Nội dung

Acta vet. scand. 2002, 43, 203-210. Persistence of Trichinella spiralis in Rat Carcasses Experimentally Mixed in Different Feed By L. Oivanen1,3, T. Mikkonen1, L. Haltia2, H. Karhula1, H. Saloniemi2 and A. Sukura1 1Department of Basic Veterinary Sciences, and 2Department of Clinical Veterinary Sciences, Faculty of Veterinary Medicine, University of Helsinki, and 3National Veterinary and Food Research Institute, Helsinki, Finland. Oivanen L, Mikkonen T, Haltia L, Karhula H, Saloniemi H, Sukura A: Persistence of Trichinella spiralis in rat carcasses experimentally mixed in different feed. Acta vet. scand. 2002, 43, 203-210. – Trichinella spiralis infected rat carcasses were incubated for 6 weeks in several animal feeds to assess how long Trichinella can present a risk for an outbreak in contaminated feeds. In groups of 6, 24 infected target rats were placed in silage, grained barley, propionic acid-preserved feed, and also into simulated pasture conditions. Test environments were sampled after one-, 2-, 4-, and 6-week-incubations. Trichinella larvae were recovered by digestion, and their infectivity was evaluated in rats. A two-week incubation reduced the number of recovered larvae, but still after 6 weeks low numbers were isolated from all feeds except from the experimental group simulating pasture conditions. After 2 weeks storage, the larvae were infective in all storage environments. However, up to 4 weeks, they survived only in the propionic acid-fermented feed and there in small numbers with reduced reproductive capability. This indicates the possibility of farm animals to get infection from rats or other infected material being hazardously mixed with hay or other feed. If silage is stored for at least one month before use, however, the risk from this forage appears to be minimized. Trichinellosis; domestic cycle; exposure assessment; feed hazard; animal hygiene; zoonosis; disease control; rat; feed contamination; carrion. Introduction Rodents have been suspected or proven to be a source of trichinellosis for pigs and farmed wild boars (Schad et al. 1987, Smith & Kay 1987, Leiby et al. 1990, Gamble et al. 1999, Oivanen et al. 2000). Pork and other pig meat products as well as carnivore/omnivore game are well recognized potential sources for human trichinellosis. However, herbivores can also transmit the infection to human beings. Since 1975, at least 3 300 people have fallen ill in 13 outbreaks due to horse meat consumption in France and Italy (Boireau et al. 2000, Touratier 2001). In addition, China has reported human outbreaks, not only due to pork but also to mutton or beef consumption (Murrell 1994, Wang & Cui 2001). All these outbreaks have raised the question of natural transmission of Trichinella to horses, or other herbivores. Two hypothesis have been proposed: grazing in pastures contaminated with infected rodent carcasses or feeding with infected flesh from pigs or wildlife (Pozio et al. 2001). In Finland, sylvatic trichinellosis is highly prevalent, and domestic trichinellosis in pigs and farmed wild boars has been repeatedly reported in meat inspection during recent decades (Oivanen et al. 2000, Sukura et al. 2001, Oivanen et al. 2002). Moreover, in Finland, Trichinella infection is commonly found in rats from dumps (Mikkonen 1998, Mikkonen et al. Acta vet. scand. vol. 43 no. 4, 2002 204 L. Oivanen et al. unpublished). These rats have been infected almost exclusively with Trichinella spiralis (Oivanen et al. 2002). Among Trichinella genotypes, T. spiralis has been the one most often involved in human outbreaks (Capó & Despommier 1996). A real risk exists for forage to become contaminated during handling and processing at farms. Fitzgerald & Prakasam (1978) tested T. spiralis survival in sewage sludge. The encysted larvae survived no longer than 96 h. Thus, such sludge seems not to offer much of a risk for pasture and field contamination. Von Köller et al. (2001) demonstrated that under laboratory condition, in steady room temperature, some species of Trichinella can survive several weeks in decaying meat and that host species and the age of infection influence on this survival. On the contrary, no information is available of longevity of Trichinella under natural conditions in different feeds or in the pasture. Data on parasite survival in feeds is needed for proper risk assessment for herbivore or omnivore domestic animals. In Finland, fresh-cut hay is used for indoor feeding of cows and horses also in the summer months if the animals are not in pastures. Silage is the common base for cow feed in Finland but is also recommended for sows (Suomi 1999) and used for horses. In swine farming, grain is commonly milled at the farm and mixed with protein concentrates. Alternative fermentation methods for grain have become more popular, such as wet preservation with acidic additives. To obtain basic data for exposure assessment, we ran an experiment on survival of T. spiralis in contaminated feeds. The experimental feeds were silage, grained barley, and propionic acidfermented feed, which were compared when mixed with decomposing rat carcasses under natural climate conditions during a Finnish summer. Acta vet. scand. vol. 43 no. 4, 2002 Materials and methods Study design Twentyfour male Wistar rats served as target rats in 4 test environments and 39 served as recipient rats to confirm the infectivity and reproduction capacity index (RCI) of isolated larvae. At the start of the experiment, these rats on average weighed 233 g (14 g; standard deviation, SD). Target rats were each infected with approximately 300 muscle larvae of T. spiralis (ISS559, code at the International Trichinella Reference Center, Rome) in minced mouse meat. This T. spiralis strain had originally been isolated from a natural infection of a Finnish pig and maintained in laboratory mice for 8 generations before the experiment. Five weeks after passage mice were euthanized and eviscerated skinned carcasses were minced for inoculum for target rats. Four weeks post-infection, all target rats were anesthetized with CO2 and euthanazied by decapitation. A digestion sample from a left hind leg was taken to confirm the initial intensity of the infection. Six rat carcasses were placed in each test environment. The environments were sampled by tests on 3 pieces of different infected rats after 1, 2, 4, and 6 weeks' incubation. The infectivity of the larvae found were confirmed by inoculation per os of recipient rats. The committee on animal experiments of the University of Helsinki has approved the study (D. no: 354/2001). Test environments Silage. Six dead infected target rats were placed in one large plastic-covered bale of silage of approximately 750 kg on the day of harvest. During silage processing, shredded hay was mixed with a formic acid-based preservation solution 5l/1000 kg of cut hay (AIV-2000, Kemira Agro Oy, Oulu, Finland), but the natural fermentation decreases the pH further. A shredder-baling machine cuts hay at the shortest into 4-cm pieces. Therefore, the dead rats Persistence of Trichinella spiralis in feed were cut into similar pieces. Each piece of rat was placed in a pouch made from polyamide pantyhose (Anette 40 den, Finnwear Oy, Tornio, Finland) and placed into the bale through a hole made by a sampling drill. The air-tight plastic cover of the bale was closed again with adhesive tape and plastic. The bale was stored outside in a similar way as in ordinary farming. Grained barley. To simulate the hazard of an infected rat being milled together with barley, the 6 skinned target rat carcasses were minced in a commercial meat mincer (LM-5, Koneteollisuus OY, Nurmijärvi, Finland), placed in the same type of polyamide pouch and placed in the grain, which was stored in 100 l plastic container inside a barn. Propionic acid-preser ved feed. By this method, seeds are not grained but flattened and mixed with commercial preservative solution 10 l/1000 kg (Propcorn7, BP Chemicals, Middlesex, Great Britain). A mixture of barley (30%) and oats (70%) was purchased as ready mixed feed. The 6 skinned target rat carcasses were minced and handled as above. Both grained barley and propionic acid-fermented feed were stored in the same room in similar plastic containers, side by side. Pasture simulation. To simulate a situation of rats having died in the pasture, and also to compare the effect of different forage-processing methods on the survival of T. spiralis, one group of 6 target rat carcasses was placed in a shaded box kept outside close to the silage bale. The box was made from plywood, well-ventilated but inaccessible to invasion by any creatures bigger than ants. The carcasses were placed in polyamide pouches but were otherwise intact except, for the left hind legs having been removed for parasitological examination before the incubation. 205 Environmental factors analyzed Outside temperature was recorded both near the shaded box and the silage bale and inside the barn close to the grain containers. Inside temperature and humidity in the shaded box were recorded by a computer based monitor (Tinytag temperature/humidity logger, Gemini Data Loggers LTD, Chichester, England). The pH was monitored in each feed, both with indicator paper and by pH meter in the liquid phase after over-night incubation in a refrigerator with distilled water added equal to 50% of the volume. The content of dry matter was also recorded after overnight drying in the incubator at 105°C, but was reported as moisture (100% – dry matter content %). At the end of the experiment, pH and moisture were monitored in the remaining target rats as well as in a fresh minced rat carcass. Parasitological examinations Intensity of infection was analyzed by artificial digestion by the HCL-pepsin method shaking either with a Jumbomix (Interscience, Saint Nom, France) or a magnetic stirrer according to recommendations of the International Commission on Trichinellosis (Gamble et al. 2000). Data from minced meat were used for those rats which were minced for the purpose of simulating a particular feed-processing (target rats in ground barley or propionic acid- fermented feed). For other target rats (shaded box or rats in silage) results from the left hind leg muscles were used. The infectivity of the harvested larvae was confirmed by inoculating isolated larvae by stomach tube into the recipient rats. The infection dose was either 300 larvae or fewer, depending on recovery of larvae from samples. Recipient rats were killed after 6 to 8 weeks of follow-up time, and the intensity of infection was analyzed as described above. To calculate the reproduction capacity index (Dick 1983, Kapel et al. 2000), the total number of Acta vet. scand. vol. 43 no. 4, 2002 206 L. Oivanen et al. Figure 1. Average lpg with standard deviation in rats incubated in different feeds. Trichinella in each rat was estimated by multiplying larvae per gram of muscle (lpg) by the animal's total weight and dividing this arbitrary value by infection dose. Data are presented as average and standard deviation. Results Per oral feeding of target rats with infected minced mouse flesh yielded variable intensities of infection. When measured in hind leg muscles, the average was 164 (60, SD) lpg and from minced meat 100 (29) lpg. Figure 2. Average reproduction capacity index with standard deviation after incubation. Acta vet. scand. vol. 43 no. 4, 2002 Persistence of Trichinella spiralis in feed Table 1. Effect of 6 weeks' incubation on target rat carcasses in different environments. Fresh rat Propionic Shaded Silage Grain pH moisture (%) 6.1 7.0 7.4 7.4 6.5 62.2 33.1 19.8 67.0 21.1 207 lated from initial inoculation of target rats with 300 larvae in minced mouse meat (Fig. 2). The target rats were badly decomposed after 6 weeks of incubation. In silage, the fleshy parts were liquefied, and only bones and hairs were left in the polyamide pouches. In grained barley, the rat carcasses were mummified, and in propionic acid-fermented feed, large moldy feed clumps surrounded the rat carcasses. In the shaded box, maggots had consumed the carcasses by 6 weeks. Maggots were found even in one-week samples, and the breeding of maggots and fur beetles was the main decaying factor in otherwise mummifying rat carcasses. Moisture had increased in carcasses incubated in silage (Table 1), but target rats in other environments were desiccated. The pH in all incubated carcasses was higher than in a fresh minced rat carcass (Table 1). The summer of 2001 was warm in Finland. During the experimental period, the maximum temperature recorded inside the shaded box was 42°C, the minimum 14°C, and the six-week average 23°C (weekly average range: 18.525.5°C). Changes in humidity followed the outside climate. The average humidity was 66% (range: 30%-93%) inside the shaded box. Because grain and propionic acid-fermented feed were kept inside a barn, the daily temperature variation was not as great as in the shaded box, which was exposed to direct sunlight and nighttime temperature drop. The pH decreased in silage from initial 5.1 to 4.5 during the first week and stayed rather constant thereafter (Table 2). In the propionic acid-fermented feed, Propionic = propionic acid-fermented feed. Shaded = shaded box simulating pasture conditions. Grain = grained barley. Trichinella recovery from target rats after different incubation periods is presented in Fig. 1. Some Trichinella were found in all environments until 4 weeks of incubation, but after 2 weeks they were found in only small quantities. After 6 weeks in the shaded box, the fleshy parts of the rats were totally decayed, with no recovery of Trichinella. In all other environments than the shaded box, small remnants of flesh with identifiable Trichinella were found also after 6 weeks of incubation. Recovery of Trichinella after a one-week incubation was sufficient to infect 4 donor rats each inoculated with 300 larvae. Later, the number of donor rats and volume of infection dose was justified based on the recovery. Still, after 2 weeks of incubation, Trichinella recovered from all environments were infective (Fig. 2) but in 4 weeks only parasites from propionic acid-fermented fodder reproduced in recipient rats. By 6 weeks, no parasites were found to be infective. Original stock infectivity was calcu- Table 2. pH and moisture (%) of feeds at different sampling times (weeks of incubation). Weeks Propionic Silage Grain 0 1 2 4 6 pH % pH % pH % pH % pH % 4.9 5.1 5.8 19.6 71.6 12.1 4.8 4.5 5.9 19.9 74.8 12.0 4.8 4.7 5.8 20.2 75.1 12.9 4.8 4.5 5.8 20.9 72.8 13.5 4.8 4.7 5.8 21.4 73.8 13.4 Acta vet. scand. vol. 43 no. 4, 2002 208 L. Oivanen et al. the pH stayed between 4.79 and 4.81 in every sampling and in grain 5.8 - 5.9 (Table 2). Moisture of the feeds was also rather constant during the 6 week study period (Table 2). Discussion Initial analyses of target rats showed lower lpg yield in minced meat than in muscles of the hind leg. It is a well-known fact that Trichinella larvae are not evenly distributed throughout the skeletal musculature (Alkarmi et al. 1990, Kapel et al. 1994, Pozio et al. 1999, Oksanen et al. 2000, Mikkonen et al. 2001). Obviously, hind leg muscles are the predilection of trichinellosis in rats, and the lower lpg yield in minced meat indicates the dilution effect of other musculature and tissue. One-week recovery yields (lpg) were on the same level or even a bit higher than in the initial samples. The small increase detected in infection intensity can be explained by the decreased moisture of the rat carcasses which caused relatively higher figures per weight. However, bigger increases may be explained by the fact that the larvea are not evenly distributed in rat bodies; there is always sample-to-sample variation in the same bodies and even in the same muscle. In the samples incubated longer than one week, the larval recovery in propionic acid-fermented feed tended to be higher than in other environments. Unlike in natural conditions, rats in the shaded box were not exposed to rain. Therefore, the carcasses appeared to dry and mummify if not consumed by colonizing maggots. The effect of proteolytic putrefaction seen as an increase in pH was strongest in silage and mildest in grain. In silage, the humid environment increased the moisture of the carcasses, but other environments dried them up. The drying effect was highest in the hot shaded box, but dry grain also took up much of the water from rat carcasses. Acta vet. scand. vol. 43 no. 4, 2002 Intriguingly, infectivity was least affected in those target rat carcasses kept in the shaded box, but because the flesh was devoured up by maggots, no larvae could be recovered, and the RCI was not confirmed after 4 weeks. The different feed processing methods all seemed to have a negative effect on the reproduction performance of Trichinella seen at the 2 weeks' sampling. Maroli & Pozio (2000) showed that Trichinella larvae can survive and be infective when ingested by maggots. Their survival in maggots depended on time and environmental temperature, but was not longer than 5 days. Silage packed in bales is often stored outdoors until used, also in wintertime; freezing does not spoil the feed. Those Trichinella species resistant to freezing can survive in contaminated fodders even during the winters of northern Europe. Stewart et al. (1990), studying the persistence of T. pseudospiralis in mouse carcasses, found them were infective for only 2 weeks when kept in 24°C, but in those mouse carcasses kept at 4°C, infectivity was preserved up to 30 days. A lower environmental temperature may thus prolong the persistence of infectivity in feeds. In pork buried in the ground T. spiralis survived infective at least for 90 days (Jovic et al. 2001). The ability of Trichinella to be infective also in different feeds for some weeks can be the explanation for unexpected herbivore hosts known to be sources of human outbreaks (Boireau et al. 2000, Touratier 2001). In an endemic area, rodents can cause a risk for trichinellosis also to indoor animals both by contaminating their feed and because these animals (such as pig) scavenge or hunt infective pest animals (Schad et al. 1987, Murrell et al. 1987). Fresh hay is used soon after harvesting. In our experiment, infectivity in the pasturecondition simulation was not at all affected in one week. For this reason, contaminated rat carrion mixed with hay may be the source of an outbreak. The typical management practice of Persistence of Trichinella spiralis in feed milling the grain at the farm and mixing it with protein concentrate does not include long storage of prepared feed. Two weeks' persistence of infectivity can thus be hazardous if rats have colonized the crop storage. Silage is recommended to be fermented for at least one month before use. In our experiment in summer temperatures, infectivity in silage was minimized by 4 weeks' incubation. It is worth noting, that after 4 weeks, infective larvae were still found in propionic acid-fermented feed. In endemic areas, rat control is important to prevent trichinellosis. Methods are minimizing direct contact and maintaining feed hygiene. Acknowledgements The authors acknowledge Ilkka Saastamoinen, Annukka Pesonen and Ilpo Forsman for help in lab and field work, and Carolyn Norris, PhD, for editing the English. This study has been supported by grants from the Walter Ehrström Foundation (LO), the Research Foundation of Veterinary Sciences (LO), the Emil Aaltonen Foundation (LO, TM) and the Marjatta and Eino Kolli Foundation (TM). References Alkarmi T, Behbehani K, Abdou S, Ooi HK: Infectivity, reproductive capacity and distribution of Trichinella spiralis and T. pseudospiralis larvae in experimentally infected sheep. Jpn. J. Vet. Res. 1990, 38, 139-146. Boireau P, Vallee I, Roman T, Perret C, Mingyuan L, Gamble HR, Gajadhar A: Trichinella in horses: a low frequency infection with high human risk. Vet. Parasitol. 2000, 93, 309-20. Capó V, Despommier DD: Clinical aspects of infection with Trichinella spp. Clinical Microbiol. Rev. 1996, 9, 47-54. Dick T: Species and infraspecific variation. In: Campbell C (editor): Trichinella and Trichinosis. Plenum Press New York 1983, pp. 31-73. Fitzgerald PR, Prakasam TBS: Survival of Trichinella spiralis larvae in sewage sludge anaerobic digesters. J. Parasitol. 1978, 64, 445-447. Gamble HR, Brady RC, Bulaga LL, Berthoud CL, Smith WG, Detweiler LA, Miller LE, Lautner EA: Prevalence and risk association for Trichinella infection in domestic pigs in the northeastern 209 United States. Vet. Parasitol. 1999, 82, 59-69. Gamble HR, Bessonov AS, Cuperlovic K, Gajadhar AA, van Knapen F, Noeckler K, Schenone H, Zhu X: International commission on trichinellosis: recommendations on methods for the control of trichinella in domestic and wild animals intended for human consumption. Vet. Parasitol. 2000, 93, 393-408. Jovic S, Djordjevic M, Kulisic Z, Pavlovic S, Radenkovic B: Infectivity of Trichinella spiralis larvae in pork buried in the ground. Parasite 2001, 8, S213-S215. Kapel CM, Henriksen SA, Dietz HH, Henriksen P, Nansen P: A study on the predilection sites of Trichinella spiralis muscle larvae in experimentally infected foxes (Alopex lagopus, Vulpes vulpes). Acta Vet. Scand. 1994, 35, 125-132. Kapel CM, Gamble H: Infectivity, persistence, and antibody response to domestic and sylvatic Trichinella spp. in experimentally infected pigs. Int. J. Parasitol. 2000, 30, 215-221. Köller von J, Kapel CMO, Enemark HL, Hindsbo O: Infectivity of Trichinella spp. recovered from decaying mouse and fox muscle tissue. Parasite 2001, 8, S209-S212. Leiby DA, Duffy CH, Murrell KD, Schad GA: Trichinella spiralis in an agricultural ecosystem: transmission in the rat population. J. Parasitol. 1990, 76, 360-364. Maroli M, Pozio E: Influence of temperature on the survival and infectivity of Trichinella spiralis larvae in Sarcophaga argyrostoma (Diptera, Sarcophagidae) maggots. J. Parasitol. 2000, 86, 633634. Mikkonen T, Oivanen L, Näreaho A, Helin H, Sukura A: Predilection muscles and physical condition of raccoon dogs (Nyctereutes procyonoides) experimentally infected with Trichinella spiralis and Trichinella nativa. Acta Vet. Scand. 2001, 42, 441-452. Mikkonen T, Oivanen L, Wihlman H, Haukisalmi V: Trichinellosis in rats (Rattus norvegicus) in Finland. – Research in 12 Finnish dumps. In: OrtegaPierres G. et al. (eds.) Trichinellosis. Proceedings of Ninth International Conference on Trichinellosis. Centro de Investigación y Estudios Avanzados del Instituto Politécnico Nacional México, D. F. México. Mexico 1998, 551-558. Murrell KD, Stringfellow F, Dame JB, Leiby DA, Duffy C, Schad GA: Trichinella spiralis in an agricultural ecosystem. II. Evidence for natural transmission of Trichinella spiralis spiralis from Acta vet. scand. vol. 43 no. 4, 2002 210 L. Oivanen et al. domestic swine to wildlife. J. Parasitol. 1987, 73, 103-109. Murrell KD: Beef as a source of trichinellosis. Parasitology Today 1994, 10, 434. Oivanen L, Mikkonen T, Sukura A: An outbreak of trichinellosis in farmed wild boar in Finland. APMIS 2000, 108, 814-818. Oivanen L, Kapel CMO, Pozio E, La Rosa G, Mikkonen T, Sukura A: Associations between Trichinella species and host species in Finland. J. Parasitol. 2002, 88, 84-88. Oksanen A, Oivanen L, Eloranta E, Tirkkonen T, Åsbakk K: Experimental trichinellosis in reindeer. J. Parasitol. 2000, 86, 763-767. Pozio E, Paterlini F, Pedarra C, Sacchi L, Bugarini R, Goffredo E, Boni P: Predilection sites of Trichinella spiralis larvae in naturally infected horses. J. Helminthol. 1999, 73, 233-237. Pozio E, Tamburrini A, La Rosa G: Horse trichinellosis, an unresolved puzzle. Parasite 2001, 8, S263-S 265. Schad G, Duffy CH, Leiby DA, Murrell KD, Zirkle EZ: Trichinella spiralis in an agricultural ecosystem: Transmission under natural and experimentally modified on-farm conditions. J. Parasitol. 1987, 73, 95-102. Smith HJ, Kay ED: Role of rats in the transmission of Trichinella spiralis spiralis to swine. Can.Vet. J. 1987, 28, 604. Stewart GL, Kennedy RR, Larsen E: Infectivity of Trichinella pseudospiralis isolated from carrion. J. Parasitol. 1990, 76, 750-751. Suomi, K: Emakon ruokinta. (Feeding of a sow) In: Sikojen ruokinta (Feeding of swine) Maaseutukeskusten liitto, Gummerus Kirjapaino Oy, Jyväskylä, Finland. 1999, pp. 57-64. Sukura A, Näreaho A, Veijalainen P, Oivanen L: Trichinellosis in farmed wild boar: meat inspection findings and seroprevalence. Parasite 2001, 8, S243-S245. Touratier LA: Challenge of veterinary public health in the European union: human trichinellosis due to horse meat consumption. Parasite 2001, 8, S252-S256. Wang ZQ, Cui J: The epidemiology of human trichinellosis in China during 1964-1999. Parasite 2001, 8, S63-S66. Sammanfattning Trichinella spiralis hållbarhet i rått-kadaver som experimentellt blandats i olika foder. Döda råttor infekterade med Trichinella spiralis inkuberades under 6 veckor i olika djurfoder för att uppskatta hur länge Trichinella utgör en smittorisk i kontaminerat foder. 24 infekterade råttor, i grupper på 6 djur, placerades i silage, säd, foder konserverat med propionsyra och på simulerat naturligt grönbete. Efter en, 2, 4 och 6 veckors inkubering togs prov från alla experimentella omgivningar. Trichinella larver återvanns ur proven genom digestion och larvernas infektivitet beprövades genom inokulering i mottagliga råttor. Två veckors inkubering minskade antalet larver, men efter 6 veckors inkubering kunde fortfarande ett litet antal larver isoleras ur alla fodertyper med undantag av den lagringsform som simulerade naturligt grönbete. Vid provtagningen efter 2 veckors inkubation var larverna infektiva i alla foder. Men efter 4 veckors lagring fanns infektiva larver endast i fodret som konserverats med propionsyra och här i ett litet antal och med reducerad förökningsförmåga. Härur slutleds att risken att råttor eller annat infekterat material blandas i hö eller annat foder kan utgöra en fara för boskapsdjur. Om silage lagras åtminstone en månad innan användning minimeras risken i detta foder. (Received June 20, 2002; accepted July 1, 2002). Reprints may be obtained from A. Sukura, Department of Basic Veterinary Sciences, Faculty of Veterinary Medicine, University of Helsinki. POB 57. FIN-00014 Helsinki, Finland. E-mail: antti.sukura@helsinki.fi, tel: +358 9 191 49526, fax: +358 9 191 49799. Acta vet. scand. vol. 43 no. 4, 2002