Abstract
Recent outbreaks of disease associated with water contaminated by the parasite Cryptosporidium have heightened public awareness of the need for pure water. Since Cryptosporidium seems to show no regard for the common methods of micro organism control domnick hunter in conjunction with a major water authority have conducted extensive trials using cartridge filtration and live oocysts to simulate realistic worst-case scenarios for cartridge filtration as an effective final barrier against cyst contamination. This introduction is intended to inform and educate and serves as part of a major report study which will be released in the near future.
Introduction
Over the last fifteen years the water industry has been much troubled by the presence in raw waters of two protozoal parasites - Cryptosporidium parvum and Giardia intestinalis. These organisms are now believed to be the most common protozoal causes of diarrhoeal disease in man. Before this time the presence of these agents in water was largely unrecognised. Whereas in most normal healthy persons, infection gives rise to a transient diarrhoeal episode, in the immunocompromised , for example organ transplantation patients or those sufferng from AIDS, it can cause a prolonged and life threatening disease. It is now known that Cryptosporidium is widespread in the environment and as a result most surface and many underground sources of drinking water are now regarded as vulnerable to contamination. In the UK sheep and cattle appear to be the main sources although contamination does occur in a range of wild animals and domestic pets. Its presence in raw waters has been reported from all parts of the world where tests have been undertaken and its occasional presence in fully treated drinking water supplies follows a similar pattern. Surface waters are also known to be contaminated with Giardia cysts. In the UK the main source appears to be sewage effluents although wildlife may also act as a host. The presence of these pathogens in water supplies presents a major problem for the water industry and its customers. This is largely due to the exceptional resistance of the infective stages to most of the disinfectants in routine use, in particular chlorine. In addition, contamination is not reliably indicated by the range of established bacterial indicators of faecal pollution such as Escherichia coli. For several years the industry has been struggling to find effective answers. Initially, nothing was known about the distribution of the parasite in the environment, sources of contamination, their survival or their susceptibility to, water treatment processes. In addition, there were no established methods for the concentration, isolation and identification of the parasites in water. Research teams, particularly in the UK and USA, have moved to fill the gaps in our knowledge, and information is now much more readily available. Recently, methods of enumeration and analysis have been developed. At the same time there is a better understanding of how and to what extent parasites can be removed by water treatment processes. Water companies are incorporating more effective processes into treatment regimes and the legislative framework, within which water companies operate, is being modified to take account of the new information. At present not all water treatment undertakings have in place process procedures robust enough to ensure that these parasites do not occur in water supplies, particularly in the developing world. Although the two parasites have many similarities and often occur together it is generally recognised that Cryptosporidium poses the greatest threat. Certainly most of the major outbreaks have been attributable to this organism rather than to Giardia. The Cryptosporidium oocyst is smaller than the Giardia cyst, 4µm to 6µm as opposed to 10µm to 16µm, and therefore more difficult to remove by filtration. The oocyst is also more resistant to disinfection; chlorine has some limited use against Giardia but is completely ineffective against Cryptosporidium. As it is generally acknowledged that treatment processes which remove Cryptosporidium will at the same time remove Giardia, research has tended to concentrate on the Cryptosporidium. Numerous outbreaks of water-borne cryptosporidiosis have been recognised worldwide - several of major public health significance. The largest to date occurred in 1993 in Milwaukee, USA, with some 400,000 infected and 100 deaths; the largest in the UK was at Oxford in 1989 (5000 at risk, 400 requiring medical treatment) and the most recent in Sydney, Australia, where over 3 million people were advised to boil all water. Fortunately in this latest incident there was little if any illness. In addition, there have been many lesser outbreaks and numerous “boil orders”, a precautionary measure issued when a water company has reason to believe that microbial contamination may be present in its distribution system. Apart from the effects on the health of those involved the economic costs of the outbreaks have been tremendous. These include factors such as rebates, compensation, modifications to plant, extra processing, adverse publicity and legal fees. Costs associated with the Sydney incident for example have been estimated at tens of millions of dollars.
Cryptosporidium
Cryptosporidium paravum is a protozoan parasite with a wide range of hosts. It can infect most warm blooded mammals, including farm animals, domestic pets and humans. Similar species occur in birds and reptiles but these are not known to be infective to humans. Infection is by the oral route and transmission via contaminated food, drinking water, swimming pool water and sexual activity has been demonstrated. On entering the body, the organism passes through the stomach and becomes established in the intestine where it multiplies to very large numbers. Its life cycle is complex with numerous stages involving both sexual and asexual reproduction, ultimately resulting in the production of resistant oocysts which are excreted in faeces. The numbers excreted are impressive, for example, one gram of faeces may contain sufficient oocysts to infect a million people. Infection in normal healthy humans usually results in prolonged diarrhoea lasting from a few days to as long as three weeks. The infection is self limiting and usually gives rise to long term immunity. However, in certain cases - in particular among the old, the weak and those with damaged immune systems the disease can be serious and even fatal. Those particularly at risk from infection are those infected with the AIDS virus or organ transplantation patients. As yet, there is no specific therapy available. The resistant form of Cryptosporidium is the oocyst. This is a spherical cyst some 3µm to 6µm in diameter with a tough cell wall and containing four banana-shaped sporozoites. The sporozoites are the infective stage and are liberated when the oocyst passes through the stomach. Identification of the oocyst is usually by means of immunofluorescent microscopy. The characteristics used to identify the oocyst are its size and shape, its staining properties - using fluorescent linked monoclonal antibodies - and its contents when viewed by differential interference contrast microscopy (DIC). Identification in environmental samples is a labour-intensive and highly skilled operation. This could result in an inaccurate measure of the concentrations of oocysts present in an outbreak or even compromise the detection of a low level outbreak. Empirical work by Thames Water Utilities Limited in the area of detection and enumeration of oocysts now offers increased levels of accuracy.
Commercial Implications
The possible presence of this organism in water supplies, presents a major threat to many commercial undertakings. In particular, those companies producing drinks, foodstuffs or medicines for human consumption must be able to ensure that the organism is not present in their finished product. The risk of contamination will vary depending on the number of organisms present, their viability, the production processes involved and the final product. The cyst is not invulnerable - it is easily killed by heat - but its presence in the water supply of, for example, a food manufacturer constitutes an avoidable risk. Such firms may consider it advantageous to have available a system capable of removing oocysts from their water supply. In this circumstance a process such as filtration which removes the organism has an advantage over treatments which merely kill it. Most detection procedures rely on microscopy and employ staining techniques specific for the oocyst wall to identify the organism. They demonstrate its presence not its viability. Hence an oocyst which has been killed by heat may still give a positive result. Viability tests are available but these are slow, difficult and expensive. They either employ animal infectivity or use dye penetration to test the integrity of the cell membranes. A further problem is that with some methods of killing oocysts - ultraviolet irradation for example - effects on the cell membranes may not be apparent until some time after treatment has finished and as a result the viability test may not give an accurate result.
Removal of Cryptosporidium
One of the problems with Cryptosporidium in water is the erratic nature of the contamination combined with the need to reduce concentrations to a very low level in treated water. The levels of contamination in raw waters often fluctuate with very low or undetectable levels most of the time but with occasional high concentrations. It is important to bear in mind that oocysts are often present in clumps and in association with other material - they do not occur naturally in a monodispered suspension. The latest regulations proposed by the UK government specify that in drinking water concentrations should be less than 100 oocysts per 1000 litres. As most major outbreaks of waterborne cryptosporidiosis have been associated with concentrations of less than one oocyst in 10 litres of water this raises a serious question as to the utility of these regulations. The traditional approach to removing microbial contaminants from water has been to employ a variety of processes to reduce them to a low level, then to apply disinfection to remove the last traces. The problem with cryptosporidium is that disinfection as traditionally used by the industry does not work. Most drinking waters are treated with chlorine at doses of about 1-3 parts per million (ppm). The doses required to kill oocysts is in the order of 1000ppm; a completely unrealistic treatment. At present the only way of ensuring removal from water is by means of high grade filtration or heat disinfection. Information on the removal efficiencies of water treatment processes is increasing rapidly.Treatments such as reservoir storage followed by slow sand filtration, as practised in the London area, are known to give a good level of protection but conventional coagulation and sedimentation may reduce concentrations by less than two log units (99% efficiency), even when operating under optimal conditions. Rapid gravity filtration in combination with the above can be effective but care has to be taken over the recycling of backwash waters. In the rapid gravity process some 2 to 5% of the filtrate is used to backwash the filter when it becomes blocked. This dirty water is normally led into a lagoon, where it is allowed to settle, and the clear supernatant fed back into the process. There is a danger that oocysts will remain in suspension and be recycled ultimately building up to a point at which they break through into the filtered water. Given that many water treatment works do not always operate under optimum conditions and that the erratic nature of the contamination can suddenly impose excessive loads, it is not surprising that there have been numerous reports worldwide of oocysts in fully treated drinking waters.
The solution is to use a high efficiency microfiltration process to guarantee 100% oocyst removal.
The domnick hunter Objective
To challenge cartridge filters with high levels of viable Cryptosporidium oocysts under realistic conditions and to measure their retention properties.
Filter Properties
Microfiltration removes suspended contamination from gas or liquid systems in the size range 0.01µm to 100µm. The filters employed to achieve this contamination removal can be divided broadly into two categories based upon their structure and methods of retention.
Depth Filters
Depth or media filters consist of melt-blown or spun bonded polymers produced in filter materials between 1000µm and 2000µm in thickness. Particle or micro-organism retention takes place both on the surface and (predominantly) within the depth of the filter medium. Retention is based upon inertial impaction and adsorption of the particles or micro-organisms as they are carried through the tortuous filter path of the media. As the depth filter retains contamination the differential pressure across the filter will increase and the flow rate may also decrease depending upon the delivery system to the filter. The high percentage of free air space (or voids volume) associated with depth filters facilitates high flow rates and excellent filter life. Depth filters are not designed to be routinely integrity tested to measure their continued ability to produce the desired level of retention.
Feed Surface
Filtrate
Typical example of depth filtration
Membrane Filters
Membrane filters have a defined pore structure producing size exclusion retention of particulate and micro-organisms. This predominantly surface retention from the homogeneous (or sponge like) pore structure results in membrane filters demonstrating lower flow rates and dirt loading capacities than depth filters. The defined pore structure in membrane filters does facilitate routine non-destructive integrity testing both before and during operation. This integrity test can be correlated to the retention of specific microorganisms during a destructive membrane filter bacterial challenge test. Efficiency ratings for depth filters are determined on their ability to retain inert particulate, for example AC Fine Test Dust. Membrane filters are rated on their ability to retain micro-organisms. Research in both the water and pharmaceutical industries has questioned the assumption that filter retention is determined by contaminant size only. For example, it is questionable to assume that a filter which can retain 100% of 5µm latex spheres will also retain 100% of Cryptosporidium oocysts or other similarly sized and shaped micro-organisms under identical challenge conditions. To provide a high degree of assurance regarding the removal of a specific micro-organism the filter should be challenged with the specific micro-organism in question. Membrane filters may be periodically integrity tested which ensures they are and they remain bacterially retentive. Cost-effective filtration is characteristically achieved through the use of a depth/media filter upstream of a membrane final filter. The depth filter removes the majority of the contaminants and the membrane filter provides a high degree of assurance regarding product quality. The membrane filter may also be routinely integrity tested. It is worth noting filtration will remove a broad range of contaminants from the water not simply the specified micro-organisms.
Feed Surface
Filtrate
Typical example of membrane filtration
