140

Journal of Food Protection, Vol. 73, No. 1, 2010, Pages 140–202

Supplement

Parameters for Determining Inoculated Pack/Challenge

Study Protocols34
ADOPTED 20 MARCH 2009, WASHINGTON, D.C.
NATIONAL ADVISORY COMMITTEE ON MICROBIOLOGICAL CRITERIA FOR FOODS

NACMCF Executive Secretariat,* U.S. Department of Agriculture, Food Safety and Inspection Service, Office of Public Health Science,
Room 333 Aerospace Center, 1400 Independence Avenue S.W., Washington, D.C. 20250-3700, USA

MS 09-287: Received 2 July 2009/Accepted 7 August 2009

ABSTRACT
The National Advisory Committee on Microbiological Criteria for Foods developed guidelines for conducting challenge
studies on pathogen inhibition and inactivation studies in a variety of foods. The document is intended for use by the food
industry, including food processors, food service operators, and food retailers; federal, state, and local food safety regulators;
public health officials; food testing laboratories; and process authorities. The document is focused on and limited to bacterial
inactivation and growth inhibition and does not make specific recommendations with respect to public health. The Committee
concluded that challenge studies should be designed considering the most current advances in methodologies, current thinking on
pathogens of concern, and an understanding of the product preparation, variability, and storage conditions. Studies should be
completed and evaluated under the guidance of an expert microbiologist in a qualified laboratory and should include appropriate
statistical design and data analyses. This document provides guidelines for choice of microorganisms for studies, inoculum
preparation, inoculum level, methods of inoculation, incubation temperatures and times, sampling considerations, and
interpreting test results. Examples of appropriately designed growth inhibition and inactivation studies are provided.

SCOPE OF DOCUMENT food, the NACMCF is asked for its guidance to clarify these
issues.
This document was prepared at the request of the
sponsoring agencies of the National Advisory Committee on 1. What are the appropriate criteria that must be
Microbiological Criteria for Foods (NACMCF). The considered for an inoculated pack/challenge study to
document is intended for use by the food industry, including determine if a food requires time/temperature control
food processors, food service operators, and food retailers; for safety (TCS)? For example, pathogen species/strain
federal, state, and local food safety regulators; public health selection, use of surrogate organism, number of
officials; food testing laboratories; and process authorities. pathogen strains, inoculation level(s), incubation tem-
The document is focused on and limited to bacterial perature(s), length of incubation/duration of study, food
inactivation and growth inhibition. The document does not product physical properties, etc.
consider toxigenic fungi or the inactivation of viruses. 2. What are the appropriate uses of mathematical growth
and inactivation models? Under what conditions can
INTRODUCTION AND STATEMENT OF CHARGE these models be used as a substitute for inoculated pack/
Statement of Charge challenge studies? Of the models currently available,
which ones are most suitable for use, and what are the
Because of the many questions raised by regulatory and limitations of these models?
industry users on the definition of potentially hazardous 3. What are the limitations for applying the results of an
food (PHF) or time/temperature control for safety (TCS) inoculated pack/challenge study conducted on one food
to another similar food?
4. Of the existing inoculated pack/challenge study proto-
* Author for correspondence. Tel: 202-690-6600; Fax: 202-690-6364; cols, e.g., those published by the American Bakers
E-mail: [email protected]. Association, NSF International, and others, which are
{ Sponsored by the U.S. Department of Agriculture, Food Safety and
Inspection Service; U.S. Department of Health and Human Services,
most suitable for application to a wide variety of foods,
Food and Drug Administration, and Centers for Disease Control and and what are the limitations of these protocols? Are there
Prevention; U.S. Department of Commerce, National Marine Fisheries existing protocols that are appropriate for specific food-
Service; and U.S. Department of Defense, Veterinary Service Activity. pathogen pairs?
This article may be reproduced without prior permission. 5. Develop a decision tree to aid in the design of an
{ Mention of trade names or commercial products in this publication is
solely for the purpose of providing specific information and does not appropriate inoculated pack/challenge study. Test or
imply recommendation or endorsement by the U.S. Department of ‘‘desk check’’ the decision tree using the following five
Agriculture and other NACMCF sponsoring agencies. foods: meat-filled puff pastry, (baked) cheese pizza,
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 141

chopped lettuce, cheese (blocks or slices), and lemon The IFT report (53) with its recommendations to the
meringue pie. FDA left a number of unanswered questions regarding the
6. Identify the basic knowledge, skills, education, training, understanding and implementation of a product assessment
experience, and abilities necessary for a multidisciplinary when based on pH and aw the producer is unable to
work group or individual to be qualified to design, determine whether TCS is required. This need for
conduct, and evaluate an inoculated pack/challenge study clarification was confirmed in a 2005 survey of stakeholders
and the pursuant results. conducted by the Conference for Food Protection (18).

THE COMMITTEE’S RESPONSE

Background
The restaurant and retail food store industry, totaling Use and Limitations of This Document
nearly 1.5 million establishments in the United States, and The primary objective of this document is to provide
their suppliers routinely use inoculation/challenge testing guidelines for challenge studies necessary to determine
to determine whether a specific food requires TCS. A food whether a variance to TCS may be granted under the Food
establishment, including restaurants, retail food stores, Code. Secondarily, the guidelines presented in this docu-
delis, caterers, and institutions or vending commissaries ment may be useful to laboratories conducting pathogen
that provide food directly to the consumer, is defined in inhibition and pathogen inactivation studies for a variety of
the U.S. Food and Drug Administration (FDA) Food Code foods for evaluation of safety prior to introduction into
(117). commerce. It may be useful to review the proposed study
When laboratory testing is used to support a change in with the appropriate regulatory agency to ensure the design
how the product is handled in a food establishment (e.g., and methods are appropriate. Studies should be completed
refrigerated to unrefrigerated holding, extending shelf life, under the supervision of and interpreted by an expert food
increasing ambient temperature storage, or eliminating the microbiologist (Table 1). One of the limitations of these
need for date marking), the data are submitted to a state or studies is the balance of statistical validity with practicality.
local regulatory agency or directly to the FDA in the form of A certain amount of variability is expected with challenge
a variance application for approval. Food establishments or studies, which can affect the validity and interpretability of
manufacturers submitting laboratory data to support their results. However, because of resource constraints, variabil-
proposals must ensure the study is appropriate for the food ity is generally addressed through the use of worst-case
and pathogen of concern and incorporate the necessary scenarios, which should provide conservative results.
elements into the study to yield a valid design and Although this document encompasses a variety of sources,
conclusion. individuals who conduct challenge studies must be aware of
A variance from any provision in the FDA Food Code the most current advances in methodologies and identifica-
must also show that no health hazard will result from the tion of new pathogens or regulatory concerns that may need
modification or waiver and that product handling is under to be considered as well as pertinent statistical issues. This
appropriate control using a hazard analysis critical control document does not make specific recommendations with
point (HACCP) plan. Examples of foods in which the need respect to public health.
for TCS was questioned include puff pastries with savory
meat, cheese, or vegetable fillings; churros (fried dough) Types of Challenge Studies
batter held unrefrigerated; sliced pasteurized processed There are several types of challenge studies that deal
cheese held at ambient temperature for more than 4 h; with validation of food safety processing procedures,
certain cheeses held unrefrigerated. State and local regula- product storage conditions, and shelf life. Shelf-life studies
tors who evaluate a variance application based on this focusing on product quality are not addressed in this report
laboratory evidence need criteria to help them determine because they are generally not related to food safety.
whether the study was adequately designed and whether the Nevertheless, many of the principles of food safety–related
conclusions are valid. challenge studies are applicable to quality shelf-life studies.
The definition of PHF or TCS food was amended in the Food safety–related challenge studies differ according to the
2005 FDA Food Code (117); chap. 1, Definitions to include objective of the study, such as a pathogen growth inhibition
pH and aw interaction tables, allowing the hurdle concept to study or a pathogen inactivation study or a combination of
be used in the determination of whether TCS is necessary. the two, and depend on the type of product, production
The two interaction tables, as well as a decision-making process, and hazard analysis of the product.
framework, were developed by the Institute of Food Food safety–related challenge studies include the
Technologists (IFT) and provided to the FDA in the report, following:
‘‘Evaluation and Definition of Potentially Hazardous
Foods’’ (53) (31 December 2001, IFT/FDA contract Pathogen growth inhibition studies. These studies are
no. 223-98-2333, task order no. 4). When the pH and aw used to evaluate the ability of a particular food product
interaction tables and the decision-making framework are formulation with a specific type of processing and
insufficient to show that a food does not require TCS, packaging to inhibit the growth of certain bacterial
further product assessment using inoculation/challenge pathogens when held under specific storage conditions
testing is likely required. (time and temperature).
142 NACMCF J. Food Prot., Vol. 73, No. 1

TABLE 1. Recommended minimum expertise needed for designing, conducting, and evaluating microbiological studies a
Category Design Conductb Evaluate

Knowledge and Knowledge of food products and Knowledge of basic Knowledge of food products and pathogens
skills pathogens likely to be encountered in microbiological techniques. likely to be encountered in different
different foods. Knowledge of the Ability to work using aseptic foods. Knowledge of the fundamental
fundamental microbial ecology of technique, to perform serial microbial ecology of foods, factors that
foods, factors that influence dilutions, and to work at influence microbial behavior in foods,
microbial behavior in foods, and biosafety level 2 (114). and quantitative aspects of
quantitative aspects of microbiology. microbiology. Knowledge of statistical
Knowledge of processing conditions analysis.c
and parameters. Knowledge of
statistical design of experiments.c
Education and Ph.D. in food science or microbiology B.S. in food science, Ph.D. in food science, microbiology, or a
training or a related field or an equivalent microbiology, or a related related field or an equivalent
combination of education and field or an equivalent combination of education and
experience. combination of education and experience.
experience. Appropriate
hands-on experience in food
microbiology is also
recommended.
Experience Two years of experience conducting Two years of experience Two years of experience conducting
challenge studies independently and conducting challenge studies challenge studies independently and
experience in design of challenge is useful; however, close experience in evaluation of challenge
studies under the guidance of an supervision by an expert food studies under the guidance of an expert
expert food microbiologist. microbiologist may substitute. food microbiologist.
Abilities Ability to conduct literature searches. Ability to read and carry out an Ability to analyze and interpret
Ability to write an experimental experimental protocol. Ability microbiological data.
protocol. to perform microbiological
techniques safely and
aseptically.
a
State or local regulatory food programs that are presented an inoculation study in support of a variance request may not have expert food
microbiologists on staff to confirm the validity of the study. Options available to them include consulting with expert food microbiologists
in their state or local food laboratories or requesting assistance from FDA food microbiologists through their regional retail food
specialist.
b
Working independently under the supervision of an expert food microbiologist.
c
It may be appropriate to consult with a statistician with applicable experience in biological systems.

Pathogen inactivation studies. These studies are used intrinsic factors such as water activity (aw) and pH that
to evaluate the ability of a particular food product affect the likelihood of the product to support growth, the
formulation, a specific food manufacturing practice, or their use of processing technologies that destroy pathogens of
combination to cause the inactivation of certain bacterial concern, and the historical record of safe use of the product
pathogens. These studies may also be impacted by food (53, 80). In 2000, the FDA requested the IFT to assemble a
storage and packaging conditions and must account for scientific panel to examine the issue of determining when
these variables. foods required refrigeration for safety. In addressing their
charge, the panel defined these foods as TCS foods and
Combined growth and inactivation studies. Com- developed a framework for determining whether time and
bined studies may be used to evaluate the ability of a temperature control is required for safety. This framework
particular food or process to inactivate certain bacterial included two tables (one for control of spores and one for
pathogens and to inhibit the growth of certain other control of spores and vegetative cells) with aw and pH value
pathogenic bacteria or to achieve a level of inactivation combinations that indicate when product assessment (e.g., a
followed by inhibition of the growth of survivors or microbiological challenge study) is needed (53). This
contaminants introduced after processing. concept was subsequently adopted as the basis for defining
when foods need refrigeration or some other form of time-
Determining When a Challenge Study Is Needed temperature control in the 2005 FDA model Food Code
The first step in determining whether a challenge study (117). These aw and pH combinations are not specific to
is needed is to describe the product and process, conduct a individual pathogens; therefore, for specific foods where the
hazard analysis to determine the significant biological pathogen of concern is established, other pH and aw values
hazards, and assess what is known about the growth or may define the need for refrigeration. Information on
inactivation of these in the product (80). Consideration parameters associated with control of growth of various
should be given to potential routes of contamination, pathogens can be found in the literature, e.g., the
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 143

International Commission on Microbiological Specifica- species/strain selection, use of surrogate organism,

tions for Foods publication Microorganisms in Foods 5. number of pathogen strains, inoculation level(s),
Characteristics of Microbial Pathogens (54). When the incubation temperature(s), length of incubation/dura-
intrinsic factors of a food are consistent with parameters that tion of study, food product physical properties, etc.
are well recognized as controlling the growth of a pathogen,
microbiological challenge studies are not needed (91). For General Factors to Consider When Designing a
example, there would be no need to assess whether a Challenge Study
product with a pH of 3.5 supported growth of Salmonella,
because this organism will not grow at pH values this low. Standardization of methods is beneficial for comparing
However, studies to determine whether Salmonella survives results among different studies, but it is not possible to
at this pH or whether it is inactivated over time may be develop a single protocol that is broadly applicable to a wide
warranted under some circumstances. It is important to use variety of food types or even to one category such as fruits
expert food microbiologists and technologists to assess the and vegetables (12). Parameters that should be considered
need for challenge testing (Table 1). when designing a microbial challenge study are outlined
A challenge study may be needed to assess whether the below (12, 53, 80, 91, 122).
pathogen can grow in the product if properties such as pH, 1.0. Obtaining expert advice and identifying a laboratory
aw, or their combination do not ensure pathogen control. For 2.0. Type of study
more details on the use of pH and aw to control the growth 2.1. Growth inhibition studies
of bacterial pathogens, consult the Compendium of Methods 2.2. Inactivation studies
for the Microbiological Examination of Foods (90). 2.3. Combination studies
Determination of the need for a challenge study is referred 3.0. Factors related to the test product
to as product assessment in the IFT and Food Code tables 3.1. Product preparation
(53, 117). 3.2. Product variability
When growth inhibition occurs due to factors other than 3.3. Competitive microflora
or in addition to pH and aw, such as the addition of 4.0. Target organism(s)
preservatives including lactate and diacetate, the literature 4.1. Identifying the pathogen(s) of concern
may provide information relevant to the pathogen and food 4.2. Use of surrogate organisms
product. However, it is necessary to ensure that the data are 4.3. Type and number of strains
applicable to the specific product and conditions of use. The 5.0. Inoculum levels
efficacy of an antimicrobial agent may be dependent on the 5.1. Growth studies
formulation of the product. For example, factors such as fat 5.2. Inactivation studies
content can decrease the efficacy of antimicrobial agents 6.0. Inoculum preparation
such as nisin (36, 58) and sorbate (82, 98). Conversely, a 7.0. Method of inoculation
low pH may potentiate the activity of antimicrobials such as 8.0. Storage conditions
sorbate and benzoate (39). These evaluations should be 8.1. Packaging
done by expert microbiologists and food technologists with 8.2. Storage and shipping
knowledge of the characteristics and the mechanism of 9.0. Sample considerations
action of microbial inhibitors. 9.1. Sampling
It is not reasonable to expect that every individual food 9.2. Sample analysis for target pathogens or toxins
product would need a microbiological challenge study. 9.3. Enumeration of indigenous microbial flora
Many food products for which the assessment tables 9.4. Determination of physical parameters
indicate that product assessment is needed have a long 10.0. Duration of study and sampling intervals
history of safe use. However, safe history of a food product 11.0. Interpreting test results
is relevant only when all conditions remain the same. Even 12.0. Elements to include in the report
apparently minor changes to a food product, process, or
packaging method may have a large impact on the safety of
1.0. Obtaining expert advice and identifying a
the product. Moreover, changes in the ecology, physiology,
laboratory
or genetic makeup of a pathogen may result in food safety
issues in products with a history of safety (31, 73, 84). Challenge studies must be designed and evaluated by
an expert food microbiologist. This expertise may or may
RESPONSE TO QUESTIONS not reside within the staff of a testing laboratory. When it
does not, it is important to choose an advisor who can work
The Committee was asked by the supporting federal
with the laboratory to conduct a proper study. Potential
agencies to answer six questions. The responses are
sources of expertise include in-house experts, university
provided in order below.
faculty, testing laboratories, and independent consultants.
1. What are the appropriate criteria that must be Once a study design has been developed it may be
considered for an inoculated pack/challenge study to appropriate to consult with a statistician with applicable
determine if a food requires time/temperature experience in biological systems and have the design
control for safety (TCS)? For example, pathogen reviewed by the regulatory body or intended recipient of
144 NACMCF J. Food Prot., Vol. 73, No. 1

TABLE 2. Potential pathogens a of concern for growth studies based on interaction of product pH and aw b
pH values:

aw values ,3.9 3.9 to ,4.2 4.2–4.6 .4.6–5.0 .5.0–5.4 .5.4

,0.88 NGc NG NG NG NG NG
0.88–0.90 NG NG NG NG Staphylococcus aureus S. aureus
.0.90–0.92 NG NG NG S. aureus S. aureus L. monocytogenes
S. aureus
.0.92–0.94 NG NG L. monocytogenes Bacillus cereus B. cereus B. cereus
Salmonella Clostridium botulinum C. botulinum C. botulinum
L. monocytogenes L. monocytogenes L. monocytogenes
Salmonella Salmonella Salmonella
S. aureus S. aureus S. aureus
.0.94–0.96 NG NG L. monocytogenes B. cereus B. cereus B. cereus
Pathogenic E. coli C. botulinum C. botulinum C. botulinum
Salmonella L. monocytogenes L. monocytogenes C. perfringens
S. aureus Pathogenic E. coli Pathogenic E. coli L. monocytogenes
Salmonella Salmonella Pathogenic E. coli
S. aureus S. aureus Salmonella
Vibrio parahaemolyticus V. parahaemolyticus S. aureus
V. parahaemolyticus
.0.96 NG Salmonella Pathogenic E. coli B. cereus B. cereus B. cereus
Salmonella C. botulinum C. botulinum C. botulinum
S. aureus L. monocytogenes L. monocytogenes C. perfringens
Pathogenic E. coli Pathogenic E. coli L. monocytogenes
Salmonella Salmonella Pathogenic E. coli
S. aureus S. aureus Salmonella
V. parahaemolyticus V. parahaemolyticus S. aureus
V. vulnificus V. parahaemolyticus
V. vulnificus
a
Campylobacter spp., Shigella, and Yersinia enterocolitica do not appear here because they are typically controlled when the pathogens
listed are addressed.
b
Data are based on the PMP (106), ComBase predictor (50), ComBase database (49), or peer-reviewed publications (11, 17, 45).
c
NG, no growth; when no pathogen growth is expected, but formulation or process inactivation studies may still be needed.

the study. Suggested modifications can then be incorporated journals. Failure to properly design the study and use valid
before the study is executed. methods and appropriate controls may render the challenge
Choosing a laboratory requires careful consideration study unacceptable and require additional time and resources
because not all laboratories have the expertise to design to repeat the study. See the questions in Appendix B for
challenge studies and the quality control procedures necessary assistance in selecting a laboratory.
to produce valid results that will be accepted by the regulatory
authority or other reviewer. Laboratories may be certified by 2.0. Type of study
various organizations and state or federal agencies for various
Challenge studies are conducted for a variety of reasons.
types of testing, e.g., water and waste water testing, ISO
The specific purpose of the study drives selection of bacterial
17025, and Grade A dairy testing. However, these certifica-
strains and inoculum level, choice of parameters tested, types
tions do not necessarily qualify a laboratory to design and
of analysis, and duration of the study as described below. For
conduct microbiological challenge studies. A laboratory
example, studies evaluating growth inhibition should consider
selected for challenge testing must be able to demonstrate
bacterial species listed in Table 2, whereas the choice of
prior experience in conducting challenge studies. It is
species for lethality or survival studies depends on the
necessary to ensure that personnel are experienced and
selection of resistant strains relative to the process and
qualified (Table 1) to conduct the types of analyses needed
technology as well as compliance with regulations for specific
for the challenge studies and that these workers will follow
foods (e.g., FDA, U.S. Department of Agriculture, Food
generally accepted good laboratory practices. Laboratories
Safety and Inspection Service [USDA-FSIS], and state laws
conducting microbial challenge studies should use test
based on the Pasteurized Milk Ordinance).
methods validated for the intended use. Some examples of
generally accepted methods are available in the most recent
2.1. Growth inhibition studies
editions of the references listed in Appendix A. In situations in
which approved methods are not available or applicable, The objective of a growth study may be to request
laboratories may consider using other widely accepted exemption (variance) from TCS or other requirements
methods, such as those that have been cited in peer-reviewed defined by the Food Code, Pasteurized Milk Ordinance, or
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 145

FDA, USDA-FSIS, national, state, provincial, or local general, larger particles take longer to equilibrate. Studies to
regulations. Other objectives may be to demonstrate safety determine growth, inactivation, or survival of a pathogen
of a current formulation during extended shelf life under present due to recontamination would involve inoculation of
normal refrigerated or ambient temperatures, to determine product after equilibration.
whether formulation or processing changes are required if
3.2. Product variability
the product is subjected to temperature abuse, or to
determine the effect of a modified formulation, process, or Knowledge of the manufacturing or production vari-
packaging technology. ability is needed to determine the appropriate test param-
eters for a challenge study. Variability within and among
2.2. Inactivation studies lots should be determined by measuring formulation factors
Inactivation studies may be used to determine whether such as pH, aw, etc. The greater the variability, the more
thermal processes provide adequate log reduction of a target samples of product need to be evaluated, e.g., more
pathogen as defined by regulations or government policy measurements need to be made to determine the upper or
(e.g., FSIS requirement for a 5-log kill of Escherichia coli lower control limits. When choosing an attribute such as pH
O157:H7 in fermented dry sausage) (110). Inactivation during the challenge test, that pH (including the uncertainty
studies may also be used to determine whether nonthermal in the measurement or manufacturing capability) becomes
technologies or combinations of pH, aw, preservatives, and the upper limit of the pH specification range for the product
holding for specified times at specific temperatures prior to subsequently manufactured.
release of product will provide sufficient lethality to render a Whenever possible, food from a commercial production
food product safe (e.g., 2-year aging of raw milk Parmesan facility (manufacturing or food service kitchen or commis-
cheese or 3-day holding at room temperature to inactivate sary) or manufactured in a laboratory that has pilot food
Salmonella in mayonnaise). processing facilities should be used for the study. The food
produced in a pilot facility should be processed to mimic
2.3. Combination studies conditions used during commercial operations (cooking
Other studies involving both verification of inactivation temperature and time, homogenization, hot fill, slicing, etc.).
and evaluation of changes in the number of microorganisms The product lots used for the challenge study should be
during extended storage combine concepts from both study representative of normal production with the exception of
types above. For example, a manufacturer of processed meat necessary adjustments to acidity, moisture, salt, aw, etc., to
wishing to have a product line classified as Alternative 1 for yield the conditions most permissive to pathogen growth or
control of Listeria monocytogenes by FSIS regulation (9 survival at each formulation control limit (worst-case
CFR 430) (113) may undertake a study to demonstrate a 2- scenario based on knowledge of manufacturing variability).
log postlethality kill step of L. monocytogenes on ready-to- Percent salt and moisture may be easier than aw to measure
eat meats by high pressure processing followed by growth and control by the producer for some products such as
inhibition by product formulation during extended refriger- processed meats, cheeses, and smoked seafood and,
ated storage. A producer of a cold-filled hot sauce with therefore, may be used for control parameters in the
pH 3.5 may wish to demonstrate a 5-log kill of acid-tolerant challenge study.
Salmonella when held at 20uC (68uF) for 3 days as well as The target limits for moisture or aw will vary depending
no recovery or growth of the pathogen during ambient- on whether the objective of the study is to verify
temperature storage for 1 year. inactivation or growth inhibition. For thermal inactivation
studies, lower moisture or aw levels should be used because
3.0. Factors related to the test product pathogens may have increased heat resistance under these
conditions (10, 24, 25, 38, 102). Similarly, increased solute
3.1. Product preparation
content has been shown to protect L. monocytogenes against
The product should be prepared under conditions most high hydrostatic pressure (43, 63). In contrast, for growth
conducive to growth or survival based on the intended challenge studies, targeting the upper limit of moisture or aw
conditions of use and expected product variability. Consid- is appropriate. For example, if the typical moisture range is
eration should be given to the physical properties (pH, aw, 56 to 58%, a thermal inactivation study should be conducted
etc.) of the prepared product and the impact that these at no more than 56% moisture, but a growth challenge study
properties can have on the results of a challenge or should be conducted at no less than 58% moisture.
inactivation study. The product should be prepared so that When pH is one of the controlling factors, the food
the critical physical properties are at the appropriate should be prepared with the lowest amount of acid allowed
minimum or maximum control limits intended for the in the formulation so that the pH is at the upper range and
finished product (see section 3.2 below on product adjustment in the laboratory is not necessary. If the target
variability). pH is 4.8 but the maximum pH observed in multiple
Multicomponent products may take days to equilibrate production batches is 5.0, a growth inhibition study or an
moisture, aw, or pH. Such products should generally be inactivation study should be conducted at a pH no lower
inoculated prior to equilibration in regions of the product than 5.0. If pH adjustment is necessary and pH is adjusted
that are considered the most permissive to growth, provided upward using sodium hydroxide, the titratable acidity prior
these are areas reasonably likely to be contaminated. In to pH adjustment should be measured and reported so it can
146 NACMCF J. Food Prot., Vol. 73, No. 1

be compared with that of the adjusted food product. If the growth or inactivation by formulation. Although many
pH of the product needs to be reduced, it is important to use pathogens are listed for some pH and aw combinations, it
the same acids that are predominant in the product. may not be necessary to evaluate each pathogen for a
Acidulants exert different degrees of antimicrobial specific food, because epidemiological attribution or
activity at the same pH. For example, acetic acid is the product characteristics may narrow the choice of appropriate
most inhibitory for many microorganisms, followed by challenge organisms. For example, a seafood product might
lactic acid, with citric acid the least inhibitory (2, 3, 28, 30, be challenged with Vibrio or Salmonella because of
83). As a result, if the challenge study were conducted on a epidemiological attribution, whereas a pasteurized product
product formulated with acetic acid (vinegar), the study may in which vegetative cells of pathogens have been eliminated
not be valid for a reformulated product containing citric acid might be challenged with pathogenic sporeformers. L.
(lemon juice) even if the final pH were the same. In some monocytogenes might be used if the study is designed to
cases, the number of challenge tests can be reduced for determine growth or inactivation due to recontamination
multiple formulations having similar proximate analysis, with this organism in a ready-to-eat product.
acidity, and aw, provided the formulation most permissive to The organism used for a challenge study to determine
growth or survival is tested. inactivation due to product formulation may need to be
selected based on the resistance of the pathogen to the
3.3. Competitive microflora bactericidal properties. For example, enterohemorrhagic E.
Competitive flora can affect the outcome of a challenge coli may be selected over Salmonella or Staphylococcus
study, particularly one determining growth of pathogens in a aureus for a food with a pH of 4.3 and an aw of 0.98 because
food product. Inoculated product should contain typical it is generally considered to be more resistant to acid.
levels of competitive microflora, including starter cultures, Ideally, when conducting a study to determine pathogen
which may interfere with consistent growth of pathogens growth in a food formulation, the fastest growing patho-
during the study. The freshest product possible, i.e., within gen(s) likely to be present would be used. Predictive models
the first 10% of its shelf life, should be used. For example, if can be useful for determining which pathogen may grow
the shelf life is ,1 month, product should be used within 1 fastest under the conditions of the study. For example, if
to 3 days of production. (For purposes of this document, predictive modeling demonstrates that Salmonella grows
shelf life is defined as the time at a specified storage better at a given pH and aw combination, then this pathogen
temperature during which product quality is considered may be considered the best choice among the organisms of
acceptable for consumption. This assessment includes concern for that product for use in a challenge study.
acceptable flavor, appearance, and functionality based on Although Table 2 is similar to Table B in the Food
chemical changes or growth of spoilage microorganisms but Code (see Appendix D) and the IFT report (53), it is not
does not necessarily infer product safety by accepted identical, and some explanation is required. First, Table 2 is
definitions in all countries.) Care should be taken during more extensive than Table B and includes both higher and
the inoculation step to not introduce atypical spoilage lower pH values and more defined categories for higher aw
microorganisms that may inhibit pathogen growth. In rare values. Second, the IFT report (53) and the Food Code (117)
cases, naturally occurring bacteria can enhance growth or are specifically focused on foods that require temperature
survival of pathogens, potentially reducing the safety of the control for safety, whereas the focus of this Table 2 is
product (19). broader. Finally, Table 2 considers time scales that may be
considerably longer than those typically of concern in retail
4.0. Target organism(s) food safety. The table should not be interpreted to suggest
that a food falling within a particular pH and aw range needs
4.1. Identifying the pathogen(s) of concern
to be challenged with a pathogen, e.g., that high aw foods
An expert food microbiologist should determine the with a pH of 3.9 need to be challenged with Salmonella.
appropriate organisms for challenge testing. There are a Although Salmonella has been shown to grow at pH values
number of issues the microbiologist must consider, as low as 3.9, these studies have been done in laboratory
including the specific product, the process used to prepare media under conditions ideal for growth, other than the pH
it, and any pathogens that are epidemiologically or value. In foods, many factors interact to support or inhibit
ecologically relevant. There are a number of resources pathogen growth. An expert microbiologist should use
available to assist in determining appropriate pathogen(s) Table 2 as a guideline to assess whether a challenge study
for a given food. Examples of assessments of the on a particular food with a specific pathogen is warranted.
appropriate challenge organism for specific food products Table 2 is useful for identifying appropriate pathogens
can be found in the IFT report on evaluation and definition of concern for particular pH and aw combinations. However,
of potentially hazardous foods (53) (specifically, see Table this information should not typically be used for the
1, Table A, Table B, Table 4-1, and Table 6-1 in that report). selection of organisms in process inactivation (e.g., thermal
For easy reference, please refer to Appendix C. inactivation) studies. The choice of organism for these types
Table 2 provides combinations of pH and aw values of studies should be based on the likelihood of pathogen
that may allow growth of pathogens of concern based on association with the specific food and pathogen resistance to
model predictions and published literature. This table may inactivation, as well as the public health objective of the
be useful in selecting organisms for use in studies to assess process and the intended use of the product. For example,
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 147

nonproteolytic strains of Clostridium botulinum might be pathogen, food, and treatment being evaluated should be
selected as the appropriate target organism for some incorporated into the final report. If no directly relevant
refrigerated foods and L. monocytogenes for others, published comparison data are available, studies need to be
depending on how likely it is that nonproteolytic C. conducted to establish the validity of using a particular
botulinum will be present, how long the product will be surrogate-pathogen-process combination.
held refrigerated, whether the product is ready-to-eat or will
be cooked prior to consumption, and other factors. 4.3. Type and number of strains
In order to account for variations in growth and survival
4.2. Use of surrogate organisms
among strains, challenge studies should generally be
Inoculation of foods with bacterial pathogens requires conducted using three to five bacterial strains either
adequate biological containment facilities and may require individually or in combination (53, 75, 91). Where there
governmental approval in the case of certain pathogens such is considerable variability among strains or there is little
as C. botulinum. Therefore, in limited cases, nonpathogenic known about the growth of the organism in a particular food
surrogate organisms are especially useful for testing product, as many as 10 strains may be used (e.g., some C.
specialized processing equipment in the plant, where the botulinum or L. monocytogenes studies).
introduction of the pathogen would pose an unacceptable Generally, using an inoculum composed of multiple
risk. Surrogates may also be useful for selecting the study strains (i.e., a cocktail) of a given pathogen is preferred
parameters before conducting the full study with the because it will help to encompass the variability among
pathogen. Care should be taken when using surrogates for organisms and may reduce the number of required tests.
in-plant challenge studies because these organisms may Prior to use in the study, the strains selected should be
have adverse sanitary or regulatory implications should they screened for antagonism that can be caused by production of
survive and contaminate the plant environment. bacteriocins or other antimicrobial factors (53). Another
Surrogates are typically nonpathogenic proxies for the approach is to screen several strains in the food matrix under
pathogen of concern that have similar or more robust investigation and determine which strain has the greatest
survival capabilities under the conditions being studied. resistance, grows fastest, etc., and then to conduct the
Such proxies may include avirulent strains of pathogens, challenge studies using that single strain (12, 91). Screening
where appropriate. The ideal surrogate should have the parameters depend on the purpose of the challenge study,
following characteristics: nonpathogenic, inactivation char- e.g., to determine inactivation or growth characteristics in a
acteristics and kinetics that can be used to predict those of product. However, there are strains with atypical resistance,
the target pathogen, similar susceptibility to injury, e.g., the extremely high moist heat resistance of Salmonella
reproducible growth, easy preparation of high-density Senftenberg 775W (79). These strains may not be
populations that are stable until used, easily enumerated appropriate for use in some studies because they are not
and differentiated, similar attachment capabilities, and representative of strains reasonably expected to be present in
genetically stable (52). the applicable foods. Whether to use an individual strain or
Clostridium sporogenes PA3679 has proven to be an cocktails of strains should be determined by an expert
excellent surrogate for C. botulinum when used in inoculated microbiologist knowledgeable in food microbiology and
pack studies to validate thermal processes for low-acid canned pathogen control.
foods. In certain cases, C. sporogenes may be suitable for Strains carrying markers such as antibiotic resistance or
reducing the number of formulations to be verified using C. green fluorescent protein may be useful for confirming that
botulinum because these organisms are culturally similar. the organisms recovered are the test organisms. When such
Formulations that support growth of C. sporogenes can be strains are used it is important to determine that they possess
excluded from further validation studies with C. botulinum. the same characteristics as the parent strain without the
However, C. sporogenes cannot be used as a direct substitute marker with respect to factors critical to the challenge study.
to validate a product for inhibition of botulinum toxin Furthermore, carriage of the resistance marker should be
production (64). Other examples of surrogate-pathogen pairs verified to be stable under stressful conditions that may be
include Listeria innocua and L. monocytogenes (99) and encountered during the challenge study.
nonpathogenic E. coli and E. coli O157:H7 (26). Isolates should be appropriate for the food product
A surrogate that works well to predict the target being challenged (53, 80, 91). This approach includes using
response for one type of process may not be an appropriate isolates from the food type, the food processing environ-
surrogate in a different type of process. For example, the ment, and clinical specimens, as appropriate. Inactivation
heat resistance of various strains of C. botulinum spores did studies should use strains that demonstrate tolerance to the
not correlate with their resistance to high hydrostatic specific process for the product under consideration, such as
pressure (71), so although C. sporogenes may be the heat or high pressure processing (16, 24, 25, 71).
preferred surrogate for C. botulinum for canning processes, Biochemical characteristics, serology, genetic profile, viru-
another organism, such as Bacillus amyloliquefaciens may lence, or toxicity should be periodically reconfirmed as
be appropriate as a surrogate for C. botulinum for high appropriate. The test strains for growth challenge studies
hydrostatic pressure studies (71, 86). should demonstrate robust growth in laboratory media or a
The choice of the surrogate needs to be justified, and similar food without inhibitors under the conditions of the
supporting documentation for its appropriate use for the study (e.g., temperature and atmosphere).
148 NACMCF J. Food Prot., Vol. 73, No. 1

5.0. Inoculum levels UV light to achieve a 5-log reduction of E. coli O157:H7 in

The inoculum level used in the challenge study depends apple cider, or to determine inactivation of a pathogen over
on whether the objective of the study is to determine growth time, e.g., the effect of preservatives on pathogen
or inactivation of a pathogen. It may be desirable to conduct inactivation during storage of a food product. In the former
case, relatively high inoculum levels are generally used, as
challenge studies using multiple inoculum levels to
noted above. However, in the latter case, lower inoculum
determine the margin of safety in the process/formulation
levels consistent with expected pathogen contamination
(91).
levels might be used because preservatives would generally
5.1. Growth studies not be expected to inactivate large numbers of pathogens,
depending upon the pH and other conditions. Studies might
When conducting studies to determine whether a
also be important to determine survival or inactivation of a
pathogen grows in a product, ideally the number of organisms
pathogen in a product that is recontaminated.
used should reflect the numbers normally expected in the
Initial inoculum levels may affect the rate of die-off in
product. Typically, an inoculum level of between 2 and 3 log
some foods (32, 95, 125), and this phenomenon needs to be
CFU/g is used, even when this exceeds expected numbers,
taken into consideration.
because this level allows enumeration by direct plating (53,
91). Lower levels may be used if documentation of low levels 6.0. Inoculum preparation
of natural contamination exists, because these lower levels
Ideally, isolates from foods should be stored in a
will more accurately represent the product’s ability to support
manner to preserve the strain characteristics with respect to
growth (91). When very low seeded populations (e.g., less
survival, growth, resistance, etc. (e.g., frozen in glycerol or
than 100 cells per sampling unit) are most appropriate,
freeze-dried). When reviving strains from the frozen or
consistent inoculation among individual samples may be
lyophilized state, there should be one to two successive
difficult to achieve. Calculating the level of organisms in the
transfers in a nonselective growth medium. Working
product from the initial inoculum suspension, increasing
cultures, e.g., refrigerated slants, may be prepared and used
sample size (e.g., from 25 to 250 g) and the number of
for a period of time (e.g., 7 to 30 days). The number of times
replicate samples analyzed (e.g., from three to six samples),
a culture is transferred to produce new working stock
and/or using enumeration methods such as the most probable
cultures should be minimized to avoid genetic changes that
number (MPN) method will increase confidence in the
affect the phenotypic properties of the organism (91).
number of organisms in the inoculum.
AOAC International guidelines for laboratories (8) indicate
The inoculum level may affect the apparent efficacy of there should be not more than five passages from the
an antimicrobial or formulation combination to inhibit primary reference material. In some instances even fewer
microbial growth. If the inoculum populations are too high, transfers may be appropriate, because organisms may
the factors inhibiting growth may be overwhelmed by the readily lose extrachromosomal elements such as plasmids
inappropriate inoculum level, leading to the incorrect or other genetic markers and phage.
conclusion that the formulation does not inhibit growth For challenge studies using vegetative cells, stationary
(53, 80). In the case of sporeformers, germination and time phase cells (18 to 24 h) grown on nonselective media under
to observable growth or toxin production may be signifi- conditions suitable for optimal growth of the specific
cantly reduced if high initial spore loads are used (69, 126). challenge culture should generally be used (53). However,
In contrast, a high inoculum level of vegetative cells (e.g., 5 in certain instances it may be desirable to precondition or
to 7 log CFU/g) in a growth study may also mimic the adapt the culture to specific conditions that may be
population nearing stationary phase, which may result in an applicable to the specific characteristics of the food product.
apparent no-growth or low-growth observation. For example, for low pH foods it may be appropriate to acid
5.2. Inactivation studies adapt cultures (34, 46, 65, 66), which can often be
accomplished by growing the culture in tryptic soy broth
When conducting inactivation studies, high numbers of with 1% glucose (14, 27). Cold adaptation at 7 to 8uC (44.6
organisms are typically used, e.g., 6 to 7 log CFU/g (53, 91), to 46.4uF) for 7 days may reduce the lag phase for
in order to quantify survivors and/or to document high pathogens (121), which may be important for assessing the
levels of inactivation. The target level of reduction, which shelf life of refrigerated ready-to-eat products. Cold
influences the inoculum level used, may depend on adaptation may be more important for challenge tests of
regulations for specific food types, e.g., a 5-log reduction foods with a short refrigerated shelf life, e.g., less than 7
of the appropriate pathogen in juice (21 CFR 120.24) (120), days. Care should be taken to avoid habituation procedures
4-log reduction for treatment of almonds (7 CFR 981) (105) that cause cells to be more sensitive to the adverse
to inactivate Salmonella, and a 7-log reduction of environment, e.g., simultaneous adaptation to cold and acid
Salmonella in poultry products (9 CFR 381.150) (112). conditions (95) or acid stressing cells prior to a heat
Laboratories conducting inactivation studies in products that treatment (87).
are subject to regulations should be aware of the most For inactivation studies, cells that are grown at higher
current requirements. than optimum temperatures may become more resistant to
Inactivation studies may be conducted to assess the heat than are cells grown at optimum temperatures (79, 96).
lethality delivered by a specific process, e.g., the ability of Increased heat resistance can also be observed with brief
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 149

exposure to sublethal temperatures (heat shock) (15, 94, 7.0. Method of inoculation
123). For either inactivation or growth studies, adaptation of Inoculation procedures for challenge studies are
cells should attempt to mimic the likely physiological state described in the IFT report (53). As that report notes,
of the organism at the time it contaminates the food. several critical considerations for the delivery of the
Prior to use, cells should be washed (e.g., in buffer or inoculum to the product include maintaining the intrinsic
carrier medium) to remove nutrients or metabolites in the or extrinsic characteristics of the product; simulating
spent medium that could have an impact on growth in the contamination that could realistically occur under manufac-
test product. Cells should then be suspended in a carrier
turing or storage conditions; and ensuring that, where
(buffer or homogenized portion of the food) to inoculate the
appropriate, each of the unique interfaces of the product
food.
components receive the inoculum.
Composites containing multiple strains should have
Two factors important to maintaining the intrinsic
approximately equal numbers of the individual strains. This
characteristics of the challenged product are minimizing
can be accomplished by previous experience enumerating
inoculum volume and matching the critical factors of the
the strains under specific growth conditions or by using
food, such as pH and aw. Typically the inoculum volume
turbidity measurements (e.g., optical density or McFarland
should be no more than 1% of the volume of the food and
standards).
when possible, less. Some methods that have been used to
Spores of pathogens such as C. botulinum, Clostridium
minimize the inoculum volume include growing the
perfringens, and Bacillus cereus can be prepared, washed,
pathogen to high populations and concentrating by
and suspended in sterile water and frozen, preferably at
centrifugation or growing the pathogen on a solid growth
220uC (24uF) or below. As with vegetative cells,
medium and then harvesting a paste for use as the inoculum.
composites should contain approximately equal numbers
When challenging food products with reduced aw or pH, the
of each strain. Spore suspensions can be enumerated to
aw or pH of the diluent can be adjusted using a humectant or
determine the number of spores and then appropriate
acidulant similar to that contained in the food (53).
volumes can be combined to prepare the inoculum.
However, preliminary analysis should verify that modified
Spore inocula are often heat shocked prior to use,
pH or aw of the buffer does not adversely affect viability of
unless they are inoculated into the product immediately
prior to heating or processing. The decision on whether or the pathogens.
not to heat shock a spore inoculum will depend on the An important extrinsic factor is the package atmosphere
expected state of the naturally occurring spores in the food (see section 8.1. below: storage conditions, packaging).
product and the conditions of use of that product. For Ideally, product should be first inoculated and then
example, spores would not be heat shocked if the packaged under an appropriate atmosphere that closely
challenge study is being conducted in a raw commodity duplicates the packaging system to be used during
that will not be heated (e.g., raw reduced-oxygen- commercial production. Alternatively, a common practice
packaged fish). When it is desirable to have a mixture of is to use a needle to inoculate through the packaging using
vegetative cells and spores, the suspension should not be some type of self-sealing rubber or silicon septum. Two
heat shocked. disadvantages of using the latter type of inoculation method
It is important to verify the numbers of viable are long-term package integrity and inoculum distribution.
organisms in the inoculum used. In addition to enumerating Also, when inoculating with a needle, culture should be
the inoculum suspension itself, the inoculated food should distributed over as large an area as possible to reduce the
be enumerated to obtain a zero-time count. If the inoculum concentration of cells, moisture, and/or nutrients in limited
level is low, an increased number of replicates of the areas. Package atmosphere (e.g., oxygen and carbon dioxide
inoculum and/or product may be necessary. Rapid and in the headspace) should be monitored during the duration
significant reductions in microbial populations are frequent- of the study to assess the integrity of the package and to
ly observed when the food includes bactericidal ingredients ensure that the effect of changes in gas composition is
such as nisin or other commercial fermentation by-products considered.
used for shelf-life extension. For example, a 0.5- to 2.5-log In general, the method of inoculation should place the
reduction in L. monocytogenes was observed immediately inoculum on or within the product in a manner that
after inoculation in fresh soft cheese and in bologna and realistically simulates potential contamination that might
ham containing lactic acid bacteria (LAB) fermentate or occur during manufacture, preparation, shipment, or display
nisin (35, 37). of the product. Liquid foods are inoculated by mixing the
A dry inoculum may be required for studies in low- inoculum throughout the product with agitation. In solid
moisture foods or when added moisture needs to be avoided. foods, the inoculum may be mixed throughout a ground
Inoculum can be prepared by freeze-drying (53, 80) or dried product or applied on the surface by dipping, aerosolizing,
on a product similar to the challenge food (53). For or spreading on the entire surface or on selected spots.
preparation of a dehydrated inoculum, the organism may Dipping the product in a liquid inoculum or using an
require several days to months to stabilize (e.g., Salmonella aerosolized inoculum will allow organisms to be spread
in skim milk powder) (59). As a result, viable populations of over the entire surface of the product, including cracks and
the stabilized dried inoculum should be determined prior to crevices. However, if an aerosolized inoculum is used,
use. inoculation should be conducted in a biological safety
150 NACMCF J. Food Prot., Vol. 73, No. 1

cabinet to protect employees from the challenge organism. 8.2. Storage and shipping
Preliminary studies should be conducted to standardize the
Storage temperatures used in the challenge study should
amount of inoculum that contacts the product.
be representative of the expected temperature range that the
Many challenge products have multiple components or product will be exposed to during commercial distribution
layers. If contamination during assembly is possible, the and storage. For refrigerated foods, NACMCF recommends
challenge inoculum should be applied to the various layers that the studies be conducted at 7uC (44.6uF) to account for
or components. Unique growth conditions can exist at the expected consumer storage temperature in the United States
interfaces between components, such as the microenviron- (75). Refrigeration studies may incorporate additional
ment between a pie crust and a pie filling. This area might temperatures (e.g., 4 to 6 or 10 to 12uC [39.2 to 42.8 or
have a unique combination of factors that will allow growth, 50 to 53.6uF]) when a better understanding of the behavior
so these areas should receive a portion of the inoculum. For of the challenge organism is desired, such as with some
this reason, the food should not simply be homogenized and antimicrobial compounds whose inhibition of microbial
inoculated. Other conditions of the microenvironment growth is temperature dependent (21, 91).
should also be considered, such as fat-water emulsions, Temperature changes may be incorporated into a
microdroplets, or partitioning. challenge study protocol if, for example, a manufacturer
Inoculating a large batch prior to packaging or distributes a refrigerated product under well-controlled
inoculating individual samples can be valid depending on conditions for a portion of its shelf life, after which the
the likely route of contamination, packaging considerations, product may be subjected to elevated temperatures imme-
and practicality. Inoculating a single batch of product will diately prior to and during use (53). For shelf-stable
minimize the variability of the starting concentration as well products, typical temperatures range from 24 to 35uC
as create a less heterogeneous distribution of the pathogen if (75.2 to 95uF) depending on expected storage room
the food can be mixed without destroying the product temperatures (21). Humidity should also be considered as
integrity. This issue is particularly critical in growth or a factor in storage conditions; for those products where the
inactivation studies in which documentation is needed to moisture content can change in response to ambient
meet a specific regulatory requirement (e.g., no more than a humidity conditions, the challenge study should be designed
1-log increase as evidence of growth inhibition of L. to incorporate representative environmental humidity vari-
monocytogenes in a deli salad or a 5-log reduction of E. coli ation (80).
O157:H7 in juice). Dividing a large inoculated batch into It is necessary to ensure that appropriate storage space
discrete portions for testing at each sampling interval is available and that proper temperatures are maintained and
reduces the risk of contamination caused by repeatedly recorded throughout the study. Temperatures during storage
resampling a large batch. Inoculating individual samples and transportation of commercially made products to the
may be more appropriate for studies representing postpro- laboratory should be monitored with continuous tempera-
cess contamination by contact (e.g., cooked frankfurters or ture recorders, data loggers, or periodic manual temperature
slices of cheese made with pasteurized milk) or when verification. Samples inoculated with pathogens should be
production cannot be readily replicated in the laboratory segregated and clearly labeled to prevent inadvertent human
(e.g., filled pastries or individual packages with unique consumption.
atmosphere and packaging materials). Inoculation methods
that result in highly variable inoculum levels or uneven 9.0. Sample considerations
distribution require a greater number of samples at each 9.1. Sampling
sampling interval and potentially additional replicate
batches to be analyzed. Sampling schemes for food microbiology experiments
are often dictated by common practice not solely based on
8.0. Storage conditions statistical design. The suggestions below reflect this
convention. The number of samples to be analyzed initially
8.1. Packaging
and at each time interval during processing and/or storage
Product packaging for the challenge study should be should be at a minimum two; however, analysis of three or
representative of typical commercial production. If the more samples is preferred. Replicates should be independent
commercial product is to be packaged under vacuum or a trials using different batches of product and inoculum to
modified atmosphere, the challenge study sample should be account for variations in product, inoculum, and other
packaged under the same conditions, including the use of factors. Generally, the number of samples and replicates
the exact gas mix used for modified atmosphere packaging, should be increased in situations of higher variability or
use of packaging material of the same gas permeability, and uncertainty. When the number of samples analyzed at each
use of similar vacuum levels for vacuum-packaged product. time interval is only two, it is better for the study to be
Specific modified atmospheres or vacuum packaging may repeated (replicated) more than two times. In studies with
be inhibitory to some microorganisms but may stimulate three or more samples tested at each time interval, two
growth or toxin production by other microorganisms (53). replicates are usually adequate. When analyzing samples for
Care should be taken to ensure that headspace volume and botulinum toxin it is appropriate to select a greater number
gas composition of the challenge study samples mimics the of samples (e.g., five or more) per time point because of the
commercial food product as closely as possible. potential variability in toxin production among samples
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 151

(74). For end-point lethality determination, 5 to 10 samples Butterfield’s phosphate buffer or buffered peptone water for
per time interval may be appropriate. If supporting data vegetative pathogens or spores. However, if the product has
from other studies exist, the need for replication may be a high salt or sugar content it may be necessary to modify
reduced (91). Appropriate statistical experimental design the dilution buffer to avoid shocking the cells. Enrichment
can improve the validity of the study. There are quantitative procedures for the target pathogen should be considered at
methods for assessing the statistical quality of a study, e.g., time points when levels of survivors are expected, or
power analysis. The study design may benefit from previously determined, to be below the experimental limit of
consultation with a statistician familiar with food microbi- detection by direct plating. Rapid detection methods that
ology studies. have been validated (see Appendix A) are appropriate when
The sample preparation method should be selected enumeration is not necessary.
based on the type and properties of the food and the method Sample analysis must be done using methods that
of inoculation, which depends on the food product and the permit the accurate and reproducible recovery of microor-
inoculation procedure (53, 91). In cases of solid foods ganisms. In all cases the amount of buffer or diluent used
inoculated on their surface and in products where the must be defined and constant among samples, and this level
contamination is expected to be localized on their surface, should be selected based on sample size, level of
samples may be swabbed or sponged, washed or rinsed, contamination expected, and minimum level of detection
and/or agitated in a liquid buffer or diluent of known desired. The sample preparation protocol and washing,
volume. After thorough mixing, the rinsate is analyzed by rinsing, or blending time should be consistent, and the time
direct plating of appropriate dilutions onto appropriate between sample processing and plating should be short and
culture media (see section 9.2 below). The results can be constant for all samples. Sample preparation temperature
expressed per unit of surface area or per sample, especially and time and conditions and variables involved in sample
for items of irregular conformation. For example, surface- preparation should be maintained constant to the extent
inoculated frankfurters may be prepared for detection of L. possible; these variables include volume or weight, surface
monocytogenes as whole links washed or rinsed with area, composition, and properties (e.g., pH) (12).
diluents, and the results may be expressed per unit surface For growth studies, pathogens should be enumerated on
area or whole link, if the links are of uniform size. appropriate selective agar (see Appendix A). Inactivation
Alternatively, surface samples may be excised and studies may result in injured cells, and direct plating of these
homogenized in diluent. The results may be expressed per cells onto selective agar can overestimate the extent of
unit of surface area or per gram. For example, a spot- death. In such cases, samples should be prepared and tested
inoculated leafy green may be sampled by cutting a surface in ways that allow repair and recovery of injured organisms.
area surrounding and greater than that inoculated and the Recovery of injured cells can be enhanced by using
sample can be homogenized or macerated to release nonselective media such as tryptic soy agar (TSA) or plate
bacterial cells. Some foods, e.g., surface-inoculated whole count agar overlaid with selective agar after 2 to 4 h of
tomatoes or melons, may be sampled with a sterile cork incubation at optimum temperature (20, 40), by using
borer, extracting a defined section from an area of the selective agar overlaid with nonselective agar (124), by
surface that was inoculated or treated. using agar underlay techniques (60, 61), or by replica
Caution should be exercised when considering analysis plating from a nonselective agar such as TSA to a selective
of composited samples in challenge studies. Compositing agar (100). Standard methods for extraction of C. botulinum
multiple samples for pathogen enumeration eliminates neurotoxins and S. aureus enterotoxins from foods can be
detection of variability among discrete samples and may found in the references provided in Appendix A.
reduce sensitivity of the analysis. Furthermore, composited
samples may dilute toxins to less than detectable levels if 9.3. Enumeration of indigenous microbial flora
these toxins are present in only one of the multiple samples. In addition to inoculated product, sometimes it is also
However, compositing samples before or pooling samples useful to test corresponding uninoculated control samples to
after an enrichment procedure may be appropriate to determine levels of background microflora surviving the
confirm absence of survivors in an inactivation study. process or changes that occur during product shelf life (53,
Pooling after enrichment can be used as a screening 91). Moreover, protocols for challenge studies to determine
procedure that will later allow determination of how many growth inhibition or inactivation based on product formu-
original samples were positive. Compositing or pooling lation should consider and address potential effects of
approaches must be validated to assure sensitivity is not lost. naturally occurring microflora on the pathogens of concern.
In addition, spoilage and the end of shelf life are usually
9.2. Sample analysis for target pathogens or toxins
associated with an increase in microbial populations. Thus it
The objective of sample preparation for microbial is recommended that microbiological numbers such as
analysis is to retrieve all microbial spores or cells of interest aerobic plate counts and number of spoilage organisms
(or toxin, where appropriate). Sample preparation should typical for the product (e.g., LAB or yeasts and molds) be
provide conditions that will allow the metabolic activity to obtained. Testing for these or other indicator microorgan-
lead to detectable colonies or other measurements indicating isms cannot substitute for pathogen testing. In addition, the
activity and allow a measurement of survival or growth presence or absence of spoilage bacteria cannot be used as
levels. It is common to use a 1:10 initial dilution in an indicator of safety.
152 NACMCF J. Food Prot., Vol. 73, No. 1

LAB are expected in fermented or cultured food year or longer. Ideally, products should be held for some
products at relatively high populations (e.g., 6 log CFU/ period beyond the end of the intended shelf life to account
g), but indigenous populations are low in most processed for users who might consume the product past the end of the
foods. This group of bacteria is known to compete well with declared shelf life and to add an additional margin of safety
low levels of pathogens for nutrients, can grow over a wide (53). Depending on the shelf life of the product, this time
range of temperatures, and can reduce the pH of the food may be 25% (e.g., for products with shelf life of 3 to 6
through acid production, and some strains can produce months) to 50% (e.g., for products with shelf life of 7 to 10
bacteriocins that may inhibit some pathogens. Relying on days) longer than the intended shelf life of the food (53, 91).
the presence of naturally occurring background levels of This additional time may be important for recovery of cells
LAB in foods is an unreliable method to control pathogens. injured by heat or by antimicrobials in the product. For
Conversely, competitive microflora may inhibit growth of some products that still have acceptable sensory properties
specific pathogens, and failure to account for this interaction at the end of the intended shelf life, it may be important to
could lead to erroneous conclusions. Thus, it may be continue studies until overt spoilage occurs, as consumers
important in some circumstances to monitor LAB growth may consume the product as long as it does not appear
during the challenge study to determine if competition may spoiled. Samples held under abuse conditions are unlikely to
contribute to inhibition of pathogens during the trial. last the full shelf life and are usually sampled for shorter
Although they may be present, molds and yeasts may time periods (53). Samples, including controls, should be
not be initially visible on the food. Deamination of food analyzed initially after inoculation (in some cases, after a
proteins by molds can produce ammonia and a localized short equilibration period) and then five to seven times over
increase in pH that can increase the potential for pathogen the duration of the study (53). For long-shelf-life products, it
growth in that microenvironment (81). Populations of molds may be necessary to have more than seven sampling points.
and yeasts can be enumerated by using a variety of selective The sampling interval should be determined based on
plating media or by other validated procedures. prior experience with similar products and in consideration
of the likely duration of survival or rate of growth or
9.4. Determination of physical parameters
inactivation. Depending on the product characteristics and
Food properties such as proximate composition (pro- expected outcomes for products with a long shelf life, it may
tein, fat, and moisture), pH, titratable acidity, aw, salt be appropriate to test on a more frequent basis early in the
content, and residual nitrite can influence the behavior of study (e.g., daily) and at longer intervals later in the study
pathogens. It may be important to measure these factors as (53).
part of the challenge study. Some parameters, such as pH, A growth inhibition study may be ended when there is
that may change during the study may need to be monitored greater than a 1- to 2-log increase in pathogen growth or
at appropriate points throughout the study in parallel with when toxin is detected in samples for two consecutive
microbial analysis. Sources of appropriate methods can be sampling intervals (indicating growth of the pathogen of
found in Appendix A. The number of samples to be concern) or there is gross spoilage such that the product is
analyzed is described in section 9.1 above. no longer fit for consumption. Care should be taken in
Changes in pH can be an indicator of microbial making this determination, because spoilage and apparent
metabolism when microbial populations are not enumerated edibility are subjective.
or if growth is not significant. The pH of foods that are When measuring pathogen inactivation, the study is
homogeneous and likely have consistent pH throughout the typically concluded when the pathogen is no longer
matrix can be measured on a representative sample. In recovered from the product. However, in some cases (e.g.,
contrast, complex foods consisting of multiple discrete thermal death time studies) it may be important to take into
components or ingredients may require multiple pH account the possibility of injured cells and to continue
measurements. For example, a sandwich may require incubation of samples until the end of product shelf life to
measuring the surface or interface pH of the components verify that injured cells do not recover and grow (91) or
in addition to a homogenized sample. produce toxin in the product over time. Alternatively,
For obvious safety reasons, no sensory assessment attempts to recover the pathogen in noninhibitory enrich-
other than changes in appearance (phase separation, ment media after a period of incubation in the product may
turbidity, texture, and gas formation) should be performed be used to verify the absence of survivors.
on challenge test samples. In some instances, the investi-
gator should make a judgment if the product would be 11.0. Interpreting test results
considered ‘‘edible’’ based on visual and olfactory obser- Interpreting the results of microbiological growth and
vations. Note that because pathogens or toxins may be inactivation studies requires evaluation by expert microbi-
present, olfactory observations may constitute unacceptable ologists who will consider all relevant factors (53, 80, 91).
risk to the laboratory worker. In determining whether a product supports growth of a
pathogen, it is rarely as simple as comparing final and initial
10.0. Duration of study and sampling intervals counts. Numbers from different sampling points may vary
Challenge studies should be conducted for at least the due to inherent variation in sampling and enumeration
intended shelf life of the product (21, 53, 122). For some procedures, particularly when foods contain antimicrobial
shelf-stable products this may mean holding products for a compounds that limit growth. It may be difficult to
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 153

determine if changes in numbers are real or due to analytical Where studies have been conducted with C. botulinum,
variability. In addition, there may be an initial die-off in detection of toxins is measured rather than growth, as toxin
some foods following inoculation, and if this die-off is can be produced without an increase in number (47). No
followed by growth that does not exceed the target inoculum toxin should be detected in the product over the duration of
level, this growth may not be recognized. This issue may be the challenge study (53). In lieu of testing for Staphylococ-
addressed by allowing a brief time (e.g., 2 h) for the cus enterotoxins, limiting growth of S. aureus to less than 3
inoculum in the product to equilibrate prior to conducting log CFU/g may be used (53). This limiting growth level was
the initial count (53). Normal sample variation may result in based on the assumption that the initial population does not
a spike at a sampling interval that may not be significant exceed 3 log CFU/g and that a minimum of 6 log CFU/g is
(122); this variation can often be addressed through testing needed to produce staphylococcal enterotoxins.
of multiple samples. Graphical representation of the data to Where multiple formulations have been challenged,
examine trends may be useful in assessing whether actual growth or toxin production in one formulation but not in
growth has occurred (53). This is particularly important in another may provide useful data on the inhibitory properties
cases where the data set contains one or more outlying data of the product with respect to pathogen growth. In this case,
points. The interpretation of inconsistent or highly variable the effect of formulation differences will help to identify
results is an important and complicated issue and should be critical factors necessary to control pathogen growth or
done by an expert microbiologist (see Table 1). toxin production. Similarly, if a product is produced by a
An increase in 1 log cycle over two or more time manufacturing process that encompasses the point of
intervals is generally considered significant by food ‘‘failure,’’ this is an indication that the manufacturing
microbiologists (122). Smaller increases may be significant variability may be too great to assure the safety of a product
depending upon the enumeration methods, number of formulated in this manner.
samples and replicates used, and the variability among data For lethality experiments, log reductions should be
points. Thus, in determining that a product does not support determined in replicate trials. The log reduction should meet
growth of a pathogen, in general less than a 1-log increase any existing regulatory performance standards that apply to
above the initial inoculum level throughout the intended the food product. Where no performance standard exists, the
shelf life of the product and across replicate trials would be lowest log reduction achieved should exceed the expected
an appropriate acceptance criterion (53, 91). This level contamination level by an amount that incorporates a margin
reflects the inherent variation that exists with enumeration of safety (a 2-log margin is often used) consistent with the
of microorganisms (53, 103). variability expected in the product and the process (91).
Statistical methods can also be used to determine The discussion above indicates that universally accept-
whether differences in counts at specific sampling points able rules for interpreting test results are not available and
indicate true growth or are simply due to sampling and points out the need for further consideration to produce clear
measurement errors. Where the repeatability and reproduc- guidance on this subject.
ibility of the enumeration method have been determined
through validation studies and the standard deviation of 12.0. Elements to include in the report
reproducibility can be calculated, a more precise determi- In order for others to assess the adequacy of a
nation of a significant difference may be made. For challenge study, it is imperative that the study report
example, Agence Française de Sécurité Santitaire des provide appropriate information, including an interpreta-
Aliments (AFSSA) (13) recommends a 0.5-log CFU/g tion of the results. The report should begin with an
increase between initial and final concentrations as introduction that includes the purpose of the study and
indicating that growth of L. monocytogenes has occurred. reviews the data supporting the experimental design. The
This value is based on an estimation of measurement report should include information characterizing the
uncertainty (55, 57), which is determined by doubling the product and process. The materials and methods should
reproducibility standard deviation. It should be noted
be described as they would be in a scientific publication.
however that the reproducibility standard deviation can
The results should include both raw and summarized data
vary. Scotter et al. (92) conducted tests to validate the
and should be clearly presented. Any statistical design and
International Organization for Standardization (ISO) method
analysis of results should be thoroughly described. If
for enumeration of L. monocytogenes in foods and found
statistical analysis was not used, that fact should be clearly
that the reproducibility standard deviation ranged from 0.17
stated and justified. A discussion should provide an
to 0.45 log CFU/g, depending on food product and level of
interpretation of the results and any limitations on the
contamination. Thus, depending on the food, inoculum
applicability of the data. The conclusions should contain
level, and method of enumeration, a difference greater than
any recommendations and should indicate the types of
0.5 log CFU/g may (or may not) be an appropriate criterion.
changes in product formulation or processing that could
It should also be noted that statistically significant
warrant a new challenge study.
differences may not always be biologically relevant. An
expert microbiologist, using available data and past 2. What are the appropriate uses of mathematical
experience, can best determine if the data represent a trend growth and inactivation models? Under what
of increasing numbers or are simply a product of the conditions can these models be used as a substitute
variation seen in enumeration studies (91). for inoculated pack/challenge studies? Of the models
154 NACMCF J. Food Prot., Vol. 73, No. 1

TABLE 3. Examples of mathematical growth and inactivation models and their applicability to different foods
Model name Reference Applicability

American Meat Institute 4 The model provides meat processors with a science-based validation tool that can be
Foundation Process Lethality used to demonstrate the effectiveness of a specific heat process to destroy
Determination Spreadsheet microorganisms of concern.
ComBase Predictor 50 ComBase Predictor models are based on observations made in culture media and
comprise a set of 20 growth models, 7 thermal death models, and 2 nonthermal
survival models. Temperature, pH, and aw (usually as a function of NaCl) are the core
factors, but for some organisms, the effect of a fourth factor, such as CO2, nitrite, etc.,
is also featured.
Isothermal-Based Prediction Tool 104 The software can be used to predict whether Salmonella, E. coli O157:H7, or S. aureus
(IBPT) will grow to a ‘‘level of concern’’ in raw beef and pork products.
Microbial Responses Viewer 78 The MRV is a new database consisting of microbial growth–no growth data derived
(MRV) for ComBase (version from ComBase. The software allows the user to rapidly view growth–no growth
beta 1) contour plots superimposed with actual ComBase data. Contours of any two of three
variables (temperature, pH, and aw) can be visualized while the third is held constant.
Opti.Form Listeria Control Model 85 The model predicts Listeria growth for both uncured and cured cooked meat products.
2007 The model will help to calculate the level of lactate and diacetate needed to control
Listeria in cured and uncured cooked meat and poultry products for their required
shelf life.
Pathogen Modeling Program 106 This predictive microbiology application was designed as a research and instructional
(PMP) tool for estimating the effects of multiple variables on the growth, inactivation, or
survival of foodborne pathogens. Most of the models are based on experimental data
of microbial behavior in liquid microbiological media.
Perfringens Predictor 51 Perfringens Predictor provides a prediction of growth of C. perfringens during the
cooling of meats. This model is part of ComBase Predictor and may give more
accurate predictions than the C. perfringens model included in PMP (89, 97).
Seafood Spoilage and Safety 77 Software includes models for relative rates of spoilage, models for growth of spoilage
Predictor (SSSP, version 3.0) bacteria in specific seafood, models to predict histamine formation by Morganella
spp., a model to predict the simultaneous growth of L. monocytogenes and lactic acid
bacteria in lightly preserved seafood, and a model to predict the growth boundary of
L. monocytogenes in lightly preserved seafood.

currently available, which ones are most suitable for less confidence in the model, then limited challenge
use, and what are the limitations of these models? studies may be warranted to verify the prediction from the
model (1).
Predictive food microbiology is a subdiscipline of food Caution should be exercised when models alone are
microbiology that uses models (i.e., mathematical equa- used to make a decision. Use of models requires experience
tions) to describe the growth, survival, or inactivation of and judgment, both in modeling and food microbiology.
microbes in food systems. Mathematical growth and When models alone are used to make a decision, those
inactivation models can always be used to help guide the models must be shown to be valid for the food in question
design of product assessments or challenge studies. In these and should take into consideration lot-to-lot variation.
cases, the challenge studies will either substantiate (i.e., Validation may be based on published or unpublished data
agree with or be conservative with respect to) the model for very similar or identical foods. The data should be
predictions or show those predictions to be invalid for the generated by a laboratory having personnel with the
specific product. An example of a conservative model appropriate knowledge, skills, and abilities in conducting
would be one that predicts a 2-log increase when the challenge studies (see Table 1) or be obtained from other
challenge study shows a 1-log increase. Two ideal uses of relevant published studies.
predictive models are for narrowing the choices for The two best-known multipathogen multifactor models
treatments to be validated for safety and for choosing the available today are the USDA Agricultural Research Service
appropriate challenge microorganisms. Eastern Regional Research Center Pathogen Modeling
Intrinsic and extrinsic factors (pH, aw, temperature, Program (PMP) (106) and the ComBase Predictor (50),
etc.) used as inputs for the model should be chosen with formerly known as the FoodMicroModel (Table 3). Both of
care. The least restrictive parameters determined for the these modeling programs make predictions for a wide array
range of processing conditions should be used. If the of foodborne pathogens and growth factors (temperature,
conditions modeled suggest that growth could occur or pH, etc.). Both programs are also based on data collected
that there is limited lethality for the product or process, primarily in laboratory media rather than foods and do not
then additional studies, product reformulation, or modifi- always cover the full range of each growth parameter
cation of target shelf life would be warranted. If there is (Tables 4 and 5). Elements of both models have been
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 155

TABLE 4. Pathogen growth ranges used in ComBase and Pathogen Modeling Programs a
ComBase PMP

Temp (uC) pH Temp (uC) pH

aw aw
Pathogen Minimum Maximum Minimum Maximum (minimum) Minimum Maximum Minimum Maximum (minimum)

B. cereus
With CO2 5 34 4.9 7.5 0.974
Aerobic 5 42 4.7 7.5 0.97
Anaerobic 10 42 5.0 9.0 0.97
C. botulinum
(growth only)
Proteolytic 14 40 4.7 7.2 0.954 15 34 5.0 7.2 0.977
Nonproteolytic 4 30 5.1 7.5 0.974 5 28 5.0 7.0 0.977
C. perfringens 15 52 5 8 0.971 19 37 6.0 6.5 0.983
E. coli O157:H7
With CO2 10 30 4.5 7 0.961
Aerobic 5 42 4.5 8.5 0.97
Anaerobic 5 42 4.5 8.5 0.97
L. monocytogenes
With CO2 1 35 4.4 7.5 0.934
Aerobic 4 37 4.5 7.5 0.928
Anaerobic 4 37 4.5 8.0 0.97
S. aureus
(growth only)
Not specified 7.5 30 4.4 7.1 0.907
Aerobic 10 42 4.5 9.0 0.911
Anaerobic 12 42 5.3 9.0 0.872
Salmonella
With CO2 7 30 3.9 7.4 0.973
Aerobic 10 30 5.6 6.8 0.974
a
Limits tested in ComBase (48) and the Pathogen Modeling Program (PMP) (106) do not necessarily represent limits for growth. See
Table 5 for growth limits.

validated (by both published and unpublished studies) to a 3. What are the limitations for applying the results of
limited degree in different food systems. an inoculated pack/challenge study conducted on
There is also a wide array of computer models one food to another similar food?
developed in laboratory media and food systems that are
not part of PMP and ComBase. Examples of several models Challenge studies on one product may sometimes be
are shown in Table 3. Some models published in the applicable to other products. However, if there are significant
scientific literature are not available in a user-friendly, differences between the intrinsic properties of the product and
downloadable form. These models require some modest those of the food on which the challenge study was conducted,
modeling or spreadsheet manipulation skills on the part of the results of the challenge study may not be applicable. If the
the user to produce a useful prediction. challenge study is conducted using parameters or conditions
Any discussion of modeling and validation of models more conducive to growth or survival than those in the food
would be remiss if it did not also mention another tool that product under consideration, then additional challenge studies
is part of the ComBase Modeling Toolbox: the ComBase may not be needed (76). For example, the results of a challenge
browser (48). The ComBase browser provides access to study for a specific pathogen in a product formulation with a
the ComBase database of microbial responses to food pH of 5.8 could be applied to a similar formulation where the
environments. At the present time the database includes primary difference is a pH of 5.4. Nevertheless, an expert
more than 35,000 observations, of which more than microbiologist should make the determination of applicability
13,000 are from food and the balance (,22,000) are from of one challenge study to additional products. The composition
culture media. Researchers publishing microbial growth or of the two foods, e.g., protein content, carbohydrate source,
survival data are requested and encouraged to submit the type of organic acid, fat, and moisture, should be considered in
data to ComBase (49). The data contained in ComBase determining the applicability of one study to another product.
may represent a useful source of published and unpub- Generally, the more similar the composition the more likely the
lished data for validating models. study will apply.
156 NACMCF J. Food Prot., Vol. 73, No. 1

TABLE 5. Limits for growth when other conditions are near optimum based on references (54) and (116)
Temp (uC) pH
aw Maximum water
Pathogen Sourcea Minimum Maximum Minimum Maximum (minimum) phase NaCl (%)

B. cereus FDA 4 55 4.3 9.3 0.92 10

ICMSF 4 55 5.0 8.8 0.93
C. botulinum (growth only, FDA 10 48 4.6 9 0.93 10
proteolytic) ICMSF 10–12 4.6 0.93 10
C. botulinum (growth only, FDA 3.3 45 5 9 0.97 5
nonproteolytic) ICMSF 3.3 5.0 0.97 5
C. perfringens FDA 10 52 5 9 0.93 7
ICMSF 12 50 5.5–5.8 8.0–9.0 0.97
Pathogenic E. coli FDA 6.5 49.4 4 9 0.95 6.5
ICMSF 7–8 44–46 4.4 9.0 0.95
E. coli O157:H7 ICMSF 8 44–45 4.5 Slow growth at 6.5,
no growth at 8.5
L. monocytogenes FDA 20.4 45 4.4 9.4 0.92 10
ICMSF 20.4 45 4.39 9.4 0.92
S. aureus (growth only) FDA 7 50 4 10 0.83 20
Aerobic conditions ICMSF 7 48 4 10 0.83
Anaerobic conditions ICMSF 5.0 0.90
Salmonella FDA 5.2 46.2 3.7 9.5 0.94 8
ICMSF 5.2b 46.2 3.8 9.5 0.94
a
FDA, U.S. Food and Drug Administration (116); ICMSF, International Commission on Microbiological Specifications for Foods (54).
b
Most serovars will not grow below 7uC (44.6uF).

4. Of the existing inoculated pack/challenge study given to challenge tests with C. botulinum, only with C.
protocols, e.g., those published by the American perfringens. The recommendations would result in unneces-
Bakers Association, NSF International, and others, sary and sometimes inappropriate challenge tests. There is no
which are most suitable for application to a wide consideration for the need to adapt the inoculum, and the
variety of foods, and what are the limitations of these inoculum size is fixed for all products. The protocol does take
protocols? Are there existing protocols that are into consideration the need to inoculate different components
appropriate for specific food-pathogen pairs? and interfaces of multicomponent products and requires
The committee agrees with an earlier assessment in the testing of duplicate samples per time point with multiple lots
IFT report (53) indicating that both the American Bakers of products. Overall, the protocol has significant limitations,
Association (ABA) and the National Sanitation Foundation even for application to the intended products.
International (NSF) testing protocols suffer from significant The ABA protocol (Industry Protocol for Establishing
weaknesses. These are briefly highlighted below; for more the Shelf Stability of Pumpkin Pie) is even more limited in
details, see Table 2 in the IFT report (53) comparing the scope (i.e., applies only to pumpkin pie intended for
NSF, ABA, and expert panel’s protocols. distribution and display without refrigeration). The objective
The NSF protocol provides test methods for determining of this protocol is to define the process that a manufacturer
that a product does not require refrigeration for safety. The can use to demonstrate the shelf stability of a pumpkin pie
NSF protocol lacks flexibility and is highly prescriptive in product in accordance with the then current edition of the
specifying microbial strains and methods. It applies to a FDA Food Code. This protocol is not an inoculated
limited number of products (breads and pastries with challenge study but rather a method for validating a cooking
vegetables or soft cheeses added prior to baking; bakery procedure (product reaches at least 82.2uC [180uF] at the
products filled or topped with cream, crème, custard, or cheese coolest point) with respect to the destruction of naturally
after baking; products filled prior to baking, such as pumpkin, occurring microorganisms, both pathogenic and nonpatho-
sweet potato, custard, or meringue pies; and toppings, glazes, genic. However, the absence of a pathogen in such a study
icings, or fillings stored without temperature control) and cannot be relied on to assess whether or not a pathogen
excludes a number of products of potential concern (e.g., would grow if present in the product, because such a
modified-atmosphere-packaged products, all products with a pathogen may or may not have been present initially.
pH , 4.6, and products stored without temperature control for Additionally, monitoring the oxidation-reduction potential
less than 24 h or more than 31 days). The aw and pH are the in the product to ascertain whether C. botulinum would
only criteria for selection of challenge test organisms, with no grow and produce toxin is inadequate to make such a
consideration of the process given the product in selecting determination. Thus, the protocol has significant limitations,
appropriate organisms. In addition, there is no consideration even for application to the intended product.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 157

The IFT expert panel report (53) is written to The protocol indicates that to assess growth potential
encompass a wide variety of foods. The guidelines provide samples need be taken only initially and at the end of the
a framework for determining whether foods need TCS. The shelf life and that for homogeneous products enumeration
document also describes guidelines for challenge tests for of only one sample is needed (three samples for
determining the ability of a food to support the growth of heterogeneous products) at each of these time points.
one or more pathogens, but it does not address inactivation (More sampling times are recommended for studies
challenge tests. The guidelines provide flexibility but result intended to assess maximum growth rate or lag time.)
in a potential for different interpretations as to what is The methods described in the AFSSA document are
appropriate for specific food types. This makes it more appropriate for L. monocytogenes in refrigerated ready-to-
difficult for those reviewing and evaluating the data to eat foods; however, the acceptance criteria differ from
determine if the study itself was adequate, and thus the those proposed here.
reviewer may need to have technical expertise for the In several documents, NACMCF has provided guid-
assessment. This is a weakness inherent to any document ance for conducting microbial challenge tests. In 1990,
that is designed to apply to a broad range of food types. NACMCF (74) made recommendations for extended-shelf-
Notermans et al. (80) developed a user’s guide to life refrigerated cooked meat and poultry products that
microbial challenge testing for food safety and stability. The included appendices on guidelines for thermal inactivation
document addresses selecting the appropriate microorgan- studies using L. monocytogenes and for C. botulinum
ism, preparing the inoculum, inoculum size, inoculation inoculation studies. Those recommendations are generally
procedure, duration of the study, and sampling times. The consistent with this NACMCF document. Although the
recommendations are generally consistent with those in this approaches used in the 1990 document are not specific to
NACMCF document, although less detailed. As with the refrigerated meat and poultry, they are specific for the
IFT expert panel report (53), technical expertise may be individual organism for which the guidance was developed.
required to interpret the adequacy of studies following these The protocols are appropriate for their intended use.
guidelines. In 2005, NACMCF (75) published an article on
Scott et al. (91) published guidelines for conducting L. considerations for establishing safety-based consume-by
monocytogenes challenge tests for foods. Their article date labels for refrigerated ready-to-eat foods; the appendix
covers guidelines for studies to evaluate both the ability of to that document contained guidance for conducting
a food to support the growth of L. monocytogenes and the microbial challenge studies to validate the safety-based
inactivation of the organism in a food. The article in large use-by date label. This guidance was specific for L.
part applies the recommendations in the IFT report (53) to monocytogenes and is consistent with the guidance in this
challenge studies involving L. monocytogenes and is thus NACMCF document. The protocol is appropriate for its
specific to a single organism. The protocols are also limited intended application (validation of a use-by date).
to those food products in which growth or inactivation of L. There are a number of good challenge test protocols
monocytogenes is a concern. The protocols in general are useful for specific purposes. This document and the IFT
consistent with those in the present document and are report (53) are the most comprehensive, broad-based
appropriate for L. monocytogenes in refrigerated ready-to- documents that can be applied to assess the adequacy of
eat foods. microbial challenge studies. Because these documents are
AFSSA, a European Union (EU) community reference not specific to a food category, technical expertise may be
laboratory for L. monocytogenes, has recently published a needed to assess the adequacy of the challenge study with
technical guidance document for conducting shelf-life respect to appropriateness of the challenge organism,
studies to determine compliance with microbiological storage temperatures, etc. However, a well-written report
criteria for L. monocytogenes in ready-to-eat foods set should provide the rationale for many of the choices, thus
out in EC regulation No. 2073/2005 (13). Similar to Scott assisting in the review to determine study adequacy.
et al. (91), the scope is limited to L. monocytogenes,
5. Develop a decision tree to aid in the design of an
including information on how to conduct experiments of
appropriate inoculated pack/challenge study. Test or
the shelf life in naturally contaminated and artificially
‘‘desk check’’ the decision tree using the following
contaminated ready-to-eat products. The document in-
five foods: meat-filled puff pastry, (baked) cheese
cludes determination of shelf life in naturally contaminated
pizza, chopped lettuce, cheese (blocks or slices), and
foods, called durability studies, which are not addressed in
lemon meringue pie.
this NACMCF document. The AFSSA document also
provides information on how to interpret the results Due to the complexity of decisions needed, the
obtained against EU L. monocytogenes regulatory criteria committee concluded that a decision tree could not be
in ready-to-eat foods (no more than 100 CFU/g at end of developed. Instead, the committee developed a template
shelf life). The document does not address inactivation of containing a series of questions to facilitate the design of an
L. monocytogenes but does address many of the same key appropriate challenge study. The template was validated
points as this NACMCF document, such as taking into using five food products (Appendices E through J).
account the product characteristics, batch variability, use The examples in Appendices E through J were
of multiple strains, adapting the challenge organisms, and developed to illustrate the thought processes that expert
simulating natural conditions when inoculating product. microbiologists use in approaching the design of microbial
158 NACMCF J. Food Prot., Vol. 73, No. 1

challenge tests. These examples should not be considered simple means of evaluating the acid tolerance of stationary-phase
complete or accurate with respect to all parameters. cells. Appl. Environ. Microbiol. 62:4009–4013.
15. Bunning, V. K., R. G. Crawford, J. T. Tierney, and J. T. Peeler.
Moreover, other approaches to conducting the challenge 1990. Thermotolerance of Listeria monocytogenes and Salmonella
studies may be applied. The pass-fail criteria used in the typhimurium after sublethal heat shock. Appl. Environ. Microbiol.
examples represent expert opinion and may need to be 56:3216–3219.
verified with the appropriate regulatory agency. 16. Chen, H., D. Guan, and D. G. Hoover. 2006. Sensitivities of
foodborne pathogens to pressure changes. J. Food Prot. 69:130–136.
6. Identify the basic knowledge, skills, education, 17. Chung, K. C., and J. M. Goepfert. 1970. Growth of Salmonella at
training, experience, and abilities necessary for a low pH. J. Food Sci. 35:326–328.
18. Conference for Food Protection. 2005. Temperature Control for
multidisciplinary work group or individual to be
Safety Committee of Council I. Feedback survey—results sum-
qualified to design, conduct, and evaluate an mary. Available at: http://www.foodprotect.org/ media/guide/
inoculated pack/challenge study and the pursuant TCS%20Committee%20Survey1.pdf. Accessed 15 June 2009.
results. 19. Cooley, M. B., D. Chao, and R. E. Mandrell. 2006. Escherichia coli
O157:H7 survival and growth on lettuce is altered by the presence of
Refer to Table 1, Question 1, Section 1.0 and Appendix epiphytic bacteria. J. Food Prot. 69:2329–2335.
B for this information. 20. Craven, S. E., and L. C. Blankenship. 1983. Increased heat
resistance of salmonellae in beef with added soy proteins. J. Food
REFERENCES Prot. 46:380–384.
21. Curiale, M. S. 1991. Shelf-life evaluation analysis. Dairy Food
1. Abdul-Raouf, U. M., L. R. Beuchat, and M. S. Ammar. 1993.
Environ. Sanit. 11:364–369.
Survival and growth of Escherichia coli O157:H7 on salad
22. Defelip, G., and L. Toti. 1984. Food poisoning caused by gram-negative
vegetables. Appl. Environ. Microbiol. 59:1999–2006.
bacteria in milk products. Microbiol. Aliment. Nutr. 2:251–256.
2. Adams, M. R., and C. J. Hall. 1988. Growth inhibition of food-
23. Delaquis, P., S. Stewart, S. Cazaux, and P. Toivonen. 2002. Survival
borne pathogens by lactic and acetic acids and their mixtures. Int. J.
and growth of Listeria monocytogenes and Escherichia coli
Food Sci. Technol. 23:287–292.
O157:H7 in ready-to-eat iceberg lettuce washed in warm chlorinated
3. Ahamad, N., and E. H. Marth. 1989. Behavior of Listeria
water. J. Food Prot. 65:459–464.
monocytogenes at 7, 13, 21, and 35uC in tryptose broths acidified
24. Doyle, M. E., and A. S. Mazzotta. 2000. Review of studies on the
with acetic, citric, or lactic acid. J. Food Prot. 52:688–695.
thermal resistance of salmonellae. J. Food Prot. 63:779–795.
4. American Meat Institute Foundation. 2009. Process lethality
25. Doyle, M. E., A. S. Mazzotta, T. Wang, D. W. Wiseman, and V. N.
determination spreadsheet. Available at: http://www.amif.org/ht/d/
Scott. 2001. Heat resistance of Listeria monocytogenes. J. Food
sp/i/26870/pid/26870. Accessed 15 June 2009.
Prot. 64:410–429.
5. American Public Health Association. 2001. Compendium of
26. Duffy, S., J. Churey, R. W. Worobo, and D. W. Schaffner. 2000.
methods for the microbiological examination of foods, 4th ed. F.
Analysis and modeling of the variability associated with UV
P. Downes and K. Ito (ed.). American Public Health Association,
inactivation of Escherichia coli in apple cider. J. Food Prot. 63:
Washington, DC.
1587–1590.
6. American Public Health Association. 2004. Standard methods for
the examination of dairy products, 17th ed. H. M. Wehr and J. H. 27. Edelson-Mammel, S. G., R. C. Whiting, S. W. Joseph, and R. L.
Frank (ed.). American Public Health Association, Washington, DC. Buchanan. 2005. Effect of prior growth conditions on the thermal
7. Andrews, W. H., and T. S. Hammack. 2007. Salmonella, chap. 5. In inactivation of 13 strains of Listeria monocytogenes in two heating
Bacteriological analytical manual online. U.S. Food and Drug Admin- menstrua. J. Food Prot. 68:168–172.
istration. Available at: http://www.fda.gov/Food/ScienceResearch/ 28. Eifert, J. D., C. R. Hackney, M. D. Pierson, S. E. Duncan, and W. N.
LaboratoryMethods/BacteriologicalAnalyticalManualBAM/ucm070149. Eigel. 1997. Acetic, lactic, and hydrochloric acid effects on
htm. Accessed 15 June 2009. Staphylococcus aureus 196E growth based on a predictive model.
8. AOAC International. 2006. Guidelines for laboratories performing J. Food Sci. 62:174–178.
microbiological and chemical analyses of food and pharmaceuticals: 29. Eimerman, P., M. Hanson, A. E. Larson, L. M. McDonnell, K. A.
an aid to interpretation of ISO/IEC 17025:2005. AOAC Interna- Glass, and E. A. Johnson. 2003. Survival of Salmonella and Listeria
tional, Gaithersburg, MD. monocytogenes on shredded cheese. J. Food Prot. 66(Suppl. A):107.
9. AOAC International. 2007. Official methods of analysis, 18th ed., 30. Farber, J. M., G. W. Sanders, S. Dunfield, and R. Prescott. 1989.
rev. 2. W. Horwitz and G. Latimer, Jr. (ed.). AOAC International, The effect of various acidulants on the growth of Listeria
Gaithersburg, MD. monocytogenes. Lett. Appl. Microbiol. 9:181–183.
10. Baird-Parker, A. C., M. Boothroyd, and E. Jones. 1970. The effect 31. Feng, P., K. A. Lampel, H. Karch, and T. S. Whittam. 1998.
of water activity on the heat resistance of heat sensitive and heat Genotypic and phenotypic changes in the emergence of Escherichia
resistant strains of salmonellae. J. Appl. Bacteriol. 33:515–522. coli O157:H7. J. Infect. Dis. 177:1750–1753.
11. Beuchat, L. R. 1974. Combined effects of water activity, solute, and 32. Flessa, S., D. M. Lusk, and L. J. Harris. 2005. Survival of Listeria
temperature on the growth of Vibrio parahaemolyticus. Appl. monocytogenes on fresh and frozen strawberries. Int. J. Food
Microbiol. 27:1075–1080. Microbiol. 101:255–262.
12. Beuchat, L. R., J. M. Farber, E. H. Garret, L. J. Harris, M. E. Parish, 33. Fonseca, J. M. 2006. Postharvest quality and microbial population
T. V. Suslow, and F. F. Busta. 2001. Standardization of a method to of head lettuce as affected by moisture at harvest. J. Food Sci. 71:
determine the efficacy of sanitizers in inactivating human patho- M45–M49.
genic microorganisms on raw fruits and vegetables. J. Food Prot. 34. Gahan, C. G. M., B. O’Driscoll, and C. Hill. 1996. Acid adaptation
64:1079–1084. of Listeria monocytogenes can enhance survival in acidic foods and
13. Beaufort, A., M. Cornu, H. Bergis, A.-L. Lardeux, and B. Lombard. during milk fermentation. Appl. Environ. Microbiol. 62:3128–3132.
2008. Technical guidance document on shelf-life studies for Listeria 35. Geornaras, I., K. E. Belk, J. A. Scango, P. A. Kendall, G. C. Smith,
monocytogenes in ready-to-eat foods. Ver. 2, November 2008. and J. N. Sofos. 2005. Postprocessing antimicrobial treatments to
Agence Française de Sécurité Sanitaire des Aliments. Available at: control Listeria monocytogenes in commercial vacuum-packaged
http://ec.europa.eu/food/food/biosafety/salmonella/docs/shelflife_ bologna and ham stored at 10uC. J. Food Prot. 68:991–998.
listeria_monocytogenes_en.pdf. Accessed 16 March 2009. 36. Glass, K. A., and E. A. Johnson. 2004. Antagonistic effect of fat on
14. Buchanan, R. L., and S. G. Edelson. 1996. Culturing enterohemor- the antibotulinal activity of food preservatives and fatty acids. Food
rhagic Escherichia coli in the presence and absence of glucose as a Microbiol. 21:675–682.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 159

37. Glass, K. A., B. B. Prasad, J. H. Schlyter, H. E. Uljas, N. Y. Farkye, 60. Kang, D. H., and D. Y. C. Fung.. 1999. Thin agar layer method for
and J. B. Luchansky. 1995. Effects of acid type and ALTA 2341 on recovery of heat-injured Listeria monocytogenes. J. Food Prot. 62:
Listeria monocytogenes in a queso blanco type of cheese. J. Food 1346–1349.
Prot. 58:737–741. 61. Kang, D. H., and G. R. Siragusa. 1999. Agar underlay method for
38. Goepfert, J. M., I. K. Iskander, and C. H. Amundson. 1970. Relation recovery of sublethally heat-injured bacteria. Appl. Environ.
of the heat resistance of salmonellae to the water activity of the Microbiol. 65:5334–5337.
environment. Appl. Microbiol. 19:429–433. 62. Koseki, S., and S. Isobe. 2005. Growth of Listeria monocytogenes on
39. Gould, G. W. 2000. Preservation: past, present and future. Br. Med. iceberg lettuce and solid media. Int. J. Food Microbiol. 101:217–225.
Bull. 56:84–96. 63. Koseki, S., and K. Yamamoto. 2006. pH and solute concentration of
40. Han, Y., R. H. Linton, S. S. Nielsen, and P. E. Nelson. 2002. A suspension media affect the outcome of high hydrostatic pressure
comparison of methods for recovery of chlorine dioxide–injured treatment of Listeria monocytogenes. Int. J. Food Microbiol. 111:
Escherichia coli O157:H7 and Listeria monocytogenes. Food 175–179.
Microbiol. 19:201–210. 64. Larson, A. E., and E. A. Johnson. 2003. Efficacy of clostridial plate
41. Hanning, I. B., M. G. Johnson, and S. C. Ricke. 2008. Precut prepack- counts as a substitute for botulinum toxin detection during botulinal
aged lettuce: a risk of listeriosis? Foodborne Pathog. Dis. 5:731–746. challenge studies of foods. J. Food Prot. 66(Suppl. A): 94.
42. Harris, L. J. 17 March 2009. Personal communication. E-mail: 65. Leyer, G. J., and E. A. Johnson. 1992. Acid adaptation promotes
[email protected]. survival of Salmonella spp. in cheese. Appl. Environ. Microbiol. 58:
43. Hayman, M. M., G. K. Kouassi, R. C. Anantheswaran, J. D. Floros, 2075–2080.
and S. J. Knabel. 2008. Effect of water activity on inactivation of 66. Leyer, G. J., L. H. Wang, and E. A. Johnson. 1995. Acid adaptation
Listeria monocytogenes and lactate dehydrogenase during high of Escherichia coli O157:H7 increases survival in acidic foods.
pressure processing. Int. J. Food Microbiol. 124:21–26. Appl. Environ. Microbiol. 61:3752–3755.
44. Health Canada. 2008. The compendium of analytical methods, vols. 67. Li, Y., R. E. Brackett, J. Chen., and L. R. Beuchat. 2001. Survival
1–5. Available at: http://www.hc-sc.gc.ca/fn-an/res-rech/analy-meth/ and growth of Escherichia coli O157:H7 inoculated onto cut lettuce
microbio/index-eng.php. Accessed 18 December 2008. before or after heating in chlorinated water, followed by storage at 5
45. Hijarrubia, M. J., B. Lazaro, E. Sunen, and A. Fernandez-Astorga. or 15uC. J. Food Prot. 64:305–309.
1996. Survival of Vibrio vulnificus under pH, salinity and 68. Lianou, A., and J. N. Sofos. 2007. A review of the incidence and
temperature combined stress. Food Microbiol. 13:193–199. transmission of Listeria monocytogenes in ready-to-eat products in
46. Hill, C., B. O’Driscoll, and I. Booth. 1995 Acid adaptation and food retail and food service environments. J. Food Prot. 9:2172–2198.
poisoning microorganisms. Int. J. Food Microbiol. 28:245–254. 69. Llaudes, M. M., L. Zhao, S. Duffy, and D. W. Schaffner. 2001.
47. Hyytia, E., S. Hielm, M. Mokkila, A. Kinnumen, and H. Korkeala. Simulation and modeling of the effect of small inoculum size on
1999. Predicted and observed growth and toxigenesis by Clostrid- time to spoilage by Bacillus stearothermophilus. Food Microbiol.
ium botulinum type E in vacuum-packaged fishery product 18:395–405.
challenge tests. Int. J. Food Microbiol. 47:161–169. 70. MacFarquhar, J., J. Dunn, D. Swerdlow, K. Jackson, W. Schaffner,
48. Institute of Food Research 2006. ComBase browser. Available at: S. Stroika, and T. Jones. 2008. Multistate investigation of
http://combase.arserrc.gov/BrowserHome/SearchOptions/Search.aspx. Escherichia coli O157:H7 infections associated with frozen pizza,
Accessed 25 September 2008. p. 238. Abstr. Int. Conf. Emerg. Infect. Dis. Available at: http://
49. Institute of Food Research. 2006. ComBase database. Available at: www.iceid.org/documents/Final_Program-2008.pdf. Accessed 10
http://www.combase.cc/faq.html. Accessed 25 September 2008. December 2008.
50. Institute of Food Research. 2006. ComBase predictor. Available at: 71. Margosch, D., M. A. Ehrmann, M. G. Ganzle, and R. F. Vogel.
http://www.combase.cc/predictor.html. Accessed 25 September 2008. 2004. Comparison of pressure and heat resistance of Clostridium
51. Institute of Food Research. 2006. Perfringens predictor. Available at: botulinum and other endospores in mashed carrots. J. Food Prot. 67:
http://www.ifr.ac.uk/safety/growthpredictor/. Accessed 15 June 2009. 2530–2537.
52. Institute of Food Technologists. 2000. Kinetics of microbial inactiva- 72. Mazurek, J., L. Holbert, M. K. Parrish, and E. Salehi. 2005. Raw
tion for alternative food processing technologies. Available at: http:// eggs—lessons learned from an outbreak of Salmonella serotype
www.cfsan.fda.gov/,comm/ift-toc.html. Accessed 18 August 2008. Enteritidis infections associated with meringue pie. J. Public Health
53. Institute of Food Technologists. 2003. Evaluation and definition of Manag. 11:201–207.
potentially hazardous foods. Comp. Rev. Food Sci. Food Saf. 73. Miao, E. A., and S. I. Miller. 1999. Bacteriophages in the evolution of
2(Suppl.):1–109. pathogen-host interactions. Proc. Natl. Acad. Sci. USA 96:9452–9454.
54. International Commission on Microbiological Specifications for 74. National Advisory Committee on Microbiological Criteria for
Food. 1996. Microorganisms in foods 5. Characteristics of microbial Foods. 1990. Recommendations of the National Advisory Commit-
pathogens. Blackie Academic and Professional, London. tee on Microbiological Criteria for Foods for refrigerated foods
55. International Organization for Standardization. 2006. Microbiology containing cooked, uncured meat or poultry products that are
of food and animal feeding stuffs—guidelines for the estimation of packaged for extended refrigerated shelf-life and that are ready-to-
measurement uncertainty for quantitative determinations. ISO/TS eat or prepared with little or no additional heat treatment. National
19036. Available at: http://www.iso.org. Accessed 18 March 2009. Advisory Committee on Microbiological Criteria for Foods,
56. International Organization for Standardization. 2009. General Washington, DC.
methods of tests and analysis for food products. ICS 67.050. 75. National Advisory Committee on Microbiological Criteria for
Available at: http://www.iso.org/iso/catalogue_ics_browse?ICS1~ Foods. 2005. Considerations for establishing safety-based con-
67&ICS2~050&. Accessed 15 June 2009. sume-by date labels for refrigerated ready-to-eat foods. J. Food
57. International Organization for Standardization. 2009. Microbiology Prot. 68:1761–1775.
of food and animal feeding stuffs—guidelines for the estimation of 76. National Advisory Committee on Microbiological Criteria for
measurement uncertainty for quantitative determinations. Amend- Foods. 2006. Requisite scientific parameters for establishing the
ment 1: measurement uncertainty for low counts. ISO/TS 19036. equivalence of alternative methods of pasteurization. J. Food Prot.
Available at http://www.iso.org. Accessed 18 March 2009. 69:1190–1216.
58. Jung D. S, F. W. Bodyfelt, and M. S. Daeschel. 1992. Influence of 77. National Institute of Aquatic Resources. 2008. Seafood spoilage and
fat and emulsifiers on the efficacy of nisin in inhibiting Listeria safety predictor. Technical University of Denmark. Available at:
monocytogenes in fluid milk. J. Dairy Sci. 75:387–393. http://sssp.dtuaqua.dk/. Accessed 15 June 2009.
59. Juven, B. J., N. A. Cox, J. S. Bailey, J. E. Thomson, O. W. Charles, 78. National Research Institute Japan. 2008. Microbial responses viewer
and J. V. Shutze. 1984. Survival of Salmonella in dry food and feed. (MRV) for ComBase (ver. beta 1). Available at: http://cbnfri.dc.
J. Food Prot. 47:445–448. affrc.go.jp/MRV/Welcome.html. Accessed 15 June 2009.
160 NACMCF J. Food Prot., Vol. 73, No. 1

79. Ng, H., H. G. Bayne, and J. A. Garibaldi. 1969. Heat resistance of 100. Sorrells, K. M., M. L. Speck, and J. S. Warren. 1970. Pathogenicity
Salmonella: the uniqueness of Salmonella senftenberg 775W. Appl. of Salmonella gallinarum after metabolic injury by freezing. Appl.
Microbiol. 17:78–82. Microbiol. 19:39–43.
80. Notermans, S., P. in’t Veld, T. Wijtzes, and G. C. Mead. 1993. A 101. Stecchini, M. L., V. Aquili, and I. Sarais. 1995. Behavior of Listeria
user’s guide to microbial challenge testing for ensuring the safety monocytogenes in mozzarella cheese in presence of Lactococcus
and stability of food products. Food Microbiol. 10:145–157. lactis. Int. J. Food Microbiol. 25:301–310.
81. Odlaug, T. E., and I. J. Pflug. 1979. Clostridium botulinum growth 102. Sumner, S. S., T. M. Sandros, M. C. Harmon, V. N. Scott, and D. T.
and toxin production in tomato juice containing Aspergillus gracilis. Bernard. 2001. Heat resistance of Salmonella typhimurium and
Appl. Environ. Microbiol. 37:496–504. Listeria monocytogenes in sucrose solutions of various water
82. Oka, S. 1964. Mechanisms of antimicrobial effect of various food activities. J. Food Sci. 56:1741–1743.
preservatives, p. 1–15. In N. Molin (ed.), Microbial inhibitors in 103. Swanson, K. M. J., R. L. Petran, and J. H. Hanlin. 2001. Culture
food. Almqvist and Wiksell, Stockholm. methods for enumeration of microorganisms, p. 53–62. In F. P.
83. Ouattara, B., R. E. Simard, R. A. Holley, G. J. Piette, and A. Begin. Downes and K. Ito (ed.), Compendium of methods for the
1997. Inhibitory effect of organic acids upon meat spoilage bacteria. microbiological examination of foods, 4th ed. American Public
J. Food Prot. 60:246–253. Health Association, Washington, DC.
84. Pupo, G. M., D. K. Karaolis, R. Lan, and P. R. Reeves. 1997. 104. University of Wisconsin–Madison. 2009. Isothermal-based predic-
Evolutionary relationships among pathogenic and nonpathogenic tion tool. Available at http://www.meathaccp.wisc.edu/ibm.htm.
Escherichia coli strains inferred from multilocus enzyme Accessed 15 June 2009.
electrophoresis and mdh sequence studies. Infect. Immun. 65: 105. U.S. Department of Agriculture, Agriculture Marketing Service.
2685–2692. 2009. Title 7—agriculture. Part 981—almonds grown in California.
85. PURAC. 2007. Opti.Form Listeria control model. Available at: http:// Code of Federal Regulations. U.S. Government Printing Office,
www.purac.com/purac_com/a5348511153c582f5bd69fd6bd64bb49. Washington, DC.
php. Accessed 15 June 2009. 106. U.S. Department of Agriculture, Agricultural Research Service,
86. Rajan, S., J. Ahn, V. M. Balasubramaniam, and A. E. Yousef. 2006. Eastern Regional Research Center. 2006. Pathogen modeling
Combined pressure-thermal inactivation kinetics of Bacillus amylo- program. Available at: http://ars.usda.gov/Services/docs.htm?docid~
liquefaciens spores in egg patty mince. J. Food Prot. 69:853–860. 11550. Accessed 25 September 2008.
87. Riordan, D. C. R., G. Duffy, J. J. Sheridan, R. C. Whiting, I. S. 107. U.S. Department of Agriculture, Food Safety and Inspection
Blair, and D. A. McDowell. 2000. Effects of acid adaptation, Service. 1998. Microbiology laboratory guidebook, 3rd ed. B. P.
product pH and heating on survival of Escherichia coli O157:H7 in Dey and C. P. Lattuada (ed.). Available at: http://www.fsis.usda.
pepperoni. Appl. Environ. Microbiol. 66:1726–1729. gov/Science/Microbiological_Lab_Guidebook/index.asp. Accessed
88. Rusul, G. 1995. Prevalence of Bacillus cereus in selected foods and 18 December 2008.
detection of enterotoxin using TECRA-VIA and BCET-RPLA. Int. 108. U.S. Department of Agriculture, Food Safety and Inspection
J Food Microbiol. 25:131–139. Service. 1999. Performance standards for the production of certain
89. Schaffner, D. W. 25 September 2008. Personal communication. meat and poultry products. Appendix A. Compliance guidelines for
E-mail: [email protected]. meeting lethality performance standards for certain meat and poultry
90. Scott, V. N., R. S. Clavero, and J. A. Troller. 2001. Measurement of products. Available at: http://www.fsis.usda.gov/OPPDE/rdad/
water activity (aw), acidity, and brix, chap. 64. In F. P. Downes and K. FRPubs/95-033F/95-033F_Appendix%20A.htm. Accessed 22 Sep-
Ito (ed.), Compendium of methods for the microbiological examination tember 2008.
of foods, 4th ed. American Public Health Association, Washington, DC. 109. U.S. Department of Agriculture, Food Safety and Inspection
91. Scott, V. N., K. M. J. Swanson, T. A. Freier, W. P. Pruett, W. H. Service. 1999. Performance standards for the production of certain
Sveum, P. A. Hall, L. A. Smoot, and D. G. Brown. 2005. Guidelines meat and poultry products. Appendix B. Compliance guidelines for
for conducting Listeria monocytogenes challenge testing of foods. cooling heat-treated meat and poultry products (stabilization).
Food Prot. Trends 25:818–825. Available at: http://www.fsis.usda.gov/OPPDE/rdad/FRPubs/
92. Scotter, S. L., S. Langton, B. Lombard, C. Lahellec, S. Schulten, N. 95-033F/95-033F_Appendix%20B.htm. Accessed 22 September
Nagelkerke, P. H. in’t Veld, and P. Rollier. 2001. Validation of ISO 2008.
method 11290 part 2. Enumeration of Listeria monocytogenes in 110. U.S. Department of Agriculture, Food Safety and Inspection
foods. Int. J. Food Microbiol. 70:121–129. Service. 2001. Performance standards for the production of
93. Seo, K. H., and J. F. Frank. 1999. Attachment of Escherichia coli processed meat and poultry products; proposed rule. Fed. Regist.
O157:H7 to lettuce leaf surface and bacterial viability in response to 66:1289–12636. Available at: http://frwebgate.access.gpo.gov/
chlorine treatment as demonstrated by using confocal scanning laser cgi-bin/getdoc.cgi?dbname~2001_register&docid~01-4420-filed.
microscopy. J. Food Prot. 62:3–9. pdf. Accessed 10 December 2008.
94. Shah, D. B., J. G. Bradshaw, and J. T. Peeler. 1991. Thermal 111. U.S. Department of Agriculture, Food Safety and Inspection
resistance of egg-associated epidemic strains of Salmonella Service. 2005. Risk assessments of Salmonella Enteritidis in shell
enteritidis. J. Food Sci. 56:391–393. eggs and Salmonella spp. in egg products. Available at: http://www.
95. Simpson, C. A., I. Geornaras, Y. Yoon, J. A. Scanga, P. A. Kendall, fsis.usda.gov/PDF/SE_Risk_Assess_Oct2005.pdf. Accessed 25
and J. N. Sofos. 2008. Effect of inoculum preparation procedure and September 2008.
storage time and temperature on the fate of Listeria monocytogenes 112. U.S. Department of Agriculture, Food Safety and Inspection
on inoculated salami. J. Food Prot. 71:494–501. Service. 2009. Title 9—animals and animal products. Part 381—
96. Smith, J. L., B. S. Marmer, and R. C. Benedict. 1991. Influence of poultry products inspection regulations. Subpart 150—requirements
growth temperature on injury and death of Listeria monocytogenes for the production of fully cooked poultry products and partially
Scott A during milk heat treatment. J. Food Prot. 54:166–169. cooked poultry breakfast strips. Code of Federal Regulations. U.S.
97. Smith, S., and D. W. Schaffner. 2004. Evaluation of a Clostridium Government Printing Office, Washington, DC.
perfringens predictive model, developed under isothermal condi- 113. U.S. Department of Agriculture, Food Safety and Inspection
tions in broth, to predict growth in ground beef during cooling. Appl. Service. 2009. Title 9—animal and animal products. Part 430—
Environ. Microbiol. 70:2728–2733. requirements for specific classes of products. Code of Federal
98. Sofos, J., and F. Busta. 1981. Antimicrobial activity of sorbate. J. Regulations. U.S. Government Printing Office, Washington, DC.
Food Prot. 44:614–622. 114. U.S. Department of Health and Human Services. 2007. Biosafety in
99. Sommers, C. H., D. J. Geveke, and X. Fan. 2008. Inactivation of microbiological and biomedical laboratories, 5th ed. Available at:
Listeria innocua on frankfurters that contain potassium lactate and http://www.cdc.gov/OD/OHS/biosfty/bmbl5/BMBL_5th_Edition.pdf.
sodium diacetate by flash pasteurization. J. Food Sci. 73:M72–M74. Accessed 25 September 2008.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 161

115. U.S. Food and Drug Administration. 2001. Bacteriological analyt- catalogue_ics_browse?ICS1~67&ICS2~050&. Accessed 15
ical manual. Available at: http://www.fda.gov/Food/ScienceResearch/ June 2009. (56)
LaboratoryMethods/BacteriologicalAnalyticalManualBAM/default.
htm. Accessed 15 June 2009.
N U.S. Department of Agriculture, Food Safety and Inspection
Service. 1998. Microbiology laboratory guidebook, 3rd ed., B.
116. U.S. Food and Drug Administration. 2001. Fish and fisheries
products hazards and controls guidance, 3rd ed. Available at: http://
P. Dey and C. P. Lattuada (ed.) Available at: http://www.fsis.
www.cfsan.fda.gov/,comm/haccp4.html. Accessed 18 August 2008. usda.gov/Science/Microbiological_Lab_Guidebook/index.asp.
117. U.S. Food and Drug Administration. 2005. Food code. U.S. Accessed 18 December 2008. (107)
Department of Health and Human Services, College Park, MD.
Available at: http://www.cfsan.fda.gov/,dms/fc05-toc.html. Ac-
N U.S. Food and Drug Administration. 2001. Bacteriological analyti-
calmanual.Availableat:http://www.fda.gov/Food/ScienceResearch/
cessed 25 September 2008. LaboratoryMethods/BacteriologicalAnalyticalManualBAM/default.
118. U.S. Food and Drug Administration. 2007. Acidified and low-acid htm. Accessed 15 June 2009. (115)
canned foods. Approximate pH of foods and food products.
Available at: www.cfsan.fda.gov/,comm/lacf-phs.html. Accessed
22 September 2008. APPENDIX B. CONSIDERATIONS FOR
119. U.S. Food and Drug Administration. 2009. Cheeses and related SELECTING A LABORATORY
cheese products. Code of Federal Regulations, Title 21, Part. 133.
Office of the Federal Register, U.S. Government Printing Office, Note: The following questions may be useful in comparing
Washington, DC. the capabilities of different laboratories. The questions are not
120. U.S. Food and Drug Administration. 2009. Title 21—food and listed in order of importance. A negative response to one or more
drugs. Part 120—hazard analysis critical control point (HACCP) of these questions does not necessarily disqualify a laboratory from
systems. Subpart 24—process controls. Code of Federal Regula- consideration. The most important considerations are associated
tions. U.S. Government Printing Office, Washington, DC.
with qualifications of personnel for designing, conducting, and
121. Uyttendaele, M., A. Rajkovic, G. Benos, K. Francois, F.
evaluating challenge studies.
Devlieghere, and J. Debevere. 2004. Evaluation of a challenge
testing protocol to assess the stability of ready-to-eat cooked meat
products against growth of Listeria monocytogenes. Int. J. Food
N Does the microbiologist in charge have experience performing
challenge studies including the food types you want to study? If
Microbiol. 90:219–236.
so, ask the laboratory to provide examples of the types of
122. Vestergard, E. M. 2001. Building product confidence with challenge
studies. Dairy Food Environ. Sanit. 21:206–209.
challenge studies performed recently.
123. Xavier, I. J., and S. C. Ingham. 1997. Increased D-values for N What is the academic education and training of the microbi-
ologist supervising the laboratory operations?
Salmonella enteritidis following heat shock. J. Food Prot. 60:181–
184.
124. Yan, Z., J. B. Gurtler, and J. L. Kornacki. 2006. A solid agar overlay
N What is the academic education and training of technicians
performing the laboratory experiments?
method for recovery of heat-injured Listeria monocytogenes. J.
Food Prot. 69:428–431.
N Is the laboratory audited periodically or accredited by an
independent third party? If so, ask the laboratory to provide a
125. Yoon, Y., M. Calicioglu, P. A. Kendall, G. C. Smith, and J. N. copy of certificates documenting the audit. If not, ask how the
Sofos. 2005. Influence of inoculum level and acidic marination on
laboratory ensures the quality of the processes and results, e.g.,
inactivation of Escherichia coli O157:H7 during drying and storage
of beef jerky. Food Microbiol. 22:423–431. appropriate positive and negative controls, and a written,
126. Zhao, I., T. J. Montville, and D. W. Schaffner. 2000. Inoculum size implemented quality control system for the laboratory opera-
of Clostridium botulinum 56A spores influences time-to-detection tions, including a corrective action plan. ISO17025 certification
and percent growth-positive samples. J. Food Sci. 65:1369–1375. is an example of a third party audit that would verify many of the
good laboratory practices that should be implemented. Accred-
APPENDIX A. SOURCES OF ACCEPTED itations and certifications do not necessarily qualify a laboratory
LABORATORY METHODS* to design and conduct microbiological challenge studies. It is
important to confirm that the laboratory has the experience and
N American Public Health Association. 2001. Compendium of expertise necessary to perform the challenge studies
methods for the microbiological examination of foods, 4th ed. N Does the laboratory use approved, validated, or widely accepted
published methods for the requested analyses? If so, what are
F. P. Downes and K. Ito (ed.). American Public Health
Association, Washington, DC. (5) the references for the methods used?
N American Public Health Association. 2004. Standard methods N Does the laboratory use certified reference materials (e.g.,
traceable positive controls) and standards (e.g., NIST calibrated
for the examination of dairy products, 17th ed. H. M. Wehr and
J. H. Frank (ed.). American Public Health Association, equipment), where applicable, to perform the requested tests?
Washington, DC. (6) N Does the laboratory use subcontractors to perform the analyses
N AOAC International. 2007. Official methods of analysis, 18th
ed., rev. 2, W. Horwitz and G. Latimer, Jr. (ed.). AOAC
in question? If so, how does the primary laboratory ensure the
subcontract laboratory produces valid results?
International, Gaithersburg, MD. (9) N If the protocol involves inoculation with a foodborne pathogen,
N Health Canada. 2008. The Compendium of analytical methods,
vols. 1–5. Available at: http://www.hc-sc.gc.ca/fn-an/res-rech/
does the laboratory have appropriate biological safety contain-
ment and practices?
analy-meth/microbio/index-eng.php. Accessed 18 December
2008. (44)
N Does the laboratory possess microbial strains that are appropriate
for the food to be challenged? How are the stocks maintained and
N International Organization for Standardization. 2009. General verified for purity and identity prior to the start of the study?
methods of tests and analysis for food products. ICS 67.050.
Listing of standards available at: http://www.iso.org/iso/
N If the protocol involves testing for a select agent (e.g.,
Clostridium botulinum or botulinum toxin), is the laboratory
approved to work with that particular agent? In the United
* Dates of references current as of publication. Use most current version States, laboratories must be approved to work with each select
available. agent on which they perform tests or research.
162 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX C. PATHOGENS OF CONCERN AND CONTROL METHODS FOR VARIOUS PRODUCT

CATEGORIES THAT MAY NEED A CHALLENGE STUDY (GROWTH INHIBITION, INACTIVATION,
OR COMBINATION) a
Pathogens of concern Examples of process controlc (alone and in
b
Product category (examples) (alphabetical order) combination, alphabetical order)

Meat and poultry: cooked (e.g., roast beef, Clostridium botulinum and C. perfringens, Cooling rate, heat treatment,d high pressure
deli-style turkey, ham) enterohemorrhagic E. coli, L. processing, preservatives, storage time-
monocytogenes, Salmonella, temperature
Staphylococcus aureus
Meat and poultry: dried and/or fermented C. botulinum, C. perfringens, aw, drying, fermentation, heat treatment,
(e.g., fermented sausage, jerky, dry enterohemorrhagic E. coli, L. humidity, nitrites and other preservatives,
cured ham) monocytogenes, Salmonella, S. aureus pH salting, storage time-temperature,
water-phase salt
Fish and seafood (e.g., smoked fish, B. cereus, C. botulinum, L. monocytogenes, aw, drying, harvest site control, heat treatment,
fresh oysters, pickled herring, pasteurized Salmonella, Shigella spp., S. aureus, high-pressure processing, nitrites, pH,
crab meat) Vibrio cholerae, V. vulnificus, V. preservatives, salting, storage time-
parahaemolyticus temperature, water-phase salt
Cultured dairy products, pH # 4.7 Enterohemorrhagic E. coli, Salmonella, Heat treatment, pH, preservatives, rate of acid
(e.g., yogurt, sour cream, buttermilk) L. monocytogenes, S. aureus production, starter culture activity, storage
time-temperature
Cultured dairy products, pH . 4.7 to #5.4 B. cereus, C. botulinum, enterohemorrhagic Heat treatment, hot fill, preservatives, storage
(e.g., cottage cheese) E. coli, L. monocytogenes, Salmonella, time-temperature
S. aureus
Cheese and cheese products (e.g., natural C. botulinum, enterohemorrhagic E. coli, aw, emulsifiers, heat treatment, hot fill,
Swiss cheese, processed cheese slices, L. monocytogenes, Salmonella, Shigella moisture content, pH, preservatives,
processed cheese spread) spp., S. aureus storage time-temperature
Butter and margarine (e.g., light salted L. monocytogenes, S. aureus, Yersinia aw, heat treatment, moisture droplet size in the
butter, whipped butter) enterocolitica water-in-oil emulsion, water-phase salt
Eggs and egg products (e.g., meringue, pooled B. cereus, L. monocytogenes, Salmonella Heat treatment, preservatives, storage time-
pasteurized egg yolks, sliced boiled eggs) temperature
Fruits and vegetables (e.g., peeled carrots, B. cereus, C. botulinum, enterohemorrhagic Heat treatment, storage time-temperature,
chopped lettuce) E. coli, L. monocytogenes, Salmonella, wash water sanitizers
Shigella spp., Y. enterocolitica
Fats, oils, and condiments (e.g., garlic B. cereus, C. botulinum, S. aureus, aw, heat treatment, pH, preservatives, salt,
in oil)e Salmonella storage time-temperature
Acidified sauces, salad dressings, and Enterohemorrhagic E. coli, Salmonella, Heat treatment, pH, storage time-temperature,
salsas S. aureus titratable acidity
High aw syrups (e.g., light maple syrup) C. botulinumf Acidification (light syrups), aw, heat
treatment, preservatives
Confectionery products (e.g., chocolate products) Salmonella aw, heat treatment
Cereal grains and related products (e.g., fresh B. cereus, C. botulinum, Salmonella, S. aw, heat treatment, pH, preservatives, storage
pasta, cooked rice) aureus time-temperature
a
Adapted from IFT report (53) Tables 4-1 and 6-1.
b
Combinations of products, storage in modified atmosphere, and use of novel preservatives or processes require special consideration.
c
Good agricultural practices where appropriate and good manufacturing practices and HACCP principles would help in reducing the
hazards.
d
Heat treatment includes processes such as cooking, pasteurization, and other thermal processes intended to inactivate pathogens.
e
Only a concern in anoxic environments.
f
Only a concern in light syrups and can be controlled by acidification.

APPENDIX D. FDA 2005 MODEL FOOD CODE operator of a food establishment or food processing plant, and
DEFINITIONS MOST RELEVANT does not offer the food for resale.
TO CHALLENGE STUDIES ‘‘Critical control point’’ means a point or procedure in a specific
food system where loss of control may result in an unacceptable
The following definitions were extracted from the 2005 FDA health risk.
Food Code (117). Note: all paragraph and section references within
Food establishment
definitions refer to paragraphs and sections in the 2005 FDA Food
Code.
(1) ‘‘Food establishment’’ means an operation that:
‘‘aw’’ means water activity, which is a measure of the free moisture (a) stores, prepares, packages, serves, vends directly to the
in the food that is available for microbial growth. It is the consumer, or otherwise provides food for human consump-
quotient of the water vapor pressure of the substance divided by tion, such as a restaurant, satellite, or catered feeding
the vapor pressure of pure water at the same temperature, and is location; catering operation if the operation provides food
indicated by the symbol aw. directly to a consumer or to a conveyance used to transport
‘‘Consumer’’ means a person who is a member of the public, takes people; market; vending location; conveyance used to
possession of food, is not functioning in the capacity of an transport people; institution; or food bank; and
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 163

(b) relinquishes possession of food to a consumer directly Packaged

or indirectly through a delivery service such as home
delivery of grocery orders or restaurant takeout orders (1) ‘‘Packaged’’ means bottled, canned, cartoned, securely
or delivery service that is provided by common bagged, or securely wrapped, whether packaged in a food
carriers. establishment or a food processing plant.
(2) ‘‘Food establishment’’ includes: (2) ‘‘Packaged’’ does not include a wrapper, carry-out box, or
(a) an element of the operation such as a transportation other nondurable container used to containerize food with the
vehicle or a central preparation facility that supplies a purpose of facilitating food protection during service and
vending location or satellite feeding location unless the receipt of the food by the consumer.
vending or feeding location is permitted by the regulatory
authority; and
(b) an operation that is conducted in a mobile, stationary, Potentially hazardous food (time/temperature
temporary, or permanent facility or location; where control for safety food)
consumption is on or off the premises; and regardless of
whether there is a charge for the food. (1) ‘‘Potentially hazardous food’’ (time/temperature control for
(3) ‘‘Food establishment’’ does not include: safety food) means a food that requires time/temperature
(a) an establishment that offers only prepackaged foods that control for safety (TCS) to limit pathogenic microorganism
are not potentially hazardous (time/temperature control for growth or toxin formation.
safety) foods; (2) ‘‘Potentially hazardous food’’ (TCS food) includes:
(b) a produce stand that offers only whole, uncut fresh fruits (a) an animal food that is raw or heat treated; a plant food that is
and vegetables; heat treated or consists of raw seed sprouts, cut melons, cut
(c) a food processing plant including those that are located on tomatoes, or mixtures of cut tomatoes that are not modified
the premises of a food establishment; in a way so that they are unable to support pathogenic
(d) a kitchen in a private home if only food that is not microorganism growth or toxin formation or garlic-in-oil
mixtures that are not modified in a way that results in
potentially hazardous (time-temperature control for safety
mixtures that do not support pathogenic microorganism
food) is prepared for sale or service at a function such as a
growth or toxin formation; and
religious or charitable organization’s bake sale if allowed
by law and if the consumer is informed by a clearly visible (b) except as specified in Subparagraph (3)(d) of this
placard at the sales or service location that the food is definition (below), a food that because of the interaction
prepared in a kitchen that is not subject to regulation and of its aw and pH values is designated as Product
inspection by the regulatory authority; Assessment Required (PA) in the Food Code Table A or
Table B of this definition:
(e) an area where food that is prepared as specified in
Subparagraph (3)(d) of this definition (above) is sold or (3) Potentially hazardous food (TCS food) does not include:
offered for human consumption; (a) an air-cooled hard-boiled egg with shell intact or an egg
(f) a kitchen in a private home, such as a small family with shell intact that is not hard boiled but has been
daycare provider; or a bed-and-breakfast operation that pasteurized to destroy all viable salmonellae;
prepares and offers food to guests if the home is owner (b) a food in an unopened hermetically sealed container that is
occupied, the number of available guest bedrooms does commercially processed to achieve and maintain commer-
not exceed 6, breakfast is the only meal offered, the cial sterility under conditions of nonrefrigerated storage
number of guests served does not exceed 18, and the and distribution;
consumer is informed by statements contained in (c) a food that because of its pH or aw value or the interaction of
published advertisements, mailed brochures, and placards aw and pH values is designated as a non-PHF and non-TCS
posted at the registration area that the food is prepared in a food in Table A or Table B of this definition;
kitchen that is not regulated and inspected by the (d) a food that is designated as Product Assessment Required
regulatory authority; or (PA) in Table A or Table B of this definition and has
(g) a private home that receives catered or home-delivered undergone a Product Assessment showing that the
food. growth or toxin formation of pathogenic microorganisms
that are reasonably likely to occur in that food is
Food processing plant precluded due to
(i) intrinsic factors including added or natural character-
istics of the food such as preservatives, antimicrobials,
(1) ‘‘Food processing plant’’ means a commercial operation that
humectants, acidulants, or nutrients;
manufactures, packages, labels, or stores food for human
consumption and provides food for sale or distribution to other (ii) extrinsic factors including environmental or opera-
tional factors that affect the food such as packaging,
business entities such as food processing plants or food
modified atmosphere such as reduced oxygen
establishments.
packaging, shelf life and use, or temperature range
(2) ‘‘Food processing plant’’ does not include a food establish-
of storage and use; or
ment.
(iii) a combination of intrinsic and extrinsic factors; or
‘‘HACCP plan’’ means a written document that delineates the (e) a food that does not support the growth or toxin formation of
formal procedures for following the hazard analysis and critical pathogenic microorganisms in accordance with one of the
control point principles developed by The National Advisory Subparagraphs (3)(a)–(3)(d) of this definition (above) even
Committee on Microbiological Criteria for Foods. though the food may contain a pathogenic microorganism or
‘‘Hazard’’ means a biological, chemical, or physical property that chemical or physical contaminant at a level sufficient to
may cause an unacceptable consumer health risk. cause illness or injury.
164 NACMCF J. Food Prot., Vol. 73, No. 1

TABLE A. Interaction of pH and aw for control of spores in food that has been heat treated to destroy vegetative cells and then packaged a
pH values:
aw values #4.6 .4.6–5.6 .5.6

#0.92 Non-PHF, non-TCS food Non-PHF, non-TCS food Non-PHF, non-TCS food
.0.92–0.95 Non-PHF, non-TCS food Non-PHF, non-TCS food PA
.0.95 Non-PHF, non-TCS food PA PA
a
PHF, potentially hazardous food; TCS food, time/temperature control for safety food; PA, product assessment required.

TABLE B. Interaction of pH and aw for control of vegetative cells and spores in food that has not been heat treated or has been heat
treated but not packaged a
pH values:
aw values ,4.2 4.2–4.6 .4.6–5.0 .5.0

,0.88 Non-PHF, non-TCS food Non-PHF, non-TCS food Non-PHF, non-TCS food Non-PHF, non-TCS food
0.88–0.90 Non-PHF, non-TCS food Non-PHF, non-TCS food Non-PHF, non-TCS food PA***
.0.90–0.92 Non-PHF, non-TCS food Non-PHF, non-TCS food PA PA
.0.92 Non-PHF, non-TCS food PA PA PA
a
PHF, potentially hazardous food; TCS food, time/temperature control for safety food; PA, product assessment required.

Ready-to-eat food dry salami or pepperoni; salt-cured meat and poultry

products such as prosciutto ham, country cured ham, and
(1) ‘‘Ready-to-eat food’’ means food that Parma ham; and dried meat and poultry products such as
(a) is in a form that is edible without additional preparation to jerky or beef sticks; and
achieve food safety, as specified under one of the (i) foods manufactured as specified in 21 CFR Part 113
following parts of the Food Code: Paragraph 3- ‘‘Thermally Processed Low-Acid Foods Packaged in
401.11(A) or (B), Section 3-401.12, or Section 3-402.11, Hermetically Sealed Containers.’’
or as specified in Paragraph 3-401.11(C); or
(b) is a raw or partially cooked animal food and the consumer
Reduced oxygen packaging
is advised as specified in the Food Code, Subparagraphs 3-
(1) ‘‘Reduced oxygen packaging’’ means:
401.11(D)(1) and (2); or
(a) the reduction of the amount of oxygen in a package by
(c) is prepared in accordance with a variance that is granted as
removing oxygen, displacing oxygen and replacing it with
specified in the Food Code, Subparagraphs 3-401.11(D)
another gas or combination of gases, or otherwise controlling
and (3); and
the oxygen content to a level below that normally found in
(d) may receive additional preparation for palatability or
the atmosphere (approximately 21% at sea level); and
aesthetic, epicurean, gastronomic, or culinary purposes.
(b) a process as specified in Subparagraph (1)(a) of this
(2) ‘‘Ready-to-eat food’’ includes:
definition (above) that involves a food for which the
(a) raw animal food that is cooked as specified in the Food hazards Clostridium botulinum or Listeria monocytogenes
Code under Section 3-401.11 or 3-401.12, or frozen as require control in the final packaged form.
specified under Section 3-402.11; (2) ‘‘Reduced oxygen packaging’’ includes:
(b) raw fruits and vegetables that are washed as specified in (a) vacuum packaging, in which air is removed from a
the Food Code under Section 3-302.15; package of food and the package is hermetically sealed so
(c) fruits and vegetables that are cooked for hot holding as that a vacuum remains inside the package;
specified in the Food Code under Section 3-401.13; (b) modified atmosphere packaging, in which the atmosphere
(d) all potentially hazardous food (TCS food) that is cooked to of a package of food is modified so that its composition is
the temperature and time required for the specific food in different from air but the atmosphere may change over
the Food Code under Subpart 3-401 and cooled as time due to the permeability of the packaging material or
specified under Section 3-501.14; the respiration of the food (modified atmosphere packag-
(e) plant food for which further washing, cooking, or other ing includes reduction in the proportion of oxygen, total
processing is not required for food safety and from which replacement of oxygen, or an increase in the proportion of
rinds, peels, husks, or shells, if naturally present, are removed; other gases such as carbon dioxide or nitrogen);
(f) substances derived from plants such as spices, seasonings, (c) controlled atmosphere packaging, in which the atmosphere
and sugar; of a package of food is modified so that until the package is
(g) a bakery item such as bread, cakes, pies, fillings, or icing opened, its composition is different from air, and continuous
for which further cooking is not required for food safety; control of that atmosphere is maintained, such as by using
(h) the following products that are produced in accordance oxygen scavengers, or a combination of total replacement of
with USDA guidelines and that have received a lethality oxygen, nonrespiring food, and impermeable packaging
treatment for pathogens: dry fermented sausages such as material;
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 165

(d) cook chill packaging, in which cooked food is hot filled ‘‘Regulatory authority’’ means the local, state, or federal
into impermeable bags that have the air expelled and are enforcement body or authorized representative having jurisdic-
then sealed or crimped closed, and the bagged food is tion over the food establishment.
rapidly chilled and refrigerated at temperatures that ‘‘Risk’’ means the likelihood that an adverse health effect will
inhibit the growth of psychrotrophic pathogens; or occur within a population as a result of a hazard in a food.
(e) sous vide packaging, in which raw or partially cooked ‘‘Variance’’ means a written document issued by the regulatory
food is placed in a hermetically sealed, impermeable bag, authority that authorizes a modification or waiver of one or more
cooked in the bag, rapidly chilled, and refrigerated at requirements of this code if, in the opinion of the regulatory
temperatures that inhibit the growth of psychrotrophic authority, a health hazard or nuisance will not result from the
pathogens. modification or waiver.
166 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX E. FOOD PRODUCT CHECKLIST: MOZZARELLA SLICES

Evaluation of Mozzarella slices packaged under modified atmosphere packaging and stored at ambient temperatures for up to 2 wk
to enhance sales
Considerations Response Additional comments

1 Determine the purpose of the study

1.a Exempt from time/temperature NAa
control for safety (no refrigeration
required)
1.b Variance from any regulatory Extended out-of-refrigeration storage of
requirements (e.g., holding for modified-atmosphere or vacuum-packaged
.4 h without temperature control) Mozzarella slices for 2 wk; Food Code
variance.
1.c Validate lethality NA (used pasteurized milk in production of
cheese).
1.d Verify that formulation will inhibit NA
microbial growth in refrigerated
foods or under mild temperature
abuse
2 Collect information regarding the product
2.a What are the ingredients? Cheese (pasteurized milk, salt, rennet, starter
cultures)
2.a.1 How consistent are the Ingredients same or similar lot to lot; pH,
ingredients from various sources, moisture, salt can vary slightly but in
lot to lot? accordance with Standard of Identity (SOI)
as defined in 21 CFR 133.155–158 (119).
2.a.2 What are the pH, aw, and pH 5.3–5.4; aw 0.96; proximate analysis at Note: aw is not measured or controlled
proximate analysis (moisture, end of production, 46–52% moisture, 1.0% in typical production but is a function
salt, fat, protein, residual nitrite, NaCl, 30% fat. Homogeneous throughout. of moisture and salt content; moisture
etc.) for product and/or is limited by SOI. Starter culture
individual components? activity (acid development, measured
by pH) is a critical control point.
2.a.3 Do any of these values change Once the cheese is sliced and packaged, the The pH will not increase during the 2-
from preparation to pH may increase from 5.4 to 5.9 during wk holding period at 23uC (73uF) at
consumption? refrigerated storage over a 3-mo period if retail.
lactic acid bacteria starter cultures are killed
by heat used in molding.
2.a.4 If applicable, what are the NA
dimensions of cuts, pieces, etc?
2.a.5 What is the normal microbial Microbial load: lactic acid bacteria starter
load (species, etc.) at the culture 7 log CFU/ml of milk; residual
beginning and end of cultures 2 log CFU/g. Reduction due to
production? heating at 70uC (158uF) during molding
step.
2.a.6 Is there likelihood that Unlikely if produced under Good
contamination may be Manufacturing Practices (GMPs), HACCP
internalized in or distributed using pasteurized milk; contamination by
throughout individual nonsporeformers would be on the surface.
components?
2.b What are the preparation steps?
2.b.1 Is the product an assembled Product is not an assembled nor a
(multicomponent) product? multicomponent product.
2.b.2 Is there a microbial reduction Pasteurization is a validated heat inactivation
step that is validated? What are step for milk used to make the cheese; no kill
the parameters associated with step for surface contamination of the cheese.
the microbial reduction step? Are
there different microbial
reduction steps for different
components?
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 167

APPENDIX E. Continued
Evaluation of Mozzarella slices packaged under modified atmosphere packaging and stored at ambient temperatures for up to 2 wk
to enhance sales
Considerations Response Additional comments

2.b.3 Is there a potential for Yes. Potential for recontamination during

recontamination? slicing and packaging.
2.b.4 What is the variability in Little variability for lethality if prepared
parameters that affect lethality or under GMPs HACCP; growth potential can
growth? vary depending on moisture and pH at the
end of production for high-moisture versus
low-moisture product.
2.b.5 How is the product packaged? After slicing or cutting, slices or blocks will
be vacuum packaged or modified-
atmosphere packaged with nitrogen–carbon
dioxide mix.
2.b.6 Is the product cultured or Product made with starter culture but
fermented? Does it contain populations reduced by heating at 70uC
starter culture intentionally (158uF) for molding step.
added?
2.b.7 Does the product contain NaCl is present but not at inhibitory levels.
antimicrobials (preservatives) or No antimicrobials are added to cheese, but
other ingredients that might be natamycin may be added to the surface of cut
inhibitory, such as spices? or shredded cheese to inhibit mold.
2.c What are the storage conditions?
2.c.1 How will the product be Slices or blocks will be packaged under
displayed for sale? Any changes vacuum or modified atmosphere (nitrogen–
to packaging for display? carbon dioxide mixture) for storage; product
may be displayed unrefrigerated for
increased sales but will otherwise be held
refrigerated to extend shelf life.
2.c.2 What temperatures (and times) During cheese production, milk will be Product quality will deteriorate rapidly
are expected during production, cultured and curd cooked at #40uC (104uF); if temperature exceeds 23uC (73uF).
preparation, and storage or curd will be heated to 70uC (158uF) for However, temperatures as high as
display? molding step; cheese cured at 3uC (37uF) for 27uC (81uF), e.g., during
up to 2 wk and distributed to retailers transportation, will have limited effect
typically at ,7uC (45uF); maximum storage on quality if the time does not exceed
at 23uC (73uF) at retail for 2 wk. 4 h.
2.c.3 What potential is there for Product is unlikely to be stored at
storage or display at temperatures greater than described;
temperatures greater than those temperatures exceeding 30uC (86uF) will
listed above in 2.c.2? result in a significant decrease in product
quality (melting, fat separation).
2.c.4 Are there other hazards that may No, but molds may grow on the surface if
be created by preparation, oxygen is present and when natamycin is not
storage, or display? used.
2.c.5 What is the estimated maximum 9 mo if stored at refrigeration temperatures;
time from production to 2 wk for unrefrigerated storage.
consumption?
2.c.6 What is the time to spoilage or 2 wk unrefrigerated storage if held at 20–
unacceptable quality? 23uC (68–73uF), shorter if temperatures
exceed 23uC; 9 mo refrigerated storage.
3 Determine if product assessment for growth or inactivation is needed
3.a Is a product assessment for growth Yes, product assessment required. Food
necessary based on pH and aw? Code Table B is applicable because of
(See Appendix D, Tables A and B.) potential recontamination and survival of
If yes, also answer 4.e and 5.a. spores. pH . 5.4 and aw 0.96.
3.b Is an inactivation study needed? If No, the purpose of this study is to determine
yes, also answer 4.f and 5.b. if pathogens likely to be present will grow in
the product if stored out of refrigeration;
milk has been previously pasteurized.
168 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX E. Continued
Evaluation of Mozzarella slices packaged under modified atmosphere packaging and stored at ambient temperatures for up to 2 wk
to enhance sales
Considerations Response Additional comments

3.c Are there any regulations Latest edition Food Code for TCS.
applicable for lethality
(inactivation) or TCS (growth)?
4 Determine pathogens of concern to include in the challenge study
4.a According to Table 2 and Given a product pH of 5.4 and an aw of 0.96,
Appendix C, which pathogens are the pathogens of concern are B. cereus, C.
of concern? If food is not seafood, botulinum, pathogenic E. coli, L.
Vibrio spp. may be excluded from monocytogenes, Salmonella, S. aureus,
consideration. Vibrio parahaemolyticus, and V. vulnificus.
4.b Considering the ecology, product, B. cereus spores survive pasteurization; The most likely vegetative pathogens
and epidemiological history, what pathogenic E. coli, L. monocytogenes, to recontaminate the product are L.
pathogens are reasonably likely to Salmonella, and S. aureus from monocytogenes and S. aureus. L.
occur? (Also see Appendix C.) postprocessing handling. monocytogenes is a more likely
pathogen to recontaminate the product
Salmonella has been associated with
due to its ubiquity in the environment.
Mozzarella due to contamination during
S. aureus is a likely contaminant from
production, not postprocess contamination;
workers’ hands.
illness associated with survival, not growth;
no outbreak has been reported with B. Vibrio spp. were excluded from
cereus, L. monocytogenes, or S. aureus (22). consideration since seafood is not a
component.
C. botulinum was excluded from
consideration because the spores are
rare in the ecology of dairy products.
4.c What pathogens are likely to Recontamination can occur as indicated
recontaminate the product after above.
the inactivation step?
4.d Are there any baseline surveys that No.
indicate prevalence of pathogens
for the target product or a related
product?
4.e For growth inhibition (TCS) studies:
4.e.1 Which pathogen(s) will grow the Please see 4.e.2, 4.e.3, and 4.e.4.
fastest? Consider gram positive
versus gram negative; vegetative
microorganisms versus spore
formers. If food is not seafood,
Vibrio may be excluded from
consideration. Use a predictive
model or cite applicable
literature. Consider growth
potential through 1.5 times the
shelf life, if appropriate.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 169

APPENDIX E. Continued
Evaluation of Mozzarella slices packaged under modified atmosphere packaging and stored at ambient temperatures for up to 2 wk
to enhance sales
Considerations Response Additional comments

4.e.2 Predictive model At pH 5.4, aw 0.96, 27uC (80.6uF): PMP 7.0 Modeling was conservatively done at
version 1.1 predicts a 3-log S. aureus the highest expected exposure
increase within 29 h (22 h without lag) temperature.
under aerobic conditions; ComBase Of the likely contaminants, L.
Predictor predicts a 3-log S. aureus increase monocytogenes and S. aureus will grow
within 18 h for the same conditions. For L. fastest at this aw and pH; S. aureus is
monocytogenes, PMP predicts a 1-log generally not a good competitor in
increase within 42 h for the same conditions cheese made with starter cultures, but
(7 h without lag); ComBase Predictor with starter cultures are reduced by heating
5,000 ppm of lactic acid predicts a 1-log L. and molding step. If B. cereus growth
monocytogenes increase within 33 h for the occurred, it would be at a slower rate
same conditions. PMP does not include B. than that of L. monocytogenes or S.
cereus predictions at aw 0.96 but ComBase aureus.
Predictor with 40% CO2 predicts a 3-log B.
cereus increase within 101 h.
4.e.3 Compare choice with literature Stecchini et al. (101) indicated a 5-log
increase of L. monocytogenes during storage
at 5uC (41uF) for 21 days (pH and moisture
not reported).
4.e.4 Any further information on Data presented at the 2003 International
growth or survival? Association for Food Protection Annual
Meeting 2003 (29) on cheese shreds for L.
monocytogenes and Salmonella
demonstrated no growth on low-moisture
Mozzarella stored at 15uC (59uF) for 2 mo
(pH 5.0–5.5; 47% moisture; aw 0.965).
4.e.5 Based on the above analysis, L. monocytogenes and S. aureus.
what challenge organisms are
chosen for growth inhibition
studies?
4.f If inactivation studies: NA
4.f.1 What is the lethal treatment?
(HPP, heat, acid, etc.)
4.f.2 Which microorganisms are most
resistant to the lethal treatment?
(HPP, heat, acid, etc.)
4.f.3 Will the lethality be delivered to
all areas of the product that may
contain the pathogen? Account
for all surface and internalized
contamination.
4.f.4 What is in the formulation that
may affect inactivation?
(Intrinsic factors may contribute
to lethality or resistance: aw,
moisture, salt, pH, fat, etc.)
4.f.5 Are there any data on pathogen
levels in the product?
4.f.6 Is there a regulatory requirement
or policy for log reduction for
this product? Cite requirement.
4.f.7 If there is no regulatory
requirement for log reduction,
use scientific basis for
determining acceptable reduction
(21, 76).
170 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX E. Continued
Evaluation of Mozzarella slices packaged under modified atmosphere packaging and stored at ambient temperatures for up to 2 wk
to enhance sales
Considerations Response Additional comments

4.f.8 Based on the above analysis,

what challenge organisms are
chosen for inactivation studies?
5 Determine appropriate time and sampling intervals for challenge study
5.a For growth inhibition (TCS) 14 days | 1.5 ~ 21 days.
studies, use 1.25–1.5 times the
shelf life as testing time
5.a.1 Maximum time from production Maximum 9 mo if refrigerated; 21 days if
to consumption not refrigerated.
5.a.2 Actual time to spoilage or Point of unacceptable quality: 21 days.
unacceptable quality
5.a.3 For growth inhibition studies, Sample at 0, 1, 2, 3, 4, 7, 14, 21 days. Based on predictive models, growth
determine appropriate sampling could occur within 24–48 h at 27uC
intervals for microbial analysis; (81uF); more than seven sampling
use five to seven (preferred) intervals are appropriate to ensure the
sampling intervals; fewer ability to identify minimum time to
sampling intervals should be growth.
justified, e.g., using results from
similar products.
5.b For inactivation studies determine NA
appropriate sampling points
considering the process and
formulation. Identify populations at
0 time and end of processing;
whenever possible, include
intermediate sampling intervals to
determine death curve.
5.b.1 When inactivation treatments
may result in sublethal injury,
repair and growth of
microorganisms during product
shelf life should be considered
(21).
6 Determine inoculation, storage, and testing procedures
6.a Determine strains for use in study L. monocytogenes and S. aureus will be
(multiple strains for each species tested individually using three-strain
are recommended; consider use of mixtures. Each mixture will include isolates
appropriate food or clinical from cheese or other dairy isolates or clinical
isolates) isolates.
6.b Determine if adaptation is No adaptation necessary; product is low
required for inoculum preparation acid, high aw at ambient temperatures.
6.c Determine method of inoculation Surface inoculation of individual 25-g slices; Using two slices per package with
(surface, mixing, dipping, liquid, two slices per package with inoculum on inoculum in between will retain
dry, etc.) inner surface between the two slices. moisture and provide a worst-case
scenario for growth.
6.d Determine size of inoculum 3 log CFU/g; 0.05 ml (50 ml) per package. Inoculum level is high considering
(populations, e.g., log CFU/g, Each organism will be inoculated likelihood of contamination but will
CFU per package, percentage of independently (separate samples) to avoid allow enumeration by direct plating
inoculum vol/wt or vol/vol) possible antagonistic effect between different and detection of growth and low levels
organisms. of inactivation by formulation during
storage; inoculum volume 1% of
sample size; preliminary data suggest
inoculum does not change pH and aw
appreciably.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 171

APPENDIX E. Continued
Evaluation of Mozzarella slices packaged under modified atmosphere packaging and stored at ambient temperatures for up to 2 wk
to enhance sales
Considerations Response Additional comments

6.e Determine packaging to be used Two inoculated slices will be used per Packaging is the same as commercial
package unit; slices will be packaged with product.
60% nitrogen and 40% carbon dioxide
mixture and sealed; packaging material will
be gas-moisture impermeable.
6.f Determine the incubation 23uC (73uF). 23uC (73uF) is the maximum
temperature for growth inhibition temperature to which the product will
studies or temperature(s) for be exposed without adverse changes in
thermal inactivation studies product quality that would deter
purchase and consumption.
6.g Determine sampling method and Entire sample (two slices) will be mixed in
sample size the bag, and 25-g portions will be removed
for microbial analysis; sample will be
homogenized with equal volume of 0.1%
peptone buffer, and serial dilutions will be
plated on selective agar as appropriate per
FDA BAM methods (115).
6.h How many replicates are needed to Two replicate (unique production) lots using
ensure confidence in data? Does highest moisture and pH combination;
variability in proximate analysis or triplicate samples per testing interval.
production warrant more than two
or three replicate trials? Will
multiple variations of similar
formulations be tested? Has a
statistical design for choosing
formulations been used (block
design, central composite, etc.)?
7 Determine other controls
7.a Is use of surrogates appropriate or Surrogates are not appropriate or necessary.
necessary? If so, justify.
7.b Are uninoculated controls needed Uninoculated controls will be used to
to assess spoilage or competitive monitor growth of molds or yeasts and other
microflora or for other purposes? spoilage microorganisms that can change pH
during testing interval and for proximate
analysis at the beginning of the study.
7.c What other controls are necessary Not required for this study; anticipate growth
(including negative or positive if samples were held for sufficient time.
growth controls)?
8 Determine pass-fail criteria
8.a What are the pass-fail criteria? No more than 1-log increase for L. A 1-log increase in L. monocytogenes
monocytogenes; no more than 3-log increase is considered significant growth, but
for S. aureus. any detectable presence of L.
monocytogenes in a ready-to-eat food
renders the product adulterated.
8.b What are the limits for use of the Data apply only to Mozzarella with the
results? maximum moisture-pH-temperature-time
limits tested in this study.
a
NA, not applicable.
172 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX F. FOOD PRODUCT CHECKLIST: CHOPPED LETTUCE

Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

1 Determine the purpose of the study

1.a Exempt from time/temperature NAa
control for safety (no refrigeration
required)
1.b Variance from any regulatory Yes. The purpose of the study is to allow
requirements (e.g., holding for chopped lettuce to be held out of temperature
.4 h without temperature control) control (at room temperature) for a period of
up to 8 h. This is a salad bar product
consumed on premises. Once the lettuce has
been removed from temperature control it
will be used or discarded within 8 h. Product
will not be re-refrigerated and offered for
service at a later time.
1.c Validate lethality NA
1.d Verify that formulation will inhibit NA
microbial growth in refrigerated
foods or under mild temperature
abuse
2 Collect information regarding the product
2.a What are the ingredients? The single ingredient is heads of whole
Romaine lettuce that are chopped and
washed in water containing a wash water
sanitizer at concentrations specified on the
label.
2.a.1 How consistent are the Total plate count on the product varies
ingredients from various sources, widely from batch to batch.
lot to lot?
2.a.2 What are the pH, aw, and The pH is estimated to be 5.8–6.2 (118).
proximate analysis (moisture, Water activity in iceberg lettuce is 0.995–
salt, fat, protein, residual nitrite, 0.998, and this is assumed to hold true for
etc.) for product and/or Romaine (33). The product is very high in
individual components? water, with minimal amounts of salt, fat, and
protein.
2.a.3 Do any of these values change The pH is not likely to change. The aw may
from preparation to decrease slightly as the product dries out, but
consumption? we have elected to ignore the impact this
drying would have on pathogen growth.
2.a.4 If applicable, what are the Heads arrive whole and are chopped into
dimensions of cuts, pieces, etc.? pieces about 5 by 5 cm.
2.a.5 What is the normal microbial No published data are available on incoming Data presented here were collected
load (species, etc.) at the heads of lettuce. Internal company data on after the lettuce had been washed
beginning and end of the lettuce after chopping shows the and cut.
production? following trends, based on several years of
sample collection, where sample size was
25 g.
Total aerobic plate counts are normally
distributed with a mean of 5.5 log CFU/g
and a standard deviation of 1.5 log CFU/g. S.
aureus has been found in 1 of 50 samples,
Salmonella in 1 of 200 samples, B. cereus in
1 of 10 samples. E. coli is generally absent,
but 1 of 20 samples had .2 log MPN/g. L.
monocytogenes was not detected in tests of
more than 200 samples.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 173

APPENDIX F. Continued
Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

2.a.6 Is there likelihood that Published laboratory data show that

contamination may be internalization in fresh cut lettuce is possible
internalized in or distributed (93). The extent to which this happens under
throughout individual real world conditions is not clear.
components?
2.b What are the preparation steps? Receive lettuce from vendor, store in cooler
until use, remove from cooler, remove and
discard outer leaves, cut off bottom end,
separate remaining leaves and wash them in
water containing wash water sanitizer at
label concentrations, spin to remove excess
water, chop into pieces approximately 5 by
5 cm.
2.b.1 Is the product an assembled The product is not assembled and is not
(multicomponent) product? multicomponent.
2.b.2 Is there a microbial reduction The wash step has been shown to result in a The microbial reduction reported here
step that is validated? What are 1- to 2-log reduction in aerobic plate count. is not considered in the design of the
the parameters associated with challenge study for this product.
the microbial reduction step? Are
there different microbial
reduction steps for different
components?
2.b.3 Is there a potential for There is a slight potential for
recontamination? recontamination. Lettuce is hand chopped in
a food service kitchen environment. Data on
actual product (see above) indicate that S.
aureus may contaminate the product but that
L. monocytogenes does not represent a
significant risk. Employees receive annual
food safety training, and managers are
certified by accredited Food Managers
Certification testing. Standard procedures are
in place to prevent cross-contamination of
this product during preparation.
2.b.4 What is the variability in The product does vary due to normal
parameters that affect lethality or biological variation. The pH and aw values
growth? are very permissive to growth, so variability
is unlikely to influence pathogen growth.
2.b.5 How is the product packaged? The product is not packaged but may be
placed in plastic bins and covered with
plastic wrap for refrigeration prior to display.
2.b.6 Is the product cultured or No.
fermented? Does it contain
starter culture intentionally
added?
2.b.7 Does the product contain There are no antimicrobials, preservatives, or
antimicrobials (preservatives) or other inhibitory ingredients.
other ingredients that might be
inhibitory, such as spices?
2.c What are the storage conditions?
2.c.1 How will the product be Display in open containers on salad bar.
displayed for sale? Any changes
to packaging for display?
174 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX F. Continued
Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

2.c.2 What temperatures (and times) Product is stored below 5uC (41uF) prior to The 8 h starts from the time of
are expected during production, preparation. Preparation takes approximately preparation unless the product will be
preparation, and storage or 2 h per batch and takes place at room rapidly cooled to 5uC (41uF) within
display? temperature (21.1uC, 70uF). Product may 4 h after preparation, in which case
either be covered with plastic wrap and the 8 h starts when the chopped
refrigerated after preparation or placed at product is removed from
room temperature for sale and consumption. refrigeration.
2.c.3 What potential is there for The restaurant is climate controlled. Our data
storage or display at show that the room temperature is usually
temperatures greater than those 21.1uC (70uF) but can in some cases increase
listed above in 2.c.2? to 23.9uC (75uF) for short periods of time.
2.c.4 Are there other hazards that may Recontamination by the consumer during
be created by preparation, serving is possible, but sneeze guards and
storage, or display? tongs are used, as per normal Food Code
practice.
2.c.5 What is the estimated maximum The maximum amount of time the product
time from production to will be out of temperature control is 8 h.
consumption?
2.c.6 What is the time to spoilage or The product is overtly spoiled after 24 h at
unacceptable quality? room temperature.
3 Determine if product assessment for growth or inactivation is needed
3.a Is a product assessment for growth Yes, product assessment is required
necessary based on pH and aw? according to Food Code Table B.
(See Appendix D, Tables A and B.)
If yes, also answer 4.e and 5.a.
3.b Is an inactivation study needed? If No.
yes, also answer 4.f and 5.b.
3.c Are there any regulations The Food Code defines this product as
applicable for lethality requiring temperature control for safety.
(inactivation) or TCS (growth)? There are no requirements for lethality on
this product.
4 Determine pathogens of concern to include in the challenge study
4.a According to Table 2 and Based on pH and aw, B. cereus, C.
Appendix C, which pathogens are botulinum, C. perfringens, L.
of concern? If food is not seafood, monocytogenes, pathogenic E. coli,
Vibrio spp. may be excluded from Salmonella, S. aureus, Shigella spp., and
consideration. Y. enterocolitica should be considered.
4.b Considering the ecology, product, Product testing shows that B. cereus and S.
and epidemiological history, what aureus are present. Epidemiological data
pathogens are reasonably likely to would suggest E. coli O157:H7 as the
occur? (Also see Appendix C.) primary concern, followed by Salmonella
and Shigella. C. botulinum and C.
perfringens were excluded based on the
nature of the finished product (loosely
packed chopped leaves). Although L.
monocytogenes will grow on chopped lettuce
(62), L. monocytogenes was excluded based
on lack of epidemiological evidence (41), as
was Y. enterocolitica and B. cereus.
4.c What pathogens are likely to See response to 2.b.3.
recontaminate the product after
the inactivation step?
4.d Are there any baseline surveys that See response to 2.a.5.
indicate prevalence of pathogens
for the target product or a related
product?
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 175

APPENDIX F. Continued
Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

4.e For growth inhibition (TCS)

studies:
4.e.1 Which pathogen(s) will grow the See response to 4.e.2 and 4.e.3.
fastest? Consider gram positive
versus gram negative; vegetative
microorganisms versus spore
formers. If food is not seafood,
Vibrio spp. may be excluded from
consideration. Use a predictive
model or cite applicable
literature. Consider growth
potential through 1.5 times the
shelf life, if appropriate.
4.e.2 Predictive model A temperature of 21uC (69.8uF), pH 6.2, and
aw 0.995 were assumed for the following
predictions:
When typical lag time values are assumed,
ComBase Predictor shows a 1-log increase
after 6.5 h (E. coli O157:H7), 8.2 h
(Salmonella), 9.4 h (S. aureus), 12 h (L.
monocytogenes), and 18.5 h (Shigella). PMP
7.0 predicted a 1-log increase (including lag)
in 9.9 h (E. coli O157:H7), 8.3 h
(Salmonella), 9.1 h (S. aureus), 9.5 h (L.
monocytogenes), and 15.9 h (Shigella).
When lag time is assumed to be zero,
ComBase Predictor shows a 1-log increase
after 3.4 h (E. coli O157:H7), 3.6 h
(Salmonella), 4.1 h (S. aureus), 4.6 h (L.
monocytogenes), and 10 h (Shigella). PMP
shows a 1-log increase (excluding lag) after
3.6 h (E. coli O157:H7), 3.0 h (Salmonella),
5.6 h (S. aureus), 3.2 h (L. monocytogenes),
and 6.7 h (Shigella).
4.e.3 Compare choice with literature Literature data (four growth rates) for E. coli
O157:H7 growth in cut iceberg lettuce were
extracted from published studies (1, 23, 67).
The four data points were fit to a simple
literature-based model, and growth rate at
21uC (69.8uF) was estimated.
The literature-based model predicted about
0.86-log increase in E. coli O157:H7 after
8 h at 21uC (69.8uF). Note that this
prediction considers only growth rate and
neglects lag time.
4.e.4 Any further information on No.
growth or survival?
4.e.5 Based on the above analysis, Results from the modeling and epidemiology The ComBase modeling analysis
what challenge organisms are show E. coli O157:H7 and Salmonella to above shows that the product could
chosen for growth inhibition represent the greatest risk. Also, modeling be of questionable safety when held
studies? results presented above demonstrate that the at room temperature for 8 h.
growth of the two organisms is similar. Literature-based model suggests that
Challenge studies will be done with E. coli the 8-h holding might be acceptable
O157:H7 due to the greatest epidemiological based on a ,1-log growth.
link to illness.
176 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX F. Continued
Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

A challenge study was justified in

order to provide a more conclusive
answer. The study will be designed to
identify the period of time the growth
remains below 1 log CFU/g.
4.f If inactivation studies: NA
4.f.1 What is the lethal treatment?
(HPP, heat, acid, etc.)
4.f.2 Which microorganisms are most
resistant to the lethal treatment?
(HPP, heat, acid, etc.)
4.f.3 Will the lethality be delivered to
all areas of the product that may
contain the pathogen? Account
for all surface and internalized
contamination.
4.f.4 What is in the formulation that
may affect inactivation?
(Intrinsic factors may contribute
to lethality or resistance: aw,
moisture, salt, pH, fat, etc.)
4.f.5 Are there any data on pathogen
levels in the product?
4.f.6 Is there a regulatory requirement
or policy for log reduction for
this product? Cite requirement.
4.f.7 If there is no regulatory
requirement for log reduction,
use scientific basis for
determining acceptable reduction
(21, 76).
4.f.8 Based on the above analysis,
what challenge organisms are
chosen for inactivation studies?
5 Determine appropriate time and sampling intervals for challenge study
5.a For growth inhibition (TCS) Assuming the product is to be held for 8 h,
studies, use 1.25–1.5 times the the product should be tested for 8 | 1.5 ~
shelf life as testing time 12 h.
5.a.1 Maximum time from production 8 h. See comment to 2.c.2.
to consumption
5.a.2 Actual time to spoilage or Prior data indicate 24 h at room temperature
unacceptable quality results in an unacceptable product.
5.a.3 For growth inhibition studies, Test at 0, 2, 4, 6, 8, 10, 12 h. If cost is an issue, a fewer number of
determine appropriate sampling time points could be evaluated (e.g.,
intervals for microbial analysis; 0, 3, 6, 9, and 12 h).
use five to seven (preferred)
sampling intervals; fewer
sampling intervals should be
justified, e.g., using results from
similar products.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 177

APPENDIX F. Continued
Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

5.b For inactivation studies determine NA

appropriate sampling points
considering the process and
formulation. Identify populations at
0 time and end of processing;
whenever possible, include
intermediate sampling intervals to
determine death curve.
5.b.1 When inactivation treatments
may result in sublethal injury,
repair and growth of
microorganisms during product
shelf life should be considered
(21).
6 Determine inoculation, storage, and testing procedures
6.a Determine strains for use in study A cocktail of marked strains will be used. E. In order to easily enumerate the E.
(multiple strains for each species coli O157:H7 strains will be a combination coli O157:H7 amid a high natural
are recommended; consider use of of human isolates and food isolates background population, the selected
appropriate food or clinical associated with sporadic illnesses or strain will be modified to express an
isolates) outbreaks where leafy greens were appropriate marker (e.g., antibiotic
implicated. resistance, green fluorescent protein).
6.b Determine if adaptation is Adaptation of inoculum not needed. (42)
required for inoculum preparation
6.c Determine method of inoculation Chopped leaves will be spot inoculated on Dip inoculation would add excess
(surface, mixing, dipping, liquid, both uncut surface and cut edges, briefly air moisture that is difficult to remove
dry, etc.) dried in a biosafety cabinet, and then stored without a salad spinner. Salad
at refrigeration temperature until the spinners used to remove moisture
following day. from inoculated lettuce generate
potentially dangerous aerosols in a
laboratory, and it is difficult to
decontaminate the spinner. The
lettuce is refrigerated at 5uC (41uF)
after inoculation to duplicate the
temperature profile of the restaurant
lettuce.
6.d Determine size of inoculum Spot inoculum (approximately 10 ml for a
(population, e.g., log CFU/g, CFU 10-g sample) will be applied in multiple
per package, percentage of (four or more) spots. The target final
inoculum vol/wt or vol/vol) concentration will be 3 log CFU/10-g
sample.
6.e Determine packaging to be used Samples will be stored in loosely sealed
plastic containers.
6.f Determine the incubation Although the product is typically held at
temperature for growth inhibition 21uC (70uF), the product will be incubated at
studies or temperature(s) for 25uC (77uF) to represent the worst-case
thermal inactivation studies condition.
6.g Determine sampling method and Each 10-g sample will be combined with
sample size 90 ml of 0.1% peptone and homogenized for
1 min at high speed prior to dilution and
plating onto appropriate selective media.
178 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX F. Continued
Evaluation to determine the absence of measurable growth (,1 log) of pathogens of concern in chopped lettuce held out of refrigeration
for up to 8 h
Considerations Response Additional comments

6.h How many replicates are needed to Two replicate trials will be conducted, and
ensure confidence in data? Does three samples will be analyzed at each time
variability in proximate analysis or point and plated in duplicate. Each trial will
production warrant more than two use fresh lettuce from a different batch and
or three replicate trials? Will fresh inoculum and will be conducted on a
multiple variations of similar different day.
formulations be tested? Has a
statistical design for choosing
formulations been used (block
design, central composite, etc.)?
7 Determine other controls
7.a Is use of surrogates appropriate or The use of surrogates is not appropriate or
necessary? If so, justify. necessary.
7.b Are uninoculated controls needed Uninoculated controls (one) will be sampled
to assess spoilage or competitive at each time point. They will be plated on
microflora or for other purposes? tryptic soy agar and on the selective agar
used for the study. The visual appearance of
the control lettuce will be described at each
time point.
7.c What other controls are necessary The concentration of E. coli O157:H7 will be
(including negative or positive determined in the freshly prepared inoculum
growth controls)? as well as the freshly inoculated lettuce at
time 0.
8 Determine pass-fail criteria
8.a What are the pass-fail criteria? Less than a 1-log increase for E. coli
O157:H7 at the end of study (12 h).
8.b What are the limits for use of the Results are applicable to similarly prepared Given the results of this study, it may
results? Romaine and iceberg lettuce. These data do not be necessary to conduct full
not apply to finely chopped or shredded studies on other leafy greens, but
Romaine and iceberg lettuce, which are some study is needed before data can
likely to support more rapid growth. be more widely applied.
a
NA, not applicable.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 179

APPENDIX G. FOOD PRODUCT CHECKLIST: MEAT-FILLED PASTRY

Evaluation of display of fully cooked meat-filled pastry for up to 12 h at room temperature
Considerations Response Additional comments

1 Determine the purpose of the study

1.a Exempt from time/temperature NAa Not a shelf-stable product.
control for safety (no refrigeration
required)
1.b Variance from any regulatory Want to hold a fully cooked meat product up Discarded if not served within 12 h.
requirements (e.g., holding for to 12 h at room temperature (assuming
.4 h without temperature control) consumption within 2 h after purchase).
1.c Validate lethality NA Processed in state or federally
inspected food processing
establishment meeting regulatory
cook and cool requirements.
1.d Verify that formulation will inhibit NA
microbial growth in refrigerated
foods or under mild temperature
abuse
2 Collect information regarding the product
2.a What are the ingredients? Ready-to-eat product that contains cooked
ground beef, spices, salt, pastry dough.
2.a.1 How consistent are the ingredients Highly consistent lot to lot. Product specifications in place,
from various sources, lot to lot? produced at a food processing
establishment under GMPs.
2.a.2 What are the pH, aw, and Beef filling: pH 6.2, aw 0.97. If this were an inactivation study,
proximate analysis (moisture, salt, Pastry dough: pH 7.0, aw at interface 0.97; percent fat content may be important;
fat, protein, residual nitrite, etc.) aw at exterior surface 0.75. not relevant for this growth study.
for product and/or individual
components?
2.a.3 Do any of these values change No change of pH. The exterior of the pastry Potential for aw to increase on external
from preparation to consumption? may increase above aw 0.75 the longer the surface if condensate forms between
product is held. the package and pastry surface.
2.a.4 If applicable, what are the NA Component dimensions consistent
dimensions of cuts, pieces, etc? with product specifications.
2.a.5 What is the normal microbial load Vegetative pathogens are inactivated during Fully cooked at processing
(species, etc.) at the beginning and cooking. There is a potential for spore- establishment.
end of production? forming pathogens to survive cooking. There
is a potential for low levels of
microorganisms (aerobic plate count of up to
2 log CFU/g).
2.a.6 Is there likelihood that Yes, spores surviving the cooking process Internal and external vegetative
contamination may be internalized could be distributed throughout the product. pathogens are destroyed by cooking
in or distributed throughout process. However, vegetative
individual components? pathogens could be introduced on
external surfaces during handling or
packaging.
2.b What are the preparation steps?
2.b.1 Is the product an assembled Yes. See product ingredients and
(multicomponent) product? description above.
2.b.2 Is there a microbial reduction step Yes. Adequate lethality and cooling to result Achieving minimum internal
that is validated? What are the in a ready-to-eat product (meets all temperature of 73.9uC (165uF),
parameters associated with the regulatory requirements for cooking and resulting in at least a 6.5-log reduction
microbial reduction step? Are cooling). One cook and cool process for the of Salmonella for lethality, see (108);
there different microbial reduction multicomponent product. for proper cooling, see (109).
steps for different components?
180 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX G. Continued
Evaluation of display of fully cooked meat-filled pastry for up to 12 h at room temperature
Considerations Response Additional comments

2.b.3 Is there a potential for Yes, L. monocytogenes is a potential Although product is individually
recontamination? recontaminant. wrapped, vegetative pathogens could
be introduced on external surfaces
during handling or packaging.
Control of this potential L.
monocytogenes postlethality
contamination is managed per 9 CFR
430 (113).
2.b.4 What is the variability in Limited variability in production of cooked
parameters that affect lethality or product due to controls in a regulated food
growth? processing establishment. Limited variability
during refrigerated distribution and storage
up to the time of display for sale.
2.b.5 How is the product packaged? Individually hand wrapped in the inspected Provides protection from moisture and
establishment in a clear plastic wrap. air.
Wrapped pastries are placed in labeled
boxes.
2.b.6 Is the product cultured or No.
fermented? Does it contain starter
culture intentionally added?
2.b.7 Does the product contain No. Low level of spices and salt would
antimicrobials (preservatives) or not likely be inhibitory to pathogen
other ingredients that might be growth.
inhibitory, such as spices?
2.c What are the storage conditions?
2.c.1 How will the product be displayed Product will remain individually wrapped.
for sale? Any changes to
packaging for display?
2.c.2 What temperatures (and times) are Delivered refrigerated at or below 5uC
expected during production, (41uF) to the retail establishment and kept
preparation, and storage or refrigerated until moved out for display.
display? Held at room temperature for display to
customers. Displayed for up to 12 h at room
temperature: 24uC (75uF). The product is
expected to be consumed or refrigerated
within 2 h of purchase.
2.c.3 What potential is there for storage Higher temperatures are possible if product A separate study may be required for
or display at temperatures greater is heated and displayed under a heat lamp. product stored under a heat lamp.
than those listed above in 2.c.2?
2.c.4 Are there other hazards that may No.
be created by preparation,
storage, or display?
2.c.5 What is the estimated maximum 7 days (refrigerated). Labeled use-by date is 7 days after
time from production to production.
consumption?
2.c.6 What is the time to spoilage or 10 days (refrigerated) or 2 days at ambient Product is to be discarded after 12 h
unacceptable quality? temperatures. of ambient display but may continue
to have an acceptable appearance and
odor at the end of the display period.
Storage under a heat lamp may lead
to unacceptable organoleptic quality.
3 Determine if product assessment for growth or inactivation is needed
3.a Is a product assessment for growth Yes, for beef filling, pH 6.2, aw 0.97. The outer pastry component with an
necessary based on pH and aw? (See aw of 0.75 does not require product
Appendix D, Tables A and B.) If assessment for growth.
yes, also answer 4.e and 5.a.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 181

APPENDIX G. Continued
Evaluation of display of fully cooked meat-filled pastry for up to 12 h at room temperature
Considerations Response Additional comments

3.b Is an inactivation study needed? If No.

yes, also answer 4.f and 5.b.
3.c Are there any regulations applicable Yes, maximum 4-h holding time limit when
for lethality (inactivation) or TCS there are no temperature controls for safety
(growth)? (Food Code).
4 Determine pathogens of concern to include in the challenge study
4.a According to Table 2 and Appendix B. cereus, C. botulinum, C. perfringens, L.
C, which pathogens are of concern? monocytogenes, pathogenic E. coli,
If food is not seafood, Vibrio spp. Salmonella, S. aureus, V. parahaemolyticus,
may be excluded from and V. vulnificus.
consideration.
4.b Considering the ecology, product, C. perfringens, C. botulinum, and B. cereus. Vegetative cells are not a concern due
and epidemiological history, what to USDA FSIS validated cooking
pathogens are reasonably likely to process. Postprocess contamination
occur? (Also see Appendix C.) would be limited to the outside of the
pastry shell, which has very low aw
and would not support growth.
Standard GMPs will also reduce
likelihood of pathogen
recontamination.
4.c What pathogens are likely to Recontamination of meat filling is not likely
recontaminate the product after the because it is encased within a pastry shell.
inactivation step?
4.d Are there any baseline surveys that No, not for meat-filled pastry products.
indicate prevalence of pathogens for
the target product or a related
product?
4.e For growth inhibition (TCS) studies:
4.e.1 Which pathogen(s) will grow the B. cereus based on predictive models (see
fastest? Consider gram positive 4.e.2).
versus gram negative; vegetative
microorganisms versus spore
formers. If food is not seafood,
Vibrio may be excluded from
consideration. Use a predictive
model or cite applicable literature.
Consider growth potential through
1.5 times the shelf life, if
appropriate.
4.e.2 Predictive model Predictive models were used to gauge All modeling included the lag phase.
comparative growth of C. perfringens, C. This was considered appropriate given
botulinum, and B. cereus in the meat filling. that spore-forming organisms require
both germination and outgrowth.
The PMP predicts a 1-log increase of C.
perfringens in approximately 32 h based on Predictive models estimate that B.
pH 6.2, aw 0.983 (lowest aw in program), cereus will grow faster than C.
and 37uC (highest temperature in program). perfringens and that both organisms
ComBase Predictor predicts a 1-log increase would grow faster than C. botulinum.
in approximately 13 h, assuming pH 6.2, aw
0.971, and 37uC (98.6uF).
182 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX G. Continued
Evaluation of display of fully cooked meat-filled pastry for up to 12 h at room temperature
Considerations Response Additional comments

The PMP predicts growth of C. botulinum in

.10 days at 26.7uC (estimated room
temperature of 80uF) based on pH 6.2, aw
0.977 (lowest aw in program). ComBase
predicts a lag time for proteolytic C.
botulinum of about 2 days, assuming pH 6.2,
aw 0.97, and 37uC (98.6uF), and a slightly
shorter lag time for nonproteolytic C.
botulinum at 30uC (86uF) and aw 0.974 (the
least permissive conditions allowed by the
model).
The PMP predicts a 1-log increase in B.
cereus in 5 h at pH 6.2, aw 0.97, and 37uC
(98.6uF) under aerobic conditions and
approximately 12 h under anaerobic
conditions. ComBase predicts a 1-log
increase in approximately 14 h at pH 6.2, aw
0.97, and 37uC (98.6uF) with 0% CO2.
4.e.3 Compare choice with literature Spices are an ingredient in the meat filling.
B. cereus is a known contaminant of spices
(88).
4.e.4 Any further information on growth No.
or survival?
4.e.5 Based on the above analysis, what B. cereus.
challenge organisms are chosen
for growth inhibition studies?
4.f If inactivation studies: NA
4.f.1 What is the lethal treatment?
(HPP, heat, acid, etc.)
4.f.2 Which microorganisms are most
resistant to the lethal treatment?
(HPP, heat, acid, etc.)
4.f.3 Will the lethality be delivered to all
areas of the product that may
contain the pathogen? Account for
all surface and internalized
contamination.
4.f.4 What is in the formulation that
may affect inactivation? (Intrinsic
factors may contribute to lethality
or resistance: aw, moisture, salt,
pH, fat, etc.)
4.f.5 Are there any data on pathogen
levels in the product?
4.f.6 Is there a regulatory requirement
or policy for log reduction for this
product? Cite requirement.
4.f.7 If there is no regulatory
requirement for log reduction, use
scientific basis for determining
acceptable reduction (21, 76).
4.f.8 Based on the above analysis, what
challenge organisms are chosen
for inactivation studies?
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 183

APPENDIX G. Continued
Evaluation of display of fully cooked meat-filled pastry for up to 12 h at room temperature
Considerations Response Additional comments

5 Determine appropriate time and sampling intervals for challenge study

5.a For growth inhibition (TCS) studies, 1.5 | 14 h (target shelf life plus up to 2 h in
use 1.25–1.5 times the shelf life as the hands of the consumer) ~ 21 h.
testing time
5.a.1 Maximum time from production to Refrigerated up to 7 days, 12 h at room
consumption temperature and 2 h to consumption after
leaving the store. May or may not be heated
prior to consumption.
5.a.2 Actual time to spoilage or Expected to have 7-day shelf life in
unacceptable quality refrigerator. At room temperature product
may appear to be spoiled after 2 days.
5.a.3 For growth inhibition studies, Times 0, 14, and 21 h. Due to the very short shelf life of this
determine appropriate sampling product and the fact that predictive
intervals for microbial analysis; models estimated limited growth in
use five to seven (preferred) the time frames of the study (e.g.,
sampling intervals; fewer approximately 1 log unit of growth),
sampling intervals should be sampling times were set at 0, 14 (the
justified, e.g., using results from end of the target shelf life), and 21 h
similar products. (1.5 times the target shelf life).
5.b For inactivation studies determine NA
appropriate sampling points
considering the process and
formulation. Identify populations at
0 time and end of processing;
whenever possible, include
intermediate sampling intervals to
determine death curve.
5.b.1 When inactivation treatments may NA
result in sublethal injury, repair
and growth of microorganisms
during product shelf life should be
considered (21).
6 Determine inoculation, storage, and testing procedures
6.a Determine strains for use in study Use at least three strains of B. cereus. This should include a composite of
(multiple strains for each species clinical strains from foodborne illness
are recommended; consider use of outbreaks as well as isolates from
appropriate food or clinical isolates) food.
6.b Determine if adaptation is required Adaptation not required.
for inoculum preparation
6.c Determine method of inoculation Injection of inoculum into the meat filling Inoculum spores mixed with meat
(surface, mixing, dipping, liquid, through the pastry. filling supplied by manufacturer.
dry, etc.)
6.d Determine size of inoculum 2–3 log CFU/g not to exceed 0.1% of filling Inoculum size is verified with time 0
(populations, e.g., log CFU/g, CFU volume. sample of filling.
per package, percentage of
inoculum vol/wt or vol/vol)
6.e Determine packaging to be used Plastic cellophane wrap.
6.f Determine the incubation Incubate at 30uC (86uF). This represents a reasonable maximum
temperature for growth inhibition ambient temperature.
studies or temperature(s) for
thermal inactivation studies
184 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX G. Continued
Evaluation of display of fully cooked meat-filled pastry for up to 12 h at room temperature
Considerations Response Additional comments

6.g Determine sampling method and Duplicate filled pastries will be sampled
sample size from each of the three replicate lots at each
time point. Each sample in its entirety will be
blended or stomached in a 1:10 dilution of
buffer. Duplicate plate counts will be run for
each sample.
6.h How many replicates are needed to Three replicate production lots are to be If different formulations, three
ensure confidence in data? Does tested, preferably lots made with separate replicates per formulation.
variability in proximate analysis or batches of ingredients or on separate days.
production warrant more than two or
three replicate trials? Will multiple
variations of similar formulations be
tested? Has a statistical design for
choosing formulations been used
(block design, central composite,
etc.)?
7 Determine other controls
7.a Is use of surrogates appropriate or Surrogates are not appropriate or necessary.
necessary? If so, justify.
7.b Are uninoculated controls needed to An uninoculated control is needed for each
assess spoilage or competitive replicate lot to monitor for natural
microflora or for other purposes? contamination.
7.c What other controls are necessary Not required for this study; anticipate growth
(including negative or positive if samples were held for sufficient time.
growth controls)?
8 Determine pass-fail criteria
8.a What are the pass-fail criteria? No more than a 3-log increase of B. cereus. The 3-log increase level selected for
B. cereus is based on the increase
suggested in the IFT report (53).
Some regulatory agencies may
consider a lower log increase to be
appropriate.
8.b What are the limits for use of the These results cannot be applied to pastries
results? held at higher than ambient temperatures,
e.g., holding under a heat lamp.
a
NA, not applicable.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 185

APPENDIX H. FOOD PRODUCT CHECKLIST: LEMON MERINGUE PIE

Validate that the formulation of lemon meringue pie will inhibit pathogen growth under nonrefrigerated conditions
Considerations Response Additional comments

1 Determine the purpose of the study

1.a Exempt from time/temperature Exempt from time/temperature control after Labeled shelf life of 3 days.
control for safety (no refrigeration opening.
required)
1.b Variance from any regulatory NAa
requirements (e.g., holding for
.4 h without temperature control)
1.c Validate lethality NA
1.d Verify that formulation will inhibit NA
microbial growth in refrigerated
foods or under mild temperature
abuse
2 Collect information regarding the product
2.a What are the ingredients? Pie crust: flour, shortening, water, salt.
Filling: water, sugar, modified food starch,
corn syrup solids, margarine, lemon juice
solids, high fructose corn syrup, sodium
citrate, agar agar, potassium sorbate, natural
flavor, locust bean gum, artificial color
(FD&C yellow no. 5).
Meringue: unpasteurized egg whites, sugar,
cream of tartar.
2.a.1 How consistent are the ingredients Very consistent, same or similar lot to lot.
from various sources, lot to lot?
2.a.2 What are the pH, aw, and Baked crust: pH 6.2, aw 0.45. Values are after baking.
proximate analysis (moisture, salt,
Cooked filling: pH 4.2, aw 0.88.
fat, protein, residual nitrite, etc.)
for product and/or individual Meringue: pH 4.6, aw 0.93.
components?
2.a.3 Do any of these values change No.
from preparation to consumption?
2.a.4 If applicable, what are the NA
dimensions of cuts, pieces, etc?
2.a.5 What is the normal microbial load After baking, aerobic plate count of ,10
(species, etc.) at the beginning and CFU/g.
end of production?
2.a.6 Is there likelihood that Yes, during slicing the contamination may
contamination may be internalized occur along the sliced edge of all three
in or distributed throughout components (crust, filling, and meringue).
individual components?
2.b What are the preparation steps? Mix dough, sheet, form, bake. Cook the Product is prepared in a commercial
filling to set the starch, fill the baked crust, manufacturing facility, cooled to room
cool to ambient temperature, spread temperature, packaged, and shipped at
meringue evenly over filling and bake. Cool ambient temperature.
to ambient temperature, package.
2.b.1 Is the product an assembled Yes.
(multicomponent) product?
2.b.2 Is there a microbial reduction step All three components have heat inactivation
that is validated? What are the steps.
parameters associated with the
microbial reduction step? Are
there different microbial reduction
steps for different components?
186 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX H. Continued
Validate that the formulation of lemon meringue pie will inhibit pathogen growth under nonrefrigerated conditions
Considerations Response Additional comments

2.b.3 Is there a potential for Yes, contamination may occur after opening
recontamination? and slicing.
2.b.4 What is the variability in Low variability.
parameters that affect lethality or
growth?
2.b.5 How is the product packaged? Paperboard box or plastic dome over an
aluminum pie plate.
2.b.6 Is the product cultured or No.
fermented? Does it contain starter
culture intentionally added?
2.b.7 Does the product contain Sodium citrate and potassium sorbate in the
antimicrobials (preservatives) or filling.
other ingredients that might be
inhibitory, such as spices?
2.c What are the storage conditions?
2.c.1 How will the product be displayed Refrigerated or ambient, no change to
for sale? Any changes to packaging.
packaging for display?
2.c.2 What temperatures (and times) are Cooled to ambient temperature after baking, Unacceptable quality at 5 days.
expected during production, shipped and displayed at ambient
preparation, and storage or temperatures 20–35uC (68–95uF) until the
display? end of labeled shelf life of 3 days.
2.c.3 What potential is there for storage Unlikely.
or display at temperatures greater
than those listed above in 2.c.2?
2.c.4 Are there other hazards that may No. However, hazards may be introduced
be created by preparation, during slicing and serving.
storage, or display?
2.c.5 What is the estimated maximum 3 days.
time from production to
consumption?
2.c.6 What is the time to spoilage or 5 days.
unacceptable quality?
3 Determine if product assessment for growth or inactivation is needed
3.a Is a product assessment for growth Yes, according to Table B, a product
necessary based on pH and aw? (See assessment is required for the meringue
Appendix D, Tables A and B.) If component but not the crust or the filling.
yes, also answer 4.e and 5.a.
3.b Is an inactivation study needed? If Yes, a separate inactivation study is being
yes, also answer 4.f and 5.b. conducted on the meringue.
3.c Are there any regulations applicable Yes, purpose of study is to get a variance
for lethality (inactivation) or TCS from need for time/temperature control for
(growth)? safety.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 187

APPENDIX H. Continued
Validate that the formulation of lemon meringue pie will inhibit pathogen growth under nonrefrigerated conditions
Considerations Response Additional comments

4 Determine pathogens of concern to include in the challenge study

4.a According to Table 2 and Appendix From Appendix C, pathogens of concern in
C, which pathogens are of concern? egg products are Salmonella and Listeria.
If food is not seafood, Vibrio spp. From Table 2, for pH 4.6 and aw 0.94, L.
may be excluded from monocytogenes and Salmonella would be the
consideration. organisms of concern.
4.b Considering the ecology, product, Salmonella and L. monocytogenes are
and epidemiological history, what known to be in retail and food service
pathogens are reasonably likely to environments.
occur? (Also see Appendix C.)
4.c What pathogens are likely to Salmonella and L. monocytogenes.
recontaminate the product after the
inactivation step?
4.d Are there any baseline surveys that There are studies documenting the presence
indicate prevalence of pathogens for of Listeria in the retail deli environment
the target product or a related (68).
product?
4.e For growth inhibition (TCS)
studies:
4.e.1 Which pathogen(s) will grow the See 4.e.2.
fastest? Consider gram positive
versus gram negative; vegetative
microorganisms versus spore
formers. If food is not seafood,
Vibrio may be excluded from
consideration. Use a predictive
model or cite applicable literature.
Consider growth potential through
1.5 times the shelf life, if
appropriate.
4.e.2 Predictive model The PMP indicates that L. monocytogenes
will not grow, but the model does not go
below pH 5.6 for Salmonella, so growth rate
under pH and aw conditions in meringue is
unknown.
4.e.3 Compare choice with literature Literature shows Salmonella and L.
monocytogenes growth can occur at pH 4.6;
most of these studies were in laboratory
media with high aw.
4.e.4 Any further information on growth No.
or survival?
4.e.5 Based on the above analysis, what Salmonella. Since we are unable to determine the
challenge organisms are chosen likelihood of growth of Salmonella
for growth inhibition studies? from predictive models or from the
literature, this organism was chosen
for a challenge study.
4.f If inactivation studies: NA
4.f.1 What is the lethal treatment?
(HPP, heat, acid, etc.)
4.f.2 Which microorganisms are most
resistant to the lethal treatment?
(HPP, heat, acid, etc.)
4.f.3 Will the lethality be delivered to all
areas of the product that may
contain the pathogen? Account for
all surface and internalized
contamination.
188 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX H. Continued
Validate that the formulation of lemon meringue pie will inhibit pathogen growth under nonrefrigerated conditions
Considerations Response Additional comments

4.f.4 What is in the formulation that

may affect inactivation? (Intrinsic
factors may contribute to lethality
or resistance: aw, moisture, salt,
pH, fat, etc.)
4.f.5 Are there any data on pathogen
levels in the product?
4.f.6 Is there a regulatory requirement
or policy for log reduction for this
product? Cite requirement.
4.f.7 If there is no regulatory
requirement for log reduction, use
scientific basis for determining
acceptable reduction (21, 76).
4.f.8 Based on the above analysis, what
challenge organisms are chosen
for inactivation studies?
5 Determine appropriate time and sampling intervals for challenge study
5.a For growth inhibition (TCS) studies, 3 days | 1.5 ~ 4.5 days (round to 5 days).
use 1.25–1.5 times the shelf life as
testing time
5.a.1 Maximum time from production to 3 days.
consumption
5.a.2 Actual time to spoilage or 5 days.
unacceptable quality
5.a.3 For growth inhibition studies, Sample at time 0, then at days 1, 2, 3, 4,
determine appropriate sampling and 5.
intervals for microbial analysis;
use five to seven (preferred)
sampling intervals; fewer
sampling intervals should be
justified, e.g., using results from
similar products.
5.b For inactivation studies determine NA
appropriate sampling points
considering the process and
formulation. Identify populations at
0 time and end of processing;
whenever possible, include
intermediate sampling intervals to
determine death curve.
5.b.1 When inactivation treatments may NA
result in sublethal injury, repair
and growth of microorganisms
during product shelf life should be
considered (21).
6 Determine inoculation, storage, and testing procedures
6.a Determine strains for use in study A mixture of at least five strains of
(multiple strains for each species Salmonella isolated from eggs or egg
are recommended; consider use of products and including at least one
appropriate food or clinical isolates) Salmonella Enteritidis strain isolated from
clinical or egg samples associated with
outbreaks.
6.b Determine if adaptation is required Not necessary for this study.
for inoculum preparation
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 189

APPENDIX H. Continued
Validate that the formulation of lemon meringue pie will inhibit pathogen growth under nonrefrigerated conditions
Considerations Response Additional comments

6.c Determine method of inoculation Inoculate the cut face of the meringue for A preliminary study should be
(surface, mixing, dipping, liquid, single slices of pie by distributing 25 ml of conducted to ensure that the dye is
dry, etc.) liquid inoculum from the filling-meringue not inhibitory to Salmonella unless
interface to the surface of the meringue; a previously documented in the
noninhibitory dye will be added to inoculum scientific literature.
to facilitate identification of the sampling
area.
6.d Determine size of inoculum Target 2–3 log CFU per site for each slice.
(populations, e.g., log CFU/g, CFU
per package, percentage of
inoculum vol/wt or vol/vol)
6.e Determine packaging to be used Packed in a ventilated plastic container that
prevents contamination of the slice but
allows exchange of air.
6.f Determine the incubation 35uC (95uF).
temperature for growth inhibition
studies or temperature(s) for
thermal inactivation studies
6.g Determine sampling method and For each sample, the entire slice
sample size (approximately 100 g) will be placed in a
sterile plastic sampling bag. The sample will
be homogenized with an equal volume of
0.1% peptone buffer, and serial dilutions
will be plated on appropriate Salmonella
selective agar using the FDA BAM method
(7).
6.h How many replicates are needed to Three replicate trials, pies made from Three replicate trials with three
ensure confidence in data? Does different batches of ingredients for each trial, samples at each interval were chosen
variability in proximate analysis or triplicate slices per trial. Separate slices will because of the inherent variability of
production warrant more than two or be assayed for each sampling interval (n ~ 9 inoculating individual slices for each
three replicate trials? Will multiple for each sampling interval). sampling time interval.
variations of similar formulations be
tested? Has a statistical design for
choosing formulations been used
(block design, central composite,
etc.)?
7 Determine other controls
7.a Is use of surrogates appropriate or No surrogates are necessary.
necessary? If so, justify.
7.b Are uninoculated controls needed to Yes. An uninoculated pie for aerobic plate
assess spoilage or competitive counts and counts of yeasts and
microflora or for other purposes? molds.
7.c What other controls are necessary NA
(including negative or positive
growth controls)?
8 Determine pass-fail criteria
8.a What are the pass-fail criteria? Must show ,1-log growth of Salmonella
throughout the 5-day testing period.
8.b What are the limits for use of the Would be applicable only to meringue pies
results? with very similar pH and aw in both the
filling and the meringue.
a
NA, not applicable.
190 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX I. FOOD PRODUCT CHECKLIST: MERINGUE TOPPING

Evaluation of the adequacy of thermal inactivation of pathogens of concern in meringue topping for lemon meringue pie
Considerations Response Additional comments

1 Determine the purpose of the study

1.a Exempt from time/temperature NAa
control for safety (no refrigeration
required)
1.b Variance from any regulatory NA
requirements (e.g., holding for
.4 h without temperature control)
1.c Validate lethality Validate lethality of meringue topping heat
treatment (baking).
1.d Verify that formulation will inhibit NA
microbial growth in refrigerated
foods or under mild temperature
abuse
2 Collect information regarding the product
2.a What are the ingredients? Pie crust: flour, shortening, water, salt. 2007 supplement to the 2005 Food
Code specifies that pasteurized egg
Filling: water, sugar, modified food starch,
white be used for meringue. This is
corn syrup solids, margarine, lemon juice
an example illustrating an
solids, high fructose corn syrup, sodium
inactivation challenge study and
citrate, agar agar, potassium sorbate, natural
could potentially be used to obtain
flavor, locust bean gum, artificial color
a variance for the use of
(FD&C yellow no. 5).
unpasteurized egg whites.
Meringue: unpasteurized egg whites, sugar,
cream of tartar.
2.a.1 How consistent are the ingredients Very consistent, same or similar lot to lot.
from various sources, lot to lot?
2.a.2 What are the pH, aw, and Baked crust: pH 6.2, aw 0.45.
proximate analysis (moisture, salt,
Cooked filling: pH 4.2, aw 0.88.
fat, protein, residual nitrite, etc.)
for product and/or individual Raw meringue: pH 4.6, aw 0.93.
components?
2.a.3 Do any of these values change aw may decrease for the meringue at the
from preparation to consumption? exposed surface.
2.a.4 If applicable, what are the NA
dimensions of cuts, pieces, etc?
2.a.5 What is the normal microbial load Before cooking: aerobic plate count ,1,000
(species, etc.) at the beginning and CFU/g.
end of production?
After baking: aerobic plate count ,10
CFU/g.
2.a.6 Is there likelihood that Salmonella may be present in unpasteurized
contamination may be internalized egg whites used for meringue topping.
in or distributed throughout
individual components?
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 191

APPENDIX I. Continued
Evaluation of the adequacy of thermal inactivation of pathogens of concern in meringue topping for lemon meringue pie
Considerations Response Additional comments

2.b What are the preparation steps? Mix dough, sheet, form, bake. Cook the Product is prepared in a commercial
filling to set the starch, fill the baked crust, manufacturing facility, cooled to
cool to ambient temperature, spread room temperature, packaged and
meringue evenly over filling and bake. Cool shipped at ambient temperature.
to ambient temperature, package.
2.b.1 Is the product an assembled Yes.
(multicomponent) product?
2.b.2 Is there a microbial reduction step Purpose of this study. All three components
that is validated? What are the (crust, filling, meringue) have heat
parameters associated with the inactivation steps, but the crust gets heat
microbial reduction step? Are treated twice, the filling gets heat treated
there different microbial reduction twice and there is an added inactivation due
steps for different components? to the pH, and the meringue gets heat treated
once.
2.b.3 Is there a potential for Very unlikely, controlled through GMPs at
recontamination? the commercial manufacturing facility.
2.b.4 What is the variability in Low variability.
parameters that affect lethality or
growth?
2.b.5 How is the product packaged? Paperboard box or plastic dome over an
aluminum pie plate.
2.b.6 Is the product cultured or No.
fermented? Does it contain starter
culture intentionally added?
2.b.7 Does the product contain Sodium citrate and potassium sorbate in the
antimicrobials (preservatives) or filling.
other ingredients that might be
inhibitory, such as spices?
2.c What are the storage conditions?
2.c.1 How will the product be displayed Refrigerated or ambient, no change to
for sale? Any changes to packaging.
packaging for display?
2.c.2 What temperatures (and times) are Crust cook: 85uC (185uF) final temperature, The cook time for the meringue is
expected during production, 15 min total cooking time in 176.7uC based on the time required to
preparation, and storage or (350uF) nonhumidified oven. achieve the characteristic browning.
display?
Filling set: 90.6uC (195uF) for 10 min.
Meringue set: 15 min total cooking time in a
preheated 176.7uC (350uF) oven.
2.c.3 What potential is there for storage NA; purpose of this study is to validate
or display at temperatures greater microbial reduction.
than those listed above in 2.c.2?
2.c.4 Are there other hazards that may No.
be created by preparation,
storage, or display?
2.c.5 What is the estimated maximum NA
time from production to
consumption?
2.c.6 What is the time to spoilage or NA
unacceptable quality?
192 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX I. Continued
Evaluation of the adequacy of thermal inactivation of pathogens of concern in meringue topping for lemon meringue pie
Considerations Response Additional comments

3 Determine if product assessment for growth or inactivation is needed

3.a Is a product assessment for growth NA
necessary based on pH and aw? (See
Appendix D, Tables A and B.) If
yes, also answer 4.e and 5.a.
3.b Is an inactivation study needed? If Yes.
yes, also answer 4.f and 5.b.
3.c Are there any regulations applicable 2007 supplement to the 2005 Food Code
for lethality (inactivation) or TCS specifies that pasteurized egg white be used
(growth)? for meringue.
4 Determine pathogens of concern to include in the challenge study
4.a According to Table 2 and Appendix From Appendix C, pathogens of concern in
C, which pathogens are of concern? egg products are B. cereus, Salmonella, and
If food is not seafood, Vibrio spp. L. monocytogenes.
may be excluded from
consideration.
4.b Considering the ecology, product, B. cereus spores would be expected to For this study, we are concerned
and epidemiological history, what survive the heat treatment but would not with pathogen survival, not growth;
pathogens are reasonably likely to grow out due to the aw of meringue. therefore, a pathogen with a low
occur? (Also see Appendix C.) Salmonella is more prevalent and present in infectious dose was chosen as the
higher numbers than L. monocytogenes in challenge organism.
unpasteurized liquid egg products.
Salmonella has been associated with
numerous products containing undercooked
egg ingredients, including meringue pie (72).
4.c What pathogens are likely to NA The objective of the study is to
recontaminate the product after the evaluate inactivation and not
inactivation step? recontamination.
4.d Are there any baseline surveys that In FSIS (111) risk assessments regarding For an inactivation study,
indicate prevalence of pathogens for eggs and egg products, Salmonella estimates quantitative levels are more
the target product or a related in unpasteurized liquid whole egg product important than qualitative
product? were 0 and 100 cells per ml, on average, in prevalence, as levels help estimate
pooled product from multiple eggs (see Fig. the amount of kill necessary to
3-45 from the FSIS risk assessment). In protect public health.
addition, estimates in unpasteurized liquid
egg white product were less than 10 cells per
ml, on average, in pooled product from
multiple eggs (see Fig. 3-44 from the FSIS
risk assessment). On some occasions MPNs
of Salmonella can exceed 1,000 CFU/ml for
both unpasteurized liquid whole egg and
liquid egg white products, but these events
appear to be rare. The FSIS identified
Salmonella Enteritidis in some but not all
samples collected and analyzed. Thus, the
Salmonella Enteritidis levels used in the
inactivation study described below would
represent a worst case.
4.e For growth inhibition (TCS) NA
studies:
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 193

APPENDIX I. Continued
Evaluation of the adequacy of thermal inactivation of pathogens of concern in meringue topping for lemon meringue pie
Considerations Response Additional comments

4.e.1 Which pathogen(s) will grow the

fastest? Consider gram positive
versus gram negative; vegetative
microorganisms versus spore
formers. If food is not seafood,
Vibrio may be excluded from
consideration. Use a predictive
model or cite applicable literature.
Consider growth potential through
1.5 times the shelf life, if
appropriate.
4.e.2 Predictive model
4.e.3 Compare choice with literature
4.e.4 Any further information on growth
or survival?
4.e.5 Based on the above analysis, what
challenge organisms are chosen
for growth inhibition studies?
4.f If inactivation studies:
4.f.1 What is the lethal treatment? Heat.
(HPP, heat, acid, etc.)
4.f.2 Which microorganisms are most Sporeformers would be most heat resistant
resistant to the lethal treatment? but have been eliminated as a risk due to
(HPP, heat, acid, etc.) inability to grow in the product. Salmonella
Enteritidis is the most appropriate challenge
organism.
4.f.3 Will the lethality be delivered to all Yes.
areas of the product that may
contain the pathogen? Account for
all surface and internalized
contamination.
4.f.4 What is in the formulation that Relatively high sugar content of the
may affect inactivation? (Intrinsic meringue will reduce the aw, potentially
factors may contribute to lethality leading to increased heat resistance.
or resistance: aw, moisture, salt,
pH, fat, etc.)
4.f.5 Are there any data on pathogen See 4.d.
levels in the product?
4.f.6 Is there a regulatory requirement No.
or policy for log reduction for this
product? Cite requirement.
4.f.7 If there is no regulatory Using 2-log reduction as worst case and
requirement for log reduction, use building in a 2-log margin of safety, the
scientific basis for determining target reduction is 4-log units.
acceptable reduction (21, 76).
4.f.8 Based on the above analysis, what Salmonella Enteritidis.
challenge organisms are chosen
for inactivation studies?
5 Determine appropriate time and sampling intervals for challenge study
5.a For growth inhibition (TCS) studies, NA
use 1.25–1.5 times the shelf life as
testing time
5.a.1 Maximum time from production to
consumption
194 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX I. Continued
Evaluation of the adequacy of thermal inactivation of pathogens of concern in meringue topping for lemon meringue pie
Considerations Response Additional comments

5.a.2 Actual time to spoilage or

unacceptable quality
5.a.3 For growth inhibition studies,
determine appropriate sampling
intervals for microbial analysis;
use five to seven (preferred)
sampling intervals; fewer
sampling intervals should be
justified, e.g., using results from
similar products.
5.b For inactivation studies determine Times 0 and 15 min. Sampling at more than three time
appropriate sampling points points would allow a D-value to be
considering the process and calculated that may be of use in
formulation. Identify populations at further defining the cook process
0 time and end of processing; but is not in the current study
whenever possible, include design.
intermediate sampling intervals to
determine death curve.
5.b.1 When inactivation treatments may NA
result in sublethal injury, repair
and growth of microorganisms
during product shelf life should be
considered (21).
6 Determine inoculation, storage, and testing procedures
6.a Determine strains for use in study A mixture of at least five strains of
(multiple strains for each species Salmonella isolated from eggs or egg
are recommended; consider use of products and including at least one
appropriate food or clinical isolates) Salmonella Enteritidis isolated from clinical
or egg samples associated with outbreaks.
6.b Determine if adaptation is required Not necessary for this study.
for inoculum preparation
6.c Determine method of inoculation Will mix concentrated, washed inoculum Because of the potential for aerosols,
(surface, mixing, dipping, liquid, into the egg whites before beating to a beating of egg whites should be
dry, etc.) meringue. Finished meringue will be done in a biological safety cabinet.
weighed and spread evenly over the surface The assumption is also made that
of a cooked lemon filling that has been the filling will not be chilled below
cooled to room temperature. The ambient temperature prior to
concentration of Salmonella in the uncooked application of the meringue.
finished meringue will be determined as
described below.
6.d Determine size of inoculum A final target level of at least 4 log CFU per
(populations, e.g., log CFU/g, CFU pie.
per package, percentage of
inoculum vol/wt or vol/vol)
6.e Determine packaging to be used NA
6.f Determine the incubation
temperature for growth inhibition
studies or temperature(s) for
thermal inactivation studies
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 195

APPENDIX I. Continued
Evaluation of the adequacy of thermal inactivation of pathogens of concern in meringue topping for lemon meringue pie
Considerations Response Additional comments

6.g Determine sampling method and The sample size will be the whole meringue
sample size from a single pie. The whole meringue will
be enriched for Salmonella using the BAM
method (7). In addition, at T0 one pie will be
used to determine the initial number of
Salmonella recovered in the meringue prior
to baking by removing the meringue from
the pie, mixing thoroughly, and taking three
10-g samples of the meringue for
enumeration.
6.h How many replicates are needed to There will be three replicate trials. Each trial
ensure confidence in data? Does will consist of three inoculated baked pies
variability in proximate analysis or plus one T0 unbaked pie; thus, a total of 12
production warrant more than two or pies will be needed for the study.
three replicate trials? Will multiple
variations of similar formulations be
tested? Has a statistical design for
choosing formulations been used
(block design, central composite,
etc.)?
7 Determine other controls
7.a Is use of surrogates appropriate or No surrogates are appropriate.
necessary? If so, justify.
7.b Are uninoculated controls needed to NA
assess spoilage or competitive
microflora or for other purposes?
7.c What other controls are necessary Temperature will be verified in several
(including negative or positive places in the oven during baking.
growth controls)?
8 Determine pass-fail criteria
8.a What are the pass-fail criteria? Must achieve .4-log reduction within Based on nondetection of Salmonella
15 min in a 176.7uC (350uF) oven. upon enrichment of the meringue.
8.b What are the limits for use of the Limitations of this study include the volume
results? and depth of the meringue on the pie. The
temperature of the filling may impact the
results. The data will apply for longer but not
shorter cook times at the oven temperature
indicated. These data could apply to other
types of filling.
a
NA, not applicable.
196 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX J. FOOD PRODUCT CHECKLIST: BAKED CHEESE PIZZA

Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

1 Determine the purpose of the study.

1.a Exempt from time/temperature NAa
control for safety (no refrigeration
required)
1.b Variance from any regulatory Holding of baked cheese (sliced) pizza
requirements (e.g., holding for .4 h without refrigeration for up to 8 h.
without temperature control)
1.c Validate lethality NA Assembled at another facility (central
commissary) and held refrigerated
until baked at retail store.
1.d Verify that formulation will inhibit NA
microbial growth in refrigerated
foods or under mild temperature
abuse
2 Collect information regarding the product
2.a What are the ingredients? Pizza crust: flour, salt, shortening.
Cheese: pasteurized milk, salt, rennet,
starter cultures.
Tomato sauce: canned tomato paste, water,
oregano, basil, garlic.
2.a.1 How consistent are the ingredients Ingredients same; relatively consistent
from various sources, lot to lot? composition of sauce and cheese but
preparation of pizza may vary considerably
with respect to amounts of ingredients.
2.a.2 What are the pH, aw, and proximate Proximate analysis: 10% protein, 8% fat,
analysis (moisture, salt, fat, protein, 1% salt, 46–49% moisture
residual nitrite, etc.) for product
pH: crust, 6.8; sauce, 4.5; cheese, 5.4
and/or individual components?
aw: crust, 0.70; sauce, 0.98; cheese, 0.95–
0.96.
2.a.3 Do any of these values change from No.
preparation to consumption?
2.a.4 If applicable, what are the NA
dimensions of cuts, pieces, etc?
2.a.5 What is the normal microbial load Microbial load: ,100 CFU/g after baking;
(species, etc.) at the beginning and primarily spore-forming microorganisms
end of production? potentially including B. cereus, C.
perfringens, and C. botulinum.
2.a.6 Is there likelihood that Yes, each of the components is likely to
contamination may be internalized contain pathogenic bacterial spores.
in or distributed throughout
individual components?
2.b What are the preparation steps?
2.b.1 Is the product an assembled Yes. Product consists of a thin crust
(multicomponent) product? covered with sauce and topped with a layer
of cheese.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 197

APPENDIX J. Continued
Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

2.b.2 Is there a microbial reduction step No further kill step after pizza is baked.
that is validated? What are the Baking has been validated to eliminate all
parameters associated with the vegetative bacterial pathogens.
microbial reduction step? Are there
different microbial reduction steps
for different components?
2.b.3 Is there a potential for Yes, there is potential for recontamination Due to the large surface area, the
recontamination? once the pizza cools from the baking pizza is expected to cool to room
process and is handled by food service temperature rapidly after baking;
workers. therefore, growth of C. perfringens
is not a concern.
2.b.4 What is the variability in parameters Little variability in parameters that affect
that affect lethality or growth? lethality if baked to an endpoint of visual
doneness.
2.b.5 How is the product packaged? Not packaged. Trays containing pizzas are
shipped from commissary to food service
establishment.
2.b.6 Is the product cultured or No.
fermented? Does it contain starter
culture intentionally added?
2.b.7 Does the product contain NaCl is present but not at inhibitory levels.
antimicrobials (preservatives) or No antimicrobials are added.
other ingredients that might be
inhibitory, such as spices?
2.c What are the storage conditions?
2.c.1 How will the product be displayed Held in an enclosed display cabinet where
for sale? Any changes to packaging the maximum temperature is 30uC (86uF).
for display?
2.c.2 What temperatures (and times) are Only the baked product holding
expected during production, temperature is relevant; in this instance,
preparation, and storage or 30uC (86uF) for up to 8 h at retail.
display?
2.c.3 What potential is there for storage There is the possibility that product will be
or display at temperatures greater held at temperatures as great as 40uC
than those listed above in 2.c.2? (104uF), but quality deterioration would
occur in less than 8 h.
2.c.4 Are there other hazards that may be L. monocytogenes contamination from the
created by preparation, storage, or environment may occur; handling can
display? result in contamination with S. aureus.
2.c.5 What is the estimated maximum time Maximum 8 h store display; 2 h from sale
from production to consumption? to consumption (total of 10 h).
2.c.6 What is the time to spoilage or Product is of acceptable quality for the
unacceptable quality? duration of the study, even though it may
appear to be dried out. Little is known
about unacceptable quality parameters for
pizza and what consumers may determine
to be of unacceptable quality. In
accordance with general food safety
practices, food should be consumed or
refrigerated within 2 h of purchase.
198 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX J. Continued
Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

3 Determine if product assessment for growth or inactivation is needed

3.a Is a product assessment for growth Yes, product assessment required; Food Multicomponent product. Crust has
necessary based on pH and aw? (See Code Table B is applicable because of low aw, but aw will be increased by
Appendix D, Tables A and B.) If yes, potential recontamination and survival of moisture from sauce. Sauce also
also answer 4.e and 5.a. spores; pH . 5.0 and aw . 0.92 in parts of lowers the pH of the crust. Moisture
the product and product not protected from loss of product occurs over time.
recontamination.
3.b Is an inactivation study needed? If No, the purpose of this study is to
yes, also answer 4.f and 5.b. determine if pathogens likely to be present
will grow in the product if stored out of
refrigeration.
3.c Are there any regulations applicable Latest edition of the Food Code for TCS.
for lethality (inactivation) or TCS
(growth)?
4 Determine pathogens of concern to include in the challenge study
4.a According to Table 2 and Appendix Based on a measured pH of 5.4 and a Vibrio spp. were excluded from
C, which pathogens are of concern? maximum aw of 0.96 for cheese, the consideration since seafood is not
If food is not seafood, Vibrio spp. organisms of concern are B. cereus, C. involved.
may be excluded from consideration. botulinum, pathogenic E. coli, L.
monocytogenes, Salmonella, and S. aureus.
Based on a measured pH of 5.3 and a
maximum aw of 0.98 at the cheese-sauce
interface, same organisms as above.
Based on a measured pH of 5.0 and a
maximum aw of 0.97 at the sauce-crust
interface, same organisms as above.
4.b Considering the ecology, product, B. cereus and C. botulinum spores survive Pathogenic E. coli and Salmonella
and epidemiological history, what baking; L. monocytogenes and S. aureus are inactivated during adequate
pathogens are reasonably likely to may be present from postprocessing baking. They are also not likely to
occur? (Also see Appendix C.) handling. be present in the environment and
therefore recontamination of the
No known illnesses have occurred from
cheese pizza with these organisms
consumption of cheese pizza. However,
is unlikely.
illnesses due to E. coli O157:H7 were
associated with frozen pepperoni pizza, C. botulinum was excluded from
although the cause of the outbreak was consideration because of the aerobic
undetermined (70). conditions and the reduced pH
levels and because spores of B.
cereus are more common and likely
to grow faster.
4.c What pathogens are likely to Study is designed to determine safety if
recontaminate the product after the recontamination should occur.
inactivation step?
4.d Are there any baseline surveys that No.
indicate prevalence of pathogens for
the target product or a related
product?
4.e For growth inhibition (TCS) studies:
4.e.1 Which pathogen(s) will grow the No one organism was determined to grow
fastest? Consider gram positive faster. See 4.e.2.
versus gram negative; vegetative
microorganisms versus spore
formers. If food is not seafood,
Vibrio may be excluded from
consideration. Use a predictive
model or cite applicable literature.
Consider growth potential through
1.5 times the shelf life, if appropriate.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 199

APPENDIX J. Continued
Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

4.e.2 Predictive model Cheese surface at pH 5.4, aw 0.96, 27uC

(80.6uF): PMP 7.0 version 1.1 predicts a 3-
log increase in S. aureus within 29 h (22 h
without lag) under aerobic conditions;
ComBase Predictor predicts a 3-log
increase in S. aureus within 18 h for the
same conditions. For L. monocytogenes,
PMP predicts a 1-log increase within 42 h
for the same conditions (7 h without lag);
ComBase Predictor with 5,000 ppm of
lactic acid predicts a 1-log increase within
33 h for the same conditions. PMP does
not include B. cereus predictions at aw
0.96, but ComBase Predictor with 40%
CO2 predicts a 3-log increase in B. cereus
within 101 h.
For the sauce-crust interface (pH 5.0, aw
0.97), PMP predicts a 3-log increase in S.
aureus within 34 h (approximately 23 h
without lag) under aerobic conditions;
ComBase Predictor predicts a 3-log
increase in S. aureus within 23 h
(approximately 16 h without lag) for the
same conditions. For L. monocytogenes,
PMP predicts a 1-log increase within 52 h
for the same conditions (approximately 9 h
without lag); ComBase Predictor with
5,000 ppm of lactic acid predicts a 1-log
increase in 34 h (within 13 h without lag)
for the same conditions. PMP predicts a 3-
log increase in B. cereus in approximately
21 h (8.5 h without lag); ComBase
Predictor with 40% CO2 predicts a 3-log
increase in B. cereus in 85 h (in 41 h
without lag) for the same conditions.
For the cheese-sauce interface (pH 5.3, aw
0.98), PMP predicts a 3-log increase in S.
aureus within 22 h (approximately 16 h
without lag) under aerobic conditions;
ComBase Predictor predicts a 3-log
increase in S. aureus within 15 h (10 h
without lag) for the same conditions. For L.
monocytogenes, PMP predicts a 1-log
increase within 22 h for the same
conditions (approximately 5 h without lag);
ComBase Predictor with 5,000 ppm of
lactic acid predicts a 1-log increase in 19 h
(within 8 h without lag) for the same
conditions. PMP predicts a 3-log increase
in B. cereus in approximately 15 h (10 h
without lag); ComBase Predictor with 40%
CO2 predicts a 3-log increase in B. cereus
within 42 h (21 h without lag).
4.e.3 Compare choice with literature
4.e.4 Any further information on growth
or survival?
200 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX J. Continued
Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

4.e.5 Based on the above analysis, what L. monocytogenes, S. aureus, Modeling results suggest that L.
challenge organisms are chosen for and B. cereus. monocytogenes, S. aureus, and B.
growth inhibition studies? cereus are all likely candidates for a
challenge study, and none could be
completely excluded from
consideration based on modeling
alone.
4.f If inactivation studies: NA
4.f.1 What is the lethal treatment? (HPP,
heat, acid, etc.)
4.f.2 Which microorganisms are most
resistant to the lethal treatment?
(HPP, heat, acid, etc.)
4.f.3 Will the lethality be delivered to all
areas of the product that may
contain the pathogen? Account for
all surface and internalized
contamination.
4.f.4 What is in the formulation that may
affect inactivation? (Intrinsic factors
may contribute to lethality or
resistance: aw, moisture, salt, pH,
fat, etc.)
4.f.5 Are there any data on pathogen
levels in the product?
4.f.6 Is there a regulatory requirement or
policy for log reduction for this
product? Cite requirement.
4.f.7 If there is no regulatory requirement
for log reduction, use scientific
basis for determining acceptable
reduction (21, 76).
4.f.8 Based on the above analysis, what
challenge organisms are chosen for
inactivation studies?
5 Determine appropriate time and sampling intervals for challenge study
5.a For growth inhibition (TCS) studies, 10 h | 1.5 ~ 15 h.
use 1.25–1.5 times the shelf life as
testing time
5.a.1 Maximum time from production to Maximum 10 h.
consumption
5.a.2 Actual time to spoilage or NA
unacceptable quality
5.a.3 For growth inhibition studies, Sample at 0, 4, 8, 10, 15 h.
determine appropriate sampling
intervals for microbial analysis; use
five to seven (preferred) sampling
intervals; fewer sampling intervals
should be justified, e.g., using
results from similar products.
J. Food Prot., Vol. 73, No. 1 GUIDANCE FOR INOCULATED PACK/CHALLENGE STUDY PROTOCOLS 201

APPENDIX J. Continued
Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

5.b For inactivation studies determine NA

appropriate sampling points
considering the process and
formulation. Identify populations at 0
time and end of processing; whenever
possible, include intermediate
sampling intervals to determine death
curve.
5.b.1 When inactivation treatments may
result in sublethal injury, repair and
growth of microorganisms during
product shelf life should be
considered (21).
6 Determine inoculation, storage, and testing procedures
6.a Determine strains for use in study Multistrain mixtures for L. monocytogenes,
(multiple strains for each species are S. aureus, and B. cereus will be used. Each
recommended; consider use of pathogen composite will be tested
appropriate food or clinical isolates) individually (i.e., inoculate one set of
samples with L. monocytogenes composite,
inoculate a different set of samples with S.
aureus composite, etc.).
6.b Determine if adaptation is required No. Although sauce pH is low, L.
for inoculum preparation monocytogenes comes from the
environment and would not be
adapted to acid. Adaptation is not a
concern for S. aureus or B. cereus.
6.c Determine method of inoculation Slice an entire pizza into 16 individual Each replicate will require 10
(surface, mixing, dipping, liquid, dry, slices (approximately 75 g each). (Assume inoculated slices (2 for each
etc.) that the pizza is approximately 1,200 g.) sampling time interval) and 5
Individual slices of pizza will be inoculated control slices for each organism
on the surface and the sliced edge with S. tested.
aureus, L. monocytogenes, or B. cereus.
6.d Determine size of inoculum Not less than 2 log CFU/g for L. Inoculum level is high considering
(populations, e.g., log CFU/g, CFU monocytogenes, S. aureus, or B. cereus, the likelihood of contamination but
per package, percentage of inoculum surface inoculated, including the cut will allow enumeration by direct
vol/wt or vol/vol) surface, delivered by spot inoculation plating and detection of growth and
(several 50-ml spots). low levels of inactivation by
formulation during storage;
As noted above, each organism will be
inoculum volume no more than 1%
inoculated independently to avoid possible
of sample size; preliminary data
antagonist effect between different
suggest inoculum does not change
organisms.
pH and aw appreciably.
6.e Determine packaging to be used Product is not packaged during typical
display but should be protected from the
environment during the study by placing in
a cardboard or plastic pizza container with
a loose fitting lid.
6.f Determine the incubation Incubated at 30uC (86uF). Maximum temperature product will
temperature for growth inhibition be exposed without adverse changes
studies or temperature(s) for thermal in product quality that would deter
inactivation studies purchase and consumption.
6.g Determine sampling method and Analyze an entire slice of pizza Slices will be tested for S. aureus,
sample size (approximately 75 g). B. cereus, and L. monocytogenes
according to methods provided in
Appendix A.
202 NACMCF J. Food Prot., Vol. 73, No. 1

APPENDIX J. Continued
Evaluation of baked cheese pizza held out of refrigeration for up to 8 h
Considerations Response Additional comments

6.h How many replicates are needed to Three replicate (unique production) lots Greatest variability likely occurs in
ensure confidence in data? Does (i.e., three whole pizzas) per organism the production of different lots of
variability in proximate analysis or tested; duplicate samples (slices) per testing pizza.
production warrant more than two or interval.
three replicate trials? Will multiple
variations of similar formulations be
tested? Has a statistical design for
choosing formulations been used
(block design, central composite, etc.)?
7 Determine other controls
7.a Is use of surrogates appropriate or No surrogates used.
necessary? If so, justify.
7.b Are uninoculated controls needed to Uninoculated controls will be used to
assess spoilage or competitive monitor other spoilage microorganisms that
microflora or for other purposes? can change pH during testing interval.
7.c What other controls are necessary
(including negative or positive growth
controls)?
8 Determine pass-fail criteria
8.a What are the pass-fail criteria? No more than a 1-log increase for L. A 1-log increase in L.
monocytogenes; no more than a 3-log monocytogenes is considered
increase for S. aureus or B. cereus. significant growth, but note that L.
monocytogenes detectable in 25 g of
a ready-to-eat food would render the
product adulterated.
Maximum 3-log increase selected for
S. aureus and B. cereus are based on
increases suggested in the IFT report
(53).
Some regulatory agencies may
consider a lower log increase to be
actionable.
8.b What are the limits for use of the These results are applicable only to cheese Minor variations in the amount of
results? pizza with tomato sauce and not to pizza cheese or tomato sauce are not likely
containing meat or vegetable toppings. to have a significant impact on
growth of the test organisms.
a
NA, not applicable.