In 2014, there was an increase in influenza activity in Europe, particularly in the U.K. and Ireland [20
]. Equine influenza was identified among horses and ponies on a total of 12 non-TB and seven TB premises in Ireland. Initial diagnosis was made on three non-TB premises in January and February. This followed an absence of any confirmed EI activity in Ireland in 2013. During October and early November, a further four non-TB premises were affected. On 23 November, clinical signs were observed and EI was diagnosed in a TB four-year-old gelding, which had been vaccinated in accordance with the Turf Club Rules (Premises 8). This horse was a new arrival on the Curragh in Co. Kildare where the majority of TB training yards and Ireland’s premier racetrack are located. It had recently travelled from a non-TB premises in the west of Ireland. A further six TB premises were affected within the coming weeks including four racing yards. A similar pattern of virus spillover from non-TB to TB populations was observed in Ireland during a large outbreak in 1989, which originated at the Royal Dublin Society horse show and spread extensively throughout the country, affecting several racing yards [21
The congregation of horses at any equestrian event is conducive to virus spread [4
], and in this outbreak, EI occurred on 11 of 18 premises following attendance at such events. The risk associated with failing to isolate horses following recent movement or the introduction of a new arrival with an unknown immunological status [1
] was evident in all the premises investigated in this study. Seven of the 11 premises were affected following the attendance of horses at a hunt meeting. Hunting has previously been implicated in EI virus spread [4
] and may contribute to the seasonal pattern of EI observed in this and previous outbreaks [16
]. The hunting “season” falls between November and April, and exposure of intrapulmonary airways to cold air [24
] coupled with the strenuous physical activity may increase the horses’ susceptibility to EI [25
], a situation exacerbated by the lack of a compulsory vaccination policy.
Of further concern is that EI occurred on three premises following the attendance of horses at a TB breeze up sale, an international horse show and a race meeting where up to date vaccination is required. During 2014 in Ireland, vaccine breakdown was identified in 30 horses vaccinated with all four commercially available products. This number was significantly higher than the three cases reported in the U.K. during the same period [8
]. Morbidity ranged from 17–100% on the 18 affected premises investigated. Analysis of passports on non-TB premises indicated that of 149 horses, only 13 (9%) had an up to date vaccination records and 59 (40%) developed clinical signs. The incidence of clinical signs among TB horses was 62/156 (40%), i.e., identical to that observed among the non-TB population even though 98/156 (63%) of TB horses included in this study had an up to date vaccination record. Virological protection in vaccinated horses has been demonstrated to correlate with the degree of antigenic relatedness between the vaccine strain and the challenge virus [9
]. FCL2 viruses have been circulating in Europe since 2003 [28
], and although FCL1 viruses were identified in Ireland in 2009 and 2010, only FCL2 viruses have been identified since [3
]. The current recommendation to include FCL2 and FCL1 representative viruses in EI vaccines was initially made by the OIE Expert Surveillance Panel in 2010 [29
]. No updated vaccine was available at the time of last vaccination for horses included in this study. An updated ProteqFlu (Merial) recombinant vaccine expressing the HA genes of FCL1 (A/eq/Ohio/03) and FCL2 (A/eq/Richmond/07) viruses became available in Ireland towards the end of 2014 [30
Analysis of HA sequence data indicated that all viruses identified during the Irish outbreaks belonged to FCL2. In Europe in recent years, two subpopulations of FCL2 viruses have been identified with amino acid substitutions at position 144 or 179 in the HAI gene [8
]. All viruses characterised in the current study contained A144V similar to viruses identified during previous Irish outbreaks [4
] and genetically distinct from an I179V group of EI viruses, which have circulated in Germany, Italy and in some parts of the U.K. [8
]. The biological significance of the substitutions merits investigation. No antigenic difference between these two FCL2 subgroups has been demonstrated using the HI assay and post-infection ferret antisera [8
]. However, this is in contrast to results observed using an egg-based virus neutralization test and equine antisera raised against the Argentinian lineage vaccine strain A/eq/La Plata/93 [31
Both HA and NA are important surface glycoproteins of influenza virus, but the selection of vaccine strains has traditionally focused only on the HA. Due to the significant increase in vaccine breakdown observed in the current study, sequence analysis of the NA gene was carried out. Similar to the results obtained for the HA gene, the NA phylogenetic tree grouped all recent Irish viruses’ within FCL2. Of note however, the NA of A/eq/Limerick/3/10 a FLC1 [3
] virus was grouped within FCL2 NA subgroup, indicating that reassortment had taken place between the clades. Such reassortment had been reported previously in the U.K. [1
Similar to previous findings, a correlation between age, time since last vaccination and number of vaccine doses received was also established [4
]. Maternal antibodies in foals born to seropositive mares generally persist for up to six months; however, in some cases, these may be more durable [32
]. In a comparative antibody study, which examined the antibody levels in different equine populations in Ireland, TB weanlings (age 6–10 months) were identified as the most susceptible [35
]. In this study, virus infection among TB weanlings in Ireland was identified for the first time (Premises 9, 10, 14). Vaccination of weanlings is currently not a requirement for attendance at TB sales irrespective of immunological status. Weanlings on two of the three premises had recently arrived from large TB sales in Ireland (Premises 9) and the U.K. (Premises 10). The index case on both premises was unvaccinated and seronegative on initial sampling.
Among vaccinated horses, infection occurred on average 239 ± 19.2 SE days (range 42–364 days), i.e., approximately eight months following last vaccination. This was observed in horses with an average age of 3.6 ± 0.37 SE years (range 1–8 years). The poor durability of the response of young horses to vaccination especially early in their vaccination career is well documented [36
]. Mathematical models based on data derived from experimental challenge studies, as well as field data indicate that six-month, rather than annual, booster vaccination reduces the risk of EI infection in young racehorses [39
]. Of 25 index cases identified in this study, seven had an up to date vaccination history.
Laboratory testing is necessary to confirm a clinical diagnosis of EI. The real-time RT-PCR primer probe-based assay was the primary method of diagnosis during this study [14
]. This assay was extensively used to detect EI in naïve horses in Australia in 2007 [40
]. The current study demonstrated its diagnostic effectiveness in seropositive partially-immune and regularly-vaccinated horses. Twenty five percent of RT-PCR positive nasopharyngeal swabs identified during this study were collected from horses with up to date vaccination records. Of 135 RT-PCR positive nasopharyngeal swabs, 125 (93%) had accompanying serum samples. On serological analysis, 22 (18%) horses were seronegative, and the 103 (82%) seropositive horses had a mean FCL2 SRH antibody level of 181 ± 4.4 SE mm2
. The maximum FCL2 antibody level detected in an RT-PCR positive horse was 297 mm2
. These high SRH values were not necessarily representative of the horses’ serological status pre-infection, but they indicate that RT-PCR is a sensitive diagnostic technique in seropositive horses.
Read et al. (2011) reported that when monitoring immunologically-naïve horses in Australia, virus was detected by RT-PCR from a single horse for a maximum period of 34 days [41
]. The maximum period for which virus was detected in two unvaccinated horses during the current study was 12 days (Premises 1 and 3). Both of these horses, a four-year-old gelding and a seven-year-old mare, had FCL2 SRH antibody levels >150 mm2
on initial sampling. On a single premises, virus was detected for a maximum period of 21 days (Premises 3). On four premises, virus was detected up to one week after the last seroconversion was identified (Premises 12, 14, 15, 16).
The serological examination of paired acute and convalescent sera for the diagnosis of EI is used by many laboratories as an adjunct test to RT-PCR/virus isolation. Hemagglutination inhibition assays are routinely used with detergent-treated viral antigens to improve assay sensitivity. However, SRH tests with untreated antigens have been shown to be more reproducible between laboratories, and a strong correlation between SRH antibody level and protection has been established [42
]. Thus, SRH is the serological test of choice for measuring vaccine efficacy. Few studies, however, have compared EI serological techniques during outbreaks [46
]. Similar to the findings by Morley et al. [46
], a good correlation between the HI and SRH test was observed in this study, although SRH demonstrated greater diagnostic sensitivity. The use of improved diagnostics greatly contributes to the detection of subclinical infection with 33 (11%) horses included in this study identified as subclinically affected. Twenty six subclinically-infected horses were identified by RT-PCR only, six by SRH only and one by RT-PCR, SRH and HI. Increased detection of subclinically-infected horses, which are shedding virus, is key to minimizing the risk of introducing EI to naïve populations through the international movement of horses.
Farm investigations carried out in the current study are a critical component to EI surveillance programmes. These investigations coupled with detailed examination of vaccination histories are an important component of the Department of Agriculture, Food and the Marine (DAFM)-funded EI surveillance programme in Ireland. They offer the opportunity to examine the pattern of disease spread among different equine populations in real time and provide data to the OIE Expert Surveillance Panel regarding the viruses circulating in the field and the effectiveness of current vaccines. This study also facilitated the evaluation of an RT-PCR diagnostic test for the detection of infection in vaccinated horses. This is pertinent as recently, the OIE together with the FEI and the International Federation of Horseracing Authorities (IFHA) has identified EI as one of six OIE listed diseases at risk of dissemination as a result of international competition horse movements. The availability of EI virus detection tests that are sufficiently sensitive to detect low quantities of virus shed by vaccinated competition horses is considered paramount for screening horses for freedom from infection and safeguarding not only the horses competing at an international event, but also the population of the host country.