West Nile Virus in Horses: Veterinary Reference

Overview and Taxonomy of West Nile Virus in Horses: Veterinary Reference

Taxonomic Classification and Virological Foundations

West Nile virus (WNV) is a positive-sense, single-stranded RNA virus belonging to the family Flaviviridae, genus Flavivirus, and is a member of the Japanese encephalitis serocomplex, which also includes St. Louis encephalitis virus (SLEV), Japanese encephalitis virus, Usutu virus, and Murray Valley encephalitis virus [1, 2, 25]. This serogroup is defined by significant antigenic cross-reactivity, a feature that profoundly complicates serological diagnosis in horses and other vertebrates [2, 4, 36]. The virion is icosahedral, enveloped, and approximately 50 nm in diameter, with a genome of roughly 11,000 nucleotides encoding a single polyprotein that is cleaved into three structural proteins (capsid, premembrane/membrane, and envelope) and seven nonstructural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, NS5) [14, 25, 27]. The envelope (E) protein is the primary target of neutralizing antibodies and is organized into three domains (DI, DII, and DIII), with domain III (EDIII) harboring virus-specific epitopes that are critical for serological differentiation among flaviviruses [8, 35]. The nonstructural protein NS1 has garnered particular attention as a candidate for DIVA (Differentiating Infected from Vaccinated Animals) assays, enabling surveillance programs to distinguish naturally infected horses from those vaccinated with whole-virus or subunit vaccines [12].

The molecular determinants of WNV virulence are distributed across multiple genomic regions, with mutations in the NS3 helicase, NS5 polymerase, and the envelope protein influencing neuroinvasiveness and neurovirulence in mammalian hosts [14, 27]. Notably, a single amino acid substitution in the NS3 gene (T249P) has been associated with enhanced virulence in American crows and may correlate with epizootic potential, though the relevance of such markers in equine pathogenesis remains incompletely characterized [14, 17]. The virus possesses all three cardinal neuropathogenic properties, neuroinvasiveness, neurotropism, and neurovirulence, enabling it to enter the central nervous system, infect neurons, and cause severe neurological disease [32, 34].

Phylogenetic Lineages and Global Distribution

WNV exhibits substantial genetic diversity, with at least eight distinct lineages recognized globally. Lineage 1 is the most widely distributed, encompassing strains responsible for the North American epizootic (introduced in 1999), European outbreaks, and isolates from the Middle East, Africa, and Australia (where the Kunjin subtype circulates) [23, 24, 27]. Lineage 2, historically considered less pathogenic and confined to sub-Saharan Africa and Madagascar, has emerged with alarming frequency in Europe since 2004, causing severe neuroinvasive disease in both horses and humans in Hungary, Austria, Greece, Serbia, Italy, and Germany [1, 6, 13, 17, 30]. The Central/Southern European cluster of lineage 2d has been responsible for recurrent equine outbreaks, and sequencing of Austrian and Serbian isolates has confirmed that these strains possess genotypic markers for neuroinvasiveness and neurovirulence, challenging the earlier assumption that lineage 2 viruses are inherently avirulent [6, 17, 30]. In the Indian subcontinent, lineage 5 viruses have been detected, while lineage 1a strains, associated with heightened virulence, have also been isolated occasionally [24]. The co-circulation of multiple lineages in the same geographic region, as observed in Italy and the Balkans, underscores the virus’s adaptive capacity and poses significant challenges for vaccine strain selection and diagnostic assay specificity [1, 5, 20].

The molecular epidemiology of WNV in the Americas reveals predominantly lineage 1 strains, with genomic surveillance of isolates from birds, mosquitoes, and mammals in New England indicating continued evolution and spatial structuring of viral populations driven by vector-host feeding preferences [23]. The emergence of lineage 2 in Europe, coupled with the northward expansion of endemic areas into Germany and Slovakia, highlights the role of climatic warming, altered precipitation patterns, and avian migration in facilitating viral range expansion [1, 13, 16, 17].

Transmission Dynamics and Vector Ecology

WNV is maintained in an enzootic cycle between ornithophilic mosquitoes, primarily of the genus Culex, and avian amplifying hosts [1, 10, 16]. In Europe, Culex pipiens (the common house mosquito) and Culex modestus serve as the principal vectors, with vector competence modulated by temperature, mosquito biotype, and intrinsic antiviral pathways [16, 17, 21]. North American vectors include Culex pipiens, Culex tarsalis, and Culex quinquefasciatus, while in Africa and the Middle East, species such as Culex univittatus and Culex neavei play dominant roles [21, 33]. The virus has occasionally been detected in Aedes, Anopheles, and Culiseta species, and, notably, in ixodid and argasid ticks in Eastern Europe and Russia, though the epidemiological significance of tick-borne transmission remains unclear [22, 29]. Hyalomma marginatum ticks have been shown to carry WNV sequences in the Danube Delta, raising concerns about alternative routes of maintenance and spillover [29].

Birds of the order Passeriformes, particularly corvids (crows, magpies, jays), are highly competent amplifying hosts, developing viremias sufficient to infect feeding mosquitoes [10, 14, 25]. Migratory birds serve as the primary long-distance dispersal mechanism, introducing WNV into naïve regions along flyways connecting Africa, Europe, and Asia [1, 15, 20, 22]. The Djoudj National Park in Senegal exemplifies the critical interface where Palaearctic migrants interact with local mosquito populations, facilitating viral exchange and subsequent northward re-introduction [21]. In Guatemala, serological evidence of WNV transmission in horses, coupled with concurrent circulation of SLEV and undifferentiated flaviviruses, illustrates the complexity of flavivirus ecology in tropical regions and the necessity for highly specific diagnostic algorithms [2].

Horses as Sentinels and Dead-End Hosts

Horses, along with humans, are considered incidental or dead-end hosts because the magnitude and duration of viremia following WNV infection are insufficient to infect feeding mosquitoes [1, 10, 25]. However, approximately 20–35% of infected horses develop clinical signs, ranging from mild febrile illness to severe neuroinvasive disease, making them exquisitely sensitive sentinels for WNV circulation [3, 7, 10, 19]. The utility of equine surveillance in a One Health framework has been repeatedly demonstrated: in Italy, passive surveillance of horses detected WNV activity 2–3 weeks before the first human neurological cases, enabling timely implementation of vector control and blood donation screening [9, 26, 31]. In Serbia, the national integrated surveillance program, encompassing serological testing of sentinel horses, mosquito trapping, and wild bird sampling, successfully identified high-risk zones during the record-setting 2018 season, with the majority of human cases preceded by signal detection in horses [11, 28]. Similarly, in Austria and Germany, horses were among the first indicators of autochthonous WNV establishment, preceding human cases by one to two transmission seasons [6, 13, 17].

The economic rationale for equine sentinel surveillance is compelling. The Lombardy region of Italy demonstrated that investing in integrated environmental and veterinary surveillance (including horses) saved an estimated 7.7 million EUR over five years compared to universal nucleic acid amplification testing of blood donors [9]. Furthermore, horses serve as valuable subjects for seroprevalence studies that reveal the spatial extent of viral circulation: in eastern Germany, WNV seroprevalence averaged 5.8%, with significantly higher rates in counties with previously registered infections [7]; in the Tuscany outbreak of 1998, the overall seroprevalence among horses in the affected wetland area reached 38%, confirming widespread exposure in the absence of prior immunity [20].

Clinical and Pathological Considerations in Equids

While a detailed clinical description falls outside the scope of this taxonomic and overview section, it is essential to note that WNV infection in horses can manifest as West Nile fever (self-limiting febrile illness) or West Nile neuroinvasive disease, characterized by ataxia, fasciculations, tremors, cranial nerve deficits, recumbency, and seizures [3, 6, 18, 30]. Neuropathological examination of euthanized horses reveals polioencephalomyelitis with neuronal necrosis, gliosis, and perivascular cuffing, predominantly affecting the brainstem, cerebellum, and spinal cord [6, 17]. The first confirmed equine cases in São Paulo, Brazil, exhibited weakness, seizures, and pelvic limb incoordination, with WNV RNA detected via nested multiplex RT-PCR from erythrocytes and central nervous system tissue [3]. In Australia, the 2011 outbreak of WNV lineage 1 (Kunjin subtype) in New South Wales affected over 300 horses, with ataxia being the most consistent clinical sign; serological diagnosis using an equine IgM ELISA proved more effective than virus detection due to the brevity of viremia [18].

Diagnostic and Serological Differentiation

Given the low-level, transient viremia in horses, serology forms the cornerstone of ante-mortem diagnosis [10, 19]. Enzyme-linked immunosorbent assays (ELISAs) for WNV-specific IgM are highly indicative of recent infection, as IgM does not cross the placenta and appears within days of infection [18, 19, 30]. IgG ELISAs, while more sensitive than virus neutralization tests (VNTs), suffer from cross-reactivity with other flaviviruses, including SLEV, USUV, TBEV, and JEV, a major challenge in regions where multiple flaviviruses co-circulate [4, 19, 35, 36]. The development of recombinant E proteins with point mutations in the conserved fusion loop domain (Equad ELISAs) has substantially improved the specificity of serological differentiation in horses, enabling discrimination between WNV, USUV, and TBEV antibodies without recourse to labor-intensive VNTs [4]. Similarly, multiplex immunoassays using microspheres coupled to recombinant EDIII antigens from WNV, JEV, and TBEV provide a powerful tool for veterinary reference laboratories, achieving high specificity and throughput [35]. The World Organisation for Animal Health (WOAH) and the European Centre for Disease Prevention and Control (ECDC) endorse the use of integrated surveillance approaches that combine entomological, avian, and equine data, with serological confirmation via VNT or plaque-reduction neutralization test as the gold standard in reference laboratories [1, 9, 34]. It is imperative that national reference laboratories harmonize their diagnostic protocols to account for cross-reactivity, especially as JEV continues its westward expansion into Europe [36] and as USUV establishes endemicity alongside WNV [4, 37].

Molecular Pathogenesis of West Nile Virus in the Equine Host

The molecular pathogenesis of West Nile virus (WNV) in the equine host represents a complex, multi-stage process that begins at the site of inoculation and culminates, in a subset of infected animals, in severe neuroinvasive disease. Understanding these mechanisms at a molecular level is critical for the development of targeted therapeutics, the refinement of diagnostic strategies, and the improvement of vaccine efficacy. Horses, along with humans, are considered the most clinically susceptible incidental hosts, yet they are dead-end hosts due to the low-magnitude and short-lived viremia that is insufficient to infect feeding mosquitoes [1, 10, 14]. This paradox, high clinical susceptibility coupled with low transmission potential, underscores the unique host-pathogen dynamics that define WNV infection in equids.

Initial Infection and Cellular Entry

The pathogenic cascade is initiated when an infected mosquito, primarily of the Culex genus, inoculates WNV into the dermis and epidermis of the horse during a blood meal [5, 16]. The virus, a positive-sense single-stranded RNA virus of the Flaviviridae family, must first overcome the physical barrier of the skin and the initial innate immune defenses [25]. The primary cellular targets at the inoculation site are keratinocytes and resident dendritic cells (Langerhans cells) [5]. The envelope (E) glycoprotein of WNV mediates viral attachment and fusion with the host cell membrane. This process is not random; the E protein interacts with specific cell surface receptors, including DC-SIGN (dendritic cell-specific ICAM-3 grabbing non-integrin), the integrin αvβ3, and the glycosaminoglycan heparan sulfate, which facilitate clathrin-mediated endocytosis [14, 25]. The low pH within the endosome triggers a conformational change in the E protein, leading to fusion of the viral and endosomal membranes and the release of the viral nucleocapsid into the cytoplasm, where translation and replication commence [25].

Viral Dissemination and the Role of the Lymphatic System

Following initial replication in keratinocytes and dendritic cells, the virus is transported to the draining lymph nodes. This trafficking is a critical step in the establishment of a systemic infection. Infected dendritic cells, acting as "Trojan horses," migrate from the skin to regional lymph nodes, where they present viral antigens and, critically, facilitate the infection of monocytes and macrophages [5, 32]. Within the lymph node, WNV undergoes a significant amplification phase. The virus replicates efficiently in these permissive immune cells, leading to a primary viremia that seeds peripheral visceral organs, including the spleen, kidneys, and liver [5, 10]. The magnitude of this primary viremia in horses is typically low (often below (10^3) plaque-forming units/mL) and transient, lasting only a few days, which is the fundamental reason why horses are considered dead-end hosts incapable of sustaining the enzootic cycle [10, 14, 25]. However, this brief viremic phase is sufficient to allow the virus to reach the central nervous system (CNS) in susceptible individuals.

Neuroinvasion: Breaching the Blood-Brain Barrier

The most consequential aspect of WNV pathogenesis in horses is its ability to invade the CNS and cause neuroinvasive disease, which manifests as meningitis, encephalitis, or acute flaccid myelitis [5, 6, 32]. The precise molecular mechanisms by which WNV crosses the blood-brain barrier (BBB) are multifaceted and likely involve several concurrent pathways. The BBB is a highly selective, semipermeable border of endothelial cells, pericytes, and astrocytic end-feet that protects the CNS from pathogens. WNV employs at least three distinct strategies to breach this barrier:

1. Direct Infection of Endothelial Cells: WNV can directly infect the microvascular endothelial cells that line the cerebral capillaries. This infection can lead to increased permeability of the BBB through the disruption of tight junction proteins (e.g., claudin, occludin, and ZO-1), allowing the virus to pass paracellularly into the brain parenchyma [14, 32].
2. "Trojan Horse" Mechanism: Infected immune cells, particularly monocytes and macrophages, which are naturally capable of crossing the BBB during immune surveillance, can carry the virus across the endothelium in a process known as transendothelial migration. Once inside the CNS, these infected cells release new virions, initiating infection of resident neural cells [5, 14].
3. Cytokine-Mediated Disruption: The systemic inflammatory response triggered by WNV infection leads to the release of pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6). These cytokines can directly increase the permeability of the BBB by modulating the expression of adhesion molecules and tight junction proteins on endothelial cells, creating a "leaky" barrier that facilitates viral entry [5, 32].

Neurovirulence and Cellular Tropism within the CNS

Once inside the CNS, WNV demonstrates a profound neurotropism, preferentially infecting neurons. The virus is highly neurovirulent, meaning it possesses the intrinsic ability to cause disease within neural tissue [32]. The primary targets are the neurons of the brainstem, cerebral cortex, thalamus, and, notably, the anterior horn cells of the spinal cord [32]. The infection of anterior horn motor neurons is the pathological correlate of the acute flaccid paralysis observed in both horses and humans, a syndrome that clinically resembles poliomyelitis [32, 34]. The molecular basis for this neuronal tropism is linked to the expression of specific entry co-factors on the neuronal surface, including the integrin αvβ3 and the laminin receptor, which facilitate viral attachment and entry [14].

The replication of WNV within neurons triggers a cascade of cytopathic effects. The virus hijacks the host cell's translational machinery, leading to the shutoff of host protein synthesis and the induction of apoptosis (programmed cell death) [14, 25]. Furthermore, the host's immune response, while attempting to clear the virus, contributes to neuronal damage. The infiltration of cytotoxic T lymphocytes (CD8+ T cells) into the CNS is essential for viral clearance, but the release of perforin and granzymes, as well as the activation of microglia (the resident immune cells of the brain), can cause collateral damage to uninfected neurons, a phenomenon known as immunopathology [5, 32]. The balance between protective immunity and immunopathology is a key determinant of clinical outcome.

Molecular Determinants of Virulence and Host Susceptibility

The severity of WNV disease in horses is not uniform and is influenced by both viral genetics and host factors. Phylogenetic analyses have identified distinct WNV lineages, with Lineage 1 (including the NY99 strain that entered North America) and Lineage 2 (which has caused significant outbreaks in Europe) being the most clinically relevant [1, 5, 6]. While both lineages can cause neuroinvasive disease, specific molecular determinants of virulence have been mapped to the viral genome. Key mutations in the envelope (E) protein and the non-structural (NS) proteins, particularly NS3 and NS5, have been associated with increased neuroinvasiveness and neurovirulence in animal models [14, 17]. For instance, a single amino acid substitution in the NS3 helicase (T249P) has been strongly linked to increased virulence in avian hosts, but its role in equine pathogenesis is less clear [14]. The NS5 protein, which functions as the viral RNA-dependent RNA polymerase, is also a critical determinant of host range and pathogenicity [14].

Host genetic factors also play a significant role. In humans, a mutation in the gene encoding the oligoadenylate synthetase (OAS) enzyme, part of the interferon-mediated antiviral response, has been associated with increased susceptibility to WNV neuroinvasive disease [5]. While a direct equine homologue has not been definitively identified, the principle that host genetic variation influences susceptibility is well established. Advanced age, immunosuppression, and concurrent infections are recognized risk factors for severe disease in horses, mirroring the situation in humans [5, 6]. The World Organisation for Animal Health (WOAH) recognizes WNV as a notifiable disease, underscoring its economic and health impact on the equine industry globally.

The Role of NS1 and Immune Evasion

The non-structural protein 1 (NS1) of WNV is a multifunctional glycoprotein that plays a pivotal role in both viral replication and immune evasion. NS1 is secreted from infected cells and can interact with the complement system, a key component of the innate immune response. By binding to complement factor H, WNV NS1 can downregulate complement activation, thereby reducing the opsonization and lysis of infected cells and viral particles [12, 14]. This immune evasion strategy allows the virus to replicate more efficiently and delay the host's clearance mechanisms. Furthermore, the presence of anti-NS1 antibodies in infected, but not vaccinated, horses has led to the development of DIVA (Differentiating Infected from Vaccinated Animals) assays, which are crucial for serological surveillance programs [12]. The detection of anti-NS1 antibodies provides a clear serological marker of natural infection, distinguishing it from the immunity induced by vaccines that primarily target the structural E protein [12].

Conclusion of Pathogenesis

In summary, the molecular pathogenesis of WNV in the equine host is a dynamic interplay between viral virulence factors and host immune defenses. The journey from a mosquito bite to a potentially fatal encephalitis involves a carefully orchestrated sequence of cellular entry, lymphatic dissemination, systemic viremia, and ultimately, neuroinvasion. The virus's ability to hijack host cell machinery, evade innate immune responses through proteins like NS1, and directly infect and destroy neurons, particularly in the spinal cord, explains the severe neurological signs observed in clinical cases. The low viremia in horses, a defining feature of their role as dead-end hosts, paradoxically highlights the extreme neurovirulence of the pathogen, as even a limited systemic infection can have devastating CNS consequences. Understanding these molecular events is not merely an academic exercise; it is the foundation upon which effective vaccines, antiviral therapies, and diagnostic strategies are built, and it is essential for the continued surveillance and control of this globally significant arbovirus.

Clinical Manifestations and Differential Diagnosis of West Nile Virus Infection in Horses

West Nile virus (WNV) infection in horses presents a complex clinical spectrum ranging from subclinical seroconversion to severe, life-threatening neuroinvasive disease. Understanding this spectrum is paramount for clinicians, as the clinical presentation is neither pathognomonic nor uniformly predictable, necessitating a rigorous and systematic approach to differential diagnosis. The horse serves as a critical sentinel species for WNV circulation within a One Health framework, and the clinical recognition of equine cases often precedes the identification of human cases in a given region [1, 6, 9, 11, 17, 26]. The World Organisation for Animal Health (WOAH) recognizes WNV as a notifiable disease, underscoring its significance in both veterinary and public health contexts.

The Clinical Spectrum: From Subclinical Infection to Severe Neuroinvasive Disease

The vast majority of WNV infections in horses are subclinical. Epidemiological studies consistently demonstrate that only a minority of infected horses develop overt clinical signs, with estimates suggesting that approximately 20-35% of seropositive animals exhibit clinical disease [3, 10, 20]. This ratio is influenced by viral lineage, host factors such as age and immune status, and environmental conditions that affect viral dose and vector exposure [14, 25]. The incubation period in horses is typically 3 to 15 days following the bite of an infected mosquito, though this can vary [10, 25].

Subclinical Infection and Seroconversion: The majority of infected horses mount a successful immune response without ever displaying clinical signs. This is evidenced by seroprevalence surveys that detect WNV-specific antibodies in healthy, unvaccinated populations. For instance, a study in Eastern Germany in 2020 found a WNV seroprevalence of 5.8% in horses from counties with known viral circulation, with the vast majority of these animals being clinically normal at the time of sampling [7]. Similarly, a serosurvey in the Tuscany region of Italy following the 1998 outbreak revealed an overall seroprevalence of 38% in the affected area, yet only 14 clinical neurologic cases were identified among the population, yielding an attack rate of just 2.8% [20]. These data highlight the remarkable capacity of the equine immune system to control WNV infection without clinical consequence in most instances.

Febrile and Non-Neurologic Syndromes: Some horses may develop a transient, non-specific febrile illness that is often overlooked or misattributed. This syndrome can include pyrexia, lethargy, anorexia, and mild depression [10, 25]. These signs are indistinguishable from a multitude of other equine infections and are rarely diagnosed as WNV in the field unless part of an active surveillance program. The viremia in horses is characteristically low-level and short-lived, typically lasting only 1 to 4 days, which further complicates molecular diagnosis during this phase [10, 19, 25]. This brief viremic window is a key reason why serology, particularly IgM detection, is the cornerstone of antemortem diagnosis [6, 18, 19, 30].

Neuroinvasive Disease: The Hallmark of WNV Infection: The most clinically significant and recognizable manifestation of WNV in horses is neuroinvasive disease, which can present as meningitis, encephalitis, meningoencephalitis, or acute flaccid paralysis [5, 6, 10, 25, 32]. The onset can be acute or peracute, and the clinical signs reflect the distribution of viral lesions within the central nervous system (CNS). WNV is a highly neurotropic flavivirus, possessing the ability to invade the CNS, infect neurons, and cause direct neuronal necrosis and inflammation [14, 25, 32].

The clinical signs of WNV neuroinvasive disease in horses are diverse and can be grouped into several overlapping categories:

1. Gait Abnormalities and Ataxia: Ataxia is the single most consistently reported clinical sign in confirmed equine WNV cases [6, 18, 30]. It is often described as a general proprioceptive deficit, leading to a wide-based stance, swaying, stumbling, knuckling of the fetlocks, and dysmetria (hypermetria or hypometria). In the first seven confirmed cases of WNV neuroinvasive disease in Austria, all horses exhibited gait abnormalities [6]. Similarly, during the 2011 outbreak in New South Wales, Australia, ataxia was the only sign consistently described in all laboratory-confirmed cases [18]. The severity can range from subtle incoordination to recumbency.

2. Muscle Fasciculations and Tremors: Fine to coarse muscle fasciculations, particularly of the muzzle, lips, neck, shoulders, and triceps, are a highly characteristic, though not pathognomonic, sign of WNV encephalitis [6, 30]. In the Austrian case series, six of seven horses exhibited fasciculations and/or tremors, with video recordings documenting these movements [6]. These tremors are thought to result from irritation of the lower motor neurons or their connections within the brainstem and spinal cord.

3. Altered Mental Status and Behavior: Encephalitis manifests as a range of behavioral changes, including depression, lethargy, somnolence, disorientation, and a dull, unresponsive demeanor [6, 30]. Conversely, some horses may exhibit hyperexcitability, agitation, head pressing, circling, and apparent blindness. The first clinical case of equine neuroinvasive WNV in Serbia was described with disorientation and loss of equilibrium [30].

4. Paresis and Paralysis: Weakness, often asymmetrical, is a common finding. This can progress to pelvic limb paresis, tetraplegia, or acute flaccid paralysis [5, 32]. The flaccid paralysis observed in WNV infection is analogous to poliomyelitis, resulting from viral destruction of the anterior horn cells of the spinal cord [32]. This can lead to recumbency, which carries a grave prognosis. In the Austrian study, the severity of clinical signs led to euthanasia in four of seven affected horses [6].

5. Cranial Nerve Deficits: Involvement of the brainstem can lead to a variety of cranial nerve deficits. These include facial nerve paralysis (drooping lip, ear, or eyelid), dysphagia, tongue weakness, and an inability to swallow [6]. Three of the seven horses in the Austrian series showed evidence of cranial nerve involvement [6]. These signs can be particularly dangerous as they may lead to aspiration pneumonia.

6. Seizures and Recumbency: In severe cases, horses may develop generalized or partial seizures [3]. Recumbency, especially when the horse is unable to rise, is a terminal event in many cases and is strongly associated with a poor prognosis and the need for euthanasia [3, 6, 20].

Prognosis and Sequelae: The prognosis for horses with WNV neuroinvasive disease is guarded to poor. Mortality rates in clinical cases can range from 20% to 40%, with euthanasia being a common outcome due to the severity of signs and poor response to supportive care [6, 10, 20]. However, horses that survive the acute phase often show remarkable recovery. Clinical improvement can be seen within days to weeks, with many horses returning to full function [6, 30]. The Belgian sports mare described in the first Serbian case improved clinically within two weeks [30]. Nevertheless, some horses may be left with residual neurological deficits, such as persistent ataxia, weakness, or behavioral changes, which can affect their athletic or working capacity [5, 10]. The literature on long-term sequelae in horses is sparse compared to human medicine, where persistent neurological and functional deficits are well-documented [5, 10]. Future studies are needed to fully characterize the post-infection period in horses [3].

Differential Diagnosis: A Systematic Approach

The clinical signs of WNV neuroinvasive disease are not unique and overlap significantly with a broad range of other neurological conditions affecting horses. A thorough and systematic differential diagnosis is essential to avoid misdiagnosis and to ensure appropriate management and reporting. The differential list must be prioritized based on geographic location, seasonality, vaccination history, and the specific constellation of clinical signs. The following conditions must be considered:

1. Other Viral Encephalitides:

• Rabies: This is the most critical differential diagnosis for any case of acute, progressive encephalitis in a horse, particularly if there is a history of a bite wound or exposure to wildlife. Rabies can present with similar signs of ataxia, altered mentation, pharyngeal paralysis, and hyperesthesia. Unlike WNV, rabies is almost invariably fatal, and the aggressive form can pose a significant zoonotic risk. Any horse with acute, rapidly progressive neurological signs should be handled with extreme caution, and rabies must be ruled out via postmortem examination of brain tissue. WNV and rabies testing are often performed concurrently on CNS samples [3].
• Eastern, Western, and Venezuelan Equine Encephalitis (EEE, WEE, VEE): These alphaviruses are endemic in the Americas and cause severe, often fatal, encephalitis in horses. The clinical signs are virtually indistinguishable from WNV, including fever, ataxia, depression, hyperexcitability, and recumbency. EEE, in particular, has a very high mortality rate (often >90% in unvaccinated horses). Geographic distribution and vaccination history are key differentiating factors. In Brazil, the differential diagnosis of arboviral encephalitis must include these viruses alongside WNV [40, 41].
• Equine Herpesvirus Myeloencephalopathy (EHM): Caused by equine herpesvirus type 1 (EHV-1), EHM is a major differential for acute-onset ataxia and paresis, often with urinary incontinence and tail hypotonia. EHM is frequently associated with a history of fever, respiratory disease, or abortion in the herd. It is not a mosquito-borne disease and can occur in outbreaks. PCR testing of nasal swabs and blood can differentiate EHM from WNV [3].
• Tick-Borne Encephalitis (TBE): In Europe and parts of Asia, TBE virus, another flavivirus, can cause neurological disease in horses. The clinical signs are similar to WNV, and serological cross-reactivity between flaviviruses (WNV, TBEV, Usutu virus) is a major diagnostic challenge [4, 35, 36]. The geographic distribution of TBE is more focal and associated with tick habitats, unlike the mosquito-borne transmission of WNV.
• Usutu Virus (USUV): This flavivirus is closely related to WNV and co-circulates in many parts of Europe [4, 37]. USUV can cause neurological disease in birds and has been implicated in rare human cases, but its clinical significance in horses is still being elucidated. Serological differentiation from WNV is difficult due to cross-reactivity, requiring advanced assays such as virus neutralization tests (VNT) or mutant E protein ELISAs [4, 37].
• Other Arboviruses: Depending on the geographic region, other arboviruses such as Japanese encephalitis virus (JEV) in Asia [35, 36], Rift Valley fever virus in Africa [38], and Shuni virus in South Africa [39] must be considered.

2. Non-Infectious Neurological Conditions:

• Cervical Vertebral Stenotic Myelopathy (CVSM or "Wobblers"): This is a common cause of ataxia in young, rapidly growing horses, particularly Thoroughbreds. The ataxia is typically symmetric and progressive, without the acute onset, fever, or altered mentation seen in WNV. Radiography and myelography can confirm the diagnosis.
• Equine Protozoal Myeloencephalitis (EPM): Caused by Sarcocystis neurona (in the Americas) or Neospora hughesi, EPM is a common cause of asymmetric ataxia and focal muscle atrophy. The onset is usually insidious, and the disease is progressive over weeks to months. CSF analysis for antibodies against S. neurona is a key diagnostic test.
• Trauma: Spinal cord or head trauma can cause acute ataxia, paresis, or altered mentation. A history of a fall, transport accident, or other injury, along with physical examination findings (e.g., swelling, pain, abrasions), can help differentiate trauma from WNV.

3. Metabolic and Toxic Conditions:

• Hepatic Encephalopathy: Liver failure from any cause (e.g., Theiler's disease, ragwort poisoning) can lead to depression, ataxia, head pressing, and seizures. Icterus and elevated liver enzymes are key differentiating features.
• Botulism: Caused by ingestion of Clostridium botulinum toxin, botulism presents as a progressive, flaccid paralysis, often beginning with dysphagia, muscle tremors, and weakness. It is afebrile and does not cause altered mentation. The progression is typically over days.
• Organophosphate or Ivermectin Toxicity: Certain toxins can cause neurological signs. Organophosphate toxicity can cause salivation, muscle fasciculations, and ataxia. Ivermectin toxicity, particularly in certain breeds (e.g., Collies, but also reported in horses), can cause depression, ataxia, and blindness.

Diagnostic Challenges and the Role of Serology

The definitive diagnosis of WNV infection in horses relies on laboratory confirmation, as clinical signs alone are insufficient. The diagnostic approach is heavily dependent on the stage of infection.

Molecular Detection (RT-PCR): Detection of WNV RNA by reverse transcription-polymerase chain reaction (RT-PCR) is the method of choice for confirming acute infection. However, its sensitivity is limited by the short, low-level viremia in horses [3, 10, 19, 25]. Whole blood, serum, or CSF can be tested, but a negative RT-PCR result does not rule out WNV infection, especially if the sample is collected more than a few days after the onset of clinical signs [30]. In postmortem cases, RT-PCR on brain tissue, particularly the brainstem and spinal cord, has high sensitivity [3, 6, 18]. Virus isolation is possible but is less commonly used due to the requirement for BSL-3 facilities and the lower sensitivity compared to RT-PCR [10, 20].

Serological Diagnosis: Serology is the primary tool for antemortem diagnosis of WNV in horses [19].

• IgM Detection: The detection of WNV-specific IgM antibodies in serum or CSF is indicative of recent or acute infection. IgM antibodies appear within a few days of infection and persist for 1-2 months. Commercial IgM capture ELISAs are widely used and have become the standard for confirming acute clinical cases [6, 18, 19, 30]. The presence of IgM in the CSF is particularly strong evidence of neuroinvasive disease.
• IgG Detection: IgG antibodies appear later and persist for years, indicating past infection or vaccination. IgG ELISAs are highly sensitive but suffer from poor specificity due to cross-reactivity with other flaviviruses (

Transmission Dynamics and Vector Ecology of West Nile Virus in Equine Populations

The Enzootic Transmission Cycle: Avian Reservoirs and Mosquito Vectors

West Nile virus (WNV) is maintained in a complex enzootic cycle primarily involving ornithophilic mosquito vectors and avian reservoir hosts, a paradigm that is central to understanding its transmission to equine populations [1, 5, 14, 25]. The virus circulates in a bird–mosquito–bird transmission cycle, with wild birds serving as the principal amplifying hosts that develop sufficient viremia to infect feeding mosquitoes [15, 21, 34]. It is within this sylvatic cycle that the virus perpetuates, and it is only when the enzootic cycle spills over into incidental hosts, such as horses and humans, that clinically apparent disease occurs [10, 14, 22]. The mosquito genera Culex, particularly species such as Culex pipiens, Culex tritaeniorhynchus, Culex neavei, and Culex univittatus, are recognized as the primary vectors globally, exhibiting pronounced ornithophilic feeding preferences that facilitate efficient amplification within avian populations [16, 21, 33]. These mosquito species are not merely passive conduits; their vector competence, the intrinsic ability to become infected, support viral replication, and transmit the virus to a susceptible host, is a critical determinant of transmission intensity and geographic spread [14, 16].

The biological mechanisms underlying vector competence are multifaceted. Following ingestion of a viremic blood meal, the virus must first infect and replicate within the mosquito midgut epithelial cells, a process that constitutes a significant physical barrier [14, 16]. The virus must then disseminate from the midgut to the hemocoel and subsequently to the salivary glands, where high-titer replication ensures transmission during subsequent feeding bouts [14]. The extrinsic incubation period (EIP), the time from ingestion of an infectious blood meal until the mosquito becomes capable of transmission, is exquisitely sensitive to ambient temperature. Warmer temperatures, a hallmark of climate change, accelerate viral replication within the mosquito, shorten the EIP, and thereby amplify transmission potential [5, 13, 15, 16]. Indeed, the exceptionally warm summer of 2018 in Germany was directly implicated in the first autochthonous WNV cases in birds and horses, demonstrating how climatic anomalies can facilitate the establishment of transmission cycles in previously non-endemic regions [13]. The ability of WNV to adapt to different mosquito species and biotypes also plays a role; studies in Europe have documented that different Culex pipiens biotypes exhibit variable vector competence, and that the origin of the mosquito population can influence infection and transmission rates [16]. This variation has profound implications for predicting outbreak risk across heterogeneous landscapes.

Vector Feeding Behavior and Bridging Vectors

The transmission of WNV from its enzootic avian cycle to horses is contingent upon the existence of "bridging vectors", mosquito species that feed on both birds and mammals, thereby facilitating spillover [1, 21]. While strictly ornithophilic Culex mosquitoes are highly efficient at amplifying the virus among birds, their reluctance to feed on mammals limits their direct involvement in equine infections. However, many Culex species exhibit opportunistic feeding behavior, shifting from avian to mammalian hosts as bird availability declines or as environmental conditions change [21, 23]. Studies in Senegal using host-baited traps demonstrated that Culex neavei was the most effective bridging vector, with significant attraction to both pigeon and horse baits, and that this attractiveness varied temporally, likely in response to migratory bird densities [21]. Similarly, Culex tritaeniorhynchus was identified as a key vector in the same region, though its activity was constrained by nightly temperatures above 20°C [21]. In North America, Culex pipiens and Culex restuans are primary enzootic vectors, but Culex salinarius and Aedes species may serve as important bridge vectors in certain ecological contexts [25, 33]. The implications for equine risk are clear: the composition, relative abundance, and feeding plasticity of local mosquito communities directly govern the probability of WNV spillover into horse populations.

Ecological and Environmental Drivers of Transmission

WNV transmission is not a static phenomenon; it is profoundly influenced by a complex interplay of ecological, environmental, and climatic factors that shape vector populations, viral amplification, and host exposure [1, 5, 27]. Landscape features such as wetlands, riparian zones, and agricultural areas provide ideal breeding habitats for Culex mosquitoes, while the presence of avian reservoirs, particularly corvids, passerines, and waterfowl, determines the local amplification potential [1, 15, 20]. The role of migratory birds in the long-distance transport of WNV cannot be overstated. These birds act as biological vehicles, introducing the virus into new geographic regions along established flyways, and then serving as local reservoirs upon arrival at breeding or wintering sites [1, 15, 21, 22]. The Danube Delta in Romania, for instance, represents a critical hub where migratory birds from Africa and Europe converge, facilitating viral exchange and amplification in a region with abundant Culex mosquito populations [29]. Similarly, the Atlantic Forest remnants in São Paulo, Brazil, and the migration routes through that region have been implicated in the introduction and maintenance of WNV, as evidenced by the first confirmed equine cases in the state [3].

Climatic variability, particularly temperature and precipitation, directly modulates transmission dynamics. Warmer summers accelerate mosquito development, increase biting rates, and shorten the EIP, while altered rainfall patterns can create or destroy mosquito breeding sites [1, 5, 13]. Drought conditions, paradoxically, can increase WNV transmission by concentrating birds and mosquitoes around limited water sources, thereby enhancing contact rates [5]. In Europe, the recurrent seasonal transmission patterns observed in countries such as Italy, Serbia, Greece, and Austria are closely linked to summer temperatures and the availability of suitable vector habitats [1, 6, 11, 20]. The 2018 outbreak in Serbia, which was the most intense ever recorded in the country, illustrated how favorable climatic conditions can synergize with established enzootic cycles to produce explosive epidemics in both horses and humans [11]. Environmental modification through urbanization also creates novel transmission foci; Culex pipiens thrives in urban environments, breeding in storm drains, catch basins, and artificial containers, bringing the virus into close proximity with horse stables and human residences [5, 33, 34].

Transmission Dynamics in Equine Populations: Sentinel Role and Dead-End Host Status

Horses occupy a unique and critical position in the epidemiology of WNV. They are highly susceptible to infection and, importantly, develop neuroinvasive disease at a rate significantly higher than many other mammals, with approximately 35% of infected horses developing clinical signs [3, 10]. This exquisite sensitivity makes them exceptionally valuable sentinels for WNV activity, often serving as an early warning system for impending human outbreaks [1, 6, 9, 10]. Integrated surveillance systems across Europe, including those in Italy, Serbia, and Austria, have repeatedly demonstrated that seroconversion in sentinel horses or the occurrence of equine clinical cases precedes human cases by two to three weeks, allowing public health authorities to implement vector control and blood donation screening measures proactively [6, 9, 11, 17, 26]. The cost-effectiveness of this approach is well-documented; in the Lombardy region of Italy, the implementation of an integrated veterinary and entomological surveillance program, which included sentinel horses, resulted in savings of 7.7 million EUR over five years by reducing the need for universal blood donation nucleic acid amplification testing [9].

Despite their susceptibility to disease, horses are true dead-end hosts for WNV [10, 14, 25, 34]. The viremia that develops in an infected horse is of low magnitude and short duration, insufficient to infect feeding mosquitoes and perpetuate the transmission cycle [10, 14, 34]. This is a critical distinction from avian reservoirs, which develop high-titer viremia that can efficiently infect naive vectors. Numerous experimental and field studies have confirmed this phenomenon; even in horses with severe neuroinvasive disease, detectable WNV RNA in blood is often fleeting or absent by the time clinical signs manifest, as was noted in the first equine case in Serbia where RT-qPCR on a blood sample collected seven days post-onset was negative [30]. This low-level viremia means that equine-to-equine or equine-to-human transmission via mosquitoes is effectively impossible. However, direct transmission to humans through necropsy or other contact with infected tissues remains a recognized occupational risk, particularly for veterinarians and laboratory workers [32, 39]. A documented case of zoonotic transmission occurred in a veterinary student who performed a necropsy on an infected horse in South Africa, underscoring the need for strict biosafety precautions when handling equine neural or visceral tissues [39].

Risk Factors for Equine Infection

Understanding the specific factors that predispose horses to WNV infection is essential for designing targeted prevention strategies. Seroprevalence studies and risk factor analyses have identified several key determinants. Geographic location is paramount; horses residing in areas with documented WNV transmission in birds, mosquitoes, or other horses have significantly higher odds of seropositivity [7, 20]. In a study in Eastern Germany, seroprevalence was 5.8% overall but was significantly higher in counties with previously registered equine infections [7]. Housing and management practices profoundly influence exposure risk. Horses kept on 24-hour turnout (pasture access) had a significantly higher risk of seropositivity compared to those stabled during peak mosquito activity [7]. The presence of outdoor shelters, while providing relief from weather, also correlated with increased risk, likely because these structures do not prevent mosquito access [7]. Conversely, stabling horses during dawn and dusk, the peak feeding times for Culex mosquitoes, can substantially reduce biting exposure. Breed type has also emerged as a risk factor, with ponies showing higher seroprevalence than other breeds in German studies, though the biological mechanism for this remains speculative [7]. Age plays a role, although the pattern varies by region; in an Italian outbreak in Tuscany, no significant age-specific differences in seroprevalence were observed, suggesting that the horse population was immunologically naïve to WNV prior to the introduction event [20]. However, in endemic areas, older horses may accumulate exposure over time, leading to higher seroprevalence in older age cohorts [7].

Owner knowledge and preventive practices represent a modifiable risk domain, yet significant gaps persist. A cross-sectional study of 227 horse owners in Romania revealed that while many had heard of WNV, fewer correctly identified mosquitoes as the primary route of infection, and awareness of the equine vaccine was low, with only a small proportion reporting vaccination [42]. Environmental measures, such as removing standing water to eliminate mosquito breeding sites, were less commonly implemented than the use of insecticides or repellents [42]. This disconnect between awareness and effective action highlights a critical need for targeted educational interventions and structured veterinary communication in endemic areas [42]. The role of the veterinarian as a trusted advisor is central, yet only a minority of owners reported discussing WNV prevention with their veterinarian, indicating a missed opportunity for risk mitigation [42].

Implications for Surveillance and Public Health

The transmission dynamics of WNV in equine populations have direct and profound implications for public health surveillance and response. The World Organisation for Animal Health (WOAH) and the World Health Organization (WHO) recognize horses as key sentinel species within a One Health framework, given their ability to signal viral circulation before human cases emerge [1, 9, 11, 17]. Integrated surveillance programs that combine entomological monitoring, avian mortality surveillance, and equine serological testing have proven highly effective in providing early warning and guiding public health interventions, including mosquito control and blood donor deferral [9, 11, 17, 26]. The serological tools available for equine diagnosis, including IgM capture ELISAs and virus neutralization tests, are robust and have been standardized across European national reference laboratories, though cross-reactivity with other flaviviruses such as Usutu virus (USUV) and tick-borne encephalitis virus (TBEV) remains a diagnostic challenge that requires careful interpretation and confirmatory testing [4, 19, 35]. The development of DIVA (Differentiating Infected from Vaccinated Animals) assays, based on detection of antibodies to the nonstructural NS1 protein, offers promise for distinguishing naturally infected horses from vaccinated ones, thereby improving the accuracy of surveillance data in areas where vaccination is practiced [12]. Ultimately, the intricate dance between mosquito vectors, avian reservoirs, environmental conditions, and equine hosts dictates the risk of WNV spillover. A deep, mechanistic understanding of these transmission dynamics is not merely an academic exercise; it is the foundation upon which effective, evidence-based prevention and control strategies for both equine and human health are built [1, 5, 25, 34].

Epidemiology of West Nile Virus in Horses: Global and European Perspectives with Emphasis on Sentinel Surveillance

Global Distribution and Emergence Patterns

West Nile virus (WNV) has undergone a dramatic global expansion over the past three decades, transitioning from a relatively obscure pathogen of the Old World to the most widely distributed arbovirus on the planet, with a geographic range now encompassing Africa, Europe, Asia, Australia, and the Americas [5, 25, 34]. The virus was first isolated from a febrile woman in the West Nile District of Uganda in 1937, and for over sixty years, it was considered a minor cause of sporadic febrile illness and encephalitis primarily in Africa and the Middle East [10, 27]. However, a series of epizootic events, beginning notably with an outbreak among horses in the Camargue region of France in the 1960s and accelerating dramatically after the 1999 introduction of WNV into New York City, fundamentally altered the perception of this pathogen’s threat to both animal and human health [25, 43]. The North American invasion, which likely originated from a lineage 1a strain from the Middle East, resulted in the largest epidemic of neuroinvasive WNV disease ever documented, with profound consequences for wild bird populations and a substantial burden of equine neurologic disease [10, 32]. This event catalyzed unprecedented research into WNV ecology, pathogenesis, and vaccine development, yet it also revealed stark contrasts in disease expression across different continents, particularly in Central and South America, where serologic evidence of widespread WNV circulation exists in horses but clinical disease remains conspicuously rare [2, 25, 40].

Horses, along with humans, are recognized as the mammalian species most frequently and severely affected by WNV infection, serving as highly sensitive sentinels for viral circulation [3, 10, 25]. The epidemiological significance of equine infections stems from several key biological realities. First, horses, like humans, are incidental or “dead-end” hosts, meaning they develop a viremia of insufficient magnitude and duration to infect feeding mosquitoes, thus playing no role in viral amplification [1, 10, 14]. Second, approximately 30-35% of infected horses develop clinical signs, a rate substantially higher than the estimated 1% risk of neuroinvasive disease in humans, making clinical equine cases a sensitive indicator of epizootic activity [3, 5]. Third, horses are large, long-lived, and often geographically stable animals, making them ideal for both retrospective serosurveys and prospective sentinel programs [3, 7]. Globally, the epidemiology of WNV in horses reflects the interplay of viral lineage, vector competence, avian reservoir community composition, and climatic conditions, with distinct patterns emerging across continents and ecoregions [1, 27].

European Epidemiology: From Sporadic Incursions to Seasonal Endemicity

Historical Context and Emergence of Lineage 2

Europe has experienced a profound epidemiological shift in WNV dynamics over the past two decades. Prior to the mid-1990s, WNV outbreaks in Europe were considered rare, isolated events, often linked to strains of lineage 1 introduced by migratory birds from sub-Saharan Africa [1, 27]. Notable early equine epizootics include the 1962-1963 outbreak in the Camargue (France) and the 1996 outbreak in Morocco, which caused severe neurologic disease in horses [10]. However, the epidemiological landscape changed dramatically with the emergence of West Nile virus lineage 2 in Europe. Initially considered a less virulent African lineage, lineage 2 strains were first detected in Hungary in 2004 and subsequently spread across Central and Southern Europe, causing increasingly severe outbreaks in both humans and horses [6, 17]. The Hungarian strain, belonging to sublineage 2d, was found to be genetically distinct from earlier African isolates, possessing similar neuroinvasive potential to highly virulent lineage 1 strains [6, 17]. The establishment of lineage 2 in Europe has been the primary driver of the continent’s current endemic situation, with the virus now overwintering in mosquito populations and circulating annually in many regions [5, 13].

The first equine cases of West Nile neuroinvasive disease in Austria, reported between 2016 and 2018, illustrate this expansion. Seven horses presented with severe neurological signs including gait abnormalities, fasciculations, tremors, and cranial nerve involvement, with WNV lineage 2 nucleic acid detected in five cases and specific neutralizing antibodies in all seven [6]. Notably, serologic evidence of infection was also found in two of fourteen in-contact horses, highlighting that subclinical infections occur even in close proximity to clinical cases [6]. Similarly, the first clinical equine case in Serbia was confirmed in July 2018 in a Belgian sports mare in Belgrade, which exhibited classic signs of ataxia, disorientation, and loss of equilibrium; detection of WNV IgM antibodies in serum strongly indicated acute infection [30]. These cases were preceded by years of documented WNV circulation in mosquitoes, birds, and sentinel horses, demonstrating that clinical disease often emerges only after a prolonged period of enzootic cycling [28, 30].

Geographic Distribution and Endemic Zones

Current European epidemiological patterns reveal a North-South gradient of transmission intensity, with the highest burdens concentrated in the Mediterranean basin, the Pannonian Basin, and specific regions of the Balkans [1, 34]. Italy, Greece, Serbia, Hungary, Romania, and Austria have become consistently endemic areas, with recurrent seasonal transmission reported annually [1, 9, 11, 17]. The 2018 transmission season in Serbia, for example, was the most intense ever recorded, with the highest number and severity of human and equine cases documented, and WNV circulation detected across the northern and central parts of the country, including Vojvodina Province, Belgrade, and surrounding districts [11, 30]. Importantly, the majority of human cases in Serbia in 2018 were preceded by detection of WNV circulation during the integrated surveillance program, validating the sentinel value of equine serosurveillance [11].

In Western Europe, the epidemiological situation has evolved more recently. Germany, which was considered WNV-free until 2018, experienced its first autochthonous cases in birds and horses during the exceptionally warm summer of that year [7, 13]. The widespread domestic Culex mosquitoes in Germany proved to be efficient vectors, and WNV circulation has been documented continuously in the affected areas of Central-East Germany, with autochthonous human cases reported in all subsequent years [13]. A seroprevalence study conducted in Eastern Germany in 2020, testing 940 equine sera, revealed a 5.8% WNV seroprevalence average, significantly higher in counties with previously registered equine infections [7]. Risk factor analysis identified breed type (pony), housing in endemic counties, continuous turnout (24-hour pasture access), and the presence of outdoor shelter as primary significant risk factors for seropositivity [7]. These findings suggest that management practices that increase exposure to outdoor mosquito vectors, rather than age or sex, are the dominant determinants of infection risk in German horses [7].

Drivers of Transmission and Climate Sensitivity

The epidemiology of WNV in Europe is exquisitely sensitive to climatic and environmental drivers, a feature that has become increasingly apparent with ongoing climate change [1, 5, 27]. The virus’s transmission cycle is fundamentally temperature-dependent, affecting both mosquito vector competence and viral replication rates. European summers with above-average temperatures, such as 2018, have been consistently associated with amplified WNV transmission [1, 13]. The shift from infrequent, sporadic outbreaks to regular, seasonal epidemics in many European regions is attributed to a combination of factors: warming summers, altered rainfall patterns that create ideal mosquito breeding habitats, and the adaptation of both vectors and virus to local ecological conditions [1, 16]. The vector competence of European mosquitoes, particularly members of the Culex pipiens complex, has been extensively characterized, with studies demonstrating that temperature, mosquito biotype, and geographic origin all influence transmission efficiency [16]. The ornithophilic Culex pipiens biotype pipiens is considered the primary enzootic vector, while the more mammalophilic Culex pipiens biotype molestus and the hybrid form may serve as bridge vectors, transmitting WNV from infected birds to incidental mammalian hosts such as horses and humans [16, 21].

Environmental variability and host community composition also play critical roles in shaping regional transmission risk, though the relative contribution of these factors remains incompletely quantified [1]. Migratory birds are the principal long-distance vectors for WNV introduction into new areas, with flyways connecting Africa, Europe, and Asia functioning as conduits for viral gene flow [15, 21, 22]. Slovakia, located at a crossroads of migration routes, has demonstrated evidence of WNV circulation for decades based on serological and molecular detection in vectors, birds, and horses, yet no major epidemics have been reported, highlighting the complex interplay of factors that determine outbreak intensity [15]. The Danube Delta in Romania serves as a critical wetland ecosystem for migratory birds, and surveillance in this region has revealed the presence of not only WNV but also other emerging arboviruses, including tick-borne viruses, underscoring the One Health importance of these ecosystems as hotspots for pathogen emergence [29].

Sentinel Surveillance: The Equine Paradigm in a One Health Framework

Rationale and Biological Basis for Equine Sentinel Programs

The use of horses as sentinels for WNV surveillance is grounded in robust biological and practical considerations that have made equine serosurveillance a cornerstone of integrated One Health programs across Europe and beyond [1, 3, 6, 9, 11]. The sentinel concept relies on the fact that horses are highly susceptible to WNV infection, develop a detectable and persistent antibody response, and serve as excellent indicators of local viral circulation without contributing to viral amplification [10, 26]. In Italy, the integrated WNV surveillance plan, operative since 2009, explicitly incorporates entomological, ornithological, and equine surveillance to identify viral presence in the environment and determine risk for human infection through potentially infected blood and organ donors [9, 26]. The economic rationale for this approach is compelling: an evaluation of the Lombardy Region surveillance program between 2014 and 2018 estimated that the application of the environmental and veterinary surveillance program allowed a reduction in costs of 7.7 million EUR over five years, compared to a scenario requiring individual nucleic acid amplification testing (NAT) of all blood donations throughout the vector season [9]. This cost-benefit analysis provides powerful support for the continued investment in equine sentinel surveillance as a public health intervention.

The biological basis for the sentinel value of horses rests on their immunologic response to WNV infection. Following natural infection, horses develop both IgM and IgG antibodies, with IgM appearing within the first week of infection and persisting for several weeks to months, providing a marker of recent viral exposure [7, 12, 30]. The detection of WNV-specific IgM in sentinel horses is therefore an early indicator of active viral transmission in a region, often preceding the identification of human cases by several weeks [11, 26]. In the Emilia-Romagna region of Italy during 2009, veterinary and entomological surveillance actions detected WNV activity by late July, approximately two to three weeks before the onset of the first human neuroinvasive case [26]. This temporal lead provides a critical window for public health authorities to intensify mosquito control measures, issue public health warnings, and institute blood donation screening protocols [1, 26]. Similarly, in the Serbian integrated surveillance program, the detection of WNV circulation in sentinel horses, wild birds, and mosquitoes consistently preceded the majority of human cases, demonstrating the predictive capacity of these surveillance components [11, 28].

Operational Frameworks and Implementation

National integrated surveillance programs in Europe have developed sophisticated frameworks that leverage equine serosurveillance as a key component within a broader One Health architecture. The Serbian national WNV surveillance program, funded by the Veterinary Directorate and operative since 2014, encompasses the entire territory and involves collaboration between the veterinary service, entomologists, and ornithologists [11, 28]. The program is based on three pillars: detection of WNV presence in wild birds (natural hosts), mosquitoes (virus vectors), and serological testing of sentinel horses for WNV-specific IgM antibodies [28]. Seronegative sentinel horses are selected from locations across the country, and serial blood samples are collected during the transmission season (typically May to October) to detect seroconversion events [28]. This indirect surveillance methodology allows for continuous and periodic monitoring of viral activity, with positive results triggering immediate notification of public health services and local authorities for implementation of mosquito control and preventive measures [11, 28].

The Austrian experience provides another instructive model of integrated human-animal-vector surveillance. During the transmission seasons of 2015 and 2016, a comprehensive surveillance program detected WNV nucleic acid in 21 samples from humans, horses, wild birds, and mosquito pools [17]. This included two equine WNND cases, two wild bird deaths due to WNND, and five Culex pipiens mosquito pools positive for WNV RNA [17]. The integrated analysis revealed that all infections were concentrated in the city of Vienna and neighboring regions of Lower Austria, with genomic sequencing demonstrating the co-circulation of several genetically distinct but closely related WNV strains belonging to sublineage 2d [17]. This level of spatial and genetic resolution is only achievable through the integration of multiple surveillance components, highlighting the power of the One Health approach to provide a comprehensive picture of WNV activity [17, 31]. The first autochthonous human case of West Nile neuroinvasive disease in Sicily (Southern Italy) in 2016 was similarly investigated using a One Health framework, with serological and molecular testing conducted on humans, horses, chickens, and dogs, alongside entomological surveillance. While mosquito pools tested negative for WNV RNA, two horses (18.2%) and two dogs (100%) on farms near the patient’s residence were positive for anti-WNV specific antibodies, confirming local viral circulation and the autochthonous origin of the human infection [31].

Diagnostic Considerations and Serological Differentiation

The effectiveness of equine sentinel surveillance depends critically on the availability of accurate, specific, and standardized diagnostic tools for detecting WNV infection and differentiating it from other flaviviruses [1, 4, 19, 35]. WNV is a member of the Japanese encephalitis serocomplex, a group of antigenically related flaviviruses that includes St. Louis encephalitis virus (SLEV), Usutu virus (USUV), Japanese encephalitis virus (JEV), and tick-borne encephalitis virus (TBEV) [2, 4, 35]. The extensive cross-reactivity among antibodies directed against these structurally similar viruses is a major challenge for serological diagnosis, particularly in regions where multiple flaviviruses co-circulate [4, 19, 35]. In Europe, the co-circulation of WNV, USUV, and TBEV is well-documented, and serosurveillance in horses must employ assays capable of distinguishing infections caused by these agents to provide meaningful epidemiological data [4, 7, 37]. For example, in the German seroprevalence study, of 106 serum samples reactive by competitive panflavivirus ELISA, only 54 (50.9%) were confirmed as WNV-positive by virus neutralization test (VNT), while 35 (33.0%) neutralized TBEV and 8 (7.5%) neutralized USUV [7]. This distribution underscores the necessity of confirmatory testing for accurate species-level diagnosis.

Recent advances in serological technology have significantly improved the discriminatory power of equine surveillance. The development of recombinant flavivirus E proteins with point mutations in the conserved fusion loop domain (Equad proteins) has enabled the reliable differentiation of WNV, USUV, and TBEV antibodies in horses without the need for time-consuming VNTs in high-biosafety-level laboratories [4]. Validation studies using panels of 136 horse sera demonstrated high sensitivity and specificity for these Equad ELISAs, representing a major advancement for conducting seroprevalence studies and routine veterinary diagnosis [4]. Similarly, multiplex immunoassays using recombinant envelope protein domain III (rEDIII) from WNV, JEV, and TBEV have been developed for equine sera, offering a powerful alternative to both ELISAs and VNTs for flavivirus diagnosis [35]. The differentiation of infected from vaccinated animals (DIVA) is another important consideration for equine surveillance in regions where WNV vaccination is practiced. Studies have shown that antibodies to the nonstructural NS1 protein can effectively differentiate WNV-infected from vaccinated horses under experimental conditions, though this differentiation may be more difficult under field conditions [12]. Standardization of diagnostic methods across European National Reference Laboratories (NRLs) has been a major priority, with inter-laboratory proficiency tests demonstrating that the adoption of commercial IgG and IgM capture ELISA kits has greatly improved the comparability and reliability of WNV serological testing across Europe [19].

Limitations and Challenges in Current Surveillance Systems

Despite the acknowledged value of equine sentinel surveillance, several limitations and challenges must be addressed to optimize its effectiveness. Reported surveillance data across Europe must be interpreted cautiously, as differences in national surveillance intensity, diagnostic capacity, and reporting frameworks influence notified case numbers [1]. The actual incidence of equine WNV infection is likely underestimated in many countries, particularly those with less robust surveillance infrastructure [1, 41]. In Brazil, for example, serological evidence of WNV circulation has been documented in horses in the Pantanal and Amazon regions since at least 2008, yet confirmed clinical cases were not reported until 2014 (human) and 2018 (equine), and the true extent of equine morbidity remains unclear [3, 40, 41]. Brazilian researchers have posited that WNV may be a “silent endemic” disease in the country, with cases going unrecognized due to lack of clinical awareness and limited diagnostic testing [40]. Similarly, in India, evidence of WNV infection has been demonstrated through seroprevalence studies and virus isolation since 1952, yet comprehensive surveillance of equine populations is lacking, and the prevalence and clinical impact of WNV in Indian horses remains poorly characterized [24].

The impact of vaccination on sentinel surveillance sensitivity is another critical concern. In areas where equine WNV vaccination programs have been implemented, vaccinated horses will develop antibodies that are indistinguishable from those produced by natural infection using standard serological assays, compromising their utility as sentinels [12, 26]. The Emilia-Romagna surveillance program noted this explicitly, observing that passive surveillance of horses would become less sensitive in the future due to an intensive vaccination program that commenced in June 2009 [26]. This challenge highlights the need for DIVA assays that can differentiate infected from vaccinated animals, as well as the strategic selection of unvaccinated sentinel horses for surveillance purposes [12]. Furthermore, the availability and uptake of equine WNV

Diagnostic Approaches for West Nile Virus in Horses: Serological and Molecular Methods

The accurate and timely diagnosis of West Nile virus (WNV) infection in horses is a cornerstone of effective surveillance, outbreak management, and clinical decision-making. Given that horses serve as highly sensitive sentinels for WNV circulation within a One Health framework [1, 9, 17], the diagnostic methodologies employed must be robust, specific, and capable of differentiating WNV from a complex landscape of co-circulating flaviviruses. The diagnostic challenge is multifaceted: the viremic period in horses is characteristically short-lived and of low magnitude, rendering direct viral detection methods time-sensitive, while the humoral immune response, though more prolonged, is plagued by the potential for serological cross-reactivity with antigenically related orthoflaviviruses such as Usutu virus (USUV), tick-borne encephalitis virus (TBEV), Japanese encephalitis virus (JEV), and St. Louis encephalitis virus (SLEV) [2, 4, 19, 35, 36]. Consequently, a comprehensive diagnostic approach necessitates a strategic combination of molecular techniques for early, acute-phase detection and serological assays for retrospective diagnosis and serosurveillance, with the definitive resolution of ambiguous cases often relying on the gold-standard virus neutralization test (VNT).

Molecular Detection: Targeting the Virus Genome

Molecular methods, primarily reverse transcription polymerase chain reaction (RT-PCR) and its real-time quantitative variant (RT-qPCR), are the principal tools for the direct detection of WNV RNA. These assays are most valuable during the acute phase of infection, typically within the first few days post-infection, when the virus is present in the blood and, more critically, in the central nervous system (CNS) of clinically affected horses [3, 18, 30]. The utility of these tests is, however, constrained by the transient and low-level viremia characteristic of horses, which are considered dead-end hosts [10, 14]. As demonstrated in the first clinical case of equine neuroinvasive WNV disease in Serbia, WNV RNA was not detectable by RT-qPCR in a blood sample collected seven days after the onset of clinical signs, despite the presence of a robust IgM antibody response [30]. This underscores the critical window for molecular sampling and the necessity of pairing negative molecular results with serological testing.

Despite this limitation, RT-PCR remains indispensable for confirmatory diagnosis, particularly from post-mortem tissue samples. The isolation of WNV RNA from the CNS of euthanized horses with neurological signs provides definitive evidence of neuroinvasion. In the 2011 Australian outbreak, a virulent strain of WNV (WNVNSW2011) was successfully detected in brain samples using a real-time RT-PCR assay, confirming the etiology of the widespread neurological disease [18]. Similarly, in the first confirmed cases of West Nile fever in horses in São Paulo, Brazil, WNV-specific RNA was amplified from erythrocytes using a nested PCR-multiplex PCR combination, demonstrating the utility of molecular techniques even when viral loads are expected to be low [3]. The use of nested PCR can enhance sensitivity, but it also carries an increased risk of amplicon contamination, requiring stringent laboratory practices.

The evolution of molecular diagnostics has also seen the development of advanced techniques such as whole-genome sequencing (WGS). While not a routine diagnostic tool, WGS is invaluable for molecular epidemiology, phylogenetic analysis, and tracking viral evolution and spread. Sequencing of WNV strains from horses, birds, and mosquito pools has been instrumental in identifying circulating lineages (e.g., lineage 1 in the Americas, lineage 2 in Europe) and tracking their geographic dissemination [6, 17, 23]. For instance, phylogenetic analysis of WNV sequences from Austria revealed the co-circulation of genetically distinct strains within the Central/Southern European cluster of WNV sublineage 2d, highlighting the dynamic nature of viral evolution [17]. The World Organisation for Animal Health (WOAH) recognizes molecular detection as a prescribed test for confirmatory diagnosis, particularly in outbreak situations where rapid and definitive identification of the agent is paramount.

Serological Methods: Interpreting the Humoral Response

Given the narrow window for molecular detection, serology forms the backbone of equine WNV diagnosis, especially for live animals presenting with neurological signs days or weeks after infection [19]. The equine immune response to WNV is characterized by an initial production of immunoglobulin M (IgM) antibodies, which appear within a few days of infection and are indicative of recent or active infection. This is followed by a sustained immunoglobulin G (IgG) response, which can persist for months or years, signifying past exposure or vaccination [12, 19].

IgM and IgG Enzyme-Linked Immunosorbent Assays (ELISAs)

ELISAs are the most widely used serological tools in veterinary diagnostic laboratories due to their relative simplicity, speed, and high throughput. The detection of WNV-specific IgM antibodies in a single serum sample is considered strong presumptive evidence of recent infection, as IgM does not cross the placenta and is not typically produced in response to vaccination with inactivated vaccines [12, 19]. Commercial IgM capture ELISAs have been rapidly adopted by national reference laboratories (NRLs) across Europe, demonstrating high specificity and sensitivity for diagnosing acute equine WNV infections [19]. In the 2011 Australian outbreak, an equine IgM ELISA was identified as the most effective tool for serological confirmation of clinical cases, outperforming other serological methods in detecting recent infection [18]. Similarly, the first clinical case of equine neuroinvasive WNV in Serbia was confirmed through the detection of WNV IgM antibodies by commercial ELISA, with results subsequently validated by VNT [30].

IgG ELISAs, while highly sensitive, are less specific due to the conserved nature of the flavivirus envelope (E) protein, which leads to cross-reactivity with antibodies against other flaviviruses such as USUV, TBEV, and SLEV [4, 19, 35]. This is a significant limitation in regions where multiple flaviviruses co-circulate. For example, in a seroprevalence study in Eastern Germany, a competitive pan-flavivirus ELISA (cELISA) detected antibodies in 106 of 940 horse sera, but subsequent VNT confirmation revealed that only 54 were specific to WNV, while 35 neutralized TBEV and 8 neutralized USUV [7]. This illustrates that a positive pan-flavivirus ELISA result is not diagnostic for WNV and must be interpreted with caution. To address this, advanced ELISAs have been developed using recombinant E proteins with point mutations in the conserved fusion loop domain (Equad proteins). These modified antigens significantly reduce cross-reactivity and enable reliable differentiation of WNV, USUV, and TBEV antibodies in horses, offering a promising alternative to the more laborious VNT [4].

Virus Neutralization Test (VNT): The Gold Standard

The VNT, also known as the plaque reduction neutralization test (PRNT), remains the gold standard for serological diagnosis of WNV due to its unparalleled specificity [2, 7, 19]. This functional assay measures the ability of antibodies in a serum sample to neutralize live virus, preventing infection of cell cultures. A four-fold or greater difference in neutralizing antibody titers between WNV and other flaviviruses (e.g., SLEV, USUV, TBEV) is used to identify the specific infecting agent [2]. In the serological survey of horses in Guatemala, PRNT90 was essential for differentiating WNV from SLEV infections, revealing that many horses previously thought to be WNV-positive by blocking ELISA were actually infected with SLEV or an undifferentiated flavivirus [2].

Despite its specificity, the VNT has several drawbacks that limit its use as a primary screening tool. It is time-consuming, labor-intensive, and requires a high level of technical expertise. Crucially, it must be performed in biosafety level 3 (BSL-3) facilities due to the use of live, infectious WNV, which restricts its availability to specialized reference laboratories [4, 35]. Furthermore, inter-laboratory proficiency tests have revealed that the sensitivity of VNTs can be variable, with some NRLs demonstrating lower sensitivity compared to IgG ELISAs [19]. Therefore, while the VNT is indispensable for confirmatory testing and resolving ambiguous ELISA results, it is not practical for large-scale surveillance or rapid clinical diagnosis.

Differentiating Infected from Vaccinated Animals (DIVA)

The widespread use of equine WNV vaccines in endemic regions creates a critical need for diagnostic assays that can differentiate naturally infected animals from those that have been vaccinated. This is essential for accurate surveillance, epidemiological investigations, and trade purposes. A common DIVA (Differentiating Infected from Vaccinated Animals) strategy targets antibodies against non-structural (NS) proteins, which are not present in most inactivated or subunit vaccines. The WNV NS1 protein has been proposed as a promising candidate for this purpose. A study evaluating the use of NS1-based assays found that while NS1 antigen could effectively differentiate infected from vaccinated horses under controlled experimental conditions, this differentiation became challenging under field conditions, likely due to variations in vaccine type, vaccination history, and the timing of sample collection relative to infection [12]. This highlights the complexity of implementing DIVA strategies in real-world settings and the need for continued refinement of these assays.

Integrated Diagnostic Algorithms and Future Directions

Given the inherent limitations of each individual diagnostic method, a hierarchical testing algorithm is recommended for the diagnosis of WNV in horses. The initial screening of samples from suspect cases should employ a highly sensitive IgM ELISA for acute infection or a pan-flavivirus IgG ELISA for serosurveillance. All reactive samples should then be subjected to confirmatory testing using a specific VNT or, where available, a second-line differential ELISA (e.g., Equad ELISA) to rule out cross-reactivity with other flaviviruses [4, 7, 19]. Molecular testing (RT-qPCR) should be prioritized for samples collected early in the clinical course, particularly from CSF or post-mortem brain tissue, to attempt direct viral detection.

The future of WNV diagnostics lies in the development of rapid, high-throughput, and highly specific multiplex platforms. A promising innovation is the microsphere-based multiplex immunoassay, which uses recombinant envelope protein domain III (rEDIII) from different flaviviruses coupled to fluorescent beads. This technology allows for the simultaneous detection and differentiation of antibodies against WNV, JEV, and TBEV in a single sample, offering a powerful alternative to sequential ELISA and VNT testing [35]. Such platforms, combined with the continued refinement of DIVA strategies and the integration of molecular and serological data within a One Health surveillance framework, will be critical for enhancing our ability to detect, monitor, and respond to the ever-present threat of WNV in equine populations [1, 9, 13]. The Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO) both emphasize the importance of robust laboratory capacity and standardized diagnostic protocols for effective arboviral disease surveillance, a principle that is directly applicable to the management of WNV in horses.

Prevention and Control Strategies: Vaccination, Vector Management, and One Health Considerations

The prevention and control of West Nile virus (WNV) in horses necessitates a multifaceted, evidence-based approach that integrates pharmacological protection through vaccination, ecological and chemical management of mosquito vectors, and a holistic, interdisciplinary framework, One Health, that recognizes the inextricable links between equine, human, and environmental health. As WNV is a zoonotic arbovirus primarily maintained in an enzootic bird–mosquito cycle, with horses and humans serving as incidental, dead-end hosts, interventions must be strategically targeted to interrupt transmission at multiple points. A critical appraisal of the literature reveals that while effective veterinary vaccines exist, their real-world application is hampered by gaps in owner awareness and inconsistent uptake [10, 42]; vector control programs require sustained investment and ecological understanding [9, 16]; and the success of any strategy is ultimately contingent upon robust, integrated surveillance and cross-sectoral collaboration under a One Health paradigm [1, 5, 11].

Vaccination Strategies for Equine WNV Prevention

Vaccination remains the single most effective tool for protecting individual horses from clinical WNV disease, and several licensed vaccines have demonstrated high efficacy in reducing morbidity and mortality [10, 43]. The development and deployment of equine WNV vaccines represent a major success story in veterinary vaccinology, catalyzed by the introduction of WNV into North America in 1999 and the subsequent rapid spread across the continent [10, 25]. The primary vaccines available for horses include inactivated whole-virus vaccines, a recombinant canarypox virus-vectored vaccine expressing the WNV prM/E proteins, and a live, attenuated yellow fever–WNV chimeric vaccine [10, 25, 43]. These vaccines predominantly induce neutralizing antibodies against the viral envelope (E) protein, which are strongly correlated with protection [44, 45]. Experimental vaccines, including a novel Newcastle disease virus-vectored construct and plant-produced virus-like particles (VLPs) displaying domain III of the E protein, have shown promise in eliciting robust humoral and cellular immune responses in both mice and horses [8, 45]. Additionally, advanced inactivation technologies, such as the HydroVax-II platform employing hydrogen peroxide, copper ions, and methisazone, have demonstrated the capacity to preserve neutralizing epitopes while achieving complete virus inactivation, resulting in 130-fold higher neutralizing antibody titers compared to earlier formulations [44].

Despite the availability of effective vaccines, vaccination coverage in many endemic and emerging regions remains suboptimal. A cross-sectional survey of horse owners in Romania, a highly endemic country, revealed that only a small proportion of respondents reported vaccinating their horses, with low awareness of the vaccine’s existence and its protective benefits [42]. This knowledge-to-action gap is attributable to several factors: insufficient veterinary communication, underestimation of disease risk, and economic constraints [41, 42]. This is particularly concerning given that WNV continues to cause severe neuroinvasive disease and mortality in unvaccinated horses, as documented in outbreaks across Europe, the Americas, and Australia [6, 18, 20, 30]. For instance, during the first confirmed equine WNV cases in Austria, four of seven affected horses were euthanized due to the severity of clinical signs [6]. Similarly, in the 2011 outbreak in New South Wales, Australia, over 300 horses exhibited neurological disease, with a virulent lineage 1 strain responsible [18].

Another critical dimension of vaccination strategy involves the ability to differentiate infected from vaccinated animals (DIVA). This is essential for surveillance programs that rely on serological data to track natural virus circulation in vaccinated populations [12]. The WNV nonstructural protein 1 (NS1) has been identified as a promising DIVA marker, as antibodies to NS1 are generated during natural infection but are absent in horses vaccinated with purified inactivated or subunit vaccines that do not contain NS1 [12]. Studies have demonstrated that NS1-based serological assays can effectively distinguish infected from vaccinated horses under controlled experimental conditions, although field validation has revealed challenges, including variable immune responses and potential cross-reactivity with other flaviviruses [12, 35]. The development of robust DIVA assays is critical for maintaining the integrity of sentinel surveillance, particularly as vaccination programs expand [9, 26]. Furthermore, the proliferation of flaviviruses such as Usutu virus (USUV) and tick-borne encephalitis virus (TBEV) that co-circulate with WNV complicates serological interpretation, underscoring the need for highly specific diagnostic tools like mutant E protein ELISAs (Equad ELISAs) that can differentiate WNV, USUV, and TBEV antibodies without requiring time-consuming virus neutralization tests (VNTs) [4, 19, 35].

Integrated Vector Management: Targeting the Mosquito Bridge

Vector management constitutes the second pillar of WNV prevention and control, directly targeting the arthropod bridge that enables transmission from the avian reservoir to horses and humans. The primary vectors for WNV are mosquitoes of the Culex genus, particularly Culex pipiens in temperate regions, which are highly competent for both lineage 1 and lineage 2 WNV strains [16, 33, 34]. Vector competence is modulated by a complex interplay of genetic, environmental, and microbial factors, including temperature, mosquito genotype, biotype, and the presence of endosymbiotic Wolbachia bacteria [14, 16]. Experimental studies have demonstrated that rising temperatures accelerate viral replication within mosquitoes, increasing the proportion of infectious vectors and shortening the extrinsic incubation period, thereby amplifying transmission risk, a finding of profound significance in the context of climate change [5, 14, 16].

Integrated vector management (IVM) strategies encompass a suite of interventions, including larval source reduction through environmental management, application of larvicides to aquatic breeding sites, targeted adulticide applications during periods of high transmission, and personal protective measures for horses such as insect repellents and stable screening [34, 42]. Larval control is widely recognized as the most cost-effective and environmentally sustainable approach, as it prevents the emergence of adult vectors before they can feed and transmit virus [32, 34]. This involves removing or treating standing water sources (e.g., water troughs, discarded tires, drainage ditches) that serve as larval habitats for Culex species [33, 42]. In Romania, a survey of horse owners found that while many used insecticides and repellents, environmental measures such as eliminating standing water were less commonly practiced, representing a critical gap in prevention [42].

Adult mosquito control using ultra-low-volume (ULV) applications of organophosphates or pyrethroids can be implemented as an emergency response during outbreak situations [32, 34]. However, the effectiveness of such interventions is contingent upon timing; vector control must be initiated early, ideally before the epidemic peaks, to significantly reduce transmission [32]. Resistance to commonly used insecticides is an emerging concern in several Culex populations, necessitating routine monitoring and rotation of active ingredients [34]. Furthermore, the ecological impact of broad-spectrum insecticides on non-target organisms, including beneficial insects and pollinators, must be carefully weighed [32, 34].

A nuanced understanding of local vector ecology is essential for tailoring control strategies. For example, different mosquito species exhibit distinct feeding preferences, with some ornithophilic species (e.g., Culex pipiens) primarily feeding on birds, while others (e.g., Culex neavei in Senegal) serve as bridge vectors by feeding on both birds and mammals, thus facilitating spillover to horses and humans [21]. In South America, similar bridging behavior has been implicated in the sporadic but widespread circulation of WNV, despite relatively low clinical disease incidence in horses compared to North America [25, 40]. The presence of alternative vectors, such as Aedes species and even ticks (Hyalomma marginatum), which have been shown to carry WNV RNA, further complicates vector management and underscores the need for a comprehensive entomological surveillance framework [22, 29].

One Health Considerations: Integrated Surveillance and Cross-Sectoral Collaboration

The One Health approach, which recognizes the interdependence of human, animal, and environmental health, is not merely an aspirational concept for WNV prevention, it is an operational necessity. WNV exemplifies a pathogen whose transmission dynamics are driven by ecological, climatic, and anthropogenic factors that transcend disciplinary boundaries [1, 22, 31]. The virus circulates in an enzootic cycle involving wild birds as amplifying hosts and ornithophilic mosquitoes as vectors, with spillover to horses and humans. Consequently, effective surveillance and control require sustained collaboration among veterinarians, entomologists, ornithologists, public health authorities, and environmental managers [1, 9, 17].

Integrated surveillance programs that simultaneously monitor WNV activity in mosquitoes, wild birds, sentinel horses, and human cases have been successfully implemented in several European countries, including Italy, Serbia, Greece, and Austria [9, 11, 17, 26, 28, 31]. These programs utilize horses as highly sensitive sentinels, as they develop detectable IgM antibodies within days of infection and are clinically susceptible to neuroinvasive disease [1, 6, 19, 26]. The detection of seroconversion or clinical cases in sentinel horses can provide an early warning of WNV circulation, often preceding human cases by two to three weeks, thereby enabling preemptive vector control and public health messaging [1, 11, 26]. In Serbia, the national integrated surveillance program detected the most intensive WNV circulation in 2018, with the majority of human cases preceded by positive findings in sentinel animals, mosquitoes, and wild birds [11]. Similarly, in Italy, the integrated surveillance program in the Lombardy region not only documented the expansion of the endemic area but also resulted in substantial cost savings, an estimated 7.7 million EUR over five years, by targeting blood donation nucleic acid amplification testing (NAT) to high-risk periods identified through veterinary and entomological surveillance [9].

From a One Health perspective, vaccination of horses not only protects individual animals but also contributes to public health by reducing the potential for local amplification and spillover, although horses are dead-end hosts and do not contribute to onward transmission [10, 45]. Nevertheless, widespread equine vaccination can indirectly reduce human exposure by eliminating clinical cases that might otherwise draw diagnostic attention and by reducing the number of viremic animals that could serve as sources for bridge vectors [26]. However, this benefit is contingent upon achieving high vaccination coverage, which remains elusive in many regions [42].

Economic and policy considerations are central to the sustainability of One Health approaches. The elimination of funding for vector-borne disease monitoring programs, as threatened by proposed budget cuts to the U.S. Centers for Disease Control and Prevention’s Division of Vector Borne Infectious Diseases in 2010, would dismantle critical infrastructure for WNV surveillance and response [46]. Such cuts would compromise the ArboNET monitoring system, which integrates data on human cases, mosquito infections, and animal sentinels, and would weaken preparedness for emerging threats like dengue and Eastern equine encephalitis [46]. International collaboration and standardization of diagnostic practices are equally important; for example, the alignment of Japanese encephalitis virus (JEV) diagnostic pipelines among European veterinary reference laboratories provides a template for cooperation that can be extended to WNV and other flaviviruses [36]. The World Organisation for Animal Health (WOAH) and the World Health Organization (WHO) have both emphasized the importance of joint human-animal-vector surveillance and the incorporation of environmental data, such as temperature, rainfall, and land use, into risk models [1, 5].

Finally, the persistent gaps in WNV surveillance and control in low- and middle-income countries (LMICs) represent a major challenge for global health equity. In regions like Central and South America, Africa, and parts of Asia, WNV circulation is documented serologically, yet clinical disease in horses and humans appears less frequent than in North America or Europe [2, 24, 40]. This discrepancy may be due to underreporting, diagnostic limitations, cross-protective immunity from other flaviviruses, or differences in circulating viral lineages (e.g., lineage 5 in India) [24, 25, 40]. In Brazil, despite evidence of WNV seropositivity in horses since the early 2000s and the first confirmed human case in 2014, equine vaccines are not yet licensed, and surveillance remains fragmented [40, 41]. The development of affordable, locally producible vaccines, such as plant-produced VLPs, and the strengthening of diagnostic capacity are urgent priorities for these settings [8, 22, 24]. A truly global One Health framework must address these inequities by fostering technology transfer, capacity building, and harmonized surveillance protocols across all regions where WNV is present or at risk of emergence.

References

[1] Nistor P, Stânga L, Iorgoni V, Cojocaru R, Gligor A, Cireșan A, et al.. West Nile Virus in Europe: Epidemiology, Vector Ecology, Environmental Drivers, and the Role of Equine Sentinel Surveillance in a One Health Framework. Pathogens. 2026. DOI: https://doi.org/10.3390/pathogens15030298

[2] Morales-Betoulle M, Morales H, Blitvich B, Powers A, Davis EA, Klein R, et al.. West Nile Virus in Horses, Guatemala. Emerging Infectious Diseases. 2006. DOI: https://doi.org/10.3201/eid1206.051615

[3] Siqueira R, Hansen VS, Martins M, Leal M, Bondan E. West Nile Fever Virus Infection in Horses in São Paulo State, Brazil. Acta Scientiae Veterinariae. 2022. DOI: https://doi.org/10.22456/1679-9216.117796

[4] Schwarzer A, Ziegler U, Fertey J, Kreuz M, Vahlenkamp T, Groschup M, et al.. Serological differentiation of West Nile, Usutu, and tick-borne encephalitis virus antibodies in birds and horses using mutant E protein ELISAs. Scientific Reports. 2025. DOI: https://doi.org/10.1038/s41598-025-14448-4

[5] Zerbato V, Rossi B, Bella SD, Bartalucci C, Cerchiaro M, Re DD, et al.. West Nile virus: epidemiology, prevention, clinical features, diagnosis, treatment, and open research questions. Annals medicus. 2026. DOI: https://doi.org/10.1080/07853890.2026.2615482

[6] Heus P, Kolodziejek J, Camp JV, Dimmel K, Bagó Z, Hubálek Z, et al.. Emergence of West Nile virus lineage 2 in Europe: Characteristics of the first seven cases of West Nile neuroinvasive disease in horses in Austria. Transboundary and Emerging Diseases. 2019. DOI: https://doi.org/10.1111/tbed.13452

[7] Ganzenberg S, Sieg M, Ziegler U, Pfeffer M, Vahlenkamp T, Hörügel U, et al.. Seroprevalence and Risk Factors for Equine West Nile Virus Infections in Eastern Germany, 2020. Viruses. 2022. DOI: https://doi.org/10.3390/v14061191

[8] Stander J, Chabeda A, Rybicki E, Meyers A. A Plant-Produced Virus-Like Particle Displaying Envelope Protein Domain III Elicits an Immune Response Against West Nile Virus in Mice. Frontiers in Plant Science. 2021. DOI: https://doi.org/10.3389/fpls.2021.738619

[9] Defilippo F, Dottori M, Lelli D, Chiari M, Cereda D, Farioli M, et al.. Assessment of the Costs Related to West Nile Virus Monitoring in Lombardy Region (Italy) between 2014 and 2018. International Journal of Environmental Research and Public Health. 2022. DOI: https://doi.org/10.3390/ijerph19095541

[10] Bosco-Lauth A, Bowen R. West Nile Virus: Veterinary Health and Vaccine Development. Journal of medical entomology. 2019. DOI: https://doi.org/10.1093/jme/tjz125

[11] Petrović T, Šekler M, Petrić D, Vidanović D, Debeljak Z, Lazić G, et al.. Intensive West Nile Virus Circulation in Serbia in 2018, Results of Integrated Surveillance Program. Pathogens. 2021. DOI: https://doi.org/10.3390/pathogens10101294

[12] Rebollo B, Sarraseca J, Lecollinet S, Abouchoaib N, Alonso J, García‐Bocanegra I, et al.. Monitoring Anti-NS1 Antibodies in West Nile Virus-Infected and Vaccinated Horses. BioMed Research International. 2018. DOI: https://doi.org/10.1155/2018/8309816

[13] Frank C, Schmidt-Chanasit J, Ziegler U, Lachmann R, Preußel K, Offergeld R. West Nile Virus in Germany: An Emerging Infection and Its Relevance for Transfusion Safety. Transfusion Medicine and Hemotherapy. 2022. DOI: https://doi.org/10.1159/000525167

[14] Fiacre L, Pagès N, Albina E, Richardson J, Lecollinet S, Gonzalez G. Molecular Determinants of West Nile Virus Virulence and Pathogenesis in Vertebrate and Invertebrate Hosts. International Journal of Molecular Sciences. 2020. DOI: https://doi.org/10.3390/ijms21239117

[15] Korytár L, Peňazziová K, Pistl J, Tichá E, Čabanová V, Csank T. Retrospective review and current knowledge on the occurrence of West Nile virus in mosquito vectors, reservoirs and hosts in Slovakia (Central Europe).. Acta virologica. 2020. DOI: https://doi.org/10.4149/av_2020_209

[16] Vogels C, Göertz GP, Pijlman G, Koenraadt C. Vector competence of European mosquitoes for West Nile virus. Emerging Microbes and Infections. 2017. DOI: https://doi.org/10.1038/emi.2017.82

[17] Kolodziejek J, Jungbauer C, Aberle S, Allerberger F, Bagó Z, Camp JV, et al.. Integrated analysis of human-animal-vector surveillance: West Nile virus infections in Austria, 2015–2016. Emerging Microbes and Infections. 2018. DOI: https://doi.org/10.1038/s41426-018-0021-5

[18] Read A, Finlaison D, Gu X, Hick P, Moloney B, Wright T, et al.. Clinical and epidemiological features of West Nile virus equine encephalitis in New South Wales, Australia, 2011. Australian Veterinary Journal. 2019. DOI: https://doi.org/10.1111/avj.12810

[19] Beck C, Lowenski S, Durand B, Bahuon C, Zientara S, Lecollinet S. Improved reliability of serological tools for the diagnosis of West Nile fever in horses within Europe. PLoS Neglected Tropical Diseases. 2017. DOI: https://doi.org/10.1371/journal.pntd.0005936

[20] Autorino G, Battisti A, Deubel V, Ferrari G, Forletta R, Giovannini A, et al.. West Nile virus Epidemic in Horses, Tuscany Region, Italy. Emerging Infectious Diseases. 2002. DOI: https://doi.org/10.3201/eid0812.020234

[21] Fall A, Diaïté A, Lancelot R, Tran A, Soti V, Etter E, et al.. Feeding behaviour of potential vectors of West Nile virus in Senegal. Parasites & Vectors. 2011. DOI: https://doi.org/10.1186/1756-3305-4-99

[22] Mikhaleva TV, Gasanov RR, Konnova SS, Lunina D. Applying One Health approach to the study of West Nile fever in the Russian Federation. Veterinary Science Today. 2026. DOI: https://doi.org/10.29326/2304-196x-2026-15-1-20-27

[23] Hyeon J, Helal ZH, Appel A, Tocco N, Hunt A, Lee D, et al.. Whole genome sequencing and phylogenetic analysis of West Nile viruses from animals in New England, United States, 2021. Frontiers in Veterinary Science. 2023. DOI: https://doi.org/10.3389/fvets.2023.1085554

[24] Gulati B, Gupta A, Kadian SK, Pratap R, Panwar PK, Malik Y, et al.. West Nile Virus infection among animals and humans in India.. Advances in Animal and Veterinary Sciences. 2014. DOI: https://doi.org/10.14737/JOURNAL.AAVS/2014/2.4S.17.23

[25] Martín-Acebes MA, Saiz J. West Nile virus: A re-emerging pathogen revisited.. World Journal of Virology. 2012. DOI: https://doi.org/10.5501/wjv.v1.i2.51

[26] Angelini P, Tamba M, Finarelli A, Bellini R, Albieri A, Bonilauri P, et al.. West Nile virus circulation in Emilia-Romagna, Italy: the integrated surveillance system 2009.. Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin. 2010. DOI: https://doi.org/10.2807/ESE.15.16.19547-EN

[27] Monini M, Falcone E, Busani L, Romi R, Ruggeri F. West Nile Virus: Characteristics of an African Virus Adapting to the Third Millennium World. Open Virology Journal. 2010. DOI: https://doi.org/10.2174/1874357901004010042

[28] Petrović T, Šekler M, Petrić D, Lazić S, Lupulović D, Lazić G, et al.. West Nile Virus surveillance program in Serbia.. . 2015. DOI: https://doi.org/10.46784/E-AVM.V7I2.130

[29] Bratuleanu B, Chretien D, Bigot T, Régnault B, Pérot P, Savuța G, et al.. Insights into the virome of Hyalomma marginatum in the Danube Delta: a major vector of Crimean-Congo hemorrhagic fever virus in Eastern Europe. Parasites & Vectors. 2024. DOI: https://doi.org/10.1186/s13071-024-06557-2

[30] Medić S, Lazić S, Petrović T, Petrić D, Samojlović M, Lazić G, et al.. Evidence of the first clinical case of equine neuroinvasive West Nile disease in Serbia, 2018. Acta Veterinaria. 2019. DOI: https://doi.org/10.2478/acve-2019-0009

[31] Macaluso G, Gucciardi F, Guercio A, Blanda V, Russa FL, Torina A, et al.. First neuroinvasive human case of West Nile Disease in Southern Italy: Results of the ‘One Health’ approach. Veterinary Medicine and Science. 2021. DOI: https://doi.org/10.1002/vms3.591

[32] Rimland D, Koplan J, Stephens D. West Nile Virus Southeast Conference. Emerging Infectious Diseases. 2003. DOI: https://doi.org/10.3201/EID0907.030254

[33] Bajwa WI, Zhou L, Shah Z. MOSQUITO SPECIES OF NEW YORK STATE: AN UPDATED CHECKLIST WITH PUBLIC AND VETERINARY HEALTH SIGNIFICANCE.. Journal of the American Mosquito Control Association. 2025. DOI: https://doi.org/10.2987/25-7246

[34] Marka A, Diamantidis A, Papa A, Valiakos G, Chaintoutis SC, Doukas D, et al.. West Nile Virus State of the Art Report of MALWEST Project. International Journal of Environmental Research and Public Health. 2013. DOI: https://doi.org/10.3390/ijerph10126534

[35] Beck C, Desprès P, Paulous S, Vanhomwegen J, Lowenski S, Nowotny N, et al.. A High-Performance Multiplex Immunoassay for Serodiagnosis of Flavivirus-Associated Neurological Diseases in Horses. BioMed Research International. 2015. DOI: https://doi.org/10.1155/2015/678084

[36] Mansfield KL, Parekh I, Rasmussen T, Lohse L, Olesen A, Dheilly N, et al.. European Preparedness for Japanese Encephalitis Virus Through Alignment of Animal Health Laboratory Diagnosis. Transboundary and Emerging Diseases. 2025. DOI: https://doi.org/10.1155/tbed/5516160

[37] Cvjetković IH, Petrović T, Petrić D, Milošević U, Radovanov J, Kovačević G, et al.. USUTU VIRUS: AN EMERGING FLAVIVIRUS IN EUROPE. Archives of Veterinary Medicine. 2019. DOI: https://doi.org/10.46784/e-avm.v10i1.79

[38] El-Harrak M, Martin-Folgar R, Llorente F, Fernández-Pacheco P, Brun A, Figuerola J, et al.. Rift Valley and West Nile Virus Antibodies in Camels, North Africa. Emerging Infectious Diseases. 2011. DOI: https://doi.org/10.3201/eid1712.110587

[39] Eeden Cv, Swanepoel R, Venter M. Antibodies against West Nile and Shuni Viruses in Veterinarians, South Africa. Emerging Infectious Diseases. 2014. DOI: https://doi.org/10.3201/eid2008.131724

[40] Vieira MACS, Aguiar AdAX, Borba AS, Guimarães H, Eulálio KD, ALBUQUERQUE-NETO LLd, et al.. WEST NILE FEVER IN BRAZIL: SPORADIC CASE, SILENT ENDEMIC DISEASE OR EPIDEMIC IN ITS INITIAL STAGES?. Revista do Instituto de Medicina Tropical de São Paulo. 2015. DOI: https://doi.org/10.1590/S0036-46652015000300017

[41] Souza JMd, Mattos Ábd, Silva BRd, Maia GR, Haidar TL. OCORRÊNCIA DA FEBRE DO NILO EM EQUINOS NO BRASIL. Revista ft. 2024. DOI: https://doi.org/10.69849/revistaft/ch10202409301258

[42] Nistor P, Stânga L, Chirila A, Iorgoni V, Cocioba V, Cojocaru R, et al.. Gaps Between Awareness and Prevention of West Nile Virus Among Horse Owners in an Endemic Country: A Cross-Sectional Study from Romania. Veterinary Sciences. 2026. DOI: https://doi.org/10.3390/vetsci13030239

[43] Iyer AV, Kousoulas K. A Review of Vaccine Approaches for West Nile Virus. International Journal of Environmental Research and Public Health. 2013. DOI: https://doi.org/10.3390/ijerph10094200

[44] Quintel BK, Thomas A, Deraad D, Slifka M, Amanna IJ. Advanced oxidation technology for the development of a next-generation inactivated West Nile virus vaccine. Vaccine. 2018. DOI: https://doi.org/10.1016/j.vaccine.2018.12.020

[45] Wang J, Yang J, Ge J, Hua R, Liu R, Li X, et al.. Newcastle disease virus-vectored West Nile fever vaccine is immunogenic in mammals and poultry. Virology Journal. 2016. DOI: https://doi.org/10.1186/s12985-016-0568-5

[46] Palmer R. Line item cut threatens monitoring of more than West Nile virus. Nature Network Boston. 2010. DOI: https://doi.org/10.1038/nm0910-938b