West Nile Virus

West Nile virus (WNV) is one of the most important arboviral diseases in the United States [1]; hundreds to thousands of cases have been reported to the CDC yearly since 2002. WNV is a positive sense, single-stranded, enveloped RNA flavivirus transmitted to humans and other mammals via the Culex mosquito [2]. Originally isolated in Uganda in 1937 [3], it was first seen in the U.S. in 1999, when a case was reported in New York. Since then, it has spread across the country and, in 2012, involved every one of the contiguous 48 states for the first time, with the largest number of deaths reported to date [4]. The basis of this particular outbreak is unknown but thought to be due to an early, hot summer that resulted in an abundance of highly infective vector mosquitoes [5, 6].

The vast majority of WNV infections are asymptomatic and, as such, are unreported. Symptomatic disease falls into two categories: neuroinvasive and nonneuroinvasive. Neuroinvasive disease, which comprises fewer than 1 % of infections [5] but for which the fatality rate is around 10 % [4], presents as meningoencephalitis, myelitis, radiculitis, and/or neuritis (Fig. 1). Symptomatically, patients suffer encephalopathy, headache, radicular pain, and/or flaccid paralysis. Risk factors for the development of neuroinvasive disease include immune suppression, age over 50, and, peculiarly, mutations in CCR5 [3]. WNV in humans is frequently lethal from respiratory insufficiency, and a recent animal model of WNV has shed light on why: Neurons with the orexin 1 receptor protein in the upper cervical cord were diminished, and infection also colocalized to medullary neurons with respiratory functions [7].

Fig. 1
figure 1

Magnetic resonance imaging in a patient with neuroinvasive West Nile virus disease. Axial fluid-attenuated inversion recovery (FLAIR, left) imaging of the cerebellum shows hyperintensity of the bilateral dentate nuclei (arrows), and T1-weighted imaging after administration of intravenous gadolinium (right) shows enhancement of the cerebellar folia (arrow heads)

Full size image

Diagnosis of WNV neuroinvasive disease can be made by WNV RNA detection in the cerebrospinal fluid (CSF); however, this has low sensitivity unless performed early in the course of infection, and thus CSF serologies are used adjunctively. In light of evidence of viral persistence in kidneys, attention has been given to the detection of WNV RNA in urine to increase the diagnostic accuracy of symptomatic disease [8], but this is not clinically readily available as yet.

No definitive treatment exists for WNV neuroinvasive disease, and long-term neurologic sequelae are common in survivors. Additionally, a recent study has connected WNV neuroinvasive disease with chronic kidney disease, although it is not clear yet whether this is an epiphenomenon [9]. Vaccine development and treatment with interferon, monoclonal antibodies, ribavirin, and specific West Nile immune globulin have been largely understudied and/or disappointing [10, 11]. Regardless, pathologic evidence of rampant inflammation teleologically supports the use of immune modulation, and several case reports support the use of WNV-specific IVIg [12]. Aggressive prevention remains the best recommendation for disease control [13]. For additional detail on WNV encephalitis, the reader is referred to the accompanying review in this issue by Venkatesan.

Dengue Virus in Asia and the U.S.

The dengue virus (family Flaviviridae, genus Flavivirus, species Dengue virus) is transmitted by the mosquito Aedes aegypti, as well as a second vector, Aedes albopictus, and is common in the southeastern U.S. Dengue virus attaches to the host cell surface and then enters the cell by receptor-mediated endocytosis. Several primary and low-affinity coreceptors have been identified. Dengue virus is one of the most common vector-borne viral diseases in the world, causing an estimated 50 million infections and 25,000 deaths each year. Recently, cases of dengue fever (DF) have increased significantly in Southeast Asia and in the U.S., and we review this emergence.

There are four serotypes of dengue virus and, thus, 12 possible sequences of consecutive infection by two different dengue serotypes. Each sequence can result in a different spectrum of disease.

Infection ranges from an asymptomatic or mild self-limited influenza-like illness (DF) to a very severe disease with spontaneous hemorrhage (dengue hemorrhagic fever, DHF) or, most seriously, to dengue shock syndrome, characterized by circulatory failure. Patients typically develop DHF when they become infected consecutively with a different strain of dengue virus. In DHF, there is greatly increased vascular permeability and plasma leakage, thrombocytopenia, and coagulopathy. This results in widespread hemorrhage throughout the skin, gut, and oropharynx.

No locally acquired cases of dengue virus had been reported in the U.S. since 1945. In 2009, however, a New Yorker who had just returned from a trip to Key West in Florida but who had not traveled outside of the U.S. developed DF. Twenty-four additional cases were later found in Key West during 2009, and additional cases continue to appear. The extent of this emerging infection in the U.S. remains uncertain, but dengue should be considered in all patients with suspected CNS infection with an appropriate travel history or potential local exposure, especially those who have impaired consciousness or seizures. In 2012, the FDA approved a new molecular test to detect the presence of dengue virus in people with symptoms of DF or DHF: the CDC DENV-1-4 Real Time RT-PCR Assay.

There is currently no specific antiviral treatment for dengue virus, but recognizing that a patient has been infected is important for trying to prevent a second infection that might lead to hemorrhagic fever. Paula Luz and colleagues modeled dengue transmission [14] and suggested that the dengue burden could be significantly reduced through high-efficacy larval control and the control of adult mosquitoes. Other potentially effective strategies include source reduction, such as targeting mosquito breeding sites and the use of larvicides or aerosolised adulticides.

Borrelia miyamotoi

Ticks of the genus Ixodes are vectors of Borrelia burgdorferi, the etiologic spirochete of Lyme disease. These ticks can also simultaneously transmit other infections, such as babesiosis and anaplasmosis. In 1995, it was discovered that they can concomitantly harbor another spirochete, Borrelia miyamotoi [15]. Although the discovery was made in Japan, ticks throughout North America and Europe are also carriers. In the U.S., the prevalence of B. miyamotoi-infected ticks is about 10 % that of B. burgdorferi [16].

Human infection by B. miyamotoi was first described in Russia in 2011, with symptoms of fever, headache, myalgias, and elevated aminotransferases [17]. In 2013, infection was reported in the U.S. [18], including a case of meningoencephalitis in an immune compromised host [19] who had xanthochromic CSF with a lymphocytic pleocytosis and spirochetes on Gram’s stain that were identified as B. miyamotoi by PCR. The patient was treated successfully for 30 days with penicillin because of ceftriaxone allergy, although ceftriaxone is the treatment of choice for CNS Borrelia infections. Notably, the relationship between immune suppression and CNS disease by B. miyamotoi remains to be determined as more is learned about this emerging infection.

Tick-Borne Encephalitis Virus and Powassan Virus

Tick-borne encephalitis virus (TBEV) and Powassan virus (POWV), like B. miyamotoi, are carried by the Ixodes tick [12]. They are genetically similar flaviviruses, with single-stranded, positive-sense RNA genomes. TBEV is the most important arboviral disease in Europe and central and eastern Asia, causing more than 10,000 hospitalizations each year [20]. Latvia has been the epicenter of infections [21]. Powassan virus is found in the northeastern U.S > and East Asia. In the U.S., 16 cases of POWV were reported in 2011 and 7 in 2012 [1, 22], but this disease is thought to be underreported [23].

TBEV and POWV clinically manifest after an incubation period of 1–4 weeks with a febrile illness that is followed days to weeks later by meningitis, brainstem encephalitis, myelitis, and/or radiculitis. Treatment is supportive. Notably, an effective vaccine is available for prevention of TBEV, but rates of vaccination throughout endemic areas remain rather low. The highest vaccination coverage is in Austria, where some 85 % of the population has received at least one dose [20]. Increased awareness, perhaps in the setting of recent addition of TBEV to the list of reportable diseases uniformly across Europe [24], is hoped to boost vaccination rates and prevent disease and death.

La Crosse Virus

La Crosse virus (LCV) is a single-stranded, negative-sense, circularized virus carried by the Aedes tresariatus mosquito and, hence, typically occurs as a summertime infection. It is endemic to the U.S.—specifically, in the Mid- and South Atlantic regions and the Midwest. In 2012, 71 cases were reported, with one death; 83 % of the cases were in children under the age of 18 [22]. The virus produces neuroinvasive disease similar to WNV of aseptic meningitis, encephalitis, and/or flaccid paralysis, with encephalitis being the most common manifestation. Diagnosis is made by demonstrating LCV IgM in the CSF, since viral detection is elusive. Treatment is supportive, and long-term neurologic deficit is the rule in survivors.

HIV-2 Encephalitis

HIV-2, a less virulent retrovirus than HIV-1, originated from the sooty mangabey in West Africa. Human HIV-2 infection is prevalent in West Africa, with an estimated 1–2 million infected [25]. The course of HIV-2 is more indolent than HIV-1, with many surviving decades asymptomatically. Eventually, however, CD4 counts drop, placing patients at risk for the same opportunistic infections as in HIV-1. HIV-2 is thought to be particularly neurotropic [26], but there has been minimal focus on CNS disease caused by HIV-2.

In 2012, Wood et al. reported on a case of biopsy-proven encephalitis due to HIV-2 in a Massachusetts patient who presented with executive dysfunction, impulsivity, and memory impairment [27]. He had emigrated from Cape Verde 30 years prior. Imaging showed a bifrontal, enhancing lesion involving the genu of the corpus callosum, which is not a typical imaging pattern for encephalitis due to HIV-1. The patient was treated with antiretrovirals and had significant immune recovery and arrest in imaging abnormalities.

Given the immigration patterns of West Africans and the indolent nature of HIV-2, there is potential for similar cases to emerge outside of West Africa [28, 29], with several implications. First, special diagnostic testing is required, since routine antibody testing can be indeterminate [27]. Next, antiretroviral regimens must be chosen carefully—specifically, not to include first-generation nonnucleoside reverse transcriptase inhibitors or enfuviritide given intrinsic resistance [28]. The significance of the imaging dissimilarity between the isolated case reported by Wood and what is typical for HIV-1 is unknown but raises concern for disparity in neurologic effects between HIV-2 and HIV-1.

Cycloviruses

The cycloviruses are a group of viruses from the Circoviridae family, with a circular genome that has recently been detected in patients with severe disease in Vietnam and in Malawi. Cycloviruses were first discovered in 2009 in fecal samples from children in Southeast Asia with acute flaccid paralysis. These viruses may also cause encephalitis. Van Dorn et al. sequenced the genetic material in CSF) samples from patients with undiagnosed CNS infections. Approximately 4 % of the samples tested positive for Circoviridae [30]. Investigations of paraplegia in Malawi disclosed that from a sample of 58 paraplegia patients, cycloviruses were found in 15 % of serum and 10 % of CSF samples [31].

Influenza A(H1N1)pdm09

Influenza is a negative sense, segmented RNA orthomyxovirus that is typed as A, B, or C and further subtyped by hemagluttinins and neuraminidases. It is transmitted via direct contact or respiratory droplets. Although the virus is ubiquitous and most commonly causes respiratory illness seasonally, periodically there is emergence of particularly virulent strains that reach pandemic status. In 2009, a particularly deadly outbreak of H1N1 that made the jump from swine to humans started in Mexico and quickly spread across the world. This particular strain gave rise to neurologic complications mainly in children [32, 33]. However, a fatal dural venous sinus thrombosis with edema and intraparenchymal hemorrhage was reported in an adult; autopsy demonstrated viral RNA positivity in brain tissue despite negative viral RNA and no o pleocytosis in CSF [34]. That being said, the exact rate and nature of neurological complications from this pandemic are largely unknown [35].

Since treatment with neuraminidase inhibitors in influenza is only variably successful, prevention is the focus, with reduction of respiratory transmission and proper hand hygiene paramount [35]. The high mortality during the outbreak of H1N1 prompted rapid development and distribution of a preventive vaccine, but this approach was accompanied by a small but measurable increased risk of Guillain-Barre syndrome (GBS) after receipt of the vaccine [36]. Notably, however, GBS is also a potential complication of the infection itself [35], and the risk of the disease-associated GBS is much greater than vaccine-associated GBS. This, combined with the risk of severe disease from influenza itself, clearly establishes that the benefits of the vaccine outweigh its risks [37].

Henipaviruses

Nipah and Hendra virus are negative sense, single-stranded, enveloped RNA paramyxoviruses whose reservoir is the Pteropus fruit bat, found in Southeast Asia, Madagascar, India, and parts of Australia [38]. Clinically, infections start with flu-like symptoms progressing to respiratory disease and/or encephalitis after an incubation time of 1–2 weeks [39, 40]. Hendra virus primarily infects horses and is highly fatal; human acquisition from exposure to infected horses has been reported in seven instances, all in Australia, with four deaths [38, 39, 41]. Nipah virus, on the other hand, primarily infects swine, but swine-to-human and human-to-human transmission, as well as acquisition from ingestion of date palm sap contaminated with Pteropus urine or saliva, have occurred [42]. Outbreaks involving hundreds of cases, with a mortality of 40 %–100 % [39, 40] have been reported. The first outbreak of 265 cases of encephalitis with 105 deaths was in Malaysia in 1998. Since, there have been three outbreaks of Nipah in India and one in Singapore, but the majority have been in Bangladesh. The most recent outbreak was in early 2012 in Bangladesh, when 6 cases—all fatal—were reported [39].

There are no effective treatments for these viruses. Ribavirin use has been evaluated in a nonrandomized study during the Malaysian Nipah outbreak and was found to reduce mortality [39], but its use in Hendra virus has not improved outcomes [39, 41]. Recently, however, a promising monoclonal antibody has been developed [39].

Alphaviruses

Chikungunya virus and Eastern equine encephalomyelitis (EEE) virus are positive-sense, single-stranded RNA viruses that belong to this genus. Both are mosquito-borne but have very disparate distributions. EEE is endemic to the northeastern United States and is carried by the Culiseta mosquito. Chikungunya is seen in Africa and Southeast Asia and is carried by the Aedes mosquito [43]. Notably, however, there has been a recent emergence of Aedes albopictus, a known chikungunya vector, in the southeastern U.S., raising concern for its introduction into this country [44].

Symptoms of chikungunya infection are typically influenza-like with an occasional rash. Encephalitis and death are rare but do occur. There are no effective treatments, although a vaccine may be in development [44].

EEE is a much deadlier disease, with a case fatality rate of up to 75 % [45]. The most common manifestation is encephalitis or meningoencephalitis, but rarely, flaccid paralysis can be seen. Over the past 10 years, there has been an overall increase in the number of cases reported from New England, ranging from 1 to 11 per year [45]; in 2012 15 cases were reported, with a 33 % case fatality rate [22]. The reasons for this are unknown, and as with other emerging encephalitis arboviruses, effective treatment is lacking, again making prevention key.

Poliovirus

Wild poliovirus cases had reached their lowest point in 2012 [46], and the World Health Organization (WHO) believed that it might finally be possible to eradicate polio, as had been done for smallpox (see Table 1). Local changes in security and a breakdown in civil authority can obviously impede vaccination efforts. For example, just very recently, new cases of wild poliovirus have been reported in the Horn of Africa, the bulk in Somalia or Kenya. Some of the newly reported cases are from south-central Somalia, where access for supplementary immunization activities has been compromised for the past 3 years because of the insecurity there. As many as 500,000 children in this area are at risk of developing polio. Efforts are ongoing to operate in this area, and vaccinations are continuing at entry and exit points to build up herd immunity levels.

Table 1 Wild poliovirus cases
Full size table

Corticosteroids in Bacterial Meningitis

In an update of the 2009 Cochrane review, Brouwer et al. examined the effect of adjuvant corticosteroid therapy versus placebo on mortality, hearing loss, and neurological sequelae in people of all ages with acute bacterial meningitis. Corticosteroids significantly reduced hearing loss and neurological sequelae but did not reduce overall mortality. Data support the use of corticosteroids in patients with bacterial meningitis in high-income countries. The authors found no beneficial effect in low-income countries [47]. Koopmans et al. found that the likelihood of a poor outcome in adults with Listeria monocytogenes meningitis is very high and has actually increased over a 14-year period, from 27 % to 61 %. The emerging L. monocytogenes genotype ST6 was identified as the principal negative predictor. Adjunctive dexamethasone should not be used in Listeria meningitis [48].

Drug-Resistant Tuberculous Meningitis

Drug-resistant tuberculosis, including drug-resistant tuberculous meningitis (TBM), is an emerging health problem in many countries and is typically associated with a high mortality. Creating difficulties for clinical identification and optimal clinical management, the pathology, clinical features, and neuroimaging characteristics of drug-resistant TBM are similar to those of drug-responsive TBM. Garg et al. reviewed the field and established that an association with Beijing strains and drug-resistance-related mutations, such as mutations in katG and rpoB genes, has been found. As is discussed below, detection methods in CSF are still problematic. The optimal treatment of multidrug-resistant TBM depends on the drug susceptibility pattern of the isolate and/or the previous treatment history of the patient. Second-line drugs with good penetration of the CSF should be preferred. Isoniazid monoresistant disease requires addition of another drug with better CSF penetration [49].

New Techniques in Identifying Encephalitis

Despite efforts to improve surveillance and rapid diagnosis of encephalitis, data from projects like the the California Encephalitis Project (CEP) emphasize that the majority of cases of suspected viral encephalitis remain undiagnosed. The CEP tests for diverse encephalitides. Thus, in the first 2.5 years of testing 334 cases (1998–2000), a confirmed or probable viral agent was found in 31 cases (9 %), a bacterial agent was found in 9 cases (3 %), and a parasitic agent was found in 2 cases (1 %). A possible etiology was identified in an additional 41 cases (12 %). A noninfectious etiology was identified in 32 cases (10 %), and a nonencephalitis infection was identified in 11 (3 %). Half of these cases were due to autoimmune conditions, such as vasculitis, Hashimoto encephalopathy, and encephalitis due to antineuronal antibodies, but 62 % of the cases remained unexplained [50]. A subsequent larger study showed a similar proportion remained undiagnosed [51]. There is clearly a need to develop and validate techniques that would improve this detection rate.

New Techniques for Detecting Tuberculous Meningitis

Detection of mycobacteria in the CSF by conventional methods (smear examination or culture) is often difficult. Nucleic acid amplification assays are better methods owing to their rapidity and high sensitivity. The Xpert MTB/RIF assay (Cepheid, CA) is a fully automated nucleic acid amplification-based diagnostic system that detects Mycobacterium tuberculosis and rifampin resistance in under 2 h with very high precision [52] and is endorsed by the WHO. A recent pediatric study examined the performance characteristics of real-time PCR in TBM diagnosed on the basis of modified Ahuja criteria. CSF for TB by real-time PCR was positive in 26/40 cases (65 %) of TBM, but only 2/40 controls. With CSF culture as the gold standard, the sensitivity of RT-PCR was 63 %; specificity was 77 % [53]. For additional discussion on this technique, see the companion article in this issue of CIDR by Chin and Mateen on CNS tuberculosis.

Digital Droplet PCR Technology

This advancement in PCR technology directly amplifies and quantifies nucleic acids such as DNA, cDNA, or RNA from human samples. Droplet PCR (dPCR) reactions separate a single clinical sample into multiple wells, and the reaction is carried out in each well individually, thereby increasing sensitivity. The droplet digital PCR (ddPCR) technology is similar but partitions a sample into 20,000 droplets and quantifies nucleic acids digitally. By using Poisson statistics, the number of DNA molecules in the original sample is directly calculated from the number of positive and negative reactions, and without the need for a standard curve. It has been used to detect MRSA, adenoviruses, enteroviruses, HHV6, and other pathogens. The ddPCR assay can determine the number of HHV-6 DNA copies per cell with precision and has been used to determine whether a patient has chromosomally integrated HHV-6 [54, 55].

New Techniques for Detecting CNS Fungal Infections

The 2012 fungal CNS disease outbreak in the United States highlighted the need for improved diagnostic accuracy for nervous system fungal infections, since current PCR and culture methods have sensitivities around 30 % and 15 %, respectively [56–58]. Detection in CSF of (1,3) β-d-glucan, a glycoprotein component of most fungal cell walls that is constitutively shed by these organisms, promises marked improvement in diagnostic sensitivity (unpublished data currently under review) of CNS fungal disease. See companion article in this issue of CIDR by Lyons for further discussion of CNS mold infections.