441

Journal of Food Protection, Vol. 53, No. 5, Pages 441-452 (May 1990) Copyright© International Association of Milk, Food and Environmental Sanitarians

Microbiological Safety of Cheese Made from Heat-Treated Milk, Part I. Executive Summary, Introduction and History

ERIC A. JOHNSON1, JOHN H. NELSON1*, and MARK JOHNSON2

Food Research Institute and the Walter V. Price Cheese Research Institute, University of Wisconsin, Madison, Wisconsin 53706 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 (Received for publication August 2, 1989)

ABSTRACT four decades, the United States cheese industry produced over 100 billion pounds of natural cheese (not including Research on pasteurization of milk for cheesemaking was cottage and related varieties). The most frequent causative begun in the late 1800's. Early equipment was crude and control factor in U.S. and Canadian cheese-related outbreaks was devices non-existent. Consequently, early pasteurization proc­ post-pasteurization contamination. Faulty pasteurization esses were not well verified. Commercial application was slow, equipment or procedures were implicated in one outbreak except in New Zealand, where almost the entire cheese industry converted to pasteurization in the 1920's. In the United States, each in the U.S. and Canada. Use of raw milk was a factor debate on the merits of pasteurization continued for years. Demand in one outbreak in each country. Inadequate time-tempera­ for cheese during World War II and foodborne disease outbreaks ture combinations used for milk heat treatment were not caused by cheese stimulated promulgation of government stan­ implicated. dards which included the options of milk pasteurization or 60 d The epidemiology of cheese-related outbreaks in the holding at a minimum temperature of 2°C (35°F). The cheese U.S., Canada, and Europe demonstrated that soft surface- industry has continued to improve technology, including that ripened cheeses, e.g. Camembert and Brie, are at signifi­ which is safety related. United States production of cheese has cantly greater risk to transmit pathogens than other cheeses. continued to expand, from just over 1 billion pounds in 1948 to No outbreaks were linked to hard Italian varieties, e.g. 5.4 billion pounds in 1987. Thirty-eight percent of the 1987 total Parmesan, Romano, and Provolone. Varieties such as comprised varieties wherein heat-treated milk is frequently util­ Cheddar and Swiss were infrequently involved. ized. Pathogens were prioritized as high, medium, or low risk in cheese. Three organisms, Salmonella, Listeria mon­ EXECUTIVE SUMMARY ocytogenes, and enteropathogenic Escherichia coli, were judged to be high risk threats to the cheese industry. The heat-treatment of raw milk can exert a significant Staphylococcus aureus was listed as low risk because growth role in producing microbiologically safe cheese. Recent, and toxin production is readily suppressed by modern lactic thorough research has affirmed that milk heat-treatment at culture technology and acidity (pH) control in cheese. 65.0-65.6°C (149-150°F) for 16-18 s will destroy virtually Recently published research (98,99,126) comprehen­ all pathogenic microorganisms which are major threats to sively defined the effect of raw milk heat-treatment on the safety of cheese. pathogen survival. Multi-strain or species mixtures of An extensive review of epidemiological literature pathogens were inoculated into raw milk at levels of 105/ identified only six illness outbreaks transmitted via U.S. ml. Inoculated milk was heat-treated in a commercial HTST produced cheese during 40 years, 1948-1988. During these pasteurizer — mean holding time 17.6 s, minimum 16.2 s. All strains of Yersinia enterocolitica, Campylobacter sp., 'Food Research Institute, University of Wisconsin, Madison. 2 E. coli 0157:H7, and all but one Salmonella species were Walter V. Price Cheese Research Institute, University of Wisconsin, destroyed at 65°C (149°F). Salmonella senftenberg was Madison. inactivated at 69°C (156.2°F). S. senftenberg is rarely isolated Note: This is Part I of a three-part manuscript in which the safety of cheese from cheese. L. monocytogenes in naturally contaminated made from heat-treated milk is examined. Part I includes the executive milk at levels of 104 organisms per ml was inactivated at summary, the introduction, and the history of the technology including statistics on pasteurization and heat treatment utilization. Part II addresses 66°C (150.8°F); laboratory-cultured inoculum at levels of 5 microbiology aspects, including epidemiology, the occurrence of foo­ 10 organisms per ml required 69.0°C (156.2°F). dborne pathogens in milk and cheese, survival and behavior of pathogens A large cheese factory which heat-treats milk for during milk heat treatment and cheese manufacture and technologies cheesemaking at 64.4°C (148-149°F) for 16 s with concur­ which may aid pathogen control in cheese. Part III includes process and product technology, the discussion, recommendations and the bibliogra­ rent concentration to 16-17% total solids provided data on phy. aerobic plate count (APC) and coliform count for several

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 442 JOHNSON. NELSON, AND JOHNSON hundred production days. The average APC of 1.4 million/ for a minimum of 60 d, or cheeses for further manufactur­ ml in raw milk was reduced 1.71 logs to an average of ing which require neither milk pasteurization nor a 60 d 28,000/ml in milk entering the vat. Coliform counts, which hold. The holding option is required when cheese is made averaged 121,000/ml in raw milk, were almost all <10/ml from raw or milk heat-treated at time-temperature combi­ in heat treated milk, with a maximum of 50/ml. All tem­ nations less than pasteurization. peratures studied were above 60°C (140°F), the minimum As the list of food pathogens has lengthened and hot food holding temperature specified in FDA's good detection methods for them have improved, there have manufacturing practice regulations. been instances when pathogens have survived for more A multiplicity of practices other than pasteurization or than 60 d in cheese made from raw milk or milk heat- heat-treatment contribute significantly to the microbiologi­ treated at less than pasteurization. Thus, the effectiveness cal safety of cheese. Some, such as milk quality manage­ of the 60 d holding period for assuring safety is in doubt. ment, lactic culture management, pH control, salt addi­ Milk heat-treatment has a long history of use. It provides tion, and controlled curing conditions are established two of the advantages of pasteurization, more consistent

technologies. Others represent potential opportunities, such control of the cheesemaking process, and more uniform Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 as natural inhibitory substances in milk, and antibacterial cheese quality. Because heat-treatment only partially inac­ substances, e.g. nisin and lysozyme. It is imperative that tivates enzymes, microorganisms, and other biologically the relationships of established and potential safety tech­ active constituents of raw milk, cheese flavor develop­ nologies be better defined to enable the articulation of ment during curing is more rapid than in cheese made proven safety systems geared to the characteristics and from pasteurized milk. Moreover, the flavor quality of safety risks of various cheese varieties. Neither pasteuri­ some varieties of cheese made from heat-treated milk is zation nor any other single technology can assure safe considered by experienced judges to be superior to pas­ cheese. teurized milk counterparts. Such varieties include Ched­ The National Cheese Institute should encourage and dar, Swiss, and hard Italian-type varieties. support research on cheese safety. Three NCI actions are Depending on the numbers and types of pathogens recommended: present and the lethality of the heat-treatment used, all (1) Establish a guideline that the minimum heat-treat­ pathogens may not be killed, particularly if numbers are ment of milk for cheesemaking be 64.4°C (148°F) high or heat-resistant species are present. However, it is for 16 s or equivalent with adequate process con­ conceivable that in the manufacture of some cheese varie­ trol. ties, e.g. Swiss and Parmesan, the cumulative effect of (2) Evaluate current safety technology and practice for milk heat-treatment, curd cooking temperature, and curing cheese manufacture. Define, prioritize, and sup­ time-temperature is sufficient to assure safety. port research with primary emphasis on the com­ The National Cheese Institute, concerned about the bined effect of heat-treatment and other current safety risks posed by the ill-defined effectiveness of heat- cheese technologies. treatment and the 60 d hold, and concerned also about the (3) Evaluate technologies not currently utilized in cheese possible demise of widely-used heat-treatment options if manufacture for safety potential. the 60 d hold provision were repealed, has commissioned this paper. I. Introduction The objectives of the assessment were to: During recent decades, the United States cheese in­ A. Review critically and comprehensively infor­ dustry has maintained a good, though not perfect record of mation on milk heat-treatment for cheesemak­ producing a wide variety of safe products. Billions of ing with major emphasis on product safety. packages of safe, wholesome cheeses and related products B. Assess the effectiveness of heat-treatment and are produced annually. However, this record is blemished other technologies in the control of pathogenic by a few serious outbreaks of foodborne illness caused by microorganisms in cheesemaking and curing. pathogen-containing cheese. Moreover, the list of known C. Recommend appropriate action to the National pathogens which can be transmitted via cheese has length­ Cheese Institute Board of Directors. ened, most notably by the addition of Listeria. The assessment begins with the history of heat-treat­ The cheese industry's proven arsenal of safety-pro­ ment technology, from first trials at the close of the nine­ moting technologies includes milk quality management, teenth century to the present. Statistics indicative of pas­ pasteurization and heat-treatment of milk, lactic culture teurization and heat-treatment used are assessed as is the technology, other good manufacturing practices, and thor­ epidemiology of cheese-borne illness since 1948. The ough sanitation programs. No single technology can assure occurrence of bacterial and viral pathogens in milk is safe cheese. A system comprising several is required. reviewed. The effect of milk heat-treatments of pathogens Pasteurization is particularly effective, for it can control a is assessed as is the behavior of pathogens during cheese­ significant pathway for pathogen transmission — raw milk. making and curing. Technologies which contribute to cheese Most U.S. Standards of Identity for natural cheeses safety are briefly evaluated. The effect of heat-treatment specify three safety options: milk pasteurization, holding on cheese characteristics and quality is addressed. Recom­ finished cheese at a minimum temperature of 2°C (35°F) mendations are made for strengthening cheese safety

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 MICROBIOLOGICAL SAFETY OF CHEESE 443 programs and for future research to resolve safety issues. B. Research and development. Research on milk "pasteurization" for cheesemaking II. Milk heat treatment for cheesemaking: a history in the United States was begun in 1907 at the Wisconsin A. Definitions. Agricultural Experiment Station by Sammis and Bruhn Modern pasteurization equipment and processes de­ (309), and Sammis (301,302). The project, a joint effort signed to inactivate known milk-borne pathogens are le­ by the University of Wisconsin and the United States gally defined and regulated in exquisite detail. Pasteuriza­ Department of Agriculture, continued for five years. Cheddar tion in this paper denotes temperature-time conditions cheese was the only variety researched. equivalent to or more rigorous than those specified in the Evaluation of "pasteurization" was prompted by the Code of Federal Regulations (37) — 21CFR33.3(d), e.g. results of trials comparing the effect of milk acidulation 62.8°C (145°F) for 30 min (vat or holder method) versus starter culture use on cheese quality. The quality of 71.7°C (161°F) for 15 s (continuous or flash method) cheese made from "overripe or tainted milk" was no better "or other time/temperature relationship which has been via either method. The investigators observed that "most

demonstrated to be equivalent thereto in microbial de­ of the defects in cheese factory milk are of bacterial origin." Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 struction." Although the primary emphasis of the project was "Pasteurization" is frequently used in early technical improvement of cheese quality, product safety was a and trade literature to describe milk heat-treatment for significant concern. The thermal sensitivity of several cheesemaking. Very few publications describe research on pathogens was cited — "tubercle bacilli," "typhoid bacil­ heat-treated milk. The "pasteurization" processes reported lus," "diphtheria bacillus," "cholera vibrio," and "dysen­ by early workers were ill-defined. Frequently, holding times tery bacillus." The investigators concluded "It is clear that were not specified. Equipment was primitive. Two types 'pasteurization' at 71°C (160°F) for one min and for most of processing were used — vat or "holder," where a batch cases for a shorter period, is effective in destroying patho­ or lot of milk was heated and held at a specified tempera­ genic bacteria in milk." ture for a specified time, and "continuous" or "flash," the Moreover, in 1908, the Chicago City Council enacted forerunner of modern HTST equipment. an ordinance: Temperature monitoring devices and controls were usually lacking. Fail-safe devices, such as flow diversion Department of Health, City of Chicago valves, were not available until the 1930's. Consequently, Ordinances, passed by City Council, some early "pasteurization" treatments almost certainly do July 13, 1908 not meet current legal criteria. To minimize confusion, "pasteurization" and derived words will be in quotes in Cheese. Be it ordained by the City Council of the City of the review of historical developments which occurred before Chicago: the end of World War 11. The term heat-treatment as currently used, denotes Section 2. It shall be unlawful to seal any such cheese the process of heating milk at time-temperature conditions in the City of Chicago unless there be stamped on the less rigorous than pasteurization. The process is also termed package in plainly legible letters of not less than one- "sub-pasteurization." The usual practice is to continuously eighth inch type: "Made of milk (or cream) from cows process milk through a high temperature — short time heat free from tuberculosis as shown by tuberculin test," or exchanger, plate or tubular, with the milk emerging from "Made from milk (or cream) pasteurized according to the the unit at setting temperature, ready for addition of starter rules and regulations of the Department of Health of the and rennet. There are neither established criteria nor stan­ City of Chicago." dards for heat-treatment. The term is thus generic. The term thermization is sometimes used to describe Section 4. This ordinance shall be in full force and milk heat-treatment. Thermization does not have an estab­ effect from and after January 1, 1909. lished definition, but an International Dairy Federation That ordinance must certainly have stimulated the inves­ document (33) describes thermization as "a system of heat- tigators! treatment . . . carried out under continuous flow condi­ tions . . . (consisting) . . . of heating to 63-65°C (145-150°F) Two technical hurdles were identified: (1) heat-treated for 15-20 s and then cooling with the result that psychro- milk coagulates more slowly than raw milk, and (2) heat- trophic bacteria are almost completely destroyed, while. . treated milk curd expels whey more slowly than raw milk . milk enzymes are unaffected." Thermization appears to curd. The former was overcome by the addition of cal­ be in limited use to reduce or prevent psychrotroph pro­ cium chloride. The latter was compounded by slower acid liferation in milk stored prior to processing. production in "pasteurized" milk than in raw milk. Post- Blanching, occasionally used to denote milk heat- "pasteurization" milk acidulation with hydrochloric acid treatment, is not an appropriate term, because it is tradi­ overcame this difficulty. "Pasteurization" and acidulation tionally used to described thermal inactivation of enzymes were considered by Sammis and Bruhn to be complemen­ and microorganisms during vegetable processing. tary. Initial cheesemaking trials with milk "pasteurized" at

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 444 JOHNSON, NELSON, AND JOHNSON

60°C (MOT) for 20 min, then acidulated, yielded good accumulated in the cheese vat until all milk had been quality cheese. "Pasteurization" was conducted either in heated, and then was cooled to setting temperature. Raw the cheese vat or in a tank-like "pasteurizer." milk bacterial counts ranged from 1.1 x 106 to 1.9 X 108/ Subsequent trials compared the 60°C-20 min treatment ml. Average percent reductions in count were 98.31, 98.65, to continuous "pasteurization" at 65.6°C (150°F), 71.1°C and 98.96 for the flash, flash-holder, and holder methods (160°F), and 76.7°C (170°F), no holding times given. Cheese respectively. Quality scores and cheese yields increased as grading revealed that 71.TC (160°F) produced slightly but average count reductions increased. Milk "pasteurized" by consistently organoleptically superior cheese compared to any of these three early methods yielded higher scoring the other treatments. cheese than did identical milk not "pasteurized." Bacterial content of milk before and after continuous Phillips (252) investigated the effect of "pasteuriza­ "pasteurization" at 71°C (160°F) was measured. Raw milk tion" of "inferior milk" on cheese quality. Flash "pasteuri­ counts of 30 daily samples, collected in August and zation" was controlled between 71.1-73.9°C (160-165°F) September, 1909 and 1910, ranged from 1 x 106 to 185 x by "an automatic temperature control apparatus" (not 6

10 organisms per ml. After "pasteurization," counts ranged described). Phillips estimated holding time to be 25 s.Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 from 15 X 103 to 1.3 X 106 organisms per ml, with count Without exception, "pasteurization" improved quality. reductions of 98.5-99.9%. Average flavor score improvement was 4-5 points; texture Temperatures of 76.7°C (170°F) and 82.2°C (180°F), improvement 3/4 point, using the USDA score card. Phil­ using the continuous, "flash" method resulted in cheese lips also observed a lower fat loss when cheese was made with little flavor and weak, crumbly body. Cheese yields from "pasteurized" rather than raw milk. were higher when milk was "pasteurized." Marquardt (208), in 1929, reported on 1923-26 studies Two makes of continuous, "flash pasteurizers" were of the relationship of milk cooling and "pasteurization" on compared at operating temperatures of 71.1-73.9°C (160- cheese quality, from pilot scale manufacture yielding three- 165°F), holding time not specified. Both units were manually pound cheeses. In 186 comparisons of cheese made from controlled as evidenced by an excerpt from start up in­ raw or "pasteurized" milk, "pasteurization" improved average structions: flavor score two points. Marquardt "pasteurized" cheese ... "A file mark on the steam valve handle is milk at 62.8°C (145°F) for 30 min. He also continuously great help in quickly adjusting the steam supply, to flash heated milk to 62.8-65.6°C (145-150°F), collecting the right point. Once adjusted and with a steady milk and holding the heated milk until all of the milk had been and steam supply, the pasteurizing temperature remains heated (time interval not specified). nearly constant, and requires only momentary inspec­ Marquardt briefly discussed milk "pasteurization" for tion, every few minutes ..." cheese varieties other than Cheddar. He mentioned the As the technology of cheesemaking with "pasteurized" benefits of skim milk "pasteurization" for Cottage cheese milk continued to be perfected, cheese was sold to con­ manufacture — improved flavor, texture, and shelf life. sumers to determine acceptance before recommending the Experimental studies on milk "pasteurization" for Brick, new process to cheesemakers. During 1909-1911, 10,126 Swiss, and Camembert were reported to have presented pounds of cheese were sold to 27 dealers in Boston, New "new problems," not further described. York, Philadelphia, Chicago, St. Louis, Minneapolis, and However, Hochstrasser and Price (164), in 1927, re­ San Francisco, as well as several cities in Wisconsin. ported that Camembert cheese could be successfully made Acceptance was good as evidenced by repeat sales and from milk "pasteurized" by the holding method of 62.8°C positive response to a letter survey soliciting buyer opin­ (145°F) for 30 min. Average quality and uniformity im­ ion. proved, although more starter culture was required than Price (257,258), in 1927, evaluated several tempera­ when raw milk was used. ture-time combinations using the holder method, and "Pasteurization" in the cheese vat, suitable for one- concluded that 62.8°C (145°F) for 30 min produced better man factories was described in the 1930 edition of Sam- Cheddar cheese than any other treatment tested. Price also mis' "Cheese Making" (305). Raw milk in the vat was found that milk "pasteurization" produced "cheese which heated to 62.8°C (145°F), held for 30 min and cooled to is not only better, but also more uniform in quality than setting temperature. The additional time required to "pas­ raw milk cheese." "Pasteurization" increased cheese yield teurize" was estimated to be 1.5 to 2 h. and improved keeping quality during storage. The make Yet another option described was "pasteurization" of procedure developed by Price used lactic starter rather curd during manufacture, devised by Sammis and Ger­ than acidulation. During the summer of 1924, 1,121,629 main (310). Immediately after milling, the milled curd pounds of cheese were made under commercial condi­ was spread evenly over the vat bottom, and twice the curd tions, using the new process, with consistently good re­ weight of water at 76.7°C (170°F) was added, without stirring. sults. After 30 min, the curd temperature was reduced to 32.2°C Price and Prickett (264), in 1928, published results of (90°C) by replacing part of the hot water with cold water. the comparison of three "pasteurization" methods — one The curd was then drained, salted and pressed as usual. holder method, a flash method, and a combination flash- Price and Call (266), during 1942-44, evaluated a wide holder method. In the latter method, flash-heated milk was range of heat-treatments for their effect on cheese quality

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 MICROBIOLOGICAL SAFETY OF CHEESE 445 and ripening. Pilot-scale portions of identical milk were Sammis authored a manual entitled "Cheese Making," which made into Cheddar cheese (a) without heating (raw), (b) he updated at 6-7 year intervals. The 1918 edition (303) after "pasteurization" by the holder method: 62°C (143°F) was the first to include information on cheese from for 30 min or 63°C (145°F) for 20 min, (c) after heating "pasteurized" milk, in a chapter entitled "Minor Sorts of to 71°C (160°F) for 15 s, (d) at a temperature over 71°C; Cheese." The risk of disease transmission via milk was between 74-87°C (165-189°F) for 15 s or (e) a temperature described, including an experiment in which tuberculosis under 71°C; between 50-59°C (122-156°F) for 15 s. Cheese organisms remained viable in cheese for seven months, was evaluated at intervals up to 1 year of age. infecting guinea pigs fed the contaminated cheese. Sam­ All cheeses improved in quality during the first 3 to mis commented, however: 6 months. From 6 months to 1 year, raw milk cheeses and "On account of the extra expense for buying pas­ cheese made from milk heated to less than 71°C decreased teurizing machinery and extra labor for operating it, in quality. All heated milk cheeses exhibited better grades it does not seem likely that any pasteurization process at every age than did raw milk cheese. Cheese made from is likely to be adopted by small, one-man cheese fac­

milk heated to 71 °C showed the greatest improvement over tories, unless required by law." Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 raw milk cheese. Cheese made from milk heated at less Sammis reported additional experiment station trials dur­ than 71°C showed the least improvement. Raw milk cheeses ing 1912 and 1913, in which 14,394 lbs. of "pasteurized" developed characteristic flavor most rapidly. Cheeses made milk cheese were manufactured, and commercial trials at from milk heated under 71°C exhibited curing changes three factories during 1914 and 1915, in which 1,113,000 similar to raw milk cheeses. lbs. of "pasteurized" milk cheese were manufactured, all with "good success." C. Early research in other countries. The 1924 edition of "Cheese Making" by Sammis The research by Sammis and Bruhn (309) (previously (304) included no new information on the Wisconsin process, described) was extensive and commercially proven, but which was included in the chapter "Other Sorts of Ameri­ not the only early work. Benson (60) and Benson and can Cheese." The "pasteurized milk cheese" made by the Evans (61) in Great Britain, Dean (102) in Canada, Fas- method is described as "mild flavored," keeping well, but cetti (127) in Italy, von Freudenreich (140) in Switzerland, "seldom, if ever acquiring the high, strong flavor of old Lunde and Holm (201) in Denmark, Haglund (153) in Cheddar cheese." Research to overcome the deficiency Sweden, and Liska (199) in Germany reported the results was reported to be unsuccessful. of their investigations between 1893 and 1915. Results Reid (275), in 1929, surveyed cheese factories in paralleled those of Sammis and Bruhn, except for Dean, Missouri concerning use of "pasteurized" milk for cheese­ who considered the quality of "pasteurized" milk Cheddar making. All the factories "pasteurized" milk for cheese­ cheese to be not very satisfactory. Dean, however endorsed making, although the total number was not reported by milk "pasteurization" for Camembert (103). Reid. "Pasteurization" temperatures were 61.1-62.8°C (142- Especially noteworthy is the rapid adoption of "pas­ 145°F) for holder treatment and 73.9-76.7°C (165-170°F) teurization" in New Zealand, described by Stevenson for flash. Conditions, practices, and opinions varied widely (327,328) and Stevenson and Grant (329). Danish-built in these small enterprises, (6,000-30,000 lbs. of milk re­ "pasteurizers" were first imported into New Zealand in ceived per day) but control of cheesemaking process and 1914. During the 1914-15 season, "pasteurizers" were better quality were consistently listed as reasons for operating in seven plants. By 1916, 32 plants were "pas­ "pasteurizing." teurizing" milk for cheesemaking; by 1920, 141 plants Sammis (305), in the 1930 edition of "Cheese Mak­ were using the process. By 1923, approximately two-thirds ing" stated that "Favorable reports were received from of all cheese made in New Zealand was made from over 50 large cheese factories in the United States, pas­ "pasteurized" milk. "Pasteurization" temperatures were 71.1- teurizing milk for American cheese manufacture, in nearly 73.9°C (160-165°F), holding time not specified. Stevenson all cases by the flash process." described "pasteurization" as " in all respects an unquali­ Price and Germain (265), in 1938, studied inexpen­ fied success," with improvement of quality foremost among sive "pasteurizing" units for small cheese factories. They the advantages obtained, which also included increased comment: yield, lower fat losses, and better control of the cheese- "More general use of 'pasteurized' milk for Ched­ making process. The major New Zealand market for cheese dar cheese in the United State has undoubtedly been was the United Kingdom — halfway around the world. discouraged by the cost of the equipment. The major­ The enthusiasm for "pasteurization" must have reflected, ity of American cheese factories in Wisconsin, for in part, the technological challenge of manufacturing cheese example, have a total value of approximately $3,500, which could withstand a voyage of several months enroute one-half of which usually represents the total invest­ from factory to consumer. ment in equipment. The usual type of pasteurizing machinery of sufficient capacity to meet the needs of D. Commercial acceptance. these factories would actually cost almost as much or Widespread acceptance of milk "pasteurization" for perhaps more than the total investment in all other cheesemaking appeared to come slowly in the United States. equipment. Such machinery would obviously beyond

JOURNAL OF FOOD PROTECTION. VOL. 53, MAY 1990 446 JOHNSON, NELSON. AND JOHNSON

the means of the small cheese factory receiving less Health Service began in 1926 (14), and in 1937, the Dairy than 10,000 pounds of milk a day." and Ice Cream Machinery and Supplies Association pub­ The 1937 and 1942 editions of "Cheese Making" by lished a resolution to promote nationwide adoption of dairy Sammis (306,307) continued to include "pasteurization in equipment specifications (14). the cheese industry" in the chapter "Other Sorts of American The 1942 (307) and 1946 (308) editions of Sammis' Cheese." It is curious that Sammis, after titling his 1912 "Cheese Making" contain illustrations of plate-type pas­ report "The Manufacture of Cheddar Cheese from Pasteur­ teurizing equipment, including holding tubes recording ized Milk," did not continue the use of the Cheddar name. thermometers and controls. In his 1918 edition, the product had no name, and was Heat treatment descriptions in early research often lack classified a "minor sort of cheese." In subsequent editions holding times, and temperature ranges are often quoted, the cheese is called "American". Did the cheese experts of implying imprecise process control. Table 1 lists the the day consider "pasteurized" milk cheese a new variety? temperature-time combinations reported by several early investigators. Relatively few heat-treatments used in the

E. Equipment and process. U.S. cheese industry have been published. Listed in Table Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 The 1930 edition of "Cheese Making" by Sammis 2 are an informal FDA recommendation (32), three treat­ (305) contained pictures and descriptions of "modern" flash ments published by Kosikowski (189), time and tempera­ "pasteurization" equipment used to process milk for cheese- ture ranges presented by Reinbold (278), and treatments making. Three sections of upright "tubular coolers" obtained via personal communication (220). comprised the unit. Cold, raw milk was pumped through Also listed is the heat-treatment temperature range, a regenerative section, where it was warmed by milk at 60-69°C (140-156.2°F) for 28 Canadian Cheddar factories "pasteurizing" temperature flowing downward by gravity (98). Holding times were not reported. over the exterior tubular cooler surface. The warmed milk then flowed through a distributor pipe onto the exterior F. Debate and controversy. surface of the steam-heated "flash" section, over the flash In 1936, nearly 30 years after Bruhn and Sammis started heater surface, then the regenerator surface, and finally their landmark investigation of milk "pasteurization" for over a surface cooler where it reached setting temperature. cheese making (previously described), the National Butter Collected in a trough below the cooler section, milk flowed and Cheese Journal, then the leading cheese industry trade by gravity to a cheese vat. The illustrations of these early journal, surveyed industry leaders' viewpoints on the subject "pasteurizers" show neither indicating nor recording ther­ (9,10,11,12,13). Among those who opposed "pasteuriza­ mometers, controllers, holding tubes, nor flow diversion tion," about one-fourth of the respondents pointed to lower valves. flavor levels and retarded ripening, and the encourage­ The effectiveness of early "pasteurizers" was unde­ ment of cheesemakers to accept poor quality milk. A northern fined until 1920 when the Bureau of Dairy Industry, United industry leader insisted that "pasteurization" by northern States Department of Agriculture developed tests for gauging factories would result in cheese indistinguishable from cheese holding times. Research on pasteurizer efficiency by the produced via "pasteurization" in southern factories. Office of Milk Investigation of the United States Public Proponents of "pasteurization" pointed to the superior

TABLE 1. Milk heat treatments for cheesemaking utilized by early investigators. Temperature Time Investigators(s). Year Reference °C 71.1-73.9 160-165 ns Sammis and Bruhn 'flash", 1912 309 60.0 140 30 m Sammis and Bruhn 'holder", 1912 309 71.1-73.9 160-165 ns Stevenson (New Zealand) "flash", 1920 327 62.8 145 30 m Price, Price and Prickett "holder", 1928 257,258,264 71.1-73.9 160-165 ns Price and Prickett "flash", 1928 264 71.1-73.9 160-165 (a) Price and Prickett "flash-holder", 1928 264 71.1-75.6 160-168 25 s Phillips, 1928 252 62.8 145 30 m Marquardt "holder", 1929 208 62.8-65.6 145-150 ns Marquardt "holder", 1929 208 61.1-62.8 142-145 ns Reid "holder" survey, 1929 275 73.9-76.7 165-170 ns Reid "flashy" survey, 1929 275 71.1-73.9 160-165 15 s Price and Germain "flash", 1938 265 62.8 145 20-30 m Price and Germain "holder", 1938 265 62.0 143 30 m Price and Call "holder", 1942 266 62.8 145 20 m Price and Call "holder", 1942 266 71.1 160 15 s Price and Call "flash", 1942 266 50-69 122-156 15-20 s Price and Call less than "flash", 1942 266 ns Not Specified. (a) Held over until heating completed, est. 22 m at not lower than 61°C (142°F).

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 MICROBIOLOGICAL SAFETY OF CHEESE 447

TABLE 2. Contemporary heat treatments. Temperalure Time, (s) Source, Year Reference

65.6 150 16 FDA, 1970 31 60.0 140 16 Kosikowski, 1977 189 62.8 145 7 Kosikowski, 1977 189 65.6 150 2 Kosikowski, 1977 189 60.0-65.6 140-150 16-17+ Reinbold, 1985 279 62.8 145 18 Middleton, 1986 220 63.9 147 16 Middleton, 1986 220 66.7 152 16 Middleton, 1986 220 68.3 155 18 Middleton, 1986 220 70.0 158 16 Middleton, 1986 220 60.0-69.0 140-156.2 ns D'Aoust et al., 1987 98 ns Not Specified. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021

keeping quality of "pasteurized" milk cheese and also the milk cheese. A second correspondent, Miollis (22) stated cleaner, milder flavor which resulted from aging. Some that Bleu cheese and Swiss cheese could be made com­ respondents who favored "pasteurization" hedged by stat­ mercially from "pasteurized" milk. He questioned whether ing that "pasteurization" was not necessary at all times. raw milk cheese should be marketed. In subsequent ar­ Product safety was stressed by three proponents. Clearly, ticles published during the post World War II euphoria of "pasteurization" of milk for cheesemaking was a contro­ 1945 (223,224) Miollis predicted, among other develop­ versial topic in 1936. ments, compulsory "pasteurization" of milk for cheese­ In 1938 Price (259,260) thoroughly assessed all major making. aspects of "pasteurization" — raw milk quality, applica­ tion for various cheese varieties, plant operations, costs, G. World War II. cheese quality, process control, marketing, and public health. The outbreak of World War II in 1939 triggered major He summarized ". . . it seems reasonable to regard pas­ changes in the United States dairy industry, including a teurization as both a safeguard for the public health and a shift from butter-powder production to cheese production protection for the quality and uniformity of the cheese ..." (15). In March, 1941, Congress passed the Lend-Lease A trade association monograph (14) on pasteurization Act which enabled major supply support to Great Britain, equipment and process published in 1938 contained a including food. Britain requested cheese but not butter. In statement that "In the South, where cheesemaking today is April 1941, eight months before the attack on Pearl Har­ relatively young, practically all milk used in cheesemak­ bor, the government announced that it would purchase 150 ing is pasteurized. In Wisconsin, our largest cheese pro­ million pounds of Cheddar cheese by June 1942 (143). ducing state, it is estimated that less than 5% of all cheese Government purchase was contingent upon the cheese milk is pasteurized. The percentage in New York is still meeting U.S. Grade 1 (later Grade A) requirements. Quotas, smaller." called "set asides" were established for Cheddar cheese The topic continued to attract interest as evidenced in manufacturers which mandated that a percent of total cheese a 1944 article by Price (261) entitled "Can Pasteurized production be offered and accepted for government pur­ Milk Cheese COMPETE with Raw Milk Cheese." Price chase. Set asides were as high as 70% during May-July, stated that it could compete and win, whether the compe­ 1943 (17), and continued until 1947. Almost overnight, tition was for milk, labor, a buyer, the consumers' dollar, the Cheddar industry was obliged to sharply increase pro­ or the approval of the health official. However, Price posed duction of consistently high quality cheese. Penalties, in­ another question: ". . . (should) cheese made from raw cluding fines and imprisonment (21) could be assessed milk be outlawed?" His answer: against management that failed to deliver. "Certainly not! Where milk is carefully produced, in­ The USDA urged cheesemakers to use "pasteuriza­ spected, and skillfully manufactured in a clean, sani­ tion" to control quality (76). The USDA and academia tary factory it has every right to command the respect joined forces to encourage and assist the industry to improve of the industry, the approval of the health official, and cheese quality. Articles by H. L. Wilson of USDA (363) a premium from the consumer." and Wilster (365), among others, promoted "pasteuriza­ Price had stated in this article that technical difficul­ tion" and other quality practices. ties barred the use of "pasteurized" milk for Swiss, Roque­ Industry response was impressive. Hundreds of mil­ fort, and Italian varieties. Bertonasco, a prominent indus­ lions of pounds of Cheddar cheese were provided to allies try technologist challenged this statement (20), asserting and the U.S. Armed Forces. Although cheese was among that all Italian varieties were routinely produced from those foods rationed during the war, civilian consumption pasteurized milk. He concurred with Price's views on raw continued to be significant.

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 448 JOHNSON, NELSON, AND JOHNSON

Near the end of the war, Wilson et al. (364) published equipment, consumption of fresh cheese, and production the results of USDA research on milk "pasteurization" for of so-called new varieties to evade wartime price controls. cheesemaking. The work had begun in 1927 and continued Before these outbreaks, other cheese-related foodborne through much of World War II. The initial phase of the disease outbreaks had occurred. Fabian (123), in a com­ USDA studies was pilot scale manufacture (800 lb milk prehensive paper entitled "Cheese and Its Relation to Dis­ per vat; 21 vat pairs) of Cheddar cheese, comparing raw ease" listed 59 United States "epidemics" attributed to milk with "pasteurized" milk. After curing for one year at cheese which occurred between 1883 and 1946, compris­ 50°F, the cheese was graded. Cheddar cheese made from ing 2,904 cases of disease and 117 deaths. Forty of these "pasteurized" milk consistently exhibited more uniform, were reported between 1935 and 1945. Fabian comments higher quality than raw milk cheese. Further experiments that cheese may have been overlooked as a source of disease! revealed that, although "pasteurization" could improve the Several states reacted promptly to the Surgeon Gen­ quality of cheese made from poor quality milk, such cheese eral's suggestion. California led the way (18) enacting within consistently exhibited inferior quality, compared to cheese a month a law requiring that all cheese sold to retailers be made from good quality milk. either: Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 The second phase of the USDA work comprised trials (a) made from pasteurized milk, in factories. Pilot scale results were confirmed, particu­ (b) pasteurized, or larly the pivotal role of milk quality. A time schedule was (c) cured for a minimum of 60 d. developed, based upon milk "pasteurization" and control New York, New York City, Colorado, Indiana, Illi­ of acid development, which enabled production of cheddar nois, Missouri, New Jersey, and Minnesota enacted similar cheese with consistently high quality, desirable flavor, and legislation (23,26). rapid curing. Via the joint sponsorship by the USDA and Industry also reacted promptly. A group of industry the University of Wisconsin, the time schedule concept of leaders convened at the University of Wisconsin under the cheesemaking evolved into a widely used, dependable auspices of the National Cheese Institute. In December method, nicknamed the "clock method" (262,263). The 1944, they issued a report (19). The first three of six principles of the "clock method" have been applied to recommendations stated: varieties other than Cheddar. " 1. Pasteurization of milk is a desirable practices for Wartime shortages impeded the fabrication of pas­ many varieties of cheese and its application to teurization equipment. Immediately after the war the number other varieties should be encouraged. of factories equipped to "pasteurize" increased sharply (28). 2. When milk is properly pasteurized by methods and In 1944, 42% of U.S. cheese factories, making 51% of the with equipment recommended or approved by health total annual cheese production were equipped to "pasteur­ officials then the cheese made from it should contain ize." By 1949, these percentages had increased to 71 and no pathogenic organisms which might have been 80% respectively. The survey did not determine whether in the original milk. Such pasteurization does not the "pasteurization" practices in each factory complied insure the cheese against subsequent contamina­ with legal criteria. tion from some other source. Some textbooks provide overviews of milk heat-treat­ 3. Holding of cheese, if it is to be used as means of ment, pasteurization, and the related technology including, protecting the public health, should be defined in Davis (101), the second edition of Cheese by Van Slyke terms of temperature of storage as well as dura­ and Price (352), and Kosikowski (189). Knutson et al. tions of holding period for each type of cheese to (186) review in detail the evolution of pasteurization. be controlled." Reinbold (279) describes the rationale for milk heat-treat­ The committee endorsed heat-treatments of milk for Cheddar ment. cheese and Swiss cheese. Gilman et al. (147), in 1946, published research on H. Concerns about the safety of cheese. the occurrence and survival of Brucella abortus in cheese, The Surgeon General of the United States, in a letter which was also reported in a trade journal by Dahlberg (248) dated June 16, 1944 to state health officers, de­ (93). Cheese made from milk inoculated with 500 B. abor­ scribed two serious outbreaks of typhoid fever in which tus organisms per ml, and cured at 4°C (40°F), contained cheese "has been the actual or suspected vehicle." He also viable B. abortus after six months but not after one year. noted other, minor, outbreaks of other diseases in which The use of milk from infected cows containing 700-800 B. cheese was implicated. He described the potential dangers abortus organisms per ml gave similar results. Further trials of the consumption of fresh, uncured cheeses and con­ were conducted in New York and Wisconsin factories using cluded by suggesting that "all cheese be either adequately pooled milk. Three Wisconsin cheeses which were posi­ ripened (e.g. cured) or made from pasteurized milk." State tive for Brucella abortus immediately after manufacture legislation to implement appropriate programs and better tested negative after 66-84 d storage. sanitation was also recommended. Friedel (139) provides perspective on conditions at I. Standards of identity the time. He attributed the outbreaks principally to war­ In early 1946; the Food and Drug Administration time conditions — a shortage of skilled cheesemakers, poor published proposed standards for several cheeses. A public

JOURNAL OF FOOD PROTECTION. VOL. 53, MAY 1990 MICROBIOLOGICAL SAFETY OF CHEESE 449 hearing on the proposal was convened in April, 1947 (24). for a period which will insure the death of all patho­ Testimony read from a prepared statement by the gens. chairman of the National Cheese Institute Research —If cheese is held for 60 d at 2°C (35°F) or higher, Committee (25) described Institute-sponsored research on it is reasonable to expect that cheese will be safe. cheese safety, including pasteurization of milk for cheese- —When cheese is made from pasteurized milk no making and holding periods for raw milk cheese. Those holding period is necessary to insure microbiological making cheese from pasteurized milk did so to control safety. "gas organisms" and to produce cheese of more uniform —A holding period may be necessary to develop iden­ quality with higher yield. "Health hazards were not part of tity characteristics which result from curing the cheese. the thinking in 1939", the chairman stated. Wartime out­ The 39 natural cheese standards promulgated by FDA breaks of typhoid fever resulting from contaminated cheese in 1949 (29,30) provided two options for producing safe had caused great concern. products; pasteurization of milk for cheesemaking or holding Professor A. C. Dahlberg of Cornell University testi­ the finished cheese for no less than 60 d at temperatures fied about his research (25) "to establish the length of time of 35°F or above. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 that selected disease-producing bacteria would survive in Table 3 lists the current U.S. cheese standards (37), Cheddar cheese." Three pathogens, the "tubercle bacillus," including whether pasteurization is required. Those origi­ Eberthella typhosa (now known as Salmonella typhi) and nally promulgated in 1949 are marked with an asterisk. Brucella abortus were considered for study. Brucella abor­ Table 4 summarizes pasteurization and curing require­ tus was selected because it could be obtained in naturally ments in the standards. infected milk. Although Dr. Dahlberg described his re­ This comprehensive promulgation of standards was a search findings that pathogens survived more than 60 d in major milestone of the Unites States cheese industry. The cheese, (refer to preceeding section) he endorsed 60 d standards provided a definitive legal framework which holding at 2°C (35°F) as adequate to protect public health, facilitated the regulation of major Unites States cheese and stressed the importance of temperature. varieties and promoted "honesty and fair dealing." Other witnesses endorsed the 60 d holding period (25). Opinion was divided on the merits of milk pasteurization J. Four Decades of Progress for cheesemaking. Milk heat-treatment was not consid­ During the 40 years since Standards promulgation, the ered, and was described as "a special case" when men­ industry has maintained a remarkable record of techno­ tioned during testimony on make procedures for Swiss logical achievement and innovation. Raw milk quality has cheese. been steadily upgraded as herd management and milk The findings of fact from those hearings (29) include collection practices have evolved. Cheesemaking facilities the statement: and equipment have continued to improve consequent to "Consumers expect, and have a right to expect, that better design and materials of construction, cleanability, manufacturers of cheese shall take reasonable precau­ product protection, and mechanization. Cheesemaking tions to render the finished cheese safe for human procedures have benefitted from research which has re­ consumption." sulted in revolutionary improvements in starter cultures, Another finding of fact delineated the status of cheese enzymes, process automation, and packaging. Product quality safety. and safety management has continued to be strengthened, Concerning pathogens in milk and cheese: currently via implementation of HACCP — //azard Analysis —Milk from infected animals may contain pathogens. Critical Control foint and SQC — Statistical Quality Con­ —It is not possible to insure the use of pathogen-free trol programs. milk for cheesemaking. —Milk can be infected by persons who handle it. Cheese production trends, 1948-1987 —Pasteurization kills pathogens; cheesemaking tem­ Annual cheese production in the United States, ex­ peratures do not cluding cottage and bakers cheese, increased almost five­ —Pathogens in milk are usually carried into cheese fold, between 1948 and 1987 (27,31). Production in 1948 and may infect consumers. was 1.098 billion lb, in 1987 5.344 billion lb. A grand —No disease outbreaks have been reported to have total of just over 100 billion lb of cheese was produced been caused by cheese held 60 d or more. during those 39 years. Note that the effectiveness of both pasteurization and American varieties (Cheddar, Colby, Washed Curd, the 60 d hold were endorsed. Another finding of fact Granular, Monterey, and High Moisture Jack) comprised concerning cheese holding contained an endorsement, but 78% of 1948 production. By 1987, the percentage of also a disclaimer about the effectiveness of the 60 d holding American varieties had declined to 51%, reflecting the period: explosive growth of Italian varieties from less than 4% of —Pathogens tend to die when cheese is held at 2°C annual production in 1948 to 34% in 1987. (35°F) or higher. Table 5 displays production statistics, 1948 vs. 1987, —The length of a "safe" holding period is uncertain. for ten cheese varieties for which milk pasteurization is —It is "unreasonable" to require the holding of cheese mandatory. Four mozzarella types and Ricotta comprise

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 450 JOHNSON, NELSON, AND JOHNSON

TABLE 3. Milk pasteurization and minimum curing periods for standardized natural cheeses (37). 21 CFR Variety Milk Minimum curing pasteurization required? time 133.102 Asiago fresh, Asiago soft* No 60 d 133.103 Asiago medium* No 6 months 133.104 Asiago old* No 12 months 133.106 Blue* No 60 d 133.108 Brick* No 60 d at 2°C (35°F) 133.109 Brick for manufacturing* No None 133.111 Caciocavallo siciliano No 90 d at 2°C (35°F) 133.113 Cheddar* No 60 d at 2°C (35°F) 133.114 Cheddar for manufacturing* No None 133.118 Colby* No 60 d at 2°C (35°F) 133.119 Colby for manufacturing* No None 133.123 Cold pack and club* All cheeses used are made from pasteurized milk or aged for not 133.124 Cold pack cheese food less than 2°C (35°F) for 60 d. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 133.125 Cold pack cheese food with fruits, vegetables, or meats. 133.127 Cook, Koch Kaese* No -- product heated to 82°C (180°F) 133.128 Cottage Yes None 133.129 Dry curd Cottage Yes None 133.131 Low fat Cottage Yes None 133.133 Cream Yes None 133.134 Cream with other foods Yes None 133.136 Washed curd, soaked curd* No 60 d at 2°C (35°F) 133.137 Washed curd for manufacturing* No None 133.138 Edam* No 60 d at 2°C (35°F) 133.140 Gammelost* No — - product heated to 62.7°C (145°F) 133.141 Gorgonzola* No 90 d 133.142 Gouda* No 60 d at 2°C (35°F) 133.144 Granular and stirred curd* No 60 d at 2°C (35°F) 133.145 Granular for manufacturing* No None 133.148 Hard grating* No 6 months 133.149 Gruyere* No 90 d 133.150 Hard* No 60 d at 2°C (35°F) 133.152 Limburger* No 60 d at 2°C (35°F) 133.153 Monterey & Monterey Jack* Yes None 133.154 High Moisture Jack Yes None 133.155 Mozzarella and Scamorza Yes None 133.156 Low moisture Mozzarella and Scamorza Yes None 133.157 Part skim Mozzarella and Scamorza Yes None 133.158 Low moisture part-skim Mozzarella Yes None and Scamorza 133.160 Muenster and Munster* Yes None 133.161 Muenster and Munster for manufacturing No None 133.162 Neufchatel Yes None 133.164 Nuworld No 60 d 133.165 Parmesan and Reggiano* No 10 months 133.181 Provolone* No 60 d at 2°C (35°F) 133.182 Soft ripened* No 60 d at 2°C (35°F) 133.183 Romano* No 5 months 133.184 Roquefort* No 60 d 133.185 Samsoe No 60 d at 2°C (35°F) 133.186 Sapsago* Milk "heated to boiling temperatures" 133.187 Semisoft* No 60 d at 2°C (35°F) 133.188 Semisoft part skim* No 60 d at 2°C (35°F) 133.189 Skim milk for manufacturing* No None 133.190 Spiced* No 60 d at 2°C (35T) 133.191 Part-skim spiced* No 60 d at 2°C (35°F) 133.193 Spiced, flavored standardized No 60 d at 2°C (35°F) 133.195 Swiss and Emmentaler* No 60 d 133.196 Swiss for manufacturing* No 60 d *Originally promulgated in 1949.

JOURNAL OF FOOD PROTFCTION, VOL. 53, MAY 1990 MICROBIOLOGICAL SAFETY OF CHEESE 451

TABLE 4. Summary of milk pasteurization and minimum curing made from heat-treated milk. 1987 production statistics periods for natural cheese varieties. list 1,188 million lb of process cheese and 673 million lb Requirements Number of foods and spreads. These tonnages were converted to Milk pasteurization mandatory 13 ingredient cheese equivalent, assuming 0.88 lb of ingredi­ Pasteurized during cheesemaking process* 3 ent cheese per lb of process cheese and 0.61 lb of ingre­ Cheese for manufacturing - neither milk pasteurization dient cheese per lb of food or spread {375). The quantities nor holding period required 7 of ingredient cheese thus calculated were 1,045 million lb Milk pasteurization or 60 d hold at 2°C (35°F) 18 for process cheese and 411 million lb for foods and spreads. Milk pasteurization or 60 d hold, no temperature The total of these two equivalents, 1,456 million lb, com­ specified 6 Milk pasteurization or 90 d hold at 2°C (35°F) 1 prises 27% of 1987 total production. Standards of Identity Milk pasteurization or 90 d hold, no temperature for process cheeses, foods, and spreads require that these specified 2 foods be pasteurized (57). Mandatory 5-12 month hold, no temperature specified 4 Thus, 65% of 1987 natural cheese production was Cold pack products — must use cheese from pasteurized comprised of the ten natural varieties (Table 5) for which Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 milk or cured not less than 60 d at 2°C (35°F) 3 pasteurization is mandatory and ingredient cheese utilized Total number of standards. 57 in pasteurized process cheeses, foods, and spreads. *Pasteurization of cheese milk optional, heating during cheese- Table 6 displays production statistics for cheese varie­ making equivalent to or exceeds pasteurization. ties wherein milk pasteurization is optional. The data reflect consistent growth in all varieties except Limburger, with the most rapid growth in hard Italian varieties.

TABLE 5. Statistics for ten cheese varieties for which milk pasteurization is mandatory — 1948 and 1987 (a). TABLE 6. Statistics for cheese varieties for which pasteuriza- Millions of lbs. tion is optional — 1948 and 1987 (a). Group/variety 1984 1987 Production, millions of lbs. Soft Italian est. 14 1,534 Group/variety 1984 1987 Mozzarella — 4 types (b) American 858 2,637 (b) Ricotta Cheddar Cream, Neufachatel 57 330 Colby Munster 8 93 Washed curd Monterey, High Moisture Jack 80 (c) Granular Totals 79 2,037 Hard Italian varieties est. 30 266 (Percent of annual production) (8) (38) Asiago (a) Data from USDA (27,31) and Heizer (161) unless otherwise Parmesan stated. Provolone (b) Mozzarella, part skim Mozzarella, low moisture Mozzarella, Romano and low moisture part skim Mozzarella. Swiss 71 227 (c) Data from Sales Area Marketing, Incorporated (SAMI), Blue mold 9 36 September 1987 - August 1988 (375). Brick, Limburger 26 29 All other varieties 4 99 Totals 998 3,374 (Percent of annual production) (91) (62) (a) Data from USDA (27,31) and Heizer (767). a second grouping, Munster is listed singly, and a third (b) Does not include Monterey and High Moisture Jack tonnage. grouping comprises Monterey and High Mositure Jack. Mandatory milk pasteurization is specified in Standards of Identity for nine of the ten varieties (57). Ricotta, for which Mild Cheddar and Colby are routinely made from there is no federal Standard of Identity, is subjected to pasteurized milk 1987-88 sales of these two categories time-temperature treatments which exceed pasteurization were: mild Cheddar, 357 million lb; and Colby, 109 million during manufacture (189). These ten varieties comprised lb (375). Blue, Brick, and Limburger are routinely made 38% of total annual production in 1987. from pasteurized milk, and 1987 production of these va­ Three cheese varieties, dry curd cottage cheese, cot­ rieties totaled 65 million lb. tage cheese, and lowfat cottage cheese, are listed sepa­ Sharp Cheddar is frequently made from heat-treated rately from other USDA cheese production statistics. milk. Sharp Cheddar sales in 1987-88 totaled 211 million Mandatory milk pasteurization is specified in the stan­ lb (375). Swiss and hard Italian varieties are predomi­ dards for all three varieties. Production in 1986 was 600, nantly made with heat-treated milk. Table 6 lists 1987 705, and 265 million lb, respectively. Swiss production of 227 million lb and of 266 million lb Process cheese, cheese foods, and cheese spreads utilize of hard Italian varieties. enormous quantities of natural cheese. The overwhelming Table 7 displays an analysis of pasteurization prac­ proportion is comprised of American varieties, frequently tices for cheesemaking in 1987. An estimated 48% of natural

JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990 452 JOHNSON, NELSON, AND JOHNSON cheese sold was made from pasteurized milk; 38% com­ duction. Cold pack cheese and cheese food production in prising varieties for which pasteurization is mandated, another 1987 totaled 185 million lb, but the proportions of pasteur­ 10% where pasteurization is optional but routinely prac­ ized milk cheese and heat-treated milk cheese could not ticed. Process cheese and related products utilized 27% of be ascertained. Substantial quantities of American cheese 1987 annual cheese production. are used in other processed foods — soups, sauces, pack­ aged dinners, and snacks. Food service is a substantial TABLE 7. An analysis of pasteurization and heat treatment market for natural American varieties. practices for cheesemaking — 1987. The "all other varieties" category, comprising 2% of Category Millions % of of lbs. annual total total production, includes Edam, Gouda, Port Salut, Havarti, Mandatory milk pasteurized — Brie, Camembert, and Feta (161). Although many of these 10 varieties 2,037 38 cheeses are routinely produced from pasteurized milk, no Utilized in pasteurized process varieties 1,456 27 statistics which define pasteurization use were found. Pasteurization optional but frequently 531 10 Cheese varieties which are frequently manufactured used from heat-treated milk totaled 704 million lb, comprising Downloaded from http://meridian.allenpress.com/jfp/article-pdf/53/5/441/1660707/0362-028x-53_5_441.pdf by guest on 30 September 2021 - Mild Cheddar, Colby, Blue mold, Swiss, hard Italian types, and sharp Cheddar, 13% of total Brick, and Limburger cheese production. Thus the exposure of consumers to direct Not defined consumption of U.S. produced cheeses made from heat- - American varieties utilized in: 517 10 - cold pack cheese and food treated milk is moderate, and substantially comprised of - other processed foods hard varieties. The make procedures for Swiss and hard - foodservice Italian varieties use relatively high cooking temperatures. All other varieties 99 2 The curd of Italian pasta filata varieties such as Provolone Milk heat treatment frequently 704 13 is heated to 57-60°C (135-140°F) during mixing and molding utilized - Swiss, Hard Italian, and (189). Sharp Cheddar The primary sources of the statistics in this section are Totals 5,344 100 the USDA Dairy Products Annual Summaries (27,31) which The "not defined" category in Table 7 comprises 12% are production statistics. Some additional production data of total production, including three utilization options for were obtained from Heizer (161). Sales data from Sales American cheese varieties comprising 10% of total pro­ Area Marketing, Inc. (SAMI) (375) were also provided.

ENVIRONMENTAL SYSTEMS SERVICE, LTD. • Testing for Listeria and other Pathogens • Dairy, Poultry and Food Product Testing • Water and Wastewater Analysis • Bioassay - Toxic Monitoring • Hazardous Waste Analysis • Sanitation Inspections and Air Quality Monitoring 218 N. Main Street 5111 College Avenue Culpeper, VA 22701 College Park, MD 20740 703-825-6660 800-541-2116 301-779-0606 V= J Please circle No. 349 on your Reader Service Cared JOURNAL OF FOOD PROTECTION, VOL. 53, MAY 1990