Waterborne disease has plagued humankind since before recorded history. However, the introduction of simple water supply technologies such as wells, filters, and chlorine disinfection has made classic waterborne diseases such as typhoid fever, cholera, and dysentery unusual in the industrialized world. However, with increased standards of living and the emergence of diseases such as AIDS, the desire to improve the level of public health is great. Furthermore, the purveyors of water to homes must be aware that there is an expectation of a quality product delivered to the point of use. The organoleptic properties should be neutral and the water should be free of microbial pathogens. Also, long-term consumption of such water should not increase the lifetime risk of cancer or other illnesses.

Figure 1. Giardia lamblia cysts.	Figure 2. Cryptosporidium parvum oocysts.

Unfortunately, since the 1980s encysted waterborne parasites such as Giardia lamblia and Cryptosporidium parvum have presented a challenge to water suppliers. These tiny microbes have forced the water industry to take notice of two important problems:

• increasing the concentration of the disinfectants during treatment may help prevent the transmission of disease by the water route; but
• by-products of the unwanted reactions between naturally occurring organic material and the chemical disinfectant may be carcinogenic in humans.

So the dilemma is to balance these difficult issues. We will leave the debate of that issue to another time. This article will focus on the issues related to the former problem: how do we meet the challenge of these modern, microbial giants?

Until human volunteers established the infectivity of G. lamblia in the 1950s, it was commonly thought to be a nonpathogenic inhabitant of the gut. Today G. lamblia is widely recognized as the most frequently identified cause of waterborne disease in North America. However, an interesting fact is that unknown agents remain the most common cause of waterborne disease in North America.

Several waterborne outbreaks of giardiasis have been identified in the past 20 years. The outbreaks have occurred as a result of a variety of conditions including use of untreated surface water, contaminated water distribution systems, and treatment deficiencies. Outbreaks and cases of giardiasis from systems using untreated surface waters or surface waters which have only been treated with chlorine, account for the majority of reported incidents of waterborne giardiasis. Some of the best documented North American waterborne outbreaks of giardiasis include those of Rome, New York in 1974-75; Camas, Washington in 1976; Berlin, New Hampshire in 1977; Banff, Alberta in 1982; and Edmonton, Alberta in 1982-83.

There are two stages in the life cycle of Giardia in warm-blooded vertebrates: the trophozoite and the cyst. The trophozoite commonly inhabits the duodenum while the cyst is found in the small and large intestine and the feces. Giardia lamblia is about 12 to 15 µm long and 6 to 8 µm wide. Trophozoites divide by longitudinal binary fission.

The cyst is the normal infective stage in the transmission of giardiasis to new hosts. Giardia cysts are ovoid in shape and have a length of 8 to 12 µm and a diameter of 7 to 10 µm depending on the species (Figure 1). The cyst wall varies from 0.3 to 0.5 µm in thickness and electron microscopy has shown it to be composed of thin fibrous elements interspersed with fine particles. After emergence from the cyst, the organism undergoes cytokinesis within 5 to 30 minutes after the initiation of excystation. Thus, each mature cyst yields two trophozoites.

Giardia lamblia have been isolated from wild animals that inhabit aquatic environments such as muskrats and beavers. In Banff, Alberta, the giardiasis outbreak that resulted from the beavers occupying the local water supply was know as "Beaver Fever". Sewage contamination of a water supply can also be an important source of concern.

The median infectious dose of Giardia cysts in humans is thought to be between 50 and 100 cysts, though some people can be infected with as few as 10. There is drug therapy that is available for giardiasis but those that have used it have reported that "the cure is worse than the disease".

Cryptosporidium has become recognized as a frequent cause of waterborne disease in humans. Cryptosporidiosis outbreaks from surface water supplies have been documented in the United States and Great Britain and it has been speculated that many other cases of waterborne outbreaks of gastroenteritis may have been caused by Cryptosporidium. The water supply conditions under which the outbreaks have occurred have been similar to those for the giardiasis outbreaks. Of interest is the fact that concurrent infections of Giardia and Cryptosporidium have been observed in some patients.

The life cycle of Cryptosporidium can be summarized by six events: excystation of the oocysts in the intestine of the host, replication within the host, gamete formation, fertilization, oocyst wall formation, and sporozoite formation. Of most interest to the water supply industry is the oocyst. The oocyst occurs in two forms, one with a thin wall which is autoinfective within the host and is not believed to survive outside of the host. The other is a thick-walled oocyst which is capable of surviving for several weeks in the environment and is the main means for transmission of the parasite. The oocyst is approximately 5 µm in diameter but can vary from this size and can be elongated depending on the species (Figure 2). It has been observed that oocysts are capable of passing membrane filters greater than 1 µm in pore size.

C. parvum appears to lack host specificity and has been shown to be able to crossinfect rodents, ruminants, and humans. Domestic cattle and pigs have been implicated in outbreaks of cryptosporidiosis as has water contamination from sewage.

Prepatent and patent periods have been reported to range from 2 to 10 days and 1 to 33 days, respectively, depending on the host. At this time, there is little known about the infective dose of oocysts required to cause infection in the hosts. One study that used immuno-competent volunteers reported that the median infectious dose was approximately 130 oocysts per volunteer. Some people can be infected with as few as 30 oocysts. There is currently no effective drug therapy for cryptosporidiosis.

It is interesting to contrast the infectious doses required to develop clinical symptoms of parasitic diseases with the bacterial diseases. Large infectious doses are required to overwhelm the immune system in the case of bacteria. Very low concentrations of parasites in the source water make them a health risk but can also make them hard to detect relative to bacteria.

The nature of parasites is that only one infectious cyst or oocyst is necessary to set up the life cycle in a host. One reason people do not display clinical symptoms is that a healthy immune system will prevent the organism from successfully completing its life cycle. If several cysts or oocysts start life cycles at once, the immune system may not respond fast enough to prevent clinical manifestation of the disease, but eventually the immune system catches up (typically one or two weeks) and the host spontaneously clears the infection.

Of particular note at this point is that while immuno-competent individuals are at little risk of serious disease from encysted parasites, a significant proportion of the population have deficient immune systems. Perhaps most well known are people with AIDS. However, once one understands that the very young and very old, cancer patients, transplant patients and others who are on immune-suppressing chemotherapy are susceptible to very low infectious doses of encysted parasites, there is a significant proportion of the community that may be at risk from waterborne parasites. Of great concern is the fact that there is no drug therapy available for patients with cryptosporidiosis: immune-deficient people may have chronic parasitic illness that lasts for weeks or months or they may even die.

Establishing risk of contracting the illness associated with waterborne parasites through the water supply requires knowledge of the numbers of infectious units in the source water and in the finished water. The following discussion highlights some of the issues related to finding a few parasites in a large volume of water.

Detection of cysts and oocysts in water samples involves two basic steps: concentration of a sample of water since cysts are typically in low numbers; and a means of identifying cysts and oocysts in the concentrated sample. The reference method for concentrating encysted parasites has been filtering large volumes of water using wound filters. Between 100 and 1000 L of water (the less pristine, the smaller the volume) are filtered through a 1 µm nominal pore size, yarn-wound, polypropylene filter at the sampling station. The filter is then sent to a specially certified laboratory where the filter cartridge is opened and the contents are washed repeatedly until all (presumably) of the captured particles are collected in the wash water. The wash water is concentrated by centrifugation.

Then a small volume of the concentrated pellet (usually about 1 mL) is resuspended and centrifuged again using a Percoll-sucrose gradient (specific gravity 1.1) whereby the parasites are separated from the water column. The layer containing the parasites is then removed and treated with a fluorescent antibody that attaches to Giardia cysts and Cryptosporidium oocysts. Microscopic examination using ultraviolet lamps and the appropriate filters reveals objects that are of the size and shape of the encysted parasites. The number of parasites in the sample water can then be obtained by relating the sample size that was counted back to the original sample volume.

The use of membrane filters has also been advocated for cysts and oocysts. Recoveries of Giardia cysts have been reported to be from 22 percent for membrane filters to 52 percent using wound filters. Filters would probably recover greater than 70 percent of the cysts but cysts are lost in intermediate processing steps. Cryptosporidium recoveries have been reported to range from 9.5 percent in river water to 59 percent in tap water.

In recent studies in the United States it was found that, when spiked samples were sent to a number of certified laboratories, there was little consistency in the final results with wide ranges on the recoveries and accuracy. One may question the usefulness of obtaining tests on the source and finished water in view of the nature of the methods that are currently available.

The detection of cysts or oocysts in a sample provides no information about the viability of the parasite and its ability to cause an infection. Several methods have been used to estimate the viability of cysts and oocysts including vital dye exclusion, in vitro excystation, cyst morphology by light microscopy, the uptake or exclusion of fluorogenic dyes, and animal infectivity models. Of these methods, only the animal models provide direct information about the ability of the organisms to cause an infection in the host.

There have been many surveys conducted throughout the world on the incidence of waterborne parasites in drinking water. Given the tenuous nature of the methodology used to count parasites in drinking water, the best that one can conclude is that encysted parasites are ubiquitous in surface waters and that drinking water treatment processes should be designed to withstand challenges from these organisms.

In our next issue, Dr. Finch will discuss control techniques for waterborne parasites. (Part II)