Transmission and Epidemiology

West Nile virus is sustained in nature through a transmission cycle that primarily involves the ornithophilic Culex species and the birds that they almost exclusively feed on. Such species of mosquitoes are referred to as amplification vectors. Other mosquito species, referred to as bridging vectors, allow the virus to emerge from this amplification cycle by transmitting it to humans, horses, and other nonavian vertebrates after feeding on viremic birds. Humans, horses, and other nonavian vertebrates rarely develop viremia of sufficient magnitude to infect feeding mosquitoes and are thus referred to as dead-end hosts. The seasonality of varying geographic locations and hence duration of the vector season have bearing on the time of year in which cases are most prevalent. Zoonotic transmission to a human from an infected horse during necropsy has been reported once. Human infections have also been documented after avian necropsies and through blood-contaminated needle puncture in laboratory workers. Nonvector routes of WNV transmission include horizontal transmission through breastfeeding, blood transfusion, organ transplantation, and oral and intrauterine infection.

West Nile virus has a vast vector and host range. Passerine species such as house sparrows and crows feature prominently in the life cycle of the virus, but evidence of WNV infection has been detected in more than 300 species of birds. Since the emergence of the disease in the United States in 1999, WNV has accounted for a significant decrease in the population of many bird species, and avian deaths have been used as an important indicator of the virus’s circulation. Likewise, the virus has been detected in approximately 30 species of mammals, more than 60 species of mosquitoes, and a wide array of amphibious and reptilian species.

As is characteristic of most single-stranded RNA viruses, the WNV genome has rapidly evolved since its introduction to the United States in 1999 (NY 1999 strain). This has given rise to novel strains that have the ability to replicate faster in avian hosts or be transmitted more efficiently by the insect vector. The NY 1999 strain arose from the highly virulent lineage 1 strain found in Europe, the Middle East, Africa, and Australia. Lineage 2 strains from Africa and Madagascar are considered less virulent. The NY 1999 strain, which has not been detected in North America since 2004, has since been replaced by a novel variant WN02. The emergence of this strain coincided with a sudden rise in equine WNV cases (15,257) in the United States in 2002, compared with 738 cases in 2001. Since 2002, there has been a gradual decline in the number of equine cases; however, at the time of this writing in autumn of 2012, 566 cases had been reported, representing an already six-fold increase in the number of cases from 2011 (87 cases).

Pathogenesis

West Nile virus has a predilection for infecting the central nervous system (CNS) in horses. Despite this neurotropism, less than 1% of flavivirus infections result in natural infection of the CNS. Following inoculation by a chronically infected vector, the virus replicates in the Langerhans and dendritic cells of the skin. From there, the virus spreads to the regional lymph nodes and into the bloodstream and peripheral organs such as the spleen and kidneys, where a second round of replication occurs. Viral replication in the host body is enhanced by the ability of the virus to negate the action of type I interferon and elude the antiviral activity of interferon-stimulated genes.

The exact mechanism of the virus’s ability to invade the CNS approximately 1 week after inoculation remains unclear. What is known is that the level of viremia directly correlates with the probability of neuroinvasion. Hypotheses include disruption of the blood-brain barrier, viral transport by infected immune cells into the CNS, direct axonal retrograde transport from infected peripheral neurons, and endocytosis across vascular endothelium into the CNS. Infection of the olfactory neurons, which do not fall under the protection of the blood-brain barrier, has also been suggested as a route of CNS invasion.

Clinical Signs

Although WNV has accounted for extensive morbidity and death in horses throughout the North American continent since 1999, experimental infection studies have shown that very few (8%) horses develop clinical signs after infection with the virus. A prospective study also reported that only 8% of unvaccinated horses that seroconverted after natural exposure to the virus developed neurologic signs consistent with WNV infection.

When the disease is manifested clinically, fever, lethargy, and loss of appetite may be the first signs observed. Acute onset of ataxia, weakness, or both may also be seen to varying degrees and are usually asymmetric. Paresis may progress to tetraplegia and recumbency. Behavioral changes ranging from aggression and hyperexcitability to somnolence or coma may also be seen.

One of the hallmarks of WNV encephalomyelitis in horses is muscle fasciculation. These are most often seen in the muscles of the face (muzzle and lips) and neck, but may also be seen in the trunk region, as well as in the triceps and quadriceps regions of the forelimbs and hind limbs, respectively. Hyperesthesia often accompanies the muscle fasciculation. Cranial nerve (VII, IX, and XII) involvement accounting for facial paralysis and dysphagia may also be seen in a small number of cases. Histologic changes in the pons and medulla oblongata account for these abnormalities.

The combination of these clinical signs varies, as does the duration and severity of signs. Most horses show clinical improvement in 3 to 7 days, but full recovery may take weeks to months, and long-term residual neurologic deficits are common. Recrudescence of disease may be seen within 7 to 10 days of the initial phase of clinical improvement; the exact cause of this recrudescence is unknown.