In 2007, an outbreak of febrile rash illness occurred on the island of Yap, a small archipelago in the Western Pacific. The cause of the outbreak was a little-known and, to that point, medically unimportant entity, Zika virus. That year, approximately 100 cases were identified; signs and symptoms included mild fever, headache, rash, and conjunctivitis. The outbreak was notable since this was the first outbreak of the virus to be identified; Zika virus had been associated with sporadic, isolated cases of mild rash illness for decades, mainly in Africa and parts of Asia. The Yap outbreak witnessed no deaths, and infected persons went on to have a complete recovery. In 2013, the epidemiology of Zika virus seemed to change drastically. First causing a large outbreak in the Pacific islands of French Polynesia, the Zika virus epidemic seemed to be associated with large increases in the incidence of Guillain-Barré syndrome (GBS); as the virus moved into South America, it caused large outbreaks of exanthematous illness also associated with extraordinarily high rates of GBS. In addition, the virus was found to be able to lead to severe congenital malformations in infants and fetuses of women infected with Zika virus early in their pregnancies. These dramatic and severe manifestations of Zika virus led the World Health Organization (WHO) to declare it a global crisis in 2014.
Zika virus is only the most recent arthropod-borne virus (arbovirus) to wreak havoc on human populations. In 1999, a little-known arbovirus of the same family as Zika virus (Flaviviridae), West Nile virus (WNV), caused a cluster of cases of meningoencephalitis and acute flaccid paralysis in New York City. This virus, which heretofore was associated with sporadic cases of mild rash illness in Africa, Asia, and the Middle East, with occasional small outbreaks, resulted ultimately in the largest arboviral outbreak in United States history.
On the other side of the globe, another flavivirus, Japanese encephalitis virus, has been causing large annual outbreaks of encephalitic illness, mainly among children, in much of South and Southeast Asia. It is, in fact, a leading cause of death in many of these areas.
This article begins by focusing on the current understanding of Zika virus and neurologic abnormalities, highlighting the correlation with GBS. The article then turns to other emerging and reemerging arboviral pathogens, including West Nile virus and Japanese encephalitis virus. As evidenced with Zika virus, West Nile virus, and Japanese encephalitis virus, arthropod-borne viruses (arboviruses) are likely to continue to emerge and result in human illness.
This section discusses the epidemiology, virology, clinical features, and future directions for research on the newly emerging virus known as Zika virus.
Zika virus is an arbovirus (a virus that is transmitted by an insect vector) within the Flaviviridae family. More specifically, it is a member of the genus Flavivirus, which includes several medically important neurotropic viruses such as Japanese encephalitis virus, West Nile virus, and St. Louis encephalitis virus. Virologically, it is most closely related to dengue virus, one of the most widely dispersed arboviruses and important in terms of burden of human illness worldwide. Zika virus is a single-stranded, positive RNA virus, with a genome of approximately 11 kilobase pairs (kb) that expresses seven nonstructural proteins and three structural proteins. The three structural proteins are responsible for the basic replication and transmission of Zika virus. The capsid protein connects viral RNA within the nucleocapsid; the two other structural proteins, the membrane protein and the envelope protein, are responsible for attaching the virus to the surface of host cells, following which the virus enters the cell by endocytosis. Following replication of negative-stranded viral RNA, which serves as the genetic template for positive-stranded genomic RNA, viruses are assembled within the endoplasmic reticulum and released from host cells by exocytosis. This process continues as long as the virus is able to withstand the human immune response, which starts with the Zika virus–specific IgM response, followed by the subsequent IgG response. The role of T-cell–mediated immunogenicity is unclear at present. At least two major lineages of Zika virus have been identified based upon nucleotide sequence analysis, the African and Asian lineages.
Ecology and Epidemiology
Zika virus has been recognized as a human pathogen for decades. The first isolation of the virus was from a febrile macaque monkey in 1948 in the Zika forest of Uganda; given the historical precedent of naming novel arboviruses after their location of discovery, the virus was termed Zika virus. The first human infection was detected in 1952. For the next several decades, Zika virus was not a significant human pathogen, being associated with sporadic cases of mild rash illness with fever, likely through a sylvatic transmission cycle in disparate areas of sub-Saharan Africa and Southeast Asia. Before 2007, approximately 13 cases of natural infection of humans with Zika virus were documented. The geographic spread of Zika virus was largely contingent on the geographic distribution of the insect vector, Aedes species mosquitoes. In urban and suburban environments, Aedes mosquitoes transmit the virus to humans in a mosquito-human-mosquito cycle. An infected Aedes mosquito feeds on human blood and transmits the virus to the human, after which the infected human, still viremic with the virus, is bitten by another mosquito, thus continuing the cycle of transmission.
Zika virus remained a relatively unimportant pathogen medically for several decades after its first isolation. The epidemiology of Zika virus seemed to change, however, in 2007, when an outbreak of the virus occurred in Micronesia, on the island of Yap. During this outbreak, 49 people with confirmed Zika virus infection and 59 cases of probable Zika virus infection occurred. Although initially thought to be an outbreak of dengue virus, serologic analysis and detection of Zika virus RNA by reverse transcriptase PCR (RT-PCR) confirmed the outbreak to be Zika virus. Notably, during the 2007 outbreak, no patients with neurologic manifestations were noted, and no deaths occurred; whether this represented a true absence of neurotropic illness or limitations in surveillance for such manifestations is unclear.
Following the 2007 Yap epidemic, Zika virus was occasionally identified in a number of South Asian and Oceanic territories. From October 2013 onward, however, reports of outbreaks of Zika virus appeared in several islands in the South Pacific. The most notable of these was a Zika virus outbreak in French Polynesia, in which more than 30,000 suspected cases of human Zika virus infection were identified, with cases occurring between October 2013 and April 2014. In the context of this outbreak, clinicians at the only tertiary hospital on the islands began to see a surprising increase in GBS cases. During the outbreak period, 42 cases of GBS were diagnosed, among a population of approximately 280,000 (FIGURE 12-1). The incidence of GBS during this time was many times greater than the expected background incidence of GBS of 1.2 per 100,000 population per year. Nearly all the patients with GBS demonstrated IgM or IgG seropositivity to Zika virus, and many GBS cases possessed neutralizing antibodies to Zika virus. A case-control study demonstrated that GBS cases were more likely to be seropositive for Zika virus than afebrile controls matched by age, sex, and island of residence; on the other hand, no difference was seen between GBS cases and controls for dengue IgG antibodies. Notable findings among the GBS cases included a rapid progression to clinical nadir (with a median of 6 days) and a short plateau phase (median of 4 days). Thirty-seven patients with GBS underwent electrodiagnostic studies, and the findings were thought to be compatible with the acute motor axonal neuropathy (AMAN) phenotype of GBS. However, less than 50% of patients demonstrated the presence of antiganglioside antibodies. The conclusion, however, was that a strong association seemed to exist between Zika virus infection and GBS.
Following the French Polynesian outbreak, Zika virus continued to circulate among Pacific islands, causing several cases of human illness in New Caledonia, Vanuatu, the Cook Islands, and Easter Island. In outbreaks in South America starting in 2015, Zika virus became hyperendemic in some areas, leading to large outbreaks of human illness. In the spring of 2015, a large outbreak of febrile rash illness occurred in Bahia in northeastern Brazil. The illness was characterized by fever, maculopapular rash, muscle and joint pain, and conjunctivitis. Zika virus was identified in several patients by RT-PCR, and chikungunya virus was identified in several people as well, indicating cocirculation of these viruses. The Zika virus identified in northern Brazil was 99% homologous with the viral envelope protein sequence from French Polynesian viruses. Very quickly, outbreaks of rash illness were identified in several separate locations in northeastern Brazil. Soon the virus had spread throughout much of Brazil, and areas with large Zika virus outbreaks began reporting higher than expected rates of GBS, similar to the outbreak in French Polynesia. As of December 2017, 440,000 to 1,300,000 Zika virus infections were estimated in Brazil. In addition, ongoing outbreaks were identified in multiple South American and Central American countries and the Caribbean.
Within this setting, increases in cases of GBS in areas with Zika virus epidemics continued to be reported from multiple countries. Several investigations were begun to systematically and objectively assess this possible association. An investigation in the city of Salvador in northeastern Brazil identified 50 cases of GBS meeting Brighton Collaboration criteria for GBS, with onset between January 1 and August 31, 2015. The Brighton Collaboration criteria were designed to assign levels of diagnostic certainty for GBS. Level 1 (representing the highest degree of diagnostic certainty) includes cases with clinical features compatible with GBS, electrodiagnostic studies consistent with GBS, and albuminocytologic dissociation (elevated CSF protein with white cell counts ≤50 cells/mm3). Level 2 includes cases with clinical GBS features and either electrodiagnostic studies consistent with GBS or albuminocytologic dissociation, while Level 3 (the lowest level of diagnostic certainty) rests on clinical findings alone. The annualized incidence of GBS in this population was 5.6 per 100,000 population, a dramatic increase over expected baseline levels. A case-control investigation showed that patients with GBS were more likely to demonstrate symptoms compatible with Zika virus infection (rash, conjunctivitis, myalgia) than age-matched and sex-matched controls. Since this investigation was a retrospective analysis, with persons having acute GBS illness approximately 6 months before the investigation, laboratory evidence of recent or prior Zika virus or dengue infection was lacking. However, being a case patient was significantly associated with evidence of recent flavivirus infection when combined with clinical criteria for suspected Zika virus disease (FIGURE 12-2).
Another case series from Colombia identified over 58,790 suspected Zika virus cases and 2630 laboratory-confirmed Zika virus cases occurring between October 2015 and March 2016. This included 270 cases of GBS with a history of Zika virus infection. The case series was able to recruit 68 of these 270 patients with GBS in whom 66 (97%) had symptoms consistent with Zika virus infection in the 4 weeks before the onset of GBS symptomatology. The other two cases did not report Zika virus–suggestive symptoms but were residents of a region experiencing a large Zika virus outbreak. Of the 68 patients with GBS, 42 (62%) had testing for Zika virus by RT-PCR. Among these, 17 (40%) tested positive for Zika virus RNA; urine was the specimen most frequently positive, but the investigators identified Zika virus RNA in CSF in three samples and in serum in one sample. In addition, 37 GBS cases were tested for the presence of anti–Zika virus antibodies; 32 (86%) had serologic evidence of a recent flavivirus infection. Based upon the laboratory testing in this investigation and including clinical compatibility with Zika virus, the classification of Zika virus infection was definite in 17 patients, probable in 18 patients, and suspected in 33 patients. One notable finding from this investigation was the period from the onset of Zika virus symptoms to the onset of GBS. The median period of time was not given in the article, but the authors noted that two patients did not have any signs or symptoms of Zika virus infection before GBS onset (suggesting asymptomatic infection could result in GBS), and two patients had the simultaneous onset of Zika virus infection and GBS.
A separate investigation assessed the occurrence of GBS in the setting of Zika virus outbreaks in seven Central American and South American countries. This investigation estimated that between April 1, 2015, and March 31, 2016, 164,237 confirmed and suspected cases of Zika virus illness and 1474 cases of GBS were reported from a number of countries. The cases of GBS from each country were compared to baseline rates of GBS historically obtained through each country’s public health surveillance. The estimates suggested the incidence of Zika virus illness during the investigation period was associated with increases in the incidence of GBS. When GBS rates were compared to pre–Zika virus baseline incidence, the increases were substantial: in Bahia State, Brazil, an increase of 172%; in Colombia, 211%; in the Dominican Republic, 150%; in El Salvador, 100%; in Honduras, 144%; in Suriname, 400%; and in Venezuela, 877%. The incidence of Zika virus illness and GBS appeared to mimic one another: increases of Zika virus disease were associated with increases in GBS incidence, and decreases in Zika virus incidence were associated with declines in GBS cases. Of course, the degree of increased incidence in some countries may have been because of underrecognition of GBS cases during the pre–Zika virus periods. Altogether, these and other recently published studies seem to indicate a strong, if not causal, association between Zika virus and GBS.
As of April 2018, over 95 countries throughout the Americas, Africa, Asia, the Caribbean, and the Pacific Islands have had evidence of Zika virus transmission or are thought to have a great risk of Zika virus transmission. Travel-related cases became a substantial problem in areas considered not at risk for Zika virus, including the United States, Canada, and countries in Europe. In the United States, sustained but limited transmission of Zika virus was identified in Miami-Dade County in Florida and in limited areas of south Texas; however, outbreaks of the magnitude seen in Central America and South America are not expected in the United States.
Modes of Transmission
The natural mode of transmission of Zika virus is through Aedes species mosquitoes. A bite from an infected mosquito to a human may result in infection; this infected person can then pass the virus on to another mosquito who takes a blood meal. This pattern continues until sufficient amplification of the virus has occurred to lead to sustained transmission, and, in areas in which people are naïve to the virus, epidemics can occur. Humans and monkeys appear to be the only viable vertebrate hosts for Zika virus; the virus has not been identified among small mammals or other animals. In addition to this natural transmission mode, other modes of transmission have been recognized. Perinatal transmission, presumably through the placenta, has been considered to be the cause of congenital Zika virus syndrome; evidence also exists that Zika virus can be transmitted through breast milk or by a blood-borne route. Sexual transmission, primarily from the semen of infected men with one report of transmission from a woman to a man, has also been documented. Zika virus appears to be able to cause sustained infection of the testes, which can then cause semen to carry the virus. This has led to recommendations that males who have traveled to a Zika virus–infected area use condoms or abstain from sex for 6 months. Women who are pregnant or are considering becoming pregnant should avoid traveling to Zika virus–endemic areas and similarly use condoms with partners or abstain from sex for 8 weeks after returning from travel.
As with most arboviruses, the vast majority of Zika virus infections are considered to be asymptomatic or mild enough that the infected individual does not require medical attention. The clinical features and presentation of Zika virus appear to have changed (or are more readily recognized) in recent years. In various investigations, nearly all the phenotypes of GBS have been described in association with Zika virus infection, including acute inflammatory demyelinating polyradiculoneuropathy (AIDP), AMAN, Fisher syndrome, and uncommon variants such as acute motor-sensory axonal neuropathy (AMSAN) and pure sensory illness. The investigation in French Polynesia described the majority of its cases as AMAN. However, less than 50% of these patients demonstrated antiganglioside antibodies. Most subsequent assessments of Zika virus–associated GBS in which electrodiagnostic studies have been performed suggest that AIDP is the most common phenotype associated with neurologic illness. The reason for this discrepancy is not clear, but it appears that in Central America, South America, and the Caribbean, most Zika virus–associated GBS is AIDP.
Zika virus–associated GBS appears to have several distinct features that differ from GBS triggered by other antecedent events. Several investigations have alluded to a short latency between the onset of Zika virus signs and symptoms and the onset of GBS. Investigations in Brazil and Colombia have estimated this latency to be a median of 6 days in Brazil and 7 days in Colombia. Another investigation in Colombia estimated the median latency period to be 7 days (interquartile range 3 to 10 days), suggesting it was shorter than expected, and, in several cases, the onset appeared to be consistent with a parainfectious process, with Zika virus signs and symptoms and GBS onset on the same day. Similarly, a shorter interval between GBS onset and clinical nadir has been suggested in other studies in Colombia, with median intervals of 4 days. Whether this represents a unique pathogenesis by Zika virus is currently unknown. Although not consistent, several investigations of Zika virus–associated GBS have found a marked preponderance of females rather than the slight male predominance typically associated with GBS. Another somewhat atypical feature of Zika virus–associated GBS is a prominent and notable increase in older age groups evident in some studies. The incidence of GBS increases with age, but an investigation in Brazil found that in age-stratified cases, the age group–specific incidence increased with age, most dramatically in older age groups. The incidence was 1.5 cases per 100,000 population for those 12 to 19 years of age, 3.9 for those 20 to 39 years of age, 7.3 for those 40 to 59 years of age, and 14.7 for those older than 60 years of age. The reason for this dramatic increase in incidence between the 40 to 59 years of age group and those 60 years or older is unknown but appears to be much larger than expected (FIGURE 12-3). This dramatic increase in incidence in the older age groups was also seen among the cases in French Polynesia (as noted by Timothee Dub, MPH, and Henri-Pierre Mallet, MD, Direction de la Sante, Bureau de Veille Sanitaire, Papeete French Polynesia, by email communication, June 2016). Another notable feature of Zika virus–associated GBS is prominent cranial nerve palsies, in particular bilateral facial nerve palsies. In Colombia, one investigation found that 50% of 66 patients with Zika virus–associated GBS had bilateral facial weakness. An assessment in Puerto Rico found that 62% of 71 patients with Zika virus–associated GBS had facial weakness, in some cases bilateral. Other cranial nerves appeared to be largely spared in these investigations (CASE 12-1).
A 65-year-old woman was admitted to the hospital because of a 2-day history of numbness and tingling in her hands and feet which progressed to weakness in her legs and arms. One week earlier she had developed headache, fever, rash on her chest, and myalgia. She was a resident of the United States but frequently traveled to Jamaica, where she had recently spent 2 weeks at her beach house.
On admission, she was noted to have a peripheral right facial droop. Cough and gag reflexes were diminished. Motor examination showed significant (3/5) weakness in both arms and legs, proximal worse than distal, with absent deep-tendon reflexes. Plantar response was flexor bilaterally. T2-weighted and fluid-attenuated inversion recovery (FLAIR) brain MRI demonstrated several small periventricular hyperintensities; postcontrast MRI of the spine noted showed hyperintense signal of nerve roots from T4 through T8. Upon admission, her pulse oximetry reading was 79%; given a concern about impending respiratory failure, she was transferred to the intensive care unit (ICU) and intubated.
CSF collected on ICU admission demonstrated a white blood cell count of 21 cells/mm3 and a protein of 256 mg/dL. Limited electrodiagnostic studies demonstrated diminished amplitude of compound muscle action potentials (CMAPs); prolonged distal latencies and F waves; and normal nerve conduction velocities in the right and left median, ulnar, and fibular (peroneal) nerves. A diagnosis of acute inflammatory demyelinating polyradiculoneuropathy (AIDP) was made, and she was started on IV immunoglobulin (IVIg) 0.5 g/kg/d for 5 days. Two days after completing the IVIg therapy, she began to have more movement in her extremities, arms greater than legs. Her breathing improved, she was extubated, and extremity strength continued to improve.
Serologies for cytomegalovirus, Epstein-Barr virus, Mycoplasma pneumoniae, and hepatitis E were negative, and a stool sample was negative for enteroviruses and Campylobacter jejuni. Given her potential exposure risks, pre-IVIg serum, urine, and CSF were sent to a public health laboratory for Zika virus testing. Serum and CSF polymerase chain reaction (PCR) were negative; urine tested PCR positive. Serum and CSF both tested IgM positive for Zika virus; she tested IgM negative and IgG positive for dengue virus. Over the next week, limb and facial weakness improved, and she was discharged to inpatient rehabilitation with a diagnosis of Zika virus–associated Guillain-Barré syndrome (GBS) with facial weakness.
This case illustrates that even in areas where Zika virus activity has not been reported, in cases of suspected GBS, it is necessary to ask for a travel history, as patients may have traveled to a Zika virus–endemic area. In all cases of suspected GBS, it is important to collect biological samples (serum, CSF, and urine) before the initiation of IVIg or plasma exchange. In addition to Zika virus, this patient was also tested for related flaviviruses (dengue, which she was also probably exposed to in Jamaica) as well as other more common antecedents of GBS.
The prognosis and overall outcome of Zika virus–associated GBS remain unclear. Anecdotal reports have suggested that Zika virus–associated GBS is more severe than GBS associated with other antecedent antigenic stimuli and that the long-term prognosis is poor. However, to date, systematic and objective investigations into the outcomes of Zika virus–associated GBS have not been completed. Some evidence from Puerto Rico suggests that long-term disability scores at 6 months do not differ between Zika virus–associated GBS cases and GBS cases associated with other stimuli. A long-term outcomes investigation conducted in Barranquilla, Colombia, assessed 40 patients approximately 1.4 years after their acute illness. Scores on the Hughes GBS Disability Scale and Overall Disability Sum Score were consistent with long-term outcomes data on GBS in published literature, and scores on the Zung Depression Scale and Health Related Quality of Life-4 (HRQoL-4) Questionnaire were not different between GBS cases and a sampling of the normal population in Barranquilla (based on data collected by Diana Walteros, MD, Instituto Nacional de Salud, Ministry of Health, Colombia, and colleagues, discussed in person, November 2017). Thus, early and preliminary data suggest that long-term outcomes and prognosis of Zika virus–associated GBS do not differ in severity from GBS due to other triggers. However, additional assessments among differing populations are needed to substantiate these findings.
Other neurologic manifestations of Zika virus infection have also been described; however, these findings have generally appeared as case reports or small case series, and the exact relationship to Zika virus infection is still questionable. It is safe to say that other neurologic syndromes reported to be temporally associated with Zika virus infection have not been found to be anywhere near the magnitude of that found with GBS. Acute myelitis was reported in a young girl from Guadeloupe 7 days after Zika virus infection; spinal MRI revealed signal hyperintensities in the cervical and thoracic cord. Zika virus RNA was detected in her serum, urine, and CSF on day 2 of neurologic onset, 9 days after acute Zika virus infection. Meningoencephalitis has also been reported in temporal association with Zika virus infection. An elderly man presented with fever and altered mental status 10 days after returning to France from a trip to New Caledonia, Vanuatu, the Solomon Islands, and New Zealand. Brain MRI demonstrated findings consistent with acute meningoencephalitis; CSF demonstrated a polymorphonuclear pleocytosis (41 cells/mm3) and slightly elevated protein (76 mg/dL). Zika virus RNA was identified in his CSF. In both these reports, identification of Zika virus RNA in CSF would seem to suggest a neurotropic potential of Zika virus. However, it is puzzling why, in the face of extremely large Zika virus outbreaks in many areas, reports of noncongenital neurotropic Zika virus infections appear to be so scarce. More surveillance is needed to determine whether Zika virus truly does lead to neuroinvasive illness.
Congenital Zika Syndrome
The other substantial neurologic issue associated with Zika virus is the apparent association between Zika virus infection in pregnant women and resultant congenital malformations in infants born to them. In September 2015, clinicians observed a substantial increase in the number of cases of microcephaly among infants born in northeastern Brazil. Shortly thereafter, other areas of Brazil with heavy Zika virus activity were found to have a higher incidence of microcephaly than expected. Virologic evidence of an association has come from viral isolation and RT-PCR positivity in the amniotic fluid of women who were pregnant with infants with microcephaly and isolation from the brains of fetuses with congenital abnormalities. One investigation in Brazil followed pregnant women who had a rash suggestive of Zika virus infection with onset in the prior 5 days, and specimens including blood and urine were obtained for RT-PCR testing. These women were followed prospectively, with clinical data collected. Altogether, 88 women were examined, of whom 72 (82%) tested positive for Zika virus in blood, urine, or both. Forty-two of the women who were Zika virus positive (58%) and all the women who were Zika virus negative were evaluated using fetal ultrasonography. Twelve of the 42 women who were Zika virus positive and none of the women who were Zika virus negative had fetal abnormalities detectable by ultrasonography. A case-control study of infants born in Paraiba, Brazil, during the Zika virus outbreak found a significant association between microcephaly and Zika virus infection in infants and Zikalike symptoms in the mother during the first trimester. Over time, in addition to microcephaly, a wide range of congenital abnormalities have been observed in infants prenatally affected by Zika virus, including decreased brain parenchymal volume, lissencephaly, ventriculomegaly, cerebral calcifications, choriomeningitis, arthrogryposis, and various ocular findings. An association seems to exist between the time of Zika virus infection during the pregnancy and the severity of congenital malformations, with infection of the mother during the first or second trimester having more serious consequences for the infant. An investigation using prenatal brain MRI of fetuses with Zika virus infection observed severe cerebral damage, with lesions similar to those seen with congenital infections with cytomegalovirus and lymphocytic choriomeningitis virus. The author concluded that Zika virus had particular tropism for the fetal brain germinal matrix. The number of infants born with these malformations is likely to result in a tremendous public health burden on countries and severe implications for the caregivers of these children. In April 2016, the Centers for Disease Control and Prevention (CDC) and the WHO concluded that a causal association exists between prenatal Zika virus infection and severe congenital abnormalities.
The diagnosis of Zika virus infection begins with clinical suspicion based upon clinical features and epidemiology. The clinical features of mild Zika virus infection are very similar to the presentations of other related arboviruses, including West Nile virus, dengue, and chikungunya virus. This challenge is only increased by the fact that these arboviruses are often in cocirculation, making clinical diagnoses more difficult. The incubation period for Zika virus is approximately 3 to 14 days, following which patients may develop an array of signs and symptoms, including fever, headache, myalgia, arthralgia, conjunctivitis, and rash. However, it is likely no clinical features can reliably distinguish Zika virus infection from other arboviral infections. Clinical suspicion of Zika virus infection should be high in persons developing these various signs and symptoms in areas with known Zika virus transmission. Suspicion of Zika virus infection should also be considered in persons with such signs and symptoms who live in or have recently traveled to areas with Zika virus transmission. In Zika virus–endemic regions, GBS should prompt clinical suspicion of Zika virus.
Laboratory diagnosis of Zika virus infection can be challenging. Viremia in humans is thought to be brief, thus limiting the utility of molecular diagnostic methods. However, despite these limitations, acute-phase diagnosis of Zika virus relies on molecular testing; RT-PCR testing may be able to detect viral nucleic acids early in the course of illness and should be performed on paired serum and urine specimens within 14 days of illness onset for individuals who are symptomatic. Serologic diagnosis rests on the identification of Zika virus–specific IgM antibodies in serum. Significant serologic cross-reactivity with other closely related flaviviruses presents challenges for serologic diagnosis. Guidance for US laboratories testing for Zika virus infection was updated in July 2017.
Treatment and Management
Currently, no specific treatments have been established for Zika virus or any other arboviral infection. Anecdotally, a myriad of treatments have been attempted, including corticosteroids, interferon alfa, ribavirin, and others. Management of acute Zika virus infection is supportive, with analgesics to relieve headache and myalgia and attention to rash to prevent cutaneous problems. Treatment of Zika virus–associated GBS is the same as for GBS associated with any other antigenic stimulus; currently no evidence shows that treatment efficacy is any different for Zika virus–associated GBS than for any other form of GBS. IV immunoglobulin (IVIg) or plasma exchange should be initiated as soon as the diagnosis of GBS is made; these treatments are more efficacious when given early in the course of GBS. Typical regimens for IVIg are 0.4 g/kg/d to 0.5 g/kg/d for 5 days; for plasma exchange, 1.5 exchange volumes every other day for 5 days is recommended.
Unanswered Questions About Zika Virus
The emergence of Zika virus raises many questions; what led to changes in epidemiology and clinical features (or newly recognized modes of transmission and clinical manifestations) in recent outbreaks, the reason for such strong associations with GBS and congenital abnormalities, and the future epidemiologic patterns of Zika virus on a global scale are only a few. Since late 2017, Zika virus transmission appears to have decreased in most areas in Central America, South America, and the Caribbean; however, the future epidemiologic patterns of Zika virus are unknown. Ongoing surveillance is needed to monitor virus prevalence and detect future outbreaks. Further investigations are needed to determine how Zika virus infection may result in GBS and how the virus manages to cross the placenta and cause congenital abnormalities in infected fetuses. In coming years, it is hoped that many questions regarding Zika virus will be answered.
OTHER SELECT ARBOVIRAL CENTRAL NERVOUS SYSTEM INFECTIONS
In addition to Zika virus, numerous other arboviruses cause significant neurologic illness in humans. A list of medically important arboviruses can be found in TABLE 12-1 . The text below focuses on two of the most important neurotropic arboviruses: Japanese encephalitis virus and West Nile virus.
Japanese Encephalitis Virus
Japanese encephalitis virus is the most common cause of arboviral epidemic encephalitis worldwide. Approximately 70,000 cases and 15,000 deaths are reported annually; however, this is almost certainly an underestimate. Japanese encephalitis virus is transmitted by Culex species mosquitoes, predominantly Culex tritaeniorhynchus, and is associated with periodic large epidemics as well as endemic transmission in much of South Asia, Southeast Asia, and the Western Pacific. The amplification cycle for Japanese encephalitis virus involves infected Culex mosquitoes and pigs, which are the primary source of viremic blood for mosquitoes, and waterbirds, which help to amplify Japanese encephalitis virus in the environment. Humans are considered to be dead-end hosts, as viremia generally never gets high enough to permit human-mosquito-human transmission.
Human illness from Japanese encephalitis virus infection occurs after a 1- to 2-week incubation period, and, as in many arboviruses, most infections are thought to be asymptomatic. In endemic areas, children are most susceptible to Japanese encephalitis virus illness, although adult cases can and do occur; adults seem to have less severe illness than children. Symptomatic infection with Japanese encephalitis virus may include a prodromic illness of fever, headache, arthralgia or myalgia, and gastrointestinal symptoms. In persons progressing to severe neurologic illness, decreased or altered mental status, sometimes progressing to coma, will follow, with features of encephalitic illness. Seizures are a common feature of Japanese encephalitis, with up to 85% of infected cases demonstrating seizures at some point in the illness. Poorer outcomes are associated with multiple seizures, prolonged seizures, or status epilepticus. A distinctive feature of Japanese encephalitis (and other flaviviral encephalitides) is prominent extrapyramidal signs, including facial masking, tremor, generalized hypertonia, and, in some cases, choreoathetosis and dystonia. These extrapyramidal signs appear to be due to a predilection of Japanese encephalitis virus for the basal ganglia and thalami, which has been demonstrated by MRI (FIGURE 12-4). A poliomyelitislike anterior myelitis has also been described, although less commonly than encephalitis. This syndrome is due to virus neurotropism for the anterior horn (motor) neurons of the spinal cord. Spinal MRI in these cases will demonstrate abnormal signal intensities of the anterior cord. Anterior myelitis may occur in the absence of encephalitis. The overall case fatality rate in Japanese encephalitis can be as high as 30%, with an additional 50% of cases experiencing long-term neurologic and functional sequelae, which can be lifelong. Children appear to have worse outcomes and more frequent sequelae than adults.
In Japanese encephalitis, CSF will generally demonstrate a moderate lymphocytic pleocytosis and moderately elevated protein. EEG is generally nonspecific, with generalized slowing indicative of encephalopathy and features of seizure activity when present; EEG may also be useful in identifying nonconvulsive status epilepticus. In cases of anterior myelitis, electrodiagnostic studies will demonstrate low-amplitude CMAPs and prolonged distal latencies with normal nerve conduction velocities, consistent with anterior horn damage. Definitive diagnosis of Japanese encephalitis virus infection can be made by detection of viral nucleic acid by RT-PCR of serum or CSF (or brain tissue in fatal cases); however, viremia in Japanese encephalitis is transient, and thus PCR is insensitive. Most diagnoses of Japanese encephalitis rest on detection of Japanese encephalitis virus–specific IgM antibodies in serum or CSF by enzyme-linked immunosorbent assay (ELISA) or a fourfold rise in antibody titers between acute and convalescent sera collected 2 to 3 weeks apart.
Treatment of Japanese encephalitis is supportive, and no definitive therapy has been identified. A therapeutic trial of interferon alfa-2a treatment for Japanese encephalitis virus infection was found to be ineffective and, in fact, led to worse outcomes. Seizures should be managed appropriately by antiepileptic medications. Increased intracranial pressure should be assessed and managed to prevent herniation. Complications such as bedsores, aspiration pneumonia, and contractures should be prevented and managed.
Prevention of Japanese encephalitis virus infection rests on avoidance of mosquito bites by minimizing exposure during peak Culex activity (dusk and dawn), wearing protective long-sleeved shirts and long pants when feasible, and using mosquito repellant. Vaccines for Japanese encephalitis virus are available and are used in endemic and epidemic areas and for travelers going to endemic regions. Older Japanese encephalitis virus vaccines, which were manufactured using inoculation of suckling mouse brains, are effective but carry risks of autoimmune neurologic adverse events. Newer Japanese encephalitis virus vaccines manufactured in a cell culture–based system have a safer profile and have been replacing older brain-derived vaccines.
West Nile Virus
West Nile virus, another flavivirus closely related to Japanese encephalitis virus, was historically associated with sporadic cases or small outbreaks of mild febrile exanthematous illness. The historical distribution of West Nile virus was throughout Africa, the Middle East, parts of Europe, Russia, and South Asia. Epidemics of West Nile virus had occurred periodically, including a large outbreak of febrile illness in South Africa in 1974, an outbreak of febrile illness and neurologic illness in Romania in 1996, and an outbreak of febrile illness in Israel in 1999 to 2000. In 1999, West Nile virus suddenly and unexpectedly emerged in North America, leading to 62 cases of illness in New York City. Subsequent years saw a dramatic and rapid spread of the virus across the United States and parts of Canada, with large numbers of cases of febrile illness and neuroinvasive disease peaking in 2003. As of 2016, over 46,000 West Nile virus disease cases (>21,000 West Nile virus neuroinvasive disease cases) and over 2000 deaths were reported to the CDC. The incidence of West Nile virus has stabilized in recent years following an outbreak in 2012.
West Nile virus is spread via an enzootic cycle involving Culex species as the principal mosquito vector, with avian (and, to a lesser extent, equine) principal amplifying hosts. As in Japanese encephalitis virus infection, humans are considered dead-end hosts for West Nile virus, as human infection does not result in human-mosquito-human transmission. Important Culex vectors include Culex pipiens and Culex quinquefasciatus in the eastern United States and Culex tarsalis in the western United States.
As with other flaviviruses, most infections with West Nile virus are thought to be asymptomatic, with approximately 20% of infected people developing mild febrile illness (West Nile fever) and less than 1% going on to develop severe neurologic illness (West Nile neuroinvasive disease). In symptomatic infections, illness occurs after an incubation period of approximately 2 to 14 days, following which most symptomatic individuals develop a febrile illness with headache, myalgia, arthralgia, nausea, vomiting, and, in some cases, rash. West Nile fever is generally self-limited and mild, but some individuals with West Nile fever require medical attention for more severe symptoms. The small percentage of people who go on to develop West Nile neuroinvasive disease may present with aseptic meningitis, including meningeal signs of nuchal rigidity, Kernig sign, Brudzinski sign, and photophobia or phonophobia. West Nile neuroinvasive disease is characterized by decreased or altered mental status, often with cranial nerve abnormalities. As in Japanese encephalitis, patients with West Nile encephalitis may develop an extrapyramidal syndrome, including masked facies, tremor, and myoclonus. A subset of patients may develop a poliomyelitislike anterior myelitis, with acute flaccid paralysis that is typically asymmetric (FIGURE 12-5, CASE 12-2). Seizures are less common in West Nile neuroinvasive disease than in Japanese encephalitis but can occur. Older individuals and persons with compromised immune status are more likely to develop West Nile neuroinvasive disease than other individuals. CSF is characterized by a mild pleocytosis, which can be predominantly neutrophilic in a surprisingly significant number of cases. MRI in West Nile encephalitis will frequently demonstrate signal abnormalities in the basal ganglia and thalami, similar to that in Japanese encephalitis (FIGURE 12-6). Clinical outcomes of West Nile virus encephalitis are generally more favorable than that of Japanese encephalitis virus, with an overall case fatality rate of 5% to 10% in most assessments. Severe neurologic sequelae are also less frequent and can be transient, although patients with anterior myelitis generally have significant ongoing limb weakness.
A 30-year-old man from Colorado presented to the emergency department because of a 1-day history of fatigue and progressive weakness. His symptoms began 2 days after a mild fever, severe headaches, and rigors. He worked outside as a construction worker, typically starting before dawn. He had no significant past medical history.
Upon presentation, his examination was notable for moderate (4/5) weakness in his proximal right arm and more severe weakness (3/5) in his proximal right leg. Deep tendon reflexes were absent on the right and diminished on the left.
Because of a concern for Guillain-Barré syndrome (GBS), he was immediately started on IV immunoglobulin (IVIg), 0.5 g/kg/d for 5 days; serum, CSF, and urine were collected before initiating IVIg. CSF demonstrated a white blood cell count of 238 cells/mm3, predominantly neutrophils, and a protein of 321 mg/dL. Testing of serum for a standard viral panel was negative. Electrodiagnostic studies on day 2 of weakness were normal. He completed the course of IVIg with no improvement. Electrodiagnostic studies repeated 8 days later demonstrated absent responses in the right leg and diminished compound muscle action potentials (CMAPs), increased distal latencies, and slowed nerve conduction velocities in the right median and ulnar nerves. Sensory nerve conduction studies were normal. Spinal MRI was performed and demonstrated signal hyperintensity involving the anterior horn region of the cord. He was transferred for inpatient rehabilitation.
The acute serum and CSF, as well as serum drawn 10 days later, were sent for West Nile virus testing. Acute serum and CSF were both positive on polymerase chain reaction (PCR) testing for West Nile virus. The acutely drawn serum was negative for West Nile virus–specific IgM; the serum drawn 10 days after admission was positive for West Nile virus–specific IgM.
Six months after discharge from the hospital, his arm weakness had greatly improved, with some remaining weakness; his leg weakness had improved as well, but he still required a cane for walking.
This is a typical presentation of West Nile virus anterior myelitis. Several things point to this as the diagnosis rather than GBS. The asymmetry of the weakness, the rapid onset and worsening of the weakness, and the CSF profile of a neutrophilic pleocytosis and elevated protein suggest a neuroinvasive illness in contrast with the albuminocytologic dissociation seen with GBS. The electrodiagnostic studies performed shortly after admission were normal; electrodiagnostic changes generally do not manifest until 7 to 10 days after the onset of neurologic illness. The physicians in this case prudently repeated the studies 8 days after onset, and the findings suggested involvement of motor axons, anterior horn cells, or both, leading to the diagnosis of anterior myelitis.
The diagnosis of West Nile virus infection is similar to that of Japanese encephalitis virus infection. Identification of viral nucleic acid by RT-PCR is generally not helpful in symptomatic human illness because of the transient viremia. Most diagnoses rest on identification of West Nile virus–specific IgM antibodies in serum or CSF or on a fourfold rise in antibody titers between acute and convalescent serum. Because of serologic cross-reactivity with other flaviviruses in persons who have been infected with or received vaccines for other flaviviruses, positive IgM results should be confirmed by neutralizing antibody testing at a state public health laboratory or the CDC. Because of potential transmission of West Nile virus through blood transfusions, all donated blood in the United States and Canada is now tested for West Nile virus through nucleic acid amplification testing.
No definitive treatment has been established for West Nile virus; monoclonal antibodies against West Nile virus showed promise for treatment if given very shortly after infection but proved to be logistically infeasible for actual treatment of patients. West Nile virus vaccines have been developed for equines, but no vaccine is available for humans. A comparison of the features of West Nile virus and Japanese encephalitis virus is presented in TABLE 12-2.
Other Neurotropic Arboviruses
A list of some of the other more common arboviruses that can lead to neurologic illness, including their associated geography and clinical syndromes, may be found in TABLE 12-1. Many of these viruses are associated with very rare, sporadic cases of human illness. Others, however, such as LaCrosse virus and tick-borne encephalitis virus, can be associated with significant human disease. LaCrosse virus is a common cause of pediatric encephalitis in the eastern United States, with several cases routinely reported to the CDC each season. Tick-borne encephalitis virus has a geographic range through Central and Eastern Europe and the Caucasus and is a common cause of encephalitis and acute flaccid myelitis among those at risk (persons spending considerable periods of time outside, with an increased risk of exposure to ticks). A vaccine is available to prevent tick-borne encephalitis virus infection. Powassan virus, another tick-borne virus, seems to be increasingly recognized in the northern United States and southern Canada. Information on the other, less common, viruses is available in the article referenced here.
Arboviruses will continue to cause human neurologic illness in the future. The unexpected rise in and clinical manifestations of Zika virus and West Nile virus in the recent past should remind us that vigilance is needed to identify future emerging arboviral pathogens. The future epidemiologic pattern of these viruses is unclear, and subsequent outbreaks can be expected. It is hoped that ongoing outbreaks of Japanese encephalitis may be mitigated by greater use of effective vaccines, but full control is likely still a way off. All three of these viruses and other etiologies of arboviral encephalitides will remain important public health problems for the foreseeable future.
- Zika virus is a member of the genus Flavivirus, which includes several medically important neurotropic viruses such as Japanese encephalitis virus, West Nile virus, and St. Louis encephalitis virus.
- In urban and suburban environments, Aedes mosquitoes transmit the Zika virus to humans in a mosquito-human-mosquito cycle in which an infected Aedes mosquito feeds on a blood meal on a human and transmits the virus, after which an infected human, still viremic with the virus, is bitten by another mosquito, thus continuing the cycle of transmission.
- Recently published studies seem to indicate a strong, if not causal, association between Zika virus and Guillain-Barré syndrome.
- The pattern of mosquito-human transmission of Zika virus continues until sufficient amplification of the virus has occurred to lead to sustained transmission, and, in areas in which people are naïve to the virus, epidemics can occur.
- Perinatal transmission, presumably through the placenta, has been considered to be the cause of congenital Zika virus syndrome; evidence also exists that Zika virus can be transmitted through breast milk or by a blood-borne route. Sexual transmission, primarily from the semen of infected men with one report of transmission from a woman to a man, has also been documented.
- Zika virus appears to be able to cause sustained infection of the testes, which can then cause semen to carry the virus. This has led to recommendations that males who have traveled to a Zika virus–infected area use condoms or abstain from sex for 6 months.
- Women who are pregnant or are considering becoming pregnant should avoid traveling to Zika virus–endemic areas and use condoms with partners or abstain from sex for 8 weeks after returning from travel.
- Nearly all of the phenotypes of Guillain-Barré syndrome have been described in association with Zika virus infection, including acute inflammatory demyelinating polyradiculoneuropathy, acute motor axonal neuropathy, Fisher syndrome, and uncommon variants such as acute motor-sensory axonal neuropathy and pure sensory illness.
- Several investigations have alluded to a short latency between the onset of Zika virus signs and symptoms and the onset of Guillain-Barré syndrome.
- The incidence of Guillain-Barré syndrome increases with age, but an investigation in Brazil found that in age-stratified cases, the age group–specific incidence increased with age, most dramatically in older age groups.
- A notable feature of Zika virus–associated Guillain-Barré syndrome is prominent cranial nerve palsies, in particular, bilateral facial nerve palsies.
- In addition to microcephaly, a wide range of congenital abnormalities have been observed in infants prenatally affected by Zika virus, including decreased brain parenchymal volume, lissencephaly, ventriculomegaly, cerebral calcifications, choriomeningitis, arthrogryposis, and various ocular findings.
- An association seems to exist between the time of Zika virus infection during the pregnancy and the severity of congenital malformations, with infection of the mother during the first or second trimester having more serious consequences for the infant.
- Acute-phase diagnosis of Zika virus relies on molecular testing; reverse transcriptase polymerase chain reaction testing may be able to detect viral nucleic acids early in the course of illness and should be performed on paired serum and urine specimens within 14 days of illness onset for individuals who are symptomatic.
- Treatment of Zika virus–associated Guillain-Barré syndrome is the same as for Guillain-Barré syndrome associated with any other antigenic stimulus; currently no evidence shows that treatment efficacy is any different for Zika virus–associated Guillain-Barré syndrome than for any other form of Guillain-Barré syndrome.
- IV immunoglobulin or plasma exchange should be initiated as soon as the diagnosis of Guillain-Barré syndrome is made; these treatments are more efficacious when given early in the course of Guillain-Barré syndrome.
- Japanese encephalitis virus is the most common cause of arboviral epidemic encephalitis worldwide.
- Humans are considered to be dead-end hosts for Japanese encephalitis virus, as viremia generally never gets high enough to permit human-mosquito-human transmission.
- In endemic areas, children are most susceptible to Japanese encephalitis virus illness, although adult cases can and do occur; adults seem to have less severe illness than children.
- Symptomatic infection with Japanese encephalitis virus may include a prodromic illness of fever, headache, arthralgia or myalgia, and gastrointestinal symptoms. In persons progressing to severe neurologic illness, altered mental status, sometimes progressing to coma, will follow, with features of encephalitic illness.
- A distinctive feature of Japanese encephalitis (and other flaviviral encephalitides) is prominent extrapyramidal signs, including facial masking, tremor, generalized hypertonia, and, in some cases, choreoathetosis and dystonia.
- A poliomyelitislike anterior myelitis has also been described with Japanese encephalitis virus, although less commonly than encephalitis. Anterior myelitis may occur in the absence of encephalitis.
- Definitive diagnosis of Japanese encephalitis virus infection can be made by detection of viral nucleic acid by reverse transcription polymerase chain reaction of serum or CSF (or brain tissue in fatal cases); however, viremia in Japanese encephalitis is transient, and thus polymerase chain reaction is insensitive.
- Most diagnoses of Japanese encephalitis virus rest on detection of Japanese encephalitis virus–specific IgM antibodies in serum or CSF by enzyme-linked immunosorbent assay or a fourfold rise in antibody titers between acute and convalescent sera collected 2 to 3 weeks apart.
- West Nile virus is spread via an enzootic cycle involving Culex species as the principal mosquito vector, with avian (and, to a lesser extent, equine) principal amplifying hosts. As in Japanese encephalitis virus infection, humans are considered dead-end hosts for West Nile virus, as human infection does not result in human-mosquito-human transmission.
- Most infections with West Nile virus are thought to be asymptomatic, with approximately 20% of infected people developing mild febrile illness (West Nile fever) and less than 1% going on to develop severe neurologic illness (West Nile neuroinvasive disease).
- The small percentage of people who go on to develop West Nile neuroinvasive disease may present with aseptic meningitis, including meningeal signs of nuchal rigidity, Kernig sign, Brudzinski sign, and photophobia or phonophobia.
- West Nile neuroinvasive disease is characterized by decreased or altered mental status, often with cranial nerve abnormalities. As in Japanese encephalitis, patients with West Nile encephalitis may develop an extrapyramidal syndrome, including masked facies, tremor, and myoclonus.
- A subset of patients with West Nile neuroinvasive disease may develop a poliomyelitislike anterior myelitis, with acute flaccid paralysis that is typically asymmetric.
- Older individuals and persons with compromised immune status are more likely to develop West Nile neuroinvasive disease than other individuals.
Dr Sejvar prepared this work as part of his official duties as an employee of the Centers for Disease Control and Prevention.