Overview of bovine viral diarrhea viruses (BVDV) and aspects of diversity

BVDV are members of the genus Pestivirus within the family Flaviviridae, The genus is composed of four recognized species, BVDV1 and 2, classical swine fever virus and border disease virus (Becher and Thiel, Reference Becher, Thiel, Tidona and Darai2002). Recently another pestivirus in cattle has been described, HoBi-like viruses (Bauermann et al., Reference Bauermann, Ridpath, Weiblen and Flores2013b).

The diversity of BVDV includes both genetic and antigenic differences. The BVDV are classified as two species, BVDV1 and BVDV2 using phylogenetic studies (Pellerin et al., Reference Pellerin, van den Hurk, Lecomte and Tussen1994; Ridpath et al., Reference Ridpath, Bolin and Dubovi1994). Using phylogenetic analysis, there are subgenotypes within these two species, 15 within BVDV1 (Vilcek et al., Reference Vilcek, Paton, Durkovic, Strojny, Ibata, Moussa, Loitsch, Rossmanith, Vega, Scicluna and Paifi2001, Reference Vilcek, Durkovic, Kolesarova, Greiser-Wilke and Paton2004) and two within BVDV2 (Flores et al., Reference Flores, Ridpath, Weiblen, Vogel and Gil2002). Regions of the BVDV genomes have been used to detect genomic diversity including: 5’ UTR (Vilcek et al., 2014; Ridpath and Bolin, Reference Ridpath and Bolin1998; Fulton et al., Reference Fulton, Ridpath, Ore, Confer, Saliki, Burge and Payton2005b: Ridpath et al., Reference Ridpath, Fulton, Kirkland and Neil2010); E2 (van Rijn et al., Reference Van Rijn, van Gennip, Leendertse, Brusche, Paton, Moorman and van Oirschot1997); Npro (Becher et al., Reference Becher, Orlich, Shannon, Horner, Konig and Thiel1997); and NS3 (Ridpath and Bolin, Reference Ridpath and Bolin1997). Knowledge of the genetic diversion was applied to various BVDV strains, especially after the severe BVDV disease in Canada during 1993–1995 (Carman et al., Reference Carman, van Dreumel, Ridpath, Hazlett, Alves, Dubovi, Tremblay, Bolin, Godkin and Andersen1998). The use of the genetic information permitted the separation of the BVDV into BVDV1a, BVDV1b and BVDV2 with the serology testing with virus neutralization separating the BVDV into two groups, 1 and 3 (Pellerin et al., Reference Pellerin, van den Hurk, Lecomte and Tussen1994). That same year another report demonstrated two BVDV genotypic groups with antigenic differences (Ridpath et al., Reference Ridpath, Bolin and Dubovi1994). Additional reporting indicated that PCR could differentiate BVDV1a, BVDV1b and BVDV2 (Ridpath and Bolin, Reference Ridpath and Bolin1998). Genetic diversity results suggested that the BVDV strains could also have antigenic differences. Using BVDV monoclonal antibodies and polyclonal antibodies from cattle receiving modified live virus (MLV) or killed vaccines revealed antigenic differences as well (Pellerin et al., Reference Pellerin, van den Hurk, Lecomte and Tussen1994; Bolin et al., Reference Bolin, Moennig, Kelso Gourley and Ridpath1988; Bolin and Ridpath, Reference Bolin and Ridpath1989, Reference Bolin and Ridpath1990; Fulton et al., Reference Fulton, Burge, d'Offay and Payton1997a; Fulton et al., Reference Fulton, Saliki, Burge, d'Offay, Bolin, Maes, Baker and Frey1997b, Reference Fulton, Ridpath, Confer, Saliki, Burge and Payton2003a; Fulton and Burge, Reference Fulton and Burge2000; Ridpath et al., Reference Ridpath, Fulton, Kirkland and Neil2010).

The BVDV diversity of infections results in various clinical forms/syndromes (Baker, Reference Baker1995; Hessman et al., Reference Hessman, Sjeklocha, Fulton, Ridpath, Johnson and McElroy2012; Fulton, Reference Fulton2013). BVDV infections can be manifested by single-organ involvement and/or multiple organs and may interact with other agents to cause clinical illness/disease. BVDV has tropism for many organs, including the respiratory tract, digestive tract, lymphoid system, reproductive tract and fetus. Another form is a severe disease syndrome called mucosal disease (MD). Thus it may simplistic to classify BVDV infections into specific diseases, except for the persistently infected (PI) fetus (Fulton, Reference Fulton2013). The PI animal is important to the epidemiology of BVDV as they are constant shedders of virus from all body excretions and secretions their entire lifetime and are quite capable of infecting susceptible cattle (Fulton et al., Reference Fulton, Hessman, Ridpath, Johnson, Burge, Kapil, Braziel, Kautz and Reck2009b). The PI is also important as the MD in cattle results with the PI calf infected with an external cytopathic (CP) virus identical genetically or the PI strain mutating to a CP form resulting in the dual infection in MD animals (Baker, Reference Baker1995; Fulton, Reference Fulton2013). The BVDV viruses have the ability to cause immunosuppression in many types of infection (Table 1), including the inapparent form, both of the innate immune system and the acquired immune system (T-cell and B-cell) as reviewed (Fulton, Reference Fulton2013; Ridpath, Reference Ridpath2013).

Table 1. Disease forms with bovine viral diarrhea viruses – immunosuppression associated with many forms

Vaccination as component of control programs

Biosecurity measures, testing for and removal/isolation of PI cattle, and vaccinations are all part of the BVDV control programs. The goals of vaccination are to control the spread of an infection if the virus is introduced and to reduce the magnitude of clinical disease caused by the virus (Ridpath, Reference Ridpath2013). When looking at expectations for BVDV vaccination as stated in vaccine label claims, two major claims are made (Ridpath, Reference Ridpath2013): (1) prevent clinical disease in cattle exposed to BVDV; and (2) prevent fetal infection that leads to PI calves (Ridpath, Reference Ridpath2013). The regulations for licensure in the USA for protection against acute challenge and prevention of fetal infection are in the USDA APHIS Center for Veterinary Biologics (CVB) website. Requirements for licensure by the USDA along with label claims are often questions posed to animal health companies and the practicing veterinarian advising clients on vaccine usage and/or expectations. These are listed in Table 2, and there are likely others. While there are requirements for licensure and label claims, producers and veterinarians, particularly those involved in feedlot processing decisions, often conduct studies to determine if a particular product, vaccine, anthelmintic, antibiotic, feed additive, etc. would result in production improvement and increased income. They weigh the financial cost of preventative/treatment measure results in an increased economic return. Studies involving vaccines may compare one vaccine to another to determine if there is an economic advantage of one vaccine to another.

Table 2. Does the licensed vaccine meet expectations of animal owner and/or veterinarian?

Implications of BVDV genetic diversity and vaccination

Of the BVDV genotypes found throughout the world, there are three principal genotypes in the USA, BVDV1a, BVDV1b and BVDV2a (Ridpath et al., Reference Ridpath, Fulton, Kirkland and Neil2010). There is one US report for BVDV2b, a case in a fatal feedlot pneumonia case (Fulton et al., Reference Fulton, Blood, Panciera, Payton, Ridpath, Confer, Saliki, Burge, Welsh, Johnson and Reck2009a). In survey of diagnostic laboratory accessions in the USA, the predominant subgenotype was BVDV1b (45.8%) with BVDV1a (28.2%) and BVDV2a (26.0%) (Fulton et al., Reference Fulton, Ridpath, Ore, Confer, Saliki, Burge and Payton2005b). Also studies of PI cattle entering a commercial feedlot found the distribution of BVDV subgenotypes: BVDV1b (77.9%), BVDV1a (11.6%), and BVDV2a (10. 5%) (Fulton et al., Reference Fulton, Johnson, Briggs, Ridpath, Saliki, Confer, Burge, Step, Walker and Payton2006b). With the BVDV being heterologous in both genetic and antigenic patterns, the question becomes how does a BVDV provide protection to heterologous strains, and if so what is the expected rate of protection?

The first BVDV immunogens in US vaccines were the CP biotypes that cause visual cytopathology in cell culture. Although noncytopathic (NCP) BVDV strains were recognized, none were ‘thought’ to be vaccine immunogens. From a production standpoint, it was assumed the CP strains were homogeneous for CP strains and free of NCP strains. Thus the move to use the CP strains such as the BVDV1 CP strains, Singer, NADL and C24 in the vaccines, which were predominantly MLV. With the recognition that there was a heterologous response with a range of neutralizing antibodies, there was concern whether these vaccines gave complete protection. An outbreak of severe disease in Ontario, Canada in the 1993–1995 was caused by a new strain, which was characterized as BVDV2 (Carman et al., Reference Carman, van Dreumel, Ridpath, Hazlett, Alves, Dubovi, Tremblay, Bolin, Godkin and Andersen1998). Thus there was a move to add the BVDV2 to existing BVDV1a vaccines. The need for BVDV2 vaccine was illustrated in a report of where several BVDV1a MLV vaccine herds developed BVDV2-related disease (Van Campen et al., Reference Van Campen, Vorpal, Huzubazar, Edwards and Cavender2000). Currently, most US BVDV vaccines have both the BVDV1a and BVDV2a in either the MLV or killed vaccines (Tables 3 and 4). Currently, there are no BVDV1b vaccines being marketed in the USA (Compendium of Veterinary Products. (2010) 12th edition). There are numerous vaccines in use today. There are many issues that producers and veterinarians face when selecting and using these licensed vaccines in a variety of production units, whether they are for the breeding herd, stocker operations, or feedlot processing. These issues are listed in Table 2 and discussed below. What may apply for one management unit such as the feedlot may not apply to the breeding herd.

Table 3. MLV BVDV vaccines

Table 4. Inactivated (killed) viral BVDV vaccines

Onset of immunity after vaccination

The onset of immunity in vaccinated cattle is determined by challenge with virulent virus at intervals after vaccination. However, measurement of serum neutralizing antibodies is most often used to ‘measure’ vaccine response. The MLV have traditionally been thought to induce more rapid immunity and have been also used as one dose. In contrast, the killed (or inactivated) BVDV vaccines require two doses. Vaccine studies measuring serum antibodies in seronegative calves often have not taken samples as early as day 7 so the actual onset of immunity may not be known. In experimental studies under controlled conditions, MLV BVDV vaccines have induced detectable antibodies by day 14 after vaccination (Fulton et al., Reference Fulton, Saliki, Burge and Payton2003b). In other studies using killed BVDV vaccines, a limited number of calves developed BVDV antibodies by days 14 and 21 after one dose (Fulton et al., Reference Fulton, Confer, Burge, Perino, d'Offay, Payton and Mock1995; Fulton et al., Reference Fulton, Burge, d'Offay and Payton1997a, Reference Fulton, Saliki, Burge, d'Offay, Bolin, Maes, Baker and Freyb). In those two studies after revaccination at day 28, there was significant increase in neutralizing antibodies at days 35 and 42. In a later study, using a bivalent killed BVDV vaccine, seronegative calves receiving the killed product responded with serum-neutralizing antibodies after 1 day at day 21 (Platt et al., Reference Platt, Coutu, Meinert and Roth2008). Serum antibodies decline after vaccination. Measuring serum antibodies in calves receiving either killed or MLV BVDV vaccines demonstrated a significant decline in BVDV antibodies from days 42 to 140 (Fulton et al., Reference Fulton, Confer, Burge, Perino, d'Offay, Payton and Mock1995). Upon revaccination of the calves with the respective vaccine, there was a significant anamnestic response with increased serum antibodies.

To determine onset of immunity to BVDV2a and BVDV1b, seronegative calves in two studies were vaccinated 3, 5 or 7 days prior to challenge with a MLV vaccine containing BVDV2a and BVDV1a (Brock et al., Reference Brock, Widel, Walz and Walz2007; Palomares et al., Reference Palomares, Givens, Wright, Walz and Brock2012). In the BVDV2a challenge study, there was reduced shedding of virus, lack of leukopenia, viremia reduction, and reduced mortality as early as 3 days with complete protection in calves receiving MLV vaccine 5 or 7 days prior to challenge (Brock et al., Reference Brock, Widel, Walz and Walz2007). For the BVDV1b challenge study, vaccination 5 or 7 days prior to challenge was effective in protecting calves from developing clinical signs associated with BVDV infection and reduced viral shedding (Palomares et al., Reference Palomares, Marley, Givens and Gallardo2013). Thus, from this study, MLV vaccination provides rapid protection against virulent viruses in an acute challenge system. This protection afforded cattle against challenge shortly after vaccination is likely the result of the MLV vaccines enhancing BVDV innate immunity.

Concerns for BVDV vaccines

Potential hazards to use of MLV BVDV vaccines would be contamination with NCP BVDV if the bovine cells used for CP-BVDV production were cultured in NCP-BVDV contaminated fetal bovine serum (FBS) (Bolin and Ridpath, Reference Bolin and Ridpath1998; Bolin et al., Reference Bolin, Matthews and Ridpath1991). The FBS may have NCP BVDV strains and these potentially would be virulent when given to the animal and especially if the heifer/cow were pregnant resulting in fetal infection. BVDV contamination in vaccines includes those for human and bovine vaccines has been reported (Kreeft et al., Reference Kreeft, Greiser-Wilke, Moennig and Horzinek1990; Barkema et al., Reference Barkema, Bartels, van Wuijckhuise, Hesselink, Holzhauer, Weber, Franken, Kock, Bruschke and Zimmer2001; Bruschke et al., Reference Bruschke, Paal and Weerdmeester2001; Giangaspero et al., Reference Giangaspero, Vacirca, Harasawa, Buttner, Panuccio, De Giuli Morghen, Zanetti, Belloli and Verhulst2001; Studer et al., Reference Studer, Bertoni and Candrian2002; Falcone et al., Reference Falcone, Cordioli, Tarantino, Muscillo, Sala, La Rosa, Archetti, Marianelli, Lombard and Tollis2003; Balint et al., Reference Balint, Baule, Palfi and Belak2005; Palomares et al., Reference Palomares, Marley, Givens and Gallardo2013). These contaminated vaccines also resulted in disease (Barkema et al., Reference Barkema, Bartels, van Wuijckhuise, Hesselink, Holzhauer, Weber, Franken, Kock, Bruschke and Zimmer2001; Bruschke et al., Reference Bruschke, Paal and Weerdmeester2001; Falcone et al., Reference Falcone, Cordioli, Tarantino, Muscillo, Sala, La Rosa, Archetti, Marianelli, Lombard and Tollis2003; Palomares et al., Reference Palomares, Marley, Givens and Gallardo2013). One example of this hazard is a study where MLV vaccine contaminated with ncp BVDV was given to pregnant heifers resulting in PI fetuses in the vaccinated heifers and the vaccine virus (BVDV1) was found in unvaccinated control heifers (Palomares et al., Reference Palomares, Marley, Givens and Gallardo2013). This underscores the point that NCP strains, whether they are the vaccine strain or contaminating strains, pose a hazard when given to pregnant and susceptible cattle.

The MLV BVDV vaccines’ potential for transmission from vaccinates to susceptible contacts is a major concern. In two studies, susceptible calves in contact with calves receiving MLV vaccines did not become infected as measured by viral isolation of blood samples or by seroconversions (Fulton et al., Reference Fulton, Saliki, Burge and Payton2003b; Kleiboeker et al., Reference Kleiboeker, Lee, Jones and Estes2003). Blood was collected after vaccination for virus isolation and the virus was cleared by 10 days post-vaccination in the vaccinated in both studies.

There have been concerns for BVDV association with immunosuppression (Baker, Reference Baker1995; Fulton, Reference Fulton2013; Ridpath, Reference Ridpath2013). Studies on bovine respiratory disease in feedlot raised concern that MLV BVDV increased the mortality rate (Martin et al., Reference Martin, Meek, Davis, Thomson, Johnson, Lopez, Stephens, Curtis, Prescott, Rosendal, Savan, Zubaidy and Bolton1980, Reference Martin, Meek, Davic, Johnson and Curtis1981). One MLV vaccine given to calves caused a leucopenia and decreased neutrophils and lymphocytes. Peripheral blood taken from the vaccinated calves had a depressed lymphocyte blastogenesis response and neutrophil function (Roth and Kaberle, Reference Roth and Kaberle1983). Thus there is experimental evidence of MLV causing immunosuppression. The additive effect of administration of MLV BVDV in calves at stocker/feedlot processing enhancing bovine respiratory disease remains an issue.

MD is the severest form of BVDV and results in high mortality. With its pathogenesis of a PI animal superinfected with a CP strain identical to the NCP strain or the existing NCP PI strain, mutating to cause the fatal interaction, there has been concern for the use of MLV BVDV contributing to the MD. However, with many PI cattle receiving MLV vaccines in stocker and feedlot processing regimens, there appears to be little evidence of MD. In differential diagnosis it is important to remember that not only MD but also acute BVDV disease can cause extensive mucosal lesions with high morbidity and mortality occurring soon after arrival in the feedlot (Hessman et al., Reference Hessman, Sjeklocha, Fulton, Ridpath, Johnson and McElroy2012). To determine if MLV vaccines caused MD in PI calves, BVDV1b-NCP PI calves were vaccinated with MLV vaccines containing BVDV1a and BVDV2a or a vaccine containing a NCP BVDV1 strain (Fulton et al., Reference Fulton, Step, Ridpath, Saliki, Confer, Johnson, Briggs, Hawley, Burge and Payton2003c). The calves did not develop MD after vaccination. All the calves receiving MLV vaccines with BVDV1a and BVDV2a seroconverted to BVDV1a and BVDV2a after vaccination, but not to BVDV1b. Calves receiving the NCP BVDV1 vaccine did not seroconvert to BVDV1a, 1b, or BVDV2a. This suggests that under controlled conditions, the MLV BVDV did not cause MD in PI calves. However, a recent report described MD in cattle after vaccination with a MLV vaccine containing BVDV1a and BVDV2a (Miller et al., Reference Miller, O'Toole, Cavender, Cornish, Dawson, Smylie, Fox, Hill, Montgomery, Vasquez and Schumaker2013). The CP strains from affected animals matched the genomic regions of the vaccine strain, 125-A. It was not possible to rule out that the endogenous BVDV mutated or there was exposure to other field strains of 125-A; however, the temporal relationship of the vaccination and disease were suggestive of a causal relationship.

Challenge studies for efficacy using heterologous BVDV strains

The value of BVDV immunity for protection against clinical disease and enhance feedlot performance of beef calves was pointed out in two studies evaluating a retained ownership program for beef ranchers evaluating their animal health program for calves prior to delivery to the feedlot (Fulton et al., Reference Fulton, Cook, Step, Confer, Saliki, Burge, Welsh, Blood and Payton2002a, Reference Fulton, Cook, Blood, Confer, Payton, Step, Saliki, Burge and Welsh2011). The calves had received various MLV or killed BVDV vaccines with BVDV1a and/or BVDV1a and BVDV2a prior to delivery. Calves with lower BVDV1a and BVDV2a serum antibodies at feedlot entry had increased treatment costs, increased clinical illness, increased number of treatments and decreased net value to the owners. In one study, the calves with higher BVDV1a and BVDV2a antibodies had higher carcass grade (Fulton et al., Reference Fulton, Cook, Step, Confer, Saliki, Burge, Welsh, Blood and Payton2002a). Thus the importance of better immunity, including BVDV, would indicate that BVDV vaccination prior to feedlot delivery would be beneficial to the rancher.

The published studies on the efficacy of BVDV vaccine indicate that the challenge strain is usually not the same strain as in the vaccine. The MLV and killed BVDV vaccine strains listed in Tables 3 and 4 are rarely used as challenge viruses. Viruses used as challenges need to cause enough disease to be recognized clinically and to replicate in the host to detect viral infection and stimulate measurable serum antibodies. Historically, the bovine viral vaccines have been tested by an acute challenge after infection, usually by intranasal or aerosol to the respiratory tract, and often with a very high concentration of virus. With the discovery that PI cattle shed high titers of virus in all discharges and excretions, PI cattle have been used as the ‘natural’ challenge method since the BVDV exposure is maintained by the PI in the population due to the high titers of shed virus and the potential that PI cattle may survive for several months (Fulton et al., Reference Fulton, Johnson, Briggs, Ridpath, Saliki, Confer, Burge, Step, Walker and Payton2006b, Reference Fulton, Hessman, Ridpath, Johnson, Burge, Kapil, Braziel, Kautz and Reck2009b). Use of PI cattle to expose vaccinates and nonvaccinates is demonstrated by two studies: (1) Southeast US calves receiving vaccine prior to shipment along with nonvaccinates to an experimental feedlot (Fulton et al., Reference Fulton, Johnson, Briggs, Ridpath, Saliki, Confer, Burge, Step, Walker and Payton2006b); and (2) calves from a ranch with vaccinated calves and unvaccinated calves shipped to the experimental feedlot (Fulton et al., Reference Fulton, Briggs, Ridpath, Saliki, Confer, Payton, Duff, Step and Walker2005a). These studies both used PI calves to expose the arriving cattle to BVDV, and to challenge the immunity in those calves. The vaccinates and controls in both studies were evaluated over an approximate 35-day period after challenge by serology and viral isolation attempts on nasal swabs and peripheral blood collected weekly.

When evaluating the efficacy of BVDV vaccines, it is important to look for relative ability to protect the vaccinated animal against experimental challenge. In the acute challenge studies, all of the vaccinates should be protected from disease and remain free of virus by virus isolation. And in challenge studies for fetal protection studies, all of the fetuses from the vaccinated heifers should remain free of the infectious virus. The desire for 100% protection in fetal challenge studies is important as one remaining PI calf would still be important as a continual source for the shedding virus and exposing other cattle.

In reviewing the literature there are 26 challenge studies done in North America that I have focused on: 10 studies for acute challenge (Table 5) and 16 studies for fetal protection (Table 6). These summaries are not inclusive from the initial US licensure, but focus on the studies subsequent to the emphasis on BVDV2 occurring in BVDV1 vaccinated herds in the 1990s. Studies have shown the addition of BVDV2 immunogens to the BVDV1a vaccines provided greater protection than BVDV1a alone when vaccinated heifers were challenged with BVDV2. These summaries in Tables 5 and 6 include the type vaccine, MLV or killed strains in the vaccine, method of administration of the challenge virus, and outcomes in evaluating protection. In reviewing published studies in the USA, the results of acute challenge and fetal protection studies are described, including subgenotypes in vaccines and challenge strains, and the results are described in vaccinates and the vaccinates' fetuses/newborns. In general, vaccines provide protection against heterologous strains, ranging from 100% to partial but statistically significant protection. There has been a move toward using PI cattle to expose the vaccinates and controls both in acute challenges and for fetal protection studies. In recent years, there have been studies on the duration of immunity stimulated by vaccines. These studies are included in the cited articles in Tables 5 and 6.

Table 5. Acute challenge studies measuring efficacy in BVDV vaccinated cattle

Table 6. Challenge to determine BVDV vaccine efficacy against BVDV fetal disease and/or infection

Planning for the future

There are several issues facing the industry and regulating agencies for BVDV vaccines, many of which apply to other bovine vaccines. Issues of contamination, especially since fetal bovine serums may be contaminated with NCP BVDV or new agents remain. The USDA CVB has issued notices and memorandums regarding BVDV vaccines. The potential for immunosuppression by MLV vaccines exists and new vaccines will be assessed in the future to prove those MLV components are not immunosuppressive by experimental studies (USDA CVB Notice 14–06). The USDA CVB memorandum no. 800.212 addressed the licensing considerations on vaccine claims for protection of the fetus against BVDV including definition of types of claims and studies to be performed to support the claim: abortion, PI calves, and fetal infection. The USDA CVB notice no. 06–06 informs the licensees and applicants for guidance for label disclosure that the type of BVDV in the vaccine and the challenge strains be included in the label… The USDA CVB Veterinary Services memborandum no. 800.110 provides the guidance for exemptions to the label warning concerning the use of infectious bovine rhinotracheitis virus MLV in pregnant cows or in calves nursing pregnant cows. This would now apply to MLV BVDV vaccines.

As one looks at other issues for BVDV vaccines, the possibility of the entry of foreign BVDV or pestiviruses into the USA such as the HoBi-like viruses is a reality (Ståhl et al., Reference Ståhl, Beer, Schirrmeier, Hoffmann, Belák and Alenius2010). Serum containing polyclonal BVDV antibodies from cattle receiving killed and MLV BVDV has been shown to protect against the HoBi-like viruses (Bauermann et al., Reference Bauermann, Harmon, Flores, Falkenberg, Reecy and Ridpath2013a). A long-desired need for BVDV vaccines is a “differentiating infected from vaccinated animals” (DIVA) vaccine that could be used in control programs to differentiate antibody responses to natural infection from vaccination (Raue et al., Reference Raue, Harmeyer and Nanjiani2011).

In summary, vaccines are crucial to the control of BVDV in cattle populations along with biosecurity and effective testing to remove BVDV-infected animals, and BVDV vaccines are needed to control diverse genetic and antigenic strains that are either currently in the USA or might be introduced into the country. Challenge studies measuring vaccine efficacy are moving towards the PI animal, which is the most important means of transmission to susceptible cattle. Protection against pregnant females developing PI calves should be the hallmark of BVDV vaccine efficacy. Vigilance is required that new MLV BVDV vaccines with additional strains are not immunosuppressive and the MLV remain free of contaminating NCP strains. Ideally, progress should be made to develop vaccines that would permit differentiation of vaccinated animals from naturally infected animals.