tor strains (90).

In acute diarrheal disease, stool enrichment is generally not required; however, when

enrichment is necessary, alkaline peptone water (1% NaCl, pH 8.5) is the most commonly

used enrichment broth for human specimens. It should be incubated at 36°C and

subcultured at 18 h. Occasionally, vibrios are recovered only after a shorter incubation (6 h),

and for these specimens longer incubation times fail to yield a vibrio, probably due to

overgrowth by other organisms (S. Abbott, personal observation).

IDENTIFICATION Back to top

Conventional Biochemicals

Biochemical properties that separate members of the Vibrionaceae from

the Enterobacteriaceae (includingPlesiomonas shigelloides) and the Aeromonadaceae are

found in Table 1, and biochemical profiles of the 12 species that occur in human clinical

specimens are given in Table 2. Generally, species are 0 to 10% positive for the following:

H2S in triple sugar iron, urea (except V. parahaemolyticus, 15%), phenylalanine deaminase

(except V. vulnificus biogroup 1, 35%), malonate, mucate, yellow pigment production, and

fermentation of D-adonitol, dulcitol, melibiose (except V. vulnificus biogroup 1, 40%),

raffinose, L-rhamnose (except V. furnissii,45%), D-sorbitol (except V. metschnikovii, 45%),

α-methyl-β-D-glucoside, and D-xylose (except for V. cincinnatiensis, 57 and 43%,

respectively). Variable reactions are seen with methyl red, growth in potassium cyanide

broth,D-galactose, glycerol, sodium acetate, DNase at 25°C, and lipase. All species are 99 to

100% positive for growth in 1% NaCl and fermentation of maltose (except G. hollisae, 0%)

and D-mannose (except V. cholerae, 78%, and V. harveyi, 50%). Many commercial standard

tube tests have sufficient salt to support growth without salt supplementation (0.5 to 1%),

but the Microbial Diseases Laboratory, California Department of Public Health, routinely adds

1% salt to all biochemicals (except for the 0% salt broth) for all NaCl-requiring species.

Voges-Proskauer, Moeller’s decarboxylases and dihydrolase, and nitrate broth may contain

no or insufficient NaCl to support growth of some NaCl-requiring strains, and these

biochemicals should always have salt added to them (to a final concentration of 1%; for

occasional strains, 3%) when these species are tested.

It should be noted that many V. cholerae O1 strains from Bangladesh and surrounding areas

and all strains ofV. cholerae O139 are resistant to both 10- and 150-μg disks of the

vibriostatic compound O/129 (Remel, Lenexa, KS), a classical test used to distinguish vibrios

(Table 1). In areas of the world where cholera is uncommon, complete biochemical testing

should be performed and all cultures identified as V. cholerae should be sent to public health

laboratories for O1 and O139 agglutination and cholera toxin testing. V. cholerae O1 isolates

should be biotyped to determine whether they are the El Tor or classical biotype. These

biotypes can be differentiated by a number of phenotypic tests, including hemolysis of sheep

erythrocytes, production of acetylmethylcarbinol (Voges-Proskauer test), and resistance to

polymyxin B, all positive for the El Tor biotype. Except for the O serotype and O/129

reaction, V. cholerae O139 strains are phenotypically similar to V. choleraeO1 El Tor. Strains

of V. cholerae that fail to agglutinate in either O1 or O139 antiserum are reported as V.

cholerae non-O1. Serotyping for non-O1 strains is available only in a limited number of

reference laboratories and is rarely complete. V. cholerae and V. mimicus are separated from

other species by growth in media lacking NaCl (Difco, BD BioSciences, Nutrient Broth is the

only broth that accurately determines NaCl requirement). Strains of V. parahaemolyticus, V.

alginolyticus, and P. damselae may be urea positive. Most vibrios isolated from humans

produce a buff or tan pigment; however, strains of V. parahaemolyticus may produce a dark

brown pigment. G. hollisae generally grows poorly, especially in Moeller’s decarboxylases and

dihydrolase broths, even after salt supplementation and produces extremely large zones of

inhibition, often necessitating the use of two plates when disk antimicrobial susceptibility

testing is performed. V. metschnikoviiis distinctive because it is an oxidase- and nitratenegative

vibrio. V. fluvialis and V. furnissii are frequently confused with Aeromonas

caviae, especially as some strains are weakly halophilic and only moderately susceptible to

O/129, and because some strains of A. caviae grow on TCBS agar. V. furnissii is the only

vibrio isolated from humans that is positive for gas production from D-glucose. Rapid, correct

identification of V. vulnificus strains is critical because of the mortality associated with this

organism. Some strains of V. vulnificusare sucrose positive, which may add to the confusion

in identifying it.

Commercial Systems

There are no recent studies evaluating commercial systems’ ability to identify organisms

covered in this chapter; however, based on individual case reports, identification of these

organisms by commercial systems remains problematic (21, 58, 69). No commercial system

includes all 12 clinical species in its database, and some manual systems do not contain any

of these species (63, 64). The most recent evaluation indicates that when tested only against

those species listed in their databases, the API 20E (bioMerieux Inc., Durham, NC), Crystal

E/NF (BD Biosciences), MicroScan Neg ID type 2 and type 3 (Siemens Healthcare Systems,

West Sacramento, CA), and Vitek GNI+ and ID-GNB cards (bioMerieux) correctly identified

only 63 to 81% of these organisms to the species level (64). Correct identification of the

three most commonly isolated species of Vibriofrom clinical samples varied. For V.

cholerae, API 20E gave the least (50%) and Crystal the most (97%) accurate identification.

For V. parahaemolyticus, Rapid Neg ID3 fared the worst (40%) and API 20E and GNI+ the

best (97% each), while for V. vulnificus biogroup 1 strains, GNI+ (50%) and Crystal E/NF

(97%) gave the lowest and highest rates (64). Only Crystal E/NF was able to correctly

identify ≥90% of V. cholerae or V. vulnificus strains, and only API 20E and the two Vitek

cards correctly identified ≥90% of V. parahaemolyticusstrains. In another study using only V.

vulnificus biotype 3 strains, MicroScan (98%) and Phoenix (90%) systems did the best in

identifying 51 well-characterized isolates to the correct species, while the identification rate

on Vitek (13.7%) was much less satisfactory (19). Croci et al. (21) found that for the

identification of V. parahaemolyticus, API 20NE exhibited greater sensitivity than API 20E (20

versus 16 of 27) but that API 20E was more specific (100% versus 82%). In another study

on a mixture of 111 clinical and environmental V. vulnificus strains, API 20E, API 20NE, and

Biolog (Biolog, Inc., Hayward, CA) correctly identified 60, 0, and 85%, respectively. Of the

above products, only Biolog is not FDA approved for use on clinical isolates (69).

Manufacturers’ instructions should be checked prior to testing of salt-requiring vibrios to

determine if salt supplementation is required. A recent publication indicated that API 20E

gave incorrect identifications for clinical isolates of V. parahaemolyticus if 2.0% NaCl was

used for the diluent, versus 0.85% (51). For clinical isolates of V. alginolyticus, either

concentration yielded identifications with high probabilities, but for V. vulnificus, the 0.85%

diluent failed to identify 1 of 2 isolates, whereas the 2.0% NaCl identified both strains with

≥94% probabilities.

Molecular Methods

Molecular identification of vibrios is commonplace in surveys and in research studies.

However, it is not commonly employed in clinical laboratories for routine identification

because vibrios are relatively rare pathogens in noncoastal areas or regions where cholera is

not endemic. Additionally, few, if any, commercial molecular products are FDA approved for

human clinical specimens, and insufficient strains are available in most laboratories to

validate in-house molecular methods in compliance with Clinical Laboratory Improvement

Amendments regulations. The use of 16S rDNA sequencing alone is less than ideal

for Vibrionaceaeidentification, as interspecies sequence differences are very small (1 to 6%)

and polymorphism has been shown to be fairly common in 16S rRNA genes (55). Tarr et al.

(77) developed a multiplex PCR assay, using primers directed at V. cholerae sodB, V.

mimicus sodB, V. parahaemolyticus flaE, and V. vulnificus hsp genes, that correctly identified

109 isolates and found an additional 4 strains of V. parahaemolyticus that either had not

been identified to species level (n = 3) or had been identified incorrectly as V.

alginolyticus (n = 1). Additionally,rpoB gene sequencing was used to identify 12 of 15

isolates not previously identified to species level by biochemical methods. In other studies

the toxR gene (V. parahaemolyticus), vvhA (hemolysin, V. vulnificus),vcggenes (virulencecorrelated

gene, V. cholerae), wbe and wbf genes (for the O1 and O139 serotypes,

respectively), and tcpA genes (specific for the El Tor and classical biotypes) have been used

for identification (21, 38, 69, 91). Matrix-assisted laser desorption/ionization time-of-flight

mass spectrometry promises to be a valuable new tool for identification of closely related

bacterial species, with a rapid turnaround time and modest test cost. At this time there are

no published studies with vibrios, but future studies should validate the usefulness of this

methodology.

V. cholerae and V. parahaemolyticus Toxin Detection

In reference laboratories, cholera toxin was traditionally detected by fluid accumulation in

animal assays or detection of a cytopathic effect in Y1 adrenal or Chinese hamster ovary cell

cultures. A reverse passive latex agglutination assay, VET-RPLA (Denka Seiken), which

detects both cholera toxin and the heat-labile toxin ofEscherichia coli, is commercially

available. The majority of human strains of V. parahaemolyticus produce a thermostable

direct hemolysin encoded by two genes, tdh and tdh2x. These toxins are rarely produced by

environmental strains of V. parahaemolyticus but have been detected in V. cholerae non-

O1, V. mimicus, and G. hollisae strains. Like cholera toxin, thermostable direct hemolysin

can be detected by a commercial latex assay (KAP-RPLA; Denka Seiken), but there are no

commercial products that detect the thermostable related hemolysin seen in V.

parahaemolyticus strains. PCR assays for both hemolysins have been developed but are not

commercially available (24).

TYPING SYSTEMS Back to top

Serotyping

Serotyping is the most widely utilized conventional procedure. Typing schemes have been

described for a number of species, including V. cholerae, V. parahaemolyticus, and V.

vulnificus; however, commercial-grade typing sera are available only for V. cholerae and V.

parahaemolyticus (73, 74). V. cholerae O1 (polyclonal, Inaba and Ogawa) and O139 antisera

are available from BD Biosciences, Denka Seiken, Remel, and New Horizons. V.

parahaemolyticus antisera (11 O groups and 9 polyvalent and 65 monovalent K groups) are

available from Denka Seiken.

Molecular Typing of Vibrios

Because it is well standardized, pulsed-field gel electrophoresis using NotI and SfiI enzymes

is probably the “gold standard” for molecular typing of vibrios (36, 80, 90). Ribotyping

appears to be less discriminatory (13,69). Kotetishvili et al. (43) found multilocus sequence

typing more discriminatory than pulsed-field gel electrophoresis for V. cholerae. Multilocus

sequence typing, arbitrarily primed PCR, group-specific PCR of thetoxRS gene, and PCR for

the orf8 sequence of phage f237 (the last two are species specific) have been used with

success for V. parahaemolyticus, particularly for serovariants of the O3:K6 clone (9, 59, 72).

Likewise, forV. vulnificus, repetitive extragenic palindromic PCR is an effective subtyping

method (34). Sequencing of genes such as ctxA, ctxB, hsp60, and recA has also been used

for molecular typing of Vibrionaceae (37).

SEROLOGIC TESTS Back to top

Reagents for the serodiagnosis of cholera are available only in specialized reference

laboratories, but titration of acute- and convalescent-phase sera in agglutination, vibriocidal,

or antitoxin tests is extremely reliable (49).

ANTIMICROBIAL SUSCEPTIBILITIES Back to top

Generally, for most V. cholerae and V. parahaemolyticus gastrointestinal infections,

treatment by rehydration is recommended over antimicrobial therapy and has the added

benefit of reducing the risk of antibiotic resistance. Antimicrobial therapy, however, can

reduce the duration of diarrhea, shedding of the organism, and the volume of rehydration

fluids needed for recovery, and patients are often treated before culture results are known

(80). Clinical and Laboratory Standards Institute (CLSI) document M45-A includes

susceptibility testing guidelines for noncholera vibrios (17). The CLSI interpretive guidelines

are limited to ampicillin, tetracyclines, folate pathway inhibitors, and chloramphenicol for V.

cholerae (18). In vitro susceptibility surveys show V. cholerae strains to be susceptible

(>90%) to aminoglycosides, azithromycin, fluoroquinolones, extended-spectrum

cephalosporins, carbapenems, and monobactams (71, 88). However, a conjugative

transposon designated the SXT element and carrying resistance to sulfamethoxazole,

trimethoprim, chloramphenicol, and streptomycin emerged first in V. cholerae O139 strains

in India and is now seen in O1, non-O1, and O139 cholera strains and V. fluvialis (1). Ahmed

et al. (1) have reported on a V. fluvialis isolate with the SXT element and a novel

aminoglycoside acetyltransferase gene encoding resistance to gentamicin; resistance to

ampicillin, furazolidone, and nalidixic acid was also reported. Multidrug-resistant strains of V.

fluvialis have also been noted among other strains from India (13). V. parahaemolyticus is

generally susceptible to most antibiotics used for traveler’s diarrhea (72). The

fluoroquinolones alone or the synergistic combination of ciprofloxacin and cefotaxime shows

excellent in vitro activity against V. vulnificusstrains (39, 76).

EVALUATION, INTERPRETATION, AND REPORTING OF

RESULTS Back to top

Isolation of V. cholerae O1 or O139 should be reported immediately to the attending

physician because of the severe dehydration that cholera can produce. The case should also

be phoned to public health authorities and the isolate sent to a public health laboratory for

confirmation and toxin testing.

When vibrios are isolated from blood or cerebrospinal fluid (bacteremia and meningitis are

associated with high mortality rates) or wound infections which cause extensive tissue

damage (V. vulnificus and P. damselae),the results should also be phoned immediately to the

attending physician so that rapid and appropriate antibiotic therapy can be initiated. This is

especially true for V. vulnificus infections, which have a high mortality rate without rapid,

appropriate intervention. The clinical significance of Vibrio strains (V. mimicus, V.

alginolyticus, G. hollisae, V. harveyi, and V. metschnikovii) in other specimens, particularly

stool, may be more difficult to determine and requires prompt consultation with the

attending physician to better understand the clinical context. Most physicians are not familiar

with many Vibrionaceae species, and a phone consultation would be mutually beneficial to

the clinician and laboratory provider. Information helpful to the physician would include the

presence or absence of other pathogens and the relative amount of growth (pure or almost

pure culture) of the vibrio. Vibrionaceae isolates should also be submitted to public health

laboratories, as they are monitored under the CDC’s International Emerging Infections

Program and Vibrio Surveillance System; they may also be needed for confirmation and toxin

testing.

Vibrionaceae species that are known to cause diarrhea should be considered clinically

significant, particularly if they are present in large numbers and no other potential pathogens

are present. Isolation of vibrios from stool in small numbers may only reflect transitory

colonization; however, species such as V. cholerae, V. mimicus, and V.

parahaemolyticus have documented virulence factors that correlate with their ability to cause

intestinal infections. Laboratory tests helpful in determining pathogenic potential are

primarily available only in reference laboratories. Vibrionaceae isolation from locations such

as the ears may represent infection, transient colonization, or merely their presence after

exposure to seawater. Again, isolation of Vibrionaceaerequires prompt consultation with the

clinician to better understand the clinical context that can help direct the need for further

laboratory investigations.

Finally, readers should also be cautioned that misidentification of Vibrio species and their

relatives can be a problem in the literature unless investigators used methods that are verysensitive in differentiating all of the species in the family Vibrionaceae (25).