---

Protozoan parasites in water and food

The recent contamination incident in Sydney water supplies highlighted how little we know about the behaviour of waterborne protozoan parasites in food systems.

Cryptosporidium parvum and Giardia lamblia are two protozoans which are significant as waterborne pathogens around the world. The main symptoms of infection are watery diarrhorea and abdominal cramps. A major waterborne outbreak of Cryptosporidium infection occurred in Milwaukee (US) in 1993 when over 400,00 cases and approximately 100 deaths were reported. During this outbreak, Cryptosporidium carried in tap water was suspected of contaminating various foodstuffs which were recalled (California Agriculture 51 (2) 1997 8-14).

There has been one documented incident of foodborne cryptosporidiosis in Australia when a mother and her one-year old child were believed to be infected by drinking unpasteurised goat milk. Unpasteurised cow's milk is the food most commonly associated with foodborne cryptosporidiosis while other foods to which infection has been attributed are fresh pressed apple juice, salad, raw meat and frozen (uncooked) tripe. Incidents of foodborne infection caused by Giardia have been attributed to fruit salad, sandwiches, fresh vegetables, noodle salad and home canned salmon (Journal of Food Protection 56 (5) 1993 451-56). In each of these giardiasis incidents, the implicated foods appear to have been contaminated by an infected food handler. In the most serious incident connected with sandwich preparation in a combined nursing home-child care centre, corrective procedures included training in hand washing procedures and the removal of all infected individuals from food service operations.

Cyclospora, another protozoan parasite which was not involved in the Sydney water scare, has been associated with foodborne infection in the United States via raspberries, basil and pesto sauce made from basil.

Resistance
The resistance of these protozoa in cyst form to standard water disinfection procedures has been well reported. Of the disinfectants commonly used to treat water, only ozone appears to be effective in destroying the oocysts of Cryptosporidium parvum which is the most resistant of the three. Korich and co-workers (Applied and Environmental Microbiology 56 (5) 1990 1423-1428) reported that 1 mg/L of ozone for 5 minutes achieved greater than 90 percent inactivation. These authors concluded that with the possible exception of ozone, the use of disinfectants alone should not be expected to inactivate C. parvum in drinking water. Filtration of water through a filter with a pore size of one micron or less is required to effectively remove cysts. Milwaukee authorities installed ozone treatment plants to supplement filtration after the 1993 incident.

Food processing operations
Few studies have been conducted on the effect of food processing operations on oocysts of Cryptosporidium or Cyclospora and Giardia cysts. Most studies in water have been conducted with Cryptosporidium which is regarded as the most resistant of these organisms.

Heating
From limited studies it is probable that standard pasteurisation procedures will inactivate these organisms. Heating Cryptosporidium cysts in water to 72.4°C within one minute or holding cysts in water at 64.2°C for two minutes rendered cysts non-infective to mice (International Journal of Food Microbiology 31 1996 1-26). The thermal death time point (the temperature at which organisms are destroyed) of Giardia cysts has been reported as 62°C (Journal of Food Protection 56 1993 451-56).

It has to be stressed that the resistance of these cysts in a complex food medium such as milk may be different from that exhibited in water. Laberge and co-authors (International Journal of Food Microbiology 31 1996 1-26) in reviewing available data conclude it is unclear whether HTST pasteurisation (72°C/15 sec) would lower the level of infective oocysts below the infective dose for humans. The level of contamination used in the study reported above (1 million oocysts/ml) is much higher than one would expect in raw milk.

Acidity
There is one reported environmental/laboratory study of cryptosporidiosis attributed to an acid food - fresh pressed apple juice (Journal of the American Medical Association 272 1994 1592-96). This is in fact the best documented outbreak attributed to food. Unfortunately the pH of the juice involved is not recorded but it may be assumed to be around 3.8-4.0.

Staff and students drank the juice the same afternoon it was prepared at a school agricultural fair in Maine, USA in 1995. State Bureau of Health workers subsequently obtained a sample of the partially fermented apple juice ten days after the fair. The juice was then frozen and stored for six weeks before Cryptosporidium oocysts were counted. Up to 750 oocysts/litre were recorded and the investigators believe the count was probably much higher. This indicates the ability of the oocysts to survive at the pH of the apple juice for several days at ambient temperature and weeks during frozen storage.

Two experimental studies confirm the relative tolerance of Cryptosporidium oocysts to low pH. Friedman and co-workers (Journal of Food Safety 17 1997 125-32) report a study in which oocysts were inoculated into beer, cola, orange juice and infant formula. The beer and cola were carbonated and the beer had a pH of 3.8 while the cola had a pH of 2.5. The pH of the orange juice was 3.9 and that of the infant formula 6.6. A pH 4.0 buffer as well as ethanol solutions were used as controls. Samples were incubated and 4°C and/or 22°C for 24 hours and viability of inoculated oocysts determined. The authors calculated that after this time there was a loss of <85% oocyst viability in beer or cola stored at 4°C while the loss of viability in water, orange juice and infant formula was equal to or less than 35 percent. The study did not permit estimation of the time required to achieve complete inactivation of Cryptosporidium oocysts.

The second study (FEMS Microbiology Letters 142 1996 203-208) used a laboratory medium with pH adjusted to 2.0 (hydrochloric acid), 4.0 (acetic acid), 6.0, 8.0 and 9.5. Cryptosporidium parvum, the common human pathogen, was one of two Cryptosporidium species tested. The initial population of cysts used was about 1 million/ml. These workers reported apparent zero viability of cysts after 60 minutes at pH 2.0 but 85 percent viability after the same time at pH 4.0. No other incubation time was studied.

Freezing
The study reported with apple juice indicates that oocysts of Cryptosporidium are able to survive at least some weeks of frozen storage in a suitable medium. An experimental study (Applied and Evironmental Microbiology 58 1992 3494-3500) using purified water as the medium showed that a small proportion of oocysts survived 750 hours at -22°C after slow freezing. Snap freezing using liquid nitrogen resulted in 100 per cent loss of viability. The initial population used in this study was again about one million cysts/ml.

The case cited above where frozen tripe acted as a vehicle for infection is further confirmation that freezing cannot be relied upon to destroy all Cryptosporidium cysts.

Drying
No published studies have investigated the effect of drying of oocysts in or on foods. However experimental studies have shown that drying of cysts suspended in water on glass surfaces at ambient temperature resulted in 97 percent loss of viability after two hours and total loss after four hours.

It is clear from the above that we need to know a lot more about the behaviour of waterborne parasites especially Cryptosporidium and Giardia in food systems. While this knowledge is being accumulated, it is essential that food processors who use mains water either for washing foods or as an ingredient in foods for which there is no terminal heat process, develop contingency plans in case of similar incidents in the future.

For further information contact:
Dr Trish Desmarchelier
Brisbane Laboratory, Telephone 07 3214 2000

---

Prepared by Keith Richardson and Beverley George
Food Science Australia
PO Box 52, North Ryde 1670. Tel +61 2 9490 8397 Fax +61 2 9490 8499
Email enquiries@csiro.au