Microbiological Profile and Storage Temperatures of Egyptian Rice Dishes

Journal of Food Protection, Vol. 8, No., Pages - (January 8) Copyright lnternational Association of Milk, Food, and Environmental Sanitarians Microbiological Profile and Storage Temperatures of Egyptian Rice Dishes M. R. EL-SHERBEENY', M. FAHMI SADDIK, HEKMAT EL-SAID ALY and FRANK L. BRYAN * Nutrition Institute, Ministry of Health, Cairo, Egypt and Department of Health and Human Services, Public Health Service, Centers for Disease Control, Atlanta, Georgia (Received for publication May, 8) ABSTRACT Cooked rice and other dishes containing rice ( ) were tested for aerobic colony count ( C), Salmonella, Shigella, Staphylococcus aureus, and presumptive Bacillus cereus. Salmonella was isolated from a sample of Oriental rice that was prepared in a five-star hotel kitchen. Shigella was isolated from a sample of boiled rice from a four-star hotel kitchen. Nineteen percent of the were contaminated by S. aureus, and % of these contained more than /g. Forty percent of the were contaminated by B. cereus, % of these contained more than /g. Rice was more frequently stored at safe temperatures (> C) in four- and five-star hotels than in any other type of establishment or that which was sold by street venders. Aerobic colony counts ( C) per g were usually quite low when rice was held at temperatures of C or higher. These counts generally became progressively higher as the temperature decreased, often reaching quantities exceeding when temperatures were C or below. This was particularly so when the temperature range was - C. These counts were lower for fried and Oriental rice than for boiled rice, rice and vegetables, kushari (a mixture of rice, macaroni and lentils), and rice and shirea (thin, wheat macaroni). Rice and dishes prepared from rice are frequently eaten by Egyptians at the mid-afternoon (main) meal. In certain regions of the Nile delta, residents eat rice most everyday and sometimes at more than one meal during the day. Rice and dishes prepared from rice are frequently served in hotels, restaurants, and small foodservice shops. Large tubs of rice are prepared in small shops or homes and held at ambient temperature on carts for several hours until sold by street vendors. Rice dishes that are prepared in Egypt include boiled rice, fried rice, kushari, rice and shirea, rice and vegetables, and certain leaves (e.g. grape) stuffed with rice. Rice is fried in butter and only salt is added. A mixture of rice, macaroni, and lentils is known as kushari; tomato sauce and fried onions are added just before serving. Rice 'Ministry of Health. Centers for Disease Control. is also mixed with shirea (which is prepared from wheat flour and is similar to spaghetti but thinner and darker in color). Cabbage leaves, grape leaves, or eggplants are often stuffed with a mixture of rice, tomatoes, onions, and greens. Oriental or Italian rice (nuts, onions, spices, and sometimes coloring added to rice) are often served in luxury-class hotels. Cooked-rice dishes in Egypt are sometimes subjected to contamination from equipment and utensil surfaces, hands of workers, flies and other insects, dust and airborne contaminants in the places they are prepared and stored. Furthermore, they are often held at temperatures that permit growth of microorganisms. Although surveillance data in Egypt have not shown an association of outbreaks of foodborne disease and rice, such associations have been made in Canada (), Finland (), Netherlands (), United Kingdom (,,), and United States (,,,). Most of these reports were about outbreaks of Bacillus cereus gastroenteritis; boiled or fried rice prepared in Chinese-style restaurants were the usual vehicles. Large numbers ( to /g) of B. cereus have been isolated from reminants of rice implicated in the outbreaks. The objective of this study was to develop a microbiological profile (with particular reference to common foodborne pathogens) of rice dishes prepared and served in different types of establishments in different locations throughout Egypt. MATERIALS AND METHODS Samples of rice and dishes containing rice were collected from hotels of all classes (five star to one star), restaurants, small cook shops, and street venders located in Alexandria, Aswan, Cairo, El-Fayoum, Giza, Hargada, Ismalia, Luxor, Port Said, Suez and many small cities and villages throughout Egypt. These were aseptically collected with sterilized spoons and put into sterile plastic (Whirl-Pak) bags. Temperature was then measured by folding the filled bag around the sensing-bulb of a thermometer and pressing the bag against it until the temperature stabilized. Sample units in which the temperature was C or higher were immediately cooled under tap water. Sample units were cooled and maintained cold by ice in an insulated box in which they were transported to the laboratory. At the laboratory, sample units were usually stored overnight in a refrigerator before analysis began. JOURNAL OF FOOD PROTECTION, VOL. 8, JANUARY 8

EL-SHERBEENY ET AL. Material from the sample units was weighed and analytical units of g were added to ml of lactose broth (to test for Salmonella), g added to ml of GN broth (to test for Shigella), g added to 8 ml of.% peptone broth (to make appropriate dilutions and for making aerobic colony counts and testing for B. cereus and Staphylococcus aureus). Plate count agar was used for the aerobic colony counts; Baird-Parker agar was used to test for S. aureus, and phenol red egg yolk agar was used to test for B. cereus. Incubation for aerobic colony counts and B. cereus was done at C and for. aureus at C. Procedures for diluting and for obtaining aerobic colony counts followed those stated by the International Commission on Microbiological Specifications for Foods (ICMSF) (). Five colonies suspected as being S. aureus were picked, inoculated into brain heart infusions broth, which was incubated and tested for coagulase according to the ICMSF (). Suspected B. cereus colonies were not confirmed. Results, therefore, refer to presumptive B. cereus. Lactose and GN broths were incubated overnight at C, afterwhich, ml of lactose broth preenrichment was inoculated into separate tubes containing ml of selenite broth. These were incubated overnight in a C waterbath. Likewise, ml from the lactose preenrichment was inoculated into tetrathionate-brilliant green broth. A loopful each of the incubated GN broth and the tetrathionate broth was streaked onto xylose lysine desoxycholate agar and incubated at C for h. A loopful of the incubated tetrathionate-brilliant green broth was streaked onto brilliant green agar and a loopful of incubated selenite broth was streaked onto bismuth sulfite agar. The brilliant green plates were incubated at C for h and the bismuth sulfite agar plates at C for 8 h. Suspect colonies (at least two from each plate containing them) were picked and stabbed into and streaked onto triple sugar iron agar and lysine iron agar. Tubes showing typical reactions for Salmonella or Shigella were tested for agglutination by polyvalent and group antisera (CDC). Procedures followed those stated in ICMSF (). RESULTS AND DISCUSSION Salmonella was isolated from a sample of Oriental rice that was prepared in a five-star hotel kitchen (Tables and ). If salmonellae had entered the kitchen on raw rice, which would be unusual, they would be killed during cooking of the rice. Contamination, therefore, would most likely be attributed to post-cooking contamination. Possible sources were either from (a) cutting onions or nuts (ingredients of Oriental rice) on a cutting board with knives that either of which had previously been used for cutting raw meat or raw poultry or from (b) cleaning these items with cloths that previously had been used in a raw preparation area. The salmonellae could have come also from workers' hands, either if a worker was infected, was shedding the organism, and had poor TABLE. Aerobic organisms and certain pathogens isolated from different kinds of rice dishes in Egypt. S. aureus B. cereus Kinds of rice Aerobic colony count ( C) Median/g Salmonella Shigella (percent) of s > (percent) of s > Rice boiled or unspecified Fried rice Oriental rice Rice & vegetables Kushari Rice and shirea 8. X. X.X.X.X.8 xlo () () K ) () () () 8 8() a () b () b () () ( ) b () () a Two not tested for B. cereus. b One sample not tested for B. cereus. TABLE. of of rice and rice-containing foods contaminated with Salmonella, Shigella, Staphylococcus aureus, and Bacillus cereus; by place prepared or served in Egypt. Place rice prepared or served Salmonella Shigella s. () aureus of s > B. cereus s () > Hotels five-star four-star three-star two-star one-star Subtotal ( ) () 8() () () () 88 8 () a () a 8() b () () () Restaurants Small, cook-shops Street vendors 8 () () () () () () a () () ''One sample not tested for B. cereus. "Two not tested for fi. cereus. JOURNAL OF FOOD PROTECTION. VOL. 8, JANUARY 8

hygienic practices, or if a worker had previously handled contaminated raw meat or raw poultry. Shigella was isolated from a sample of rice from a four-star hotel kitchen (Tables and ). Most likely the rice was handled by a carrier because Shigella is hostadapted to man. It is also possible that a vector, such as a fly, transmitted shigellae from human fecal matter to the rice. The sample contaminated with Shigella had an aerobic colony count ( C) of and yielded x S. aureus and X B. cereus per g. Therefore, the rice most likely was either grossly mishandled or improperly stored or both. Outbreaks of salmonellosis and shigellosis have followed the eating of meals consisting of Oriental, Japanese, Chinese or Polynesian food at which rice dishes were probably served (). The potential for transmission of these organisms by rice in Egypt is shown by the isolation made during this study. This potential exists in five- and four-star restaurants inspite of their better facilities, layout, and equipment. Food source, hygiene of workers, time-temperature control, and procedures compatible with food safety are of utmost importance in control of these illnesses. Nineteen percent of of cooked rice contained S. aureus (Table ). The prevalence of contamination was similar for al kinds of rice dishes, except it was higher in kurshari. Counts were surprisingly low in many of Oriental rice even though ingredients would be chopped and handled, but this dish was seldom served in establishments other than four- and five-star hotels, and these establishments had better equipment for, and practices of, storing cooked rice. Counts were somewhat lower in collected from four-star and five-star hotel kitchens than in collected elsewhere, and it was highest in collected in one-star hotel kitchens (Table ). This difference in levels of contamination was probably related to storage temperature as well as hygienic practices within the establishments. Seventythree percent of the that contained S. aureus had counts that exceeded per g. This suggests the possibility of multiplication during storage. In comparison, Schiemann did not isolate S. aureus from rice in Canada (). Rice dishes have been reported as vehicles of staphylococcal food poisoning (). TABLE. Relationship of storage temperature of 8 a of rice Temperature range ( C) > - - - - - - < () CHARACTERISTICS OF EGYPTIAN RICE DISHES Forty percent of of cooked rice contained presumptive B. cereus (Table ). The prevalence of contamination was similar in all types of establishments and from street carts, except lower in five-star hotels and higher in one-star hotels (Table ). This observation suggests contamination of raw rice by heat-resistant spores, which is compatible with data of Gilbert and colleagues (-). Only one sample of uncooked rice was collected; it contained. XlO B. cereus per g. Thirty-one percent of of cooked rice that were for B. cereus had more than per g. This is much higher than the levels found in Chinese-style rice by Bryan et al. () and Schiemann (). By classifying temperatures into - and -degree intervals (including intervals of - C and C degree and higher at which foodborne pathogens do not grow; - C at which many foodborne pathogens grow quite rapidly; - C which encompasses the temperature that is optimum for growth of B. cereus; and - C at which foodborne pathogens multiply quite slowly), an interesting association between temperature and aerobic colony counts is observed (Table ). Aerobic colony counts ( C) per g were usually < ; frequently <, when rice was held at C or above. s of organisms in rice generally increased as the rice was held at temperatures that progressively declined below C and as the duration of holding after cooking increased. Counts frequently exceeded, usually or greater, when the temperature at collection was between and C. Counts usually exceeded when the temperature at collection was between and C. The correlation coefficient (r) between temperatures and log of aerobic colony counts was -. (k =.; r =.). Thus an inverse relationship exists between temperatures (from C C to C) and log of aerobic colony counts. This correlation is not exactly a high negative, but it is well above a low negative. From this correlation procedures, % of the variance in counts can be accounted for by temperature (and 8% cannot). Therefore, there is a % reduction in predictive error over what it would be if the correlation (r) were. Possible reasons for the error variance could be the different types of rice products sampled, limitations of collecting only one sample from a production batch, variation in laboradishes to aerobic colony ( C) count. and percent of having aerobic colony count ( C) per gram -< () a Four not tested for aerobic colony counts. -< (8) -< () -< 8 () -< () > 8 () 8 () () () (8) () ( ) ( ) JOURNAL OF FOOD PROTECTION, VOL. 8, JANUARY 8

EL-SHERBEENY ET AL. tory media prepared on different days, and procedures (e.g. weighing, mixing, presence of clumps, diluting, incubation time and temperature, counting and other normal laboratory errors). Rice was prepared in a variety of ways, by different equipment, in the various establishments. Different methods of cooking and whether spices, nuts, or other ingredients were added would have altered microbial counts. Some ingredients might have altered ph, which would influence bacterial growth. Also, whether lids or other covers were used and method of storage would have affected the water activity, which in turn influences bacterial growth. Stored rice may have been subjected to contamination by dust, a likely possibility of rice sold by street vendors and small cook shops. Little control of these differences could be made. Had there been, however, it is hypothesized that the negative correlation would have been greater. Rice was more frequently stored at temperatures greater than C in four- and five-star hotel kitchens than in the other types of establishments or on street carts (Table ). Rice in three-star hotels was usually within a temperature range of C to C. Rice in one- and two-star hotels was usually at temperatures between C and C. Rice sold by street venders always ranged between C and C, usually at the ambient air temperature. The median counts from collected in fourand five-star hotels were quite low compared to median counts from all other types of establishments and rice sold by street vendors. Counts tended to become progressively higher as the establishment rating decreased. Microbial counts of street-vended rice had a particularly high median (8.xl ). In England, a Public Health Laboratory Service Working Party on Microbiological Specification for Foods ( found a larger percentage of of rice to have aerobic colony counts in excess of per g from establishments in which rice was stored at room temperature than from establishments in which rice was stored deepfrozen, refrigerated or hot, or that served rice freshly cooked. These percentages were highest in rice that had been held at room temperature for more than h. Similar results were observed in Egypt. Aerobic plate counts ( C) were usually lower for fried and Oriental rice than for other rice dishes (Table ). Fried rice is reheated after boiling so lower numbers in this product are not surprising. Staphylococcal enterotoxin and B. cereus emetic toxin, however, are quite heat-resistant and would be expected to survive the heat generated in rice during frying, although vegetative cells of these organisms would be killed. Oriental rice was frequently served in four- and five-star hotels where storage facilities and practices were better than in the other types of establishments. CONCLUSION Rice may be responsible for diarrheal diseases in Egypt as it is elsewhere. Foodborne pathogens were isolated from of rice, although usually in small numbers. Aerobic colony counts, on the other hand, were frequently quite high. Rice was frequently stored at safe (> C) temperatures in four- and five-star hotel kitchens because adequate hot-holding facilities were available and usually properly used in these establishments. Three-star establishments often had such facilities, but they were seldom used properly. With slight modification, additional heating elements (such as individual gas burners under iron griddles) could hold cooked rice and other foods at safe temperatures (J C) in one-, two-and three-star hotels, small restaurants and small cook shops. Hazards associated with cooked rice can be monitored by measuring the temperature of rice. Rice found to be at temperatures lower than C, particularly that found at temperatures between to, is likely to have large quantities of microorganisms. Prevention of foodborne disease problems from rice in foodservice establishments must be based on (a) preparing small volumes of rice several times throughout the day, (b) keeping cooked rice at temperatures of C or above, (c) cooling leftover rice or rice prepared ahead of planned service in refrigerators, and (d) frying rice to TABLE. Storage temperature and median aerobic colony counts ( C) or rice and dishes containing rice prepared in food servic establishments or sold by street vendors in Egypt. of within temperature range ( C) Place rice prepared or sold Hotels five-star four-star three-star two-star one-star Subtotal Restaurants Small, cook-shops Street vendors 8 - - 8-8 - 8 - - + Median count.ox. XlO.x 8.X. x. xlo.xl.8xl 8.x".X JOURNAL OF FOOD PROTECTION, VOL. 8, JANUARY 8

CHARACTERISTICS OF EGYPTIAN RICE DISHES temperatures that exceeded C. (See Bryan et al. () Gilbert () and Gilbert et al. (8) for details about preventive measures.) Many of the above procedures are impractical for use by street vendors. In Egypt, usually hot and dry, the water activity of street-vended rice might decrease in time, as suggested by Bryan et al. (), so as to slow or perhaps even prevent growth of pathogenic bacteria. Either growth of some organisms or dust-blown contaminants, most likely accounts for the large quantities of aerobic organisms that were observed in street-vended rice. A practical means of controlling this product, other than prohibiting its sale, has not yet been devised. Health risk associated with large masses of rice that are kept at room or outside ambient temperatures must be realized and efforts initiated to change such practices. ACKNOWLEDGMENTS Thanks is given to Dr. Ahmed El-Akkad for his administrative support to this project and to Brince M. Mousa, Merfat M. Hossani and Essam M. Farid for their assistance in the laboratory. Thanks is given to Dr. George K. Morris, Centers for Disease Control, for confirming the isolate of Salmonella and assisting in training laboratory staff. Thanks is also given to Judy M. Gantt and Grady Waters (CDC) for consultation on satistical calculations. REFERENCES. Beckers, H. J.. The bacteriological quality of boiled and fried rice from Chinese restaurants in relation to Bacillus cereus food poisoning. Arch. Lebensmittelhyg. :-8.. Beuchat, L. R., C. F. A. Ma-Lin, and J. A. Carpenter. 8. Growth of Bacillus cereus in media containing plant seed materials and ingredients used in Chinese cookery. J. Appl. Bacteriol. 8:-.. Bryan, F. L., C. A. Bartleson, and N. Christpherson. 8. Hazard analysis in reference to Bacillus cereus of boiled and fried rice in Cantonese-style restaurants. J. Food Prot. :-.. Centers for Disease Control. -8. Foodborne and waterborne disease surveillance, Annual summaries. -8. Centers for Disease, Atlanta.. Gilbert, R. J.. Bacillus cereus. In H. Riemann and F. L. Bryan (eds.), Foodborne infections and intoxications, nd ed., Academic Press, New York.. Gilbert, R. J., and J. M. Parry.. Serotypes of B. cereus from outbreaks of food poisoning and from routine foods. J. Hyg. 8:-.. Gilbert, R. J., and A. J. Taylor.. Bacillus cereus food poisoning pp. -. In F. A. Skinner and J. G. Can (eds.), Microbiology in agriculture, fisheries and food. Academic Press, New York. 8. Gilbert, R. J., M. F. Stringer, and T. C. Pierce.. The survival and growth of Bacillus cereus in boiled fried rice in relation to outbreaks of food poisoning. J. Hyg. :-.. International Commission on Microbiological Specifications for Foods (ICMSF). 8. Microorganisms in foods.. Their significance and methods of enumeration, nd ed., University of Toronto Press, Toronto.. Lefebre, A., C. A. Gregoire, W. Brabant, and E. Todd.. Suspected B. cereus in fried rice. Epidemiol Bull. (Canada) :8.. Mortimer, P. R., and G. McCann, G.. Food poisoning episodes associated with Bacillus cereus in fried rice. Lancet :-.. Public Health Laboratory Service Working Party on Microbiological Specification for Foods.. Summary on colony counts and predominant organisms in of boiled and fried rice from Chinese and Italian restaurants and "take-away" shops. Environ. Health 8:-.. Raevoure, M., T. Kiutamo, A. Niskanen, and K. Salminen.. An outbreak of Bacillus cereus food poisoning in Finland with boiled rice. J. Hyg. :-.. Schiemann, D. A. 8. Occurrence of Bacillus cereus and the bacteriological quality of Chinese "take-out" foods. J. Food Prot. :-.. Terranova, W., and F. A. Blake. 8. Bacillus cereus food poisoning. N. Engl. J. Med. 8:-. JOURNAL OF FOOD PROTECTION, VOL. 8, JANUARY 8