Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

number of factors, especially the climatic conditions.

Intensive livestock production has created problems

of excreta disposal. The number of animals per unit

area of land has vastly increased in many countries

and it has been necessary to increase application

rates of animal waste on to the land. One survey

found that 10% of cattle slurry samples contained

Salmonella, with survival periods ranging from 11

to 12 weeks.

0025 Biological treatment of human sewage removes

most of the suspended and dissolved organic matter,

but it does not necessarily remove pathogenic bac-

teria. Where sewage sludge is used as a fertilizer,

many samples have been found to be contaminated

with Salmonella. There are numerous reports of the

isolation of Salmonella from rivers and streams and

of their pollution by sewage effluent and slurry run-

off. Further spread from contaminated environments

may also take place by wild animals and birds.

Transmission to Humans

0026 Typhoid fever usually follows the ingestion of drink-

ing water contaminated by human carriers of S. typhi

or food prepared or handled by such carriers who do

not take adequate hygienic precautions.

0027 In contrast, acute gastroenteritis (foodborne dis-

ease/poisoning) is usually acquired from consumption

of foods which may be directly or indirectly contam-

inated with Salmonella. A wide variety of animal

species have been shown to be capable of harboring

the organisms, and in the developed world turkeys,

chickens, swine, and cattle have often been found to

be infected in abattoir studies. Cross-contamination

may readily occur during transport to the abattoir

and during slaughter. Modern poultry-processing

plants are designed with emphasis on speed, and

during the slaughtering process there are many points

where cross-contamination may occur. Recent

surveys have found that up to 30% of chicken car-

casses may be contaminated with Salmonella. Table

eggs have been found to be an important vehicle for

the transmission of S. enteritidis to humans and

undercooked or raw eggs or dishes made from them,

such as mayonnaise or mousse, have been responsible

for many outbreaks of food poisoning.

0028 The progressive trend toward mass processing and

distribution of food products has been an important

factor in the increased incidence of Salmonella food-

borne disease. Milk has also been shown to be an

important source of infection and has been associated

with massive outbreaks of human salmonellosis.

Most outbreaks have been associated with the con-

sumption of raw milk, although pasteurization plant

failure has also resulted in large outbreaks. Salads and

other vegetables which have been contaminated in

washing or ‘freshening’ with polluted water have

also been shown to be associated with human sal-

monellosis. Many other products, e.g., beansprouts,

chocolate, have been found to be the source of Sal-

monella infection and dried milk became contamin-

ated with Salmonella when the production machinery

was contaminated. Whatever the source, many out-

breaks occur as a consequence of the delay between

preparation/cooking and consumption of the food,

which allows the numbers of Salmonella to increase

rapidly. The important food-associated risk factors

are inadequate cooking or reheating, storage at

inappropriate temperatures, and cross-contamination

from raw to cooked food.

0029Person-to-person spread has been demonstrated on

many occasions and may be of importance in young

children and groups living under poor socioeconomic

conditions where effective sanitation is lacking. In

these situations, flies may serve as vectors for Salmon-

ella transmission. Dog and wild animal feces on

footwear have been implicated as a route by which

infection can be carried into households. Person-

to-person spread may also occur in hospitals, nursing

homes, and mental institutions, where a number of

large outbreaks have occurred. Amplification of in-

fection in these institutions may occur from contam-

inated food or from asymptomatic carriers; babies are

at special risk. Likewise, contaminated equipment

has been shown to be vehicles of infections.

Causation of Disease

0030Salmonella cause disease by invading the intestinal

mucosa and multiplying in the gut-associated lymph-

oid tissue. From the infected tissue the organisms are

drained to the regional lymph nodes, where further

spread is usually prevented if the host defense

mechanisms are successful. In humans nontyphoidal

Salmonella serovars usually cause a localized disease

which is manifest as an acute gastroenteritis. If, on

the other hand, the host’s defense mechanisms are

unable to prevent the spread of the organism, sys-

temic disease results, as commonly occurs with

typhoid. The etiology of diarrheal disease is poorly

understood, and although several putative toxins,

e.g., enterotoxins, cytotoxins, have been described,

their significance is far from clear. (See Salmonella:

Salmonellosis.)

0031Whether infection follows exposure to any patho-

genic microorganisms depends in part on the number

of organisms that enter the host’s tissues. From vol-

unteer studies in humans, it has been suggested that

large inocula of Salmonella are necessary to cause

illness. However, investigations of outbreaks of Sal-

monella foodborne disease have suggested that the

SALMONELLA

/Properties and Occurrence 5077

infective dose is often low; in six of the 11 outbreaks

the actual dose ingested was calculated as < 10

organisms. However, many factors may affect the

infective dose, not least of which are strain variations,

e.g., S. choleraesuis, although a rare cause of human

infection, may cause serious disease. Children and

older people are more susceptible; other concurrent

illnesses (e.g., acquired immune deficiency syndrome

(AIDS)) may increase susceptibility. The use of anti-

biotics for other indications may predispose to food-

borne salmonellosis, even when the Salmonella are

sensitive to that antibiotic. There is a fivefold increase

in risk when the Salmonella are resistant.

Prevention of Transmission

0032 The overall quality of meat and edible byproducts

depends heavily on their microbiological quality.

This is very dependent on the contamination level of

the production equipment, as well as on the skin

and intestinal microflora of animals and on the

technology and techniques used in slaughtering and

processing. The fecal microflora of animals is the

predominant source of pathogens on the final dressed

carcass, and the modern slaughter line offers many

risks of cross-contamination. In general, such cross-

contamination is much less pronounced with car-

casses of beef and sheep than with pork and poultry

carcasses. A system to insure quality and safety must

be designed for the entire chain of operations, from

the farm to the consumer. This system must include

the identification of potential microbiological prob-

lems and control measures at all stages of the entire

meat production chain, as well as the development of

Hazard Analysis Critical Control Point procedures.

Although the hazards may differ from country to

country and vary with animal health, slaughter

procedures, type of meat products, and storage

and distribution, the main areas of focus are the

slaughter process itself and disinfection of tools and

equipment.

0033 Pasteurization will prevent milk-borne salmonel-

losis, although outbreaks have occurred on a number

of occasions when faults have developed in plants.

One of the most effective end-product treatments is

gamma-irradiation, which can destroy Salmonella

on poultry carcasses and prolong shelf-life, although

possible changes in flavor and texture may occur.

There may be some consumer resistance toward the

use of nuclear energy.

0034 Equally important is the education of caterers and

domestic consumers in all aspects of hygienic prepar-

ation of food. Cross-contamination may occur

readily in kitchens, especially when raw and cooked

foods are prepared on the same surfaces.

Public Health Aspects

0035In many countries the results of tests which identify

the presence of Salmonella in animals have to be

reported to the appropriate government department

and, in addition, separate orders may relate to animal

feeds.

0036Likewise, on the human side, it may be necessary to

report if a patient is suffering from or is suspected of

having foodborne disease. Following the reporting of

Salmonella incidents in both humans and animals,

further investigations may be carried out. Regulations

may also apply to the production and sale of food.

Economic and Social Impact

0037The financial costs of Salmonella infections are not

only associated with investigations, treatment, and

prevention of human illness; they may also affect the

whole food production chain.

0038In the public sector, resources are necessary for

surveillance, treatment of patients, and investigations

of outbreaks. Financial burdens may be imposed on

industry, especially the food industry. Likewise, the

cost to affected individuals may be extremely high,

especially should death result from Salmonella infec-

tions. The costs associated with Salmonellosis may be

subdivided into tangible costs, which can be meas-

ured in monetary terms (e.g., medical costs and loss of

work), and intangible costs, such as loss of leisure,

pain, discomfort, etc., which cannot be readily

costed. A study of infectious intestinal disease in

England between 1993 and 1996 suggested that the

true rate of salmonellosis in the community was 220/

100 000 population and the estimated cost per human

case was £660.

0039An analysis of five Salmonella outbreaks in North

America gave direct costs which ranged from $36 400

to $62 million. In the UK, an outbreak of S. ealing

infection associated with powdered milk was esti-

mated to have cost £14.6 million in the late 1980s.

0040There have been few studies into the costs and

benefits of preventing Salmonella infection, but it

has been suggested that for every £1 spent on investi-

gation and curtailment of an outbreak, there is a

saving of £5.

See also: Food Poisoning: Classification; Salmonella:

Detection; Salmonellosis

Further Reading

Doyle MP (ed.) (1989) Foodborne Bacterial Pathogens.

New York: Marcel Dekker.

Ewing WH (ed.) (1986) Edwards and Ewing’s Identifica-

tion of Enterobacteriaceae, 4th edn. New York: Elsevier.

5078

SALMONELLA

/Properties and Occurrence

Guinee PAM and van Leeuwen WY (1987) Phage typing

of Salmonella. In: Methods in Microbiology, vol. 11,

pp. 158–190. New York: Academic Press.

Le Minor L and Popoff MY (1987) Request for an opinion.

Designation of Salmonella enterica sp. nov. as the type

and only species of the genus Salmonella. International

Journal of Systemic Bacteria 37: 465–468.

Reeves MW, Ewins GM, Heiba AA et al. (1989) Clonal

nature of Salmonella typhi and its genetic relatedness

to other salmonellae as shown by multilocus enzyme

electrophoresis and proposal of Salmonella bongori

comb. nov. Journal of Clinical Microbiology. 27:

313–320.

Saeed AM (1999) Salmonella enterica Serovar enteritidis in

Humans and Animals: Epidemiology, Pathogenesis and

Control. Ames, Iowa: Iowa State University Press.

Varnam AH and Evans MG (1991) Foodborne Pathogens:

An Illustrated Text. London: Wolfe.

WHO Collaborating Centre for Reference and Research on

Salmonella (1997) Antigenic formulas of the Salmonella

Serovars. Paris: Institut Pasteur.

Wray C and Wray A (eds) (2000) Salmonella in Domestic

Animals. Wallingford, Oxon: CABI Publishing.

Detection

T Humphrey, PHLS Food Microbiology Research Unit,

Heavitree Exeter, Devon, UK

P Stephens, Oxoid Limited, Basingstoke, Hampshire,

Detection

0001 In the developed world, the majority of cases of

human salmonellosis is zoonotic and results from

the colonization/infection of food animals and result-

ant contamination of products derived from them.

Salmonella are also ubiquitous, and a diverse range

of foodstuffs – including dried milk, fermented saus-

age, beansprouts, chocolate, and yeast-based flavoring

– have been implicated as vehicles for human salmon-

ellosis. In the UK, and elsewhere in the last decade,

most cases or outbreaks of Salmonella infection have

resulted from the consumption of contaminated eggs

or poultry meat. Natural water systems, particularly

those receiving human and animal waste, are also

frequently contaminated.

0002 The ubiquity of Salmonella and the wide range of

materials with which they have been associated have

meant that numerous media and techniques have

been developed for their isolation. It is not the inten-

tion in this chapter to go into detail on the possible

virtues or disadvantages of individual methods.

Attention is focused instead on the fundamental

aspects of the isolation and detection of Salmonella,

particularly from food and environmental samples.

Special emphasis is given to sublethal injury and its

impact on recovery of the bacteria and the possible

inhibition by other organisms. The application of

rapid methods is also discussed. (See Food Poisoning:

Tracing Origins and Testing.)

Sublethal Injury and its Impact on Detection

0003Salmonella are primarily enteric organisms and,

whether in the gut or on contaminated foodstuffs,

they are usually present as part of a mixed microbial

population. Exceptions to this are seen with invasive,

host-specific strains such as S. typhi in humans and

S. enteritidis in poultry. The latter organism, for

example, can be isolated in pure culture from both

the tissues of infected chickens and the contents of

intact shell eggs. It is generally necessary, however, to

apply techniques which selectively encourage the

growth of Salmonella and to use agars which permit

the organism to produce characteristic colonies. The

similarity in antibiotic resistance patterns between

Salmonella and other enterobacteria has largely pre-

cluded the use of antibiotics and selection has relied

upon the use of chemicals such as selenite, tetra-

thionate, and deoxycholate.

0004When microorganisms, particularly Gram-negative

bacteria, are exposed to unfavorable conditions such

as high or low temperature, they may become sub-

lethally injured. This can be defined as ‘a sensitivity to

either selective agents or conditions to which normal

cells are resistant.’ In Salmonella, injury can be mani-

fested by, for example, lowered resistance to the chem-

icals mentioned above. The organisms may also be less

able to grow, particularly in selective media, at ele-

vated incubation temperatures, such as 43



C, which

are used to improve the selectivity of isolation media

when samples like sewage are examined.

0005Techniques used routinely for clinical specimens

when large numbers of the target pathogen may be

present and where it is relatively unaffected by its

environment are not suitable for use with either food

or environmental samples. These might have been

exposed to conditions damaging to bacteria and

they may also contain only low numbers of Salmon-

ella. The direct inoculation of food or environmental

samples into selenite broth or another selective

medium may lead to false-negative results.

0006The principal site of injury in the bacterial cell is

the outer membrane, and alterations in permeability

allow the ingress of selective agents that are normally

excluded. If cells are exposed to more extreme condi-

tions, damage can also be caused to the chromosome

and ribosomes. Much work remains to be done on the

SALMONELLA

/Detection 5079

response of Salmonella to adverse conditions and on

the full impact of the exposure of these bacteria to

different environments and their subsequent ability

to grow in culture media. Salmonella, like other

bacterial cells, are able to repair the lesions causing

sublethal injury, either partially or completely, if they

are incubated in a nonselective medium at 37



C. Two

media, buffered peptone water (BPW) and lactose

broth, are the most commonly used, with the buffer-

ing capacity of BPW giving it an advantage over the

other medium – particularly when foods containing

fermentable substrates are examined.

0007 Studies have demonstrated that the isolation rate

from selective broth cultures is greatly improved if

they are inoculated with an actively growing popula-

tion of Salmonella. It is thus essential that initial

incubation in BPW allows the organism both to over-

come the effects of sublethal injury and to multiply.

0008 The period of recovery or repair is often referred to

as preenrichment. Incubation is usually for 18–24 h,

although some workers have reported improved isol-

ation rates when preenrichment is extended to 72 h.

This is dependent on the material under investigation.

Salmonella do not appear to be able to compete well

with some other bacteria. Recovery media, by neces-

sity, are nonselective and permit the growth of the

majority of aerobic and facultative organisms present

in the sample. Where this encourages the prolifer-

ation of either Proteus or Pseudomonas spp. – as

with liquid raw egg and chicken meat – or lactobacilli

– as with dairy products – Salmonella may be

inhibited. This can be particularly pronounced when

incubation is prolonged (Table 1).

0009 The addition to BPW of novobiocin and cefsulodin,

which inhibit Proteus or Pseudomonas spp., respect-

ively, but have little effect on the growth of Salmon-

ella, was found to improve the isolation rate of

Salmonella from naturally contaminated samples of

raw liquid egg (Table 1). The recovery of Salmonella

from dried milk can also be improved if, after 2–4h

initial incubation, brilliant green at a final concen-

tration of 0.0002% is added to the preenrichment

culture.

0010The presence of any number of Salmonella in a

processed or cooked food is regarded as being poten-

tially hazardous and most microbiological guidelines

suggest an absence of the bacteria in 25 g of product.

There is also evidence that, with some foodstuffs,

particularly those rich in fat, and with certain vulner-

able individuals such as the very young or the elderly,

the infective doses of Salmonella may be low. Recent

studies have also demonstrated that sublethally

injured cells of Campylobacter jejuni are capable of

repair in the intestinal tract. Salmonella may also be

able to do this and it is important that sensitive tech-

niques are used for their detection in the laboratory.

In this respect, the importance of effective preenrich-

ment cannot be overemphasized. As a general rule,

it is prudent, when examining food and environmen-

tal samples from any source, to carry out primary

incubation in BPW at 37



C for a full 24 h. The ratio

of sample to medium should be at least 1:10 and,

where appropriate, the growth of competing flora

may be suppressed by the addition of antibiotics.

Selective Culture

0011The necessity for selective culture techniques in

microbiology has been recognized for over 100 years

and much of the early work was centered upon

Salmonella. Selective isolation is usually in two

stages: growth in an enrichment broth followed by

plating on a selective agar. Selective media make use

of either dyes or chemicals. A formulation much in

use in clinical laboratories is based on selenite. It is

believed that this compound inhibits bacteria either

by reaction with sulfhydryl groups or by the forma-

tion of analogs of sulfur amino acids. Selenite broth

has been used successfully for many years and modi-

fications to improve the isolation rate have included

the addition of cystine and the incorporation of fer-

mentable sugars, such as either mannitol or lactose,

which prevent increases in the pH. Selenite is less

selective at alkaline pH values.

0012Selenite can be used for direct inoculation with

specimens such as feces or used following preenrich-

ment. It is probably less suitable, however, for the

latter when compared to other media. The medium

can inhibit many bacterial types, but two important

exceptions – Proteus and Pseudomonas spp. – can

cause problems, particularly as they are capable of

preventing the growth of Salmonella and can also

resemble them on selective agars.

0013Salmonella have been shown to possess the enzyme

tetrathionate reductase, which is absent in many

tbl0001 Table 1 The influence of period of preenrichment and the

addition of novobiocin and cefsulodin on the isolation of

Salmonella from samples of liquid raw egg

Culture medium No. of Salmonella-positive

samples

after incubation

for

24 h 48 h

Egg/BPW only 32 14

Egg/BPW plus novobiocin (5 mgml

1

cefsulodin (10 mgml

1

)

74 70

110 samples were examined. Rappaport–Vassiliadis broth was used as

the selective medium.

BPW, buffered peptone water.

Data from Humphrey TJ and Whitehead A (1992).

5080

SALMONELLA

/Detection

other bacteria. This provides a selective advantage

and has been used in the development of another

commonly used enrichment medium, tetrathionate

broth. Tetrathionate is usually employed in combin-

ation with brilliant green and bile salts, the medium

being known as Mu

ller–Kauffman tetrathionate

broth. It is probably more selective than selenite, but

more care may be required in the control of incuba-

tion temperatures.

0014 Malachite green, in combination with high concen-

trations (4%) of magnesium chloride and a medium

pH of 5.0–5.2, has been shown to be effective for the

isolation of Salmonella. The medium was originally

developed for use with feces but modifications in

the last 10–15 years, which have seen a reduction

in the malachite green concentration from 0.012%

to 0.004% and use of soya peptone rather than tryp-

tone, have meant that the formulation, usually known

as Rappaport–Vassiliadis (RV) broth, is more suitable

for selective culture following preenrichment. This

medium is probably the one of choice in the examin-

ation of nonclinical samples and numerous studies,

on many different materials, have confirmed the

superiority of RV broth over both selenite and

tetrathionate.

0015 The selective nature of the above broths is often

finely balanced and can be disturbed by either a large

inoculum, relative to broth volume, or overlong incu-

bation. A change in incubation temperature can also

profoundly affect the toxicity of the medium.

0016 An inoculum-to-broth ratio of 1:10 is recom-

mended for both selenite and tetrathionate broths,

although following preenrichment culture there

would appear to be no reduction in sensitivity if

smaller volumes of BPW are used. The size of the

inoculum is critical with RV broth and its ratio to

the medium should be at least 1:100. With this

medium, studies have indicated that more samples

are Salmonella-positive after 48 h rather than 24 h

incubation, although the differences are often not

significant. In contrast, extended incubation signifi-

cantly reduces the effectiveness of tetrathionate and

selenite broths and, with the latter, overgrowth of

competing flora occurs frequently. These media

can also become more toxic to the target organism

and Salmonella may die if incubation is continued

much beyond 24 h. Care also has to be taken in

controlling incubation temperature. Selenite and

tetrathionate can be used at either 37



Cor43



and are more selective at the higher temperature.

Some studies have shown, however, that Salmonella

will not always survive in these media at 43



C. Initial

studies on RV broth recommended 43



C but it is

now accepted that incubation at 41+0.5



C increases

sensitivity.

0017Studies with particular samples or in particular

situations have demonstrated that one selective

medium may be superior to another. No one medium,

however, can be relied upon always to maximize the

isolation rate with all samples. There may thus be

advantages in using two selective broths (Figure 1).

0018An important disadvantage with traditional culture

techniques is the time taken to confirm whether a

food sample is Salmonella-positive or -negative.

There have been many attempts to make Salmonella

testing more rapid and some of the newer methods

are discussed below.

0019At first sight, it would seem possible to reduce the

length of either preenrichment or selective broth cul-

ture, but in the past this approach has been largely

unsuccessful. A number of publications have stated

that recovery from injury is complete within 8 h at



C. This is dependent on the culture medium used

and the degree of cellular damage. In some systems,

lag times of up to 72 h have been reported, and other

studies have indicated that repair of damaged cells is

not complete until some hours after growth has begun.

Thus, subculture of preenrichment broths after 6 h

rather than 24 h incubation was found to reduce the

isolation rate. Newer work, however, has suggested

that it may be possible to shorten traditional culture

methods without any loss of sensitivity by using a

combination of delayed release of selective agents

into an optimized recovery medium. The new method

makes use of a peptone-based medium that contains

oxygen-consuming sterile outer membrane fragments

from Escherichia coli, which protect bacteria by

reducing oxidative stress. This allows damaged

Salmonella cells to recover more rapidly than with

conventional media. The method also makes use of

delayed release capsules that release the selective

components of RV broth after about 5 h at 42



Trials to date indicate that this method is at least

equivalent in performance to standard methods.

0020Selective agars should be so formulated that the

competing flora is suppressed while the target organ-

ism is able to form discrete, characteristic colonies. A

degree of selectivity is achieved by the addition of

either dyes like brilliant green or chemicals such as

sodium deoxycholate. The resistance of Salmonella to

selective agents is similar to that of many other bac-

teria and this can create difficulties in media formula-

tion. Thus, the detection of Salmonella on selective

agars relies heavily on diagnostic aspects. Principal

among these is the inability of most strains to ferment

lactose and the production of hydrogen sulfide from

thiosulfate at a neutral pH (leading to black colonies).

More recently, the advent of affordable chromogenic

substrates has led to the development of a number of

new selective differential media for Salmonella. These

SALMONELLA

/Detection 5081

new media often utilize a combination of traditional

diagnostic markers with new substrates for the detec-

tion of galactosidase, glucoronidase, and esterase ac-

tivities. Additionally, old diagnostic tools such as acid

production from carbohydrate utilization have been

applied to new substrates, e.g., propylene glycol util-

ization by Salmonella.

0021 As with enrichment broths, many selective agars

are available, but none has been shown definitively to

be suitable for all applications and the best course of

action would be to use two agars in parallel. The most

commonly used media appear to be bismuth sulfite,

xylose lysine deoxycholate (XLD), and brilliant green

agar (BGA).

0022 Many workers have attempted to improve selectiv-

ity by the incorporation of antibiotics and/or other

selective agents. Of these, novobiocin, which prevents

the growth of Proteus spp., seems to be the most

useful.

Confirmatory Tests

0023 Other organisms, particularly members of the family

Enterobacteriaceae, can resemble Salmonella on

selective agars and confirmation of ‘Salmonella-like’

colonies is necessary. Tests take the form of biochemical

confirmation, which usually involves assessment of

urease and lysine decarboxylase activity, fermentation

of dulcitol, indole production, growth in the presence

of potassium cyanide, utilization of sodium malonate

and, more recently, pyrrolindonylarylamidase (PYR)

activity. These reactions, in combination with sero-

logical tests, are usually sufficient for identification,

but additional tests may sometimes be necessary. If

these are to form part of a laboratory routine, the use

of commercial kits may be cost-effective. Antibodies

against somatic (O) and flagella (H) antigens are used

to confirm/identify Salmonella-like isolates. Somatic

antigens are composed of polysaccharide, while those

from the flagella are proteinaceous. Testing will usually

comprise slide agglutinations with polyvalent O and H

antisera, followed by the use of sera raised against

specific antigens.

Rapid Methods

0024The time taken to confirm or deny the presence of

Salmonella, especially in food and environmental

samples, can create difficulties and, with perishable

foods, can mean that products may be distributed

and consumed before test results are known. It

is not surprising that considerable attention has

Test portion (1 part, typically 25

Buffered peptone water (9 parts)

Incubate at 37

C for 18

h  2

Sub-culture 1

ml into 10

MKTTnmedium

Incubate at 37

C for 24

h  3

Plate onto: XLD agar + second

agar

Sub-culture 0.1 ml into 10 ml

Rappaport-Vassiliadis medium

Incubate at 41.5

C for 24

h  3

Plate onto: XLD agar + second

agar

Incubate at 37

C for 24

h  3

Select 5 typical colonies from

each

late for confirmation

Incubate at 37

C for 24

h  3

Select 5 typical colonies from

each

late for confirmation

fig0001 Figure 1 International Standards Organization (ISO) standard method for Salmonella isolation (ISO 6579:1997). BGA, brilliant green

agar; XLD, xylose lysine deoxycholate.

5082

SALMONELLA

/Detection

been paid to the development of rapid methods of

detection.

0025 The new methodologies can be divided into two

broad categories. There are those designed to detect

or isolate viable cells and those that detect cellular

components such as DNA. There are concerns that

the latter methods may detect either dead cells or

DNA released from them and that cultural techniques

may lack the required sensitivity.

0026 Conventional culture techniques, as discussed

earlier, can be long-winded, cumbersome, and expen-

sive. How much better it would be if it were possible

to detect pathogenic bacteria by the direct application

of a test to a food sample. Salmonella spp., and pre-

sumably other foodborne pathogens, appear to have

unique DNA sequences that could allow detection.

Salmonella-specific DNA probes and polymerase

chain reaction (PCR) methods have been developed.

0027 These techniques compare well to conventional

culture when used on enrichment broths. The above

studies were designed to obviate the need for selective

culture and serological/biochemical identification.

There are other examples, and various automated

PCR-based systems are now available. Most of them

require an enrichment period which delays results but

sensitivity can be of the order of 3 CFU per 25 g of

food.

0028 Other new approaches include flow cytometry.

These methods have faced the criticism that they are

still too slow for the rapid pace of modern processed-

food production methods. Tests would be much

easier to perform and much more rapid if the modern

technologies could be applied directly to samples.

Such an approach, which continues to be investi-

gated, is not without problems. One is due to the

fact that DNA is heat-stable. Thus, intact DNA will

be present in processed foods. DNA in both a free

form and within dead cells of Salmonella can also

survive for considerable periods in sea water. One

possible way to overcome this problem is to include

a culture step as a means of detecting viable cells. Not

only will this increase the length of the test but it may

not be sufficient to allow the growth of more severely

damaged cells, as discussed earlier.

0029 An alternative nucleic acid target to determine via-

bility would be an RNA molecule that exists as many

copies in a cell and which exhibits rapid turnover.

Messenger RNA (mRNA) may prove to be a suitable

target as it has a reported half-life of 2 min or less,

and PCRs have been published for the detection of

viability in Escherichia coli and the thermophilic

Campylobacter spp. The rate of degradation of the

mRNA is influenced by the method of inactivation,

however, and this has to be considered when assessing

food-processing conditions.

0030‘Rapid’ methods clearly have a place for the detec-

tion of Salmonella spp. in food samples. With many

of the current tests, time is saved because selective

plating and biochemical and/or serological confirma-

tory tests are not necessary. ‘Rapid’ tests would

appear to offer advantages over traditional culture

systems. It is likely that improvements will be made

in both speed and sensitivity. This may encourage

a wider use of the new technologies. If this is

the case, it is to be hoped that people continue to

bear in mind the vital roles of proper sample handling

and the importance of using initial culture systems,

which facilitate the recovery and growth of injured

organisms.

See also: Escherichia coli: Occurrence; Detection; Food

Poisoning; Occurrence and Epidemiology of Species

other than Escherichia coli; Food Poisoning by Species

other than Escherichia coli; Food Poisoning: Tracing

Origins and Testing

Further Reading

Bailey JS, Cox NA and Blankenship LC (1991) Comparison

of an enzyme immunoassay, DNA hybridisation, anti-

body immobilisation and conventional methods for re-

covery of naturally occurring Salmonella from processed

broiler carcases. Journal of Food Protection 54: 354–

356.

Barrow GI and Feltham RKA (1993) Cowan and Steel’s

Manual for the Identification of Medical Bacteria, 3rd

edn. Cambridge: Cambridge University Press.

Candish AAG (1991) Immunological methods in food

microbiology. Food Microbiology 8: 1–14.

Cudjoe KS and Krona R (1997) Detection of Salmonella

from raw food samples using Dynabeads anti-Salmon-

ella and a conventional reference method. International

Journal of Food Microbiology 17: 55–62.

de Boer E (1998) Update on media for isolation of entero-

bacteriaceae from foods. International Journal of Food

Microbiology 45: 43–53.

Fierens H and Huyghebaert A (1996) Screening of Salmon-

ella in naturally contaminated feeds with rapid methods.

International Journal of Food Microbiology 31:

301–309.

Howie A (1999) A new development in Salmonella

testing. European Food and Drink Review, winter

issue, 53–56.

Mansfield L and Forsythe S (1996) Collaborative ring-trial

of Dynabeads anti-Salmonella for immunomagnetic

separation of stressed Salmonella cells from herbs and

spices. International Journal of Food Microbiology 29:

41–47.

Reissbrodt R, Vielitz E, Kormann E, Rabsch W and Kuhn H

(1996) Ferrioxamine E-supplemented pre-enrichment

and enrichment media improve various isolation

methods for Salmonella. International Journal of Food

Microbiology 29: 81–91.

SALMONELLA

/Detection 5083

Saeed AM (ed.) (1999) Serovar Enteritidis in Humans and

Animals: Epidemiology, Pathogenesis and Control.

Ames, Iowa: Iowa State University Press.

Speck ML (1984) Compendium of Methods for the Micro-

biological Examination of Foods, 2nd edn. Washington,

DC: American Public Health Association.

Salmonellosis

C Wray, Greendale Veterinary Laboratories Ltd,

Knaphill, Woking, UK

C A Hart, University of Liverpool, Liverpool, UK

Introduction

0001 Salmonellosis is the general term for human illness

caused by members of the genus Salmonella; however

from the epidemiological point of view it is necessary

to differentiate the enteric fevers caused by S. typhi

and paratyphi, A, B, and C from foodborne gastro-

enteritis caused by any of the other Salmonella sero-

vars. The infectious nature of typhoid fever has been

known for more than a century and the causative

organism was one of the earliest to be isolated and

characterized. The transmission of S. typhi occurs

most often by water which has been contaminated

by feces or urine from infected human carriers or food

prepared and handled by such carriers who do not

take adequate hygienic precautions. The organism

appears to be only a parasite of humans and the

higher apes and the link between poor sanitation

and other epidemiological aspects of the disease was

elucidated by William Budd before the opening of the

bacteriological era.

0002 In contrast, gastroenteritis (foodborne disease) is

usually acquired from consumption of foods which

may be directly or indirectly contaminated with Sal-

monella. Whether infection follows exposure to any

pathogenic microorganism depends in part on the

number of organisms (infective dose) that enter the

host’s tissues. Factors that may predispose to human

salmonellosis are given in Table 1 and the following

section. There are four main clinical patterns of

Salmonella infection in humans: (1) typhoid (enteric

fever); (2) acute gastroenteritis; (3) septicemia, with

or without localized infection; and (4) asymptomatic

infection.

Enteric Fevers

0003 Typhoid and paratyphoid fevers usually follow inges-

tion of drinking water contaminated with feces

containing S. typhi/paratyphi A, B, or C, or food

prepared or handled by a person who is a carrier of

the organism. Direct transmission from one individ-

ual to another may occur via the fecal–oral route,

especially in children, because of their poor attention

to personal hygiene. While large-scale epidemics of

typhoid fever have been traced to human fecal con-

tamination of food, poor sanitary practice resulting in

contamination of drinking water by fecal material

containing S. typhi appears to be a major factor in

many countries. Of foodborne outbreaks, 504 cases

of typhoid occurred in Aberdeen; this was caused

by the ingestion of tinned corned beef from Argen-

tina. The meat was contaminated in Argentina, where

river water containing unprocessed sewage was used

to cool the containers after sterilization. The water

seeped through the seams of the cans and reached

the contents. In Aberdeen, infection developed in

individuals who ate the corned beef, but also in those

who ate meat that became contaminated by the

machine that had been used to slice the corned beef.

0004In Mexico, in the early 1970s a large outbreak of

typhoid was caused by a chloramphenicol-resistant

strain of the organism, which had contaminated

water. The lack of susceptibility of the organism to

chloramphenicol was responsible for the death of

many patients who were infected with it. The resist-

ance determinant was a useful epidemiological

marker when the organism caused disease in the

USA, and a number of European countries.

Clinical Findings: Typhoid Fever

0005Typhoid fever is the best-studied form of salmonel-

losis and is well-described in the literature. It is po-

tentially fatal for young children, particularly during

the first year of life, and for the elderly, but in endemic

areas a degree of immunity may develop in the popu-

lation. Clinically, paratyphoid tends to be less severe

than typhoid.

tbl0001Table 1 Factors affecting the infectious dose of Salmonella

Bacteria-associated factors

Serovar, genotype

State and nature of contaminated material ingested (dry, moist,

liquid, and does it allow bacterial multiplication?)

Resistance to antibacterials

Stress status of the bacteria

Host-associated factors

Age of host (very young and aged are most susceptible)

Effect of therapy (prior treatment with antibacterials,

antacids, etc.)

Presence of other disease (malaria, human immunodeficiency

virus (HIV), etc.)

Immune status of the host

Genetic status of the host

5084

SALMONELLA

/Salmonellosis

0006 Illness usually occurs 7–10 days after consumption

of infected water or food, and the initial symptoms

tend to be both vague and variable. Headaches tend

to be common, as does a general feeling of lethargy,

along with abdominal pain/discomfort, and constipa-

tion is more usual than diarrhea. The fever increases

progressively during the first week, and then de-

creases, although a biphasic response may be seen.

Rose spots representing cutaneous vasculitis may be

seen on the abdomen and enlargement of the spleen

may be felt on abdominal palpation. The initial symp-

toms resemble those of acute gastroenteritis and Sal-

monella may be isolated from the feces. Then for a

week or more the organism may not be isolated from

the feces, although blood cultures are positive. This is

then followed by further fecal shedding. In about

25% of patients Salmonella may be isolated from

the urine. Fatalities usually arise from perforation

of the intestine after the bacteria have been localized

in the Peyer’s patches, and/or pneumonia, but other-

wise the disease runs a slow course over 5–6 weeks

and then leaves the patient in a longer period of

gradual convalescence.

0007 The pathogenesis of the disease is of interest. From

the small intestine the organisms pass via the lympha-

tics to the mesenteric lymph nodes, where, after a

period of multiplication, they invade the blood

stream via the thoracic duct and the liver, gall bladder,

spleen, kidney, and bone marrow become infected

during this bacteremic phase in the first 7–10 days of

the disease. From the gall bladder a further invasion

of the intestine occurs and lymphoid tissue, particu-

larly Peyer’s patches, are involved in an acute inflam-

matory reaction, followed by necrosis, sloughing, and

the formation of the characteristic typhoid ulcers.

Hemorrhage of varying degree may occur and, less

frequently, perforation through a necrotic Peyer’s

patch may complicate the illness and result in death.

S. typhi is present in large numbers in the inflamed

tissues in the ulcers and it may localize in the kidneys

and be found in other lesions that occur as a com-

plication or sequela of typhoid fever, e.g., acute sup-

purative periosteitis, renal and hepatic abscesses,

bronchopneumonia, and ulcerative endocarditis.

0008 In 2–5% of convalescents the organism persists in

the body, sometimes for an indefinite period, and in

such chronic carriers the organism is most commonly

found in the gall bladder or, more rarely, in the urin-

ary tract.

Diagnosis

0009 Typhoid fever should be suspected in any patient who

has visited an endemic area; the diagnosis is con-

firmed by recovering the organism from the feces,

urine, blood, or other tissues. Blood cultures are

positive in about 80% of patients and in the early

stages of illness, i.e., 7–10 days, it is the most reliable

diagnostic test. Bone marrow cultures and skin biop-

sies of rose spots are, however, more likely to be

positive for S. typhi than blood cultures. S. typhi

and paratyphi can be isolated from the feces through-

out the illness but are most frequently found during

the second and third weeks. Repeated examination of

the feces may be required before isolation is success-

ful. The tube agglutination test (Widal reaction) for

detecting antibodies to the somatic, flagellar H and

the envelope Vi antigen has been used for the sero-

logical diagnosis and positive titers occur about day

7–10 after infection. Occasionally positive titers de-

velop earlier, but they may be delayed and a negative

result at an early stage of the illness therefore may be

inconclusive (cross-reactions with other bacteria and

previous typhoid may give false-positive results).

Other serological tests, such as enzyme-linked immu-

nosorbent assays (ELISAs) have also been developed

to detect different antigens of the bacillus. Vaccin-

ation may give rise to false-positive serological reac-

tions and, if the patient has a history of vaccination, it

is advisable to take a second sample a week or so later

to determine whether there is a rising titer.

Chemotherapy

0010The treatment of enteric fevers necessitates the use

of antibacterial drugs. Chloramphenicol, ampicillin/

amoxicillin, trimethoprim/sulfamethoxazole are the

drugs of choice against sensitive Salmonella. Multiple

drug-resistant (MDR) isolates of S. typhi, resistant to

the first three antimicrobials (chloramphenicol, ampi-

cillin, and co-trimoxazole) have emerged in South-

east Asia and in 1998 in Kenya. Therapeutic options

include fluoroquinolones or azithromycin. Thus,

because MDRs of S. typhi have been reported, all

isolates should be tested for their antimicrobial sus-

ceptibility. Proper management of the fluid and elec-

trolyte balance is important in all patients, especially

the young and the old.

Immunoprophylaxis

0011Vaccines have been available for the prevention of

enteric fevers for many years. The classic inactivated

vaccine TAB protects against both typhoid and para-

typhoid A and B and extensive field trials in endemic

areas have shown a good degree of protection, al-

though a high percentage of vaccinates develop local

and systemic reactions.

0012In recent years, two new typhoid vaccines have

become available. A vaccine prepared from the Vi

capsular polysaccharide antigen has given good results

in field trials. One dose protects for 3 years and it has

a low incidence of side-effects. It is not licenced for

SALMONELLA

/Salmonellosis 5085

children less than 18 months of age because they may

have a suboptimal response.

0013 A live oral preparation produced from an attenu-

ated strain (Ty 21a) of S. typhi is available in enteric-

coated capsules; it is administered in three doses on

alternate days. Significant side-effects have not been

reported, but transient mild nausea, vomiting, ab-

dominal cramps, diarrhea, and urticaria may occur

in less than 1% of vaccinates. In mass trials it has not

proved as effective as the previous vaccines.

Acute Gastroenteritis

0014 Acute gastroenteritis is usually acquired from con-

sumption of foods which may have been directly or

indirectly contaminated with Salmonella. A wide

variety of animal species have been shown to be

capable of harboring the organisms without showing

clinical signs, and in the developed world, turkeys,

swine, chickens, and cattle have often been found to

be infected in abattoir studies. Animal products such

as milk and eggs have also been shown to be import-

ant vehicles of infection. Many other foodstuffs, in-

cluding vegetables and salads, have also been found

to be associated with Salmonella outbreaks. It should

also be remembered that person-to-person spread

may also be of importance.

Clinical Findings

0015 The incubation period ranges from 8 to 48 h

following the ingestion of contaminated food. The

most common symptoms are diarrhea, abdominal

pain, fever, and headache. Extremely fluid stools per-

sist for 3–4 days, and these may contain mucus and

occasionally blood. The temperature may be slightly

elevated in most patients but usually returns to

normal within 1–2 days. In the young and the elderly

the symptoms may be more severe and last for a week

or more.

0016 Salmonella gastroenteritis is usually a self-limiting

illness and fatalities are uncommon. Sigmoidoscope

and biopsy examinations have demonstrated that the

colon is the major site of infection. The changes in the

colon range from edema of the lamina propria with

a focal or diffuse inflammatory infiltrate to a more

intense inflammation with disruption of the surface

epithelium and multifocal microabscesses. In more

severe cases, edema, vascular congestion, infiltration

of the lamina propria with polymorphonuclear leuko-

cytes, and abscess formation have been recorded.

Chemotherapy

0017 Although Salmonella are usually sensitive in vitro to

many antibiotics, their use for the treatment of un-

complicated gastroenteritis until recently has generally

been contraindicated by the lack of favorable effect on

the course of disease and by a prolongation of Salmon-

ella shedding. Following the introduction of the fluor-

oquinolones, a number of clinicians have advocated

their use for the treatment of Salmonella gastroenter-

itis because of their efficacy in reducing the severity of

symptoms, the duration of illness, and Salmonella

shedding. However, this has been followed by increas-

ing development of resistance.

Salmonella Bacteremia (Septicemia)

0018A transient bacteremia may occur in patients with

gastroenteritis caused by Salmonella, although

periods of longer persistence occur in the blood of

individuals with one of the forms of enteric fevers or

those suffering from underlying disorders. Chronic

bacteremia is commonly associated with S. cholerae-

suis infection, and the fatality rate is usually two to

three times that caused by typhoid fever. Metastatic

infectious foci, of which the most important are intra-

vascular lesions, osteomyelitis, and meningitis, are

common complications of disease caused by S. cho-

leraesuis. This organism may lie dormant in some

tissues, usually bone marrow, and then later become

activated.

0019The serovar most commonly associated with bacter-

emia is S. typhimurium. This is perhaps not surprising

because it is the most frequent cause of salmonellosis

in humans and other animals. The high-grade bacter-

emia suggests the presence of focal intravascular infec-

tions in either the large arteries or the endocardium,

but whether this is a primary or secondary effect is

not certain. In sub-Saharan Africa, non-typhi Sal-

monella (NTS) are the most common isolates from

blood culture in children (< 50% of isolates). This is

not related to human immunodeficiency virus (HIV)

infection but occurs in the malaria season and espe-

cially in children with severe malarial anemia. In

HIV-infected adults NTS bacteremia is an early ac-

quired immunodeficiency syndrome (AIDS)-defining

illness: relapse after therapy commonly occurs.

0020Focal infections, such as osteomyelitis, arthritis,

and meningitis, are not uncommon sequelae to Sal-

monella infections. Any area of the skeleton, espe-

cially the long bones, may be invaded. Two types of

arthritis have been described: bacteremia may give

rise to suppurative arthritis with the organism in the

joint fluid, or a sterile inflammatory reaction involv-

ing the synovia may occur as a sequel in about 2.5%

of cases of gastroenteritis.

Chemotherapy

0021Chemotherapy is usually similar to that for typhoid

fever, although prolonged antibiotic therapy for

5086

SALMONELLA

/Salmonellosis