Bacterial growth in seafood on restaurant premises1 NIRMALA VENKATARAMAIAH A N D A . G. KEMPTON

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

Depnrttnetlt of Biology, University of Waterloo, Waterloo, Ontario Accepted July 18, 1975

V E N K A T A R A M A INA.H, and , A. G . KEMPTON.1975. Bacterial growth in seafood o n restaurant premises. Can. J. Microbiol. 21: 1788-1797. Operational guidelines for a chain of the newer type of seafood restaurants and take-out stores were derived from plotsofbacteriological and trimethylamine changesduringthe preparation and storage ofcod fillets, shrimp, and clams. Iffish is to be defrosted at room temperature it should be washed and portioned within 14 h. Subsequent storage at 5 "C should not exceed 72 h and once removed from the refrigerator it should be fried within 3 h. New stores should consider improved methods of defrosting. Trimethylamine content was a more sensitive measure ofquality loss than bacterial growth. Shrimp quality can be improved by emphasizing the way shells are removed because most of the contamination was external. Shrimp shelled while still frozen had the best potential keeping quality. Clam meat is handled less and is always held at 5 OC; but bacterial growth indicated that it should not be held on the premisesfor more than 3 days including the time necessary for defrosting. Batter prepared fresh daily can be left at room temperature. Frying can obliterate poor handling procedures, but adequate cooking is essential under any conditions. Clams are a gourmet item but the practice of cooking them lightly at customer request proved inadvisable. V E N K A T A R A M A IN.. A H .et A. G. KEMPTON.1975. Bacterial growth in seafood o n restaurant premises. Can. J. Microbiol. 21: 1788-1797. Les modes operationels d'une chaine de type nouveau de restaurants ou d e comptoirs d e vente de mets d e rner proviennent de rapports sur les changements bacteriologiques et en trim6thylamine qui prennent place au cours d e la preparation et de I'entreposage des filets de morue, des crevettes e t des palourdes (clams). Si ces aliments doivent t t r e decongeles B la temperature de la piece, ils doivent t t r e laves et repartis en portions en dedans d e 14 h. Un remisage subsequent i 5 "C ne devrait pas exceder 72 h et une fois retires du rkfrigerateur, ils doivent stre frits en dedans de 3 h. Les nouveaux comptoirs d e mets de mer devraient considerer I'application de mithodes ameliorees de decongelation. La teneur en trimethylamine est une mesure plus sensible d e la perte de qualite que la croissance bacterienne. La qualitk d e s crevettes peut stre amilioree en insistant sur le mode de decoquillage vu que la plupart des contaminations sont externes. Les crevettes dCcoquillees alors qu'elles sont encore gelCes ont un meilleur potentiel de conservationde qualite. Lachair des mollusques subit moinsde manipulationset elle est toujours maintenue a 5 "C; mais la croissance bacterienne indique que cette chair ne devrait pas t t r e gardee sur les lieux plus de 3 jours, incluant le temps necessaire a la decongklation. Des portions quotidiennes fraichement prkparees peuvent t t r e conservees a la temperature de la piece. La friture peut camoufler les mauvaises procedures de preparation, mais une cuisson adequate est essentielle sous toutes les conditions. Les mollusques constituent un met de gourmet mais la pratique de les cuire legerement au gofit du client n'est pas recommandable. [Traduit par le journal]

Introduction Family-operated 'fish and chip' shops in Canada have been largely displaced by a growing number of multistore corporations. These new outlets have several problems not encountered by the family shop that warrant the concern of both the owners and food microbiologists. (1) They are open for longer hours to cater to the snack trade in addition to the mealtime business. Typical hours are from 11 :00 a.m. to 12:OO p.m. 6 or 'Received February 3, 1975.

7 days a week. (2) Most offer a variety of seafoods including shrimps, clams, and lobsters in addition to the conventional fish and chips. In this respect they are moving against the trend to less complex, limited-menu restaurants. All products except lobster are obtained in the frozen state and each requires a separate handling regime and possesses different keeping qualities. (3) Cooking may not be as thorough as a result of the displacement of halibut by round fish such as cod and increased emphasis on texture by the consumer. Gourmet items such as clams are frequently

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

VENKATARAMAIAH A N D KEMPTON: BACTERIAL GROWTH IN SEAFOOD IN RESTAURANTS

cooked very lightly at the request of the customer. (4) All operations must be scheduled within a normal working day involving part-time workers or more than one shift, whereas the family store was operated by family members working long hours. (5) Although experienced food handlers are employed, they may lack the intuitive knowledge of the established owner-operator. This work was done with the cooperation of the owner of a chain of modern seafood restaurants and take-out stores to explore the apparent gap in our knowledge that covers the time that seafood is on such premises. The exhaustive studies done on virtually every other aspect of the microbiology of seafood have been fully discussed elsewhere (Chichester and Graham 1973). Bacteriological and sensory changes during defrosting, refrigerator storage, counter storage, and frying of fish, clams, and shrimps were plotted. The results were used to establish operational guidelines for this particular chain of restaurants and take-out stores, but the findings were basic enough for broader application.

Materials and Methods Quantitoriue Bacteriological Methods A 30-g sample of food was homogenized in a Waring Blendor with 270 ml of 1.0% (w/v) peptone water for 2 min. The 30-g sample represented no more than 5 Ib of product and care was taken t o make it as composite as possible. Mesophilic and psychrophilic (or psychrotrophic) bacteria were enumerated on Standard Methods agar (SMA) obtained from Baltimore Biological Laboratories (BBL) after incubation at 30 "C for 18 h and 5 "C for 5 t o 7 days, respectively. The number of coliforms was obtained using violet red bile (VRB) agar (Difco), and enterococci were counted on K F streptococcal agar (BBL). These two media were incubated at 37 "C for 24 h. SPS agar (BBL) was incubated anaerobically in a n Anaerobic Jar Gas Pac (BBL) and suspected colonies of Clostridium perfringens were confirmed by nitrate reduction. Surfaceinoculated plates of Baird-Parker agar (B-PA) (Difco) were used t o count staphylococci, and the coagulase test was always used to confirm the presence of Staphylococcus aureus. Salmoneila were sought by the method outlined in Edwards and Ewing (1972). Identification of Spoilage Organisms Representatives of the most numerous colonial types on SMA incubated at both 5 " C and 30 "C were selected from plates prepared from the highest countable dilution. The isolates were purified on SMA at the original temperature of incubation. Shewan et a/. (1960a, 19606) found that the organisms responsible for the spoilage of fish were primarily members of the genera Pserrdomonas and Achromobacter, and they developed a determinative scheme for the identification of these and related Gramnegative bacteria based o n a limited number of tests. Shewan's scheme was used for Gram-negative bacteria

1789

except that a positive 2,3-butanediol test (Collins 1967) was used t o identify Aeromonas. Conventional IMViC tests were used to separate further enterobacteria isolates and t o confirm the reactions of VRB. The Hugh and Leifson test (Collins 1967) was used to separate micrococci from staphylococci, after the oxidase and catalase tests. I n addition, use was made of colonial morphology and pigmentation as well as the shape, arrangement, and staining characteristics of the cells. Nitrate reduction, arginine hydrolysis, starch hydrolysis, the urease test, and the reactions on triple sugar - iron agar were among other biochemical tests used where applicable. Physical and Chemical Tests The rate of change of temperature during defrosting was measured by a Fluke Digital millivoltmeter attached t o a Honeywell electronic 19 recorder using a copper thermocouple. One arm of the thermocouple was inserted deep into the sample and the other into a thermos flask containing ice. The millivolt recorder was set t o read between - 20 "C and 25 "'C. The line of fit was established by regression equations (Snedecor and Cochran 1967). Trimethylamine (TMA) was determined by the colorimetric method developed by Dyer (1945).

+

Results and Discussion Fish Fillets The chain of restaurants under investigation uses cod fillets which are received frozen and held in a central plant at - 18 "C or colder as recommended by the AFDOUS (Anonymous 1962). Individual stores receive their fish in 50-lb frozen blocks which are left to defrost at room temperature from midnight until after noon the next day. The blocks consisted of 10 individually wrapped 5-lb units. That this 12- to 14-h interval should fit the work schedule was fortuitous but quite critical, as shown in Fig. 1. There was no significant increase in bacterial numbers during the first 14 h when a 5-lb unit was defrosted at room temperature (Fig. la). After 14 h, all major groups of bacteria proliferated rapidly. Streptococcusfaecalis appeared first after 16 h and when the defrosted fish was left at room temperature for 24 h Clostridium perfringens appeared. For practical purposes, fish is 'thawed' when the large fillets can be cut into portion-sized pieces. On the temperature plot (Fig. Ib), this point was not reached for 12 h, at which time the innermost part of a 5-lb unit was 15 "C. It is noteworthy that the bacterial count did not increase throughout the defrosting period but only after complete thawing had occurred. The use of TMA as an index of fish quality is not universally applicable. For example, Tarr and Sunderland (1940) found that it was possible

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

1790

CAN. J. MICROBIOL. VOL. 21, 1975

. - .

THAWED

F

O

0

. 2

O 4

I

u ~ I I I 6 8 10 12

I 14

I 16

10 h of defrosting (Fig. 1c). This was found to be due to the 'drip' of proteinaceous material. When trichloroacetic acid was added to the collected dripping, protein was precipitated and a T M A test showed that it contained 4 mg of TMA/100 ml. The TMA content increased rapidly after the first 10 h, slightly ahead of complete thawing and the onset of bacterial growth. The difference in temperature between the warehouse and the fish as received in the store could be reduced. By extrapolation of the temperature curve (Fig. 1 b) to - 18 "C, it is estimated that an additional 4 h would be required for defrosting. This would upset the work schedule by imposing the washing and cutting of the thawed fish on the busiest part of the working day. However, should the initial temperature be above - 9 "C on occasion, thawing would be complete before the store was open. If done at room tem-

I I I I 18 20 22 24

HOURS FIG.1. Changes during the defrosting of cod fillets at room temperature. (a) Bacterial numbers, (b) temperature, (c) trimethylamine content. Symbols: A, total count (5 "C); M, total count (30 "C); @, S. aureus; A, coliforms; 0 , S. faecalis; 0 , C. perfringens.

to obtain very stale fish without the formation of appreciable amounts of TMA and Cobb et al. O I2 24 % 48 60 72 84 96 [OB 1Z3 132 HOURS (1 973) reviewed the deficiencies of the Dyer during method of TMA analysis. ~ ~ of de- ~ FIG.2. Changes ~ ~ storaged at 5 "C ofl cod fillets~ previously defrosted at room temperature. (a) bacterial ficiencies in the method, in this work it was numbers, (b) trimethylamine content. Symbols: A, total significant that the TMA content dropped from count (5 'C); total count (30 "C); @, S. aureus; A, 12 to less than 4 mg/100 g of fish during the first coliforms; 0 , S. faecalis; 0 , C. perfringens.

.,

1791

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

VENKATARAMAIAH AND KEMPTON: BACTERIAL GROWTH I N SEAFOOD IN RESTAURANTS

perature, the defrosting operation requires the careful attention of the store owner because the interval between complete thawing and the onset of spoilage is too short to permit errors. In practice, thawed blocks of fillets are washed, cut into serving portions, and placed under refrigeration at 5 "C in trays fitted with wire screens to keep the fish out of any water that may accumulate. Since -demand is not predictable, the shelf life of the refrigerated portions is important. According to Fig. 2 the maximum shelf life was 3 days. Shelf life was judged to be terminated soon after the sudden increase in TMA content that occurred at 72 h (Fig. 2b) rather than by bacterial growth (Fig. 2a); Spoilage was obvious organoleptically at 84 h when the TMA content was 80 mg/100 g of fish. The total number of bacteria increased only by 1 log cycle over 5 days (108 h), while the number of coliforms and staphylococci remained relatively constant. Dyer and Dyer (1949) found that the shelf life of fish fillets could be increased from 3 to 8 days by lowering the temperature from 5 to 0 "C. Three days should be sufficient for most retail operations; but, as with the defrosting operation, the end point was sudden rather than gradual. The variability encountered in a study on a commercial site is evident in a comparison of Fig. 1 with Fig. 2. The tray of fish pieces used to derive Fig. 2 was taken from the restaurant during an unsheduled visit. Had the TMA value been the only parameter studied, it would be concluded that the fish was cut after 10 h of defrosting whereas the bacteriological results would indicate that the defrosting interval was about I8 h. It was considered advisable to conduct a study of the next phase of the operation, during which fish is left at room temperature (about 20 "C consistently) on a counter adjacent to the fryer, on a tray that had been kept in the refrigerator for the absolute maximum time of 84 h. Although spoilage is evident by odor after 84 h, it is still within the '3rd day' when restaurant hours are considered. The initial values in Fig. 3 proved to be close to those predicted from the 84-h values in Fig. 2. When removed from the refrigerator, the portion-sized pieces of fish warmed to room FIG.3. Changes during counter display at room temperature of cod fillets previously defrosted a t room temperature and then stored at 5 "C for 84 h. (a) Bacterial numbers, (b) temperature, (c) trimethylamine content. Symbols: A, total count (5 "C); total count (30 "C); 0, S.aureus; A,coliforms; 0 ,S.faecalis; 0, C.perfringens.

.,

(a) I-

6 8~ L

k

5-

5 4-

5

.?

n

0 25

"u

o

o

n

-

n

( b)

20-

W

52

15-

5 I-

lo 5

o

I

I

I

I

110-

I

(c

2 I

k

70-

8

0

I

2

3 HOURS

4

5

6

1792

CAN. J . MICROBIOL. VOL. 21, 1975

TABLE 1 The effect of frying batter-dipped seafoods at 160 "C for varying periods of time on bacterial numbers and trimethylarnine content Media used to count survivors

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

Sample Fish

Shrimp

Clams

Cooking time

VRB

KF

B-PA

SPS

TMA, SMA (30"C) SMA (5"C) rng/100 g

Raw 1 min 45 s 2 min 25 s 3 min 25 s

8x lo8* 0

6x lo6 0

8x10' 3x102

0 0

8 x lo8 < 10

10 x lo8 0

80 1.7

0

0

0

0

100

0

1.6

0

0

0

0

0

0

1.7

Raw 1 rnin 30 s 2 min 3 min 10 s Raw 1 min 1 rnin 20 s 2 rnin 15 s

3x103 0

3x10' 380

0 0

5x lo6 lox lo3

5 x lo7 13

-

10 0

0 0

60 10

0 0

-

2 . 8 lo5 ~ 60 40

0 20 0

2 x lo5 20 7

-

2

-

0 0

7.5x103 10 0 0

0 0 0

16x lo2 100 < 10

0

20

0

0

2.5~ lo5 lo3 40 20

-f

'Bacteria per gram. tNot done.

temperature in 2 h (Fig. 3b). During this period, 2 rnin 25 s, but TMA was negligible after only the TMA content dissipated from 90 to 30 mg/ 1 rnin 45 s at 160 O C . Undue bacterial growth and TMA production 100 g of fish (Fig. 3c) and the bacterial numbers apparently declined (Fig. 3a). After 3 h the TMA can be prevented by defrosting in the refrigerator content rose sharply, which proved to be a better for 24 h, as shown in a laboratory experiment estimate of a change in organoleptic properties illustrated by Fig. 4. Except for Streptococcus (odor) than the relatively sluggish changes in faecalis, all groups of organisms were within one bacterial numbers. Once removed from the re- log cycle of their initial values after defrosting, 3 frigerator, the tray of fish should not be kept at days of refrigerator storage, and 6 h at counter room temperature for more than 3 h, regardless temperature. However, this procedure would require additional refrigerator space. Armstrong et of demand. Immediately before serving, the fish is dipped al. (1960) compared six defrosting procedures in batter and deep fried in vegetable shortening and found that submersion in tap water for 50 at 160 "C. Table 1 contains the results of three rnin consistently resulted in a palatable cooked frying regimes on fish that had been defrosted for product. Jason (1964) suggested that a small re14 h at room temperature, stored under refrigera- tail or catering establishment could adopt a tion for 84 h, and then left at room temperature method involving electrically heated electrodes. for 33 h. The initial bacterial load and TMA Adoption of either of these methods could decontent were as anticipated from Fig. 3. Al- crease the amount of refrigerator storage rethough this sample had been subjected to storage quired since the defrosting rate is rapid enough t o times in excess of recommended maxima, the respond to unforeseen changes in consumer denormal 3 rnin 25 s frying time sterilized the fish mand within a single working day. and virtually eliminated TMA. A few aerobic Barring the installation of additional equipspore-forming organisms survived frying for ment for defrosting, the conclusions were sum-

1793

VENKATARAMAIAH A N D KEMPTON: BACTERIAL GROWTH IN SEAFOOD IN RESTAURANTS

-

5 %STORAGE

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

THAWED WASHED

,&-.

W Le

R

5

: 2 , : d

2-

/'

//

//

R / \

\\ \\

'

\\

/

\,

b'

//

I

"90

/'

,

\

/'

\ \\

\

h, ,'/

FIG.4. Changes in cod fillets during defrosting for 24 h at 5 "C, storage at 5 "C for 72 h and counter display at room temperature for 6 h. (a) Bacterial numbers, (b) trimethylamine content. Symbols: A, total count ( 5 "C) total count (30 "C); 0, S. arrrecrs; A, coliforms; 0, S. faecalis; , C. perfringens.

.,

gram. The initial total count was composed primarily of micrococci, coliforms, and Proteus with lesser proportions of Achromobacter, Ftavobacter, and Pseudomonas (Fig. 50). Raj et al. (1961) reported that fish products often contain large numbers of faecal coliforms which Varga and Anderson (1968) suggested may not originate directly from human faecal matter since they survive and proliferate on improperly sanitized working surfaces. Salmonellae were never detected. Stapl~ylococcusaureus was not isolated from SMA plates even though the use of a selective medium showed that it was present in relatively high numbers. This observation is consistent with the reputation that staphylococci have as poor competitors. Clostridium perfringens was not detectable initially but appeared later in the handling process even when the fish was kept under refrigeration during defrosting. The initial T M A content of the block of cod fillets from

(bl 100 W

'As a result of this study, this particular restaurant now buys 10-lb blocks of cod fillets and defrosts them at 5 "C.

80

0 J

'

60

p

40

f =

20

8

lnarized as follows. Defrosting at room temperature should be completed within 14 h. Subsequent refrigerator storage should not exceed 3 days; and, once removed from the refrigerator, a tray of fish should be used within 3 h.2 These guidelines naturally reflect the quality of fish that can be obtained on the wholesale market and the store owner has n o control over the initial bacterial load or T M A content. Although the total bacterial counts were below the lo5 per gram recommended by AFDOUS (Anonymous 1962), the numbers of coliforms and staphylococci were significantly higher than their respective recommended maxima of 200 and 100 per

ACHROMOBACTER

VI

2

MICROCOCCI PSEUDOMONADS BACILLUS COLIFORMS

n PROTEUS

FLAVOBACTERIUM

0

FIG.5. Predominant groups of bacteria on seafood as received in a retail establishment and isolated from Standard Methods agar. (a) Cod fillets, (b) shrimp, (c) clams.

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

1794

C A N . J . MICROB IOL.

which Fig. 1 was derived grossly exceeded the value of 1 to 4 mg/100 g of tissue suggested as normal by Dyer and Mounsey (1945). The frozen fish used in the laboratory experiment was within the recommended limits (Fig. 4), which shows that the retail merchant has to accept some variation in this regard. However, unless the criteria for defrosting, refrigerator storage, and counter exposure are adhered to strictly, T M A will exceed the 30 mg/100 g limit that Castell et al. (1954) used to indicate spoilage. It has been shown that frying obscures the previous history of both bacterial growth and TMA development. Merchants must regard adequate cooking as a necessary safety measure but hot as a means of covering up faulty procedures. The discovery of overt pathogens such as Staphylococcus aureus, ~lostridiumperfringens, and Streptococcus faecalis at various stages of the retail process justifies the establishment of cautious guidelines. However, in this content it is worth reporting that no food poisoning incidents have been reported during the 4 years this chain of restaurants has been in operation.

VOL.

oh 0

21, 1975

n 12

I

24

n 36

I

n

I

48

60

72

n

84

HOURS

FIG.6. Bacterial numbers during storage of shrimps a t 5 "C that had previously been defrosted at room tempera, total count (30 "C); ture for 2 h a n d washed. Symbols: . a, S. aureus; A, coliforms; 0, S. faecalis; 0,C . perfiinge~zs.

Shrimps Shrimps are received at the retail outlet in a frozen block in a 5-lb carton. The heads had been 1949c) and one by Williams et al. (1952) indiremoved, which according to Fieger (1950) could cated that a large proportion of the total flora reduce the bacterial count of whole shrimp by as and coliforms in particular were carried on the much as 75%. Defrosting was accomplished by shells and appendages. Consequently, two laborimmersing the entire 5 lb in tap water at room atory experiments were performed in which emtemperature for about 1 h, after which the shrimp phasis was placed on the short defrosting and was shelled by hand and transferred to a fresh washing cycles. In Fig. 7a the shrimps were detray of tap water in the refrigerator. As required, frosted and shucked in the normal manner. The individual shrimps were blotted dry on a piece of coliform count increased during the 2-h defrostcloth, dipped into batter, and deep-fried at ing period during which the temperature rose 160 "C. from - 5 "C to + 10 "C, but after being shelled Since the defrosting period was minimal, em- and washed the coliform level dropped to less phasis was first placed on the refrigerated storage than 10 per gram throughout a further 9 h at 5 " C period of the shelled product. The results shown storage. However, the total count on SMA inin Fig. 6 were unexpected. At the start of the cubated at 30 " C remained near lo6 per gram. storage period, Fieger and Novak (1961) would When the shrimps were shelled while still frozen, have rated the quality of these shrimps between it was possible to eliminate coliforms (Fig. 76). fair and poor on the basis of bacterial numbers In addition, the total count was reduced to lo5 (between 10.7 x lo6 and 19.0 x lo6 per gram). per gram for a t least the first I 1 h of storage. As In less than 2+ days they were badly spoiled. All with fish, the predominate organisms comprising shellfish, whether crustacean or molluscan, con- the total count of washed, shelled shrimps were tain far greater amounts of free amino acids than micrococci (Fig. 56). And again, as with fish, fish, which greatly facilitates bacterial growth frying could be used to overcome deficiencies in and probably explains their ready spoilage handling (Table 1). (Velankar and Govindan 1957, 1958; Ranke It is recommended that shrimps be shelled as 1959). A series of papers by Green (1949a, 19496, soon as possible, preferably before they are corn-

1795

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

VENKATARAMAIAH AND KEMPTON: BACTERIAL GROWTH IN SEAFOOD IN RESTAURANTS

Clams Molluscan shellfish differ in their chemical composition from both teleost fish and crustacean shellfish in that they contain a significant amount of carbohydrate and a lower total quantity of nitrogen in their flesh. The carbohydrate is primarily glycogen, hence fermentative activities may contribute to microbial spoilage (Fieger and Novak 1961). It is also acknowledged that the microflora of molluscan shellfish varies considerably depending on the quality of water from which they are taken because of their mode of feeding. It was established early (King et 01. 1945) that the degree of washing influenced the course of decomposition. A thorough cleanwater washing removes putrefying organisms preferentially over fermentative bacteria which enhances the natural tendency for acid formation. Therefore, while Shewan (1961) lists Pseudomonas and other Gram-negative bacteria as the organisms primarily responsible for the spoilage of teleost fish, Bacillus was recognized as a predominant organism in the spoilage of molluscan shellfish as early as 1929 (Rice 1929). Clams also differ from other seafood products in that they are usually removed from their shells before being frozen in blast or tunnel freezers to - 18 There is an apparent lack of knowledge on the effect of frozen storage on clams, the effect of defrosting, and the nature of the organisms causing spoilage of defrosted clams. The retail outlet obtains clam meat in 2-lb polythene bags, which can be thawed at 5 for 4 h or in tap water in 15 to 30 min. After they are thawed, the clams are portioned into plastic cups S O 0 O and kept under refrigeration. They are dipped in 0 1 2 3 4 5 6 7 8 9 1 0 1 1 batter before frying at 160 "C. Figure 8 shows HOURS the growth of bacteria during defrosting and subFIG.7. (a) Bacterial changes during the normal sequent storage at 5 "C. The total count rehandling of shrimp during which they are shelled after defrosting for 1 h and then stored a t 5 "C. (b) Bacterial mained below the recommended USPHS maxichanges during storage of shrimp at 5 "C that had been mum of 50 000 per gram (Anonymous 1954) for shelled while frozen. Symbols: .,total count (30 ' C ) ; 0,S. only 48 h. Unlike fish and shrimps, the bacteria aureus; A, coliforms; 0 ,S. faecalis; 0, C. perfringens. proliferated steadily during refrigerated storage, reaching an unacceptable level in 4 days. Coliforms were absent, as were salmonellae, but the pletely defrosted, and that special care be taken growth of staphylococci from an initial undetectin their washing. Care should be taken not to able level to the predominant organism is cause drop any shells into the shelled tray. These mea- for concern. Once again, frying for 2 min 15 s sures are aimed at reducing the external contami- eliminates most organisms (Table I). However, nation. Defrosted shrimps should not be kept in clams cooked for only 1 min contained a residual the refrigerator for more than 48 h, and they number of bacteria, hence this practice should be should be removed from the refrigerator only as discontinued. The initial population of bacteria needed. capable of growing on SMA at 30 and 5 O C .

O

O

C

C

O

C

1796

CAN. J. MICROBIOL. VOL. 21, 1975

used within 3 days, including the time required for d e f r ~ s t i n g . ~

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

Batter

The batter was prepared in 3-lb lots each morning and left on the counter during the day. The mix contained yellow corn flour (2 lb), high loaf flour (1 Ib), monosodium glutamate, pepper, and salt as flavoring agents and enough baking powder to add crispness to the final product. Water was added and mixed with the other ingredients with a wire wisk to produce the desired consistency. Monitoring the bacterial content of batter over a 12-h work day provided the data for Fig. 9. Although the total count exceeded lo3 per gram only after 12 h, coliforms, Staphylococcus a u r e u s , Streptococcus o F ~ A ~ + T + T + ? faecalis, ~ ? Aand fungi (counted on Saboraud's dexh ; N &r nr *n g p * r wn m r ng ho l ~ n * s trose agar) a t levels of 100 to 200 were present. HOURS Whether this was due to the raw ingredients or FIG. 8. Growth of bacteria in clam meat during de- unsanitary practices was not determined at this frosting and storage at 5 OC. Symbols: A, total count (5 time because the relatively small number of these "C); B, total count (30 "C); a, S.aweus; A, coliforms; organisms in the batter compared with the num0, S. faecalis; , C. perfringens. ber of organisms in the basic seafood did not appear to be cause for alarm.

Acknowledgment This investigation was supported by National Research Council of Canada grant A-3644.

n 0 1

l

n

i

I

n~

I

n~

l

r

l

2 3 4 5 6 7 8 9 1 0 1 1 1 2 HOURS

FIG. 9. Bacterial numbers in frying batter prepared and held- at room temperature for 12 h. Symbols: A, total count (5 "C); B, total count (30 "C); a, S. aureus; A, coliforms; 0 , S. faecalis; 0, C. perfiingens; 0, fungi.

(Fig. 5c) contained a large proportion of Bacillus as predicted by Rice (1929). The shelf life of frozen clam meat is very limited, even though it is always kept at 5 "C or below. It was recommended that this product be

ANONYMOUS. 1954. Recommended procedures for bacteria counts of shucked oysters. Shellfish Branch, Milk, Food and Shellfish Sanitation Program. Division of Sanitary Engineering Services. U.S. Public Health Serv. Publ. ANONYMOUS. 1962. AFDOUS Frozen Food and Drug Code. Assoc. Food Drug Off. U.S. Q. Bull. 26: 2 5 4 2 . ARMSTRONG, I. L., E. W . PARK,B. A. MCLAREN, and D. S. PARKER. 1960. The effect of time and temperature of cooking on the palatability and cooking losses of frozen atlantic cod fish fillets. J. Fish. Res. Board Can. 17: 1-7. CASTELL, C. H., W. A. MACCALLUM, and H. E. POWER. 1954. The influence of the quality of the fish in the fish plant on the subsequent keeping time in retail stores in Ontario. Atl. Fish. Exp. Stn. Note No. 144. 1973. Microbial CHICHESTER, C. O., and H. D. GRAHAM. safety of fishery products. Academic Press, Inc., New York. COBB,B. F., I. ALANIZ,and C. A. THOMPSON. 1973. Biochemical and microbial studies on shrimp: volatile nitrogen and amino acid analysis. J. Food Sci. 38: 431436. COLLINS, C. H. 1967. Microbiological methods. 2nd ed. Butterworth & Co. (Publishers) Ltd., London. 3The restaurant studied now uses partially cooked, prebreaded, preportioned clams that are kept frozen. They are cooked for 2 min at 160 OC as required.

Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 11/13/14 For personal use only.

VENKATARAMAIAH AND KEMPTON: BACTERIAL GROWTH IN SEAFOOD IN RESTAURANTS

DYER,F. E., and W. J. DYER.1949. Changesin palatability of cod fillets. J. Fish. Res. Board Can. 7 : 449460. DYER,W. J. 1945. Amines in fish muscle. I. Colourimetric determination of trimethylamine as the picrate salt. J. Fish. Res. Board Can. 6 : 35 1-358. DYER,W. J., and Y. A. MOUNSEY. 1945. Amines in fish muscle. 11. Development of trimethylamine and other amines. J. Fish. Res. Board Can. 6 : 359-367. EDWARDS, P. R., and W. H. EWING.1972. Identificationof Enterobacteriaceae. 3rd ed. Burgess Publishing Company, Minneapolis, Minn. FIEGER, E. A. 1950. Problems in handling fresh and frozen shrimp. Food Technol. 4: 409413. FIEGER, E. A., and A. F. NOVAK.1961. Microbiology of shellfish deterioration. Irr Fish as food. Vol. I. Edited by G. Borgstrom. Academic Press, Inc., New York. pp. 561-61 1 . GREEN, M. 19490. Bacteriology of shrimp. 11. Quantitative studies of freshly caught and iced shrimp. Food Res. 14: 372-383. 19496. Bacteriology of shrimp. 111. Quantitative studies of frozen shrimp. Food Res. 14: 384-394. -1949c. Bacteriology of shrimp. IV. Coliform bacteria in shrimp. Food Res. 14: 395400. JASON,A. C. 1964. Selection of thawing methods. Reprinted from Proc. Meet. Fish Technol. Fish Handl. Pres. Scheveringer. pp. 191-202. KING,W. H., F. F. FLYNN, and J . N. GOWANLOCH. 1945. Experimental studies on decomposition of oysters used for canning. J. Assoc. Off. Agric. Chem. 28: 385-398. RAJ, H., W. J. WEIBE,and J . LISTON.1961. Detection and enumeration of faecal indicator organism in frozen sea foods. 11. Enterococci. Appl. Microbial. 9: 295-303. RANKE,B. 1959. Uber die nicht-eiweissgebundenen und eiweissgebundenen Aminosaurenbestande von Fischen,

1797

Mollusken und Krebsen. Arch. Fischereiwiss. 10: 117-159. RICE,C. E., 1929. The decomposition of clam muscle in acid solutions. Contrib. Can. Biol. Fish. 4: 95-105. SHEWAN, J . M. 1961. The microbiology of sea-water fish. In Fish as food. Vol. I. Edited by G. Borgstrom. Academic Press, Inc., New York. pp. 487-560. S H E W A NJ., M., G. HOBBS,and W. HODGKISS. 19600. A determinative scheme for the identification of certain genera of gram-negative bacteria, with special reference to the Pseudomonadaceae. J. Appl. Bacteriol. 23: 379-390. 1960b. The Pserrdornor~ns and Achrornobocter groups of bacteria in the spoilage of marine white fish. J. Appl. Bacteriol. 23: 463468. SNEDECOR, G. W., and W. G. COCHRAN. 1967. Statistical methods. Iowa State Univ. Press, Ames, Iowa. pp. 135-171. TARR,H. L. A,. and P. A. SUNDERLAND. 1940. The comparative value of preservatives for fresh fillets. J. Fish. Res. Board Can. 5: 148-163. VARGA,S., and G. W. ANDERSON. 1968. Significance of coliforms and enterococci in fish products. Appl. Microbiol. 16: 193-196. VELANKAR, N. K., and T . K. GOVINDAN. 1957. The free amino acid nitrogen content of the skeletal muscle of some marine fishes and invertebrates. Curr. Sci. 26: 285-286. 1958. A preliminary study of the distribution of non-protein nitrogen in some marine fishes and invertebrates. Proc. Indian Acad. Sci. Sect. B, B47: 202-209. WILLIAMS, D. B., L. L. CAMPBELL, JR., and H. B. REES, JR. 1952. The bacteriology ofgulfcoast shrimp. 11. Qualitative observationson the external flora. Texas J. Sci. 4: 53-54.

Bacterial growth in seafoood on restaurant premises.

Bacterial growth in seafood on restaurant premises1 NIRMALA VENKATARAMAIAH A N D A . G. KEMPTON Can. J. Microbiol. Downloaded from www.nrcresearchpre...
555KB Sizes 0 Downloads 0 Views