Public Health Aspects of Cream-filled Pastries. A Review

Transcription

1 289 J. Milk Food Techno/. Vol. 39, No. 4, Pages (April, 1976) Copyright 1976, International Association of Milk, Food, and Environmental Sanitarians Public Health Aspects of Cream-filled Pastries. A Review FRANKL. BRYAN U.S. Department Health, Education and Welfare, Public Health Service Center for Disease Control, Bureau of Training Atlanta. Georgia (Received for publication September 8, 1975) ABSTRACT In the United States food borne disease outbreaks for which creamfilled pastry was identified as the vehicle have declined from 17.8% in the 1930's to 2.3o/oin the 1970's. Cream fillings or cream-tilled pastries were usually contaminated with staphylococci and Salmonella typhi by workers and with salmonellae by ingredients such as eggs and milk. These contaminants multiplied as a result of favorable nutrients, water activity, temperature, and ph; and they survived the effects of competing organisms and processes. Prevention and control of such outbreaks have been based on formulating the product so it will not mpport bacterial growth, using pathogen-free ingredients, thorough cooking and reheating, sanitary handling of fillings and finished pastry, rapid cooling and storing at low temperature, establishing microbiological standards for the finished products, training food-processing and food-service personnel, and educating the consumer to refrigerate products after purchase. EPIDEMIOLOGY Cream-filled pastries have a notorious history of being vehicles of food borne illnesses, particularly staphylococcal food poisoning (16, 18, 44). If this history is examined for years in which records are available, a decreasing percentage of outbreaks in which cream-filled pastries were vehicles occurred during each successive decade (fable 1)_ From 1961 to 1971, for instance, cream-filled TABLE 1. Percent of foodborne disease outbreaks in which cream ;tilled pastries were vehicles in reported outbreaks from I 938 to 1973a Outbreaks Years Number Total Pereent 1938, , aoata for 1940 and 1951 unavailable. blncludes outbreaks in which vehicles were and were not detected. pastries were vehicles in only SO (9.So/o of) reported outbreaks of staphylococcal intoxication (Fig. 1). Whereas in previous decades they were usually the most commonly reported vehicles of this disease (J 8). For the years covered in Table 1, cream-filled pastries were frequently vehicles of staphylococcal intoxication outbreaks (fable 2). A number of outbreaks of salmonel- Egg and egg products Milk and milk products Potato and macaroni salad Cream-filled pastry Other foods and mixtures* Chicken Turkey Beef Pork Combinations *59 outbreaks in which the incriminated food was not determined have been om1tted. Figure 1. Foods incriminated in staphylococcal food poisoning outbreaks reported from 1961 to TABLE 2. Diseases conveyed by contaminated cream-filled pastry in the United States (selected periods from 1938 to 1972 in which surveillance data was available and outbreaks reported in literature from 1916) Number of Disease Outbreaks Percent Staphylococcal intoxication Salmonellosis (caused by various serotypes) Typhoid fever Enterococcal gastroentritis Viral hepatitis Food borne illness of unknown etiology Total 573 losis, a few outbreaks of typhoid fever and enterococcal gastroenteritis, and one outbreak of viral hepatitis were also attributed to these vehicles_ A number of outbreaks were classified as unknown etiology. Many of these reports gave clinical pictures typical of staphylococcal intoxication, but they contained insufficient laboratory confirmation to classify the disease. FACTORS CONTRIBUTING TO OUTBREAKS Conditions necessary for cream-filled pastries to become vehicles for foodborne pathogens and to allow these pathogens to multiply to such an extent that illness results when they are ingested are: (a) pastries or their

2 290 BRYAN ingredients become contaminated with pathogens; (b) they contain nutrients which support growth of pathogenic bacteria; (c) the filling or the cake or crust has a water activity that is high enough to permit growth of pathogenic bacteria or the water activity differential of these causes migration of water to create pockets of moisture that permit such growth; (d) ph of the filling is such that pathogenic bacteria can survive and can multiply; (e) oxidation-reduction potential of the filling or topping is within a range so that pathogenic bacteria can initiate growth; ~1 temperatures at which the filling or the finished pastries are held are within a range that pathogenic bacteria multiply for sufficient time to permit development of large numbers of these bacteria or elaboration of enterotoxins; (g) pathogenic bacteria can successfully compete with non-pathogenic microbial flora in the product; and (h) pathogenic bacteria (or toxins) must survive all operations to which they are exposed. If these conditions are met and the cream-filled pastry is ingested by susceptible persons, foodborne illness will result. The likelihood of the eight conditions occurring can be assessed by looking at each in greater detail. Sources of contamination Foodborne pathogens that reach cream filling or pastry made from such filling come from two major sources~workers and foods of animal origin. The relative importance of sources of contamination of cream-filled pastry as indicated by outbreak investigations is shown in Table 3. Enterotoxogenic TABLE 3. Factors contributing to 367 outbreaks in which cream-jilled pastries were vehicles (covers periods from 1920's to 1972)a Inadequate n"n"'"ra1nnn Infected worker Prepared a day or more before serving Insanitary plant Workers with poor personal hygiene handling food Inadequately cleaned equipment Contaminated raw ingredients (eggs and milk) Cooking or reheating failures Rodent contamination(?} construction a Many reports did not give complete information of the factors that contributed to the outbreak, and laboratory analysis was often not done. strains of Staphylococcus aureus are usually introduced into cream fillings by infected workers after the pastry has been cooked. The workers may have infected burns or cuts on their hands or they may have boils; but usually they have no lesions, they just carry the staphylococci in their nares (23, 34). In a few outbreaks reported before 1950, staphylococci were introduced into fillings when raw milk was added. Salmonellae have frequently entered cream fillings or custards with raw ingredients. Broken, cracked, or checked eggs and unpasteurized frozen and dried eggs have been the main sources. Other raw ingredients such as cocoa, dry milk, or coconut, however, may also be contaminated with salmonellae. Cross contamination of cream from contaminated dried eggs was suggested as the source of Salmonella typhimurium in creamfilled sponge cakes that were epidemiologically and laboratory incriminated during an outbreak investigation (15). In reports made in the 1920's and 1930's, rodents were presumed to be sources of salmonellae; but these reports appeared in a period when presence of certain serotypes of Salmonella in foods was attributed to rodents by some investigators whether or not there was epidemiologic or laboratory evidence to support such a hypothesis. This hypothesis has not been substantiated in recent years. Hepatitis-A virus and Salmonella typhi have been introduced into cream fillings after cooking during handling by workers. Nutrients Cream fillings used in pastries may be made of eggs, milk, shortening, sugar, cornstarch or flour, salt, vanilla, and water or of just a "synthetic'' mix consisting of homogenated vegetable oil, emulsifying agents, sugar, salt, and water. If the water activity is high enough, the first type of filling and some brands of the second type of filling will support growth of S. aureus (13), Salmonella. and other foodborne pathogenic bacteria. S. aureus requires up to eleven amino acids, and thiamine and nicotinic acid to grow, but salmonellae only require simple sugars and inorganic nitrogen compounds to grow. When water was substituted for milk when reconstructing dry-mix pudding, S. aureus growth was inhibited (13). When milk, eggs, or butter was added to synthetic cream fillings, staphylococcal growth was enhanced (13, 17, 22). Water activity Availability of water in a food or medium to microorganisms is known as water activity <aw). The water activity value is the point where the relative humidity of an atmosphere around a food neither gains nor loses water. Thus, the two humidity values are in equilibrium, and the vapor pressure of water in the food is the same as that in the atmosphere. A more detailed description of water activity can be found in papers by Scott (39) and Troller (45). Water activity is decreased by substances that dissolve in water. Items of high molecular weight, such as flour, eggs, and milk have relatively little effect on reducing water activity because their molecules are large and the number dissolved in a unit of water is small. Sucrose can appreciably decrease water activity because its molecular size is relatively small. The same amount of dextrose will reduce water activity even more because its molecule is about half the size of the sucrose molecule. The same amount of sodium chloride will decrease water activity even more because of the small size of its molecule and consequently more surface area. Solutes decrease vapor

3 PUBLIC HEALTH ASPECTS OF PASTRIES 291 pressure so that fewer water molecules go into the air above a solution. Thus, the air above such a solution has a lower relative humidity than the air above pure water. Because of the amount of sugar that is added to cream filling, the water activity is relatively low; too low to support growth of many microorganisms. A water activity of below will prevent growth of Salmonella. A water activity below 0.86 will prevent growth of S. aureus and below 0.90 will prevent enterotoxin production by this organism. Fillings made of cooking fat, emulsifying agents, sugar, and salt (no protein) did not support growth of S. aureus (22). When water only was added to some brands of synthetic fillings, however, the fillings supported growth of S. au reus (17. 32). When these fillings were put in pie crusts, S. aureus grew profusely at the interface between the crust and the filling (17). Silliker and McHugh (41) observed that S. aureus could grow neither in the crumb portion of devil's food cake nor in butter-cream fillings containing as low as 1.8 parts sucrose to one part water. Water activity in both was low enough to inhibit growth of S. aureus. But staphylococci grew in cream fillings containing as much as 2.7 parts sucrose, dextrose, or invert sugar to one part water when these fillings were put between layers of the devil's food cake. This growth was attributed to migration of water from the cake to the interface of the cake and filling and to creation of localized areas of high water activity. Rolfes (37) found that staphylococci grew in inoculated fillings of chocolate cup cakes when the sugar to water ratio was 1.83:1 or lower but not at 2.20:1 and of sponge cake fillings when the sugar to water ratio was 1.53:1 but not as 1.83:1. There is marked dehydration on the top surface of custard pies. Top surface inoculum of S. aureus showed a 24-h lag before growth commenced (36). Cut surfaces also showed dehydration, but this caused less inhibition. In general, it is difficult to predict whether or not particular fillings, icings, or toppings will support staphylococcal growth. because migration of water can result in creating areas of higher water activity than that of either of the components (40). ph The optimum ph for growth of salmonellae and staphylococci is near 7. The ph of commercial custard mixes used by bakeries is close to the optimum ( ) for these organisms (12). Growth of staphylococci is prevented at ph 5, and growth of salmonellae is prevented at a ph of 4.5. Oxidation-reduction potential Staphylococci and salmonellae are aerobic, facultatively anaerobic in regard to their oxygen needs. Thus, the oxidation-reduction potential of cream fillings should be within a suitable range for these pathogens to initiate growth. Temperature The temperature range for which staphylococci can grow in custard is 46 to 114 F (6). Salmonellae did not grow in custard between 40 and SO F, but they grew at 114 F (6). Optimal temperature for multiplication of these pathogens is near 100 F. Surkiewicz (43) reported that coliform, coagulase-positive staphylococci, and aerobic colony counts increased greatly when imitation-cream pies were stored at room temperature for 24 and 48 h. Post et al. (35) reported that S. aureus increased in cream filling to between 10 5 to 10 7 /g when stored under conditions to simulate delivery temperaturt: for 10 h. If cream-filled pastries or their fillings are left in the temperature range in which foodborne pathogens can grow for sufficiently long periods, these pathogens multiply and in the event of enterotoxigenic strains of S. aureus produce enterotoxin. Inadequate cooling of cream fillings or inadequate cold storage of cream-filled pastries are the factors that contribute to most reported food borne outbreaks (fable 3). Another important factor associated with inadequate cooling is to prepare cream-filled pastries a day or more before serving during which time pastries or their fillings are improperly stored. Competition Many microorganisms compete with and outgrow staphylococci in the ingredients that go into cream fillings; but after these ingredients are blended and cooked, competing organisms as well as staphylococci are killed. Staphylococci can be readily reintroduced if the cooked fillings are handled by infected workers after cooking, and these organisms will subsequently grow without competition from other organisms. The ratio of competitive organisms to pathogen in the mix and the phase of growth that contaminating organisms are in when they reach the product will, along with the other factors mentioned, determine which organisms will predominate. Survival Cooking and reheating failures have permitted pathogens to survive in cream fillings and toppings (fable 3). Browning (heat treatment of approximately 450 F for about 6 min) of meringue topping, for instance, offers too little heat treatment to kill bacteria except those in the top layer. Meringue is an excellent insulator because of air bubbles entrapped in egg white foam. Experiments conducted by the American Institute of Baking (3) revealed that the temperature attained at the meringue-filling interface was less than 112 F during browning. EXAMPLES OF OUTBREAKS Circumstances leading up to two outbreaks in which either cream-filled pastry or meringue-topped cream pies were vehicles are presented below. Although both outbreaks occurred several years ago, similar patterns have repeated with monotonous regularity; and, unfortunately will be repeated as long as cream-filled pastries are mishandled by uninformed or careless bakers,

4 292 BRYAN food-service operators, and consumers. Cream-filled eclairs which were implicated in an outbreak of staphylococcal intoxication (16) were prepared in a bakery in the following manner. Cream filling was made of milk, sugar, starch, eggs, salt, and vanilla. (A blend which would readily support rapid and progressive grmvth of bacteria.) A mixture of milk and sugar was boiled. Then other ingredients were added. This mixture was poured into several 2-gal. pots which were placed on the floor of an unheated room to cool for 4 h. Of contamination had occurred during the transfer, staphylococci would have probably passed from their lag phase and begun exponential growth.) Then the cream filling was poured into a mechanical cream-puff filler. Prepared pastry shells were filled one at a time by pushing the shell upon the nozzle of the filler by hand. This process took about an hour. (During the investigation, it was observed that parts of the filler which were not easily accessible for cleaning had not been cleaned.) After filling, a chocolate covering was sprayed over the shells. Eclairs were left at room temperature for S h until they were delivered. Delivery took up to S more h. (If contamination had taken place during filling, there would have been ample time for staphylococcal growth and toxin production during holding and delivery.) Nose and throat cultures of most of the employees showed S. aureus. This organism was also isolated from a purulent infection on a finger of a worker who poured cream filling into the filler and handled each eclair during filling. It was also isolated from the nozzle and base of the filling machine as well as from samples of the cream filling. Emetic responses were observed in kittens tested. (This epidemic occurred before immunogel-diffusion tests for enterotoxin in food and phage typing of isolates from cultures of specimens and samples were available.) Chocolate and banana-cream pies were incriminated as vehicles in Salmonella outbreaks arising from meals served at two separate food service establishments (2). The incriminated cream pies were manufactured at a local pie bakery. The chocolate- or banana-flavored pudding made of dried skimmed milk, sugar. water, salt, cornstarch, and flavoring was boiled; then unpasteurized dried egg white (a likely source of Salmonella) was folded into the mixture. Pie shells were filled and a meringue made from unpasteurized frozen egg white (a likely and suspected source of Salmonella), sugar, agar stabilizer, and water was added by hand. Meringue-topped pies were browned in an oven for S min at 450 F. (Hardly enough time to kill Salmonella.) Pies were not refrigerated while stored at the bakery or during delivery. This practice provided ample time and suitable temperature conditions to allow salmonellae to proliferate. About a year later, lemon meringue pies baked in the same bakery were vehicles in a second outbreak. This time frozen egg whites were considered the source of Salmonella for the pies. CONTROL Prevention of foodborne illness is based on preventing or limiting contamination, killing pathogens, and preventing multiplication of bacteria. These principles must be applied at appropriate steps of cream-filled pastry manufacture, distribution, and storage. Product formulation and ingredients Synthetic cream fillings have been formulated to alleviate the inherent problem of supporting growth of pathogens and to prolong product shelf life. Without protein, staphylococci are unable to grow; without eggs the likelihood of contamination by salmonellae is lessened. Staphylococci can grow in some brands of synthetic cream fillings when just water has been added. Growth was enhanced, however, by the addition of milk, eggs, or butter. Control of water activity is the most important factor to consider when formulating creams so that they will prevent growth of microorganisms. But even in creams that are formulated to have water activity values too low for growth to occur, water may migrate and form pockets of moisture which can allow bacterial growth. The pockets form particularly at the interface of cream and pastry. Acid ingredients such as lemon may slow bacterial growth, but buffering properties of milk and eggs require that an excessive amount of acids are needed to lower the ph of a tilling to a value that would prevent growth of pathogens (13, 38). Lemon, orange, pineapple, apricot, raspberry, peach, and strawberry pure fruit fillings inhibited growth of salmonellae and staphylococci (13, 38). Chocolate or cocoa also inhibited staphylococcal gro\'iih in cream fillings (12). This inhibition was attributed to nonfat components of chocolate and to ph. Certain additives have inhibited bacterial growth in cream-filled pastries. Castellani (10) showed that DL -serine had bacteriostatic action on staphylococci (but not salmonellae) in cream-pastry fillings. Both coconut and cocoa, however, neutralized the bacteriostatic effect of serine on staphylococci (II). Subtilin and terramycin prevented growth of bacteria in cream fillings (21), but their use is prohibited. Hydrogen peroxide has been effectively used to inhibit bacterial growth in synthetic creams (22). Preonas et al. (36) reported that 80 mg of undissociated sorbic acid per gram in combination with 100 mg of undissociated propionic acid per gram prevented growth of S. aureus on top surfaces of custard pies. Eggs are a potentially hazardous ingredient in cream-filled pastry. Broken or checked eggs and unpasteurized frozen or dried eggs have been the source of salmonellae in these products (Table 3). Only eggs that have been pasteurized or otherwise treated so as to destroy all salmonellae should be used in fillings. As new developments in processed foods occur, changes take place in formulation of fillings and other

5 PUBLIC HEALTH ASPECTS OF PASTRIES 293 components of cream-filled pastry. These new ingredients sometimes permit incorporation of more water. High ratio shortenings, for example, permit incorporation of more water in a cake batter than was previously possible. If this increase in water is not accompanied by a similar increase in sugar, the resulting cake may not be safeguarded from growth of pathogens. Changes in formula should only be made after effects of changes on safety of the product have been evaluated (.5). Cooking and reheating Foodborne pathogens that have caused cream-filled pastry associated outbreaks are killed by routine methods of cooking cream fillings. Dack et al. (19) found that when sponge cakes were heated to temperatures above 170 F, staphylococci inoculated in the cake were destroyed. Cathcart et al. (14) observed that S. aureus and Salmonella were killed by the proper baking of custards and that bringing custard to a boil after adding thickening ingredients killed these organisms. Husseman and Tanner (24) reported that 10 8 staphylococci/g were killed when heated in cream filling to a temperature of 167 F. Temperatures of 185 to F rendered puddings and custards containing eggs free of S. aureus, but this organism survived at the minimum temperature for eggs to coagulate (25). In observations of commercial bakery operations a temperature of at least 180 F was attained in the center of custard when 9-inch pies were baked at 425 F for 25 to 30 min (3). Angelotti et al. (7) reported that 53 to 59 min were required to kill 10 7 staphylococci/gin custard heated at 140 F and that 5.2 to 6.6 min were needed to kill the same number when this product was heated at 150 F. They also reported that from 19 to 78 min at 140 F were required to kill the same number of two serotypes of Salmonella in custard and that 3.5 to 11.5 min were needed to kill them when this product was heated at 150 F. An internal temperature of 165 F in a cream mix should be adequate to kill any expected levels of contamination of S. aureus or Salmonella in less than a minute. The American Institute of Baking (4) recommends that pie shells be filled wtth cream fillings after the filling has been cooked and while it is still hot. The hot process of making meringue causes a marked reduction of bacteria (3). This method consists of adding hot sugar syrup to whipped egg whites and whipping until the desired consistency is reached. Browning has not been effective in killing pathogens. At one time, reheating cream"filled pastry after filling was recommended (20, 42). It was suggested that cream puffs or eclairs be put into a pan, covered with an inverted pan, and heated in an oven at a temperature of 420 F for 20 min. Although this procedure was effective in killing salmonellae and staphylococci (and preventing staphylococcal intoxications if it was used before enterotoxin was formed), it was objected to by industry because of the fear of overcooking, and it became antiquated as refrigeration became more readily available and more frequently used. Handling Because man is a reservoir of S. aureus and principal sources are his nose, hands, clothing, and lesions, attempts to exlude S. aureus from environments where cream-filled pastry is processed are impossible and attempts to prevent contamination of foods are extremely difficult. Because animals are frequent reservoirs of salmonellae, as well as staphylococci, they and unpasteurized foods produced by them or manufactured from their products can harbor these pathogens and transport them to food processing and preparation establishments. Dack et al. (19) showed that cooking killed staphylococci in cream filling and baking killed these organisms in sponge cakes. but contamination occurred after cooking. Therefore, workers who have open lesions. recurrent boils, acute sinus infections, diarrhea, or jaundice should not handle cooked fillings or prepared pastries. Regulations that prohibit ill or infected persons from handling food are difficult to enforce, and they do not entirely solve the problem because healthy people are just as apt to carry S. aureus and contaminate cream-filled pastry. Thorough washing can effectively remove transient organisms-those that are acquired from touching the nose or other body parts and from handling food-from hands, but washing is ineffective at removing resident staphylococci (29, 30). The use of disinfectants after handwashing unfortunately does not yield dramatic results in the control of S. aureus (9. 28). Gloves act as a barrier between contaminated hands or infected lesions and food, but they can become as contaminated with transient organisms as hands are if they are worn very long or while handling contaminated foods. Bacteria multiply rapidly on hands that are covered by gloves and hands so covered perspire freely so that punctures and tears in gloves, which readily occur, result in leakage of large numbers of bacteria, including resident staphylococci. Rubber gloves are seldom washed as frequently as hands or as well as other pieces of equipment at the end of the day's operation. Use of disposable plastic gloves or holding pastry between waxed paper, however, may be valuable in limiting contamination of cream-tilled pastry by contaminated hands during packaging. Their use should be limited to their intended purpose, and they should be changed if unrelated tasks are done or if they become soiled or torn. So even with the best intentions of personal hygiene and hygienic handling of fillings and pastry, contamination still occurs. Although adequate cooking will kill vegetative bacterial pathogens that may be in the filling or on pastry shells, there is a possibility of incorporating Salmonella into products such as meringue topping which are not adequately cooked. Such contamination can occur via air currents; by use of equipment used consecutively for unpasteurized products and then for cooked products; and by handling of unpasteurized products and then touching cooked cream filling, meringues, or finished

6 294 BRYAN pastries. However, if unpasteurized or particularly broken eggs are used in a bakery, they should be used in an area separate from cooked cream filling or meringue and should be handled only by workers who do not have access to areas where cooked cream filling or finished pastries are handled or stored. Their use should be limited to items that get thoroughly cooked. cans should not be used to store any food product other than the eggs originally packed in them. They are not designed for reuse. Organisms may lodge in the inner seam and be difficult to remove by washing. Frozen eggs should be thawed either in a refrigerator for several days or in a tank with cold running water for approximately 18 h. During mixing of batters or whipping of eggs, very fine droplets are produced. These aerosols may be carried by air currents to other areas and contaminate other products (4). Also, dust is generated from powdered ingredients. This dust may settle across a room and, if contaminated, spread pathogens. Ventilation should, therefore, be designed so that flow of filtered air is from finished product area to production area. Eclair filling machines should be constructed of easily cleanable materials and be so designed as to permit easy access for cleaning. These machines as well as storage pans, spoons, scoops, spatulas, ladles, and other utensils that touch the tilling should be washed with hot water and detergent. rinsed, and disinfected after use. Equipment sanitation is more effective if a cleaning schedule (which defines cleaning responsibility, frequency, time, and method) is followed. In a survey of 12 plants and 453 samples of imitation cream pie and 350 line samples, frozen pies produced in plants operating under good sanitary conditions (as determined by Food and Drug Administration inspections) had fewer than 3 Escherichia coli per gram, fewer than 50 coliforms per gram, fewer than 25,000 aerobic.plate colonies per gram, and no coagulase-positive staphylococci in 0.1-gram portions (43). Good sanitary conditions included freedom from gross sanitary deficiencies, periodic and thorough cleaning and disinfection of equipment. and use of hand sanitizing dips. Plants operating under poor sanitary conditions produced pies that had as many as 15 E. coli per gram, 1,100 coliforms per gram, 2,700,000 aerobic plate colonies per gram, and coagulase-positive staphylococci were found in 0.01-gram portions. In conclusion to a study of the bacteriology of cream-filled pastry, Abrahamson et al. (J) stated that the keystone of control rests in proper sanitary technique and that refrigeration is a secondary consideration. (Perhaps the order should have been reversed, but the two most important control measures were emphasized.) Other control measures stated were adjustment of ph and reheating of finished bakery products, But they felt that these methods were not as well accepted by industry because of consumer rejection. Storage Contemporary practices of personal hygiene and sanitation of bakeries do not ensure that foodborne pathogens will be kept out of cream-filled pastries. Therefore, foods must be treated as if they are contaminated by at least staphylococci if not by other foodborne pathogens. In an early report on the control of foodborne illness from cream-tilled pastry. the need for refrigeration in bakeries and homes was emphasized (48). The most crucial procedure to prevent bacterial foodborne diseases is to keep pathogenic bacteria from multiplying. This can only be done by: (a) altering cream fillings to make them unable to support growth of pathogenic bacteria, (b) allowing filling or cooked products to remain within the temperature range at which pathogenic bacteria can grow for such time that these organisms stay in the lag phase, (c) rapidly lowering temperatures of foods and storing them at temperatures at which pathogenic bacteria cannot grow, or (d) holding hot foods above temperatures that allow pathogenic bacteria to grow. The first method has already been discussed, and the last is a preventive measure that is probably not practicable for storage of cream-filled pastry. Discussion will, therefore, be on the second and third listed methods. Bulk cream filling or cream-tilled pastry should not be held within the temperature range 42 to 126 F that permits growth of foodborne disease bacteria for more than 3 cumulative hours, including chilling time. This objective is difficult to meet unless special procedures are taken to cool fillings rapidly because custard and similar fillings cool slowly. If cream fillings are stored for less than 3 days, they should be kept at temperatures of 45 F or below (46). If they are stored for a longer time, temperatures of 40 F or less should be used. Prompt cooling of cream filling is essential for preventing gro"wih of pathogenic bacteria. Batches of cream tilling should never be cooled to room temperature before being refrigerated. In this period. considerable bacterial gro"wih can occur. Cooked fillings which are to be used at some later time should be put in a refrigerator within a half hour after cooking. Black and Lewis (8) observed that custard cooled faster when immediately refrigerated (at 170 F) than when allowed to cool to room temperature before refrigerating. Very high bacterial colony counts were observed in custards held at room temperature before cooling, but only a small increase was noted in custard that was put in a refrigerator immediately after cooking. The time it takes filling to cool to temperatures at which food borne pathogens cease multiplication is increased as the time is extended at which fillings are held at room temperature before refrigeration (27). The size and shape of the container and its fill greatly influence cooling time. The farther heat must travel from the center of a food to the food surface or container wall where it is taken up by the cooling medium, the longer it takes the food to cool. The internal temperature of food

7 PUBLIC HEALTH ASPECTS OF PASTRIES 295 cooled in shallow pans falls faster than that of foods cooled in large, deep containers. For instance, McDivitt and Hammer (31) observed that cornstarch pudding cooled slower and S. aureus grew in it when it was stored in 100-serving quantities in 20-quart and 14-quart stock pots in a refrigerator (40 F), but it cooled relatively rapidly and no bacterial growth occurred when it was stored in a shallow pan at the same temperature. Cooling rate is also affected by the material of which containers are made and by the method of cooling. Miller and Smull (33) reported that 3 gal. of pie filling cooled faster in enamel pans than in earthenware crocks. This filling cooled faster in refrigerators than at room temperature; it cooled faster in ice or water baths than in either refrigerators or freezers. Agitation and stirring also enhanced the rate of cooling cornstarch pudding (31) and custard (26). Some companies and health agencies have advocated discontinuing preparation of cream-filled pastry during the warmer months. Such a practice, however, does not offer a high degree of safety because bakeries, food service establishments, and homes are relatively warm year round and if cream filling or cream-filled pastries are left at room temperature (winter or summer) bacterial contaminants will multiply. Proper refrigeration can prevent this from happening. The practice of discontinuing manufacture of cream-filled pastry in the summer seems to be an antiquated control measure left over from a period before refrigerated storage equipment, refrigerated trucks, or synthetic fillings were readily available. Microbiological standards Microbial standards of quality for frozen ready-to-eat banana. coconut, chocolate, or lemon cream-type pies have been proposed by the Food and Drug Administration (47). These standards have been based on a survey of samples from manufacturers of such products. The standards are that the geometric mean of 10 representative samples of a lot be equal to or less than 50,000 aerobic plate count per gram, and equal to or less than SO coliform count per gram. Lots falling below this standard must be labeled with the statement "Below standard in quality-contains excessive bacteria." Such standards should assist in providing incentive to keep contamination to a minimum level, to maintain a reasonable level of plant sanitation, and to have good operational procedures. Training Prevention of foodborne disease from cream-filled pastry must be based on training and education. Training of professional public health workers, managers of bakeries, and employees and supervisors of baking and food service operations is necessary. They should be informed about specific disease hazards associated with preparation of cream-filled pastry, of factors that have contributed to outbreaks, and of critical control points of the operation. They should develop skills in their specific preventive roles whether it be teaching, inspection of operations, purchasing safe ingredients, managing operation, cooking cream filling, filling eclairs, or storing prepared fillings or pastries. And they should be motivated to perform these skills so that a good, safe, profitable product is produced. For prevention of staphylococcal intoxication, for instance, food workers and their supervisors must not only know that staphylococcal contamination results primarily from infected (but usually healthy) workers touching foods, but that foods so contaminated must be handled in a way that will prevent staphylococci from multiplying and, thus from elaborating enterotoxin. They must know that unpasteurized egg products, cracked and checked eggs, and the egg shell itself may harbor salmonellae which end up in finished products as a result of inadequate cooking or from cross contamination. They must know that foodborne disease outbreaks can be prevented by application of the following three food protection principles: (a) preventing contamination by purchasing safe foods, by careful handling of foodstuffs, by personal hygiene, and by cleaning and disinfection of kitchen equipment; (b) killing pathogens by thorough cooking; and (c) preventing bacteria from multiplying by processing foods rapidly while bacteria are in a lag phase, by rapidly cooling foods, and by holding foods at temperatures at above or below those at which pathogens multiply. The aim of training is to provide habitually good day-to-day operational procedures in the production of cream-filled pastry. Education of the purchaser so that he will not abuse cream-filled pastry by letting the small numbers of bacteria that will invariably be in it multiply is a necessary part of a cream-filled pastry protection program. Tapes or labels on boxes informing the purchaser to refrigerate cream-filled pastry have assisted in obtaining this goal. SUMMARY Achievements by industry, research institutions, and official food protection agencies in developing an awareness of the problem and in devising and applying preventive measures are notable. Few foodborne diseases transmitted by specific vehicles (other than milk) have shown such a marked decline in the last four decades. Disease surveillance with appropriate follow up, microbiological testing, and training and education are the approaches that will prevent cream-filled pastry from becoming vehicles of food borne disease outbreaks in the future. ACKNOWLEDGMENTS Presented at the Annual Meeting of the New York State Association of Milk and Food Sanitarians, September 20, REFERENCES 1. Abrahamson, A. E.. R. Field, L. Buchbinder, and A. V. Catelli A study of the control ofthe sanitary quality of custard-tilled

8 296 BRYAN bakery products in a large city. Food Res. 17: Ager. E. A., K. E. Nelson, M. M. Galton, J. R. Boring III, and J. R. Jernigan Two outbreaks of egg-borne salmoneliosis and implications for their provention. J. Am. Med. Assoc. 199: American Institute of Baking Practices designed to prevent the growth of food poisoning bacteria in meringues. Special Bull. No. 84. Am. Inst. of Baking, Chicago, Ill. 4. American Institute of Baking Salmonellosis: Bacterial food poisoning caused by salmonellosis. Bull. No Am. Inst. Baking, Chicago, Ill. 5. American Institute of Baking Safety of 'cream filled' cakes: 'butter cream' type fillings. Bull. No Am. Inst. Baking, Chicago, Ill. 6. Angelotti. R., M. J. Fater, and K. H. Lewis, Time-temperature effects on salmonellae and staphylococci in foods. I. Behavior in refrigerated foods. II. Behavior at warm holding temperatures. Am. J. Public Health 51: Angelotti, R., M. J. Fater, and K. H. Lewis Time-temperature effects on salmonellae and staphylococci in foods. II. Thermal death time studies. Appl. Microbial. 9: Black, L. C., and M. N. Lewis Effect on bacterial growth of various methods of cooking cooked foods. J. Am. Dietet. Assoc. 24: Brodie, J Hand hygiene. Scotish Med. J. 10: Castellani, A. G Inhibiting effects of amino acids and related compounds upon the growth of enterotoxigenic micrococci in cream pastry. Appl. Microbial. 1: Castellani, A. G., R. Makowski, and W. B. Bradley The inhibiting effect of serine upon the growth of indigenous flora of cream tilling. Appl. Microbial. 3: Cathcart. W. H., and A. Merz Staphylococci and salmonella control in foods. Ill. Effect of chocolate and cocoa fillings on inhibiting growth of staphylococci. Food Res. 7: Cathcart, W. H., W. H. Godkin, and G. Barnett Growth of Staphylococcus au reus in various pastry fillings. Food Res. 12: Cathcart, W. H., A. Merz. and R. E. Ryberg Staphylococci and salmonella control in foods. IV. Effect of cooking bakery custards. Food Res. 7: Cook, G. T Food poisoning associated with infected dried egg. Mon. Bull. Min. Health Public Health Lab. Serv. (Gr. Brit.) 6: Coughlin, F. E., and B. Johnson Gastroenteritis from cream-filled pastry. Am. J. Public Health 31: Crisley, F. D., R. Angelotti, and M. J. Fater Multiplication of Staphylococcus au reus in synthetic cream tillings and pies. Public Health Rept. (U.S.) 79: Dack, G. M Food poisoning, 3rd ed. Univ. Chicago Press, Chicago, Ill. 19. Dack, G. M., 0. Woolpert, I. Noble, and E. G. Halliday Experimental study of possible routes of contamination of cakes with food-poisoning staphylococci. J. Prev. Med. 5: Gilcreas, F. W., and M. B. Coleman Studies of rebaking cream-filled pastry. Am. J. Public Health 331: Godkin, W. J., and W. H. Cathcart The complementary action of subtilin and terramycin in preserving custard fillings. Food Techno!. 7: Hobbs, B. C., and M. E. Smith The control of infection spread by synthetic cream. J. Hyg. 52: Hodge, B. E Control of staphyiococcal food poisomng. Public Health Rept. (U.S.) 75: Husseman, D. L., and F. W. Tanner Relation of certain cooking procedures to staphylococcus food poisoning. Am. J. Public Health 37: Kintner, T. C., and I. M. Mangel Survival of staphylococci and salmonellae in puddings and custards prepared with experimentally inoculated dried eggs. Food Res. 18: Longree, K Cooling fluid food under agitation. J. Am. Dietet. Assoc. 44: Longree, K., and J. C. White Cooling rates and bacterial growth in food prepared and stored in quantity. I. Broth and White Sauce. J. Am. Dietet. Assoc. 31: Lowbury, E. J. L Skin disinfection. J. Clin. Pathol. 14: Lowbury, E. J. L., H. A. Lilly, and J.P. Bull Disinfection of hands: Removal of resident bacteria. Brit. Med. J. 2: Lowbury, E. J. L., H. A. Lilly, and J.P. Bull Disinfection of hands: Removal of transient organisms. Brit. Med. J. 2: McDivitt, M. E., and M. L. Hammer Cooling rates and bacterial growth in cornstarch pudding. J. Am. Dietet. Assoc. 34: McKinley, T. W., and E. J. Clarke Imitation cream fillings as a vehicle of staphylococcal food poisoning. J. Milk Food Techno!. 27: Miller, W. A., and M. L. Smull Efficiency of cooling practices in preventing growth of micrococci. J. Am. Dietet. Assoc. 31: Munch-Petersen. E Staphylococcal carriage in man. Bull. World Health Organ. 24: Post, F. J.. A. H. Bli~s, and W. B. O'Keefe Studies on the ecology of selected food poisoning organisms in food. I. Growth of Staphylococcus aureus in organisms in food. II. Growth of Staphylococcus aureus in cream and a cream product. J. Food Sci. 26: Preonas, D. L., A. I. Nelson, Z. J. Ordal, M. P. Steinberg, and L. S. Wei Growth of Staphylococcus aureus M F 31 on the top and cut surfaces of southern custard pies. Appl. Microbial. 18: Rolfes, T. J An evaluation of butter-cream-type fillings for cakes. (Cited in American Institute of Baking, Bull. No. 119, 1966.) 38. Ryberg, R. E., and W. H. Cathcart Staphylococci and salmonella in foods. I. Effect of pure fruit fillings. Food Res. 7: Scott, W. J Water relations of Staphylococcus au reus at 30 C. Australia J. Bioi. Sci. 6: Silliker, J. H Some guidelines for the safe use of fillings, toppings, and icings. Baker's Digest 43(1): Silliker, J. H., and S. A. McHugh Factors influencing microbial stability of butter-cream-type fillings. Cereal Sci. Today 12(1):63-65, Stritar, J., G. M. Pack, and F. G. Jungewalter The control of staphylococci in custard-filled puffs and eclairs. Food Res. 1: Surkiewicz, B. F Bacteriological survey of the frozen prepared food industry. I. Frozen cream-type pies. Appl. Microbial. 14: Tanner. F. W., and L. P. Tanner Food-borne infections and intoxications, 2nd ed. Garrard Press, Champaign, Ill. 45. Troller, J. A The water relations of food-borne bacterial populations: A review. J. Milk Food Techno!. 36: U.S. Department Health, Education, and Welfare, Public Health Service Food service sanitation manual. Public Health Pub!. No U.S. Department Health, Education, and Welfare; Food and Drug Administration Certain foods for which there are no standards of identity. (21 CFR Part II). Federal Register 37(No. 186, Sept. 23, 1972): U.S. Public Health Service Preventing typhoid and bacterial food poisoning from cream-filled pastry. Public Health Rept. (U.S.) 52: