Microbiological Shelf-life and Product Spoilage: Achieving Objectives and Investigating Failures PDF Free Download
1 / 22/22
100%
Microbiological Shelf-life
and Product Spoilage:
Achieving Objectives and
Investigating Failures
A EUROFINS WHITE PAPER
J. David Legan and Patricia Quinn
This document by Euro ns
is licensed under a Creative
Commons Attribution 3.0
Unported License.
Table of Contents
page 3 Introduction
page 4 Designing for shelf-life
page 11 Investigating the failure
page 16 Determining the root cause
page 16 Recovering the future
page 17 A better way
page 18 Where to get help
page 18 References
Page 3 of 22
Introduction
Product spoilage rarely makes the headlines but can still be very costly for a producer in
terms of lost product, damaged reputation, and loss of future sales. Prevention is the best
approach to avoid these costs but if prevention fails, we must investigate, take corrective
action, and improve controls to prevent reoccurrence.
This white paper will provide a brief overview of both prevention and investigation of
spoilage. By focusing on principles, it will not be able to answer every question, but rather
will illustrate questions to ask during product design and failure investigations.
Designing for shelf-life
The shelf-life of a food, ingredient, or dietary supplement is a function of many different factors.
Together they determine what kind of spoilage occurs, and how quickly that spoilage is likely to
happen. This has been taught in food microbiology as the “hurdle concept” since 1976 (10) and
imagines that each factor is a hurdle to microbial growth. Together, multiple hurdles combine to
limit the occurrence of microbiological spoilage within a defined period of time.
In this paper, we will consider only microbiological spoilage, but products may also spoil through
physical changes such as loss of texture or acceptable appearance, or chemistry-driven
organoleptic changes such as development of rancidity. This paper is likely to be more relevant
to “manufactured” products that have undergone more than minimal processing than to
minimally processed foods such as produce or raw meat, where shelf-life is very short. When
we get to the sections on diagnosing spoilage, it will be important to remember that
microbiological activity can generate many metabolites with strong odors. These might mislead
us into suspecting chemical adulteration, hence it will be prudent to keep an open mind. First, let
us review the factors that determine product shelf-life.
Intrinsic Factors
Intrinsic factors are those that result from the product formulation, and generally are related to
the desired sensory properties of the product. In most cases, intrinsic factors can be modified to
extend shelf-life, but only within a limited range of values before the desired product sensory
qualities are sacrificed.
Water activity
Water activity (aw) is a measure of the “availability” of water. The definition of aw is “the vapor
pressure of water surrounding the product in a sealed system, divided by the vapor pressure of
pure water at the same temperature”. It is expressed as a decimal, e.g., 0.97. Sometimes,
instruments measure the equilibrium relative humidity (ERH) around the product in a sealed
chamber and express it as a percentage. It is easy to convert between these two quantities
because ERH = aw x 100. Anything dissolved in water will lower its vapor pressure and since
Page 4 of 22
food products usually contain a blend of water and other ingredients, it is the other ingredients
that cause a reduction in aw.
Table 1: Minimum aw for growth of selected microorganisms. Modified from Fontana,
2007 (7).
Microorganism*
Water
Activity
Bacteria
Molds
Yeasts
0.97
Clostridium botulinum E
Pseudomonas fluorescens
0.95
Escherichia coli
Clostridium perfringens
Salmonella
spp.
Vibrio cholerae
0.94
Clostridium botulinum A, B
Vibrio parahaemolyticus
Stachybotrys chartarum
0.93
Bacillus cereus
Rhizopus stolonifer
0.92
Listeria monocytogenes
0.91
Bacillus subtilis
0.90
Staphylococcus aureus
(anaerobic)
Trichothecium roseum
Saccharomyces cerevisiae
0.88
Candida
0.86
Staphylococcus aureus
(aerobic)
0.85
Aspergillus clavatus
0.84
Byssochlamys nivea
0.83
Penicillium expansum
Debaromyces hansenii
0.80
Penicillium citrinum
Zygosaccharomyces bailii
0.78
Aspergillus flavus
0.75
Aspergillus restrictus
Aspergillus candidus
0.71
Eurotium chevalieri
0.70
Aspergillus amstelodami
0.62
Zygosaccharomyces rouxii
0.61
Xeromyces bisporus
<0.60
- - - - - - - - - - - - No microbial growth - - - - - - - - - - - - - - - - - - - - - -
*Mold and yeast names current according to Species Fungorum (2)
Page 5 of 22
Water activity matters because as it becomes lower, the number of different microorganisms
able to grow reduces (7) (Table 1) and the growth rate of those microorganisms becomes
slower. Hence, lower aw translates into reduced spoilage risk and increased microbiological
shelf-life, as well as changes in the potential spoilage microflora. Traditional preservation
techniques of drying and salting work by reducing aw. Once aw drops below 0.6, foods are
protected from microbiological spoilage indefinitely, provided that aw does not increase again for
any reason. Table 2 shows the ranges of aw values in selected food examples.
Table 2: Examples of typical aw ranges for food items. Modified from Schmidt et al,
2020 (18).
CATEGORY: Subcategory; Item(s)
a
w
range
Low
High
Dairy: milk, fresh; whole, 2%, fat free, 1.5%, skim.
0.988
0.996
Fruits, fresh: limes, plums, apricots, nectarines, oranges, apples,
peaches, melon, strawberries 0.980 0.997
Meat, raw: pork, lamb, ground beef,
0.980
0.992
Dairy: yogurt; various manufacturers and flavors.
0.973
0.988
Beverages, ready- to-drink: margarita mix; branded carbonated
beverages, regular and diet. 0.972 0.999
Meat and poultry, ready to eat: ham, smoked turkey, corned beef
(canned), meat paste (canned) 0.971 0.984
Fruits: fruit juices; grape, apple, cherry, orange, tomato, pineapple
0.970
0.990
Vegetables, fresh: onions, green beans, parsnips, Brussels sprouts,
carrots, asparagus, cucumbers, spinach 0.970 0.988
Fish, canned: tuna, sardines, mackerel, salmon.
0.968
0.974
Dairy, cheese, fresh: Cheddar, Swiss
0.950
0.960
Dairy, cheese, processed.
0.910
0.946
Baked goods: bread; multiple brands and varieties.
0.900
0.970
Jams, jellies, preserves: grape, peach, raspberry, strawberry
0.802
0.852
Fruits: fruit juice concentrate; orange
0.800
n.a.
Baked Goods: cake; multiple brands and varieties
0.720
0.945
Dairy: cheese, fresh; parmesan
0.693
0.881
Meat: dried; beef jerky, salami, pepperoni
0.691
0.860
Fruits: dried; raisins
0.510
0.540
Breakfast cereals: multiple brands and varieties
0.157
0.472
Cereal grains, legumes, nuts and seed.
0.147
0.750
Milk: dried*
0.137
0.277
Baked Goods: cookies and crackers; multiple brands and varieties
0.112
0.649
*Outlier recorded at a
w
= 0.75, in line with dairy protein powders.
Page 6 of 22
pH
The pH of a material is a measure of acidity or alkalinity on a scale from 0 to 14, where 0 is the
most acidic, 14 is the most alkaline and 7.0 is neutral. Most foods are neutral to acidic (Table 3).
Similar to aw, the more extreme the pH value becomes, the fewer microorganisms are able to
grow (Table 4), and the slower the growth rate of those that can. The traditional preservation
techniques of lactic acid fermentation and pickling work by acidifying the product and reducing
the pH.
Table 3: Typical pH of common foods. Adapted from extension sources (1, 13).
Approximate pH
Approximate pH
Item
Low
High
Item
Low
High
Eggs, whole, frozen
8.5
9.5
Beef (ground)
5.1
6.2
Cake, chocolate
7.2
7.6
Sugar
5.0
6.0
Egg solids, whole
7.1
7.9
Molasses
5.0
5.5
Egg whites
7.0
9.0
Oysters
4.8
6.3
Crackers
7.0
8.5
Peppers
4.7
5.5
Freshwater fish (most)
6.9
7.3
Eggplant
4.5
5.3
Cake, pound
6.8
7.1
Tomatoes
4.3
4.9
Shrimp
6.8
7.0
Tomatoes, puree
4.3
4.5
Fish (most fresh)
6.6
6.8
Cucumbers, pickled
4.2
4.6
Egg solids, whites
6.5
7.5
Mayonnaise
4.2
4.5
Chicken
6.5
6.7
Tomatoes, juice
4.1
4.6
Broccoli
6.3
6.9
Corn starch
4.0
7.0
Avocados
6.3
6.6
Ketchup
3.9
3.9
Milk
6.2
7.3
Cherries, Royal Ann
3.8
3.8
Butter
6.1
6.4
Olives, green fermented
3.6
4.6
Herring
6.1
6.4
Plums, Red
3.6
4.3
Olives, black
6.0
7.0
Tomatoes, paste
3.5
4.7
Brussels sprout
6.0
6.3
Mangoes, green
3.4
4.8
Flour
6.0
6.3
Apricots
3.3
4.8
Ham
5.9
6.1
Orange juice
3.3
4.2
Carrots
5.9
6.4
Peaches
3.3
4.1
Mangoes, ripe
5.8
6.0
Apple, eating
3.3
4.0
Squash, yellow, cooked
5.8
6.0
Pineapple
3.2
4.0
Beans
5.6
6.5
Cucumbers, dill pickles
3.2
3.7
Spinach
5.5
6.8
Rhubarb
3.1
3.4
Lamb
5.4
6.7
Strawberries
3.0
3.9
Potatoes
5.4
5.9
Grapefruit
3.0
3.8
Pork
5.3
6.9
Grapes, seedless
2.9
3.8
Beets
5.3
6.6
Plums, Blue
2.8
3.4
Bread
5.3
5.8
Ginger ale
2.0
4.0
Tuna fish
5.2
6.1
Vinegar
2.0
3.4
Papaya
5.2
6.0
Lemon juice
2.0
2.6
Page 7 of 22
Table 4: pH ranges for growth of different groups of microorganisms. Modified from
Lund & Eklund, 2000 (12).
pH range
Species name Minimum Optimum Maximum
Bacteria (most)
4.5
6.5-7.5
9.0
Acetobacter spp.
3.0
5.0-6.0
Alicyclobacillus
2.0
3.5-4.0
7.0
Alicyclobacillus acidocaldarius
2.5
3.4-4.0
5.5
Enterobacter agglomerans
3.6
Escherichia coli (non-pathogenic)
4.0
Clostridium butyricum
Gluconobacter spp.
3.0
5.5-6.0
Lactobacillus spp.
2.9
6.3-6.8
Pathogenic bacteria
Bacillus cereus
5.0
6.0-7.0
8.8
Clostridium botulinum
4.6
Clostridium perfringens
5.0
9.0
Salmonella spp.
3.8
7.0-7.5
9.5
Yeasts (general range)
1.5-3.5
4.5-6.8
8.0-8.9
Candida krusei
1.3
Saccharomyces cerevisiae
1.6
Pichia membranifaciens
1.9
Zygosaccharomyces bailii
2.2-2.5
>7.0
Molds (general range)
1.5-3.5
4.5-6.8
8.0-11.0
Moniliella acetoabutans
<3.3
Aspergillus flavus, A. parasiticus, A. niger
2.0
5.0-8.0
>11.0
Fusarium equiseti
<3.3
5.0-8.0
>10.4
Penicillium verrucosum
<2.1
6.0-7.0
>10.0
Preservatives and other microbial inhibitors
Many food components have some antimicrobial activity that goes beyond their effect on aw or
pH. Organic acids, such as acetic, lactic, and citric acids, are often important components of the
flavor system of the product through their acidifying action. Others such as sorbic, propionic and
benzoic acids are primarily added as antimicrobials. In all cases, these acids become more
antimicrobial as the pH drops, with this effect being in addition to any reduction in microbial
growth attributable solely to the pH value. Other additives including alcohols, nitrites or sulfites
have utility in particular applications. Many of these ingredients and additives are regulated, and
specific restrictions can be found in the Substances Added to Food database (22) from FDA
formerly known as EAFUS (Everything Added to Foods in the United States). Whether these
compounds are derived naturally, or “artificially” makes no difference to their activity but does
affect the way that they are shown on product labels because consumers have, for decades,
expressed a preference for “natural” ingredients. Hence, product labels may show ingredients
such as “cultured corn sugar”, which contains lactic acid and/or sodium lactate, or “cultured
Page 8 of 22
celery juice,” which contains sodium nitrite as a “friendlier” or “cleaner” alternative to artificial
preservatives.
Structure
Food structure can significantly affect shelf-life by affecting where and how microorganisms can
grow, and the distribution of nutritive and inhibitory factors. For example, butter is a water-in-fat
emulsion where individual water droplets are so small that only a small amount of growth can
occur in any individual droplet before it is “full,” and the microbial growth is physically
constrained. Salted butter contains about 11-17% water and 1.0-1.6% salt (median values
16.1% and 1.3% respectively) (19). However, the salt partitions into the water droplets, so that
the concentration of salt on water is around 8% and the range of aw is from 0.83 to 0.95 (18),
which constrains most bacterial growth.
Extrinsic factors
Extrinsic factors are those pertaining to the environment around the product. We have great
capacity to manipulate extrinsic factors to our advantage but cost or product factors may
constrain our ability to maximize their effect on shelf-life.
Temperature
Probably the first factor that comes to mind for most of us is temperature. We all have personal
experience of food keeping for longer under refrigeration than at room temperature, and for very
long periods when it is frozen. All microorganisms have an optimum temperature range for
growth. For most food spoilage microbes that optimum temperature range is around 20-40 °C
(68-104 °F). As temperature moves away from that range, either higher or lower, growth rates
are reduced and the number of different microorganisms able to grow becomes smaller.
Principles for use of temperature for safe storage of foods have been developed over many
years and the current view of the US Department of Agriculture is that for safety, foods should
be stored below 40 °F (4.4 °C) or, if being kept hot for serving, above 140 °F (60 °C) (21).
Further, for safety, foods should be cooled from processing temperatures to below 70 °F within
two hours and to below 40 °F within a further two hours.
The shelf-life of foods held above 140 °F is a matter of hours at most, and spoilage is through
some degradation of taste, texture, appearance, or other quality attributes. We will not consider
this temperature range further. Foods stored below 40 °F but not frozen may have a shelf-life of
days to weeks or months, depending on the intrinsic factors of the product and its microflora. If
foods are frozen and stored at 0 °F (-18 °C) or lower, they may last for months or years and
spoilage is through physical and/or chemical changes.
Page 9 of 22
Atmosphere
The atmosphere immediately surrounding the food can affect both the type and rate of
microbiological growth that can occur. We can extend the shelf-life of many foods by choosing
to remove or modify the atmosphere within the package.
Foods with no internal spaces can be vacuum-packed, in which all the air is removed before the
package is sealed. This leaves the residual oxygen concentration at extremely low levels,
effectively limiting the growth of aerobic microorganisms such as molds and many bacteria and
promoting an anaerobic spoilage microflora that can grow in the absence of oxygen. Most
yeasts and many bacteria can grow anaerobically, though to cause spoilage these anaerobes
must also be able to tolerate the product’s intrinsic factors and the storage temperature.
Vacuum packaging will crush any product with internal voids, including many bakery items, or
foods with soft textures or soft components. For these foods, we can replace some or all of the
air in the package with a custom gas blend. Such blends often contain carbon dioxide (CO2) and
nitrogen or other inert gases. The CO2 is actively antimicrobial but tends to dissolve into any
water present in the food. The inert gas helps to exclude oxygen while protecting the package
from collapsing as the CO2 dissolves.
Packaging barrier properties
The most obvious function of packaging is to protect products against gross contamination and
physical damage, but it does much more than that. For long shelf-life items, the moisture-barrier
properties prevent products from drying out in low humidity environments and from absorbing
moisture from high humidity atmospheres. If some form of modified atmosphere is used,
whether vacuum packaging or use of a replacement gas mixture, the oxygen and CO2 barrier
properties are also important. Glass and metal containers have the highest barrier properties
(and highest cost and weight). Plastic structures cover a wide range of barrier properties. In
general, for higher barrier properties plastics require laminated structures and are more
expensive than those with low barrier properties. Pinhole leaks and poor seals can compromise
the performance of any package and cause loss of any expected shelf-life enhancement from
the use of a modified atmosphere. Selection of appropriate packaging involves a tradeoff of
cost, weight, convenience, and performance.
Processing
In the context of this whitepaper, we are concerned with the ability of food processing to provide
a treatment lethal to microorganisms that could otherwise sicken consumers or spoil the
product. However, it is important that whatever process we select does not adversely affect the
desired product qualities of taste, texture, and appearance. To simplify matters we can think of
“sterilizing” processes that destroy all viable microorganisms, pasteurizing processes that
destroy vegetative microbial cells but not bacterial spores and non-lethal treatments that
redistribute microorganisms but have no lethal effects. Process lethality is a function of
temperature and time, in the case of thermal processing.
Page 10 of 22
Canning is the classical example of a sterilizing process. A canning process delivers
commercial sterility, which is defined in the US Code of Federal Regulations (4) as
“…the condition achieved -
(i) By the application of heat which renders the food free of -
(a) Microorganisms capable of reproducing in the food under normal
nonrefrigerated conditions of storage and distribution; and
(b) Viable microorganisms (including spores) of public health significance; or
(ii) By the control of water activity and the application of heat, which renders the food
free of microorganisms capable of reproducing in the food under normal nonrefrigerated
conditions of storage and distribution. “
Other, less lethal, thermal processes kill some fraction of the vegetative microorganisms present
but not any bacterial spores, thereby reducing the starting population and extending the time
until the population grows large enough to cause spoilage. When applied deliberately for this
purpose, or to improve safety, we call these pasteurization processes. However, the same effect
can be achieved when the thermal process is used to cook the food to achieve desired
functional and organoleptic properties.
Processes that redistribute microorganisms can have either positive or negative effects on
microbiological shelf-life but rarely in a manner that can be exploited.
Lethal processes commonly use heat, but several non-thermal lethal processes such as high
pressure, pulsed electric field, ultraviolet light, ionizing radiation and treatment with antimicrobial
gases, are all used in limited applications.
Microbial factors
If microorganisms present can grow given the intrinsic and extrinsic factors discussed above,
we will often see characteristic spoilage patterns, for example:
• Low aw products tend to have mold spoilage if solid or yeast spoilage if liquid.
• High aw, neutral or nearly neutral products tend to exhibit bacterial spoilage.
• High aw, acidic products tend to show mold and/or yeast spoilage but a limited range of
bacterial spoilage is also possible.
• In vacuum-packaged products, anaerobic bacteria or yeasts tend to dominate spoilage and
will often, though not always, produce gas leading to visible pack bloating and the condition
known as “blown pack spoilage.”
Page 11 of 22
Realistic expectations
The interplay of all the factors described to this point determine the shelf-life and spoilage
population in any particular product. After understanding the intrinsic and extrinsic factors and
processing pertaining to a product, an experienced food microbiologist will be able to assess the
most probable characteristics of spoilage. They will also likely be able to estimate shelf-life.
However, there are so many variables at play, that it is wise to confirm an accurate shelf-life
through stability and challenge studies. The key point to remember is that every product spoils
eventually if not consumed first. In exceptional cases, this could take several decades (6, 11)
and not all spoilage is caused by microorganisms. If failure to achieve the intended shelf-life is
because of microbiological spoilage, and within a timescale that is reasonable or typical for the
product given the factors discussed above, then the underlying problem may be an unrealistic
expectation of the achievable shelf-life, possibly based on incomplete information.
Investigating the failure
The Swiss cheese conceptual model is used in many areas of failure analysis (16, 17). It sees
every barrier to failure (in this case spoilage) as imperfect – like a slice of Swiss cheese with
holes. It is illustrated in Figure 1 showing that failure occurs if the holes in all the layers align.
This model helps us to remember that when a product spoils more quickly than was expected,
an investigation to determine the root cause must ask many questions and make many
observations. Only then will we have the information to determine steps needed to correct the
problem and prevent it from reoccurring.
Figure 1: "Swiss cheese" model of product failure. Each slice represents an
incomplete barrier. When all barriers are breached, a spoilage event (failure) can
occur.
Page 12 of 22
Symptoms
Observations of symptoms are the most logical place to begin.
• Appearance: What does the product look like?
o Bubbly? (Or pack bloating)?
o Slimy?
o Moldy? (Mold spores are readily inhaled and may trigger allergies or infections so we
should not deliberately smell a moldy product.)
• Smell: If it’s not visibly moldy, what does it smell like?
o Alcoholic / yeasty?
o Sour?
o Putrid?
o Rancid / paint-like?
o Rotten eggs?
o Dirty feet?
• Taste: We will not deliberately taste a spoiled product but sometimes spoilage is discovered
when spoiled food is inadvertently eaten, and that information can provide insights.
Product, package, and distribution
Next, we can examine the intrinsic and extrinsic factors discussed above looking for conditions
that permit the observed spoilage symptoms, for example:
• What is the aw? If the product has multiple components, is aw uniform across all of them?
• What is the pH? Do all components of the product have the same pH (if relevant)?
• What is the product form: solid, liquid, emulsion, multi-component. And where is the spoilage
physically appearing? In one component? At the junction between two components?
• What are the product’s ingredients? Does the ingredients list include a preservative?
• Is the package damaged? If yes, is that common to all spoiled units?
• If the product is vacuum- or modified-atmosphere packed, what are the oxygen, CO2 and
moisture barrier properties?
• At what temperature is the product stored, distributed and sold?
o Are we certain that the product experienced no temperature-abuse?
o If not, what was the temperature and time history of the product?
Diagnosis: microbiological testing
The answers to the questions above may be enough to determine the cause of spoilage, but
more commonly they suggest microbiological tests that would shed more light on the problem.
For example,
• Common causes of pack bloating, bubbling, or other forms of gassy spoilage include
yeasts, coliforms, heterotrophic lactic acid bacteria and Clostridium species.
• Slime formation is often caused by some species of Leuconostoc, Pseudomonas, or
Bacillus.
Page 13 of 22
• Sour notes can be caused by acid-producing bacteria including Lactic acid bacteria,
Enterobacteriaceae, or some species of Bacillus or Clostridium.
• Putrid notes are caused by microorganisms capable of breaking down proteins including
some Pseudomonas and Clostridium species.
• Rancid notes are often the result of purely chemical oxidative processes but can be
caused by bacteria including some Pseudomonas and Clostridium species, some
yeasts, and many molds. Some microbial enzymes are robust enough to survive thermal
processing and accelerate chemical processes in otherwise shelf-stable, commercially
sterile products.
This is not an exhaustive list but intended to illustrate that symptoms can help us to design a
rational testing approach for diagnosis of the cause of the spoilage. At this point, standard
methods are not always adequate because they were not designed for this purpose, so
sometimes we must consider method modifications or even alternative methods. It is always
best, if possible, to run any tests alongside the same tests run on an unspoiled product since
the comparison will highlight differences between the two. A few hypothetical examples will
illustrate the thought process but should not be considered an exhaustive list.
Example 1
A vacuum packed, refrigerated, ready-to-eat cooked meat product (high aw, near-neutral pH)
with an expected shelf-life of 8 weeks presents after only 4 weeks with an expanded package
indicating gas production. On aseptically opening the package, a sour odor is noticed. This
suggests the presence of gas-producing microorganisms. We hypothesize the presence of
heterofermentative lactic acid bacteria from the smell. We will test for heterofermentative lactic
acid bacteria and, from the same sample, Enterobacteriaceae, yeasts and anaerobic spore
formers. All tests will be run at the “standard” incubation temperature for said test, and with
incubation at a non-standard 15-20 °C in case the spoilage organism is only able to grow at low
temperatures. In this example, all the tests are quantitative methods or “counts”. Running all
tests in parallel means that we do not lose time or sample if our initial hypothesis is wrong.
We find very high counts of heterofermentative lactic acid bacteria, with Enterobacteriaceae,
yeasts and anaerobic spore formers all below the limit of detection. These results indicate lactic
acid bacteria as the immediate cause of the gas production in the package. Since lactic acid
bacteria are not particularly heat resistant, hence would not have survived a validated and
correctly operated cook step, the root cause investigation must confirm correct operation of the
cook step and investigate potential sources of cross-contamination during slicing and
packaging.
Example 2
A fruit juice product, subjected to an ultra-heat treatment (UHT) process, aseptically filled into
cartons, and distributed at ambient temperature, presents with a medicinal odor when opened.
There is no distortion of the carton to indicate gas production. The aw is high, and the pH is low.
Page 14 of 22
The spoilage observed strongly suggests that outgrowth of Alicyclobacillus spores was the
cause, however we need to eliminate the possibility of other common spoilers of fruit juice
including yeasts, lactic acid bacteria and acetic acid bacteria. None of these three groups are
heat-resistant, hence we would not expect them to be present in this product. If we did find
them, their presence would point to either a process failure or an issue with post-process
recontamination. Since this product is distributed at ambient temperature, there is no need to
consider incubation at reduced temperatures.
In this case, testing reveals a high count of Alicyclobacillus. Moreover, the Alicyclobacillus test
demonstrates that the strain present can produce guaiacol, which is the direct cause of the
medicinal odor. Since Alicyclobacillus is a spore former it may have survived the UHT process
and further investigation should look at the condition of raw materials, the sanitation of
processing equipment, and both the validation and correct operation of the thermal process.
Example 3
Retail packages of raisins arrive at the laboratory with a note indicating that there is a white
bloom on the surface of many of the individual fruits. There is no indication that the packages
are bloating or damaged. Upon opening a package there is no noticeable odor other than that of
the raisins themselves and when we aseptically sample a few of the fruits into an empty Petri
dish there is a low, white bloom, on the wrinkled surface of the raisins. Examination under a low
powered stereomicroscope reveals that the white bloom is slightly furry and, therefore, most
likely a mold. Since the mold is visible, there is little point in a standard “mold count” and an
alternative method can be employed. One option is to place contaminated raisins directly on the
surface of a convenient medium for growing molds. Since the raisins have a low aw, we place
additional raisins on the surface of a medium with a very high (50%) glucose concentration to
support growth of molds that prefer to grow at lower aw. Plates are incubated upright at 25, 30
and 35 °C and are reviewed every 3 days until there is clear mold growth present. We also take
samples from each package for aw measurement.
After a few days we see the beginnings of white growth across the surface of the high-glucose
medium and we have results for water activity of the raisins. The measured value ranged from
0.67 to 0.72 across the samples tested. We have demonstrated that the white bloom was
evidence of mold growing on the raisins. We also have a measured aw, which is unusually high
for this product and suggests that either the raisins were not completely dried, or they
encountered moisture at some time after drying. Further investigation should examine the drying
process and downstream opportunities for moisture ingress.
Diagnosis: microbiome analysis
One of the more useful modern tools for spoilage diagnosis is the ability to characterize the
entire microbiome of a spoiled sample using Next Generation Sequencing (NGS) technologies.
These technologies have dramatically increased the speed, and cut the cost of sequencing,
thereby offering diagnostic capability that was cost-prohibitive only a few years ago. If applied to
Page 15 of 22
DNA extracted directly from spoiled and unspoiled product, NGS can quickly indicate important
differences between the two, including qualitative differences in the identities of any
microorganisms present, and semi-quantitative differences in the relative abundance of those
microorganisms. However, NGS cannot show whether any organisms identified are alive or
dead, or the absolute numbers present. Hence, NGS is ideally used to supplement or guide
cultural approaches. Figure 2 illustrates the ability of microbiome analysis to show differences in
microbial populations. It shows the major components of the microbiome of cold smoked
salmon, before and after aerobic incubation at 4 °C for 30 days. The population changed from
one dominated by Carnobacterium species to one dominated by Brochothrix thermosphacta, a
lactic acid bacterium common in spoilage of chilled raw meats. The plots are simplified from
supplementary data of Jarvis et al (8).
Figure 2: Microbiome comparison: relative abundance of dominant microorganisms on
cold-smoked salmon before and after 30 days’ aerobic incubation at 4 °C. The
population, initially dominated by Carnobacterium, became dominated by Brochothrix
species. Pseudomonas, Enterococcus, Lactococcus, and Serratia species were also
present. Data from Jarvis et al, 2022 (8).
Diagnosis: interpretation
The cause of spoilage is normally determined by reviewing the results from the investigational
steps above and comparing them with the reported spoilage observations. This typically results
in a solid understanding of what microorganism(s) were the immediate cause of spoilage and
what intrinsic or extrinsic product factors enabled this spoilage to occur. There is no standard
confirmation procedure. If the microbiological investigation leads to an isolated microorganism, it
Page 16 of 22
is possible to inoculate unspoiled product with that isolate, then package and incubate to mimic
the conditions of the original spoiled product. If the observed spoilage condition is recreated in
the inoculated product but not in an uninoculated control, that will provide strong evidence that
we isolated the causative microorganism. However, if an inoculation study does not reproduce
the observed spoilage, we cannot conclude that the microorganism isolated was not the
causative agent – only that we do not fully understand all the circumstances. Considering the
time needed for this additional step, it is more common to act without the information it may
generate.
Determining the root cause
To this point, we’ve looked at factors that contribute to product shelf-life and approaches for
identifying the immediate cause of product spoilage that might occur within the labeled shelf-life
of the product. Sometimes, this stage of the investigation will reveal that the labeled shelf-life
was unrealistically long for the product and the solution is to shorten the labeled shelf-life or
modify some aspect(s) of the product and/or distribution system. With products that are
relatively new to market, this may be because the initial product testing was not adequate to
accurately establish a reliable shelf-life. With long-established products we sometimes see that
multiple small changes have been made over an extended period and each had an intended
beneficial effect: a cost saving, improved palatability through a small increase in product
moisture content (raising aw), or an improved preference score through a small decrease in
product acidity (raising pH). The net result of these changes is that the product drifts so far away
from the conditions under which its shelf-life was established that it can no longer support the
labeled shelf-life.
Other times, the root cause may be less obvious and require additional effort to evaluate steps
in the sourcing, manufacture, and distribution of the product upstream of the point when the
spoilage was observed. Those steps may not all be under the control of the same organization,
hence our goal must be to learn enough to prevent the problem from reoccurring. This stage of
the investigation may involve review of documents and specifications for indications that
something unusual happened. It also may involve physically reviewing the upstream operations
in sequence looking for sources of contamination, evidence of inadequate sanitation, signs of
temperature abuse, etc. We should be aware that there may not be a single point of failure.
Small adverse changes in several elements can happen inadvertently. Or a series of small but
deliberate changes that individually have little impact can add up with premature spoilage as the
unintended consequence.
Recovering the future
To prevent the spoilage condition from reoccurring, we must address the root cause(s) of failure
identified in the investigation. If they are within our control, we can directly implement necessary
corrective and preventive actions. When the root causes are out of our control, we must require
Page 17 of 22
that our suppliers take appropriate actions and, if that fails, be prepared to change suppliers.
Then we must use appropriate controls to keep the operation in its new, more secure state.
A better way
Rather than experience the pain and cost of investigating the cause of a spoiled product and
taking remedial action, it is better to take steps to design and make a product with the required
shelf-life and then take further steps to manufacture and distribute it consistently in line with the
design parameters.
Product design
We considered the intrinsic product factors and extrinsic environmental factors that determine
product shelf-life in the section “Designing for shelf-life”. During product design, we need to
balance the desired product sensory qualities with the intrinsic and extrinsic factors required to
achieve our desired shelf-life. This can be a difficult task involving much trial and error.
Experienced product developers have good intuition for how to approach this process efficiently
and some modeling resources are available to help with parts of this process (5, 14, 15, 20).
Note that models are not a substitute for expert knowledge or appropriate testing but are
valuable for guiding study design and reducing product development time.
Challenge and stability testing
A prototype product can be deliberately inoculated with potentially important pathogens and/or
spoilage microorganisms and evaluated over a period of storage in the intended distribution
conditions and under “abuse” conditions. Periodic evaluation of the microbial population will
show how long the product survives before significant changes occur. This microbiological
challenge study allows us to understand what is likely to happen if some sanitary failure occurs
during manufacture or packaging of the product.
A parallel sensory study can evaluate the appearance, smell, and taste of uninoculated
prototype samples under the same storage conditions. At this stage of the product development
cycle, we would want to do some pathogen testing of the prototype before tasting.
The results of these two studies can tell us what shelf-life to expect when the product is made
and distributed under control, and when some element of the overall operation is out of control.
They can also show us the most likely spoilage mode and help to estimate the pathogen risk.
A short list of questions in appendix 1 will help when preparing to engage a laboratory to help
with challenge or shelf-life testing.
Page 18 of 22
Specifications and verification testing
Once we have a product that delivers on all expectations for taste, texture, appearance and
shelf-life, the key to avoiding future problems is to manufacture and distribute it consistently,
meeting all of the design criteria. That is best done by following the principles of good
manufacturing practice (3) and using appropriate routine testing to verify that all aspects of
the operation are under control. Such testing could include simple and quick verification of
product aw and pH and microbiological testing of both product and environmental samples
for presence and levels of the most likely spoilage microorganisms. Specifications for
acceptable results should be based on what can be achieved when the operation is under
control. These tests can easily be added to the routine pathogen testing that is already
being done. Guidance on microbiological specifications is described elsewhere (9).
Where to get help
Eurofins has many resources to help at all stages of product development and failure analysis.
• The National Food Laboratory can help with product design and development.
• Eurofins Assurance can help with developing specifications, standard operating procedures,
training, and in-plant investigations.
• Eurofins Microbiology Laboratories in New Berlin, WI and Fresno, CA can help with the
microbiological issues discussed in this whitepaper including spoilage diagnosis, challenge
studies and shelf-life studies.
• The Eurofins Microbiology Laboratory based in Madison, WI, can assist with microbiome
analysis using Next Generation Sequencing.
These resources can be accessed through a single point of contact with your local Eurofins
laboratory or online at https://www.eurofinsus.com/food-testing/contact-us/
Acknowledgements
Thanks to Andrzej Benkowski, Dan DeMarco, and Chris Crowe for helpful comments on the
manuscript.
References
1. Anon. pH Values of Common Foods and Ingredients. Clemson Co-operative Extension. Food
Systems and Safety. Available at:
https://www.clemson.edu/extension/food/food2market/documents/ph_of_common_foods.pdf
Accessed 06 October 2023.
2. CABI Databases. Species Fungorum. https://www.speciesfungorum.org/Names/Names.asp
Accessed 18 October, 2023.
3. Code of Federal Regulations 21 CFR § 110 Current Good Manufacturing Practice in
Manufacturing, Packing, or Holding Human Food. Available at: https://www.ecfr.gov/current/title-
21/chapter-I/subchapter-B/part-110#110.5 Accessed 04 October 2023.
Page 19 of 22
4. Code of Federal Regulations 21 CFR § 113.3 Definitions. Available at
https://ww.ecfr.gov/current/title-21/chapter-I/subchapter-B/part-113/subpart-A Accessed 24
August 2023.
5. Combase. A Web Resource for Quantitative and Predictive Food Microbiology.
https://www.combase.cc/index.php/en/ Accessed 19 October 2023.
6. Dell ‘Amor, C. 2017. 100-Year-Old Fruitcake Found in Antarctica Is ‘Almost’ Edible. National
Geographic. August 10, 2017. Available at:
https://www.nationalgeographic.com/adventure/article/antarctica-fruitcake-scott-terra-nova
Accessed 29 September 2023.
7. Fontana Jr., Anthony J. 2020. Appendix D: Minimum Water Activity Limits for Growth of
Microorganisms. In Water Activity in Foods: Fundamentals and Applications, 2nd edition. Edited
by G.V. Barbosa-Cánovas, A.J. Fontana Jr., S.J. Schmidt &, T.P. Labuza, pp. 571-572. Wiley,
Available at: https://onlinelibrary.wiley.com/doi/full/10.1002/9780470376454.app4 Accessed 18
October 2023, 2023.
8. Jarvis, K.G., C-K Hsu J.B. Pettengill, J. Ihrie, H. Karathia, N.A. Hasan & C.J. Grim. 2022.
Microbiome Population Dynamics of Cold-Smoked Sockeye Salmon during Refrigerated Storage
and after Culture Enrichment. Journal of Food Protection, 85, (2), 238–253.
9. Legan, J.D. 2023. Microbiological Specifications in Food Operations. A Eurofins Whitepaper.
Available at: https://www.eurofinsus.com/food-testing/resources/microbiological-specifications-in-
food-operations/ Accessed 06 October 2023.
10. Leistner, L. & Rödel, W. 1976. The stability of intermediate moisture foods with respect to micro-
organisms. In Intermediate Moisture Foods. Davis, R., Birch, G.G., & Parker, K.J., eds. 120-137.
Applied Science Publishers, London.
11. Lewis, D. 2015. Arctic Explorers Uncover (and Eat) 60-Year-Old Food Stash. Smithsonian
Magazine. October 22, 2015. Available at: https://www.smithsonianmag.com/smart-news/arctic-
explorers-uncover-60-year-old-food-stash-180956936/ Accessed 29 September, 2023.
12. Lund, B.M. & T. Eklund. 2000. Control of pH and use of organic acids. In The Microbiological
Safety and Quality of Food, Vol 1 pp 175-199. Edited by B.M. Lund, T.C. Baird-Parker & G.W.
Gould. Gaithersburg, MD. Aspen Publishers.
13. McGlynn. M The Importance of Food pH in Commercial Canning Operations. Food Technology
Fact Sheet FAPC 118. Robert M. Kerr Food & Agricultural Products Center. Oklahoma State
University. Available at: https://extension.okstate.edu/fact-sheets/print-publications/fapc-food-
and-agricultural-products-center/the-importance-of-food-ph-in-commercial-canning-operations-
fapc-118.pdf Accessed 06 October 2023.
14. Mullan, W.M.A. (2009). Predicting the safety and stability of products preserved with acetic acid
using the CIMSCEE model. [On-line]. Updated 2013, 2022 Available
at: https://www.dairyscience.info/index.php/food-model/177-cimscee-model.html Accessed: 19
October 2023.
15. Mullan, W.M.A. (2015) Results of the mould free shelf life prediction. [On-line] Available
from: https://www.dairyscience.info/newCalculators/mould1.asp Accessed: 19 October 2023.
16. Reason, J. 1990. The Contribution of Latent Human Failures to the Breakdown of Complex
Systems. Philosophical Transactions of the Royal Society of London. Series B, Biological
Sciences 327:475-484.
17. Reason, J.T., Carthey, J., & de Laval, M.R. 2001. Diagnosing “vulnerable system syndrome”: an
essential prerequisite to effective risk management. Quality in Health Care, 10 (Suppl II): ii21–
ii25.
18. Schmidt, S.J. & A.J. Fontana. 2020. Appendix E. Water Activity Values of Select Food
Ingredients and Products. In Water Activity in Foods: Fundamentals and Applications, 2nd edition.
Page 20 of 22
Edited by G.V. Barbosa-Cánovas, A.J. Fontana Jr., S.J. Schmidt &, T.P. Labuza, pp. 573-591.
Wiley, Available at: https://onlinelibrary.wiley.com/doi/full/10.1002/9780470376454.app5
Accessed 18 October 2023.
19. U.S Department of Agriculture. Food data Central Available at: https://fdc.nal.usda.gov/fdc-
app.html#/ Accessed 18 October 2023.
20. U.S. Department of Agriculture. Agricultural Research Service. Overview of Predictive
Microbiology. Available at: https://portal.errc.ars.usda.gov/Overview.aspx Accessed 19 October
2023.
21. U.S. Department of Agriculture Food Safety and Inspection Service. “Danger Zone” (40 °F – 140
°F). Available at: https://www.fsis.usda.gov/food-safety/safe-food-handling-and-preparation/food-
safety-basics/danger-zone-40f-
140f#:~:text=%22Danger%20Zone%22%20%2840%20%C2%B0F%20-
%20140%20%C2%B0F%29%201,40%20%C2%B0F.%20Place%20food%20in%20containers%20
on%20ice. Accessed 18 October 2023
22. U.S. Food & Drug Administration. 2022. Substances Added to Food (formerly EAFUS). Available
at:https://www.cfsanappsexternal.fda.gov/scripts/fdcc/?set=FoodSubstances&sort=Sortterm_ID&
order=ASC&startrow=1&type=basic&search=antimicrobial Accessed 18 October, 2022.
Appendices
When engaging with a laboratory during product development for shelf-life testing, or during a
spoilage investigation, compiling answers to the questions in the following appendices will
speed the process and aid in determining appropriate testing.
Page 21 of 22
Appendix 1: Questions to consider when planning for a shelf-life or
stability study.
• What is the Product? (Name and description)
• What is in it? Provide a complete ingredients list including Primary ingredients and
preservatives. The concentration of any preservatives is also important.
• How is it made? Describe the process flow including intermediate steps such as mixing,
intermediate hold times and temperatures, main process (baking, pasteurization, freezing,
pressurization, hot fill, cold fill, etc.) and finishing processes.
• How is it packaged? (Packaging materials and processes, such as MAP or Vacuum
Packing)
• What is the finished product state? (Ready to eat, par-cooked, raw, etc.)
• How will it be distributed and stored condition? (Refrigerated, ambient, frozen, or other)
• List what you know about product characteristics.
o pH
o Water activity
o Moisture
o Estimated shelf-life
o Target shelf-life
• Have you had a shelf-life analysis performed for this product or one similar? (Y/N)
• What is the goal for testing?
o Verification of current shelf-life
o New product – establishing shelf-life
o Change in ingredients, process, packaging
o R&D
o Other (Explain)
• Do you want to simulate ideal/recommended storage conditions via a real-time shelf-life
study, or are you interested in an accelerated shelf-life study with modeling?
• Which stability parameters are you concerned about? (Microbiological, chemical, or
organoleptic/sensory)
Page 22 of 22
Appendix 2: Questions to consider when initiating a microbiological
spoilage investigation.
• What is the Product? (Name and description)
• What “symptoms” of spoilage are you noticing?
• What is in it? Provide a complete ingredients list including Primary ingredients and
preservatives. The concentration of any preservatives is also important.
• How is it made? Describe the process flow including intermediate steps such as mixing,
intermediate hold times and temperatures, main process (baking, pasteurization, freezing,
pressurization, hot fill, cold fill, etc.) and finishing processes.
• How is it packaged? (Packaging materials and processes, such as MAP or Vacuum
Packing)
• What is the finished product state? (Ready to eat, par-cooked, raw, etc.)
• How is it distributed and stored condition? (Refrigerated, ambient, frozen, or other)
• List what you know about product characteristics.
o pH
o Water activity
o Moisture
o Estimated shelf-life
o Target shelf-life
• Is your product packaged at a co-man, or do you package on-site?
• Do you currently utilize an Environmental Monitoring Program?
• If yes, would you be willing to submit environmental monitoring samples to bolster data
procured during the investigation of your product?
• Have you noticed anything unusual in your environmental monitoring results?
• Has anything changed recently (or shortly before the spoiled lot was made?)
