Dogs and N. caninum

Dogs can be both definitive and intermediate hosts of N. caninum – the reasons that determine which role a particular dog assumes when infected are not understood, nor are those that determine whether a dog will progress to clinical disease or not. In a number of case reports on clinical canine neosporosis, some infected litter mates progress to severe clinical manifestation, while others remain healthy (Reichel et al. Reference Reichel, Thornton, Morgan, Mills and Schares1998).

Disease in dogs caused by N. caninum was first diagnosed in Norway (Bjerkås et al. Reference Bjerkås, Mohn and Presthus1984) and thereafter has been reported around the world (Gasser et al. Reference Gasser, Edwards and Cole1993; Ruehlmann et al. Reference Ruehlmann, Podell, Oglesbee and Dubey1995; Barber and Trees, Reference Barber and Trees1996; Reichel et al. Reference Reichel, Thornton, Morgan, Mills and Schares1998).

Clinical reports in dogs, however, seem to be the exception rather than the rule. Dogs may suffer from a variety of clinical manifestations of N. caninum infection (ranging from dermatitis to pneumonia), but as in cattle, abortions, stillbirths and neonatal deaths are more frequent occurrences (Barber and Trees, Reference Barber and Trees1996), while an ascending paralysis of the hindlegs is almost pathognomonic in puppies less than 6 months of age (Ruehlmann et al. Reference Ruehlmann, Podell, Oglesbee and Dubey1995; Reichel et al. Reference Reichel, Ellis and Dubey2007).

In 1998, dogs were also shown to be a definitive host in the life cycle of N. caninum (McAllister et al. Reference McAllister, Dubey, Lindsay, Jolley, Wills and McGuire1998) and subsequently other canids, such as coyotes (Gondim et al. Reference Gondim, McAllister, Pitt and Zemlicka2004c ), dingoes in Australia (King et al. Reference King, Šlapeta, Jenkins, Al-Qassab, Ellis and Windsor2010) and grey wolves (Dubey et al. Reference Dubey, Jenkins, Rajendran, Miska, Ferreira, Martins, Kwok and Choudhary2011). Foxes (Vulpes vulpes) have also been suggested as definitive hosts (Wapenaar et al. Reference Wapenaar, Jenkins, O'Handley and Barkema2006) and identified as a risk factor associated with infection/abortion in cattle (Wouda et al. in Hemphill and Gottstein, Reference Hemphill and Gottstein2000) (for further details see also Table 1).

Table 1.

Reports for an epidemiological association of Neospora caninum infection and/or abortion with canids (dogs/foxes/coyotes) (odds ratios and other reported epidemiological associations)

Canids are infected through the consumption of infected bovine tissues and start excreting oocysts about 1 week after infection (Gondim et al. Reference Gondim, Gao and McAllister2002). The numbers of oocysts produced by dogs may vary, depending on the tissue consumed that initially harboured the tissue cysts, with bovine or buffalo tissue-derived cysts resulting in higher numbers being shed than when N. caninum cysts were derived from experimentally infected mice (Gondim et al. Reference Gondim, Gao and McAllister2002; Rodrigues et al. Reference Rodrigues, Gennari, Aguiar, Sreekumar, Hill, Miska, Vianna and Dubey2004). The oocysts need to sporulate first to become infective (McAllister et al. Reference McAllister, Dubey, Lindsay, Jolley, Wills and McGuire1998). The proportion of dogs shedding oocysts at any one time may be small, as only 7 in >24 000 dogs coprologically examined in a large-scale study in Germany were found to be shedding N. caninum (Schares et al. Reference Schares, Pantchev, Barutzki, Heydorn, Bauer and Conraths2005). Puppies tend to produce more oocysts than adults (Gondim et al. Reference Gondim, McAllister and Gao2005), and this experiment, albeit with a small number of dogs, also suggested that following the initial infection dogs may become refractory to oocyst shedding for a period of between 8 to 18 months.

The seroprevalence reported from dogs varies across countries and continents (Barber et al. Reference Barber, Gasser, Ellis, Reichel, McMillan and Trees1997; Reichel, Reference Reichel1998; Ortuño et al. Reference Ortuño, Castella and Almeria2002; Andreotti et al. Reference Andreotti, Oliveira, Silva, Oshiro and Matos Mde2006; Collantes-Fernandez et al. Reference Collantes-Fernandez, Gomez-Bautista, Miro, Alvarez-Garcia, Pereira-Bueno, Frisuelos and Ortega-Mora2008; Dubey et al. Reference Dubey, Stone, Kwok and Sharma2008; Ploneczka and Mazurkiewicz, Reference Ploneczka and Mazurkiewicz2008; Sharma et al. Reference Sharma, Bal, Meenakshi, Kaur, Sandhu and Dubey2008; King et al. Reference King, Brown, Jenkins, Ellis, Fleming, Windsor and Šlapeta2012). However, dogs do not always seroconvert soon after infection and may also revert back to seronegative status, despite the shedding of oocysts (McAllister et al. Reference McAllister, Dubey, Lindsay, Jolley, Wills and McGuire1998; Cavalcante et al. Reference Cavalcante, Monteiro, Soares, Nishi, Alves Neto, Esmerini Pde, Sercundes, Martins and Gennari2011), thus these estimates probably underestimate the true prevalence of infection. In summary, the presence of canids is a risk factor for infection of cattle. Many epidemiological studies show a strong statistical correlation between the presence and density of dogs and seroprevalence and probability of abortion in cattle (Table 1).

Abortions and disease in cattle

The observation that N. caninum causes abortions in cattle was first made in the late 1980s in the USA (Thilsted and Dubey, Reference Thilsted and Dubey1989), only a few years after clinical disease was first described in dogs. There are earlier reports of congenital infections in calves (O‘Toole and Jeffrey, Reference O'Toole and Jeffrey1987; Parish et al. Reference Parish, Maag-Miller, Besser, Weidner, McElwain, Knowles and Leathers1987). Observations in other countries soon followed, and particularly in countries such as Australia and New Zealand where bovine brucellosis was eradicated, neosporosis quickly became the most often diagnosed infectious cause of abortions in cattle (Thornton et al. Reference Thornton, Thompson and Dubey1991; Obendorf et al. Reference Obendorf, Murray, Veldhuis, Munday and Dubey1995). In New Zealand, it remains the most commonly diagnosed cause of abortion, at around 20% of abortions, according to information collated from veterinary diagnostic laboratories (Bingham et al. Reference Bingham, Rawdon and Watts2012).

In experimental models, cattle abort from about 3 weeks after infection (Williams et al. Reference Williams, Guy, Smith, Ellis, Björkman, Reichel and Trees2007) when infected with intravenously applied doses of N. caninum tachyzoites at 70 days of gestation. In challenge trials that used the (more realistic) infection route of an oral application of oocysts, abortions occurred later, about 6 weeks after infection (Gondim et al. Reference Gondim, McAllister, Anderson-Sprecher, Björkman, Lock, Firkins, Gao and Fischer2004a ; McCann et al. Reference McCann, McAllister, Gondim, Smith, Cripps, Kipar, Williams and Trees2007). These trials also give an estimate of the infective dose of N. caninum capable of effecting abortion, which, in the McCann trial, was estimated to be as low 127 viable N. caninum oocysts.

Under field conditions, cows appear to abort generally in mid-term, at 5–7 months of gestation (Dubey et al. Reference Dubey, Buxton and Wouda2006).

Infection of the foetus later in gestation, in the third trimester, more often than not results in the subclinical infection of the offspring, which is born clinically normal, but congenitally infected (Pereira-Bueno et al. Reference Pereira-Bueno, Quintanilla-Gozalo, Seijas-Carballedo, Costas and Ortega Mora2000; Quintanilla-Gozalo et al. Reference Quintanilla-Gozalo, Pereira-Bueno, Seijas-Carballedo, Costas and Ortega Mora2000; Williams et al. Reference Williams, Guy, McGarry, Guy, Tasker, Smith, MacEachern, Cripps, Kelly and Trees2000; Innes et al. Reference Innes, Andrianarivo, Björkman, Williams and Conrad2002; Gondim et al. Reference Gondim, McAllister, Anderson-Sprecher, Björkman, Lock, Firkins, Gao and Fischer2004a ).

Life cycle

Dogs (canids) are at the heart of the epidemiology of N. caninum infections and abortions, closing the part of the life cycle that includes sexual replication in canids, and subsequent infection of cattle. However there is another life cycle that just involves cattle, and which relies on the efficient vertical transmission of the parasite from chronically infected dams to their offspring (Paré et al. Reference Paré, Thurmond and Hietala1996; Schares et al. Reference Schares, Peters, Wurm, Barwald and Conraths1998; Hall et al. Reference Hall, Reichel and Ellis2005). Even though some reports claim that transmission is close to 90–100% efficient (see Transmission), theoretical modelling suggests that both parts of the life cycle are necessary to maintain N. caninum in cattle populations (French et al. Reference French, Clancy, Davison and Trees1999). The literature provides strong evidence that the presence of dogs/canids is associated with infection with N. caninum in cattle (Table 1).

Transmission

An important aspect of the parasite is its ability to observe two modes of transmission in cattle.

Other risk factors

Wild and domestic birds have been shown to be able to be infected with N. caninum experimentally and in field observations (Baker et al. Reference Baker, Morishita, Brooks, Shen, Lindsay and Dubey1995; McGuire et al. Reference McGuire, McAllister, Wills and Tranas1999; Costa et al. Reference Costa, Santos, Uzeda, Pinheiro, Almeida, Araujo, McAllister and Gondim2008; Mineo et al. Reference Mineo, Carrasco, Marciano, Werther, Pinto and Machado2008). Epidemiologically, the presence of poultry has been linked with N. caninum sero-positivity (Otranto et al. Reference Otranto, Llazari, Testini, Traversa, Frangipane di Regalbono, Badan and Capelli2003) and abortions (Bartels et al. Reference Bartels, Wouda and Schukken1999). Wild rodents have been shown to be infected with N. caninum (Huang et al. Reference Huang, Yang, Watanabe, Liao and Ooi2004; Barratt et al. Reference Barratt, Al Qassab, Reichel and Ellis2008) and dogs will shed oocysts after the feeding of experimentally N. caninum infected mice (Gondim et al. Reference Gondim, Gao and McAllister2002), but the biological significance of these observations has not been established. Pond water as a source of water supply was also identified as a risk factor for seropositivity, as was the presence of ducks and rabbits (Ould-Amrouche et al. Reference Ould-Amrouche, Klein, Osdoit, Mohammed, Touratier, Sanaa and Mialot1999).

Other domestic ruminants

A number of other species have been reported to be susceptible to N. caninum infections and they might potentially act as reservoirs of continued infection – most commonly sheep and goats are reported as infected. The reported seroprevalences for small ruminants appear to be low in Australia and New Zealand (Reichel et al. Reference Reichel, Ross and McAllister2008; Bishop et al. Reference Bishop, King, Windsor, Reichel, Ellis and Slapeta2010) and parts of South America such as Argentina (Hecker et al. Reference Hecker, Moore, Manazza, Unzaga, Spath, Pardini, Venturini, Roberi and Campero2013a ), but also occasionally higher in other parts of that continent, for example in Brazil (Figliuolo et al. Reference Figliuolo, Kasai, Ragozo, de Paula, Dias, Souza and Gennari2004). In Spain, the presence of N. caninum and Toxoplasma gondii infections in abortion cases in small ruminants submitted for diagnosis has been recently studied. Lesions suggestive of protozoal infection were observed in 10·8% (8/74) and 15·4% (4/26) of ovine and caprine abortions respectively. Neospora caninum and T. gondii infection was detected by PCR in 6·8% (5/74) and 5·4% (4/74) of sheep foetuses, respectively, of which five showed protozoal-associated lesions. Neospora caninum DNA was detected in 11·5% (3/26) of goat foetuses, of which two showed protozoal-associated lesions, whereas T. gondii DNA was detected in one goat foetus with no lesions. The simultaneous presence of N. caninum and T. gondii DNA was detected in one sheep foetus with severe lesions (Moreno et al. Reference Moreno, Collantes-Fernandez, Villa, Navarro, Regidor-Cerrillo and Ortega-Mora2012).

As their risk to cattle is mediated through the completion of the life cycle in dogs, similar measures should be limiting a dog's ability to complete the N. caninum life cycle through ingestion of ovine or caprine tissues. An added bonus of this approach would be the control of other parasites, some of which have zoonotic potential, such as the cestode Echinococcus granulosus (Barnes et al. Reference Barnes, Deplazes, Gottstein, Jenkins, Mathis, Siles-Lucas, Torgerson, Ziadinov and Heath2012).

Wildlife and other bovids

There is strong evidence that N. caninum is a common parasite of wildlife, with a major sylvatic life cycle in wild canids and ruminants (Gondim, Reference Gondim2006; King et al. Reference King, Jenkins, Ellis, Fleming, Windsor and Šlapeta2011a ). In North America, high seroprevalence has been documented in white tailed deer (Odocoileus virginianus), moose (Alces alces), wolves (Canis lupus) and coyotes (Canis latrans) (Gondim et al. Reference Gondim, McAllister, Mateus-Pinilla, Pitt, Mech and Nelson2004b ). Neospora caninum has been isolated from deer tissues and dogs shed N. caninum oocysts after ingestion of infected deer tissue (Gondim et al. Reference Gondim, McAllister, Mateus-Pinilla, Pitt, Mech and Nelson2004b ). Similarly, coyotes shed oocysts after ingestion of infected bovine tissues (Gondim et al. Reference Gondim, McAllister, Pitt and Zemlicka2004c ). Oocysts have also been observed in faeces of wild wolves (Dubey et al. Reference Dubey, Jenkins, Rajendran, Miska, Ferreira, Martins, Kwok and Choudhary2011) and wild red foxes (Wapenaar et al. Reference Wapenaar, Jenkins, O'Handley and Barkema2006). In several studies from North America, up to half of deer samples had antibodies to N. caninum (Dubey et al. Reference Dubey, Hollis, Romand, Thulliez, Kwok, Hungerford, Anchor and Etter1999; Lindsay et al. Reference Lindsay, Little and Davidson2002). Wolves have a seroprevalence among the highest of any studied canine population (Gondim et al. Reference Gondim, McAllister, Mateus-Pinilla, Pitt, Mech and Nelson2004b ), suggesting that a sylvatic life cycle of N. caninum and its potential spill over into cattle is a concern (Rosypal and Lindsay, Reference Rosypal and Lindsay2005). Additional life cycles are also possible, for example feral and wild rodents have been found to harbour N. caninum as detected by PCR (Ferroglio et al. Reference Ferroglio, Pasino, Romano, Grande, Pregel and Trisciuoglio2007; Jenkins et al. Reference Jenkins, Parker, Hill, Pinckney, Dyer and Dubey2007; Barratt et al. Reference Barratt, Al Qassab, Reichel and Ellis2008) and also by isolation in vivo (Huang et al. Reference Huang, Yang, Watanabe, Liao and Ooi2004). The suspected involvement of red fox would most likely necessitate smaller infected prey such as rodents or birds. Small Australian marsupials, such as the fat-tailed dunnart (Sminthopsis crassicaudata), are exquisitely susceptible to N. caninum infection and produce large numbers of tissue cysts (King et al. Reference King, McAllan, Spielman, Lindsay, Hurkova-Hofmannova, Hartigan, Al-Qassab, Ellis and Šlapeta2011b ), which in turn might infect canid predators that could perpetuate the N. caninum life cycle.

The ‘Do nothing’ option

Low levels of infection with N. caninum may well make the identification/control of the infected animals (veterinary and laboratory testing costs) or eradication from the herd (replacement cost of seropositive animals, loss of calves) more expensive than the economic losses incurred by the disease (abortion, early culling, possibly loss in milk yield etc.). For example, in South Australia, less than 5% of beef and dairy cattle appear to be infected across the state (Nasir et al. Reference Nasir, Lanyon, Schares, Anderson and Reichel2012b ), and very few farms demonstrate a within-herd seroprevalence of greater than 15%. Economic modelling (Reichel and Ellis, Reference Reichel and Ellis2006) has previously suggested that it may be the preferred course of action in such herds to ‘live with the disease’, as long as the within-herd prevalence remains below a threshold of 18–21%. The cost of identifying those few infected individual animals (professional veterinary and diagnostic testing cost, replacement of infected individuals) will exceed the benefit (reduction in abortions, stillborn and congenitally defected calves, improved milk yield).

In other states of Australia, particularly on the Eastern seaboard, the prevalence of N. caninum infection appears to be higher than this threshold, at ∼20–30% (Hall et al. Reference Hall, Reichel and Ellis2006; Landmann et al. Reference Landmann, Gunn, O'Donoghue, Tranter and McGowan2011). Even in New Zealand, where the national seroprevalence of dairy cattle appears to be generally below 10% (Reichel, Reference Reichel1998), and less than 5% in beef cattle (Tennent-Brown et al. Reference Tennent-Brown, Pomroy, Reichel, Gray, Marshall, Moffat, Rogers, Driscoll, Reeve, Ridler and Ritaven2000), the within-herd N. caninum prevalence on infected dairy farms generally exceeds 30% (Thornton et al. Reference Thornton, Gajadhar and Evans1994; Cox et al. Reference Cox, Reichel and Griffiths1998; Schares et al. Reference Schares, Conraths and Reichel1999; Pfeiffer et al. Reference Pfeiffer, Williamson, Reichel, Wichtel and Teague2002), making eradication economically attractive. Provided that vertical transmission is the predominant route of transmission and that risk from postnatal (exogenous) transmission is negligible, then a significant reduction of N. caninum-infected cattle or eradication from the herd is achievable if economically feasible by not retaining replacement breeding heifers from seropositive dams (Hall et al. Reference Hall, Reichel and Ellis2005; Landmann et al. Reference Landmann, Gunn, O'Donoghue, Tranter and McGowan2011).

It seems that breaking the life cycle of N. caninum at any one or several points should lead to a reduction of infection with (and thus abortion from) N. caninum in cattle.

Efforts can be aimed at either reducing endogenous transmission and/or at preventing exogenous transmission, depending on the presumed predominant mode of transmission (Fig. 1). However, veterinarians should emphasize to clients that no matter what approaches may be employed, there will always be some risk of future reintroduction of infection via ingestion of contaminated pastures, feedstuffs, or water. The goal is risk reduction, but complete risk elimination will seldom be feasible.

Fig. 1.

General schematic for approaches to the control of N. caninum infections and abortions. Pathways showing ways for the reduction of the number of N. caninum-infected cows (largely controlling endogenous infection) and pathways showing ways for managing exposure to dogs and canids (managing exogenous infection) and preventing transmission of N. caninum to cattle.

Managing exposure of cattle to canids – enhanced biosecurity measures

Reichel et al. (Reference Reichel, Alejandra Ayanegui-Alcerreca, Gondim and Ellis2013) estimate that globally only 16·1% of dairy and 11·5% of beef cattle are already N. caninum infected. Thus the use of gates, doors, containers and fences to prevent dogs from having access to (and potentially contaminating) feedstuffs is probably the biggest risk-mitigation strategy for preventing point-source transmission.

As dogs and other canids (coyotes, dingoes, gray wolves, but possibly also foxes) are identified as definitive hosts, it would seem prudent to reduce their ability to ingest domestic ruminant (and cervine) tissue material or to contaminate cattle feed (pasture or otherwise) with oocysts. This could include measures such as the limiting/preventing domestic dogs and wild canids gaining access to domestic ruminant material (muscle, foetal material, placenta and colostrum; Uggla et al. Reference Uggla, Stenlund, Holmdahl, Jakubek, Thebo, Kindahl and Björkman1998), and the heat or freezing treatment of dog food that has been prepared from raw domestic ruminant tissue. All domestic ruminant tissues, also those derived from adults, must be assumed to be potentially N. caninum-infected, as the predilection sites for N. caninum encystation are not well known. Ruminant brain, placenta, skeletal muscle, heart and liver tissue, however have been demonstrated to harbour sufficient tissue cysts in infected cattle to infect dogs and result in oocyst shedding (Dijkstra et al. Reference Dijkstra, Eysker, Schares, Conraths, Wouda and Barkema2001b ; Cavalcante et al. Reference Cavalcante, Monteiro, Soares, Nishi, Alves Neto, Esmerini Pde, Sercundes, Martins and Gennari2011). Stray dogs should be controlled.

Although this may be uncommon, dogs may shed oocysts for protracted periods of time (McGarry et al. Reference McGarry, Stockton, Williams and Trees2003) and as such all dog faeces might potentially be a source of infection to cattle. On the other hand, there is evidence that shedding by experimentally infected dogs usually only lasts for about a week (Gondim et al. Reference Gondim, Gao and McAllister2002; Rodrigues et al. Reference Rodrigues, Gennari, Aguiar, Sreekumar, Hill, Miska, Vianna and Dubey2004) and thus this risk period might be short. Nevertheless, it seems prudent to avoid whelping dogs or raising pups near pregnant cattle.

Although the sylvatic cycles are noteworthy, this knowledge does not easily lead to practical control measures for cattle. Some risk of infection of cattle is inherent when grazing fields or drinking from ponds – thus complete eradication of neosporosis is an unlikely and questionable goal. However, a few practical steps can be taken to limit the access that wildlife has to stored feedstuffs, particularly those that will be used in mixed rations. Also, farms should have animal carcass disposal practices that do not give wild carnivores ready access to freshly dead stock or offal. In both instances, the primary protective measure may simply be careful attention to the placement and design of fences and gates.

Water buffalo (Bubalus bubalis) appear to be quite susceptible to infection with N. caninum, and high infection rates are being reported from a number of countries (Guarino et al. Reference Guarino, Fusco, Savini, Di Francesco and Cringoli2000; Gennari et al. Reference Gennari, Rodrigues, Viana and Cardoso2005; Meenakshi et al. Reference Meenakshi, Sandhu, Ball, Kumar, Sharma, Sidhu, Sreekumar and Dubey2007; Nasir et al. Reference Nasir, Ashraf, Khan, Yaqub, Javeed, Avais and Akhtar2011). Particular care should be taken in managing buffalo–dog interactions, especially as dogs fed water buffalo-derived tissue cysts appear to shed large number of oocysts (and thus presumably are in turn more likely to infect water buffalo again) (Rodrigues et al. Reference Rodrigues, Gennari, Aguiar, Sreekumar, Hill, Miska, Vianna and Dubey2004). In comparison, it is reported that dogs shed lower numbers of oocysts when fed murine or bovine tissue-derived cyst material (Gondim et al. Reference Gondim, Gao and McAllister2002). That particular study, however, did not directly compare the relative oocyst output of dogs derived from buffalo vs cattle sources.

Because they have been shown to be intermediate hosts of N. caninum infection, the control of rodents and poultry (and thus prevention of the infection of dogs) might also be beneficial in reducing the risk of post-natal infection of cattle.

Monitoring/testing herds and introductions

If the transmission from one generation of cattle to the next can be prevented (i.e. endogenous transmission), future generations of cattle will be uninfected and should have a reduced risk of abortion provided that further exposure to oocysts can be prevented (see above).

Neospora caninum-infected cattle are assumed to be infected for life and infected cattle are at about three times increased risk of abortion (Paré et al. Reference Paré, Thurmond and Hietala1997; Moen et al. Reference Moen, Wouda, Mul, Graat and van Werven1998); even their seropositive offspring are at continued increased risk of abortion (Thurmond and Hietala, Reference Thurmond and Hietala1997; Wouda et al. Reference Wouda, Moen and Schukken1998). Testing for N. caninum antibodies will allow for the monitoring of replacement heifers; if possible, offspring should only be retained from seronegative cows. Introductions of new stock can be tested and seropositive animals excluded.

Embryo transfer

Embryo transfer (ET) has been shown to be highly effective in breaking the life cycle of N. caninum transmission. Harvesting and transplanting uninfected embryos from seropositive dams has been highly successful (Baillargeon et al. Reference Baillargeon, Fecteau, Paré, Lamothe and Sauve2001; Landmann et al. Reference Landmann, Jillella, O'Donoghue and McGowan2002; Campero et al. Reference Campero, Moore, Lagomarsino, Odeon, Castro and Visca2003) and could provide an avenue for cattle of high genetic value, where the high cost of ET would not matter. It should be remembered though that the valuable offspring is still susceptible to post-natal infection.

Selective breeding

As vertical transmission is cited as highly efficient, it also provides an opportunity to predict the outcome of a gestation – heifer calves from seropositive dams are likely to be infected themselves, and rather than breeding from them (and continuing the life cycle), they could be used for meat production. It might not be economical to test all female offspring of seropositive dams (which would have to be done pre-colostrally or 6 months after birth) to identify those (often few) that are seronegative.

‘Test-and-cull’

Eliminating individual females that test positive should decrease the risk of abortion occurring in a herd and also have the potential to break endogenous transmission of the parasite, potentially reducing/eliminating the infection from the herd (Hall et al. Reference Hall, Reichel and Ellis2005). This should, if complemented by enhanced biosecurity measures (see above) reduce the risk of abortions occurring (Frössling et al. Reference Frössling, Uggla and Björkman2005).

There is good evidence that selectively excluding N. caninum-seropositive, infected dams can quickly reduce the within-herd prevalence of infection (Hall et al. Reference Hall, Reichel and Ellis2005; Landmann et al. Reference Landmann, Gunn, O'Donoghue, Tranter and McGowan2011). The former study reduced the within-herd prevalence by more than 50% in a year, the latter by ∼80% over a 9-year period. A long-term Swedish study showed the potential for pro-active management (culling of seropositive cattle in low-prevalence herds, not retaining, for breeding, the offspring from seropositives in other herds) to reduce infection rates in infected herds over a 2–3 year period (Frössling et al. Reference Frössling, Uggla and Björkman2005). Equally, not retaining the offspring of seropositive cows should reduce the prevalence of infection in herds where the vertical transmission pattern is predominant.

Others (Innes et al. Reference Innes, Buxton, Maley, Wright, Marks, Esteban, Rae, Schock and Wastling2000; McAllister et al. Reference McAllister, Björkman, Anderson-Sprecher and Rogers2000) have presented evidence that already infected dams are at reduced risk of abortion in the event of repeated environmental exposure (essentially ‘naturally’ vaccinated, analogous to the live vaccination mentioned above). There might thus be a case for retaining these cattle in the herd – there is evidence that the risk of abortion in infected animals reduces over time (Thurmond and Hietala, Reference Thurmond and Hietala1997; Stahl et al. Reference Stahl, Björkman, Emanuelson, Rivera, Zelada and Moreno-Lopez2006), however the relative risk (compared with seronegative animals) still remains above 1. These dams are also likely to pass on their infection vertically, thus maintaining infection in a herd, and deferring the risk of abortion to the next generation (Thurmond and Hietala, Reference Thurmond and Hietala1997). Congenitally affected calves may also be affected clinically (Barr et al. Reference Barr, Conrad, Breitmeyer, Sverlow, Anderson, Reynolds, Chauvet, Dubey and Ardans1993), although this may be more common in exogenous outbreaks. Regardless of the cause, there are good management arguments to cull any aborting cow from a commercial herd following good sanitary practices in order to minimize pathogen pollution, or the perpetuation of low reproductive performance.

Reducing concomitant infections

Both infectious bovine rhinotracheitis (IBR) and bovine viral diarrhoea (BVD) (or ‘bovine pestivirus’; Lanyon et al. Reference Lanyon, Hill, Reichel and Brownlie2014) and mycotoxins may exert immune-suppressive effects in cattle that are already infected with N. caninum, which is hypothesized to lead to subsequent recrudescence of the parasite and vertical transmission, and/or abortion. The absence of at least BVD infection has been shown to have a protective effect against N. caninum infection (Vanleeuwen et al. Reference Vanleeuwen, Haddad, Dohoo, Keefe, Tiwari and Scott2010). Control of rodents, while hypothesized to reduce transmission risks of Neospora, will also reduce risks from leptospirosis and salmonellosis, and reduce loss of feedstuffs. Thus control practices for one problem may yield additional benefits. Observations from Switzerland, where bovine neosporosis, on the one hand, is notifiable, and secondly, BVD has recently essentially been eradicated (Presi et al. Reference Presi, Struchen, Knight-Jones, Scholl and Heim2011), should provide evidence soon on whether an absence of BVD infection impacts positively on a reduction in N. caninum abortions. This should also provide evidence on whether vaccination for concomitant infections is not only good husbandry practice but would also make a positive contribution towards N. caninum abortion risk management.

Anti-coccidial treatment

In vitro screening has identified a number of compounds as efficacious against N. caninum (tachyzoites) in cell culture (Lindsay and Dubey, Reference Lindsay and Dubey1989; Lindsay et al. Reference Lindsay, Butler, Rippey and Blagburn1996, Reference Lindsay, Butler and Blagburn1997). Sulphadiazine has also had some effect experimentally in mice (Lindsay and Dubey, Reference Lindsay and Dubey1990) and potentiated sulphonamides and clindamycin remain the drugs of choice in the treatment of clinical cases of neosporosis in dogs (Ruehlmann et al. Reference Ruehlmann, Podell, Oglesbee and Dubey1995; Barber and Trees, Reference Barber and Trees1996; Reichel et al. Reference Reichel, Thornton, Morgan, Mills and Schares1998, Reference Reichel, Ellis and Dubey2007).

Treatment with a coccidiostat (toltrazuril) was successful at eliminating an intravenous, experimental infection with N. caninum tachyzoites in calves when they were treated immediately after infection (Kritzner et al. Reference Kritzner, Sager, Blum, Krebber, Greif and Gottstein2002). This approach has also been found to be an economically viable option in modelling carried out in Switzerland (Häsler et al. Reference Häsler, Regula, Stärk, Sager, Gottstein and Reist2006), requiring low cost inputs, returning the highest positive cost-benefit ratio and high net present value. The authors concluded, however, that considerations regarding food residues and unknown safety profile of the coccidiostat might make it difficult to find acceptance in the general public (or with regulatory authorities). The efficacy of this treatment in the field still needs to be proven. This hypothetical method is aimed at preventing new infections, and would not likely clear established infections.

Treatment with monensin in Canada was identified as being negatively associated with N. caninum seropositivity (Vanleeuwen et al. Reference Vanleeuwen, Haddad, Dohoo, Keefe, Tiwari and Scott2010) in a large-scale epidemiological study of 240 dairy herds, but not statistically associated with better outcomes in a randomized, clinical trial (VanLeeuwen et al. Reference VanLeeuwen, Greenwood, Clark, Acorn, Markham, McCarron and O'Handley2011).

Effective treatment of field infections hence is yet to be demonstrated, and the regulatory hurdles for treatment in a food-producing animal, as much as the need to demonstrate efficacy of such treatment may make this avenue impossible to achieve.

Vaccination

Vaccination with a range of products derived from N. caninum has been trialled for 15 years – early promise came from a crude lysate administered in mice that appeared to prevent completely infection of the offspring (Liddell et al. Reference Liddell, Jenkins, Collica and Dubey1999). This was closely followed by the application in cattle (Choromanski and Block, Reference Choromanski and Block2000). This ultimately led to the commercialization of the only ever registered vaccine for N. caninum (based on inactivated tachyzoites of N. caninum), Neoguard^®, that was recently withdrawn from worldwide sales. The limited commercial success may well be attributed to the low efficacy of the vaccine, recently estimated to be as low as 25% (Weston et al. Reference Weston, Heuer and Williamson2012).

Live vaccine approaches have shown promise and higher efficacy, up to 100% in experimental infections (Williams et al. Reference Williams, Guy, Smith, Ellis, Björkman, Reichel and Trees2007; Rojo-Montejo et al. Reference Rojo-Montejo, Collantes-Fernandez, Perez-Zaballos, Rodriguez-Marcos, Blanco-Murcia, Rodriguez-Bertos, Prenafeta and Ortega-Mora2013; Weber et al. Reference Weber, Jackson, Sobecki, Choromanski, Olsen, Meinert, Frank, Reichel and Ellis2013), but have not yet become commercially available because of perceived difficulties with distribution and product stability. Some live vaccine trials, while demonstrating protection of cattle from abortions, have however shown that transmission to the foetus may still be possible (Hecker et al. Reference Hecker, Moore, Quattrocchi, Regidor-Cerrillo, Verna, Leunda, Morrell, Ortega-Mora, Zamorano, Venturini and Campero2013b ). Sub-unit vaccines have been trialled in mouse models (for details see the reviews by Reichel and Ellis, Reference Reichel and Ellis2009; Hemphill et al. Reference Hemphill, Debache, Monney, Schorer, Guionaud, Alaeddine, Mueller and Mueller2013), but have not yet shown efficacy in cattle trials.

There are thus, at the present moment, no commercial vaccines available for the prevention of infection with, or abortion by N. caninum.

Ideally a vaccine for the prevention of N. caninum infection and/or shedding in dogs is needed and should be sought, rendering dogs ‘safe’. Some attempts at producing a vaccine for N. caninum in dogs have been reported earlier (Nishikawa et al. Reference Nishikawa, Ikeda, Fukumoto, Xuan, Nagasawa, Otsuka and Mikami2000a , Reference Nishikawa, Kousaka, Fukumoto, Xuan, Nagasawa, Igarashi, Fujisaki, Otsuka and Mikami b ), but thus far have not produced a viable vaccine. Rendering dogs ‘safe’ would allow them to resume their other important functions on farms, such as warding of wild canids, thus reducing the possibility of infection from the sylvatic part of the life cycle.

The Netherlands

The Netherlands was a pioneer country in implementing control programmes for neosporosis in dairy herds with reproductive problems. In 2005, an estimate of 3000 farmers used serological testing of aborting cows and regular bulk milk screening (three times per year) as tools in the control programmes. In herds with a positive bulk milk test (validated to over 15% within-herd seroprevalence) and abortion problems, seropositive dams were selected based on a whole herd blood screening and herd-specific measures were adopted. In herds with a negative bulk milk test but with abortion problems, selective culling of heifer calves of seropositive dams was done through testing the relatives of seropositive aborting cows. Finally, herds with a negative bulk milk test with no abortion cases or seronegative aborting cows, were advised to test purchased animals in the future. Results on high seroprevalence herds showed a reduction of seroprevalence from 38·2% to 24·4% in a 3-year period (Dijkstra et al. Reference Dijkstra, Bartels and Wouda2005).

Spain

In Spain, control programmes for neosporosis have been conducted on a voluntary basis through the livestock producer health groups (Agrupaciones de Defensa Sanitaria Ganadera, ADSG) and developed at a regional level. ADSG extended its reach from 2003–2004 mainly in the northern parts of the country (Galicia, Asturias and Santander), which have a high percentage of the Spanish dairy cattle population. Initially, efforts were focused on (1) establishing the epidemiological situation of the different regions and herds involved, to establish guidelines for herd monitoring, by means of serological analysis, (2) providing training programmes for veterinarians and farmers and (3) the control of all traded cattle. In addition to neosporosis, the Health Programme included also other relevant diseases of cattle that impact production, such as IBR, BVD and paratuberculosis. From an economical point of view, and as far neosporosis is concerned, the subsidy scheme consisted of: (a) support for the recruitment of veterinarians responsible for the programmes and (b) a reduction in the prices of laboratory tests. This was the basis for achieving a gradual reduction in the prevalence of N. caninum, by gradually removing seropositive cattle and replacing them with seronegative animals. As an example of the success of these programmes, individual prevalences have been reduced in cattle belonging to ADSG from 16·2% and 12·5% in 2003 to 8·5% and 9·2% in 2010 in the Spanish regions of Galicia and Asturias, respectively (Ortega-Mora et al. Reference Ortega-Mora, Aguado-Martínez, Rojo-Montejo and Alvarez-García2010). However, there are no data regarding any concomitant reduction in the number of N. caninum abortions.

Portugal

In Portugal, a privately funded programme promoted by farmers, veterinarians and some pharmaceutical companies was initiated in 2009 and involved more than 250 herds and 23 000 animals (Bovicontrol^®). At the starting point of the programme, more than 80% of the herds had seropositive cattle and within-herd seroprevalences were as high as 80%. This programme follows the same basic scheme described above for Spain and also includes neosporosis along with paratuberculosis, IBR and BVD (Niza Ribeiro et al. Reference Niza Ribeiro, Pereira, Afonso and Pacheco2011).

Belgium

In Belgium, the two livestock producer health groups (ARSIA; l'Association Régionale de Santé et d'Identification Animales and DGZ; Dierengezondheidszorg Vlaanderen) also include neosporosis together with IBR, BVD and paratuberculosis in their herd health schemes (Cantaloube, Reference Cantaloube2010). Support for the laboratory diagnosis of abortion events (including neosporosis) is given to farmers and veterinarians.

Switzerland

Finally, in a country such as Switzerland where bovine neosporosis has been a notifiable disease since 2001, no national control programme has been established so far. As Häsler and coworkers pointed out, resources for these initiatives are limited and it is necessary first to fully understand the population dynamics and economic consequences of control options to support the decision making process of policy makers (Häsler et al. Reference Häsler, Regula, Stärk, Sager, Gottstein and Reist2006).