INNLEGG FRA MØTER I FORENINGEN
Analysis of water for protozoan parasites is expensive – why not use indicator organisms?
Av Lucy J. Robertson
Lucy J. Robertson er professor i parasittologi ved Institutt for mattrygghet og infeksjonsbiologi, Norges miljø- og biovitenskapelige universitet.
Innlegg på fagtreff i Norsk vannforening 4. ridium. Such analyses are not necessarily to november 2013. determine whether a contamination event has already happened, for by the time the analysis is Why analyse water samples for completed the water might well be already dis protozoan parasites? tributed to consumers, but also for a range of The protozoan parasites, Cryptosporidium spp. other reasons, including to be used for input to and Giardia duodenalis, have been associated risk assessment, to determine contamination with hundreds of waterborne outbreaks of sources in catchments, to assess whether suffi disease, some of which have involved hundreds cient barriers are in place to ensure safe supply or even thousands of individuals (see: Karanis et to consumers, and to follow up if an outbreak al, 2007; Baldersson and Karanis, 2011). In addi has occurred and to ensure that implemented tion to the direct medical issues, such outbreaks measures or barriers are working effectively. are also a substantial economic drain in terms of lost productivity, tracking down and addressing Standard Methods: the cause of the outbreak, and also in terms of why are they so expensive? possible treatment and hospitalization costs. Standard methods for analyzing water for proto Although countries with high endemic pre zoan parasites were first produced in around valence of these parasites and with poor infra 1980 in USA and have since been developed and structure regarding water supply and sewerage improved; the most commonly used Standard are perhaps more likely to experience outbreaks Methods today are US EPA Method 1623.1 (most of waterborne protozoan disease, it seems that updated version published in 2012 and developed no country can consider itself invulnerable, and from earlier versions) and ISO Method 15553 communitywide outbreaks of waterborne giar that was published in 2006. These Standard diasis and cryptosporidiosis have been reported Methods basically consist of 3 stages: concentra from Scandinavia in recent years (Robertson et tion (often by filtration), isolation (by immuno al, 2006; Robertson & Huang, 2012). Our aware magnetic separation – IMS), and detection (by ness of these outbreaks, as well as the fact that microscopy using specific fluorescent labels to the parasites are difficult to kill or remove from assist in identification). Each of these stages is water supplies, has led to the need for analyzing expensive – some of the filters recommended for water samples for both Giardia and Cryptospo the first stage in the analysis cost approximately
Vann I 01 2014 77 INNLEGG FRA MØTER I FORENINGEN
600 kr each (around 70 €); there is currently only ridium can infection humans, and at least one of a single company that supplies an IMS kit that is these, C. cuniculus commonly found in rabbits, suitable for isolating both Cryptosporidium has been associated with a waterborne outbreak oocysts and Giardia cysts, and the costs are usu (Chalmers et al, 2009), other species are of limi ally over 400 kr (around 50 €) per test, but may ted or negligible public health significance. be higher for more turbid water samples; and Similarly, with Giardia duodenalis, the species is although the fluorescent labels for detection are subdivided into different genotypes or Assembl not expensive, a fluorescence microscope for exa ages that are genetically distinct from each other mining the sample concentrates is, and such and also have different host-specificity. Thus, analyses are best performed by a properly trained only G. duodenalis isolates in Assemblage A or and well-experienced operator. Together, all B are infective to humans, but other genotypes these factors mean that the analysis of a single are commonly found in other animals; for water sample is likely to cost, at the very least, example cattle and sheep are often infected with around 4000 kr (480 €), and very probably more. G. duodenalis of genotype E that is not of public In addition, it is that the laboratory that under health significance. Therefore, identification of takes such analyses is established and respected species or genotype of parasites found in water in this field, with appropriate QC data, successful is also of relevance from a water safety perspec and regular participation in ring-testing, and, tive, but is an additional procedure and adds preferably, accreditation for the particular analy further to the cost of the analysis. tical method to be used. All these documentation and competence demonstration steps are expen The use of indicator organisms sive, and this cost will be passed onto the custo with respect to contamination mer providing the water sample. with Cryptosporidium and Giardia In addition, if parasites are detected, then the With analysis of water samples directly for Crypto water supplier and consumer are likely to be sporidium and Giardia being so expensive, it is interested in whether the species or genotype is not surprising that the water industry has long infectious to humans, and thus of public health been searching for a more cheaply and readily significance. There are about 25 named species detected indicator organism that may provide of Cryptosporidium, of which two, C. parvum information on when contamination with proto and C. hominis, are most associated with human zoan parasites is likely to have occurred. The US infection, and one of which, C. parvum is of zoo EPA lists seven criteria to be fulfilled for notic importance, being a cause of diarrhoeal organisms that may be useful to use as indicator disease in various animals, particularly calves organisms for Cryptosporidium and Giardia, see and lambs. Although other species of Cryptospo table 1.
Criteria for indicator organisms for Cryptosporidium and Giardia 1. The organism should be present whenever Cryptosporidium/Giardia are present. 2. The organism should be useful for all types of water. 3. The organism should have a longer survival time than Cryptosporidium/Giardia. 4. The organism should not grow in water. 5. The organism should be found in warm-blooded animals’ intestines. 6. The testing method should be easy to perform. 7. The density of the indicator organism should have some direct relationship to the degree of pollution. Table 1. Criteria for indicator organisms for Cryptosporidium and Giardia.
78 Vann I 01 2014 INNLEGG FRA MØTER I FORENINGEN
Finding an organism that fulfils all of these fulfil the first criterion in table 1, and therefore criteria is not easy. For example, Cryptospori is not necessarily appropriate as indicator of dium oocysts are notoriously robust, surviving contamination or potential contamination with weeks or months in damp, cool conditions and Cryptosporidium and/or Giardia. also surviving exposure to various disinfectants In the paragraphs above, specific publications to which other microorganisms are susceptible. have been used to demonstrate reasons why Thus, many of the most common faecal indica there are currently no appropriate indicator tor organisms do not fulfill criterion number 3. organisms for Cryptosporidium and/or Giardia. For example, a study was conducted in Spain in However, consideration of more studies may which four water reclamation facilities were provide more convincing data. In a publication monitored over 2 years to determine the occur by Wu et al (2011), data were collected on the rence and concentrations of a set of microbial relationships between indicator organisms and indicators (total coliforms, Escherichia coli, ente pathogens (including Cryptosporidium and rococci, spores of sulphite reducing Clostridia, Giardia) published in scientific journals between somatic coliphages, and some phages), along 1970 and 2009 (over a 40 year period). These data with those of two pathogens, one of which was were then sorted, with only data retained where Cryptosporidium (Costan-Longares et al, 2008). the methods of statistical analysis, correlation The results obtained demonstrated that several coefficients and p-values were reported and regression models and discriminant functions where correlation analyses were not grouped that were able to predict both the presence and across water types. In addition, due to the low concentrations of enteroviruses. However, neither detection of pathogens in groundwaters and index functions nor a predictive relationship treated drinking waters, these data were exclu were observed between any of microbial indica ded from the study. This resulted in 92 data tors and viable Cryptosporidium oocysts (Cos points for Cryptosporidium and 59 for Giardia; tan-Longares et al, 2008). logistic regression analysis demonstrated that of Another potential indicator organism, which the 92 Cryptosporidium cases, 69 (75%) found no is also more robust than those used in the study correlation with indicators and only 23 (25%) of Costan-Longares et al, (2008) is Clostridium found a correlation, while for Giardia 40 cases perfringens, as the spores of this species of bac (68 %) were uncorrelated and 19 (32%) were cor teria are very robust and can withstand various related. Although C. perfringens, faecal coli environmental p ressures, including high tem forms and total coliforms were suggested as peratures. It is also a normal component in the possible indicators for protozoan parasites, they environment and is found in the intestinal tract were not significantly better than others (Wu et of humans and other animals. However, the con al, 2011). The study concludes that indicator clusion from a three-year study in Europe was organisms cannot with certainty signal the pre that defining C. perfringens as a conservative sence of pathogenic contamination for a water indicator for total faecal pollution monitoring in sample, but long-term monitoring may indicate water was no longer justified; although C. perf the potential degree, and thus a relative risk ringens was shown to be a conservative indicator score. However, much of the problem is due to for faecal excreta from non-herbivorous wildlife insufficient data and sizeable site-specific moni and human-associated sewage, it was not suit toring efforts are necessary to define local public able for herbivorous wildlife (Vierheilig et al, health risk accurately (Wu et al, 2011). 2013). Due to the zoonotic nature of both Crypto sporidium and Giardia, and that species and The role of faecal source tracking genotypes of public health significance of both As all Cryptosporidium and Giardia in water ori parasites occur in herbivorous wildlife, these ginate from human or animal faeces, faecal findings indicate that C. perfringens may not source tracking has also been a field that has been
Vann I 01 2014 79 INNLEGG FRA MØTER I FORENINGEN considered with respect to water monitoring for using both catchment knowledge and epidemio these parasites. In a review article considering logical data, and then, in the light of a structured faecal source tracking, indicators and water qua risk assessment, use baseline surveying of indi lity (Field and Samadpour, 2007), the authors cators, targeted faecal source tracking, and spe make the point that emphasis on faecal indicator cific pathogen monitoring when appropriate to bacteria in legislation has resulted in the patho monitor the risk of contamination under diffe gens themselves being overlooked. The authors rent scenarios. Our arsenal of approaches to add suggest that a more rational approach to regula ress this issue of waterborne protozoan transmis ting water quality with respect to contamination sion is increasing, and includes not only the would start with using the available epidemiolo approaches described above but also more gical data to identify pathogens of concern in a advanced molecular techniques, use of Geogra particular water body (e.g. cattle operations phic Information Systems, and improving moni might suggest a potential for C. parvum contami toring of infection in the human and animal nation). When levels of the appropriate indica communities within the water catchment area. tors rise, then targeted monitoring for pathogens could be undertaken, thus minimising expensive References shotgun approaches, but ensuring that health Baldursson S, Karanis P. 2011. Waterborne transmis threats were being monitored. In addition, targe sion of protozoan parasites: review of worldwide out breaks - an update 2004-2010. Water Res. 45(20): ted source tracking could be used as required to 6603-6614. identify the probable source of pathogens, again avoiding expensive shotgun approaches (Field Chalmers RM, Robinson G, Elwin K, Hadfield SJ, and Samadpour, 2007). Targeted pathogen moni Xiao L, Ryan U, Modha D, Mallaghan C. 2009. Cryp tosporidium rabbit genotype, a newly identified toring coupled with targeted faecal source track human pathogen. Emerg. Infect. Dis. 15:829-830 ing and baseline monitoring of indicators would become just one tool among many. Costan-Longares A, Montemayor M, Payan A, Mendez J, Jofre J, Mujeriego R, Lucena F. 2008. Microbial indicators and pathogens: removal, rela Conclusion tionships and predictive capabilities in water recla Although monitoring of water for protozoan con mation facilities. Water Res. 42(17):4439-4448 taminants has improved in efficacy over the last decades, with improved filtration methods, new Field KG, Samadpour M. 2007. Fecal source tracking, the indicator paradigm, and managing water quality. isolation techniques, and molecular characteri Water Res. 41(16):3517-3538. zation possibilities readily available at many labo ratories, the analytical technique has not become Karanis P, Kourenti C, Smith H. 2007. Waterborne cheaper and there is still a need for trained and transmission of protozoan parasites: a worldwide review of outbreaks and lessons learnt. J Water experience laboratory personnel and proper qua Health. 5(1):1-38. lity control and accredited use of methods. Alt hough the use of an indicator organism to replace Robertson LJ, Hermansen L, Gjerde BK, Strand E, the requirement for such analyses remains elu Alvsvag JO, Langeland N. 2006. Application of geno typing during an extensive outbreak of waterborne sive, this does not mean that such approaches giardiasis in Bergen, Norway, during autumn and should not be used when considering public winter 2004. Appl Environ Microbiol. 72(3):2212- health and safe drinking water within the context 2217. of contamination with Cryptosporidium oocysts Robertson LJ, Huang Q. 2012. Analysis of cured meat and/or Giardia cysts. A shotgun approach is both products for Cryptosporidium oocysts following pos expensive and inefficient, and a more useful way sible contamination during an extensive waterborne of considering any drinking water catchment outbreak of cryptosporidiosis. J Food Prot. 75(5):982- with respect to these protozoan parasites is first 988. to identify potential sources of contamination,
80 Vann I 01 2014 INNLEGG FRA MØTER I FORENINGEN
Vierheilig J, Frick C, Mayer RE, Kirschner AK, Reis Wu J, Long SC, Das D, Dorner SM. 2011. Are micro cher GH, Derx J, Mach RL, Sommer R, Farnleitner bial indicators and pathogens correlated? A statistical AH. 2013. Clostridiumperfringens is not suitable for analysis of 40 years of research. J Water Health. the indication of fecal pollution from ruminant wild 9(2):265-278. life but is associated with excreta from nonherbivor ous animals and human sewage. Appl Environ Microbiol. 79(16):5089-5092.
Vann I 01 2014 81