Microbiological hazards and melons

Report prepared for: Codex Committee on Food Hygiene Working Group on the
development of an Annex on melons for the Code of Hygienic Practice for Fresh
Fruits and Vegetables (CAC/RCP 53-2003)

Date: 27 June 2011

 2

Executive summary
This report was prepared in response to a request to FAO/WHO from the Codex Committee
on Food Hygiene (CCFH). In undertaking new work on an Annex to the Code of Hygienic
Practices for Fresh Fruit and Vegetables (CAC, 2003) for melons, CCFH requested a Call for
Data and an evaluation of the pathogen-specific hazards associated with various types of
melons, the role of various agricultural and manufacturing practices in enhancing or
mitigating these hazards, and influences of marketing and consumer handling. The report
includes a literature review of microbial safety and the melon supply chain and data received
from the FAO/WHO call.

Melons are widely consumed in the human diet. There are many varieties of melons known
by common names, the most popular ones being watermelon, cantaloupe (rockmelon) and
honeydew. Melon consumption, production and international trade in melons have increased
over the last decade. In addition, foodborne illness attributed to melons has become a
significant public health concern in some countries with significant negative consequences
for trade in this commodity.

Epidemiological evidence of foodborne illness linked with melons is based on outbreaks.

Between 1950 and May 2011, 85 outbreaks were identified, mainly in North America. The
most common aetiological agent reported was Salmonella enterica (47.1%) followed by
Norovirus (22.4%), Escherichia coli O157:H7 (5.9%), Campylobacter jejuni (3.5%), Shigella
sonnei (2.4%), Listeria monocytogenes, Cyclospora sp. and a suspected combination of
Staphylococcus aureus and Bacillus cereus. The cases numbers per outbreak varied from 2
to 600 with actual case numbers likely ≥100 fold higher. Two deaths were recorded in 3
salmonellosis and a listeriosis outbreak.

The epidemiological data emphasised several points:

• Investigation of illness associated with melons is complicated by the variety in their
culinary use, their distribution and availability of traceback information
• The nature of melons and their popular use in food service, pre-cut and in mixes with
other foods renders them vulnerable to contamination from the rind to edible flesh, via
food handlers and the preparation environment
• S. enterica is the most common aetiological agent and the netted varieties of
cantaloupe either alone or mixed with other melons and other food in meal/dishes is
the most common melon type in recorded outbreaks
• Cross-contamination, poor washing, infected food handlers and poor hygiene
together with poor control of holding temperature contribute to outbreaks in particular
• Norovirus outbreaks result from preparation of melons by infected food handlers with
poor hygiene while other pathogens appear to more often arise from the intact
melons, contamination of the flesh during preparation and poor temperature
management to control growth

Melons have specific characteristics that are important in their interaction with foodborne
pathogens and managing food safety risks, namely:
• Melon rind topography influences the attachment and protection of microorganisms.
Netted rinds, such as on cantaloupe, provide a waxy and highly hydrophobic surface
matrix where microorganisms attach and can be protected from removal by washing
and the effects of sanitizers.
• Foodborne bacterial pathogens have the potential to grow and /or survive on melon
rinds and melon flesh. Growth is arrested at refrigeration temperatures with the
exception of Listeria monocytogenes. Growth can occur rapidly at 20-300C, the
exception being Campylobacter spp., viruses and parasites that can survive
 3

• Microorganisms have been shown experimentally to infiltrate the root system of

melon vines or the fruit. The former is considered transient and of lesser importance
in the field while the latter can occur through a negative temperature differential
during immersion in contaminated water, via wounds caused by physical damage or
pests, splits, the ground spot and the stem scars.

Melons can be contaminated throughout the food chain in a similar manner to other fresh
produce. As fresh melons may be contaminated from their source and there is no further
process that will eliminate the hazards, it is necessary to minimise contamination at primary
production and reduce or at least not increase risk through to the consumer. Important points
at primary production include:
• Favourable conditions for the growth of melons are also favourable for the presence
of wildlife and other pests that may forage in growing areas for the high sugar content
fruits therefore pest management requires special attention
• Melon vines are grown along the ground mostly where the melons are directly or
indirectly (irrigation water, heavy rain splashes) exposed to soil; Use of drip irrigation
and assessing risks at harvest time (e.g. after rain) is required.
• Melons can be exposed to human contamination in the field particularly if they are
turned by hand to minimise the ground spot and sunburn and when placed on cups or
mats; personal hygiene and hygienic use of cups etc and tools is necessary
• After harvest further contact with soil should be avoided.

Following harvest melons can be field packed or sent to packing houses where they may be
washed, sanitized, treated to extend shelf life and cooled before distribution. Processes
should be designed to control factors such as water quality and temperature, duration of
immersion, sanitizer efficacy, personal and equipment hygiene. Sanitising in general
maintains the quality of the wash water rather than sanitizing the melons. Melon growing can
be seasonal and facilities and equipment left vacant and unused in the off season should be
thoroughly cleaned and disinfected before re-use.

There is strong epidemiological evidence that during preparation of melons for consumption
there is a potential for increasing the risk of foodborne illness. Contamination at this
operation point can be introduced from the melon rind, from food handlers, the preparation
environment (cutting boards, knives) and cross-contamination with other melons or foods.
Poor temperature control between contamination during preparation and consumption can
amplify the risk for several bacterial pathogens. The following are important in processing or
value adding:
• While the evidence is variable, washing, scrubbing and sanitizing whole melons
before preparation will result in some decrease although not ensure elimination of
pathogens. Similar results are found with melon pieces and additional reduction
strategies are the use of edible coatings or essential oils
• It has been demonstrated experimentally that pathogens were transferred from the
melon rind to the edible flesh during cutting and cutting after rind removal resulted in
less contamination than cutting before rind removal
• Storage temperature of cut melons and the duration between cutting and
consumption is critical to control growth of bacterial hazards on melon tissue.

Melons have become popular as healthy, fresh, convenient and delicious foods that are
hugely diverse in their use in dishes that appeal to all age groups in all cultures. However,
consumers frequently do not appreciate there is any risk for fruits with an inedible skin.
Melons are low acid fruits and their soft texture makes them appealing for the young, elderly
and infirm who have been shown epidemiologically to be at increased risk when their food is
prepared in an institutional setting. Education in safe handling of melons is required through
 4

chain. Specific characteristics for emphasis include washing, scrubbing and sanitizing before
use, contamination during cutting and serving and the need for temperature and time control.
Industry, retailers and food service suppliers of fresh-cut products need to provide clear
instructions for end users of their products on safe storage, shelf life and handling of their
products.

The available evidence suggests that there will be a low risk melons can be contaminated in
the field and that through the food chain this risk should be decreased or at the least not
increased.
 5

Table of Contents
1.0 Background ..................................................................................................................... 6
1.1 Scope .......................................................................................................................... 6
1.2 International production and trade................................................................................ 6
2.0 Foodborne illness associated with melons....................................................................... 7
3.0 Melon cultivation.............................................................................................................10
4.0 Melons and association with microbial foodborne pathogens .........................................11
4.1 Melon structure ...........................................................................................................11
4.2 Melons and microbial growth and survival...................................................................12
4.2.1 Whole melons and rind.........................................................................................12
4.2.2. Flesh and pulp.....................................................................................................13
4.3 Infiltration of microorganisms into melons ...................................................................13
5.0 Melon production chain and risks of contamination with foodborne pathogens ...............14
5. 1 Production and harvest operations.............................................................................14
5.1 Growing site ............................................................................................................15
5.2 Soil and soil amendments .......................................................................................15
5.3 Water and irrigation .................................................................................................16
5.4 Human activity.........................................................................................................16
5.5 Harvest....................................................................................................................17
6.0 Post-harvest ...................................................................................................................17
6.1 Packing facilities .........................................................................................................17
6.2 Washing and sanitising ...............................................................................................18
6.3 Processing ..................................................................................................................19
7.0 Consumers.....................................................................................................................20
8.0 Microbiological sampling ................................................................................................20
Conclusions..........................................................................................................................21
9.0 References.....................................................................................................................22
Annex 1 Foodborne illness outbreaks associated with melons .............................................26
Annex 2 FAO/WHO Call for data: Summary .........................................................................33
 6

1.0 Background
In 2006 the Codex Alimentarius Commission (CAC) through the 38th session of the Codex
Committee on Food Hygiene (CCFH) sought scientific advice to support the development of
commodity-specific annexes for the Codex Alimentarius “Code of Hygienic Practices for
Fresh Fruit and Vegetables” (CAC, 2003). As a result a priority list of fresh fruits and
vegetables was developed and an Annex for the first priority group of leafy vegetables and
herbs was developed (CAC, 2010). At its 42nd Session, the CCFH proposed to undertake
new work to address the specific problems associated with the control of microbiological
hazards on melons that was included within the second priority group of concern.

The CCFH therefore requested the FAO/WHO to issue a Call for Data and to evaluate the
pathogen-specific hazards associated with various types of melons and the role of various
agricultural and manufacturing practices in enhancing or mitigating these hazards. How
these products are marketed and handled by consumers and the impact of this on foodborne
disease should also be taken into consideration.

1.1 Scope
Melons include fleshy fruits of a vartiety of members of the plant family including gourds or
cucurbits. The plants grow as vines and the flowers following pollenation produce berries that
develop into the mature melons.

Melons belong to the Family Curcubitaceae with two genera of melons widely consumed in
the human diet, Citrullus in which C. lanatus includes commonly known watermelons, and
Cucumis in which C. melo includes many melon varieties some commonly known as
muskmelons or by other common or varietal names. The watermelon differs in producing
berries with a harder rind when compared with the other melons. There are more than 1,200
varieties of watermelon grown in many regions of the world and these can vary in shape and
colour. Among the C. melo varieties there are also notable differences in the melon rind with
the smooth skinned honeydew, crenshaw and casaba melons, the netted skinned cultivars
such as cantaloupe, Santa Claus or Christmas melon and some partly netted varieties such
as the Persian melon, Chinese Hami melons and Charentais. Different cultivars of melons
had been crossbred to improve suitability for commercial production (e..g robustness for
transport, disease resistance) and consumer appeal (e.g. seedless and sweeter cultivars).
Common usage has resulted in some confusion with common names, for example,
muskmelon and cantaloupe or cantaloupe and rockmelon being used synonymously.

This report addresses melons, watermelons, muskmelons and their varieties for human
consumption that are consumed fresh and sold either whole or sliced or combined with other
ingredients as fresh-cut products. It does not include melon seeds, juice and skin (e.g. used
pickled or stir fried) that are also consumed in the human diet. The melon food chain is
considered from primary production to retail and includes information on marketing and
consumer handling where this impacts on foodborne disease.

1.2 International production and trade

FAOSTAT world statistics on agricultural commodities includes a category for watermelons
and uses the term “other melons” collectively to include cantaloupe and other varieties for
human food consumption (FAOSTAT, 2011). World production of all melons in both 2008
and 2009 was approximately 126 million tonnes (Table 1). Of the total production, 78% of the
melons were watermelons in both years.
 7

Table 1 World production of melons 2008 and 2009 (FAOSTAT, 2011)

Production (tonnes)
Year other melons watermelons Total
2008 27,637,248 98,439,589 126,076,837
2009 27,726,563 100,687,056 128,413,619

China was the top world producer of melons in 2008 accounting for 52% of world production
and producing 4.7 times more watermelons than other melons. Other top producers were
responsible for a far less proportion of world supply of 6.4% and less each and included
Turkey, the Islamic Republic of Iran, Spain and the USA. For the decade to 2009, the
production of total melons approximately doubled by 2005 after which it steadied. The
increase in other melons has been double that of watermelons.

Country rankings for production and trade in melons differed in 2008. For example China’s
production would appear to have been predominantly for domestic consumption. The key
exporters in 2008 for watermelons were Mexico (23%) and Spain (17%) followed by the
U.S.A. and Panama and for other melons were Spain (17%) followed by Brazil (10%) and the
U.S.A. (10%). The top importers of melons were Northern American and European Union
countries with the U.S.A. importing almost 19% world imported watermelons and 30% of the
other melons.

The association of foodborne illness with melons can have a significant impact on production
and trade. An example is the melon industry in Mexico between1999 and 2005 where after
reports of illness in North America linked to cantaloupes imported from Mexico, cantaloupe
exports declined by 92% and production declined by 24% (Avendano et al, 2009). These
trade gaps are then readily filled by other countries and the original position may or may not
ever be recovered.

2.0 Foodborne illness associated with melons

Information on foodborne illness was collected from submissions forwarded to JEMRA in
response to a call for data, literature searches, peer reviewed publications and websites. A
Table listing outbreaks and related epidemiological data is provided in the Annex. It is noted
that it was not possible to confirm in outbreaks if the role of melons and the causative agent
were confirmed by laboratory detection or suspected based on statistically significant
epidemiological evidence or both. Where data was suspected this information is included in
the table; however, for many outbreaks it cannot be assured these were all confirmed.

Fresh melon can be eaten alone; however, melons are frequently included in salads (e.g.
dishes combining cut fruits, vegetables and meats etc with optional condiments), and are
used as garnishes on other foods. They have become popular in buffets, salad bars and
catered meals where multiple foods can be consumed by a few or a very large numbers of
people at a common source at the same or over periods of time. The salad can be prepared
in a central facility for distribution to multiple retail or food service outlets. Melons can be
purchased individually and in bulk and a single whole fruit may be only partially consumed at
one sitting. These factors can lead to widespread distribution of outbreak cases and can
make epidemiological investigation and attribution of food difficult. To meet the increased
consumer popularity of melons and to provide year round supply, melons are traded
internationally particularly from countries with warmer climates suitable for production. This
has led to difficulty in tracing back to the primary source of melons and in collecting relevant
epidemiological data on the individual fruit(s) along the supply chain.
 8

Between 1950 and May 2011, 85 outbreaks were identified. The outbreaks were reported by
the U.S.A., Australia, Brazil, Canada, New Zealand, Singapore, Sweden and the United
Kingdom. In four outbreaks, cases occurred in more than one country e.g. U.S.A. and
Canada and were distributed in multiple states in these countries. In the CCFH call for data
from member countries, 8/13 responding countries reported they had no data on the
occurrence of foodborne illness linked with melons. The U.S.A. reported the most significant
number of outbreaks attributed to melons. The U.S.A. Food and Drug Administration
reported that between 1996-2008 in the U.S.A., 82 microbial foodborne disease outbreaks
were linked with fresh produce of which 13 (15.9%) were attributed to melons (USDA, 2009).

The aetiological agent was recorded for 72 of the 85 outbreaks. Salmonella enterica serovars
were most commonly reported (47.1%) followed by Norovirus (22.4%) and Escherichia coli
O157:H7 (5.9%; Annex 1). Much less frequent were Campylobacter jejuni (3.5%) and
Shigella sonnei (2.4%) outbreaks.

Listeria monocytogenes, Cyclospora sp. and a combination of Staphylococcus aureus and

Bacillus cereus were suspected each in a single outbreak. The number of cases per
outbreaks varied from 2 to 600 with estimations the actual number of cases in one outbreak
could have been as much as 100 fold higher. The number of cases did not appear related to
the aetiological agent specifically. Two deaths were recorded for each of 3 outbreaks of
salmonellosis caused by serovars Miami, Chester and Poona, and 2 deaths were reported in
an outbreak of listeriosis.

Eighteen S. enterica serovars were reported in the 40 outbreaks of salmonellosis linked with
melons. Some, notably Javiana, Litchfield, Newport and Poona, were linked with 5 outbreaks
each and Oranienberg with 4 outbreaks.

Table 2 Serovars of Salmonella enterica recorded for 40 outbreaks of salmonellosis linked with
melons and meals/dishes containing melon between 1950 – May 2011.

Number of Number of
Serovar outbreaks Serovar outbreaks
Anatum 1 Muenchen 1
Bareily 1 Newport 5
Berta 1 Oranienberg 4
Carrau 1 Panama 1
Chester 1 Poona 5
Enteritidis 2 Saintpaul 1
Heidelberg 1 Saphra 1
Javiana 5 Typhimurium 1
Litchfield 5 Weltevreden 1
Miami 1 no serotype 1

Some of these serovars were uncommon in human illness databases at the time of the
outbreak. For example, serovar Poona causing several outbreaks in the U.S.A. was rare
among humans at the time and known to be associated with contact with reptiles (MMWR.
2002; MMWR, 1999). This observation subsequently led to the hypothesis that contamination
occurring during production could have been a result of wildlife activity.

Cantaloupe also referred to as muskmelon and rockmelon, honeydew melon and watermelon
were the common melons specified as the food vehicle in outbreaks. In about 6% outbreaks,
the type of melon was not specified. Cantaloupe either alone (24.7%) or included with other
foods (including other melon types) in a meal/dish (29.4%) was most commonly implicated
 9

(Table 3). Watermelon was the next most common melon (16.5%) although it was not so
commonly reported in combination with other foods (4/14 outbreaks) other than when
combined with other melons (10/14 outbreaks). In comparison, honeydew alone was linked
to only 2 (2.4%) outbreaks. In 16 (18.8%) outbreaks meals/dishes contained honeydew
although in 14 of these other melons were present also. In 5 (5.9%) outbreaks the type of
melon was not specified and in 9 (10.6%) outbreaks combinations of melon types were
included among the suspect food vehicles and were not identified specifically.

This information supports the experimental observations described in Section 4 that the
netted varieties of melons such as cantaloupe present a greater risk of pathogen
transmission.

Table 3 Food vehicles associated with 85 foodborne illness outbreaks occurring between 1950
– May, 2011, where melons were implicated

Food vehicles including melons Number of outbreaks (%)

cantaloupe 21 (24.7)
honeydew 2 (2.4)
watermelon 14 (16.5)
melons (not specified) 5 (5.9)
meal/disha including cantaloupe 25 (29.4)
meal/dish including honeydew 16 (18.8)
meal/dish including watermelon 14 (16.5)
meal/dish including melons (not specified) 9 (10.6)
a
meal or dish can include other melons as well as other foods

In 10 outbreaks the melons were reported to have been imported of which 8 were linked with
cantaloupe and one with watermelon. The data gathered on the outbreaks were in many
cases brief and it was not possible to draw overall conclusion on associations with seasonal
effects. However, Bowen et al (2006) in an extensive review of 23 cantaloupe associated
outbreaks in the U.S.A. between 1984 and 2002 found outbreaks occurred in each calendar
month although salmonellosis outbreaks occurred more frequently in December to June with
a peak in May and Norovirus outbreaks occurred in June, September and December (Bowen
et al, 2006).

The outbreak setting was recorded for 58 outbreaks. Food service type facilities such as
restaurants, conference venues and a camp constituted 25 (43%) of these. Other community
based settings included homes (26%), churches, temples, picnics, schools and a day care
centre (17%) and hospital and care facilities (15.5%). The attributed food was widely
distributed in 13 (22.4%) of outbreaks resulting in multiple outbreak settings being involved.

Food service and catering establishments were the most common places where melons
were prepared for consumption, accounting for 60.4% and 15.1% respectively of the 53
outbreaks where this information was provided. Bowen et al (2006) had a similar finding for
28 cantaloupe-related outbreaks in the U.S.A. where 61% and 14% for these facilities were
identified respectively. Grocery stores, supermarkets and a food stall (17% of 53) were also
important while homes (3.8% of 53) and both a processor and a distributor were identified.

Pre-cutting of melons or inclusion of melons in a fruit or other type of salad or buffet type dish
where they would have been sliced or chopped makes these foods particularly vulnerable to
contamination during preparation. Pre-cutting or mixing was noted specifically in 12 (14.1%)
outbreaks (Table 4). This process would have allowed the opportunity for transfer of
contamination from the rind to the edible flesh or contamination by infected food handlers
and the preparation environment. In 4 salmonellosis outbreaks it was noted the melons were
 10

unwashed or poorly washed, including one attributed to watermelon and 3 to cantaloupe. In

one outbreak of salmonellosis attributed to watermelon use of contaminated wash water was
found (Ooi et al 1997).
Table 4 Contributing factors, where identified, among 85 outbreaks of foodborne illness
associated with melons between 1950 - May 2011

Contributing factor Number of outbreaks (%)

Pre-harvest/transport contamination 3 (3.5)
Unwashed melons 4 (4.7)
Pre-cut and/or mixed dish 12 (14.1)
Infected food handler 10 (11.8)
Poor hygiene, bare hands 8 (9.4)
Cross-contamination 2 (2.4)
Poor temperature control 13 (15.3)

Infected food handlers were listed as a contributing factor for 10 (11.8%) outbreaks (Table 4).
Eight of these were caused by Norovirus. In one salmonellosis outbreak an infected food
handlers was identified and for a 2 further outbreaks bare hand contact with food was noted.
Gloved hands were also noted in 2 Norovirus outbreaks in food service settings that
emphasises that even when gloves are used it must be in a hygienic manner. Two of the
outbreaks caused by E. coli O157:H7 were likely caused by cross-contamination with red
meat during preparation or storage. Poor hygiene during preparation was recorded in 6
outbreaks caused by several of the pathogens.

Poor temperature control (15.3% outbreaks) that would have provided the opportunity for
bacterial growth, included both holding at ambient temperature (7 outbreaks) and poor cold
storage (3). These were identified as contributory factors for the S. aureus/B.cereus
(suspected), an E. coli O157:H7 and S. enterica outbreaks.

The epidemiological data emphasises several points:

• Investigation of illness associated with melons is complicated by the variety in their
culinary use, distribution and traceback
• The nature of melons and their popular use in food service pre-cut and in mixes with
other foods renders them vulnerable to contamination from the rind, food handlers
and the preparation environment
• S. enterica is the most common aetiological agent and the netted varieties of
cantaloupe either alone or mixed with other melon and other food in meal/dishes is
the most common melon type in recorded outbreaks
• Cross contamination, poor washing, infected food handlers and poor hygiene
together with poor control of holding temperature contribute to outbreaks in particular
• Norovirus outbreaks resulted from preparation of melons by infected food handlers
with poor hygiene while other pathogens appear to more often arise from the intact
melons, contamination of the flesh during preparation and poor temperature
management to control growth.

3.0 Melon cultivation

Melons grow optimally in warm to hot, sunny locations with fertile, well drained soils. They
grow on vines along the ground in a trailing and scrambling manner. If plants are trained over
a trellis the fruits require support as they enlarge to avoid damage to the plant. Warm and, in
non arid regions, humid conditions, favourable for melon growing can also be favourable for
the survival and growth of human pathogens, and the presence of wildlife and pests. Melons
 11

are susceptible to a wide range of insect pests and microbial diseases. Soil amendments and
agricultural chemicals may be used in conventional primary production and the former in
organic systems.

As melon vines grow horizontally when the fruit enlarges and increases in weight it rests on
the ground. Melons will be exposed to soil contamination directly or via water splashing such
as during heavy rain, spray irrigation or flooding. Plastic mulch if often used to cover the
ground to provide various production advantages for growers by increasing the quality and
quantity of fruit yields and decreasing the growing time compared with growth on bare
ground. Where plastic mulch is used, this will also reduce direct soil exposure. The term
“ground spot” is used to describe the particular area on the rind where the melon sits in
contact with the ground or mulch. This area is thin and underdeveloped in melons and more
susceptible to fungal and bacterial growth (Castillo et al, 2009). Muskmelons can be placed
on plastic cups to raise them away from direct soil contact and melons can be turned during
growth to limit the occurrence of the ground spot and sunburn as these cause discolorations
of the rind.

Determining the maturity of melons varies among the varieties and may include any of visual
or sensory indicators and sugar content. When cantaloupes are ripe the stem pulls away
easily from the fruit with the formation of an abscission zone or “slips” from the vine leaving a
dish-shaped scar. Industry uses terms to describe the stage of maturity and slip development
e.g. “half slip” or “¾ slip”. Melons such as honeydew do not slip from the vine and ripeness is
indicated by softness at the flower end of the fruit. Watermelons have a small curled tendril
on the end that becomes brown and dies when the fruit is ripe and in addition the ground
spot changes colour and the skin changes hue. In deciding when to harvest the time
between harvest and consumption is critical, for example, cantaloupes may be harvested
between ¾ and full slip and watermelons destined for distant markets are harvested when
mature but not fully ripe to counter against handling damage and loss of quality attributes
(Boyhan et al, 2000).

Melons vary in storage life after harvest and can be only a few weeks for some cantaloupe
varieties in uncontrolled storage conditions (Krarup et al, 2009). Soon after harvest it is
necessary to remove the “field” heat and to control temperature and relative humidity until
reaching markets or the quality begins to deteriorate and the shelf life is reduced. Melons can
be susceptible to chilling injury for more than short periods at very low temperatures. Melons
are cooled using cold water, cold air, or ice and cooling methods vary with the melon types
and available facilities.

4.0 Melons and association with microbial foodborne

pathogens
Melons can become contaminated with foodborne pathogens at any point along the food
chain in a similar manner to other fresh fruit and vegetables (FAO/WHO MRA 14). In
addition, there are specific characteristics of melons that influence the risk of pathogen
contamination and the potential for growth and survival of pathogens that are important
considerations in assessing and managing food safety risks.

4.1 Melon structure

The melon rind protects the internal flesh eaten as fresh product in human diets. Melons are
marketed either whole with the rind intact, portioned with the rind intact or with the rind
removed and flesh sliced or chopped into pieces. The structure of melon rind differs among
the species and varieties and is loosely divided into 2 groups based on the rind topography.
Watermelons and some other melons have a smooth surface while others have a rough,
 12

corrugated surface referred to as “netted” (Gerchikov et al, 2008). The net is a network of
suberized periderm tissue formed in response to natural cracking of the fruit surface during
its enlargement. The degree of netting can vary among netted varieties.

In studies of the efficacy of washing melons it was observed that the rough netted skinned
cantaloupe retained inoculated S. enterica serovars to a greater extent than smooth skinned
honeydew (Castillo et al, 2009; Parnell et al, 2005). The netted melon surface is waxy and
highly hydrophobic and has been associated with enhanced attachment and resistance to
detachment of Salmonella (Ukuku and Fett, 2006). The strength of attachment of bacterial
pathogens such as Salmonella, E. coli (O157:H7 and non-O157:H7), and L. monocytogenes
on cantaloupe rinds involves a linear correlation between bacterial cell surface
hydrophobicity, negative charge and positive charge (Ukuku and Fett, 2002). Further, the
concentration of competing natural bacterial flora plays a role as an increase in attachment
of Salmonella (Ukuku, 2006) and L. monocytogenes and a slower decline in L.
monocytogenes was observed on inoculated cantaloupe rind following sanitization (Ukuku et
al, 2004).

4.2 Melons and microbial growth and survival

4.2.1 Whole melons and rind

Pathogens are able to survive and grow on melon rind and flesh. Del Rosario and Beuchat
(1995) inoculated E. coli O157:H7 on watermelon and cantaloupe rinds and found after
storage at 50C significant population decreases occurred up to 4d then a slower decline until
8-14d (Del Rosario and Beuchat, 1995). When the inoculated rind was held at 250C growth
was observed, more on cantaloupe than watermelon, increasing in the first 4d and then
remaining constant for a further 14-21d. However, the authors noted that the nutrients in the
suspending medium for the inocula together with high humidity may have influenced the
result. They also note this may equate to a scenario with faecal contamination. Beuchat and
Scouten, (2004) reported S. Poona inoculated onto intact cantaloupe rind, wounds and stem
scars, survived unchanged when held between 2h and 24h at 40C and grew in wound scars
within 24h when held at 210C and 370C.

Annous et al (2004) reported similar behaviour of S. Poona inoculated on cantaloupe rinds

where at 40C, more than a 1log cfu/cm2 decrease occurred by 72h. At 200C about 2log
cfu/cm2 increases occurred in the first 24h and the population stabilizing during a further 48h
(Annous et al, 2005a). They suspended their inoculum in water and therefore considered
added nutrients were not essential for growth. In the same inoculation experiments, a generic
E. coli strain did not grow at 200C suggesting E. coli may not be a suitable surrogate for the
presence of Salmonella. Behrsing et al (2003) found S. Salford and E. coli inoculated on
whole cantaloupe (70C for 7d) and honeydew melons (120C for 1d then 80C for 7d) survived
with no growth. In contrast, L. innocua increased about 2log cfu/mL after 7d at 80C (Behrsing
et al, 2003) although growth was arrested by refrigeration below 40C.

In experimental studies washing melons when inocula of S. enterica were dried on the rinds
for 1 h at ambient temperature (14oC) a log greater reduction was observed on honeydew
compared with cantaloupe rind (Parnell et al, 2005). Annous et al (2005b) demonstrated
evidence of fibrillar material formed by S. Poona inoculated on cantaloupe rind after holding
at 2h and biofilm formation by 24h at 200C and at 100C (Annous et al, 2005b). The meshwork
of the netted rind observed by scanning electron microscopy revealed a large number of
attachment sites, crevices and pits that could protect inoculated microorganisms (Annous et
al, 2005b; Parnell et al, 2005).
 13

4.2.2. Flesh and pulp

Bacterial pathogens are able to survive and grow in the low acid environment of melon flesh
or pulp. Fredlund et al (1987) in the 1980’s demonstrated the ability of S. sonnei to grow
rapidly when injected into watermelon reaching 8.0 – 9.0 log cfu/g in 3d at 20 or 300C (cited
by Castillo et al, 2009). Most studies have been based on S. enterica serovars inoculated on
fresh-cut cantaloupe, honeydew and watermelon and in general the inocula survived at 4 or
50C, growth was retarded at 100C and from about 200C inocula could reach hazardous levels
in 4-6h and up to 7days depending on the initial contamination level (Ukuku and Sapers,
2001; Golden et al, 1993; Escartin et al, 1989). E. coli O157:H7 behaved similarly when
inoculated on cantaloupe and watermelon cubes (Del Rosario and Beuchat, 1995). High
levels can be attained before spoilage is apparent.

Following inoculation of muskmelon (av. pH 5.87) and watermelon pulp (av. pH 5.50), S.
Enteritidis populations increased at 10, 20 and 300C on muskmelons with generation times of
7.31, 1.69, 0.69h respectively and on watermelon with generations times of 7.47, 1.60, 0.51h
respectively (Penteado and Leitão, 2004a). Suspensions of S. Typhi inoculated onto
watermelon suspended in water and held at 220C were reported by Escartin et al (1989) to
have a similar generation time of 1.3 (cited by (Penteado and Leitão, 2004a).

Differing from the Enterobacteriaceae is C. jejuni that has been shown to survive although
not grow on watermelon stored at 25-290C for 6h (Castillo and Escartin, 1994). In contrast,
psychrophilic L. monocytogenes inoculated on honeydew (pH 5.8) increased 4.6 logs at 100C
after 7 days (Leverentz et al, 2003). Generation times of 7.12, 1.74 and 0.84h and lag times
of 24, 6,and 4h have been reported when 2 log cfu/g L. monocytogenes was inoculated in
melon (C. melo var Valenciano amarelo) pulp and held at 10, 20 and 300C respectively
(Penteado and Leitão, 2004b). The same authors reported generation times in watermelon
pulp were 13.03, 2.17 and 1.0h and lag times of 24, 28 and 4h at 10, 20 and 300C
respectively. The average pH of the melon and watermelon pulps was 5.87 and 5.50.

Relative humidity has an additional impact on microbial survival at a particular temperature.

When inoculated pieces of produce were held at 18-260C in a controlled environmental
chamber with low (mean 45.1 – 48.4%) and high (mean 85.7 – 90.3%) relative humidity
microbial survival was significantly greater on edible melon pieces than lettuce and bell
peppers (Stine et al, 2005). Among the inoculated organisms, E. coli and feline calicivirus
had the highest and HAV, coliphage and Clostridium perfringens had the lowest inactivation
rates on edible melon flesh.

4.3 Infiltration of microorganisms into melons

The ability of microorganisms to infiltrate the integument of fresh fruits and vegetables has
become of increasing interest as internalisation of the organisms has the potential to offer a
medium for amplification or survival and protection from removal by washing and from
exposure to sanitizers (Delaquis & Austin, 2007). Infiltration is referred here to lodgement of
the microorganisms within the subsurface tissues or further internalisation into the flesh.

Suslow et al (2010) studied the feasibility of the uptake of inoculated Salmonella via the root
system of cantaloupe and honeydew vines and contamination of the fruit under experimental
conditions with extraordinary inocula concentrations (≥7log cfu) not expected in natural field
conditions (Suslow et al 2010). Systemic dose dependent uptake by vines was demonstrated
in the greenhouse; however, no transfer to fruit was observed, and the internalised
Salmonella died off over 2 weeks. Uptake was variable with cultivar and growth conditions
and the researchers could not rule out the possibility of cells entering a viable but non-
culturable state. Under field conditions internalisation was not demonstrated in vines or fruit
although under the extraordinary levels of experimental contamination of soil, contamination
 14

of fruit occurred. Similar transient uptake following exposure to extraordinary numbers of E.

coli O157 was observed (Suslow et al, 2008).

Direct entry points for fruit include wounds caused by physical damage or pests, splits and
fissures and the stem scar. Infection of cantaloupe with phytopathogens on the rind surface
and wound sites has been shown to enhance the survival and internal migration of S. Poona
(Richards and Beuchat, 2005a). The phytopathogens effectively raised the pH of the tissues
with the distance from the rind surface and S. Poona populations per sample increased up to
4logs during storage at 200C over 14d (Richards and Beuchat, 2005b). In a small survey of
market cantaloupe, those with soft rot (87% of 8 samples) were found to be Salmonella
positive more often compared with healthy cantaloupes (47% of 17; Wells and Butterfield,
1997).

Infiltration of pathogens into whole fruit can be enhanced as a result of a negative

temperature differential for example where the temperature of the fruit is higher than the
temperature of the pathogen contaminated water in which it is immersed. Infiltration of dye
into cantaloupes during hydrocooling and of S. Typhimurium into cantaloupes during post-
harvest processing has been demonstrated (Castillo et al, 2009). Infiltration of S.
Typhimurium up to 5mm under the rind has been demonstrated primarily through the ground
spot where the netting is underdeveloped, and, secondarily, through the stem scar (Suslow,
2004 cited by Castillo et al, 2009). However, Richards and Beuchat, (2004) found conflicting
results with temperature dependent infiltration with experimentally inoculated cantaloupes
with varying density of netting and concluded the effect of temperature alone was obscured
by the complexity of the netting and bacterial interactions (Richards and Beuchat, 2004).

While studying approaches to disinfection of melon pieces, Perni et al (2008) found E. coli
were able to migrate through melon tissue with an estimated velocity of around 300µm /min.

5.0 Melon production chain and risks of contamination with

foodborne pathogens
The main operational units in the melon supply chain are as follows:

1. Production and harvest operations

2. Post-harvest operations
3. Fresh-cut/value added processing
4. Distribution/ transport
5. Consumer/retail/food service

These units are discussed with regard to microbial food safety risks and their control specific
to melons. The Risk Assessment Series Report, MRA14, (FAO/WHO, 2008) on microbial
hazards in fresh leafy vegetables and herbs should also be consulted as it contains both
specific information on that commodity as well as information generally applicable to all fresh
produce on each of these topics.

5. 1 Production and harvest operations

Several authors have reviewed the potential sources of pathogen contamination and
preventive approaches to their control pre-harvest for fresh produce in general (Doyle and
Erickson, 2011; FAO/WHO, 2008) and specifically for melons (Castillo et al, 2009, Bowen et
al, 2006). The most important inputs in the production environment for fresh produce include
wildlife, livestock, human activity and wastes, water, soil and soil amendments, seeds, plant
stocks and equipment (FAO/WHO, 2008). Application of Good Agricultural Practices (GAPs)
is appropriate for control in this sector and should be considered together with specific
 15

guidance for melons. The draft Annex for melons in principle follows closely the Annex for
Leafy Vegetables and Herbs (CAC, 2010) for which the FAO/WHO (2008) MRA series 14
provides the scientific evidence. Specific risk factors, available evidence and mitigations for
melons are provided.

5.1 Growing site

• Environmental conditions favourable for melon cultivation are favourable for a variety
of wildlife and insect pests and favourable also for the survival and possible growth of
microbial pathogens. Melons and associated field waste are attractants to wildlife as
a ready source of food (Castillo et al, 2009).
• In an investigation of melon growing sites wildlife has been found to carry S. enterica
and nearby river water has been shown to have positive bacterial faecal indicators
and to be contaminated with S. enterica serovars (Aguillar et al, 2005; Gagliardi, et al,
2003). An assessment of the risks associated with the growing site location should be
undertaken with particular attention given to flooding (particularly in high rainfall
areas) and run off from high risk sites, to evidence of wildlife and insect pest
presence, and to the proximity to wildlife and pest reservoirs. Re-location or
measures to eliminate or reduce the risk may be required (FAO/WHO, 2008).
• Growing areas should be protected and maintained to avoid attraction of wildlife and
pests for example water puddles and waste accumulation. Pathogens could survive
and grow in waste from prior or current harvests or vine maintenance. In melon
growing environments S. enterica serovars have been detected in reptiles such as
iguanas that were attracted to and fed on the melons and at the same time defecate
in the vicinity (Aguillar et al, 2005). This pathogen has also been detected in water
puddled in furrows (Gagliardi et al, 2003).
• Monitoring for the presence of wildlife, pests and domestic animal intrusion should be
undertaken regularly and at harvest and reported by field staff. If detected, decisions
are required on whether to harvest melons from affected areas (FAO/WHO, 2008).

5.2 Soil and soil amendments

• Pathogens such as S. enterica have been detected in soil in melon growing fields
where the distribution can vary widely (Espinoza-Medina et al, 2006; Gallegos-
Robles et al, 2009).
• The unique characteristic of melon rind and the netted surface of some varieties
give them a special ability to attach microorganisms and subsequently protect
them from removal as described in section 4.2. Pathogen infiltration is possible
especially at the ground spot and wounds or abrasions to the rind, and fruit
nearing and at maturity can support pathogen growth and/or survival (Section
4.3). Pathogen uptake via roots appears feasible although considered an unlikely
and transient event (Suslow, 2010). However, if contaminated faecal material is
deposited randomly with a high pathogen load this could lead to a sporadic
contamination event.
• Foodborne pathogens can survive in soil e.g. L. monocytogenes, S. enterica and
E. coli O157, and viruses for up to 8, 23 and 3 weeks respectively (Bowen et al,
2006). Pathogens can be introduced in soil amendments and their presence can
be facilitated by insects and soil creatures such as nematodes.
• Melon vines are often grown on plastic mulch or the fruit on plastic pads or cups
to reduce soil contact and ground spots. At the same time this reduces direct
exposure to soil contaminants. These devices must be maintained in a sanitary
manner if used.
• Use of an appropriate irrigation system such as furrow or drip can minimise soil
contamination and irrigation water and melons is discussed further below
(FAO/WHO, 2008).
 16

• Suslow et al (2010) demonstrated in field studies with soils experimentally

inoculated with exceptional large inocula via furrows that fruit could be surface
contaminated after a significant rain event. Following heavy rain the risk should
be assessed before deciding on harvest time, and on the operating condition
required for washing in the presence of an increased soil load.

5.3 Water and irrigation

• Irrigation water can be a vehicle for exposure of vines, fruit or root systems to
pathogens. Suslow et al, (2010) inoculated large numbers of S. enterica into soil via
furrow and drip irrigation systems during cultivation of cantaloupes and found while
the inocula could be detected for the duration of the growing season in soil the
inoculated bacterium could not be detected in the vines or fruit at harvest. Rind
surface of furrow irrigated fruit was contaminated during heavy rain only.
• Evidence is available that irrigation water for melons can be contaminated with faecal
indicator bacteria and S. enterica at both the source and in holding ponds (Aguillar et
al, 2005; Castillo et al, 2004; Gagliardi, et al, 2003). Surface waters, poorly
maintained wells and irrigation canals were shown to be contaminated.
• Duffy et al (2005) detected E. coli in 39.4% of 179 irrigation water sources for
cantaloupe, parsley and oranges with a mean count of 0.4 log cfu/ml. Well water was
most frequently positive (10/10) followed by waters from reservoir (15/30), and
riverine (9/30) sources. The well and reservoir waters had the highest E. coli counts
of 0.7±0.3 and 1.0 ± 0.7 log cfu/ml. Cement irrigation canals were significantly less
contaminated than dirt canals.
• Castillo et al (2004) similarly detected S. enterica and E. coli, 15% and 16%
respectively, in irrigation water sources on 6 cantaloupe farms. A farm using water
from an irrigation canal accounted for most of the positives compared with the others
using well or pond water. Farms drawing water from the same primary riverine
source were similarly contaminated. Filtering was effective in reducing the number of
E. coli positives in water at some farms although the method was not described.
• Ground water (1/11), irrigation water (4/17), soil (2/24) and in-field cantaloupes (9/35)
were positive for S. enterica using enrichment culture and the Polymerase Chain
Reaction (PCR) in investigation of 5 commercial farms in Mexico in 2003-4
(Espinoza-Medina et al, 2006)
• In the study of Duffy et al (2005) S. enterica was detected in 16 irrigation water
samples, the frequency of source being reservoir, dirt canals, furrow, cement canals
and no positive well or riverine irrigation waters.
• Interestingly the S. enterica serovars detected in irrigation water and those on
melons at the same farm can be different as can the serovars in washing water
(Duffy et al, 2005; Castillo et al, 2004). This raises questions of the contamination
source pre- and post-harvest and whether methodological insensitivities due to low
prevalence and concentration may limit investigations.

5.4 Human activity

Where field workers turn melons attached on the vine as it matures to avoid ground
spots developing and sunburn an opportunity arises for introduction of pathogens in
conditions of poor hygiene.
• Field workers hands have been shown to be contaminated with E. coli (Castillo et al,
2004). In a survey of a small number of field staff working in a field with contaminated
soil and cantaloupes no evidence of Salmonella was found in 24 hand samples
(Espinoza-Medina, 2006).
 17

• Gloves may be used; however, the use of wool or cotton gloves have been shown to
be a source of contamination and a program for maintaining hygiene is required
where they are used (Castillo et al, 2009).
• Care should be taken at harvest, whether manual or mechanical, to avoid
environmental contamination or damage to melons that would introduce pathogens
and increase the risk of microbial infiltration, survival or growth (See Section 4.2).

5.5 Harvest
• At melon harvest an stem scar or any unintentional damage to the rind during
handling can allow invasion and proliferation of pathogens (See section 4.2). Melons
such as watermelons at maturity can be heavy and cumbersome to manually handle.
Careful handling is required to minimise contamination by handlers, water, soil and
other environmental sources.

6.0 Post-harvest
Post-harvest practices vary between regions and countries and with the type of melon.
Watermelons are generally packed and sent to market without pre-cooling although cool
ambient storage temperatures are preferred. Others such as cantaloupe and honeydew
varieties, depending on the stage of maturity, are cooled soon after harvest. Further storage
conditions of temperature and relative humidity can depend on the maturity and ripening
method and are optimised to maintain quality and prolong shelf life. Melons can be either
packed in the field or transferred to a designated packing facility where cooling, rinsing, and
washing and sanitizing can take place (Castillo et al, 2009).

6.1 Packing facilities

• Wildlife such as rodents, reptiles and birds frequent these facilities which may be of a
temporary open-air design and intermittently used where production is seasonal.
Damaged melons and waste are attractants for pests and can also be a reservoir
where bacterial pathogens can grow (Duffy, 2005; Castillo et al, 2004). This
emphasises the need for hygiene maintenance, pest control and a program with
Standard Operating Procedures (SOPs) established. The fittings and environment of
cantaloupe packing facilities have been found to be contaminated by S. enterica.
These include walls and floors of cooling rooms (Castillo et al, 2004) and surfaces to
which washed melons are exposed (Duffy et al, 2005) E. coli had been detected on
surfaces such as boxing ramps, conveyor belts, plastic bags and bins used for
harvesting, receiving hopper, transport trailer and an unloading ramp (Duffy et al,
2005).
• Fruit should be culled to remove those with damage or fungal rot and handled to
prevent further damage due to the risk of microbial infiltration and growth (See
section 4.2).
• Workers are a potential source of contamination. The hands of 3/60 field and packing
plant workers surveyed on cantaloupe farms were contaminated with E. coli although
salmonellas were not detected (Castillo et al, 2004). In another small study neither E.
coli nor S. enterica as detected on hands or gloves of 10 workers (Duffy et al, 2005).
Espinoza-Medina et al, (2006) detected Salmonella using the PCR on the hands of
4/24 packing house workers handling melons also found to be contaminated (7/34)
• Poorly controlled hydrocooler water was observed to have significant levels of faecal
indicators and to contaminate cantaloupe rinds with up to 3.4 log cfu/g (Gagliardi et
al, 2003). The process may result in infiltration of microorganisms (See section 4.3).
 18

6.2 Washing and sanitising

As the rind of whole melon, in particular in netted varieties, is identified as a major source of
the most common bacterial pathogens in outbreaks, considerable effort has been applied to
reduce the risk at this point in the supply chain. Washing removes loose soil and can improve
visible cleanliness; however, it can provide a mechanism for both the spread (Parnell et al,
2005) and introduction of pathogen contamination (Bowen et al, 2006). The need for washing
varies with the farm location; for example, in more arid regions fruit is field packed while in
humid areas washing and fungicide application that helps control plant pathogens is
practiced (Gagliardi et al, 2003).
• In several studies it has been found pathogens (S. enterica) were introduced and the
bacterial load of aerobic bacteria (Akins et al, 2008), E. coli (Duffy et al, 2005; Castillo
et al, 2004), faecal coliforms and faecal enterococci (Gagliardi et al, 2003) on
cantaloupe melons was increased between pre- and post- harvest. Whether
processing released bacteria from the netted rinds and the extent of introduction of
contamination or both were involved is not clear (Duffy et al, 2005).
• Water is used in abundance and can be a source of microorganisms if not treated to
approximate potable quality. Contamination of source water has been mentioned
above. Gagliardi et al (2003) found much of the contamination during cantaloupe
processing could be traced to primary wash tanks and hydrocoolers.
• Cross-contamination between melons is a high risk when melons are co-mingled in
dump tanks. Factors such as high biological load, turbulent flow, temperature and pH
can impact on the concentration and efficacy of residual sanitizers and have to be
monitored and controlled (Gagliardi, 2003). Washing melons in water in the absence
of a sanitizer has been demonstrated experimentally to spread contamination in wash
tanks (Parnell et al, 2005). This has resulted in some risk managers choosing to
eliminate use of dump tanks in certification conditions (Alvarado-Casillas, 2010).
• Control of the temperature differential between wash water and melons is important
to avoid risks of infiltration of pathogens into melons and has been discussed in the
section 4.3.
• Experimentally sanitized and inoculated cantaloupe rind has been shown to retain
higher populations of Salmonella (Ukuku, 2006) and L. monocytogenes (Ukuku et al,
2004) than untreated controls and the more netted melon varieties retain larger
concentrations. Therefore once sanitization has reduced microbial populations, there
is an increased risk of contamination of melon rinds emphasising the importance of
maintaining the cleanliness and hygiene all contact surfaces post-sanitization.
• Soaking cantaloupe and honeydew inoculated with S. enterica in water for 60 sec
resulted in 0.7 and 2.8 log decreases in the inocula respectively (Parnell et al, 2005).
The greater amount of organic matter associated with netted rinds is believed to
interfere with sanitizers (Parnel et al, 2005) and experiential use of a surfactant with
sanitizers significantly increased washing efficiency (Bastos et al, 2005).
• There have been numerous studies of the use of sanitizers and alternate approaches
to decontaminating melons. It is generally agreed that sanitizers control microbial
populations in the wash water rather than on the melons (Castillo et al, 2009;
FAO/WHO, 2008). Sanitizers have some effect and add to the through chain risk
reduction although cannot be relied on to eliminate pathogens and are least effective
for the netted varieties. High risk of re-contamination after washing and sanitizing has
been mentioned.
• Mechanical removal of pathogens with scrubbing has been experimentally evaluated
for both industrial and domestic use (Parnell et al, 1005). Scrubbing is more effective
than soaking alone and scrubbing for 60 sec reduces the bacterial load. However,
unless a sanitizer is added to wash water spread to other melons and sites on a
melon will occur. Washing and scrubbing has been recommended for domestic
households by authorities (Parnell et al, 2005). For industrial use water disinfection
 19

and sanitary maintenance of brushes and equipment is also essential to prevent

cross-contamination.
• Approaches to decontaminating whole fruit include using chemical agents in wash
water, gaseous ozone, non-chemical (heat, irradiation, cold atmospheric plasma),
bacteriocins, biocontrol using bacteriophages and lactic acid bacteria, and new
approaches continue to evolve (Bowen et al, 2005; Castillo et al, 2009). In modified
atmosphere cold storage of cantaloupes S. enterica growth was inhibited (Salgado
et al, 2009); however, L. monocytogenes would need to be considered. The
combination of agents can have an additive hurdle effect and increase efficacy;
however, these have to be assessed considering economical and practical factors.
The efficacy will depend on the bacterial load of the incoming melons, melon type and
the facility and validation of any process will be necessary.
• The time between whole fruit decontamination and processing for fresh-cut products
has been shown to be important. Sanitizing cantaloupe reduced transfer of S.
enterica to cut pieces; however, if the melons were sanitized then stored for several
days at 4 or 200C transfer of S. enterica was observed (Ukuku and Sapers, 2002).
This has implication for process scheduling in the fresh-cut industry and requires
further assessment (Castillo et al 2009).

6.3 Processing
There is strong epidemiological evidence that during preparation of melons for consumption
there is a potential for increasing the risk of foodborne illness (Section 2.0). There are no
further risk reduction steps as these products receive no further microbiocidal treatment
before consumption. This supports the rationale for managing the risk through chain and
preventing contamination at production to the greatest extent possible.

Contamination at this operation point can be introduced from the melon rind, from food
handlers, the preparation environment (cutting boards, knives) and cross-contamination with
other melons or foods. Poor temperature control between contamination during preparation
and consumption can amplify the risk for several bacterial pathogens. Some important
evidence follows.
• While the evidence is variable washing, scrubbing and sanitizing melons before
preparation will result in some decrease although not ensure elimination of
pathogens.
• It has been demonstrated experimentally that pathogens can be transferred from the
rind to the internal edible flesh of the melon (Ukuku et al, 2005; Ukuku and Sapers,
2001). Where sanitizers had lowered rind contamination the transferred population
was similarly reduced. Studies have relied on inoculated samples and how this
relates quantitatively to naturally contaminated samples is less clear. As the netted
cantaloupes have the highest potential for rind contamination and microbial
concentration they present the greatest risk and this is supported epidemiologically.
• Castillo et al (2009) quote Vadlamundi (2004) who found experimentally that the
sequence of skin removal and cutting was important as cutting after rind removal
resulted in less contamination than cutting then removing the rind (Castillo et al,
2009)
• Decontamination of melon pieces for the fresh-cut market is a consideration with
similar approaches taken as for whole fruit with the need for additional consideration
of maintaining sensory quality. An approach suited to cut product is the use of edible
coatings containing essential oils where some are considered to have potential and
accepted sensory effects (Raybudi-Massilia et al, 2008).
• Storage temperature of cut melons and the duration between cutting and
consumption is critical and growth and survival on melon tissue has been discussed
(See section 4.2). Cantaloupe has been reported to support survival of S. enterica
better than honeydew and watermelon. Holding contaminated fresh-cut product at
 20

220C for 3h prior to refrigerated storage was concluded to increase risk of proliferation
of salmonellas (Ukuku and Sapers, 2007) emphasising cut products of all melons
should be chilled as soon as possible.

7.0 Consumers
Melons have become popular as healthy, fresh, convenient and delicious foods that are
hugely diverse in their use in dishes that appeal to all age groups in all cultures. Given their
utility, they are a popular choice in the home, food service and catering industries and are
retailed whole, portioned or as ready-to eat salads or meals. The available evidence
suggests that there will be a low risk melons can be contaminated in the field and that
through the food chain this risk should be decreased or at the least not increased.

Melons are low acid fruits and their soft texture makes them appealing to the young, elderly
and infirm. The epidemiological evidence provided indicates that these vulnerable groups are
at increased risk when their food is prepared in an institutional setting (Section 2.0). Among
the end users of this commodity, these vulnerable groups and those preparing their food are
a priority in education on safe handling of melons.

As melons have a protective rind that is peeled and not eaten it is not surprising that
consumers assume the edible flesh could not be contaminated. In a survey of 2,000
consumers in the U.S.A. 35% indicated they did not wash melons before preparation, 9%
thought they were already clean and 16% thought it was not necessary as they did not eat
the skin (Li-Cohen and Braun, 2000).

Authorities have taken initiatives often in response to foodborne illness outbreaks linked with
melons to inform consumers and provide guidance to targeted food service industries on
safety handling of melons through consumer information channels such as websites with
guidelines and fact sheets. While most food safety programs will address the hazards and
their management during preparation of melons, there are specific characteristics that need
emphasis. In particular, these include washing, scrubbing and sanitizing before use,
contamination during cutting and temperature control.

Industry, retailers and food service suppliers of fresh-cut products need to provide clear
instructions for end users of their products on safe storage, shelf life and handling of their
products (FAO/WHO, 2008).

8.0 Microbiological sampling

Under GAPs, Good Hygienic and Manufacturing practices, and food safety programs, the
presence of pathogenic microorganisms in fields, water and melons appears of low incidence
and concentration and with heterogeneous distribution. Whether the approach to sampling of
foods can be applied in sampling soils and melons in fields in not clear.

Observations have been made on bacterial indicators for faecal contamination and in
particular the presence of Salmonella in melons. Annous et al, (2005a) observed S. enterica
survived better than E. coli at 4 and 200C and suggested E. coli was not a suitable surrogate.
Stine et al, (2005) investigated the use of C. perfringens as an indicator of faecal
contamination of fresh produce including melons, lettuce and bell peppers. They concluded
that C. perfringens may be a more acceptable indicator of bacterial contamination and
survival in various environments and different types of crops.

The methodology used for pathogen detection with melons varies widely in experimental
studies and surveys. Sampling methods include sponging, incising, including skin and/or
 21

pulp, blending, massaging and rinsing. Hammack et al (2004) compared sampling methods
and recommended the rinse method used in regulatory control in the U.S.A. Melons are
soaked in a non-selective broth at 350C for 24± 2hr before selective enrichment. Molecular
methods such as the PCR have been developed and compared with conventional culture for
pathogen detection on melons. PCR for Salmonella detection in selective enrichments can
yield larger number of positive results although the viability of the target bacterium may be
unknown (Gallegos-Robles et al, 2009; Espinoza-Medina et al, 2006).

Gallegos-Robles et al (2009) collected cantaloupes with soil still attached in quadrants of 4

fields, (25 per field) in Mexico in 2005. They detected S. enterica in 9 surface washings by
conventional enrichment culture and 11 using a PCR of the enrichment. The distribution of
positive melons varied between the quadrants of each field and between fields with 5, 5 and
2 positive samples from 3 fields by either method and none from the fourth. The application
of GAPs in the latter field may have been the reason for the result although the study was not
designed to assess this.

Surveys of pathogens in melons from the field often fail to detect their presence while low
detection rates are reported after packing. In a study of cantaloupes on farms in the U.S.A.
and Mexico, S. enterica was detected in 1/475 field samples and 2/325 samples from the
cooler or after packing. E. coli at counts too low for analysis were detected in 12/475 and
42/325 of the same samples as tested for S. enterica respectively (Castillo et al, 2004).
Similarly in another study, no salmonellas were detected in 100 field collected cantaloupes
while 3/100 were positive in the packing shed; however, the mean log E. coli cfu/melon were
similar, 2.2 ± 0.8 and 2.1± 0.7 (Duffy et al, 2005). Espinoza-Medina et al, (2006) detected
Salmonella in in-field cantaloupes (9/35) and packed cantaloupe (7/34) using PCR although
not using conventional culture suggesting the concentration was low.

Surveys have been conducted on domestic and imported produce including melons. Where
the sample size is small the probability of detection of pathogens may be small under good
hygienic practices. In a survey of imported foods in New Zealand, 50 samples were found of
satisfactory quality as ready-tot-eat products (McIntyre and Cornelius, 2009). In a U.S.A.
FDA survey of imported fresh produce in1999, 11 of 151 cantaloupes were contaminated
with Shigella (3) and Salmonella (8; FDA, 2011). Ongoing surveillance in the U.S.A. of
domestic and imported melons includes significantly larger sample sizes and between 2005
and 2010 less than 5 and in two exceptions 16 and 17 Salmonella positives were detected
when testing from 1,000 to >2,000 samples per year (U.S.A. JEMRA response).

Conclusions
Melons are a popular fruit included in the human diet worldwide. Their taste, texture,
versatility and healthy characteristics along with the convenience of fresh melon have
resulted in increased consumption, production and trade. Melons have inherent
characteristics that render them susceptible to contamination and potential vehicles for
foodborne illness transmission. Fresh melons receive no processing along the food chain
that will totally eliminate any contaminating foodborne hazards. Control has to begin at
primary production and continue through to the consumer such that the low level of risk that
may be present on farm is reduced or at the least is not increased before consumption.
 22

9.0 References
Aguillar, G.A.F., Ramirez, M.G. Garcia, A.M., González, R.Y, Navarrete, Escutia, M.C.S. and
Madrigal, J.C. 2005. Identification of Salmonella spp. in water, cantaloupe melons and
iguana faeces in a melon orchard. Med. Int. Mex. 21:255-258
Akins, E.D., Harrison, M.A., Hurst, W. 2008. Washing practices on the microflora on Georgia-
grown cantaloupes. J. Fd. Prot. 71:46-51
Alvarado-Casillas, S., Ibarra-Sanchez, L.S., Martínez-Gonzáles, N.E., Rodríguez-García,
M.O., Castillo, A. 2010. Validation of a washing and sanitizing procedure for
cantaloupes at a Mexican packing facility. J. Fd. Prot. 73:362-365
Annous, B.A., Sapers, G.M., Jones, D.M., Burke, A. 2005a. Improved recovery procedure for
evaluation of sanitizer efficacy in disinfecting contaminated cantaloupes. J. Fd. Sci. 70:
M242-M247.
Annous, B.A., Solomon, E.B., Cooke, P.H., Burke, A. 2005b. Biofilm formation by Salmonella
spp. on cantaloupe melons. J. Fd. Safety. 25: 276-287.
Annous, B.A., Burke, A., Sites, J.E. 2004. Surface pasteurization of whole fresh cantaloupe
inoculated with Salmonella Poona or Escherichia coli. J. Fd. Prot. 76:1976-1885.
Avendano, B., Narrod, C., Tiongco, M. 2009. Food Safety Requirements for Cantaloupe
Exports from Mexico and their Impact on Small Farmers’ Access to Export Markets.
Internat. Fd. Pol. Res. Instit. IFPRI Discussion Paper, May 2009.
Bastos, M.D.S.R., de Fátima Ferreira Soares, N., José de Andrade, N., Cristina Arruda, A.,
Elesbão Alves, R. 2005. The effect of the association of sanitizers and surfactant in the
microbiota of the Cantaloupe (Cucumis melo L.) melon surface. Fd. Contr. 16:369-373.
Behrsing, J., Jaeger, J., Horlock, F., Kita, N., Franz, P., Premier, R. 2003. Survival of Listeria
innocua, Salmonella Salford and Escherichia coli on the surface of fruit with inedible
skins. Postharvest Biol. Technol, 29: 249-256.
Beuchat, L.R., Scouten, A.J. 2004. Factors affecting survival, growth, and retrieval of
Salmonella Poona on intact and wounded cantaloupe rind and in stem scar tissue. Fd.
Microbiol. 21:683-694.
Bowen, A., Fry, A., Richards, G. and Beuchat, L. 2006. Infections associated with cantaloupe
consumption: a public health concern. Epid. Infect. 134: 675-685.
Boyhan, G.E., Granberry, D.M., Kelley, W.T. 2000. Commercial watermelon production,
Cooperative Extension Service, The University of Georgia College of Agricultural and
Environmental Sciences.
http://www.agmrc.org/media/cms/B996_B3D54FD90A36C.pdf).
Castillo, A., Escartin, E.F. 1994. Survival of Campylobacter jejuni on sliced watermelon and
papaya. J.Fd.Prot. 57:166-168.
Castillo, A., Martinez-Téllez, M.A., and Rodríguez-García, M.O. 2009. Melons. Chapter 9. In
Sapers, G.M., Solomon, E.B. and Matthews, K.R. (eds) The Produce Contamination
Problem: Causes and solutions. Elsevier, Burlington, USA.
Castillo, A., Mercado, I., Lucia, L.M., Martínez-Ruiz, Y., Ponce De León, J., Murano, E.A.,
Acuff, G.R. 2004. Salmonella contamination during production of cantaloupe: A
binational study. J. Fd. Prot. 67:713-720.
CAC..2010. Alinorm 10/33/13 Appendix II. Proposed draft annex on fresh leafy vegetables
(Annex to the Code of Hygienic Practice for Fresh Fruits and Vegetables). p37.
Delaquis, P. and Austin, J.W.2007. The effect of heat treatments on the fate of foodborne
pathogens in horticultural produce. Stewart Post Harvest Review. Review 3, art. no. 3.
Del Rosario, B.A., Beuchat, L.R. 1995. Survival and growth of enterohemorrhagic
Escherichia coli O157:H7. J. Fd. Prot. 58:105-107.
Doyle, M. P., Erickson, M. 2011. Opportunities for mitigating pathogen contamination on-
farm. Internat. J. Fd. Microbiol. doi:10.1016/j.ijfoodmicro.2011.02.037.
Duffy, E.A., Lucia, L.M., Kells, J.M., Castillo, A., Pillai, S.D., Acuff, G.R.. 2005.
Concentrations of Escherichia coli and genetic diversity and antibiotic resistance
 23

profiling of Salmonella isolated from irrigation water, packing shed equipment, and
fresh produce in Texas. J. Fd. Prot. 68:70-79.
Escartin. E.F., Ayala, A.C., Lozano, J.S. 1989. Survival and growth of Salmonella and
Shigella on sliced fruit. J. Fd. Prot. 52:471-472 (Cited by Penteado and Leitão, 2004a).
FAOSTAT 2011. FAO statistical databases data sets. Available at http://faostat.fao.org/.
FAO/WHO. 2008. Microbiological hazards in fresh leafy vegetables and herbs: Meeting
report. Microbiological Risk Assessment Series 14. FAO/WHO, Rome.
FDA 2001. Survey of Imported Fresh Produce FY 1999 Field Assignment. January 30, 2001.
(Cited 24/05/11 at http://www.fda.gov/Food/FoodSafety/Product-
SpecificInformation/FruitsVegetablesJuices/GuidanceComplianceRegulatoryInformatio
n/ucm118891.htm
Fredlund, H., Back, E., Sjoberg, L., and Tornquist, E. 1987. Watermelon as a vehicle of
transmission of shigellosis. Scand. J. Infect. Dis. 19:219-221.
Gagliardi, J.V., Millner, P.D., Lester, G., Ingram, D. 2003. On-farm and post harvest
processing sources of bacterial contamination to melon rinds. J. Fd. Prot. 66: 82-87.
Gallegos-Robles, M.A, Moraes, A., Alvarez-Oieda, G., Osuna-Garcia, J.A., Martinez, I.O.,
Morales-Ramos, L.H., Fratamico, P. 2009. PCR detection and microbiological isolation
of Salmonella spp. from fresh beef and cantaloupes. J. Fd. Sci. 74:M37-40.
Gerchikov, N., Keren-Keiserman, A., Perl-Treves, R., Ginzberg, I.2008. Wounding of melon
fruits as a model system to study rind netting. Scientia Horticulturae. 117: 115-122.
Golden, D.A., Rhodehamel, E.J., Kautter, D.A.1993. Growth of Salmonella spp. in
cantaloupe, watermelon, and honeydew melons. J. Fd. Prot. 56:194-196.
Hammack, T.S.Valentin-Bon, I.E., Jacobson, A.P., Andrews, W.H. 2004. Relative
effectiveness of the bacteriological analytical manual method for the recovery of
salmonella from whole cantaloupes and cantaloupe rinses with selected preenrichment
media and rapid methods. J.Fd. Prot. 67:870-877.
Krarup, C., Tohá, J. and González, R. 2009. Symptoms and sensitivity to chilling of
cantaloupe melons during postharvest. Chilean. J. Agric. Res. 69:125-133.
Leverentz, B., Conway, W.S., Camp, M.J., Janisiewicz, W.J., Abuladze, T., Yang, M.,
Saftner, R., Sulakvelidze, A. 2003. Biocontrol of Listeria monocytogenes on fresh-cut
produce by treatment with lytic bacteriophages and a bacteriocin. Appl. Environ.
Microbiol. 69:4519-4526.
Leverentz, B., Conway, W.S., Alavidze, Z. Janisiewicz, W.J., Fuchs, Y., Camp, M.J.,
Chighladze, E., Sulakvelidze, A. 2001. Examination of bacteriophage as a biocontrol
method for Salmonella on fresh-cut fruit: A model study. J. Fd. Prot. 64: 1116-1121
Li-Cohen, A.E., Bruhn,C.M. 2002. Safety of consumer handling of fresh produce from the
time of purchase to the plate: A comprehensive consumer survey. J. Fd. Prot. 65:1287-
1296.
Little, C.L., Mitchell, R.T. Microbiological quality of pre-cut fruit, sprouted seeds, and
unpasteurised fruit and vegetable juices from retail and production premises in the UK,
and the application of HAACP. (2004) Communicable disease and public health /
PHLS, 7 (3), pp. 184-190.
McIntyre, L. Cornelius, A. 2009. Microbiological survey of retail fresh produce of imported,
domestic, conventional and domestic organic origin. ESR Client report FW 09064.
Cited 24/05/11 at http://www.foodsafety.govt.nz/elibrary/industry/microbiological-
survey-retail-research-.
projects/FW09064Produce_Survey_Final_Report_30_Sept_09.pdf
MMWR. 2002. U.S. Multistate outbreaks of Salmonella Poona infections associated with the
eating cantaloupe from Mexico, United States and Canada, period 2000-2002. Morbid.
Mort. Wk. Rep. 51:1044-1047.
MMWR. 1999. Reptile-Associated Salmonellosis -- Selected States, 1996-1998. 48: 1009-
1013.
Ooi P.L. Goh K.T., Neo K.S. and Ngan C.C. 1997. A shipyard outbreak of salmonellosis
traced to contaminated fruits and vegetables. Ann. Acad. Med. Sing. 26: 539-443.
 24

Parnell, T.L., Harris, L.J., Suslow, T.V. 2005. Reducing Salmonella on cantaloupes and
honeydew melons using wash practices applicable to postharvest handling,
foodservice, and consumer preparation. Internat. J. Fd. Microbiol. 99: 59-70.
Penteado, A.L., Leitão, M.F.F. 2004a. Growth of Salmonella Enteritidis in melon, watermelon
and papaya pulp stored at different times and temperatures. Fd. Control. 15: 369-373.
Penteado, A.L., Leitão, M.F.F. 2004b. Growth of Listeria monocytogenes in melon,
watermelon and papaya pulps. Internat. J. Fd. Microbiol. 92:89-94.
Perni, S., Shama, G., Kong, M.G. 2008. Cold atmospheric plasma disinfection of cut fruit
surfaces contaminated with migrating microorganisms. J. Fd. Prot. 71:1619-1625.
Raybaudi-Massilia, R.M., Mosqueda-Melgar, J., Martín-Belloso, O. 2008. Edible alginate-
based coating as carrier of antimicrobials to improve shelf-life and safety of fresh-cut
melon. Internat. J. Fd. Microbiol. 121:313-327
Richards, G.M., Beuchat, L.R. 2005a. Infection of cantaloupe rind with Cladosporium
cladosporioides and Penicillium expansum, and associated migration of Salmonella
Poona into edible tissues. Internat. J. Fd. Microbiol. 103:1-10.

Richards, G.M., Beuchat, L.R. 2005b. Metabiotic associations of molds and Salmonella
Poona on intact and wounded cantaloupe rind. Internat. J. Fd. Microbiol. 97:327-339.
Richards, G.M. and Beuchat, L.R. 2004. Attachment of Salmonella Poona to cantaloupe rind
and stem scar tissues as affected by temperature of fruit and inoculum. J. Fd. Prot. 67:
1359-1364
Salgado, P.L., Hernández Anguiano, A.M., García, J.C., Aguilera, G.M., Quiroz, C.C. 2009.
Survival of Salmonella typhimurium on 'Cantaloupe' melon during cold storage under
controlled atmospheres [Sobrevivencia de Salmonella typhimurium en melón
'Cantaloupe' durante el almacenamiento refrigerado en atmósferas controladas].
Revista Fitotecnia Mex., 32:209-215.
Stine, S.W., Song, I., Choi, C.Y., Gerba, C.P. 2005. Effect of relative humidity on preharvest
survival of bacterial and viral pathogens on the surface of cantaloupe, lettuce, and bell
peppers. J. Fd. Prot. 68: 1352-1358.
Suslow, T. Sbodio, A., Lopez, G., Wei, P., Tan, K.H. 2010. Melon food safety: 2010 Final
Report. California Melon Research Board (cited 09 May, 2011 at
http://www.cmrb.org/documents/files/20110110085901.pdf).
Suslow, T. D’lima, C.,Tan, K.H. 2008. Assessment of pre-harvest attachment and
internalization of bacteria into melons from irrigation water. Project Report 2008.
California Melon Research Board (cited at 09 May, 20011
http://www.cmrb.org/documents/files/20081222152813.pdf).
Ukuku, D.O. 2006. Effect of sanitizing treatments on removal of bacteria from cantaloupe
surface, and re-contamination with Salmonella. Fd. Microbiol. 23:289-293.
Ukuku, D.O., Bari, M.L., Kawamoto, S., Isshiki, K. 2005. Use of hydrogen peroxide in
combination with nisin, sodium lactate and citric acid for reducing transfer of bacterial
pathogens from whole melon surfaces to fresh-cut pieces. Internat. J. Fd. Microbiol.
104:225-233.
Ukuku, D.O., Fett, W.F. 2006. Effects of cell surface charge and hydrophobicity on
attachment of 16 Salmonella serovars to cantaloupe rind and decontamination with
sanitizers. J. Fd. Prot. 69:1835-1843.
Ukuku, D.O., Fett, W.F. 2002. Relationship of cell surface charge and hydrophobicity to
strength of attachment of bacteria to cantaloupe rind. J. Fd. Prot. 65:1093-1099.
Ukuku, D.O., Fett, W.F., Sapers, G.M. 2004. Inhibition of Listeria monocytogenes by native
microflora of whole cantaloupe. J. Fd. Safety. 24: 129-146.
Ukuku, D.O., Sapers, G.M. 2007. Effect of time before storage and storage temperature on
survival of Salmonella inoculated on fresh-cut melons. Fd. Microbiol. 24:288-295.
Ukuku, D.O., Sapers, G.M. 2001. Effect of sanitizer treatments on Salmonella Stanley
attached to the surface of cantaloupe and cell transfer to fresh-cut tissues during
cutting practices. J. Fd. Prot., 64:1286-1291.
 25

USDA. 2009. Guidance for industry: Guide to minimize microbial food safety hazards of
melons; Draft Guidance. (Cited 16/05/2011 at http://www.food-label-
compliance.com/Sites/5/Downloads/USFDA-Draft-Guidance-MELONS-issued-073109-
complete-text.pdf)
Wells, J.M., Buterfield, J.E. 1997. Salmonella contamination associated with bacterial soft rot
of fresh fruit and vegetables in the marketplace. Plant Dis. 81:867-872.
Annex 1 Foodborne illness outbreaks associated with melons

ng factors
contributi

Source of
Aetiology

Outbreak
prepared
outbreak

(if stted0
Possible
(deaths)
Country

setting/
Type of

a
+other
melon

foods

No. ill

Origin

place
Year

Data
of

0 b Roadside Gaylor et al, 1955. Cited by

1950 U.S.A. watermelon S. Bareily 6 D Held AT C Castillo et al, 2009
stall
17 Gaylor et al, 1955. Cited by
1954 U.S.A. watermelon S. Miami D Sliced, wrapped Supermarket Home Castillo et al, 2009
(1)
Pre-cut damaged fruit;
1979 U.S.A. watermelon S. Oranienberg D plastic film cover; MMWR, 1979
0
Held AT C possible
1984 U.S.A. cantaloupe unknown 12 Bowen et al, 2006
MMWR, 1986 (cited by
1985 U.S.A. cantaloupe unknown 77 Bowen et al 2006)
1985 U.S.A. cantaloupe C. jejuni 16 Bowen et al, 2006
Purchased whole;
1987 Sweden watermelon S. sonnei I suspect food fraud, Home Fredlund et al, 1987
inoculation with water
United c
1987 melon Norovirus 206 Infected food handler ACMSF, 2005
Kingdom
cantaloupe +
1989 U.S.A. honeydew and unknown 101 Bowen et al, 2006
pineapple
245
(2)
[esti I or Cut, unwashed, served
1989 U.S.A. cantaloupe S. Chester Caterer Reis et al, 1990
m. > D in salad bars
2500
0]
1991 U.S.A. cantaloupe unknown 21 Bowen et al, 2006
Contamination during Indoor picnic,
1991 U.S.A. watermelon S. Javiana 39 transport?; in-school Blostein, 1993
sliced unwashed; party
 27

0
stored AT C; leftovers
held 24h
>400
U.S.A.
US; Multiple
1991 Canada cantaloupe S. Poona D Pre-cut, fruit salad MMWR, 1991
72 locations
Can
cantaloupe +
1993 U.S.A. unknown 140 Bowen et al, 2006
honeydew
melon + Food service CSPI; CDCd 2003
1993 U.S.A. C. jejuni 48 Restaurant
strawberries
Cross-contamination Food service
1993 U.S.A. cantaloupe E. coli O157:H7 27 Restaurant CSPI
raw beef possible
Private
1993 U.S.A. watermelon S. Javiana 27 CSPI; Del Rosario et al, 1995
home; church
cantaloupe +
1995 U.S.A. unknown 24 Bowen et al, 2006
icecream
cantaloupe +
1995 U.S.A. unknown 27 Bowen et al, 2006
watermelon
watermelon +
Singapor Contaminated wash
1996 papaya + S. Weltevreden 27 Food stall Workplace Ooi et al, 1997
e water
pineapple
melon + lemon
1997 U.S.A. E. coli O157:H7 9 Home CDC 2003
bars
Contamination pre-
harvest, most
1997 U.S.A. cantaloupe S. Saphra 5 I unwashed, poor Multiple sites Mohle-Boetani et al, 1999
refrigeration in
distribution
Sliced,
1998 Canada cantaloupe S. Oranienberg 22 I 0 . Deeks et al, 1998
held AT C several hrs
honeydew + Infected food handler, Food service
1998 U.S.A. ? 41 Restaurant CDC
strawberries bare hands
cantaloupe,
infected food handler, Food service
1999 U.S.A. honeydew, Norovirus 61 Restaurant CDC; CDC 2003
served at salad bar
watermelon
watermelon Food service
1999 U.S.A. Norovirus 23 Poor hygiene Restaurant CSPI
+pineapple
1999 U.S.A. cantaloupe Norovirus 5 Infected food handler Food service Restaurant CDC
 28

(sus.) Caterer
honeydew, 0
1999 U.S.A. S. Enteritidis 82 Held at AT C School CDC; CDC, 2003a
watermelon
Multiple
1999 U.S.A. Melon S. Javiana 11 Pre-cut locations or CSPI
unknown
cantaloupe +
2000 U.S.A. turkey Norovirus 33 Infected food handler Caterer CDC
sandwich
cantaloupe +
2000 U.S.A. Norovirus 20 Bowen et al, 2006
turkey
S. aureus
Poor hygiene, Church,
2000 U.S.A. Melon (sus.), 55 Caterer CDC, CSPI
poor cold storage caterer
B. cereus (sus.)
Buffet salad, poor
hygiene, bare hand Food service
2000 U.S.A. Melon S. Heidelberg 4 Restaurant CDC
contact,
cross-contamination
Possible grower Food service Restaurant, ,
2000 U.S.A. cantaloupe S. Poona 46 primary source, Grocery store nursing CDC
I 0
held AT C Caterer home, home
Poor hygiene, Food service
2000 U.S.A. watermelon E. coli O157:H7 23 Restaurant CDC
poor cold storage
2001 U.S.A. melon not reported 33 Poor cold storage Caterer CDC
2001 U.S.A. cantaloupe S. Anatum USDA JEMRA response
cantaloupe,
honeydew Food service Restaurant
2001 U.S.A. Norovirus 36 CSPI; CDC
+pineapple+ buffet
grapes
cantaloupe
2001 U.S.A. Norovirus 100 CDC
+pineapple
melon+ Food service
2001 U.S.A. ? 42 Restaurant CDC
strawberry, fruit
50 Food service Multiple
2001 U.S.A. cantaloupe S. Poona I Poor hygiene CDC; CDC 2002
(2) grocery store locations,
honeydew,
Food service Restaurant, ,
2001 U.S.A. musk melon, S. Poona 23 Salad handling CSPI
grocery store home
watermelon
 29

S. enterica, no Food service Restaurant,

2001 U.S.A. cantaloupe 2 CDC
serotype nursing home
cantaloupe Food service
2002 U.S.A. Norovirus 15 Restaurant CDC
+pineapple
watermelon,
2002 U.S.A. cantaloupe S. Berta 29 Caterer Church/ CSPI
+grapes
Nursing
Food service
2002 U.S.A. cantaloupe S. Poona 26 I Food stored in advance home, CDC
grocery store
private home,
Daycare
cantaloupe, Food service
2003 U.S.A. S. Muenchen 58 centre, CDC
honeydew grocery store
private home
Food service
68 Poor holding Restaurant,
2003 U.S.A. honeydew S. Newport caterer CDC
(2) temperature hospital
grocery store
Restaurant or
2003 U.S.A. honeydew S. sonnei 56 CSPI
deli
cantaloupe +
Food service
2003 U.S.A. banana + Norovirus 16 Infected food handler nursing home CDC
pineapple
cantaloupe
2004 U.S.A. S. Oranienberg USDA JEMRA response
(suspect)
cantaloupe,
Food service Banquet
2004 U.S.A. honeydew, Norovirus 100 Salad CDC
facility
watermelon
watermelon, Church,
2004 U.S.A. Norovirus 34 Infected food handler CDC
honeydew temple
Melon
Food service Nursing
2004 U.S.A. +strawberries+ Norovirus 62 CSPI
home
grapes + salad
watermelon,
honeydew, Norovirus Food service Conference
2004 U.S.A. 30 Gloved hands CDC
cantaloupe + (sus.) facility/
fruit
6
2004 U.S.A. cantaloupe E. coli O157 H7 [HUS ISID, 2004
]
 30

Food service
2005 U.S.A. watermelon Norovirus 18 Gloved hands Camp CDC

cantaloupe +
chicken
2005 U.S.A. S. Enteritidis 126 Home Private home CSPI
(barbeque) +
corned beef
cantaloupe +
2005 U.S.A. S. Newport 24 Unknown CSPI
beef (ground)
melon + fruit
2006 U.S.A. E. coli O157:H7 USFDA JEMRA response
salad
73% cases
US39
U.S.A., honeydew & Processing served at
2006 S. Oranienberg , Can MMWR (2007)
Canada cantaloupe plant health care
2
facility
2006 U.S.A. watermelon C. jejuni 15 Picnic CSPI
watermelon
2006 U.S.A. Norovirus 14 Grocery store Picnic/ CSPI
+fruit
Melon,
Multiple
honeydew,
2006 U.S.A. S. Newport 12 locations or CSPI
plum, pizza,
unknown
cheese
Infected food handler,
watermelon + Food service
2006 U.S.A. S. Newport 20 bare hands Restaurant CDC
fruit
cantaloupe
2006 Australia S. Saintpaul 115 Inadequate washing Munnoch et al (2009)
(rockmelon)
honeydew Food service Banquet
2007 U.S.A. Norovirus 19 CSPI
+caramel rolls facility
Food service Restaurant,
2007 U.S.A. honeydew S. Litchfield 11 MMWR, 2008
home
cantaloupe +
grapes + fruit
Food service
2007 U.S.A. salad, green S. Litchfield 30 Restaurant CSPI
salad
(suspect)
melon + mixed Food service,
2007 U.S.A. Norovirus 44 Infected food handler Restaurant CDC
fruit Distributor
2008 U.S.A. cantaloupe Norovirus 23 Infected food handler Food service Restaurant CDC
 31

Infected food handler,

Central
2008 U.S.A. watermelon S. Javiana 594 poor storage Multiple sites CDC
kitchen
temperature
Multiple
cantaloupe
2008 U.S.A. S. Javiana 10 D locations or CSPI
(suspect)
unknown
US Restaurant;
cantaloupe
U.S.A. 53 Food service Private
2008 +fruit S. Litchfield I CSPI; CDC
Canada Can home;
suspected
9 Hospital
2008 U.S.A. melon + fruit S. Litchfield 5 CDC
Restaurant/
Food service
2008 U.S.A. melon + fruit S. Litchfield 5 grocery store/ CDC
home
cantaloupe,
2008 U.S.A. S. Newport 3 Salad mix Home Private home CSPI
watermelon
cantaloupe,
US32
U.S.A. honeydew,
2009 S. Carrau , Can ? Community PHAC, 2009
Canada watermelon
35
(suspect)
New Watermelon S. Typhimurium ESR JEMRA response;
2009 19 McCullum et al (2009)
Zealand (unwashed) PT1
Hospital
L. (possible),
cantaloupe or Food service OzFoodNet Working Group
2010 Australia monocytogene 9 (2) immuno- (2010)
honeydew
s compromised
patients
cantaloupe,
mint and OzFoodNet Working Group
2010 Australia Cyclospora sp. 314 (2010)
lettuce
(suspect)
2011 U.S.A. cantaloupe S. Panama 13 I CDC, 2011
2007
3 outbreaks
- Brazil watermelon Brazil response to JEMRA
no data
2022
a
D=domestic; I=imported; b Ambient temperature; c “Melon” indicates type unspecified; d CDC data provided in U.S.A. response to JEMRA
 32

References used in Table

ACMSF (Advisory Committee on the Microbiological Safety of Food). 2005. Information Paper. Microbiological status of ready to eat fruit and vegetables.
ACM/745. (Cited 11/05/11at http://www.food.gov.uk/multimedia/pdfs/acm745amended.pdf).
Blotstein, J 1993. An outbreak of Salmonella Javiana associated with consumption of watermelon. J. Envir. Hlth. 56:29-31.
Bowen, A., Fry, A., Richards, G. and Beuchat, L. 2006. Infections associated with cantaloupe consumption: a public health concern. Epid. Infect. 134: 675-
685.
CDC. 2011. Investigation Update: Multistate Outbreak of Salmonella Panama Infections Linked to Cantaloupe. Cited 26/04/11 at
http://www.cdc.gov/salmonella/panama0311/032911/index.html.
CDC. 2008. Investigation of outbreak of infections caused by Salmonella Litchfield. Cited May 2011 at http://www.cdc.gov/salmonella/litchfield/.
Deeks, S, Ellis, a Cieben, B et al. 1998. Salmonella Oranienberg, Ontario. Can. Commun. Dis. Rep. 24:177-179.
Del Rosario, B. A. and Beuchat, L. R. 1995. Survival and growth of enterohaemorrhagic Escherichia coli O157:H7 in cantaloupe and watermelon. J. Fd. Prot.
58:105-107.
Gayler, G. E. MacCready, R.A., Reardon, J.P. and McKernan, B.F. 1955. An outbreak of Salmonellosis traced to watermelon . Pub. Hlth. Rep. 70(3):311
(Cited by Castillo et al, 2009).
McCallum L., Torok M., Dufour, M.T., Hall A, and Cramp G. 2009. An outbreak of Salmonella Typhimurium phage type 1 associated with watermelon in
Gisborne. N. Z. Med. J. 10;123:39-45.
Mohle-Boetani, J.C., Reporter, R., Werner, S.B., et al. An outbreak of Salmonella serogroup Saphra due to cantaloupes from Mexico. J. Infect. Dis.
1999;180:1361-1364.
MMWR, 2008. Salmonella Litchfield outbreaks associated with a hotel restaurant – Atlantic City, New Jersey, 2007. Morbidity and mortality weekly report.
57:775-779.
MMWR. 2007. Salmonella Oranienburg infections associated with fruit salad served in health-care facilities--northeastern United States and Canada, 2006.
Morbidity and mortality weekly report. 56:1025-8.
MMWR. 2003. U.S. Outbreak of Salmonella serotype Javiana infections, Orlando Florida, June 2002. Morbidity and mortality weekly report. 51:683-684.
MMWR. 2002. U.S. Multistate outbreaks of Salmonella Poona infections associated with the eating cantaloupe from Mexico, United States and Canada,
period 2000-2002, November, 2002. Morbidity and mortality weekly report. 51: 1044-1047.
MMWR. 1991. Epidemiologic notes and reports: multistate outbreak of Salmonella Poona infections. United States and Canada. Morbidity and mortality
weekly report. 40:549-552.
MMWR. 1986. Aldicarb food poisoning from contaminated melons – California. Morbidity and mortality weekly report. 35:254-258.
MMWR, 1979. Salmonella Orangeburg gastroenteritis associated with consumption of pre-cut watermelons. Morbidity and mortality weekly report. 28:522-23.
Munnoch, S.A., Ward, K., Sherida, N. S., Fitzsimmons, G.J., Shadbolt, C.T., Pilspanen, J.P., Wang, Q., Worgan, T.L.M., Oxenford, C., Musto, J.A., McAnulity,
Durrheim, D.N. 2009. A multi-state outbreak of Salmonella Saintpaul in Australia associated with cantaloupe consumption. Epid. Inf. 137:367-375.
OzFoodNet Working Group. 2010 OzFoodNet Quarterly Report, 1 April to 30 June 2010. Communicable Diseases Intelligence 34:345-354 (Cited at
http://www.health.gov.au/internet/main/publishing.nsf/Content/cda-cdi3403-pdf-cnt.htm/$FILE/cdi3403o.pdf.
PHAC (Public Health Agency of Canada). 2009. International outbreak of Salmonella Carrau Infections - Final Epidemiological Report, Center for Food-borne,
Environmental and Zoonotic Infectious Diseases, PHAC.
Ries A.A., Zaza S., Langkop C., et al. 1990. A multistate outbreak of Salmonella Chester linked to imported cantaloupe [Abstract]. In: Programs and abstracts
of the 30th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, DC: American Society for Microbiology. (Cited in MMWR,
2002).
Annex 2 FAO/WHO Call for data: Summary
FAO/WHO extended a call for data on the identification and control of microbiological
hazards associated with melons to support the development of an Annex to the Code of
Hygienic Practices for Fresh Fruit and Vegetables. The data received at May 2011 is
summarised in this annex.

Thirteen countries responded to the call including Colombia, Mexico, Brazil, Argentina,
Guatemala, France, United States of America (U.S.A.), Canada, Australia, New Zealand,
China, Lithuania and the Republic of Armenia.

Only a couple of countries provided detailed information and references. There was
insufficient data from enough countries to make comments on generalisations or trends in
most data listed in the call. Key points in managing food safety hazards were listed for all
sectors along the melons supply chains indicating the need for a through chain approach.

Melons and the link to foodborne illness

Five of the responding countries reported evidence of illness associated with melons within
their country based on outbreak investigations. Not all countries had surveillance systems
where data could be found or were confident their surveillance system would detect an
outbreak related to fresh produce. These outbreaks and associated details have been
incorporated in the Annex 1 together with data from the literature research.

Countries where outbreaks occurred or, countries exporting melons that had been linked to
outbreaks in the importing countries, had developed follow-up actions. These included:
• Review of domestic, import and export risk management
• Development of industry guidelines specifically for melon production or amendments
to existing guidelines for fresh fruits and vegetables to strengthen attention to risk
mitigation in melon production
• Attention to safety of foods for vulnerable populations
• Risk communication with education packages for industry and consumers
• Research and surveillance of hazards in melons to support risk managers

Production practices
The responding countries produce the common watermelon, cantaloupe and honeydew and
a myriad varieties of each. They are mostly grown in open fields. China in addition reports
production of hami melons. They also report greenhouse production is some regions and
growing some varieties of melon for longer periods and to a much larger size and maturity.
The supply chain flow varies between countries mainly in the presence or otherwise of field
packing.

Melons are produced in both large scale operations, especially for distant and export
markets, and in small holder operations that could be pooled for distant locations or
marketed locally.
For those countries reporting production guidelines or codes of practice for melon production,
these include GAP, GlobalGAP, national or regional guidelines for fruit production or those
specific for melons or a melon type (e.g. watermelons, cantaloupe), customer driven
requirements (e.g. import country specifications).