MICROBIOLOGICAL STUDIES OF MACARONI AND VEGETABLE SALADS IN WAAKYE , A LOCAL STREET FOOD

The microbiological quality of macaroni and vegetable salads served with waakye , was investigated. Aerobic mesophiles counts (AMC), coliforms counts (CC) and moulds and yeasts counts (MYC) were estimated, and the coliform profiles for different samples of macaroni (raw, local/ imported, laboratory-cooked) served with waakye , and vegetable salads served with waakye were determined. Raw macaroni (local and imported) had AMC of 3.6 and 3.0 log 10 CFU/g, MYC of 1.9 and 1.0 log 10 CFU/g and no CC, respectively. Laboratory- cooked local samples had AMC of 2.4 log 10 CFU/g and 3.3 log 10 CFU/g (after 4 h storage) and no MYC. Macaroni obtained from vendors had AMC mean of 3.1-8.4, CC mean of 2.5-7.3 and MYC mean of 0- 4.1 log 10 CFU/g depending on time of sampling. Vegetable salads sampled at early and late morning had AMC of 6.9 and 7.6, CC of 5.7 and 6.4, MYC of 4.9 and 5.4 log 10 CFU/g, respectively. Six coliforms were detected on macaroni and three were detected in addition to Salmonella spp. on vegetable salads. No significant difference was recorded in the microbial load of raw local and imported macaroni. Cooking improved the microbial quality of raw macaroni (AMC of 2.4 log 10 CFU/g). Generally, there were increases of 3-5 log cycles in the AMC, CC and MYC in macaroni sampled from waakye vendors in the morning (early and late) compared to those at dawn. Although the nature of raw macaroni and its cooking are adequate,


Introduction
Street foods (ready-to-eat foods sold in the informal sector) form an important and wellestablished sector of the food industry in Ghana. They are readily available, inexpensive, nutritionally-balanced and also provide a source of income for the vendors (Ekanem, 1998;Swanepoel et al., 1995Dawson & Canet, 1991. Despite these benefits, concerns have been raised about their safety and quality because most of the vendors lack training in basic food safety practices concerning raw material acquisition, food preparation, storage, handling, and final delivery to the consumer (Moy, Hazzard & Kaferstein, 1997;Bryan et al., 1988).
Information on street foods in developing countries is not readily available. However, studies on street-vended foods in USA, Asia, and a few African countries have revealed high bacterial counts and presence of foodborne bacterial pathogens (Mosupuye & von Holy, 1999;Bryan et al., 1997;FAO/Ghana, 1997). Aerobic mesophilic count (AMC) exceeding 4×10 5 CFU/g, Staphylococcus aureus count exceeding 3×10 4 CFU/g and Bacillus cereus count exceeding 2×10 5 CFU/g have been reported for vegetable salads and pepper sauce served with street foods in Ghana (FAO/GHANA, 1997). According to the World Health Organisation, effects of microbiological hazards such as Salmonella, Campylobacter jejuni and enterohaemorrhagic Escherichia coli on food safety is now a major public health concern worldwide (WHO, 2002).
Despite the poor hygienic quality and potential food safety problems, street foods such as waakye (co-boiled rice and beans), jollof and fried rice containing macaroni continue to enjoy tremendous patronage owing to their availability and affordability. Reports indicate that there is persistent contamination of macaroni and other pasta products with Staphylococcus spp., Escherichia coli, and Bacillus spp. during manufacture (Lopez et al., 1998;Castelvetri, Abroggi & Gola, 1986;Pasolini, Aloi & Ceralli, 1981;Lee, Staples & Olson., 1975). However, few studies have been conducted on the impact of macaroni and other minimally-processed accompaniments such as vegetable salads on the overall hygienic quality of street foods in the country. A public health risk may be associated with the consumption of such foods. There is, therefore, the need to investigate the potential sources of microbiological contamination of such accompaniments that are served with street foods. This could provide baseline information for effective microbiological risk assessment of street foods. Development of a national strategy to reduce food-related risks requires knowledge about levels of potential pathogens and factors influencing them. The aim of this study is to investigate the factors influencing microbiological quality of macaroni and vegetable salads served with waakye, a local street food.

Experimental design
Samples of cooked macaroni (locally produced) and corresponding vegetable salads (lettuce, tomato, onion, cabbage) used for the study were bought at different times in the morning from waakye vendors in three suburbs of Accra (Dome, Haatso and Madina). Samples of raw local and imported macaroni were also obtained from the markets (Madina and Dome). A portion of the raw local sample was cooked in the laboratory by boiling 150 g in 500 ml of distilled water for 5 min. The cooked macaroni was held under ambient laboratory conditions (26-28 0 C) up to 4 h. In all instances, samples were procured in sterile polyethylene sachets, kept at 3-5 0 C before analysis within 2 h. Dawn samples were obtained just after cooking without bulk display. Samples procured from vendors at dawn, early morning and late morning had average holding periods of ½, 2½ and 4½ h, respectively.
The seven samples of macaroni and two samples of vegetable salads used were:

Microbiological analysis
Five grammes of each sample were added to 45 ml peptone water (1% peptone water + 0.5% NaCl) and homogenised with a blender (Waring Laboratory Blender, Christison, Germany) for 5 min. Microbial load determination was carried out using standard decimal dilution and plate count methods (APHA, 1976). Aerobic mesophilic counts were estimated on Plate Count Agar (Oxoid, UK) at 36 0 C/48 h. Coliforms were estimated on Violet red bile agar (Merck, Germany) at 36 0 C/48 h. Moulds and yeasts were estimated on Oxytetracycline (0.01%) Glucose Yeast Extract Agar (Merck, Germany) at 28 0 C/4 days.
For the detection of Salmonella and Shigella, 25 g of the sample were pre-enriched in 225 ml of buffered peptone water for 24 h at 37 0 C, followed by selective enrichment in Rappaport Vassiliadis Soy broth for 24 h at 41 0 C. Isolation was done on Xylose Lysine Deoxycholate agar for 24 h at 37 0 C followed by biochemical confirmation. For the detection of other coliforms, representative colonies from the coliform plates were purified by sub-culturing, and identified using morphological characteristics and standard biochemical tests.
The tests used were Gram stain, citrate, catalase, oxidase, motility, nitrate, carborhydrate fermentations, Triple sugar iron and the IMViC tests with reference to Biochemical Tests for Identification of Medical Bacteria. (MacFaddin, 1980).

Microbial load
The results for the microbial count and coliform isolates of macaroni and vegetable salads are shown in Tables 1 and 2. Raw local and imported samples of macaroni had aerobic mesophilic count (AMC) of 3.6 and 3.0 log 10 CFU/g, moulds and yeasts count (MYC) of 1.9 and 1.0 log 10 CFU/g and no coliform count (CC), respectively (Table  1). Laboratory-cooked samples had AMC of 2.4 log 10 CFU/g and 3.3 log 10 CFU/g (after 4 h storage). No MYC and CC were recorded for laboratory samples.
Macaroni obtained from waakye vendors had varying counts depending on the time samples were obtained. Samples of freshly cooked macaroni obtained at dawn had a fairly low AMC of 3.1 log 10 CFU/g, CC of 2.5 log 10 CFU/g and no MYC. Macaroni obtained early or late morning had high AMC mean of 7.3-8.4 log 10 CFU/g, CC mean of 5.6-7.3 log 10 CFU/g and MYC of 2.9-4.1 log 10 CFU/g. Early and late morning samples of vegetable salads also had AMC mean of 6.9-7.6 log 10 CFU/g, CC mean of 5.7-6.4 log 10 CFU/g and MYC mean of 4.9-5.4 log 10 CFU/g.

Coliform isolates
No coliforms were isolated from raw macaroni and laboratory-cooked macaroni (Table 1). A variety of coliforms were, however, isolated from the macaroni obtained from waakye vendors. Two, four and six isolates were, respectively, detected

Discussion
Microbial analysis of a complete meal of waakye as well as macaroni, vegetable salads, stew, gari and fried fish revealed high total viable counts of 8.00 log 10 CFU/g. (Nketsia-Tabiri, Adu-Gyamfi & Owusu-Biney, 2004). In that study, vegetable salads and macaroni were identified as the main accompaniments responsible for the high count in view of their equally high total viable count of 8.18-8.37 log 10 CFU/g. With the exception of macaroni sold at dawn, this study has also revealed a high mean AMC of 6.9-8.4 log 10 CFU/g for macaroni and vegetable salads obtained from vendors. However, a similar study on waakye without vegetable salads and macaroni reported a maximum AMC of 3.00 log 10 CFU/g (FAO/Ghana, 1997).

Microbial load of raw macaroni and effect of cooking
The processes of preparation, display and sale of macaroni sold with waakye involve microbial hazards at various critical points according to the Hazard Analysis and Critical Control Points (HACCP) concept (ICMSF, 1988). In spite of the fact that cereal products usually have a rich microflora of bacteria, moulds and yeasts, the results of this study showed that the microbial load of raw macaroni is low. The study did not establish any significant difference in the microbial load of raw local and imported macaroni (Table 1). Cooking, which has been identified as a critical control point, destroys vegetative cells and most microbial spores. The effect of cooking has been demonstrated by this study, which has indicated that fresh laboratory-cooked samples of macaroni had lower AMC of 2.4 log 10 CFU/g compared to the AMC of 3.6 and 3.0 log 10 CFU/g recorded for raw local and imported macaroni, respectively.

Effect of holding and bulk display
Bulk display, holding and serving of waakye are critical points during which germination of spores, growth of vegetative cells and cross contaminations occur in macaroni. While holding a relatively smaller quantity of freshly-cooked macaroni in the laboratory under ambient conditions for 4 h resulted in only one log-cycle increase in AMC, macaroni from waakye examined 2½ and 4½ h later had significantly higher AMC of 8.4 and 7.1 log 10 CFU/g, respectively. Coliforms were not detected in the uncooked and laboratorycooked macaroni. However, all vendors sampled had coliforms with a steady increase from 2.5 log 10 CFU/g at dawn to 7.3 log 10 CFU/g after only 2½ h followed by a decrease to 5.6 log 10 CFU/g after 4½ h. Moulds and yeasts also increased steadily to 2.9 log 10 CFU/g after 2½ h and 4.1 log 10 CFU/g after 4½ h. These results generally indicated increases of 2-5 log cycles in the AMC, CC and MYC in macaroni sampled from waakye vendors in the morning (early and late) compared to those at dawn. In addition, six coliform isolates (four of which were not present in the dawn samples) were detected in the morning (early and late) macaroni. These findings emphasised the roles of holding and bulk display as important critical points in determining the microbial quality of macaroni as has also been reported in other studies on street foods. (Bryan et al., 1988;Bryan et al., 1992;Mosupuye & von Holy, 1999).

Cross-contamination by vegetable salads
As can be seen from the high microbial counts in the vegetable salads, it is evident that they contributed immensely to the poor quality of VOL. 47 macaroni, possibly through cross contamination during serving by vendors. The high microbial counts in vegetable salads shown in Table 2 support the study by Nketsia-Tabiri, Adu-Gyamfi & Owusu-Biney (2004), which also suggested cross-contamination from vegetable salads to macaroni. The poor microbial quality of vegetables such as lettuce and cabbage has been reported in other studies (Farkas et al., 1997;Beuchat, 1995Beuchat, , 1996WHO, 2002;Tano-Debrah, 2004), and attributed to their origin, cultivation practices and minimal processing. Careful consideration of the results revealed the similarity of the coliform profiles of macaroni and vegetable salads. The presence of Escherichia coli, Klebsiella spp. and Citrobacter spp. on both macaroni and vegetable salads sold by vendors in the morning further confirms the role of cross contamination by the vegetable salads.

Presence of pathogens
The isolation of Salmonella, Escherichia coli, Klebsiella spp. and Serratia spp. from macaroni and the vegetable salads poses food safety problems since they are all enterotoxigenic and cause gastroenteritis (Stewart & Beswick, 1977;Klipstein, Engert & Short, 1977;Anon., 1996). Other studies have also identified pathogens including Salmonella spp. on other street foods and their accompaniments in South Africa (Mosupuye & von Holy, 1999) and Zambia (Bryan et al., 1997). While Salmonella sp. causes salmonellosis and typhoid fever, E. coli O157:H7 causes severe illness and deaths, especially among children in several countries (WHO, 2002).

Conclusion
The nature and cooking of macaroni are adequate to ensure acceptable hygienic quality in waakye. Cross-contamination from vegetable salads and poor handling during the holding and bulk display periods cause deterioration in microbial quality of macaroni. The presence of potential foodborne pathogens in macaroni and vegetable salads raises serious food safety concerns and calls for urgent action. In addition to the provision of infrastructure and enforcement of bye-laws and codes of practice on street foods, emphasis should be placed on educating vendors on simple preventative steps of keeping food hygienically safe.

Recommendations
Aside the major recommendations of providing basic infrastructure and enforcement of bye-laws and codes of practice on street foods, the education of vendors should emphasise the importance and observation of the following simple steps in keeping food safe from harmful microorganisms: i. Using properly washed good quality raw materials. ii. Washing hands before meals are prepared and frequently during the sale of food. iii. Cooking food adequately and not re-using leftover food. iv. Keeping foods separate to prevent cross contamination between raw (or minimallyprocessed) and cooked foods.