- Email: [email protected]
Giardia and Cryptosporidium in backwash water from rapid sand filters used for drinking water production
Zbl. Bakt. 284,107-114 (1996) © Gustav Fischer Verlag, Stuttgart· Jena . New York
Giardia and Cryptosporidium in Backwash Water from Rapid Sand Filters Used for Drinking Water Production PANAGIOTIS KARANISl, DIRK SCHOENEN 2 , and HANNS MARTIN SEITZ l 1 Institute 2
for Medical Parasitology Hygiene-Institute, University of Bonn, Sigmund-Freud-StralSe 25, D-53127 Bonn, Germany
Received October 4, 1995 . Accepted November 14, 1995
Summary Backwash water from rapid sand filters of a treatment plant using surface water from small rivers for drinking water production was examined with the aim of determining the degree of their potential contamination with Giardia cysts and Cryptosporidium oocysts. Simultaneous investigations were carried out for both protozoans from November 1993 to February 1994. Water samples were concentrated by continuous flow centrifugation (11 backwash water samples) or by using polypropylene cartridge filters (12 raw water samples and 39 backwash water samples). Parasites were identified by the direct immunofluorescence assay. Ten out of 12 raw water samples tested were positive for Giardia (range: 21031100 L) and 8 out of 12 were positive for Cryptosporidium (range: 0.8-1091100 L). Eight of 11 backwash water samples collected by continuous flow centrifugation were positive for Giardia (range: 3-86/100 L) or Cryptosporidium (range: 1-69/100 L). Out of 39 samples collected using cartridge filters, 34 were positive for Giardia (range: 1.4-3741100 L) and 33 for Cryptosporidium (range: 0.8-2521100 L). Overall, Giardia, Cryptosporidium, or both were detected in 92 % of the backwash water samples. The results have clearly shown that backwash waters were contaminated with Giardia and Cryptosporidium and the supernatant returned to the raw water after the sedimentation process was not free from cysts and oocysts.
Introduction With respect to surface water supplies that are used for production of drinking water, two parasites, Giardia lamblia and Cryptosporidium parvum, have received considerable attention in the past few years in the USA, Canada and European countries, because they are the etiological agents of water-borne diseases. Cryptosporidium and Giardia are ubiquitous and they are responsible for cases of diarrhoea in humans and animals (1,2,3,4,5,9,12,15,18,19). The dormant stage of G.lamblia (G. duodenalis, Lamblia intestinalis) is the cyst. G.lamblia cysts measure lO!tm in length and 7!tm in width. They are capable of surviving for three to six months in the environment outside the host.
108
P. Karanis. D. Schoenen, and H. M. Seitz
Oocysts of Cryptosporidium are approximately 5 [tm large and represent the longlasting form of the parasite. Oocysts are able to survive in the environment for a twoyear period. Both parasites (Giardia and Cryptosporidium) are ingested by humans and animals with contaminated drinking water, impure food, or by direct contact. When infection has become established in the host, cysts or oocysts are released into the environment with faeces (3,12). Consequently, all surface water supplies used for drinking water are subject to parasitic contamination through excrements of humans and animals (e. g. muskrats, beavers, rodents, cattle and other animals). When the raw water is highly contaminated, the filtration barriers may suffer a break-through. To minimize this risk, the initial contamination level of the raw water should be as low as possible. For this reason it is necessary to know the sources of parasitic contamination. In 1989, an outhreak of cryptosporidiosis in Swindon and in Oxfordshire, England had occurred in a system that used rapid sand filtration (15). In the supernatant from the backwash water recycled to the raw water, 1000 oocysts/L were detected (Colbourne 1989, cited in 18]). The aim of the present study was to determine the extent of contamination with Giardia and Cryptosporidium in backwash waters from sand filters of a surface water treatment plant. Because backwash water is commonly returned to the raw water it may be responsihle for an increased risk of introducing the parasites into the drinking water. On the hasis of the findings, steps can be taken to protect the quality of the raw water used to produce drinking water.
Materials and Methods Water Treatment Plant and the Water Treatment Process
The water treatment plant studied is supplied with surface water drawn from three small rivers. The water is treated by sedimentation, direct filtration with phosphate precipitation, destabilisation, agglomeration and multi-media filtration. The raw water is flocculated using iron chlorine. and subsequently passed through multilayer filters. Water treated at this plant is not supplied to customers, but the treatment process can be used generally to produce drinking water from surface water supplies. The treated water is pumped to a lake lIsed as drinking water reservoir. The backwash waters resulting from backwashing the sand filters are pumped into special basins for sedimentation. Following sedimentation (La. 2-3 h). the supernatant is returned to the raw water reservoir, thus reentering the treatment process. The remaining sludge containing the iron hydroxide is finally deposited in
The procedure of water sample collection, particle separation, and detection of parasites is generally similar to the "EPA method" (6) which has been used in several studies in the USA and England (1, 8,11.13,16,17). Samples from the sedimentation basins were collected during the sedimentation phase of backwash water. Fifty-one samples (39 backwash water samples and 12 raw water samples) were collected using cartridge filters (model: Cuno Microwynd, cartridge filter, nominal porosity 1 ~lm). Another 11 backwash water samples were concentrated using continuous centrifugation (model: cepa-continuous flow centrifuge). The method of continuous flow centrifugation has been described previously by Renoth et al. (14). A generator and a wa-
Giardia and Cryptosporidium in Backwash Water
109
ter pump were used to pump 200 L into storage tanks for continuous flow centrifugation or to filter 500 L through the polypropylene cartridge filters. Flow rates during filtration through the cartridge filters were adjusted to S-16 Umin. Samples were collected on 02/091 16/23/30.11.1993 (trials I-V), 14/21.12.1993 (trials VI-VII), 04111/25.01.1994 (trials VIII-X), and 01/0S. 02.1994 (trials XI-XII) (Tables 1,2,3). The backwash water samples were taken in sedimentation basins at depths of 1, 2, 3, and 3.3 m.
Particle separation, purification, identification, and detection of parasites: For clarification of the samples and purification of Giardia and Cryptosporidium 1.5 M sucrose was used. The iron flakes present in the backwash water were dissolved using 1 M citric acid (20). Identification of cysts and oocysts in the purified pellet followed immediately after spreading the pellet on nitro cellulose membrane filters (pore size 1.2 11m, Sartorius) by the direct immunofluorescence test (IF test). Identification of Giardia and Cryptosporidium was based on the size and morphology of the cysts or oocysts. In the direct immunofluorescence test, Giardia cysts showed a bright apple-green fluorescence and were identified as oval in shape, measuring approximately 7-14 11m. Cryptosporidium oocysts appear as spheres between 4 and 6 11m in diameter often showing a suture line. Cysts and oocysts enumerated on the membrane filters by the IF test were calculated as number of cysts and oocysts per 100L of water sample. Positive controls were prepared with unfixed slides of Giardia cysts and Cryptosporidium oocysts supplied by Cellabs. In addition, positive controls were prepared with Giardia cysts from human sources (7) or with cysts produced in vitro (10) and stored in 10% formalin. Cryptosporidium oocysts for positive controls were harvested from fresh faeces of infected calves using a discontinuous sucrose gradient after sample filtration and sedimentation (14).
Table 1. Detection of Giardia cysts and Cryptosporidium oocysts in raw water using cartridge filters No. of samples
I II
III
IV V VI VII
VIII
IX X XI XII pos./exam. m.v. neg. = negative pos. = positive exam. = examined m. v. = median value
Number of Giardial 100 L 3.6 5.S
85
103 neg. 2 neg. 9.6 2.8 2.4 22.4
8.8 10/12 24.5
Number of Cryptosporidiuml
100 L 1.4 3 109 62 4 2 neg. neg.
0.8
neg. 3.2 neg.
8/12 23.2
110
P. Karanis, D. Schoenen, and H. M. Seitz
Table 2. Detection of Gia rdia cysts and Cryptosporidium oocysts in backwash water using continuous flow centrifugation No. of samples
I II
Depth (m)
2 3 3
III IV
V
VI VII VIII IX X XI XII pos./exam. m. v.
2 1
3 n.d. 3 1 2 3
Time (min)
Number of Giardial 100 L
Number of Cryptosporidiuml100 L
45 75 75 65 50 75 45 n.d. 75 95 105 60
7 3 86 27 3 neg. neg. n.d. 3 21 28 neg.
19.5 2 69 24 1 neg. 1 n.d. 4 neg. 2 neg.
8/11 22.3
8/11 15.3
n.d . =not determined neg. = negative m.v. = median value depth = depth at which the samples were taken in the sedimentation basins time = time at which samples were collected Results Concentrations of Giardia cysts and Cryptosporidium oocysts in raw and in backwash waters have been summarized in Tables 1-3. Recovery of parasites is influenced by the quality of the water sample and the efficiency of the collection procedure. Table 1 shows the numbers of Giardia and Cryptosporidium in the raw water samples per 100 L. Giardia was observed in ten out of twelve raw water samples (2-103 cystsl100 L) and Cryptosporidium in eight of twelve raw water samples (0.8-109 oocystsll 00 L). Tables 2-3 show the number of Giardia cysts and Cryptosporidium oocysts in backwash water per 100 L, the number of trials, the depth and the time at which samples were taken in the sedimentation basins and the method of sample collection. The number of Giardia cysts detected in backwash water ranged from 1.4 to 3741100 L (Tables 2,3) and the number of Cryptosporidium oocysts was 0.8 to 2521100 L (Tables 2, 3). Eight samples were negative fo r Giardia and nine were negative for Cryptosporidium.
Discussion The most important observations from this study are firstly the high detection rate of Giardia and Cryptosporidium in the backwash waters and secondly that the supernatant returned to the raw water was not free from cysts and oocysts after the sedi-
1.9
neg.
2.3
11/13 6.4
11.7
6.9
11 /13 21.2
0.8
2.4
56
5.6
21
5
7.6
Abbrevations see ta ble 1 and 2.
pos./exam. m. v.
VIII IX X XI XII
5.2
5.2
145
14 5
105
145
130
105
65
145
0.8
1.6
12/13 55.1
92.8
9.8
374
5.2
16.4
4 1.4
neg.
2.8
8.2
130
neg.
1.8
26.6
19.8
IV V VI VII
105
69.4
10/13 22.2
2.9
1.6
125
neg.
2.4
27 .6
neg.
1.8
n.n.
30
28
1.2
37.4
22.4
95.6
III
1.6
1.4 2.2
55 85
Giardia
Cryp tosporidium
Depth = 2 m
mm
Time
120
neg.
0.8
Cryptosporidium
neg.
2.4
Giardia
Depth = 1 m
II
No . of sam ples
Paras ites/ 100 L
110
110
135
120
90
105
65
120
10.4
8110 24.6
20
22.4
n.d.
9/10 48.4
1.8
22.4
n.d .
1.8
n.d.
252
neg. n.d.
neg.
3.4
2. 8
neg.
100
42.2
109.4
93 .2
1.4 1.2
42.2
Time
68
175
90
3/3 17.4
n.d. n.d. 3/3 36.9
n.d.
n.d . n.d.
n.d. n.d.
n.d.
n.d.
n.d.
n. d.
n. d.
90
n.d.
n.d . n.d. n.d.
150
1.2 n. d.
110 175
6
60 45
Time
CryptomIn sporidium
45
n. d.
9.6
18
Giardia
Depth = 3.3 m
83.2 75
195
145
120
Cryptomm spo ridium
2 .2
3.8
Giardia
Depth = 3 m
95
150
11 5
85
mIn
Time
la ble 3. Detection ot Giardia cysts and Cryptosporidium oocysts in backwash wa ter using cartridge fil ters
....
"
~
:r
"'
,..,"' :0;:E
I:l:l
it· :! S·
I:l..
:::!.
0
e" ~
0-
'< ~
(J
...
;:3
"'0-
~
t..,
()
112
P. Karanis, D. Schoen en, and H. M. Seitz
mentation process. Overall, Giardia, Cryptosporidium, or both were detected in 92 % of the backwash waters. The detection rate in raw water samples was 91.7%. A negative result does not mean that these parasites were not present in the water. We suppose that the level of contamination with Giardia and Cryptosporidium in these samples was lower than the detection limit for cysts and oocysts (see below). When backwash waters are channelled into sedimentation basins, it can be assumed that all Giardia cysts and Cryptosporidium oocysts enter the backwash sediment with the sedimenting flakes. This is only partially valid for Giardia and Cryptosporidium. In our study, both parasites (Giardia and Cryptosporidium) were regularly detected in backwash waters. 84% of the backwash water samples were Giardia-positive and 82 %, Cryptosporidium-positive. The mean value for Giardia was 32.9 cystsl100 Land the mean value for Cryptosporidium was 22 oocysts/100L (collection of the samples with cartridge filters and with continuous flow centrifugation at different depths of the sedimentation basins; all positive samples irrespective of the depth at which samples were taken). Giardia cysts level was 1.34 times higher in the backwash water samples than in the raw water whereas the level for Cryptosporidium oocysts in backwash water was the same as in raw water (23.2 and 22 oocysts respectively, 100 L). It is not surprising that cysts or oocysts are observed in different concentrations in the backwash water samples because the samples were taken at different depths of the basins and at different times after sedimentation. In the USA (Rose 1989, cited in 1), 40% of the backwash waters from 15 water treatment plants were positive for oocysts. The mean value of oocysts was 2100/1 00 L and maximum number of oocysts 10000/100 L. In another study in the USA, Le Cheval/ier et al. (8) detected Giardia in only 14 of 34 (41.2 %) backwash samples and Cryptosporidium in 22 of 34 (64.7%) backwash samples. The lower detection rate in backwash water compared to the raw water in the study of Le Cheval/ier et al. was due to the differences in detection sensitivity. The average limit of detection for backwash samples was more than 6 times higher than for raw water samples. In our study, there were no differences in the detection rate in backwash water and raw water samples (92% and 91.7%, respectively were positive of Giardia or Cryptosporidium or both). It is reasonable to assume that the differences in the numbers of parasites detected in water samples depend on the varying distribution of the parasites in the environment, the number of samples examined and the detection limit of the sampling method. Recovery rates of parasites are different for both methods. Recovery with continuous flow centrifugation was 16% for Cryptosporidium oocysts (14) and up to 47% (Karanis, unpublished data) for Giardia cysts in tap water samples. Recovery rates for Giardia and Cryptosporidium using cartridge filters vary, depending on the water quality (11, 13, 16, 17). Le Chevallier et al. (8) reported high recovery rateS for Giardia (48-69%) and Cryptosporidium (25-42%) in spiked water samples using filterite filters. Generally, backwash water samples are difficult to examine because they are very dirty and contain large amounts of iron floes. It becomes apparent from our investigations that a considerable proportion of cysts and oocysts is still present in the supernatant even after 3 h of sedimentation. This supernatant contains non-settling material, including free Giardia cysts and Cryptosporidium oocysts. Non-settling cysts and oocysts are returned to the raw water, thus adding to the natural contamination level of the raw water. The reentry of Giardia and Cryptosporidium into the raw water could be a great problem for water treatment plants. Experts on the Badenoch committee recommend the treatment of supernatant,
Giardia and Cryptosporidium in Backwash Water
113
if recycled, by specialised filtration techniques or disinfection. Whereas chlorine is an ineffective disinfectant for oocysts, ozone or chlorine dioxide may be suitable (1). Because no data are available on backwash water treatment, especially with respect to the elimination of Giardia cysts and Cryptosporidium oocysts by sludge disposal, further research is needed to determine the distribution of cysts and oocysts in the sludge after conventional processing and to define the sources of contamination of raw water by both parasites. Acknowledgement. We thank the DVGW (Deutscher Verband fur Gas- und Wasserwirtschaft), EschbornlGermany, for supporting this study.
References 1. Badenoch, ].: Cryptosporidium in water supplies. Report of the Group of Experts U. Badenoch, ed.), pp.1-230. London, HMSO (1990) 2. Craun, G. F.: Waterborne giardiasis. In: Giardiasis (E. A. Meyer, ed.), pp.267-293. Elsevier, Amsterdam. New York, Oxford (1990) 3. Current, W. L.: Cryptosporidium: Its biology and potential for the environmental transmission. Crit. Rev. Environ. Control 17 (1986) 21-51 4. Hayes, E. D., T. D. Matte, T. R. O'Brien, T. W. McKinley, G. S. Logsdon, ]. B. Rose, B. p. L. Ungar, D. M. Word, P. F. Pinsky, M. L. Cummings, M. A. Wilson, E. G. Long, E. S. Hurwitz, and D. D. Juranek: Large community outbreak of cryptosporidiosis due to contamination of a filtered public water supply. N. Eng!. J. Med. 320 (1989) 1372-1376 5. Herwaldt, L. B., G. F. Craun, L. S. Stokes, and D. D. Juranek: Outbreaks of waterborne disease in the United States: 1989-1990. J. AWWA 84 (1992) 129-135 6. Jakubowski, W. and T. H. Ericksen: Methods for detection of Giardia cysts in water supplies. In: Waterborne Transmission of Giardiasis (w. Jakubowski and]. C. Hoff, eds.), pp.193-210. EPA-60019-79-001, US Environmental Protection Agency, Cincinnati, Ohio (1979) 7. Karanis, P., W. A. Maier, D. Schoenen, and H. M. Seitz: Studies on the in vitro cultivation of Giardia lamblia isolates from humans faecal specimens. Zbl. Bakt. 325, pp. 46, Abstracts of the 15th Meeting of the Deutsche Gesellschaft fur Parasitologie, MarchI April 30.03.-03.04.1992 Berlin, Germany (1993) 8. Le Cheval/ier, M. w., W. D. Norton, R. G. Lee, and]. B. Rose: Detection and treatment of Giardia and Cryptosporidium in water supplies. AWWA Research Foundation, 6666 West Qincy Avenue Denver, CO 80235 (1991) 9. MacKenzie, R. w., N.]. Hoxie, M. E. Proctor, M. S. Gradus, K. A. Blair, D. E. Peterson, ].]. Kazmierczak, D. G. Addis, K. R. Fox, J. B. Rose, and]. P. Davis: A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public wa-
ter supply. N. Eng!.
J. Med. (1994) 161-167
10. Monar, G., P. Karanis, W. A. Maier, D. Schoenen, and H. M. Seitz: Intraspecific variation of the in vitro encystation of Giardia lamblia isolates from humans and animals.
IX International Congress of Protozoology, Abstracts, pp. 87, July 25-31 Berlin, Germany (1993) 11. Madore, M.]., ]. Rose, C. P. Gerba, M.]. Arrowood, and C. R. Sterling: Occurrence of Cryptosporidium oocysts in sewage effluents and selected surface waters. J. Parasitol. 73 (1987) 702-705 12. Meyer, E. A.: Giardia as an Organism. In: Giardia: From Molecules to Disease (R. C. A. Thompson,]. A. Reynoldson, and A.]. Lymbery, eds.),pp. 3-13. Cab International (1994) 13. Musial, C. E., M.J. Arrowood, C. R. Sterling, and C. P. Gerba: Detection of Cryptosporidium in water by using polypropylene cartridge filters. Appl. Environ. Microbiol. 53 (1987) 672-676 8
Zbl. Bakt. 284/1
114
P. Karanis, D. Schoenen, and H. M. Seitz
14. Renoth, S., P. Karanis, D. Schoenen, and H. M. Seitz: Recovery efficiency of Cryptosporidium from water with a crossflow system and continuous flow centrifugation: a comparison study. Zbl. Bakt. (in press) 15. Richardson, A.j., R. A. Frankenberg, A. C. Buck, ]. B. Selkon, j. S. Colbourne, j. W. Parsons, and R. T. Mayon- White: An outbreak of waterborne cryptosporidiosis in Swindon and Oxfordshire. Epidemiol. Infect. 107 (1991) 485-495 16. Rose, j. B., A. Cirrino, M. S. Madore, C. P. Gerba, C. R. Sterling, and M.]. Arrowood: Detection of Cryptosporidium from wastewater and freshwater environments. Water Sci. Technol. 18 (1986) 233-239 17. Rose, j. B., D. Kayed, M. S. Madore, C. P. Gerba, M.j. Arrowood, C. R. Sterling, and ]. L. Riggs: Methods for the recovery of Giardia and Cryptosporidium in environmental waters and their comparative occurrence. In: Advances in Giardia Research (P. M. Wallis and B. R. Hammond, eds.), Calgary: University of Calgary Press (1988) 18. Smith, H. v., W. ,. Patterson, R. Hardie, L. A. Greene, C. Benton, W. Tulloch, R. A. Gilmour, R. W. A. C;irdwood,]. c. M. Sharp, and G. I. Forbes: An outbreak of waterborne cryptosporidiosis caused by posttreatment contamination. Epidem. Inf. 103 (1989) 703-715 19. Wallis, P. M.: Abiotic Transmission - Is water really significant? In: Giardia: From Molecules to Disease (R. C. A. Thompson, j. A. Reynoldson, and A. j. Lymbery, eds.), pp. 125-126. Cab International (1994) 20. Walter, R. and S. T. Riidiger: Ein zweistufiges Verfahren zur Virusanreicherung aus Losungen mit geringem Virustiter (z. B. Trinkwasser). ]. Hyg. Epidemiol. Microbiol. Immunol. 25 (196 I) 71-81 Dr. rer. nat. Panagiotis Karanis, Institute for Medical Parasitology, University of Bonn, Sigmund-Freud-StraiSe 25, D-53127 Bonn, Germany
Giardia and Cryptosporidium in Backwash Water from Rapid Sand Filters Used for Drinking Water Production PANAGIOTIS KARANISl, DIRK SCHOENEN 2 , and HANNS MARTIN SEITZ l 1 Institute 2
for Medical Parasitology Hygiene-Institute, University of Bonn, Sigmund-Freud-StralSe 25, D-53127 Bonn, Germany
Received October 4, 1995 . Accepted November 14, 1995
Summary Backwash water from rapid sand filters of a treatment plant using surface water from small rivers for drinking water production was examined with the aim of determining the degree of their potential contamination with Giardia cysts and Cryptosporidium oocysts. Simultaneous investigations were carried out for both protozoans from November 1993 to February 1994. Water samples were concentrated by continuous flow centrifugation (11 backwash water samples) or by using polypropylene cartridge filters (12 raw water samples and 39 backwash water samples). Parasites were identified by the direct immunofluorescence assay. Ten out of 12 raw water samples tested were positive for Giardia (range: 21031100 L) and 8 out of 12 were positive for Cryptosporidium (range: 0.8-1091100 L). Eight of 11 backwash water samples collected by continuous flow centrifugation were positive for Giardia (range: 3-86/100 L) or Cryptosporidium (range: 1-69/100 L). Out of 39 samples collected using cartridge filters, 34 were positive for Giardia (range: 1.4-3741100 L) and 33 for Cryptosporidium (range: 0.8-2521100 L). Overall, Giardia, Cryptosporidium, or both were detected in 92 % of the backwash water samples. The results have clearly shown that backwash waters were contaminated with Giardia and Cryptosporidium and the supernatant returned to the raw water after the sedimentation process was not free from cysts and oocysts.
Introduction With respect to surface water supplies that are used for production of drinking water, two parasites, Giardia lamblia and Cryptosporidium parvum, have received considerable attention in the past few years in the USA, Canada and European countries, because they are the etiological agents of water-borne diseases. Cryptosporidium and Giardia are ubiquitous and they are responsible for cases of diarrhoea in humans and animals (1,2,3,4,5,9,12,15,18,19). The dormant stage of G.lamblia (G. duodenalis, Lamblia intestinalis) is the cyst. G.lamblia cysts measure lO!tm in length and 7!tm in width. They are capable of surviving for three to six months in the environment outside the host.
108
P. Karanis. D. Schoenen, and H. M. Seitz
Oocysts of Cryptosporidium are approximately 5 [tm large and represent the longlasting form of the parasite. Oocysts are able to survive in the environment for a twoyear period. Both parasites (Giardia and Cryptosporidium) are ingested by humans and animals with contaminated drinking water, impure food, or by direct contact. When infection has become established in the host, cysts or oocysts are released into the environment with faeces (3,12). Consequently, all surface water supplies used for drinking water are subject to parasitic contamination through excrements of humans and animals (e. g. muskrats, beavers, rodents, cattle and other animals). When the raw water is highly contaminated, the filtration barriers may suffer a break-through. To minimize this risk, the initial contamination level of the raw water should be as low as possible. For this reason it is necessary to know the sources of parasitic contamination. In 1989, an outhreak of cryptosporidiosis in Swindon and in Oxfordshire, England had occurred in a system that used rapid sand filtration (15). In the supernatant from the backwash water recycled to the raw water, 1000 oocysts/L were detected (Colbourne 1989, cited in 18]). The aim of the present study was to determine the extent of contamination with Giardia and Cryptosporidium in backwash waters from sand filters of a surface water treatment plant. Because backwash water is commonly returned to the raw water it may be responsihle for an increased risk of introducing the parasites into the drinking water. On the hasis of the findings, steps can be taken to protect the quality of the raw water used to produce drinking water.
Materials and Methods Water Treatment Plant and the Water Treatment Process
The water treatment plant studied is supplied with surface water drawn from three small rivers. The water is treated by sedimentation, direct filtration with phosphate precipitation, destabilisation, agglomeration and multi-media filtration. The raw water is flocculated using iron chlorine. and subsequently passed through multilayer filters. Water treated at this plant is not supplied to customers, but the treatment process can be used generally to produce drinking water from surface water supplies. The treated water is pumped to a lake lIsed as drinking water reservoir. The backwash waters resulting from backwashing the sand filters are pumped into special basins for sedimentation. Following sedimentation (La. 2-3 h). the supernatant is returned to the raw water reservoir, thus reentering the treatment process. The remaining sludge containing the iron hydroxide is finally deposited in
The procedure of water sample collection, particle separation, and detection of parasites is generally similar to the "EPA method" (6) which has been used in several studies in the USA and England (1, 8,11.13,16,17). Samples from the sedimentation basins were collected during the sedimentation phase of backwash water. Fifty-one samples (39 backwash water samples and 12 raw water samples) were collected using cartridge filters (model: Cuno Microwynd, cartridge filter, nominal porosity 1 ~lm). Another 11 backwash water samples were concentrated using continuous centrifugation (model: cepa-continuous flow centrifuge). The method of continuous flow centrifugation has been described previously by Renoth et al. (14). A generator and a wa-
Giardia and Cryptosporidium in Backwash Water
109
ter pump were used to pump 200 L into storage tanks for continuous flow centrifugation or to filter 500 L through the polypropylene cartridge filters. Flow rates during filtration through the cartridge filters were adjusted to S-16 Umin. Samples were collected on 02/091 16/23/30.11.1993 (trials I-V), 14/21.12.1993 (trials VI-VII), 04111/25.01.1994 (trials VIII-X), and 01/0S. 02.1994 (trials XI-XII) (Tables 1,2,3). The backwash water samples were taken in sedimentation basins at depths of 1, 2, 3, and 3.3 m.
Particle separation, purification, identification, and detection of parasites: For clarification of the samples and purification of Giardia and Cryptosporidium 1.5 M sucrose was used. The iron flakes present in the backwash water were dissolved using 1 M citric acid (20). Identification of cysts and oocysts in the purified pellet followed immediately after spreading the pellet on nitro cellulose membrane filters (pore size 1.2 11m, Sartorius) by the direct immunofluorescence test (IF test). Identification of Giardia and Cryptosporidium was based on the size and morphology of the cysts or oocysts. In the direct immunofluorescence test, Giardia cysts showed a bright apple-green fluorescence and were identified as oval in shape, measuring approximately 7-14 11m. Cryptosporidium oocysts appear as spheres between 4 and 6 11m in diameter often showing a suture line. Cysts and oocysts enumerated on the membrane filters by the IF test were calculated as number of cysts and oocysts per 100L of water sample. Positive controls were prepared with unfixed slides of Giardia cysts and Cryptosporidium oocysts supplied by Cellabs. In addition, positive controls were prepared with Giardia cysts from human sources (7) or with cysts produced in vitro (10) and stored in 10% formalin. Cryptosporidium oocysts for positive controls were harvested from fresh faeces of infected calves using a discontinuous sucrose gradient after sample filtration and sedimentation (14).
Table 1. Detection of Giardia cysts and Cryptosporidium oocysts in raw water using cartridge filters No. of samples
I II
III
IV V VI VII
VIII
IX X XI XII pos./exam. m.v. neg. = negative pos. = positive exam. = examined m. v. = median value
Number of Giardial 100 L 3.6 5.S
85
103 neg. 2 neg. 9.6 2.8 2.4 22.4
8.8 10/12 24.5
Number of Cryptosporidiuml
100 L 1.4 3 109 62 4 2 neg. neg.
0.8
neg. 3.2 neg.
8/12 23.2
110
P. Karanis, D. Schoenen, and H. M. Seitz
Table 2. Detection of Gia rdia cysts and Cryptosporidium oocysts in backwash water using continuous flow centrifugation No. of samples
I II
Depth (m)
2 3 3
III IV
V
VI VII VIII IX X XI XII pos./exam. m. v.
2 1
3 n.d. 3 1 2 3
Time (min)
Number of Giardial 100 L
Number of Cryptosporidiuml100 L
45 75 75 65 50 75 45 n.d. 75 95 105 60
7 3 86 27 3 neg. neg. n.d. 3 21 28 neg.
19.5 2 69 24 1 neg. 1 n.d. 4 neg. 2 neg.
8/11 22.3
8/11 15.3
n.d . =not determined neg. = negative m.v. = median value depth = depth at which the samples were taken in the sedimentation basins time = time at which samples were collected Results Concentrations of Giardia cysts and Cryptosporidium oocysts in raw and in backwash waters have been summarized in Tables 1-3. Recovery of parasites is influenced by the quality of the water sample and the efficiency of the collection procedure. Table 1 shows the numbers of Giardia and Cryptosporidium in the raw water samples per 100 L. Giardia was observed in ten out of twelve raw water samples (2-103 cystsl100 L) and Cryptosporidium in eight of twelve raw water samples (0.8-109 oocystsll 00 L). Tables 2-3 show the number of Giardia cysts and Cryptosporidium oocysts in backwash water per 100 L, the number of trials, the depth and the time at which samples were taken in the sedimentation basins and the method of sample collection. The number of Giardia cysts detected in backwash water ranged from 1.4 to 3741100 L (Tables 2,3) and the number of Cryptosporidium oocysts was 0.8 to 2521100 L (Tables 2, 3). Eight samples were negative fo r Giardia and nine were negative for Cryptosporidium.
Discussion The most important observations from this study are firstly the high detection rate of Giardia and Cryptosporidium in the backwash waters and secondly that the supernatant returned to the raw water was not free from cysts and oocysts after the sedi-
1.9
neg.
2.3
11/13 6.4
11.7
6.9
11 /13 21.2
0.8
2.4
56
5.6
21
5
7.6
Abbrevations see ta ble 1 and 2.
pos./exam. m. v.
VIII IX X XI XII
5.2
5.2
145
14 5
105
145
130
105
65
145
0.8
1.6
12/13 55.1
92.8
9.8
374
5.2
16.4
4 1.4
neg.
2.8
8.2
130
neg.
1.8
26.6
19.8
IV V VI VII
105
69.4
10/13 22.2
2.9
1.6
125
neg.
2.4
27 .6
neg.
1.8
n.n.
30
28
1.2
37.4
22.4
95.6
III
1.6
1.4 2.2
55 85
Giardia
Cryp tosporidium
Depth = 2 m
mm
Time
120
neg.
0.8
Cryptosporidium
neg.
2.4
Giardia
Depth = 1 m
II
No . of sam ples
Paras ites/ 100 L
110
110
135
120
90
105
65
120
10.4
8110 24.6
20
22.4
n.d.
9/10 48.4
1.8
22.4
n.d .
1.8
n.d.
252
neg. n.d.
neg.
3.4
2. 8
neg.
100
42.2
109.4
93 .2
1.4 1.2
42.2
Time
68
175
90
3/3 17.4
n.d. n.d. 3/3 36.9
n.d.
n.d . n.d.
n.d. n.d.
n.d.
n.d.
n.d.
n. d.
n. d.
90
n.d.
n.d . n.d. n.d.
150
1.2 n. d.
110 175
6
60 45
Time
CryptomIn sporidium
45
n. d.
9.6
18
Giardia
Depth = 3.3 m
83.2 75
195
145
120
Cryptomm spo ridium
2 .2
3.8
Giardia
Depth = 3 m
95
150
11 5
85
mIn
Time
la ble 3. Detection ot Giardia cysts and Cryptosporidium oocysts in backwash wa ter using cartridge fil ters
....
"
~
:r
"'
,..,"' :0;:E
I:l:l
it· :! S·
I:l..
:::!.
0
e" ~
0-
'< ~
(J
...
;:3
"'0-
~
t..,
()
112
P. Karanis, D. Schoen en, and H. M. Seitz
mentation process. Overall, Giardia, Cryptosporidium, or both were detected in 92 % of the backwash waters. The detection rate in raw water samples was 91.7%. A negative result does not mean that these parasites were not present in the water. We suppose that the level of contamination with Giardia and Cryptosporidium in these samples was lower than the detection limit for cysts and oocysts (see below). When backwash waters are channelled into sedimentation basins, it can be assumed that all Giardia cysts and Cryptosporidium oocysts enter the backwash sediment with the sedimenting flakes. This is only partially valid for Giardia and Cryptosporidium. In our study, both parasites (Giardia and Cryptosporidium) were regularly detected in backwash waters. 84% of the backwash water samples were Giardia-positive and 82 %, Cryptosporidium-positive. The mean value for Giardia was 32.9 cystsl100 Land the mean value for Cryptosporidium was 22 oocysts/100L (collection of the samples with cartridge filters and with continuous flow centrifugation at different depths of the sedimentation basins; all positive samples irrespective of the depth at which samples were taken). Giardia cysts level was 1.34 times higher in the backwash water samples than in the raw water whereas the level for Cryptosporidium oocysts in backwash water was the same as in raw water (23.2 and 22 oocysts respectively, 100 L). It is not surprising that cysts or oocysts are observed in different concentrations in the backwash water samples because the samples were taken at different depths of the basins and at different times after sedimentation. In the USA (Rose 1989, cited in 1), 40% of the backwash waters from 15 water treatment plants were positive for oocysts. The mean value of oocysts was 2100/1 00 L and maximum number of oocysts 10000/100 L. In another study in the USA, Le Cheval/ier et al. (8) detected Giardia in only 14 of 34 (41.2 %) backwash samples and Cryptosporidium in 22 of 34 (64.7%) backwash samples. The lower detection rate in backwash water compared to the raw water in the study of Le Cheval/ier et al. was due to the differences in detection sensitivity. The average limit of detection for backwash samples was more than 6 times higher than for raw water samples. In our study, there were no differences in the detection rate in backwash water and raw water samples (92% and 91.7%, respectively were positive of Giardia or Cryptosporidium or both). It is reasonable to assume that the differences in the numbers of parasites detected in water samples depend on the varying distribution of the parasites in the environment, the number of samples examined and the detection limit of the sampling method. Recovery rates of parasites are different for both methods. Recovery with continuous flow centrifugation was 16% for Cryptosporidium oocysts (14) and up to 47% (Karanis, unpublished data) for Giardia cysts in tap water samples. Recovery rates for Giardia and Cryptosporidium using cartridge filters vary, depending on the water quality (11, 13, 16, 17). Le Chevallier et al. (8) reported high recovery rateS for Giardia (48-69%) and Cryptosporidium (25-42%) in spiked water samples using filterite filters. Generally, backwash water samples are difficult to examine because they are very dirty and contain large amounts of iron floes. It becomes apparent from our investigations that a considerable proportion of cysts and oocysts is still present in the supernatant even after 3 h of sedimentation. This supernatant contains non-settling material, including free Giardia cysts and Cryptosporidium oocysts. Non-settling cysts and oocysts are returned to the raw water, thus adding to the natural contamination level of the raw water. The reentry of Giardia and Cryptosporidium into the raw water could be a great problem for water treatment plants. Experts on the Badenoch committee recommend the treatment of supernatant,
Giardia and Cryptosporidium in Backwash Water
113
if recycled, by specialised filtration techniques or disinfection. Whereas chlorine is an ineffective disinfectant for oocysts, ozone or chlorine dioxide may be suitable (1). Because no data are available on backwash water treatment, especially with respect to the elimination of Giardia cysts and Cryptosporidium oocysts by sludge disposal, further research is needed to determine the distribution of cysts and oocysts in the sludge after conventional processing and to define the sources of contamination of raw water by both parasites. Acknowledgement. We thank the DVGW (Deutscher Verband fur Gas- und Wasserwirtschaft), EschbornlGermany, for supporting this study.
References 1. Badenoch, ].: Cryptosporidium in water supplies. Report of the Group of Experts U. Badenoch, ed.), pp.1-230. London, HMSO (1990) 2. Craun, G. F.: Waterborne giardiasis. In: Giardiasis (E. A. Meyer, ed.), pp.267-293. Elsevier, Amsterdam. New York, Oxford (1990) 3. Current, W. L.: Cryptosporidium: Its biology and potential for the environmental transmission. Crit. Rev. Environ. Control 17 (1986) 21-51 4. Hayes, E. D., T. D. Matte, T. R. O'Brien, T. W. McKinley, G. S. Logsdon, ]. B. Rose, B. p. L. Ungar, D. M. Word, P. F. Pinsky, M. L. Cummings, M. A. Wilson, E. G. Long, E. S. Hurwitz, and D. D. Juranek: Large community outbreak of cryptosporidiosis due to contamination of a filtered public water supply. N. Eng!. J. Med. 320 (1989) 1372-1376 5. Herwaldt, L. B., G. F. Craun, L. S. Stokes, and D. D. Juranek: Outbreaks of waterborne disease in the United States: 1989-1990. J. AWWA 84 (1992) 129-135 6. Jakubowski, W. and T. H. Ericksen: Methods for detection of Giardia cysts in water supplies. In: Waterborne Transmission of Giardiasis (w. Jakubowski and]. C. Hoff, eds.), pp.193-210. EPA-60019-79-001, US Environmental Protection Agency, Cincinnati, Ohio (1979) 7. Karanis, P., W. A. Maier, D. Schoenen, and H. M. Seitz: Studies on the in vitro cultivation of Giardia lamblia isolates from humans faecal specimens. Zbl. Bakt. 325, pp. 46, Abstracts of the 15th Meeting of the Deutsche Gesellschaft fur Parasitologie, MarchI April 30.03.-03.04.1992 Berlin, Germany (1993) 8. Le Cheval/ier, M. w., W. D. Norton, R. G. Lee, and]. B. Rose: Detection and treatment of Giardia and Cryptosporidium in water supplies. AWWA Research Foundation, 6666 West Qincy Avenue Denver, CO 80235 (1991) 9. MacKenzie, R. w., N.]. Hoxie, M. E. Proctor, M. S. Gradus, K. A. Blair, D. E. Peterson, ].]. Kazmierczak, D. G. Addis, K. R. Fox, J. B. Rose, and]. P. Davis: A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public wa-
ter supply. N. Eng!.
J. Med. (1994) 161-167
10. Monar, G., P. Karanis, W. A. Maier, D. Schoenen, and H. M. Seitz: Intraspecific variation of the in vitro encystation of Giardia lamblia isolates from humans and animals.
IX International Congress of Protozoology, Abstracts, pp. 87, July 25-31 Berlin, Germany (1993) 11. Madore, M.]., ]. Rose, C. P. Gerba, M.]. Arrowood, and C. R. Sterling: Occurrence of Cryptosporidium oocysts in sewage effluents and selected surface waters. J. Parasitol. 73 (1987) 702-705 12. Meyer, E. A.: Giardia as an Organism. In: Giardia: From Molecules to Disease (R. C. A. Thompson,]. A. Reynoldson, and A.]. Lymbery, eds.),pp. 3-13. Cab International (1994) 13. Musial, C. E., M.J. Arrowood, C. R. Sterling, and C. P. Gerba: Detection of Cryptosporidium in water by using polypropylene cartridge filters. Appl. Environ. Microbiol. 53 (1987) 672-676 8
Zbl. Bakt. 284/1
114
P. Karanis, D. Schoenen, and H. M. Seitz
14. Renoth, S., P. Karanis, D. Schoenen, and H. M. Seitz: Recovery efficiency of Cryptosporidium from water with a crossflow system and continuous flow centrifugation: a comparison study. Zbl. Bakt. (in press) 15. Richardson, A.j., R. A. Frankenberg, A. C. Buck, ]. B. Selkon, j. S. Colbourne, j. W. Parsons, and R. T. Mayon- White: An outbreak of waterborne cryptosporidiosis in Swindon and Oxfordshire. Epidemiol. Infect. 107 (1991) 485-495 16. Rose, j. B., A. Cirrino, M. S. Madore, C. P. Gerba, C. R. Sterling, and M.]. Arrowood: Detection of Cryptosporidium from wastewater and freshwater environments. Water Sci. Technol. 18 (1986) 233-239 17. Rose, j. B., D. Kayed, M. S. Madore, C. P. Gerba, M.j. Arrowood, C. R. Sterling, and ]. L. Riggs: Methods for the recovery of Giardia and Cryptosporidium in environmental waters and their comparative occurrence. In: Advances in Giardia Research (P. M. Wallis and B. R. Hammond, eds.), Calgary: University of Calgary Press (1988) 18. Smith, H. v., W. ,. Patterson, R. Hardie, L. A. Greene, C. Benton, W. Tulloch, R. A. Gilmour, R. W. A. C;irdwood,]. c. M. Sharp, and G. I. Forbes: An outbreak of waterborne cryptosporidiosis caused by posttreatment contamination. Epidem. Inf. 103 (1989) 703-715 19. Wallis, P. M.: Abiotic Transmission - Is water really significant? In: Giardia: From Molecules to Disease (R. C. A. Thompson, j. A. Reynoldson, and A. j. Lymbery, eds.), pp. 125-126. Cab International (1994) 20. Walter, R. and S. T. Riidiger: Ein zweistufiges Verfahren zur Virusanreicherung aus Losungen mit geringem Virustiter (z. B. Trinkwasser). ]. Hyg. Epidemiol. Microbiol. Immunol. 25 (196 I) 71-81 Dr. rer. nat. Panagiotis Karanis, Institute for Medical Parasitology, University of Bonn, Sigmund-Freud-StraiSe 25, D-53127 Bonn, Germany