International Journal

of Food Microbiology
International Journal of

ELSEVIER

Food Microbiology 22 (1994) 277-289

A comparison of six different plating media used

in the isolation of Salmonella
Donald W. Warburton a,* Bruce Bowen a Anne Konkle a
Carol Crawford b, Samir Durzi c, Roger Foster a, Cathy Fox e,
Lorraine Gour f, Gail Krohn g, Pierre LaCasse h, Gaetan Lamontagne i,
Shelagh McDonagh J, Victoria Arling J, Janet Mackenzie k,
Ewen C.D. Todd h, John Oggel ~, Robert Plante m Susan Shaw n,
N.P. Tiwari o, Yvon Trottier v, Brenda Daly Wheeler q
a Evaluation Division, Bureau of Microbial Hazards, HPB Health and Welfare Canada, Ottawa,
KIA OL2 Canada;
b FieM Operations Division, HPB Health and Welfare Canada, Burnaby, V5G 4P2 Canada;
Fisheries and Oceans Canada, Toronto, M6N 3E3 Canada;
d Field Operations Directorate, HPB Health and Welfare Canada, Winnipeg, R2J 3YI Canada;
e Field Operations Directorate, HPB Health and Welfare Canada, Dartmouth, B2Y3Z7 Canada;
f Field Operations Directorate, HPB Health and Welfare Canada, Longueuil, J4K 1C7 Canada;
g Regina Veterinary Laboratory, Saskatchewan Agriculture and Food, Regina, $4S OBI Canada;
n Laboratoires d'expertises et d'analyses alimentaires de Sainte-Foy MAPAQ, Sainte-Foy,
G1P 2W8 Canada;
i Laboratoires d'expertises et d'analyses alimentaires de Sainte-Hyacinthe MAPAQ, Sainte-Hyacinthe,
J2S 2M2 Canada;
J Lab Services Division West, Agriculture Canada, Calgary, T2L 2L1 Canada;
k Research Division Bureau of Microbial Hazards, HPB Health and Welfare Canada, Ottawa,
K1A OL2 Canada;
l Laboratory Services Division, Agriculture Canada, Ottawa, K1A 0C6 Canada;
m Service de l'environnement, Communaute urbain de Montreal, Montreal, H4N 2T2 Canada;
n Fisheries and Oceans Canada, Halifax, B3J 2S7 Canada;
o Food Laboratory, Services Branch, Alberta Agriculture, Edmonton, T6H 4P2 Canada;
P Laboratoire d'hygiene veternaire, Agriculture Canada, Sainte-Hyacinthe, J2S 8E3 Canada;
q Fish Inspection Laboratory, Fisheries and Oceans Canada, St. John's A1C 5X1, Canada

(Received 15 December 1993; accepted 21 February 1994)

Abstract

Seventeen Canadian Federal, Provincial and Public Health Laboratories took part in
different phases of a comparative/collaborative study that evaluated rapid methods to the

* Corresponding author.
0168-1605/94/$07.00 1994 Elsevier Science B.V. All rights reserved
SSDI 0168-1605(94)00017-Z

278

D.W. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

standard Health Protection Branch (HPB) method for the detection of Salmonella. A
variety of commercial media were tested, including Brilliant Green Sulpha Agar, Bismuth
Sulphite Agar, Hektoen Enteric Agar, Xylose Lysine Deoxycholate Agar, EF-18 Agar and
Rambach Agar. Each laboratory compared up to six of these different plating media.
Plating of 123 salmonellae cultures and 28 artificially-inoculated foods showed the recovery
of Salmonella spp. on the six plating media to be within one log. Therefore, quantitative
testing of the media showed them to be comparable in the recovery of salmonellae.
Qualitative testing of the six media during the comparative/collaborative study of various
methods showed that EF-18 Agar recovered the greatest number of isolates. Hektoen
Enteric Agar ranked second, with the other agars being comparable in their recovery of
Salmonella spp. Problems with the various media are summarized. Based on our results and
those of other researchers, it is recommended that Bismuth Sulphite Agar be compulsory
and that at least one other agar be used for newly developed cultural procedures.
Key words: Comparative/collaborative study; Plating media; Isolation procedure; Salmonella

I. Introduction

The use of selective and differential plating media is a very important part of
standard cultural methods for the isolation of Salmonella from foods and environmental samples. A wide variety of media have been developed, differing in their
composition, use and performance characteristics.
Brilliant G r e e n Sulpha Agar (BGS; HPB, 1978), Bismuth Sulphite Agar (BIS;
Andrews et al., 1984; HPB, 1978), Hektoen Enteric Agar (HEK; Andrews et al.,
1984; Anon., 1993) and Xylose Lysine Desoxycholate Agar (XLD; Andrews et al.,
1984) are used in standard methods of different agencies in Canada, the United
States and in Europe (Flowers et al., 1992). BGS uses sodium sulfapyridine as the
selective agent and brilliant green and phenol red dyes as the indicators of
carbohydrate utilization (Anon., 1984). BIS uses bismuth sulfite and brilliant green
as inhibitory agents against competing microrganisms and ferrous sulfate as the
indicator of H2S production (Difco, 1984). H E K relies on the use of bile salts for
selectivity, bromothymol blue and acid fuschin as indicators of carbohydrate
utilization, and ferric iron as an indicator of formation of H2S from thiosulphate
(Anon., 1993b). X L D uses sodium desoxycholate, sodium thiosulfate, ferric ammonium citrate and phenol red as selective agents, indicators of H2S production and
carbohydrate utilization, respectively (Anon., 1984)
EF-18 Agar (EF-18; Dickson and Anderson, 1991; Entis, 1990; Entis and
Boleszczuk, 1991; Todd et al., 1992) and Rambach Agar (RAM; Anon., 1993,
Bankes, 1992; Feng, 1992; Freydiere and Gille, 1991; Gruenewald et al., 1991;
Manafi and Sommer, 1992; Rambach, 1990) are newly developed media associated
with the recently developed methods (cited above). EF-18 uses sulfapyridine, bile
salts, crystal violet, novobiocin and incubation at 42C as differential/selective
agents, with sucrose and lysine decarboxylase utilization as its differential reactions
(Entis, 1990). R A M uses the selective agent desoxycholate, the formation of acid
from propylene glycol by Salmonella, and a chromogenic indicator of/3-galactosi-

D.W. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

279

dase to differentiate Salmonella from Proteus and other Enterobacteriaceae

(Anon., 1993a).
These different commercially available plating media were evaluated as part of
a comparative and collaborative study comparing rapid methods for the detection
of Salmonella to the standard Health Protection Branch (HPB) method (HPB,
1978) involving Canadian Laboratories.

2. Materials and methods

Seventeen Canadian Federal, Provincial and Public Health Laboratories took

part in different phases of a comparative/collaborative study that evaluated rapid
methods to the standard HPB method for the detection of Salmonella. During the
different parts of this study a variety of commercially available media were tested,
including BGS (Difco Laboratories, Detroit, MI), BIS (Merck, Darmstedt, Germany; Difco Laboratories), H E K (Merck; Difco Laboratories; Unipath Inc., Nepean, Ont.), XLD (Merck; Difco Laboratories), EF-18 (QA Life Sciences Inc., San
Diego, CA) and RAM (Merck). The first four media are associated with standard
methods (Flowers et al., 1992) while the last two are associated with methods being
tested in this study. Each laboratory compared up to six of these different plating
media.

2.1. Microorganisms
One hundred and seventeen Salmonella and 49 non-Salmonella cultures were
used (Table 1). These cultures were used to artificially contaminate samples, and
as positive or negative controls for testing both media and methods.

2.2. Samples tested

Seven hundred and thirty-five naturally-contaminated and 551 artificially-contaminated food and environmental samples were tested in this study (Table 2). In
all cases the foods were analysed by the HPB method and one other method.
Often, several methods were used simultaneously. In other cases, samples found
positive by one method, were often re-tested by other methods so that 1622
naturally-contaminated and 1271 artificially-contaminated samples were tested by
the six methods.

2.3. Artificial contamination of the foods

The method used was the same as that followed for an earlier study with
Listeria (Warburton et al., 1992) and is summarized below: The Salmonella
cultures were grown in Tryptose Phosphate Broth (TPB) for 24 h at 35C. After a
second subculture in TPB (24 h at 35C), the cultures were serially diluted in 0.1%
peptone water and 0.5 ml of the 10 -s dilution was inoculated into 25 g of food (so

~.80

D.W. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

Fable 1
List of cultures used (numbers of isolates)
~almonellae used
~almonellae(13) Agriculture Canada and HPB
retyped isolates,
~almonellae Grp B(4) chicken fluff, chicken,
:urkey and pork isolates,
~almonellae Grp C(2) chicken isolates,
~almonellae rough(2),
~. alachua
~. albany
~. agona,
~. anatum,
~. arizonae(4),
~. arizonae 1,
~. arizonae II(2),
g. arkansas,
g. bareilly(2),
5. bredeny,
5. braenderup,
5. cerro(3),
5. chester,
5. cholera-suis,
5. dublin(2),
S. ealing(2),
S. eastbourne(2),
S. enteritidis(2) 1 H 2 S - ,
S. flint(2),
S. gallinarum,
S. git,e,
S. godesburg,
S. haardt,
S. hadar(6),
S. hal~ana(2),
S. heidelberg,
S. indiana H2S-Suc+
S. infantis(5),
S. johannesberg(5),
S. kentucky(2) 1 Lac +,
S. landow,
~. mbandaka(5),
S. montet;ideo,
S. minnesota(2),
S. muenster,
S. newington,
S. newport,
S. ohio,
S. oranienburg,
S. pinza,
S. pullorum,

S. reading,
S. schwarzengrund(4)
S. senftenberg(6) 1 H2S-Lys-, 1 775W, 1 Suc+
S. st.-paul(2),
S. taksony(2),
S. thomasL,ille(2),
S. thompson(5),
S. typhi,
S. typhimurium(4), 1 ATCC 14028,
S. urbana,
S. wassenaar Grp II,
S. welteureden,
S. worthington(2),
Other species
Achromobacter xylosoxidans,
Citrobacter dit,ersus,
C. freundii(3) 1HzS +,
Enterobacter aerogenes(3),
E. agglomerans (5),
E. cloacae(2),
E. georgenes,
E. hafnia,
E. taylorae,
Enterococcus faecal&,
Erwinia carotoL,ora(2),
Escherichia coli(5),
Hafnia ah,ei(3),
Klebsiella oxytoca,
K. pneumoniae(2),
Oerskot~ia sp.,
Proteus sp.,
P. mirabilis(3),
P. morganii(2),
P. rettgeri(2),
P. t,ulgaris(2),
ProL,idencia spp.,
Providencia alcalifaciens,
Pseudomonas aeruginosa,
P. fluorescens,
Serratia fonticola,
S. marcescens(2),
Shigella boydii,
S. dysenteriae,
S. flexneri,
S. sonnei,
Staphylococcus aureus,

D.W. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

281

that the inoculation was < 1 c f u / g ) and blended (or mixed using a stomacher) with
the preenrichment broth. To ensure a high competitor level, either 1 ml of a
non-Salmonella culture or 2 g of soil (previously tested and found negative for
Salmonella) was added to the food mixture. Incubation and analysis continued as
appropriate for the method studied (described below).
To determine the inoculum level, 0.1 ml of the 10 -6 to 10 -8 dilutions were
plated onto BIS and the other agars. The agars were incubated at 35C for 18-48
h, except for EF-18 which was incubated at 42C. At 24 h, plates were examined
and confirmation started on typical colonies. As needed, BIS and other agars were
re-incubated overnight, and visible colonies counted.
2.4. Methods tested (qualitative testing)

During the comparative/collaborative testing six methods were used. Five

methods (the modified 1-2 Test (Feng, 1992; Nath et al., 1989; Oggel et al., 1990;
BioControl Systems Inc., Bothell WA), EF-18 and H G M F method (Entis, 1990;
Entis and Boleszczuk, 1991; Feng, 1992; Gelman Sciences Inc.), the E L A method
(Todd et al., 1993), Gene-Trak (Feng, 1992; B D H Inc.), the Merck Salmosyst broth
and tablet method (Ossmer, 1992; B D H Inc.)) were compared to the HPB
standard method MFHPB-20 (HPB, 1978). The results of the c o m p a r i s o n /
collaborative testing of the methods will be published elsewhere (Warburton et al.,
1994b,c,d,e). Each laboratory could streak out the selective broths, from the HPB
and other methods tested, onto at least BIS and BGS, plus the media stipulated
for that method and any other agars they wished to test. A representative number
of typical colonies (up to five per plate) and atypical colonies, if desired, were

Table 2
Foods tested and number of samples found to contain Salmonella sp. by one or more methods
Sample
NaturallyArtificiallycontaminated
contaminated
Controls
ND
124
Dairy: cheese
ND
52
milk/cream
ND
87
Egg/egg powder
3
17
Environmental
15
3
Feeds
366
10
Seafood
17
2
Meat
15
11
Other
35
21
Poultry
62
ND
Spices
221
7
Vegetables/fruit
ND
10
Water
ND
10
Grain/flour
1
197
Total
ND, not done.

735

551

282

D.W.. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

purified for confirmation. Biochemical testing and serology were done according to
the HPB method (HPB, 1978). Other confirmation tests included the use of
API-20E (API Laboratory Products Ltd, St-Laurent, Que.), Micro-ID (Organon
Teknika Inc., Scarborough, Ont.), Enterotube (Roche Diagnostic Systems, Nutley,
N J), as well as the mini-Vidas and the Vitek (bioMerieux USA, Hazelwood MI).
The ability of standard non-Salmonella to grow on the plating media was tested
as follows. The cultures, listed in Table 1 as "other species", were subcultured and
streaked onto the plating media which were incubated as described above.

2.5. Quantitative testing of the media

One hundred and twenty-three salmonellae cultures and 28 inoculated food
samples were plated on BIS, BGS, EF-18, H E K and RAM. An additional 46
salmonellae were plated on these agars and XLD. The foods consisted of nine
inoculated homemade chocolate and egg liqueur samples, as well as 19 inoculated
distilled water samples (Warburton et al., 1993, 1994a,b). The salmonellae in these

Table 3
Recovery of Salmonellae by different selective plating media
Plating medium tested

BGS vs BIS vs H E K vs
EF-18 vs RAM vs XLD
BGS vs BIS vs H E K vs
EF-18 vs RAM
BGS vs BIS vs H E K vs
RAM
BGS vs BIS vs RAM
BGS vs BIS vs EF-18 vs
XLD
BGS vs BIS
BGS vs BIS vs H E K
BGS vs BIS vs H E K vs
RAM vs XLD
BGS vs BIS vs EF-18
Range of % recovery

Number of positive plates (% recovery) a

BGS

BIS

HEK

EF-18 b

RAM

XLD

139
(93.3 a)
596
(95.5)
102
(85.7)
238
(94.8)
91
(94.8)
1125
(100)
301
(94.7)
114
(97.4)
26
(72.2)
72.2100

136
(91.3)
605
(97.0)
119
(100)
251
(100)
92
(95.8)
1119
(99.5)
267
(84.0)
53 c
(45.3)
34
(94.4)
45.3100

142
(95.3)
613
(98.2)
116
(97.5)
ND

149
(100)
624
(100)
ND

132
(88.6)
ND

ND

96
(100)
ND

130
(87.3)
602
(96.5)
95
(79.8)
249
(99.2)
ND

ND
318
(100)
114
(97.4)
ND
95.3100

ND

ND
ND

ND

96
(100)
ND

ND

ND

ND

ND

100
(85.5)
ND

117
(100)
ND

79.899.2

88.6100

36
(100)
100

AS compared to the media with the highest number of positive plates. Plating media were streaked
from either the side-arm of the 1-2 Test, the Salmosyst broth, Tetrathionate brilliant green broth,
Mueller-Kaufman Tetrathionate broth, Selenite cystine broth or GN broth.
b The results for the EF-18 agar are from streaked plates, the results using the H G M F s will be
presented elsewhere.
c One lab was unfamiliar with BIS, and had problems with false-positives and false-negatives (see text).
ND, not done.
a

D.W. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

283

foods were stressed due to exposure to alcohol, and being stored in a low nutrient
environment for 60 days, respectively. Counts were obtained by plating serial
dilutions of the foods.
Quantitative recovery from broth cultures was studied using the method used
for L i s t e r i a (Warburton et al., 1992) and is summarized as follows. After being
subcultured twice in TPB, 24 h at 35C, the S a l m o n e l l a cultures were then serially
diluted in 0.1% peptone water and 0.1 ml of the 10 -6 to 10 -8 dilutions were
plated onto BIS and the other agars, which were incubated as above.

3. Results

Table 3 shows the results of the qualitative testing of the six media on the
different foods tested (Table 2). The % recovery was determined by comparing the
number of positive plates recovered by each medium to the highest number of
positive plates possible in each comparison. The range of the % recovery of
S a l m o n e l l a varied between plating media (Table 3). Streaked plates of EF-18 agar
recovered the greatest number of isolates, showing 100% recovery. H E K ranked
second with 95.3-100% recovery, and the others as follows: X L D with 88.6-100%,
R A M with 79.8-99.2%, BGS with 72.2-100% and BIS with 45.3-100% recovery.
One laboratory was unfamiliar with BIS, despite the fact that they tested for
S a l m o n e l l a . [Other laboratories noted some problems (Table 5) with the different
media, but not to this extent]. Subsequently, this laboratory had problems differen-

Table 4
Microorganisms givingfalse-positive reactions on the plating media a
Plating media
BGS
M C. malonaticus
i
C
r
0
0
r

g
a

C. freundii
E. agglomerans
E. taylorae
K. oxytoca
Kluyvera spp.
P. mirabilis
P. morganii
Pseudomonas spp.

BIS

HEK

EF-18

RAM

XLD d

Aeromonas spp.
C. freundii
E. agglomerans
E. eoli
H. alvei
K. pneumoniae
P. mirabilis

C. diversus
C. freundii
P. mirabilis

C. freundii
E. aerogenes
E. agglomerans
E. coli
E. fergusonii
H. alveii
Providencia
P. fluorescens

E. taylorae

C. freundii

H. alvei
Proteus spp. b
Providencia c
Pseudomonas

spp.

i
S

m
S

a Microorganisms giving false-positives were isolated from this study, and were part of the qualitative
testing (Table 1).
b,c Clear colonies (typical of some Salmonella): b p. morganii and P. rettgeri; c p. alcalifaciens.
d Some false-positives that were red (no H2S) that might be picked included /~ fluorescens, S.
dysenteriae, A. xylosoxidans, E. hafnia and P. morganii.

284

D.W. Warburton et al. /International Journal of Food Microbiology 22 (1994) 277-289

Table 5
Summary of the questionaire rating the plating media
Media

BIS

BGS

RAM

XLD

HEK

EF- 18

Company
(# labs using
product)
Selectivity
Rating

Difco(8),
Merck(2),
Unipath(l)
High(4),
Medium(6)

Difco(10)

Merck(7)

Difco(4),
Merck(2)

QA Life
Sciences(6)

High(2),
Med-high(2),
Medium(3)
High(4),
Medium(3)

Medium(5),
Low(l)

Difco(2),
Unipath(2),
Merck(3)
Medium(4),
Low(3)

High(l),
Medium(6),
Low(4)
Differentia- High(3),
High(2),
tion rating
Medium(6), Medium(7),
Low(I)
Low(1 )
Used before'? Yes(9) High, Yes(9) High,
No(I)
No(l)
Overall
rating

High(3),
Medium(7)

Would you
use it
routinely?

Yes(7),
Maybe(l),
No(l),
Only with
others(I)
a

Comments?

High(2),
Medium(5),
Low(3)
Yes(4),
No(4),
Only with
others(I)
b

High(2),
Medium(3),
Low(1)
Yes(3) Low to Yes(4) Low to
high,
high,
No(4)
No(2)
High(4),
High(l),
Med-high(1), Medium(3),
Medium(2)
Low(l)
Yes(7)
Yes(3),
Maybe(l),
No(2)

High(3),
Medium(3),
LOw(1)
Yes(5) Low to
high,
No(2)
High(3),
Medium(2),
Low(2)
Yes(4),
No(3)

High(l),
Medium(3),
Low(2)
High(2),
Medium(3),
Low( 1)
No(5)
High(l),
Medium(4),
Low(l)
Yes(2),
Maybe(2),
No(2)

Fussy preparation and storage, short expiry, detects atypical salmonella ( H 2 S - , Lac+), S. typhimurium at times difficult to detect, isolations easy to make, deteriorates with age quickly, need to
reincubate plates, too many false-positives.
b Lactose fermenters a problem, had few Salmonella colonies compared to the others, many mimics, not
suitable for S. typhi, Pseudomonas a problem, not vary selective.
c Looks promising, still being evaluated, expensive, clear differentiation between non- and salmonellae,
atypical reactions for S. typhi and paratyphi, a good medium for a novice.
d Discontinued its use, salmonellae very characteristic.
Discontinued its use, isolations were made easier with this agar, least number of false-negatives,
extremely easy to differentiate between non- and salmonellae, false-positives a problem.
f Colour may change at 4C storage, repicking due to false-positives, extremely good differentiation
between non- and salmonellae, Pseudomonas a problem.
NB: not all participants completed the evaluation forms. Ratings above were assigned by the participating labs.
a

Table 6
Other comparisons of various plating media (from D'Aoust 1989)
Food

Enrichment broths

Ranking

Frogs legs
Raw meats

TBG/SC
TBG
SC
TBG/SC
TBG/SC
TBG/SC
TBG/SC/RV
TBG/SC

BIS > HEK = XLD

XLD > HEK
HEK > BGS > XLD
BIS > BGS
BIS > BGS
BIS > XLD > HEK
BIS = BGS
BIS = HEK = XLD

Miscellaneous

Meats and animal feeds

Meats and dry products

D.W. Warburton et al. /International Journal of Food Microbiology 22 (1994) 277-289

285

tiating between Salmonella and other bacteria, so that a number of false-positives

and false-negatives were noted with this media, but not with others. If the lowest
recovery rate for BIS (45.3%) is regarded as an anomaly (due to the problems the
one lab had with this medium) then the % recovery range for BIS is 84.0-100%.
During the quantitative testing, there was a < 1 log difference between the
counts on five media (BIS, BGS, EF-18, H E K and RAM) with 120 (out of 123)
salmonellae cultures (data not shown). Testing of an additional 46 Salmonella
cultures on six media (the above including XLD) resulted in 43 with a < 1 log
difference in counts. Re-testing of the six cultures (with differences in counts > 1
log) on the six media resulted in a < 1 log difference (data not shown). Similarly,
plating the 19 water and the nine liquor samples containing stressed bacteria
resulted in 25 having < 1 log difference on these five media (data not shown).
Therefore, the quantitative recovery of Salmonella on the six media appears
comparable.
The stock cultures (Table 1) and isolates found to give false-positive reactions
on the various plating media are listed in Table 4. All media had from three to
nine different non-Salmonella that gave false-positive reactions. Although the list
of the false-positives is not exhaustive, it does show that each medium may have
problems with competitive microorganisms.

4. Discussion

Table 5 gives a summary of the ratings given to the various media by the
participating laboratories. This Table includes ratings for selectivity, differentiation
and overall performance with comments showing the pros and cons of each media.
BIS and RAM had the best overall ratings (medium to high) while the ratings of
other media varied from low to high. D'Aoust (1989) summarized the performance
of various plating media, with the pertinent information being shown in Table 6.
The importance of BIS in the isolation of Salmonella from various test materials
under different enrichment conditions is clearly evident from Table 6 (D'Aoust,
1989). However, based on the overall information provided by this Table BIS,
BGS, H E K and XLD are fairly comparable in their ranking.
BGS, H E K and XLD are similar in their response to carbohydrate (lactose and
sucrose) utilization, have a medium to low selectivity, and can contain numerous
false-positives (D'Aoust, 1989; Table 5). These agars, as well as RAM and EF-18,
do not readily identify atypical lactose-positive (lac + ) and sucrose-positive (suc + )
salmonellae (Table 7). This fact supports the use of BIS in standard methods due
to its nonsaccharide differential system (D'Aoust, 1989). However, BIS is not
without its problems (Tables 4 and 5) and is unable to detect some strains of
H2SSalmonella ( H E K and XLD may have problems detecting H 2 S Salmonella as well). Problems with preparation, storage and aging of BIS before
use have been noted (D'Aoust, 1989). After completion of this study, the potential
use of Novobiocin-Brilliant Green-Glucose Agar (Devenish et al., 1986) to detect

typical-red
with black
center
typical-red
with black
center
atypicalcreamy red
atypicalyellow

typical-red
with black
center
typical-red
with black
center
typical-red
with black
center

Salmonella sp.
Lys-

S. enteritidis
(green on BIS)

typical-black

typical-black

atypical-green

typical-red

typical-red

typical-red

RAM

typical-red

typical-dark green
with black center
typical-dark green
with black center
typical-dark green
with black center

typicalblue/green
atypical-yellow

atypicalcolourless to
pale pink
typical-red

atypical-blue

typical-red

atypical-blue

typical-red

typical-red

typical-red

atypical-creamy pink
with black center

atypical-green

typical-black

atypical-light green

atypical-light green

typical-dark green
with black center

atypical-creamy pink
with black center

atypical-light green

HEK

atypicalyellow/green
to typicalblue/green ~
typicalblue/green

atypical-yellow

typical-red

atypical-light
green to
typical-black
typical-black

typical-red

atypical-yellow
to typicalblue/green ~
atypical-yellow

atypical-light
green to
typical black
typical-black
typical-red

EF- 18
atypical-light
green
atypical-yellow

BGS
atypical-lime
green
atypical-lime
green

typical-black

BIS

After further incubation all colonies developed the typical blue/green coloration.

S. newington,
S. senftenberg,
S. ealing

S. flint

S. senftenberg
H:S-LysS. newport
Lac +

S. arizonae 1

XLD

atypicalyellow
atypicalyellow

Serotype

S. indiana H 2 S Suc +
S. senftenberg
Suc +

Table 7
The reactions of Salmonella on the six plating media

6,a

,,.q
-q

,K

g~

e~

o~

D.W. Warburton et al. / International Journal of Food Microbiology 22 (1994) 277-289

287

lactose-positive Salmonella was noted. Further comparisons of this agar to BIS

would be beneficial.
Despite these problems, various combinations of BIS, BGS, H E K and XLD are
recommended for use by various standard methods organizations 1 in Canada and
the United States (Flowers et al., 1992): AOAC, A P H A and F D A recommend BIS,
H E K and XLD; HPB recommends BIS and BGS; ICMSF and ISO recommend
BIS and brilliant green agar (BGA) with a third optional agar; NAS recommends
BIS and BGA; while the U S D A recommends BIS, BGS and XLD.
No studies have been published comparing either EF-18 or RAM to other
standard media in their ability to isolate salmonellae from foods, however, there
are a few papers that discuss the isolation of salmonellae from stools and marine
waters on RAM. Rambach and co-workers (Laudat and Rambach, 1991; Prere et
al., 1992) have compared RAM to H E K in the isolation of Salmonella from stool
samples in several studies. They found that RAM had a sensitivity of 98% and a
specificity close to 100%, was able to increase the number of Salmonella isolates,
and had less false-positives. Dusch and Altwegg (1993) compared RAM to H E K
(and another media not used in this study) in their testing of 504 stool samples via
direct plating for Salmonella. The sensitivities and specificities were 69 and 98%
for RAM and 100 and 79% for HEK. They concluded that RAM should not be
used for primary plating of stool samples because of its lack of sensitivity.
However, they also concluded that RAM might be used for subculturing of
enrichment broth cultures because of its high specificity (compared with HEK) and
that this could substantially reduce the work load in a laboratory. During this
study, it was found that RAM was one of the easiest media to use in purifying
isolates and stock cultures.
Alonso et al. (1992) found that, in general, RAM performed as well as H E K in
the isolation of Salmonella while testing marine waters and that there was no
significant difference in Salmonella recovery between the two media. RAM was
more efficient than H E K when there was low fecal contamination but, in the
presence of high levels of competing bacteria, H E K showed greater selectivity.
Dusch and Altwegg (1993) found RAM to be very specific, having obtained only
10 false-positives from 504 stool samples. These included Escherichia coli, four
Citrobacter freundii isolates, three Citrobacter amalonaticus isolates, Klebsiella
oxytoca and E. agglomerans. Freydiere and Gille (1991) found that Pseudomonas
and Acinetobacter gave false-positives reactions on RAM. Many of these nonSalmonella give false-positive reactions on the other media as well (Table 4).
Flowers et al. (1992) state that none of the generally used agars (BIS, BGS,
HEK, XLD and others) are ideal for all situations, justifying the recommendation
in many reference methods for the use of two or more agar media. Based upon our

1 AOAC, Association of Official Analytical Chemists; APHA, American Public Health Association;
FDA, US Food and Drug Administration; HPB, Health Protection Branch; ICMSF, International
Commission on Microbiological Specifications for Food; ISO, International Organization for Standardization; NAS, National Academy of Sciences (US); USDA, US Department of Agriculture).

288

D.W. Warburton et al. /International Journal of Food Microbiology 22 (1994) 277-289

r e s u l t s a n d t h o s e f r o m o t h e r s t u d i e s , a n d t h e f o l l o w i n g f a c t s : (1) B I S is c u r r e n t l y a
m e d i u m o f c h o i c e i n C a n a d i a n s t a n d a r d m e t h o d s d u e t o its n o n - s a c c h a r i d e
d i f f e r e n t i a l s y s t e m ; a n d (2) n e w l y d e v e l o p e d o r t e s t e d m e d i a m u s t b e e x t e n s i v e l y
f i e l d t e s t e d b e f o r e i n c o r p o r a t i o n i n t o C a n a d i a n s t a n d a r d m e t h o d s u s e d in c o m p l i ance activity, our recommendation
f o r n e w l y d e v e l o p e d l a b o r a t o r y p r o c e d u r e s is
that use of BIS be compulsory and at least one other agar (BGS, HEK, XLD,
R A M a n d E F - 1 8 ) b e u s e d as well.

Acknowledgements
A t h a n k y o u is e x t e n d e d t o all s t a f f w h o a i d e d in t h e p r e p a r a t i o n

of media and

the subculturing of samples to additional test media.

References
Alonso, J.L., Botella, M.S., Amoros, I. and Rambach, A. (1992) Salmonella detection in marine waters
using a short standard method. Water Res. 26(7), 973-978.
Anon. (1984) Difco Manual. Tenth Edn. Difco Laboratories. Detroit, MI.
Anon. (1993a) Rambach agar. Int. J. Food Microbiol. 17, 243-244.
Anon. (1993b) Hektoen enteric agar. Int. J. Food Microbiol. 17, 234-236.
Andrews, W.H., Poelma, P.L. and Wilson, C.R. (1984) Isolation and identification of Salmonella
species. Ch. 7. In: FDA Bacteriological Analytical Manual. AOAC, Arlington, VA.
Bankes, P. (1992) Evaluation of a new Salmonella isolation medium: Rambach agar. Technical
Memorandum No. 651. The Campden Food and Drink Research Association. Chipping Campden,
UK.
Bisciello, N.B.Jr. and Schrade, J.P. (1974) Evaluation of Hektoen Enteric Agar for the detection of
Salmonella in foods and feeds. J. Assoc. Off. Analyt. Chem. 57(4), 992-996.
D'Aoust, J.-Y. (1989) Salmonella. ln: Foodborne Bacterial Pathogens. M.P. Doyle (Ed.). Marcel
Dekker, Inc. NY. pp 328-446.
D'Aoust, J.-Y, Sewell, A.M. and Warburton, D.W. (1992) A comparison of standard cultural methods
for the detection of foodborne Salmonella. Int. J. Food Microbiol. 16(1), 41-50.
Devenish, J.A., Ciebin, B.W. and Brodsky, M.H. (1986) Novobiocin-Brilliant Green-Glucose Agar: new
medium for isolation of salmonellae. Appl. Environ. Microbiol. 52(3), 539-545.
Dickson, J.S. and Anderson, M.E. (1991) Control of Salmonella on beef tissue surfaces in a model
system by pre- and post-evisceration washing and sanitizing, with and without spray chilling. J. Food
Prot. 54(7), 514-518.
Dusch H. and Altwegg, M. (1993) Comparison of Rambach agar, SM-ID medium, and Hektoen enteric
agar for primary isolation of non-typhi Salmonellae from stool samples. J. Clin. Microbiol. 31(1),
410-412.
Entis, P. (1990) Improved Hydrophobic Grid Membrane Filter method, using EF-18 agar, for detection
of Salmonella in foods: collaborative study. J. Assoc. Off. Anal. Chem. 73(5), 734-742.
Entis, P. and Boleszczuk, P. (199l) Rapid detection of Salmonella in foods using EF-18 agar in
conjunction with the Hydrophobic Grid Membrane Filter. J. Food Prot. 54(12), 930-934.
Feng, P. (1992) Commercial assay systems for detecting foodborne Salmonella: a review. J. Food Prot.
55, 927-934.
Freydiere, A.-M. and Gille, Y. (1991) Detection of Salmonellae by using Rambach agar and by a C8
esterase spot test. J. Clin. Microbiol. 29(10), 2357-2359.
Flowers, R.S., D'Aoust, J.-Y., Andrews, W.H. and Bailey, J.S. (1992) Salmonella. Ch. 25. In: Corn-

D. V Warburton et al. /International Journal of Food Microbiology 22 (1994) 277-289

289

pendium of Methods for the Microbiological Examination of Foods. Third Edn. American Public
Health Association. Washington, DC. pp. 371-422.
Gruenewald, R., Henderson, R.W. and Yappow, S. (1991) Use of Rambach Propylene glycol containing
agar for identification of Salmonella spp. J. Clin. Microbiol. 29(10), 2354-2356.
Health Protection Branch (HPB). (1978) Methods for the isolation and identification of Salmonella
from foods. HPB method MFHPB-20. In: Compendium of Analytical Methods. Vol. 2. Polyscience
Publications Inc. Montreal.
Laudat, P. and Rambach, A. (1991) Rambach agar: a new plate medium for rapid and faciliated
identification of Salmonella spp. 5th European Congress of Clinical Microbiology and Infectious
Diseases. September 9-11, (1991) Oslo, Norway. p. 177.
Manafi, M. and Sommer, R. (1992) Comparison of three rapid screening methods for Salmonella spp.:
MUCAP Test, MicroScreen Latex and Rambach agar. Lett. Appl. Microbiol. 14, 163-166.
Moats, W.A. (1978) Comparison of four agar media with and without added novobiocin for isolation of
Salmonellae from beef and deboned poutry meat. Appl. Environ. Microbiol. 36(5), 747-751.
Moats, W.A. (1981) Update on Salmonella in foods: selective plating media and other diagnostic media.
J. Food Prot. 44(5), 375-380.
Nath, E.J., Neidert, E. and Randall, C.J. (1989) Evaluation of enrichment protocols for the 1 2 Test for
Salmonella detection in naturally contaminated foods and feeds. J. Food Prot. 52(7), 498-499.
Oggel. J.J., Nundy, D.C. and Randall, C.J. (1990) Modified 1-2 Test system as a rapid screening
method for the detection of Salmonella in foods and feeds. J. Food Prot. 53(8), 656-658.
Ossmer, R. (1992) Salmosyst and Rambach agar: a rapid alternative for the detection of Salmonella.
Salmonella and Salmonellosis Symposium. September 15-17, 1992, Ploufragan/Saint-Brieuc, France.
Prere, M.F., Rambach, A. and Lareng, M.B. (1992) Rambach agar: a very useful plate medium for the
detection of Salmonella. Salmonella and Salmonellosis Symposium. September 15-17, 1992;
Ploufragan/Saint-Brieuc, France.
Rambach, R. (1990) New plate medium for facilitated differentiation of Sahnonella spp. from Proteus
spp. and other enteric bacteria. Appl. Environ. Microbiol. 56(1), 301-303.
Todd, E.C.D., MacKenzie, J.M. and Peterkin, P.I. (1993) Development of an enzyme-linked antibody
Hydrophobic Grid Membrane Filter method for the detection of Salmonella in foods. Food
Microbiol. 10, 87-99.
Warburton, D.W., Farber, J.M., Powell, C., Tiwari, N.P., Read, S., Plante, R., Babiuk, T., Laffey, P.,
Kauri, T., Mayers, P., Champagne, M.-J., Hunt, T., LaCasse, P., Viet, K., Smando, R. and Coates, F.
(1992) Comparison of methods for the optimum detection of stressed and low levels of Listeria
monoeytogenes. Food Microbiol. 9, 127-145.
Warburton, D.W., Harwig, J. and Bowen, B. (1993a) The survival of salmonellae in homemade
chocolate and egg liqueur. Food Microbiol. 10, 405-410.
Warburton, D.W., McCormick, J.K. and Bowen, B. (1994a) Survival and recovery of Aeromonas
hydrophila in water: development of methodology for testing bottled water in Canada. Can. J.
Microbiol. 40, 145-148.
Warburton, D.W., Oggel, J., Bowen, B., Crawford, C., et al. (1994b) A comparison study of the
modified 1-2 Test and the HPB standard method in the isolation of Salmonella. Food Microbiol. In
press.
Warburton, D.W., Feldsine, P.T. and Falbo-Nelson, M.T. (1994c) Modified immunodiffusion method to
detection Salmonella in raw flesh and highly contaminated food types: collaborative study.
J.A.O.A.C. submitted for publication.
Warburton, D.W., Arling, V., Worobec, S., MacKenzie, J., Todd, E., et al. (1994d) A comparison of
EF-18 agar/Hydrophobic Grid Membrane Filter (HGMF) method and the enzyme linked antibody
(ELA)/HGMF method to the HPB standard method in the isolation of Salmonella. Int. J. Food
Microbiol. Submitted for publication.
Warburton, D.W., Bowen, B., Durzi, S., Lamontagne, G., et al. (1994e) A comparison study of the
Salmosyst method and the HPB standard method in the isolation of Salmonella. In preparation for
publication.