PREVALENCE OF HISTAMINE AND PATHOGENIC MICROORGANISMS IN FISH
CONSUMED IN RURAL MALAWIAN HOUSEHOLDS
By
Connel Ching’anda

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
FOOD SCIENCE – MASTER OF SCIENCE
2013

ABSTRACT
PREVALENCE OF HISTAMINE AND PATHOGENIC MICROORGANISMS IN FISH
CONSUMED IN RURAL MALAWIAN HOUSEHOLDS
By
Connel Ching’anda
Fish is a major source of protein in Malawi, contributing about 70% of total animal protein
consumed. However, this food resource is typically subjected to poor handling, storage and
transportation in Malawi as it changes hands from fishermen through sellers and finally to
consumers. Such poor handling exposes the product to contamination by foodborne pathogens
and may predispose the fish to formation of histamine. The aim of this study was to determine
the prevalence of selected foodborne pathogens and concentration of histamine in common fish
species consumed in Malawi and to evaluate handling practices that may be associated with the
incidence of these hazards. Raw samples (n=40) collected from Lakeshore and Blantyre markets
and cooked samples (n=40) from Mbayani Township households were analyzed for presence of
Salmonella spp., Staphylococcus aureus, Listeria spp. and E. coli O157:H7. Fish were frequently
contaminated with Salmonella spp., Staphylococcus aureus, or E. coli O157:H7 in Lakeshore
and Blantyre markets (50% incidence) and Mbayani Township households (63%). Raw/market
samples and cooked samples from homes had similar pathogen incidence (P>0.05). Sixty-six
percent of the samples had histamine concentrations exceeding 50 mg/kg, the guidance level for
histamine identified by the US Food and Drug Administration. Preservation method did not
influence histamine concentration within fish species (P>0.05) No correlations between
incidence of food safety hazards and handling practices in homes or after harvest were identified.

ACKNOWLEDGEMENTS

I would like to express my sincere thanks to my supervisor, Dr. Bourquin, for his continual
guidance, advice and support. Also, I would like to thank my committee members, Dr. Ustunol
and Dr. Linz, for supporting this research. Thanks to Dr. Ng for his continued guidance for the
whole MS program. All staff and fellow graduates in the Department of Food Science and
Human Nutrition for their day to day support. I would like to thank Dr. Karl Seydel for allowing
me to use the Blantyre Malaria Project (BMP) Research Lab and help from Jimmy Valeta and
other colleagues at BMP and the Polytechnic. I also acknowledge support from staff at
Department of Fisheries in Monkey Bay. I received so much support from so many people too
numerous to mention and I appreciate all.
I would like to thank my family for their support and for patiently bearing the times I had to
commit to this work.

iii

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................................................... vi
LIST OF FIGURES ................................................................................................................................... viii
1

INTRODUCTION............................................................................................................................... 1

2

LITERATURE REVIEW................................................................................................................... 3
2.1

Biogenic Amines .......................................................................................................................... 3

2.2

Histamine Effects in Humans..................................................................................................... 5

2.3

Histamine Formation in Fish ..................................................................................................... 6

2.4

Fish handling and risk of contamination ................................................................................ 11

2.5

Contamination in the home ...................................................................................................... 13

2.6

Summary.................................................................................................................................... 14

3

RATIONALE AND SPECIFIC AIMS ............................................................................................ 16

4

EXPERIMENTAL DESIGN AND METHODS............................................................................. 19
4.1

Fish sample collection ............................................................................................................... 19

4.2

Objective 1: Determination of prevalence and levels of exposure to histamine .................. 22

4.2.1

Sample preparation........................................................................................................... 22

4.2.2

Sample extraction and histamine quantification ............................................................ 23

4.3

Objective 2: Determination of prevalence of pathogenic microorganisms in raw and
cooked fish .................................................................................................................................... 23

4.3.1

Neogen Reveal for Salmonella .......................................................................................... 24

4.3.2

Neogen Reveal for Listeria ................................................................................................ 24

4.3.3

Neogen Reveal for E.coli O157:H7 .................................................................................. 25

4.3.4

Test for Staphylococcus aureus ........................................................................................ 25

4.4
4.5
5

Objective 3: Documentation of food handling and storage practices ................................... 25
Statistical Analyses.................................................................................................................... 27

RESULTS .......................................................................................................................................... 28
5.1

Prevalence of histamine in raw fish ......................................................................................... 28

5.2

Prevalence of pathogenic microorganisms in raw and cooked fish ...................................... 30

5.2.1

Prevalence of foodborne pathogens in raw samples ...................................................... 31

5.2.2

Prevalence of foodborne pathogens in cooked samples ................................................. 32
iv

5.3

Fish handling and storage practices in markets and food processors .................................. 36

5.3.1

Storage practices ............................................................................................................... 40

5.3.2

General market conditions ............................................................................................... 41

5.4

Food preparation and handling practice in the home ........................................................... 43

5.4.1

Food storage practice ........................................................................................................ 45

5.4.2

General condition of the home ......................................................................................... 46

5.4.3

Relationship between practice and level of contamination ........................................... 48

6

DISCUSSION .................................................................................................................................... 51

7

SUMMARY AND CONCLUSIONS ............................................................................................... 56

APPENDICES ........................................................................................................................................... 57
APPENDIX A: QUESTIONNAIRE FOR ASSESSING HOME FOOD HANDLING AND
PREPARATION PRACTICES AND STORAGE PRACTICES IN RURAL HOMES OF
MALAWI ............................................................................................................................................... 58
APPENDIX B: QUESTIONNAIRE FOR ASSESSING FISH HANDLING AND STORAGE
PRACTICES IN RURAL FISH MARKETS AND PROCESSORS ................................................ 63

REFERENCES .......................................................................................................................................... 67

v

LIST OF TABLES

Table 1. Numbers of fish samples obtained from various sources for foodborne pathogen surveillance. .. 19

Table 2. Sample codes. ............................................................................................................................... 28

Table 3. Least square mean concentration of histamine in selected fish species in Malawi....................... 29

Table 4. Prevalence of Salmonella spp., Staphylococcus aureus, Listeria spp. and E. coli O157:H7 in all
fish samples................................................................................................................................................. 30

Table 5. Summary of contamination in all samples. ................................................................................... 35

Table 6. Time taken for smoked fish to reach market................................................................................. 38

Table 7. Time taken for sundried fish to reach market. .............................................................................. 38

Table 8. Time taken for fresh fish to reach market. .................................................................................... 38

Table 9. Summary of maximum time taken for each fish form to reach consumer once caught. ............... 40

Table 10. Common fish storage utensils in markets. .................................................................................. 41

Table 11. Sources of water at markets. ....................................................................................................... 41

Table 12. Use of proper bench at markets. ................................................................................................. 42

Table 13. Distribution of food preparation and handling practices in the homes and incidence of all
pathogens. ................................................................................................................................................... 44

Table 14. Distribution of food storage practice and incidence of all pathogens in the homes.................... 46

vi

Table 15. General condition of the home. ................................................................................................... 47

Table 16. Relationship between practice and incidence of pathogens in lakeshore and Blantyre Markets.50

vii

LIST OF FIGURES

Figure 1: Percentage of fish samples obtained from various sites that were contaminated with more than
one bacterial pathogen. ............................................................................................................................... 32

Figure 2: Prevalence of Salmonella spp. Staphylococcus aureus, E. coli 0157:H7 and Listeria spp. in Lake
shore markets, Blantyre markets and Mbayani households ........................................................................ 33

Figure 3: Prevalence of Salmonella spp. in raw and cooked fish samples.................................................. 34

Figure 4: Prevalence of S. aureus in raw and cooked fish samples ............................................................ 34

Figure 5: Contamination within the samples .............................................................................................. 36

viii

1

INTRODUCTION
In Malawi, cases of foodborne illness are numerous and most of them go unreported. Life

expectancy is very low, estimated at 57 years for males and 58 years for females with an under
five mortality rate of 112 for every 1000 live births (WHO, 2011; National Statistics Office,
2010). A number of factors contributing to this low life expectancy have been identified such as
HIV-AIDS, malaria and diarrheal diseases. Despite diarrheal diseases being one of the major
causes of low life expectancy and high infant mortality rate, there is little surveillance to assess
impact of food contamination in Malawi.
A 2005 report by the Food and Agriculture Organization (FAO) of the United Nations
provided a situation analysis of food safety systems in Malawi. This report noted that cholera
and dysentery were the only foodborne diseases that were properly monitored while other
potential outbreaks arising from Salmonella, E. coli and Staphylococcus aureus were left
undocumented due to the absence of a comprehensive food safety and control program
(FAO/WHO, 2005) To the best of my knowledge, to date there have been no major changes to
improve the monitoring of other potential foodborne illness outbreaks in Malawi since the 2005
FAO/WHO report. Lack of a surveillance system for foodborne diseases also has been
documented as a general problem throughout the African region which limits assessment of food
contamination and its impact on public health (De Waal and Robert, 2005).
Personal communication with Mr. Kadziputa and Mr.Katukana, medical officers at
Monkey Bay and Ulongwe health centers, respectively, indicated that they handle on average
about 300 diarrhea-related illnesses per month. These health centers serve a combined population
of over 70,000. These medical officers also mentioned that the causative agents for most of
1

these cases were not clearly identified and in most cases they assume patients had consumed
contaminated water. In addition, other potential causes of foodborne diseases such as toxicity
resulting from contaminated food haven’t been studied extensively in Malawi.
Eighty percent of Malawi’s 14 million people live in rural areas, which are characterized
by high prevalence of poverty and food insecurity. In these rural communities, it is a common
practice for families to store leftover food at ambient temperature in their homes due to lack of
refrigeration equipment (Taulo et al, 2008). Fish cooked for consumption in the home is one of
the foods typically stored at ambient temperature. Given the tropical climate in Malawi, storage
of perishable foods at ambient temperature is conducive to the rapid growth of microorganisms
including bacterial pathogens.
Going by these observations, it appears that most Malawians consuming fish may be at
risk of illnesses associated with consumption of foodborne pathogens as well as histamine and
other biogenic amines. This research will be the first study in Malawi to address the potential
risk of biogenic amines associated with fish consumption, and one of the few studies to assess
the incidence of microbial pathogens in fish as consumed in Malawi. The study seeks to
determine the incidence of foodborne pathogens and concentrations of histamine in fish
consumed by Malawians before cooking and during storage of left-over cooked fish. Incidence
of foodborne pathogens and histamine concentrations will be correlated with food handling and
storage practices assessed by interviewing Malawian consumers.

2

2
2.1

LITERATURE REVIEW
Biogenic Amines
Fish is a major source of protein in Malawi, where it is estimated that 60-70% of the

protein consumed is provided by fish (Chimatiro, 1998). The development of biogenic amines in
fish after capture and growth of microorganisms during post-harvest handling and storage of fish
(both raw and cooked) at ambient temperature are potential food safety risks. Biogenic amines
can develop in certain foods, particularly certain species of fish, due to contamination by bacteria
which produce enzymes that decarboxylate amino acids to form toxic biogenic amines. Nine
forms of biogenic amines are primarily formed by bacterial action and are responsible for
foodborne illness and toxicity. Histamine, putrescine and cadeverine contamination has been
associated with toxicity in human beings. Biogenic amines also have been implicated as
precursors for formation of the carcinogenic compound nitrosamine (Shalaby, 1996).
Most technologies used in Malawi to preserve fish, such as open sun drying, smoking,
pre-boiling and drying before marketing to the consumer, pose risks of contamination by both
pathogenic and biogenic amine forming bacteria. For example, there usually are relatively few
hygiene controls in place to limit exposures associated with fomites such as flies and other
potential sources of bacterial contamination. Also, extended exposure of fish at ambient
temperatures in the market places could facilitate proliferation of pathogenic microorganisms as
well as microorganisms responsible for biogenic amine formation. Collectively, these risk factors
suggest there is a high likelihood that consumers in Malawi purchase fish that may be
contaminated with foodborne pathogens, histamine and other toxic biogenic amines. In addition,
poor hygiene and temperature abuse during preservation, marketing and food preparation and
handling in the homes pose high risks of further microbial contamination of fish. Work by Taulo
3

et al (2008) showed high levels of food contamination with microbial pathogens in rural areas of
Malawi. This suggests that there could be significant contamination along the preservation and
processing and preparation chain before the food is actually consumed. Unlike histamine and
other biogenic amines, most pathogenic microorganisms are destroyed by high temperatures
during cooking thereby reducing the risk of contracting foodborne illness. However, histamine
and other biogenic amines are reportedly stable at high temperatures (Yung-Hsiang et al, 2005).
Refrigeration or freezing effectively controls formation of biogenic amines in food.
Veciana-Nogues et al (1997) showed that storage of fresh fish at freezing temperatures can
inhibit formation of histamine and other biogenic amines. However, few Malawians can afford
refrigeration due to poverty. A study by Taulo et al (2008) revealed that 99% of rural Malawians
do not own a refrigerator. In addition, their study revealed that 79% of rural Malawians will
typically store leftover food at ambient temperature for a period of 1 to 3 days prior to
consumption. This ambient temperature storage of food likely allows proliferation of pathogenic
microorganisms in the foods thereby increasing the risk of foodborne illness. Furthermore, it is
known that high levels of histamine can be formed before the food appears spoiled or before it is
organoleptically unacceptable (Veciana-Nogues et al, 1997). Thus, consumers cannot reliably
determine the potential presence of histamine contamination by smell or other senses.
Biogenic amines are formed in foods by the action of bacteria in a process known as
amino acid decarboxylation. Consumption of high levels of biogenic amines causes toxicity in
both human beings and animals. Histamine, cadaverine, putrescine, tyramine, tryptamine, βphenylethylamine, spermine and spermidine are the most common biogenic amines formed in
food. These biogenic amines may form in a wide range of foods such as fish, meat, eggs,
vegetables (fermented) and soya (Shalaby, 1996).
4

In fish the presence of biogenic amines is known to cause a type of food poisoning
termed scombrotoxin poisoning which has been associated with the formation of histamine in
Scombridae and Scomberesocidae families of fish, which include tuna, bonito and mackerel. In
addition, some fish species outside the families Scombridae and Scomberesocidae are known to
be susceptible to histamine formation. Although scombrotoxin is characterized by high histamine
concentrations, some studies have shown that the presence of cadaverine and putrescine
increases the level of reaction to scombrotoxin poisoning (Bjeldanes et al 1978; Lehane and
Olley, 2000).
2.2

Histamine Effects in Humans
As a defense mechanism, the human body is able to detoxify low levels of histamines.

However, studies indicate that consuming foods containing relatively high levels of histamine
(above 200 mg/kg) is responsible for causing illness (Lokuruka, 2009). Once histamine is
released in the blood stream various symptoms can be observed affecting the heart, the
gastrointestinal system, and the respiratory system. The most common symptoms are tingling,
rash, drop in blood pressure, vomiting, headache, dizziness, nausea, vomiting, diarrhea, heart
palpitation and respiratory distress (Taylor, 1986; Shalaby 1996; Lehane and Olley, 2000).
Documented mortalities associated with scombrotoxin poisoning are rare, with only one death in
the US attributed to scombrotoxin poisoning between 1983 and 2008 (CDC, 1990; CDC, 1996;
CDC, 2000; CDC, 2013).
Due to this toxicity, different countries have set legal limits for histamine concentration
in food for human consumption. Examples of these limits are the 50 mg/kg “guidance level” in
the USA (FDA, 1998), 100 mg/kg in Europe (EC, 2003), 200 mg/kg in Australia (Australian

5

Food Standards Code, 2001) and 100 mg/kg in South Africa (South African Bureau of Standards,
2001).
The mechanism whereby histamine causes illness in humans is not clearly understood.
Some studies have proposed that illness due to histamine poisoning is a result of histamine being
absorbed in small intestines and released in the blood stream where it binds to cardiovascular and
secretory system receptors (Shalaby, 1996; FAO/WHO, 2005). However, studies examining the
effects of oral administration of histamine at levels similar to those observed in foods associated
with scombroid food poisoning did not cause illness in humans (Taylor, 1986; FAO/WHO,
2005). It has been hypothesized that the presence of other biogenic amines, cadaverine and
putrescine, that inhibit histamine-metabolizing enzymes in the small intestine, may be potentiate
the toxic effects of histamine (Bjeldanes et al 1978; Lehane and Olley, 2000).
2.3

Histamine Formation in Fish
For histamine to develop in fish three conditions must be met, 1) temperature abuse of the

fish, 2) presence of free histidine in fish, and 3) presence of bacteria that produce histidine
decarboxylase (Lehane and Olley, 2000). Development of histamine in fish is partly due to
temperature abuse of raw fish. Histamine formation is very rapid when raw fish is stored at
temperatures above 21.1oC compared to when it is held at temperatures below 7.2oC (FDA,
2011). Therefore, refrigeration is an effective method for controlling histamine formation.
Amino acid decarboxylases of bacterial origin are the enzymes responsible for formation
of biogenic amines including histamine. These decarboxylases act on free amino acids present in
the food to produce the biogenic amines. Histidine decarboxylase converts free histidine to

6

histamine. Similarly, cadaverine and putrescine are formed by decarboxylation of lysine and
ornithine, respectively (FAO, 2012).
Several bacterial species have been identified to be responsible for histamine formation. In
general, formation of histamine and other biogenic amines is a result of the action of mesophilic
bacteria rather than psychrotrophic bacteria (Frank et al., 1985). Veciana-Nogues et al. (1997)
showed high correlation between histamine formation and presence of Enterobactericeae and
coliforms. Tsai et al. (2006) isolated Bacillus coagulans and Bacillus megaterium as weak
formers of histamine. Yatsunami and Echigo (1991) found Staphylococcus spp., Vibrio spp. and
Pseudomonas spp. as microorganisms capable of forming histamine. Importantly, bacterial
species capable of producing the amino acid decarboxylases associated with biogenic amine
formation are ubiquitous in the gastrointestinal tract of fish (Dawood,1988; FAO/WHO, 2005).
Once the amino acid decarboxylase enzyme is formed, it continues to produce histamine
even if the microorganisms have been deactivated. Once formed, the decarboxylase enzymes
maintain activity even at refrigeration temperatures.

Freezing is an effective control for

histamine formation as it prevents proliferation of the microorganisms responsible for forming
the decarboxylase enzymes and also inhibits activity of pre-formed decarboxylase enzymes.
Cooking of fish deactivates both the decarboxylase enzyme and the microorganisms responsible
for its formation, but it is important to note that cooking does not destroy histamine (FDA, 2011;
Yung-Hsiang et al, 2005).
The species of fish has a very strong influence on the potential for scombrotoxin
formation. Early studies identified Scombridae and Scomberesocidae as the fish families most
commonly responsible for histamine poisoning. The fish in these families includes many species

7

that are commonly consumed including bonito, mackerel, tuna, saury and wahoo. Subsequent
research has indicated that other species of fish outside the Scombridae and Scomberesocidae
are commonly predisposed to scombrotoxin formation. In most cases the species identified as
being predisposed to scombrotoxin formation are marine fish having a high free histidine
content. A comprehensive list of fish species which are considered high risk for histamine
formation was included in a recent FAO report (FAO, 2012).

Additional susceptible species

include amberjack (yellowtail), anchovy, herring, mahi-mahi, salmon, sardine and other species.
Peculiar to these species is the presence of high concentration of free histidine which serves as a
substrate for histamine formation.
Over many years, the focus of histamine poisoning has been on marine fish, as these have
been predominantly associated with scombrotoxin food poisoning outbreaks. Very little is
known about the potential of fresh water fish species to be associated with histamine poisoning.
Few published studies have assessed the potential for histamine formation in fresh water fish.
Chytiri et al (2004) assessed the formation of biogenic amines in whole and filleted freshwater
catfish stored on ice for up to 18 days. They observed that putrescine and spermidine were the
major biogenic amines formed under these conditions, with relatively little histamine (< 2
mg/kg) formed even after 18 days of storage on ice. Catfish fillets had significantly greater
concentrations of biogenic amines compared to whole, uneviscerated fish held under these
conditions.
Rezaei et al (2007) studied the formation of biogenic amines in whole farmed rainbow
trout stored in ice for 18 days. They observed significant increases in putrescine, cadavarine and
histamine concentration with storage time. However, the maximum histamine concentration

8

observed after 18 days (1.61 mg/kg) was lower than the FDA 50 mg/kg guidance level for
histamine.
Another study by Coban and Patir (2008) observed an average histamine concentration of
12.21 mg/kg in rainbow trout studied between the months of September 2003 to April 2004.
Furthermore, histamine concentrations decreased significantly (P<0.05) with decrease in
temperature during winter such that they observed a maximum histamine concentration of 16.37
mg/kg in September and the lowest concentration of 7.66 mg/kg in February. Histamine
concentration in fresh water fish was lower than that of marine fish (Coban and Patir, 2008).
A study by Chakrabarti (1998) on shelf life and histamine content of ten fish species
stored at ambient temperature revealed the ability of two freshwater catfish species to form
histamine. Chakrabarti (1998) observed that catfish was organoleptically acceptable when help at
tropical ambient temperatures up to 6 hrs. The histamine level in this catfish after 6 h of holding
at ambient temperature was below 100 mg/kg which was deemed to be acceptable based on
regulatory requirements in most countries. These results suggest that postharvest histamine
formation may not be a significant hazard in temperature abused freshwater fish. However, the
observation also should be treated with caution as ambient temperature holding times in Malawi
may exceed six hours, particularly in rural settings where refrigeration equipment is lacking.
A similar study by Dawood et al. (1988) assessed the effect of holding freshly caught
whole rainbow trout (Salmo irideus) at different temperatures (10, 20 and 30o C) for 6 h prior to
chilling (0o C for up to 14 days). Dawood et al. (1988) observed increased levels of histamine,
putrescine and cadavarine with increase in storage time. The observed concentrations of biogenic
amines after 14 days of chilled storage did not exceed the safe limit, with maximum observed
histamine levels being approximately 13 mg/kg and the other biogenic amines typically being
9

detected at concentrations less than 10 mg/kg. Eviscerated fish contained lower concentrations of
amines than fish that were not eviscerated prior to refrigerated storage. It should be noted again
that this study was conducted in controlled refrigeration environments and with minimal
contamination which does not simulate conditions in rural areas in Malawi.
The effect of temperature on histamine formation was determined in a study by VecianaNogues et al (1997). In their study, tuna samples were stored at 0, 8 and 20o C, and high
histamine levels (100 mg/kg) were observed in samples stored at 8 and 20o C. In addition, this
study demonstrated that, given a consistent fish source and microbial flora, histamine formation
escalates dramatically with increase in storage temperature. This suggests that when cold storage
is inadequate for fish susceptible to histamine formation, the potential of a toxicological hazard
will be high. This observation could relate very well to a rural Malawi setting where refrigeration
facilities are lacking and hygiene practices are relatively poor. Therefore, scombrotoxin food
poisoning could be a concern in rural Malawi if the fish typically harvested and consumed are
shown to be susceptible to histamine formation.
Lake Malawi and other fresh water lakes in central Africa contain a rich diversity of fish
species. Lake Malawi has over 1000 different species and only half have been identified (Weyl
and Weyl, 2001). Some of the families identified are Cichlidae, Cyprinidae, and Alestiidae
which are also highly consumed both by urban and rural populations. Research available on
Malawian fish species has focused on nutrition content mainly protein, fat and minerals. The
precursors for biogenic amine formation such as histidine have not been extensively studied in
these species.

10

Cases of histamine poisoning resulting from consumption of fish from Lake Malawi and
other fresh water lakes are not commonly reported. Auerswald et al. (2006) noted this
underreporting occurs across the African continent, and speculated this underreporting was due
to a lack of laboratory facilities to analyze histamine as well as histamine poisoning possibly
being mistaken for seafood allergy. For example, as of 2009 there had been no cases of
scombrotoxin food poisoning reported in Kenya even though consumption of scombroid fish
species is high in that country. This may be due to ignorance by consumers regarding the
symptoms of scombrotoxin food poisoning, as the illness has transient symptoms that typically
appear 2-8 h after consumption (Lokuruka, 2009).

2.4

Fish handling and risk of contamination
In Malawi, fresh fish is mostly captured by gillnetting (Weyl and Weyl, 2001). A large

proportion (90%) of the total catch is preserved. Fifty percent of the fish are preserved by sun
drying while 40% are preserved by smoking (Chimatiro, 1998). A small proportion of the catch
is sold fresh as observed by Mumba and Jose (2005). Sun dried, smoked parboiled fish are ideal
for rural populations in Malawi where refrigeration equipment is often unavailable. The
preservation methods for sun drying and smoking are usually done with homemade equipment
making the process very slow. This slow preservation process may allow proliferation of both
spoilage and pathogenic bacteria. If the fish contain high concentrations of free histidine and
come in contact with histidine decarboxylase forming bacteria, there is a significant likelihood of
histamine being formed.

11

Auerswarld et al., (2006) observed higher histamine levels in sundried fish and smoked
fish compared to fish preserved by canning. Those elevated levels of histamine were a
consequence of continued favorable conditions for bacterial growth during the sun drying or
smoking processes that allowed more histamine to form before a substantial reduction in water
activity was reached and inactivation of both histamine forming enzymes and bacteria occurred.
Similarly, Lehane and Olley (2000) observed that in Africa it is common for air drying or sun
drying to precede smoking and, in the course of sun drying, bacteria can multiply in the moist
interior of the fish (which has higher water activity) causing the inside to spoil while the outside
appears satisfactory. Furthermore, whereas the common fish preservation practices (sun drying
or smoking) may control growth of microorganisms and activity of decarboxylase enzymes,
these practices will not impact the concentrations of histamine already formed prior to
preservation.
Poor transportation and marketing facilities also increase the potential for fish spoilage
and biogenic amine formation. Temperature and moisture control during transportation and
retailing of fish are lacking in Malawi. A number of studies discussed before in this review have
stressed the necessity of continuous temperature control of fish starting right after the catch
through preparation and consumption (Auerswald et al, 2006; Dawood et al, 1998; VecianaNogues et al, 1998). In Malawi, processed and fresh fish are typically transported by public
transport. Preserved fish is carried in cartons and baskets with little temperature control
(Est.ryukoku, 2001). Fresh fish is typically covered with ice, although sometimes this ice is not
of sufficient quantity to maintain a safe temperature for long distances. In addition, there is often
a greater chance of pathogen contamination at the market place where fish is exposed openly for

12

sale at ambient temperature allowing contamination by flies and other potential sources of
pathogen contamination.
2.5

Contamination in the home
The way food is prepared and handled in the home may increase the risk of

contamination with microbial pathogens. As noted by Unusan (2006), the home has no legal
regulations for food preparation and storage. Home food safety is controlled through education
of the consumer. In Malawi there have been a number of consumer education campaigns through
radio and other media by various Non-Governmental Organizations (NGOs) and the government.
However, a study by Taulo et al. (2008) revealed high incidences and concentrations of
bacteria in cooked fish sampled from households near coastal waters of Lake Malawi. Fish
samples obtained from these households (n=37) had incidences of nonspecific E. coli, E. coli
O157:H7, Salmonella spp. and Staphylococcus aureus of 49, 5, 30 and 51%, respectively.
Median concentrations of Staphylococcus aureus, nonspecific E. coli, and Salmonella in these
fish samples were 3.90, 4.01, and 3.36 log10 CFU/g, respectively. The high microbial loads were
thought to be a result of poor handling practices in the homes (Taulo et al., 2008).
Another study by Kapute et al. (2012) also isolated both pathogenic and spoilage
microorganisms in fresh Oreochromis spp. (tilapia, local name Chambo) sold at local markets
and supermarkets in Malawi suggesting that contamination of fish also occurs at the market
place. In their study, they observed occurrences of Corynebacterium (20%) Micrococcus (16%),
Pseudomonas (15%), Bacillus (15%) Flavobacterium (15%) and nonspecific E. coli (10%).
Incidences of Salmonella, Staphylococcus and Shigella were below 5%. However, in some
samples, the concentrations of Salmonella, Vibrio and E.coli were observed to be above
13

8

10 CFU/g. Furthermore, they observed that fish from local markets had significant higher
8

5

bacterial counts (P<0.01), 9.5x10 , and 2.7x10 CFU/g respectively.

2.6

Summary
Fish represents a significant source of protein for people living in Malawi. Much of the

fish consumed in Malawi is sourced from Lake Malawi and poor post-harvest handling of these
fish predisposes them to contamination with spoilage microorganisms and bacterial pathogens.
Improper handling practices and lack of refrigeration in the homes of residents in rural Malawi
may exacerbate contamination arising during post-harvest handling of fish. Relatively few
studies have assessed the incidence of contamination of fish consumed in Malawi households
with bacterial pathogens. Those studies that have been conducted (Taulo et al., 2008; Kapute et
al., 2012) have indicated that contamination with bacterial pathogens is significant in cooked fish
and other foods commonly consumed in Malawi households.
To date, no published studies have assessed the incidence or concentrations of histamine
in fish consumed in Malawi households. Recent studies have indicated the potential of fresh
water fish and non scombroid fish to form histamine when held under improper conditions such
as temperature abuse. It is important to explore the potential of fish commonly consumed in
Malawi to cause histamine poisoning. From the ongoing discussion, it appears that conditions
permissive for formation of histamine and other biogenic amines are present given the normal
fish handling practices in Malawi. The likelihood of histamine poisoning in Malawi could be
high despite a lack of clinical data indicating that histamine toxicity is prevalent. Surveillance for
foodborne illness in Malawi is limited (FAO/WHO, 2005) and cases of histamine poisoning
14

could be overlooked due to a lack of recognition of the symptoms of histamine poisoning by
health workers or consumers (Auerswald et al., 2006; Lokuruka, 2009). My research aims to fill
this data gap by analyzing the incidence and concentrations of histamine in fish commonly
consumed in Malawi.
In addition, the poor post-harvest handling conditions common in Malawi are also
permissive for contamination and proliferation of pathogenic microorganisms on fish.
Pathogenic bacteria and biogenic amines both have a high likelihood to cause foodborne illness
if the fish is subjected to contamination and temperature abuse after harvest. Preservation
methods such as smoking and cooking may deactivate pathogenic bacteria while histamine and
other biogenic amines remain stable under these conditions and may cause foodborne illness. In
addition, poor food handling practices in the distribution chain and in the home may result in
recontamination of the fish by pathogenic bacteria.

15

3

RATIONALE AND SPECIFIC AIMS
Exposure to histamine and other biogenic amines and pathogenic microorganisms is a

health risk. Adverse health effects resulting from histamine exposure range from skin rashes to
cardiovascular effects. Foodborne illness outbreaks associated with histamine poisoning have
been reported globally for several decades. As a consequence of these outbreaks, various
government regulatory bodies have established safety limits for histamine in fish. For instance,
the US FDA has established 50 mg/kg histamine as the guidance level.
While it is generally known that consumption of histamine contaminated fish is a health
concern in developing countries where fish is consumed in large quantities and hygiene practices
are typically poor, clinical data are scanty in Malawi and levels of biogenic amines in fish are not
known. In addition, most studies have concentrated on histamine formation in marine fish while
little is known concerning whether freshwater fish have the capacity to form histamine and other
biogenic amines of medical importance. It is believed that most cases of histamine poisoning are
either misdiagnosed or are unreported, meaning patients suffer in silence or may receive
improper treatments. In addition, studies have indicated that histamine is resistant to heat
treatment and its presence is not readily detectable by sight or smell. Therefore, it is important to
know the prevalence and levels of exposure to biogenic amines in Malawi from fish commonly
consumed in the diet.
Similarly, since studies have shown frequent contamination of fish by bacterial pathogens
it is important to explore the prevalence of these pathogens in the fish supply chain and to
document handling and processing practices that may increase contamination. This knowledge
will help in planning low-cost intervention strategies to reduce the risks both during handling and

16

consumption of fish. As recommended by Auerswald et al. (2006), such knowledge will help
form a basis for setting legal limits for foodborne hazards in fish.
My long term objective is to determine the prevalence and concentrations of food safety
hazards (chemical and microbiological) in foods commonly consumed by persons in rural
Malawi, and to develop practical guidance and interventions to minimize exposure of Malawian
consumers to these food safety hazards. The objectives of this research were to determine the
prevalence and concentrations of histamine and selected bacterial pathogens (Salmonella spp.
Staphylococcus aureus, Listeria spp. and E. coli O157:H7) in raw and cooked fish consumed by
residents in rural Malawi, and to correlate the prevalence and concentrations of these hazards
with food handling, preparation, and storage practices by fish markets, processors, and
consumers in rural Malawi. I pursued these objectives by completing the following specific
aims:
1. Quantify the concentration of histamine in raw and preserved fish sampled in fish
markets. The working hypothesis is that fish commonly consumed in rural Malawi is
subject to temperature abuse and improper handling, and therefore is at high risk of
contamination with histamine.
2. Determine the prevalence of pathogenic bacteria in raw and preserved fish sampled in
fish markets and in the homes of rural Malawians. The working hypothesis is that fish
commonly consumed in rural Malawi is subject to temperature abuse and improper
handling, and therefore is at high risk of contamination with bacterial pathogens.
3. Correlate food handling, processing and storage practices with the prevalence and
concentrations of histamine and bacterial pathogens in raw and preserved fish consumed
in rural Malawi. The working hypothesis is that the prevalence and concentrations food
17

safety hazards in fish is associated with specific food handling practices during postharvest handling, preservation, distribution, or home preparation of fish.
By completing these specific aims, I will have completed an initial survey of the
prevalence and concentrations of key food safety hazards in fish commonly consumed in Malawi
and also identified major factors associated with the presence of these hazards. This information
will be essential for determining the potential public health impact of food safety hazards in fish
on Malawian citizens as well as for guiding the development of appropriate intervention
strategies to minimize risk from these hazards.

18

4
4.1

EXPERIMENTAL DESIGN AND METHODS
Fish sample collection
Samples of raw and cooked fish were obtained from the lakeshore, markets and homes

for this research. The plan used to collect fish samples is presented in Table 1. The sampling
plan was balanced with the exception that Kampango was not sampled from Mbayani homes and
Mlamba was not sampled from the Lakeshore or Blantyre markets.

19

Table 1. Numbers of fish samples obtained from various sources for foodborne pathogen
surveillance.
Fish type and form

Lakeshore (source)

Blantyre
1
Markets

Mbayani homes

Usipa fresh

2

2

4

Usipa sundried

2

2

4

Usipa smoked

2

2

4

Utaka fresh

2

2

4

Utaka smoked

2

2

4

Utaka sundried

2

2

4

Chambo fresh

2

2

4

Chambo smoked

2

2

4

Kampango fresh

2

2

Mlamba fresh

4

Mlamba smoked

2

2

4

Total

20

20

40

1

An additional four fresh fish samples were obtained from Limbe (Blantyre) market and were

analyzed for histamine content.

The sampled fish were representative of the major species consumed from Lake Malawi
fisheries. Raw samples included “fresh” (fish preserved under ice), smoked and sundried species
20

of “Utaka” (Copadichromis spp), “Usipa” (Engraulicypris sardella; alternative common name:
Lake Sardine), “Kampango” (Bagrus meridionalis; a type of catfish), “Chambo” (Oreochromis
lidole; a type of tilapia) and “Mlamba” (Clarias gariepinus; alternative common name: African
Sharptooth Catfish).
Sampling of raw fish was done in two stages. First stage samples were collected from
lakeshore areas and second stage samples were from Blantyre markets. Lakeshore markets
included Chidzale fishing village, Monkey Bay market in Monkey Bay, Maldeco area and
Mangochi Boma market. Blantyre markets included Limbe and Mbayani markets. Fish samples
weighing about 500 g were collected into sterile Whirl-Pak sampling bags (Nasco; Fort
Atkinson, Wisconsin, USA) and were stored in cooler boxes loaded with ice and transferred to
Blantyre for histamine and microbiological analyses. Samples were collected from fishermen,
fish processors (smoking and drying) and vendors.
The sampling sites were identified with recommendations from the Fisheries Department
in Monkey Bay and are active fish markets in the area. Sampling was done randomly depending
on fish species and form that was available on the market. Consumers selected to obtain cooked
fish were also sampled randomly though this was difficult since houses/homes are not properly
planned in Mbayani Township. Therefore, depending on the willingness of participant at least
every fifth house was targeted.
Cooked fish samples for all species studied were collected from 40 homes of Mbayani
Township around Kabula and Msikawanjala areas. The cooked fish included those which were
originally purchased as fresh, smoked or sundried. Collected samples were fish that were “ready
to be consumed” at meal times. Individuals were asked to pick samples from their plates into

21

sterile Whirl-Pak sampling bags using their utensils. Samples were transported in cooler boxes
under ice to the Polytechnic microbiology lab for microbiological analyses. Samples were
analyzed on the same day they were collected.

4.2

Objective 1: Determination of prevalence and levels of exposure to histamine
Of 44 raw fish samples, 40 were subjected to microbial testing (Table 1) and an

additional 4 samples collected from Limbe market were tested for prevalence of histamine using
a Neogen Histamine Veratox ELISA Kit (Neogen Corporation; Lansing, Michigan, USA).
4.2.1 Sample preparation
Smoked and fresh samples of “Kampango” (Bagrus meridionalis), “Chambo”
(Oreochromis lidole) and “Mlamba” (Clarias gariepinus) were deboned and the head section
was removed before being homogenized by a Waring food blender for 5 minutes while samples
of “Utaka” (Copadichromis spp) and “Usipa” (Engraulicypris sardella) were homogenized
without deboning because of their smaller sizes. All fresh samples of Kampango and Chambo
were eviscerated before blending. Other species were not eviscerated prior to homogenization as

22

these small fish species are commonly consumed without evisceration. The blender was washed
and sterilized by flame before another sample was analyzed.
4.2.2 Sample extraction and histamine quantification
Samples were prepared and tested according to the manufacturer’s instructions. Ten
grams of each homogenized fish sample was added to 90 mls of distilled water in a plastic
extraction bottle. The mixture was vigorously shaken for about 15 to 20 seconds and allowed to
stand for 5 minutes. The mixture was shaken again for 15 to 20 seconds and allowed to stand for
another 5 minutes. The mixture was then filtered through glass wool in a syringe and the filtrate
was collected into a plastic collection tube. Samples were diluted up to 20 times following
preliminary results. Histamine was quantified using the Neogen Veratox ELISA assay with
quantification by measuring absorbance at 630 nm using a BioTek ELx800 96 microwell plate
reader (BioTek Instruments, Inc.; Winooski, Vermont, USA). Histamine quantification
calculations were conducted using Neogen Veratox Software where a standard curve was
produced as reference for sample histamine concentrations.

4.3

Objective 2: Determination of prevalence of pathogenic microorganisms in raw and
cooked fish
A total of 40 raw and 40 cooked fish samples were tested for presence of Salmonella

spp., E. coli O157:H7, Listeria spp. and Staphylococcus aureus using Neogen reveal 2.0 for
Salmonella ELISA Kit, Neogen Reveal for E. coli O157:H7 ELISA Kit, Neogen Reveal for
Listeria spp. (Neogen Corporation; Lansing, Michigan, USA) and the StaphTEX Blue Kit

23

(Hardy Diagnostics; Santa Maria, California, USA), respectively. The distribution of samples
analyzed for microbiological contamination is presented in Table 1.
o

Samples were stored in a refrigerator (2 to 6 C) between collection and analysis.
Samples were tested within 48 h of collection. Prior to analysis, fish samples were homogenized
as whole (without deboning nor removal of head section) using a Waring Food Blender for two
minutes. The blender was washed and flame sterilized before processing subsequent samples.
These homogenates were then tested for the various bacterial species using the test kits listed
above.
4.3.1 Neogen Reveal for Salmonella
Enrichment media was prepared following the manufacturer’s instructions. Twenty-five
o

grams of homogenized sample was added to 200 ml media and incubated at 42 C for 24 h. The
enriched samples were tested for Salmonella presence using the Salmonella test devices after
incubation at ambient temperature for 15 minutes. The test is sensitive to 1 CFU/g or 10

6

CFU/ml post enrichment.
4.3.2 Neogen Reveal for Listeria
Twenty-five grams of homogenized sample was added to 225 ml of media prepared using
the manufacturer’s instructions. The mix was tested for presence of Listeria spp. following
o

incubation at 30 C for 30 h using test devices provided by the manufacturer. Results were read
from the test device. A double line in both control and test zone after 20 minutes was considered
positive while a single line in control area only indicated a negative result. The test is sensitive to
6

detect 1 CFU per g or 10 CFU/ml post enrichment.
24

4.3.3 Neogen Reveal for E.coli O157:H7
Powdered enrichment media (10.7 g; Neogen Corporation) was weighed and added to
o

325 ml of sterilized water held at 42 C. Sixty-five grams of homogenized sample was added to
o

the media and the mix was incubated at 42 C for 20 h. The enriched samples were tested for the
presence of E. coli O157:H7 using the test devices provided. The test is designed to detect as
4

low as 1 CFU per 375g of sample or 10 CFU/ml post enrichment.

4.3.4 Test for Staphylococcus aureus
Twenty-five grams of sample was added to 225 ml sterilized water. Dilutions from the
mix were cultured on Plate Count Agar (Biolab; UK) supplied by Merck, South Africa. The
composition of the Plate Count Agar in g/l was 5.0, 2.5, 1.0 and 14.5 for tryptone, yeast extract,
o

dextrose and agar respectively). Plates were incubated at 37 C for 24 h. Each sample was
cultured on a separate plate. Multiple colonies from the culture were tested for agglutination
following the manufacturer’s procedure. Agglutinated colonies were read as positive and nonagglutinating colonies were interpreted as negative. The test has 98.8% sensitivity and 99.5%
specificity.

4.4

Objective 3: Documentation of food handling and storage practices
Two questionnaires (Appendices A and B) with multiple and structured questions were

used to document food handling and storage practices by consumers in their homes and by the
fishermen, fish processors and fish marketers. The instruments were translated to Chichewa, a
25

local language, and were initially pretested before use.

The questionnaire in Appendix A

targeted homes while the questionnaire in Appendix B was used for fish processors, fishermen
and vendors in markets. In general, the questions in both tools were based on similar practice
questionnaires used by Taulo et al. (2008) and Redmond and Griffith (2003) with some
modification.
For homes, three areas of interest were assessed - food preparation and handling practices, food
storage practices, and the general environment of the home. Hand washing practices, equipment
used for preparation of food, and the source of water for both cleaning utensils and cooking were
considered under food preparation and handling. Storage of leftover food and storage length
were considered under storage practices. The condition of the home was evaluated based on
exposure to domesticated animals, the presence of children with diapers in the home, and
assessing the persons responsible for cooking and cleaning of utensils. The questionnaire in
Appendix B assessed fish handling practices at the market place and fish handling during
processing.
Questionnaires were administered by direct one-to-one interviews after obtaining
informed consent. An interview followed after collection of sample from the seller or home
owner. A total of forty interviews were completed for the markets and forty for the homes. To
guarantee anonymity of responses and protection of identity of respondents, randomly selected
numbers were assigned to each questionnaire. Survey procedures were conducted with the
approval of the MSU Human Subjects Institutional Review Board.

26

4.5

Statistical Analyses
Histamine concentrations were calculated using Neogen’s Veratox Software version 3.3.

Questionnaire and foodborne pathogen data were analyzed using SPSS (Version 20.0).
Descriptive statistics were conducted for prevalence of both histamine and pathogenic
microorganisms. Difference in prevalence of foodborne pathogens with location and fish form
were conducted using a Chi-Square test. Analysis of variance was conducted to test differences
in histamine concentration with form of fish species. Significant differences were observed at
(P<0.05).

27

5

RESULTS
For presentation of results the samples were coded as shown in Table 2 below.

Table 2. Sample codes.
Scientific
Name
Engraulicypris
sardella
Engraulicypris
sardella
Engraulicypris
sardella
Copadichromis
spp.
Copadichromis
spp.
Copadichromis
spp.
Oreochromis
spp.
Oreochromis
spp.
Bagrus
meridionalis
Clarias
gariepinus

5.1

Common
Name
Usipa

Form
Fresh

Sample
Code
ESF

Usipa

Sundried

ESD

Usipa

Smoked

ESS

Utaka

Fresh

CSF

Utaka

Sundried

CSD

Utaka

Smoked

CSS

Chambo

Fresh

OSF

Chambo

Smoked

OSS

Kampango Fresh

BMF

Mlamba

CGS

Smoked

Prevalence of histamine in raw fish

Histamine concentrations (mg/kg) observed in raw fish samples for the various species studied
are presented in Table 3

28

Table 3. Least square mean concentration of histamine in selected fish species in Malawi.
Specie Name

Code

LS Mean
A

Usipa - fresh

ESF

344

Usipa - sundried

ESD

382

Usipa - smoked

ESS

300

Utaka - fresh

CSF

95

Utaka - sundried

CSD

281

Utaka - smoked

CSS

126

Chambo - fresh

OSF

80

Chambo - smoked

OSS

71

Kampango - fresh

BMF

74

Mlamba - smoked

CGS

A

AB

CD

ABC
BCD
D
D
D
D

17

SEM
67
60
60
77
67
67
60
60
60
67

Average histamine concentrations for samples (n=44) ranged from 17 to 382 mg/kg. The
lowest average concentration was observed in smoked Mlamba while highest average
concentration was observed in sundried Usipa. Sixty-six percent of the fish samples obtained
from fishermen or markets in this study had histamine concentrations that exceeded the FDA
guidance level of 50 mg/kg. It is of interest to note that all smoked Mlamba samples contained
histamine concentrations below the FDA guidance level while all fresh Usipa and sundried Usipa
samples contained histamine concentrations exceeding the FDA guidance level. Three fish
samples (7%), all of which were Usipa, had histamine concentrations exceeding 500 mg/kg.

29

A one-way ANOVA revealed that there were no significant differences in histamine
concentrations for fish preserved by different preservation methods, (ice, sundrying or smoking)
within the samples (P>0.05).

5.2

Prevalence of pathogenic microorganisms in raw and cooked fish
The prevalence of Salmonella spp., Staphylococcus spp., E. coli O157:H7 and Listeria

spp. was assessed in both raw and cooked fish samples (n=80). Data are summarized in Table 4.
Salmonella spp. was detected in 28% of all samples, Staphylococcus aureus was observed in
36% of all samples, and E. coli O157:H7 was present in one sample (1%). Listeria spp. was not
detected in any samples.

Table 4. Prevalence of Salmonella spp., Staphylococcus aureus, Listeria spp. and E. coli
O157:H7 in all fish samples.
Pathogen

Numbers Present/Absent

Percentage

aureus

Listeria spp.

E. coli O157:H7

58

72.5%

22

27.5%

Absent

51

63.7%

Present

29

36.2%

Absent

80

100.0%

Present

Staphylococcus

Absent
Present

Salmonella spp.

0

0.0%

Absent

79

98.8%

Present

1

1.2%

30

5.2.1 Prevalence of foodborne pathogens in raw samples
Fifty percent (50%) of raw samples (lakeshore samples and Blantyre markets samples)
(n=40) showed the presence of Salmonella spp., Staphylococcus aureus, or both pathogens.
Salmonella spp. was detected in 23% of the samples while Staphylococcus aureus was detected
in 35% of the samples. Listeria spp. and E. coli O157:H7 were not detected in any of the raw
samples.
Samples which were obtained at the Lakeshore (n=20) had lower detection rates for
microbial pathogens compared to the stage two samples obtained from Blantyre markets. Thirtyfive percent of Lakeshore samples were contaminated with either Salmonella spp. (5%) or
Staphylococcus aureus (30%). Alternatively, in samples obtained from Blantyre markets (Limbe
market and Mbayani Market; n=20) contamination was detected in 65% of all samples with 40%
being contaminated with Salmonella spp. and 40% being contaminated with Staphylococcus
aureus. As stated previously, no contamination by Listeria or E. coli O157:H7 was detected in
any samples procured from the Lakeshore or Blantyre markets.
Contamination with more than one pathogen (Figure 1) was observed in 9% of all
samples (7/80). One sample was contaminated by both Salmonella and E. coli O157:H7 while
six samples (8% of the total) involved contamination by both Salmonella and Staphylococcus
aureus. In this research, no sample obtained from Lakeshore areas was contaminated with more
than one pathogen. However, double contamination was observed in 15% of Blantyre market
samples (3/20) and in 10% of cooked samples obtained from Mbayani households (4/40).

31

Figure 1. Percentage of fish samples obtained from various sites that were contaminated
with more than one bacterial pathogen.

16
14

% contamination

12
10
8
6
4
2
0
Lakeshore Mkts

Blantyre Mkts

Mbayani Homes

5.2.2 Prevalence of foodborne pathogens in cooked samples
For cooked samples collected from Mbayani Township in Blantyre, pathogen
contamination was in 63% of samples (25/40).

Thirty-three percent of samples were

contaminated with Salmonella, 38% with Staphylococcus and 3% with E. coli O157:H7 (Figure
2). Furthermore, while E. coli O157:H7 was detected in this sample group, no Listeria was
detected.
Pathogen prevalence in samples obtained from the lakeshore markets, Blantyre markets
and Mbayani homes was compared, as there appeared to be a general increase in prevalence of
pathogen contamination as the fish traveled from the Lake shore to the markets and homes
(Figure 2). A Chi-square test for prevalence of Salmonella indicated a significant increase in
prevalence of Salmonella with location (P<0.05), with samples obtained from the lakeshore
32

being less likely to be contaminated with Salmonella compared to those obtained from Blantyre
or Mbayani. However, no statistically significant differences were noted for prevalence of
Staphylococcus and E. coli O157:H7 in fish samples obtained from the different locations.
Figure 2. Prevalence of Salmonella spp., Staphylococcus aureus, E. coli 0157:H7 in lakeshore
markets, Blantyre markets and Mbayani households.
45

%

35

Sample contamination

40

30
25

Salmonella

20

Staphylococcus
E.coli 0157:H7

15
10
5
0
Lakeshore Mkts

Blantyre Mkts

Mbayani Homes

The prevalence of Salmonella spp. and Staphylococcus aureus in raw/market samples and
cooked fish samples is summarized in Figures 3 and 4, respectively. The prevalence of
Salmonella spp. and Staphylococcus aureus did not differ between raw and cooked samples.

33

Figure 3. Prevalence of Salmonella spp. in raw and cooked fish samples.
35

% contamination

30
25
20
15
10
5
0
Raw/Market samples (n = 40)
Cooked samples from homes (n=40)
Form of sample

Figure 4. Prevalence of S. aureus in raw and cooked fish samples.
38
37.5
% contamination

37
36.5

36
35.5
35
34.5
34
33.5
Raw/Market samples (n = 40)
Cooked samples from homes (n=40)
Form of sample

34

The percentages of fish samples within each species and treatment group that were
contaminated with pathogens are presented in Table 5 and Figure 5. In summary, CGF and CSD
were the most contaminated form with 100% of samples contaminated while OSS and ESF had
the lowest levels of contamination (13%).
Table 5. Summary of contamination in all samples.
no of
E. coli
Double
%
ID
samples Salmonella Staphylococcus 0157:H7 contamination contamination
BMF
4
2
1
0
75
CGF
4
2
3
0
1
100
CGS
8
4
3
0
1
75
CSD
8
5
4
0
1
100
CSF
8
1
2
0
38
CSS
8
1
2
0
38
ESD
8
1
3
0
50
ESF
8
0
1
0
13
ESS
8
1
6
0
1
75
OSF
8
4
3
1
2
75
OSS
8
1
1
0
1
13

35

Figure 5. Percentage of fish samples (by species and preservation method) containing bacterial
pathogens.
120
100

Maximum%

80
60
40
20

0
BMF CGF

5.3

CGS

CSD

CSF

CSS

ESD

ESF

ESS

OSF

OSS

Fish handling and storage practices in markets and food processors
Fish handling and storage practices in markets and by food processors were documented

by using a questionnaire tool (Appendix B) and by observations targeting fish sellers, processors
and fishermen in their respective places of business. The questionnaires were completed by
conducting direct interviews of the subjects and noting practices as described by the subjects or
as observed by the researcher. The respondents also provided samples for microbiological
analysis.
Twenty interviews were completed for the lakeshore markets and twenty interviews for
Blantyre markets. It should also be noted that for the lakeshore, there was interrelation of
activities such as fishermen being involved in fish drying and smoking and furthermore others
would transport their products from Lakeshore to Blantyre markets (especially Limbe market),

36

an activity that is also done by middlemen. When fish arrived at Limbe market it was typically
sold to Blantyre market fish retailers who would finally sell to the consumer.
At the lakeshore, it was observed that fish was generally caught during the night and fish
landing times were mainly in the morning around 5:30 am to 7:00 am. Once the fish was caught,
a portion was often sold to middlemen who would preserve it in ice and transport to inland
markets including Blantyre.

Alternatively the fisherman would carry out the preservation

themselves and sell to inland markets. Some fresh fish would enter the preservation process
(sundrying or smoking) and this was also done by either the fishermen who caught the fish or by
middlemen who could purchase fresh fish from fishermen. Preservation by sundrying would take
about 2 to 3 days while smoking was completed within 1- 2 h. However, the smoking process
would continue for 1-3 days until the processor had adequate volumes for sale. Once enough
volume was achieved for both dried and smoked fish the processor would either transport and
sell the fish to inland markets themselves or he might decide to sell it to a middleman who
would transport it to inland markets including Blantyre.
Fishermen, processors and sellers (n =20) were asked to provide information on the time
required for the fish (fresh, smoked, and dried) to be transported from the lakeshore market to
Blantyre market (Limbe) once the fish was made ready. Sixty-five percent of smoked fish was
transported to Limbe market on the same day (Table 6). Responses for sundried fish indicated
that 65% transported the fish to Limbe market at least two days after drying (Table 7).
For fresh fish, 95% of the respondents indicated that they take their produce to the
lakeshore market on the same day while 5% sold their produce on the second day (Table 8).

37

Table 6. Time taken for smoked fish to reach market.

Location
Respondents
Lakeshore
% of
markets
respondent
Respondents
Blantyre
markets

% of
respondent

Number of days taken for
smoked fish to reach
market
1
2
4
13
6
1
65.00%

20

30.00% 5.00% 100.00%

17
85.00%

Total

3

0

20

15.00% 0.00% 100.00%

Table 7. Time taken for sundried fish to reach market.

Location of respondent
Lakeshore Respondents
markets
%
respondents
Blantyre
markets

Respondents
% of
respondents

Number of days taken for sundried
fish to reach market
1
2
3
7
Total
8
5
4
3
20
40.0% 25.0% 20.0% 15.0% 100.0%
14
70.0%

6
30.0%

0
0.0%

Table 8. Time taken for fresh fish to reach market.
Number of days
taken to get fresh
fish to the market
Location of respondent
1
2
Total
Respondents 19
1
20
Lakeshore % of
markets
respondents 95.0%
Respondents 17

5.0%
3

100.0%
20

Blantyre
markets

15.0%

100.0%

% of
respondents

85.0%

38

0
20
0.0% 100.0%

The responses from survey participants in Blantyre were slightly different from those of
lakeshore sellers. Eighty-five percent of Blantyre sellers (n=20) brought their smoked fish to the
market on the same day while the remaining 15% waited until the following day. For fresh fish,
85% of the respondents indicated that they begin selling their produce on the same day they
purchase it while 15% didn’t begin to sell until the second day. Sale of sundried fish began on
the same day for 70% of the sellers, while 30% waited until the following day.
The time it took for all procured fish products to be sold from the market was also
documented for both lakeshore sellers (n=20) and Blantyre sellers (n=20). Sixty-five percent of
smoked fish for lakeshore sellers was sold on the same day it was procured while interviews
revealed that the remaining 35% would take up to 5 days to be completely sold. Fresh fish sellers
indicated that 95% of fish would be sold within 24 h whereas 5% indicated that sales of fresh
fish could continue for up to 48 h. Sundried fish was sold less quickly, with only 10% sold on
the same day it was procured while the rest would remain on the market for up to 5 days.
Interviews with Blantyre sellers indicated that same day sale occurred for 85% of fresh
and smoked fish sold in Blantyre markets and for 70% of sundried fish. All fresh fish was sold
within 2 days. Smoked fish could stay on the market for up to 5 days. The volumes sold by
Blantyre sellers were observed to be smaller as compared to their lakeshore market counterparts,
which likely contributed to the reduced time required to sell all products.
A summary of the maximum time taken for fish to reach consumers in Blantyre are
presented in Table 9. Fresh fish took a maximum of 4 days while sundried and smoked fish took
a maximum of 13 and 14 days, respectively.

39

Table 9. Summary of maximum time taken for each fish form to reach consumer once caught.
Form

Fishing hours

Fresh

12

Smoked

12

Sundried

12

Processing
(drying /
smoking)

Lakeshore
Markets

Blantyre
Markets

Max time
to reach
Consumer

1day

1 -2 days

4 days

1-2days

1-5days

1- 5 days

13 days

2-3days

1-5days

1-5days

14 days

5.3.1 Storage practices
Common storage utensils for Lakeshore and Blantyre sellers were also analyzed. The
summary is presented in Table 10.Canoes and basins were common for fishermen who would
leave their fish in canoes waiting for buyers. Lakeshore middle men mainly used cartons (45%)
for smoked and dried fish, and wooden or plastic cooler boxes (20%) for fresh fish. It was also
noteworthy that 20% of respondents who stored sundried fish indicated that they store their
products on the floor.
Fresh fish in Blantyre was commonly stored in Styrofoam or plastic cooler boxes (40%).
Respondents for sundried and smoked fish mainly mentioned cartons (30%) and plastic bags
(30%). Moreover, it was discovered that Limbe market fish sellers had special lockable
cupboards within the market where they would store their product after the market day while
their equivalents in Mbayani would store the remaining product at their homes.

40

Table 10. Common fish storage utensils in markets.

Location of respondent
Count
Lakeshore
% of
markets
respondents

Fish storage aftermarket day
Plastic
Cooler
Cartons
bags
boxes
Sale all
9
0
4
3
45%
0%
20%
15%

Floor
4
20%

Total
20
100%

6
30%

6
30%

8
40%

0
0%

0
0%

20
100%

% of Total

Blantyre
markets

Count
% of
respondents

15%

15%

20%

0%

0%

50%

5.3.2 General market conditions
The general market conditions which included source of water at the market and use of proper
bench (e.g. a table or other elevated display for the fish) were also documented and are
summarized in Tables 11 and 12.
Table 11. Sources of water at markets.
Source of water at the
market
Location of respondent
Lakeshore markets
Count

River Potable
Borehole or Lake water
11
9
0

% of
respondents

55%

41

0%

100%

0

20

20

0%

% of
respondents

45%

0

Count
Blantyre markets

Total
20

0%

100%

100%

Table 12. Use of proper bench at markets.
Use of proper
display bench
Location of respondents
Count
Lakeshore
markets
% of

Yes
13

No
7

Total
20

65%

35%

100%

11

9

20

55%

45%

100%

respondents
Count
Blantyre
markets

% of
respondents

Analysis of the source of water (Table 11) for lakeshore market sellers (n=20) revealed
that 55% used borehole water while 45% used water from the lake to clean their utensils and
fishing equipment. On the other hand, all respondents from Blantyre indicated that they used
potable water for cleaning of their utensils.
General assessment of the market conditions indicated that all markets observed in this
study, either at the lakeshore or Blantyre, had a separate fish section where fish was sold
separately from other products. All markets observed were exposed to animals. In addition it
was observed that in all markets the sellers would use their hands to display fish or arrange fish.
Sixty five percent of lakeshore sellers used proper benches to display their products while 35%
used a mat on the floor (Table 12). In Blantyre, 55% of the sellers used proper benches while the
remaining 45% used a mat on the floor.

42

5.4

Food preparation and handling practice in the home
Common food preparation and handling practices in Mbayani homes and their

association with pathogen contamination were documented and are summarized in Table 13.

43

Table 13. Distribution of food preparation and handling practices in the homes and incidence of
all pathogens.
Kitchen utensils

Frequency

Percent

Winnowing basket and knife

12

Incidence of all
pathogens/frequency
7 (58)
30

Plastic basin and knife

14

35

11 (78)

winnowing basket, basin and
knife

14

35

11 (78)

Before preparation of food

40

100

29 (73)

After visiting toilet

40

100

29 (73)

After changing a child’s
diaper

40

100

29 (73)

Washing with soap

40

100

29 (73)

Water kiosk

29

72

22 (76)

Taps within reach

11

28

7 (64)

Cold water and soap

35

88

24 (68)

Hot water and soap

3

8

3 (100)

Cold water only

2

5

2 (100)

Washing of hands

Sources of water

Method of cleaning utensils

44

The common kitchen utensils used for food preparation by survey respondents in
Mbayani Township homes (n=40) were distributed in this order (Table 13); 35% used plastic
basins and a knife, 35% used plastic basins, winnowing baskets and knives, and the remaining
30% used winnowing baskets and a knife. All respondents (100%) indicated that they wash their
hands with soap before food preparation, after visiting a toilet, and after changing a child’s
diapers.
When questioned about sources of water for cleaning utensils and cooking, all
respondents (n=40) used portable water supplied by the Blantyre Water Board. However, it is
revealing to note that 73% used strategically positioned water kiosks while 27% had water taps
within their reach.

For cleaning utensils, 87% indicated that they used cold water and soap

while 5% used only cold water with no soap.
5.4.1 Food storage practice
Interviews with consumers about their food storage practices (Table 14) revealed that
95% of the homes stored leftover fish for 1 to 2 days with the majority (95%) keeping it for only
one day. It was further revealed that 87% of the respondents stored their leftover food at ambient
temperature while the remaining 13% had refrigerators.

45

Table 14. Distribution of food storage practice and incidence of all pathogens in the homes.
Storage of left over fish

Frequency

Percent

Yes
No

38
2

Incidence of all
pathogens/frequency
95
27 (71)
5
2 (100)

Time food storage (days)
One day
Two days

36
2

95
5

26 (72)
1 (50)

Place of food storage
Ambient
Refrigerator

33
5

87
13

24 (73)
3 (60)

5.4.2 General condition of the home
Assessment of the place of food preparation (Table 15) showed that only 22% (9/40) of
the homes had a designated kitchen while 78% (31/40) did not. When those without a designated
kitchen were questioned further, consumers in 60% of homes mentioned using either the
verandah or preparing food in the open and 18% mentioned using the living room. Assessment
of animal exposures found that 35% of the homes kept either chickens, dogs or cats. Some
respondents mentioned keeping chickens in their living rooms or store rooms.

46

Table 15. General condition of the home.
Frequency

Percent

Incidence of
pathogens /frequency

Place of food preparation
Designated kitchen
Verandah
In the open
In the living room

9
21
3
7

22
52
8
18

6 (67)
15 (71)
3 (100)
5 (71)

Exposure to animals
Yes
No

14
26

35
65

10 (71)
19 (73)

Exposure to children with
diapers
Yes
No

22
18

55
45

16 (73)
13 (72)

Person responsible for
cooking
Adult

40

100

29 (73)

Person responsible for
cleaning cooking utensils
Adult
Child

39
1

98
2

28 (72)
1 (100)

Illness associated with fish
for past 3 months
Yes
No

13
27

32
68

9 (69)
20 (74)

In all the homes visited (n=40), food preparation was conducted by female adults, while
only one household involved children in cleaning of kitchen utensils. It was also noted that 55%
of the homes visited had children in diapers. When consumers were questioned about previous

47

foodborne illnesses that they associated with fish consumption, 33% of the interviewees (13/40)
indicated that persons in their homes had been affected by such an illness in the previous 3
months.
5.4.3 Relationship between practice and level of contamination
The relationship between level of pathogen contamination with practices (incidence of
pathogens/frequency of respondents) was calculated for the homes (refer to Tables 13, 14 and
15). For food preparation and handling practices, the common use of plastic basin and knife and
use of winnowing baskets, plastic basin and knife contributed 11 (78%) incidences each. Despite
all respondents mentioning that they wash hands regularly before food preparation, after visiting
the toilet, and after changing a child’s diapers, 29 (73%) incidences were observed in this
category. Use of water kiosks was associated to 22 (76%) incidences compared to water taps
within reach which was associated to 7 (64%) incidences. Mode of cleaning of kitchen utensils
revealed use of hot water and soap and use of cold water only to be associated to 3 (100%) and 2
(100%) incidences respectively compared to use of cold water and soap which was associated to
24 (68%) incidences.
For food storage practice, storage of left over fish was associated to 27 (71%) incidences.
Storage of leftover food at ambient temperature was associated to 24 (73%) incidences compared
to use of refrigeration with 3 (60%) incidences.
A general assessment of the homes revealed that observed practices generally were not
significantly associated with incidences of pathogens. Homes that prepared food in a designated
kitchen had a 67% incidence of pathogen contamination in the fish that was sampled, whereas
homes preparing food on the verandah, in the open, or in the living room had pathogen

48

incidences of 71, 100 and 71%, respectively. For homes which were exposed to animals, 10
(71%) incidences were observed though homes with no animals also revealed high incidences of
pathogens (13 incidences; 73%). Homes with children in diapers were associated to 16 (73%)
incidences which was not different compared to homes which did not have children in diapers
(13 incidences; 72%).

Overall, few observed practices in the homes were correlated with

increased incidence of bacterial pathogens in fish sampled from the homes.
The relationship between practices and incidence of pathogens for both lakeshore and
Blantyre markets is summarized in Table 16. Lack of proper benches/display cases at lakeshore
markets contributed to 4 (57%) incidences of pathogens while in Blantyre markets it contributed
to 6 (57%) pathogen incidences. In Blantyre markets presence of a proper display bench did not
seem to reduce the incidences as this was associated to 10 (91%) incidences.

49

Table 16. Relationship between practice and incidence of pathogens in lakeshore and Blantyre
Markets.
Frequency

Sources of water
Lakeshore
Blantyre
Use of proper bench
Lakeshore
Blantyre

Fish storage utensils
Lakeshore

Blantyre

Percentage

Incidence of
pathogens/frequency
(%)

Borehole
River or lake
Potable water

11
9
20

55
45
100

4 (36)
3 (33)
16 (80)

Yes
No
Yes
No

13
7
11
9

65
35
55
45

3 (23)
4 (57)
10 (91)
6 (67)

Carton (boxes)
Cooler boxes
Sale all
floor
Carton (boxes)
Plastic bags
Cooler boxes

9
4
3
4
6
6
8

45
20
15
20
30
30
40

5 (56)
0 (0)
0 (0)
1 (25)
5 (83)
3 (50)
8 (100)

For fish storage utensils, use of carton boxes at lakeshore markets was associated with 5
incidences. On the other hand at Blantyre markets cooler boxes were associated with 8 (100%)
incidences followed by carton boxes (5 incidences; 83%) and plastic bags (3 incidences; 50%).

50

6

DISCUSSION
It is noteworthy that fish species in this study showed significant concentrations of

histamine despite the species sampled in this study being generally considered as not susceptible
to scombrotoxin formation. A study by Auerswald et al. (2006) found 399 mg/kg histamine in
yellowtail, a non scombroid fish. Most studies have correlated increased storage time and higher
temperatures with increased histamine concentrations (Dawood et al., 1988; FDA, 1998;
Auerswald et al., 2006). Thus, the high histamine values observed in this study are indicative of
temperature abuse of raw fish during the initial stages of storage and preservation. In the area
under study, it was observed that most fresh fish were sold at ambient temperatures. While fresh
fish could be transported on ice, it was exposed at ambient temperature for many hours during
sale. This observation may explain the higher histamine values observed in ESF in this study.
Furthermore, it was observed that some fish processors opted to preserve their catch by
either sun drying or smoking as a second option if fresh fish did not seem to fetch good money in
the market or if they had unsold fish by close of the market day. This suggests that high
histamine concentrations observed in processed fish could have developed during initial
exposure during presentation for sale as fresh fish. Secondly, preservation by sun drying is a
slow process, is weather dependent, and can require two to three days even in ideal weather
conditions in Malawi. Thus, high histamine levels observed in ESD, ESS and CSD likely could
have developed during their preservation processes. Long transit times from the catch, through
preservation to the market could also influence changes in histamine concentrations in the fish.
Histamine concentrations greater than 500 mg/kg which were observed in ESD are of concern as
such concentrations are reported to cause sickness (Taylor, 1989; FDA, 1998). Since this study
did not look at the other parameters that influence histamine development in fish, such as free
51

histidine content or presence of microorganisms responsible for histamine formation, there is a
need to carry out these tests for conclusive results in the species studied.
The presence of Salmonella spp., Staphylococcus aureus and E. coli O157:H7 in the
species studied is of concern to human health. These microorganisms enter the food chain by
cross contamination during handling, processing, and sale. Salmonella spp. is commonly
associated with poultry (Adams and Moss, 2008; Parry et al., 2002). The presence of Salmonella
spp. suggests possible cross contamination between poultry and food products. All the markets
visited for this study had a separate section for sale of fish, but the poultry section in these
markets where live chickens were sold were often close enough (about 3 to 5 meters) for cross
contamination to occur (insects, wind, etc). Furthermore, cross contamination could have been
caused by buyers in the market who would visit both the poultry and fish sections. In this study
it was observed that both sellers and buyers frequently handled fish with bare hands, so this
could be a source of cross contamination especially for those buyers who have touched poultry
and fish products. Buyers were frequently seen using hands to select fish while sellers were also
using bare hands to sort their product. In addition, poor sanitary handling of fish could be another
factor of concern. Kapute et al., (2013) observed that unsanitary handling practices and
conditions at the local markets were the reasons for occurrences of both spoilage and disease
causing microorganisms in fresh Chambo sold in Malawian markets.
The significantly lower prevalence of Salmonella spp. observed in lakeshore markets
compared to those in Blantyre suggest that cross contamination between poultry (or other
Salmonella sources) and fish was less common in lakeshore areas. It was observed in this study
that most fish products were sold in bulk in lakeshore markets and at this stage of sampling the
fish were only handled by fishermen and processors. For instance, fresh fish was sold directly
52

from the canoe to the dealers and traders. In comparison, fish sold in Blantyre had been
subjected to handling by middlemen and exposed during transportation from the lakeshore to
Blantyre. The retailing practices used by traders in Blantyre also allowed exposure to
contamination by flies and other fomites that could be a source of bacterial pathogens. Lack of
proper benches in Blantyre markets could also be another factor contributing to higher pathogen
prevalence as it was observed that 45% of fish sellers did not have an elevated bench (i.e. at least
a table) compared to 35% observed in lakeshore markets. Those persons who lacked benches
typically kept their fish on the floor in the market.
Observation of food handling practices in homes found common usage of winnowing
baskets (68%) which are made of wood, keeping of chickens in the homes (35%), keeping of
food at ambient temperature (87%), and use of contaminated raw products from the market as
some of the risk factors potentially leading to the high prevalence of Salmonella in the home.
Use of wooden food contact materials such as winnowing baskets and common usage of plastic
basins are considered potential risk factors as wooden and scratched plastic basins are reported to
absorb and retain viable bacteria (Kohl et al., 2002). Other studies (Parry et al., 2002; Kohl et al,
2002; Mitakatis et al, 2004) identified handling of free range chickens as significant risk factors
associated with Salmonella. Furthermore, keeping contaminated food at ambient temperatures
would likely result in rapid proliferation of the microorganisms resulting in the high prevalence
of Salmonella observed in this study.
High prevalence of Staphylococcus aureus in markets and homes is indicative of poor
personal hygiene and temperature abuse of fish and fish products (Le Loir et al., 2003).
Staphylococcus spp. is known to inhabit the nose and skin of human beings (Adams and Moss,
2008). Common use of hands at the market places and improper washing of hands by individuals
53

involved in preparation of food in the homes are the likely factors for the observed prevalence of
Staphylococcus aureus. Despite all respondents mentioning that they wash hands regularly, it
could be that hands were not sufficiently washed to reduce bacterial populations. In addition,
sneezing and coughing during food preparation and storage of food at ambient temperature
(87%) in homes could also be sources of contamination as observed by Sudershan et al. (2012) in
studying microbiological hazard identification and exposure assessment of poultry products sold
in various localities of Hyderabad, India.
The occurrence of E. coli O157:H7 in one cooked fish sample from a home in Mbayani
suggests further compromise in sanitation practices.
The high prevalence of Salmonella and Staphylococcus observed in CGF (fresh Mlamba),
CSD (dried Utaka), OSF (fresh Chambo), and ESS (smoked Usipa) is striking. In this study, the
higher prevalence observed in the “dried” products (sundried and smoked) is supported by
considering the time taken for those products to be ready for the market. Because of reduced
perishability, processed fish were kept relatively longer in the market (up to 14 days before sale)
while fresh fish were sold within a maximum of 4 days. It is therefore suggested that a long
processing and storage period increased chances for contamination for CSD and ESS.
Since the study found no significant difference in prevalence of Salmonella,
Staphylococcus and E. coli O157:H7 between raw fish (from the markets) and cooked fish (from
homes), this suggests either that the fish was not cooked properly to destroy pathogens and/or
that cooked fish was rapidly re-contaminated after the cooking process in most homes. Further
study is necessary to assess the specific reasons for the very high frequency of pathogen
contamination in cooked fish products in this study.

54

Correlations between handling/processing practices and the incidence of pathogens in
homes showed no clear distinctions between practices and their associated contribution to
incidence of pathogens. However, it is clear that some practices identified in this study such as
keeping of animals, lack of designated kitchens, distant water supplies, and storage of food at
ambient temperature could contribute to cross contamination and proliferation of pathogenic
microorganisms. Similarly, for the lakeshore and Blantyre markets, lack of proper benches, sale
of fish in the open at ambient temperature, and poor storage utensils/facilities could be
contributing factors for the incidences observed despite no distinct association with pathogen
incidence being identified with these practices.

55

7

SUMMARY AND CONCLUSIONS

In this study, it has been demonstrated that there is a possibility of histamine being
present in fresh, smoked and sundried fish species under study and that temperature abuse during
sale, handling, preservation and processing are likely causes of the observed high histamine
concentrations. Histamine levels observed in ESD are alarming as they have a potential to cause
human illness. Future research should assess other parameters associated with histamine
formation such as free histidine concentrations and presence of histamine forming bacteria in the
fresh fish species that were most susceptible to histamine contamination.
This study has demonstrated that contamination of fish sold at lakeshore markets and
Blantyre markets with Staphylococcus and Salmonella is common. The prevalence of Salmonella
was observed to be significantly higher in Blantyre markets compared to lakeshore markets. In
addition, the fish appeared to be more contaminated as it moved from the source (Lake Malawi)
to homes. Based on the high prevalence of pathogens in cooked fish sampled from Mbayani
Township homes, we consider recontamination of cooked fish to be common.
In this study no clear correlations between practices and incidence of pathogens could be
identified. However, the study has revealed some practices which were generally associated
with cross contamination and rapid growth of microorganisms in both the markets and homes.

56

APPENDICES

57

APPENDIX A

QUESTIONNAIRE FOR ASSESSING HOME FOOD HANDLING AND PREPARATION
PRACTICES AND STORAGE PRACTICES IN RURAL HOMES OF MALAWI
Part A: Food handling practice
Q1. What type of utensils do you use for food preparation?
Check here
A

Winnowing baskets and knife

B

Plastic basin and knife

C

Metallic basin and knife

D

Hands and knife

E

Hands only

Q2. Do you wash your hands at the following times? Please check all that apply.
Yes
A

Before food preparation?

B

After using the toilet?

C

After changing child’s
diapers?

D

With soap?

58

No

Q3. How do you clean your kitchen utensils?
Check here
A

Cold water and soap

B

Hot water and soap

C

Cold water only

D

Hot water only

E

Wiping only

E)Other……………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………
Q4. What is the source of water for cleaning of your utensils?
Check here
A

Protected well

B

Unprotected well

C

Borehole

D

River or Lake

59

Q5. What is the source of water for cooking?
Check here
A

Protected well

B

Unprotected well

C

Borehole

D

River or Lake

Part B: Storage Practice
Q6. Do you store left over fish and other foods?

Yes

No

If no skip question 7 and 8

Q7. How long is that food stored?
Check here
A

1 day

B

2 days

C

More than 2 days

60

Q8. How is the food stored?

Answer here

Part C: Condition of the home
Q9. Where do you prepare your food?
Check here
A

Separate kitchen

B

Verandah

C

In the open

Q10. Are you exposed to the following?
Yes
A

Domesticated animals

B

Children with diapers

61

No

Q11. Have you or your family member got sick

Yes

suspected to be due to consumption of fish?

Q12. Who is responsible for cooking of food in
your home?

Answer here

Q13. Who is responsible for cleaning of cooking
utensils in your home?

Answer here

62

No

APPENDIX B

QUESTIONNAIRE FOR ASSESSING FISH HANDLING AND STORAGE PRACTICES
IN RURAL FISH MARKETS AND PROCESSORS
Part A: Fish handling at the market
Q1. How long does it take to get the fish to the market on average?
Hours or days (Fill below)
A. Smoked fish
B. Fresh fish
C. Dried fish

Q2. On average, how long does the fish stay on the market before being sold to a consumer?
Number of storage hours or days (Fill below)
A. Fresh fish
B. Smoked fish
C. Dried fish

Q3. How is the fish stored after market-day?

63

………………………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………

Q4. What is the source of water for cleaning of your utensils at market?
Check here
A

Protected well

B

Unprotected well

C

Borehole

D

River or Lake

E)Other……………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………
Q5. General market conditions: please check appropriately.
Check below
A.

Separate fish section

Yes

No

B.

Exposure to flies/

Yes

No

Yes

No

Yes

No

rodents or other animals
C.

Use of hands to display,
arrange/select fish

D.

Proper display bench

64

E)Other……………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………

65

Part B: Fish handling during processing
Q6. How long does it take to sell the fish after catch or preservation?
Time
A. Fresh
B. Smoked
C. Dried

Q7. How do you store the fish awaiting buyers?
………………………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………
Q8. What is the source of water for cleaning your utensils during processing or preservation?
Check here
A.

Protected well

B.

Unprotected well

C.

Borehole

D.

River or Lake

E. Other………………………………………………………………………………………
………………………………………………………………………………………………………
………………………………………………………………………………………………………

66

REFERENCES

67

REFERENCES

Adams.M.R.and Moss.M.O. 2008. Food Microbiology. Third edition. Cambridge. Royal Society
of Chemistry. Pg 235-257.

Auerswald L., Morren C., and Lopata L.A. 2006. Histamine levels in seventeen species of fresh
and processed South African sea food. Journal of Food Chemistry 98:231-239.

Australian Food Standards Code. 2001. Part D: Fish and fish products. Standards D1 and D2.
Version 18.

Bjeldanes L.F., Schultz M.M., and Morris M.M. 1978. On the aetiology of scombroid food
poisoning: Cadavarine potentiation of histamine toxicity in the guinea pig. Food cosmetic
toxicology.16:157-159.

CDC. 1990. Foodborne disease outbreaks, 5-year summary, 1983-1987. (Accessed 2013
December 8). Available from: www.cdc.gov/mmwr/preview/mmwrhtml/00001597.htm

CDC.1996. Surveillance for Foodborne-Disease Outbreaks -- United States, 1988-1992
(Accessed
2013
December
8).
Available
from:
www.cdc.gov/mmwr/preview/mmwrhtml/00044241.htm

CDC.2000. Surveillance for Foodborne Disease Outbreaks --United States, 1993-1997. Accessed
2013 December 8). Available from: www.cdc.gov/mmwr/preview/mmwrhtml/ss4901a1.htm

CDC.2013. Surveillance for foodborne outbreaks – United States 1998 -2008. (Accessed 2013
December
8).
Available
from:
www.cdc.gov/mmwr/preview/mmwrhtml/ss6202a1.htm?s_cid=ss6202a1_w

Chakrabarti R. 1998. Shelf-life and histamine content of ten species of fish stored at tropical
ambient temperature. J. Food Sci. Technol. 35:62-5.

68

Chimatiro S.K. 1998. Aquaculture production and potential for food safety hazard in sub Saharan
Africa: with a special reference to Malawi. Int. J. Food Sci. Technol. 33:169-176.

Chytiri S., Paleologos E., Savvaidis I., and Kontominas M.G. 2004. Relation of biogenic amines
with microbial and sensory changes of whole and filleted freshwater rainbow trout
(Onchorynchus mykiss) stored on ice. J. Food Protect. 67:960-965.

Çoban O.E., and Patir B. 2008. Investigation of histamine level and some chemical quality
parameters in fish consumed in Elazig City. Int. J. Sci. Technol. 3:59-65.

Dawood A., Karkalas J., Williams S.C., and Roy N.R. 1988. The occurrence of non-volatile
amines in chilled stored Rainbow Trout (Salmo irideus). J. Food Chem. 27:33-35.
De Waal C., and Robert N. 2005. Food safety around the world. Center for science in the public
interest. Washington, DC. Pg 23-29.

Doyle, M. 1991. Escherichia coli O157 H7 and its significance in foods. Int. J. Food Microbiol.
12:289-301.

EC 2003. Commision recommendation of 10 January 2003 concerning a coordinated programme
for the official control of foodstuffs for 2003 (2003/10/EC). Official Journal of the European
Commission 7:76-81.

Est.ryukoku
[internet].ryukoku
2001.
[Accessed
from:www.est.ryukoku.ac.jp/est/yuhma/FLM-pamph

2012

May

31].

Available

FAO. 2012. Meeting Report: Joint FAO/WHO Expert Meeting on the Public Health Risks of
Histamine and Other Biogenic Amines from Fish and Fishery Products. 23-27 July 2012. FAO
Headquarters, Rome, Italy.

FAO/WHO. 2005. Conference Proceedings on Food Safety for Africa. Harare, Zimbabwe, 3-6
October 2005. Situation analysis of food safety systems in Malawi. Agenda 1

69

FDA. 1998. FDA and EPA guidance levels. In fish and fishery products hazards and controls
guide (pp. 245–248, Appendix 5). Department of Human Health and Human Service, Food and
Drug Administration, Center for Food Safety and Applied Nutrition, Office of Seafood,
Washington, DC.

FDA. 2011. Fish and fishery products hazards and controls guidance. (Accessed 2012 May 31).
Available from: www.usfda.gov

Frank H.A., Yoshinaga D.H., and Nip W. 1981. Histamine formation and honeycombing during
decomposition of skip jack tuna, Katsuwonus pelamis at elevated temperatures. Marine Fisheries
Review 43:9-14.

Hwang D.F., Chang S.H., Shiua C.Y., and Chai T. 1997. High-performance liquid
chromatographic determination of biogenic amines in fish implicated in food poisoning. Journal
of Chromatography B. 69:23-30.

Kapute, F., Likongwe, J., Kang’ombe, J., Kiiyukia, C., and Mpeketula, P. 2012. Quality
assessment of fresh Lake Malawi tilapia (Chambo) collected from selected local and super
markets in Malawi. Internet Journal of Food Safety.14: 112-120

Kohl, K.S., Rietberg, K., Wilson, S., and Farley, T.A. 2002. Relationship between home foodhandling practices and sporadic salmonellosis in adults in Louisiana, United States.
Epidemiology and Infection 129:267-276.

Le Loir, Y., Baron, F., and Gautier, M. 2003. Staphylococcus aureus and food poisoning. Genet
Mol. Res. 2:63-76.

Lehane L., and Olley J. 2000. Histamine food poisoning revisited. J. of Food Microbiol. 58:1-37.
Lokuruka M.N. 2009. Scrombotoxicosis in African fisheries-its implications for international
fish trade: An overview. African Journal of Agriculture Nutrition and Development 9:16171634.

Mitakakis T.Z., Wolfe R., Sinclair M. I., Fairley C.K., Leder K., and Hellard M.E. 2004. Dietary
intake and domestic food preparation and handling as risk factors for gastroenteritis: a casecontrol study. Epidemiology and Infection, 132:601-606.
70

Mumba P.P., and Jose M. 2005. Nutrient composition of selected fresh and processed fish
species from Lake Malawi: A nutritional possibility for people living with HIV/Aids. Journal of
Consumer Studies 29:72-77.

National Statistics Office, 2011. Malawi Health and Demographic Survey 2010 Preliminary
Report . Pg 15-24.

Parry S.M., Palmer S.R., Slader J., and Humphrey T. 2002. Risk factors for Salmonella food
poisoning in the domestic kitchen-a case control study. Epidemiology and Infection, 129:277285.

Redmond C.E. and Griffith C.J. 2003. Consumer food handling in the home: A review of food
safety studies. J. Food Protect. 66:130-161.

Rezaei M., Montazeri N., Langrudi H.E., Mokhayer B., Parviz M., and Nazarinia A. 2007. The
biogenic amines and bacterial changes of farmed rainbow trout (Oncorhynchus mykiss) stored in
ice. Food Chemistry 103:150-154.

Shalaby A. 1996. Significance of biogenic amines to food and human health. Food Research
International 29:675-691.

South African Bureau of Standards. 2001. Regulations governing microbiological standards for
foodstuffs and related matters. Government Notice No. R 490.

Tsai, Y. H., Lin, C. Y., Chien, L. T., Lee, T. M., Wei, C. I., and Hwang, D. F. (2006). Histamine
contents of fermented fish products in Taiwan and isolation of histamine-forming bacteria. Food
Chemistry 98(1): 64-70.

Sudershan R.V., Naveen Kumar R., Kashinath L., Bhaskar V., and Polasa K. 2012.
Microbiological hazard identification and exposure assessment of poultry products sold in
various localities of Hyderabad, India. The Scientific World Journal.2012:1-7

71

Taulo S., Wetslen A., Abrahamsen R., Kululanga G., Mkakosya R., and Grimason A. 2008.
Microbiological hazard identification and exposure assessment of food prepared and served in
the rural households of Lungwenya, Malawi. Int. J. Food Microbiol. 125:111-116.

Taylor S.L. 1986. Histamine food poisoning: Toxicology and clinical aspects. Critical Reviews
in Toxicology 17:91-128.

Unusan N. 2007. Consumer food safety and practice in the home in Turkey. Journal of Food
Control 18:45-51.
Veciana-Nogues M.T., Marine-Font A., and Vadal-Carou M.C. 1997. Biogenic amines as
hygienic quality indicators of tuna. Relationships with microbial counts, ATP-related
compounds,volatile amines, and organoleptic changes. J. Agric. Food Chem. 45:2036-2041.

Weyl O.L.F., and Weyl M.V. (editors). 2001. Lake Malawi fisheries management symposium
proceedings; 2001 June 4-9; Lilongwe, Malawi. Organizing Committee of the Lake Malawi
Fisheries Management Symposium, Lilongwe, Malawi, pg 95-104.

WHO. 2013.Countries statistics: Malawi. (Accessed 2012 May 31). Available from:
www.who.int/countries/mwi/en/

Yatsunami, K., andEchigo, T. (1991). Isolation of salt tolerant histamine-forming bacteria from
commercial rice-bran pickle sardine. Bulletin of the Japanese Society of Scientific Fisheries. 57,
1723–1728.

Yung-Hsiang T., Chueh-Yueh L., Liang-Tan C., Tsong-Ming L.,Cheng-I W., and Deng-Fwu H.
2006. Histamine contents of fermented fish products in Taiwan and isolation of histamineforming bacteria. Journal of Food Chemistry 98:64-70.

72