Journal of Infection (2014) xx, 1e7
www.elsevierhealth.com/journals/jinf
West Nile virus: Should pediatricians care? Jennifer C. Smith a, Tim Mailman b, Noni E. MacDonald c,* a
Paediatric Infectious Diseases, Paediatrics Dalhousie University, IWK Health Center, 5850/5980 University Ave, Halifax, Nova Scotia B3K 6R8, Canada b Paediatrics and Microbiology, Dalhousie University, IWK Health Center, 5850/5980 University Ave, Halifax, Nova Scotia B3K 6R8, Canada c Paediatrics, Dalhousie University, IWK Health Center, 5850/5980 University Ave, Halifax, Nova Scotia B3K 6R8, Canada
Available online - - -
KEYWORDS Arbovirus; West Nile virus; West Nile virus epidemiology; West Nile virus fever; Encephalitis; Meningitis; West Nile poliomyelitis
Summary Given the recurrent serious outbreaks of West Nile Virus (WNV) in the United States over the past decade, the spread to Canada and South America, the recurrent outbreaks in Europe, and the potential for serious neurological disease even in children under 18 years, paediatricians in affected areas must consider WNV in the differential diagnosis of all children presenting with aseptic meningitis, encephalitis and acute flaccid paralysis. Additionally, given that WNV encephalitis can occur after WNV infection, suspicion for neurological WNV disease must remain high even after otherwise benign febrile illnesses if the child lives in or has traveled to an affected region. Under-diagnosis in the pediatric population is likely a serious problem, necessitating further educational efforts. More follow-up studies of WNV neurological disease in children and youth are needed to better understand the potential long-term sequelae during vulnerable times of neurodevelopment and neural remodeling. Similarly, more research is need on short and long-term fetal outcomes of maternal WNV infection. ª 2014 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
Introduction West Nile Virus (WNV) has gained increased attention in North America and parts of Europe over the past 15 years because of large outbreaks, predominately affecting adults, and for the virus’ propensity to cause neurological complications among symptomatic individuals. The question arises: should
paediatricians care? This overview provides evidence for why pediatricians should, indeed, care.
Virus details WNV is an enveloped, single-stranded, positive-sense RNA virus that belongs to the genus Flavivirus and the family
* Corresponding author. Tel.: þ1 902 470 8799; fax: þ1 902 470 7232. E-mail addresses: [email protected] (J.C. Smith), [email protected] (T. Mailman), [email protected] (N.E. MacDonald). http://dx.doi.org/10.1016/j.jinf.2014.07.019 0163-4453/ª 2014 The British Infection Association. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Smith JC, et al., West Nile virus: Should pediatricians care?, J Infect (2014), http://dx.doi.org/ 10.1016/j.jinf.2014.07.019
2 Flaviviridae. Classified as an arbovirus (arthropod-borne virus), WNV is transmitted by infected mosquitoes, and possibly by ticks, to vertebrates causing disease in both humans and animals. WNV is related to the Japanese encephalitis virus complex that includes the infectious agents responsible for Japanese encephalitis, Murray Valley encephalitis, and St. Louis encephalitis.1 The WNV genome measures approximately 11 kB in length and encodes a single polyprotein, that is post-translationally cleaved by host and virus proteases into three structural (capsid (C), premembrane (prM)/membrane (M), and envelope (E)) and seven nonstructural (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) proteins.2 The use of phylogenetic analysis, based on nucleotide sequence data, has identified the existence of at least two and potentially five distinct WNV genetic lineages3 (Fig. 1).
Epidemiology/history WNV likely evolved in Africa during the past millennium, separating from other members of the Japanese encephalitis virus complex and subsequently diverging into individual WNV lineages.4 WNV was first isolated from a woman with an undiagnosed febrile illness in the West Nile region of Uganda in 1937. During the 1950se1980s, WNV caused several outbreaks of relatively mild febrile illness in regions of Africa, the Middle East, India, and Australia. However, starting in the mid-1990s more frequent outbreaks, with increased severity of associated neuroinvasive disease, were observed in the Middle East and Eastern Europe.5 WNV first appeared in the United States in 1999, as an outbreak of 62 cases in New York City.6 Genomic sequencing
Figure 1 West Nile Virus phylogenetic tree.3 (Figure 3 from Brault AC. Changing patterns of West Nile virus transmission: altered vector competence and host susceptibility. Vet Res. 2009; 40: 43).
J.C. Smith et al. of the virus demonstrated genetic similarity to a WNV strain isolated from Israel in 1998.7 Subsequent to the original outbreak in 1999, human WNV cases were reported across the United States, with a peak of 9862 cases reported in 2003.8 In 2012, 5674 WNV cases were reported, the highest number since 2003.9 Since 2001, WNV has also spread northward to Canada as well as southward into Latin America and the Caribbean.10,11 In contrast to other locales with WNV, bird mortality has only been reported for WNV transmission in North America12 and Israel.13 The ability to alternate replication between vertebrate and invertebrate hosts, termed the “trade-off hypothesis”, may help explain the adaptation success of WNV upon arrival in the United States. This hypothesis proposes that WNV exchanges superior fitness in a single host for the ability to replicate in two disparate hosts.14 Although WNV, like other RNA viruses, is more prone to replication errors and higher mutation rates than DNA viruses,4 the use of a two-host system counters the rapid genetic evolution of WNV.14
Epidemiologic triangle The traditional model of infectious disease causation, known as the epidemiologic triangle, can be useful in understanding the epidemiology of WNV disease (Fig. 2). WNV is generally transmitted in a bird-mosquito cycle with mosquitoes serving as the vector and birds, through developing prolonged high-level viremia, as the amplifying host.15 While some female mosquitoes are ornithophilic (i.e. they only feed on birds), other mosquitoes feed on birds and other species transmitting WNV to humans, horses, and other mammals. The resultant viremia tends to be mild (e.g. short-duration, low titer) and insufficient to infect mosquitoes. Therefore, mammals, including humans are incidental rather than true WNV hosts.16 While mosquito-to-human transmission is the primary route of human infection, transmission has also occurred through blood transfusion,17 organ transplantation,18 transplacental transmission19 and possibly, breastfeeding.20 Transmission may also occur through dialysis21 and occupational exposure to WNV.22 Screening of blood products in many countries has served to substantially decrease the risk of WNV transmission23 while the screening of transplant organs for WNV remains controversial.24 Multiple factors, rather than a single mechanism, likely supported the recent spread of WNV in North and South America including: the availability of competent mosquito vectors and susceptible avian hosts, viral adaptation, and climatic conditions (e.g. warmer summers attributable to climate change).3,25,26 Given this degree of complexity, it is difficult to predict future changes in WNV epidemiology.26 Hence, surveillance efforts and public health measures (e.g. education of the public, timing of blood product screening) must remain highly sensitive to changes in patterns of WNV epidemiology.
Pathogenesis While the pathogenesis of WNV infections is not completely understood, a 2013 review by Suthar and colleagues
Please cite this article in press as: Smith JC, et al., West Nile virus: Should pediatricians care?, J Infect (2014), http://dx.doi.org/ 10.1016/j.jinf.2014.07.019
West Nile virus: Should pediatricians care?
3
Figure 2 West Nile Virus classic epidemiological triangle [adapted from Ref. 3].
provides a suggested pattern with host responses at different stages.2WNV is capable of evading the innate immune system by blocking toll-like receptors, disrupting type I interferon signaling and interfering with complement activation pathways.2 As well, mosquito saliva can counteract the host immune system response by making the host cytokine milieu more favorable to viral survival.27 Exactly how WNV evades the immune system to enter the brain and spinal cord is unclear but once inside the body WNV directly invades neurons. Resultant neuronal damage likely occurs from a combination of direct viral-mediated destruction and indirect host-mediated immune responses.15 Within the brain, a combination of interferon responses, alongside CD4þ and CD8þ involvement, helps clear WNV.2
WNV infection Asymptomatic infection and West Nile Fever (WNF) The clinical patterns associated with WNV infection are listed in Table 1. Like other arboviruses, 70e80% of WNV infections are asymptomatic followed by presumed lifelong immunity.15 However, unlike many arboviruses, WNV does
Table 1
West Nile viral infections [adapted from Ref.
15,29,32,35
not cause hemorrhagic fever. The most common manifestation of WNV is “flu”-like illness, West Nile Fever (WNF), which is seen in about 20% of infected individuals.15 The incubation period for WNV is 2e15 days but can be longer in immunocompromised hosts. WNF is characterized by abrupt onset of high fever, headache, myalgia and fatigue. In about 25% of affected individuals, nausea, vomiting and diarrhea may be prominent, sometimes leading to dehydration. On physical examination, a transient macular rash lasting less than a week and lymphadenopathy may be found. Acute symptoms usually last for only 3e7 days but fatigue may be prolonged. The median time for full recovery is 60 days. As with asymptomatic infection, lifelong immunity is presumed to follow WNF. Longer follow-up of patients with WNF has shown decreased quality of life at 18 months, as well as some functional impairment, but no objective cognitive sequelae.28
West Nile Neurological Disease (WNND) While less than 1% of cases of WNV infection are complicated by neurological involvement (Table 1), WNV has become the leading cause of neuroinvasive arboviral disease in the United States.29 West Nile Neurological Disease (WNND) is less common in children and youth less than 18 years of age than in adults. Adults with underlying conditions such as chronic renal disease, cancer, alcohol abuse, diabetes and/or hypertension have higher rates.30 Additionally, an association has been found between certain genetic differences in the interferon response pathway and WNND.31 Patients with WNV meningitis (WNM), which comprises about 35e40% of WNND, have typical findings of viral meningitis with headache, fever, stiff neck, and predominance of lymphocytes in the cerebrospinal fluid although neutrophils may also be present. In WNV encephalitis (WNE), which makes up roughly 55e60% of WNND, the findings are more global with marked lethargy, confusion, and decreased level of consciousness with or without focal neurologic deficits such as limb paralysis, cranial nerve palsies, tremors or movement disorders.29 Seizures however, are uncommon. WNV poliomyelitis syndrome (WNP) is very uncommon, making up an estimated 5e10% of WNND. WNP tends to
]. West Nile virus neurological disease (WNND)
% of WNV infection % WNND Estimated decreased quality of life and functional loss at 18 months post Estimated Fatality Rate Post infection presumed lifelong immunity
Asymptomatic infection West Nile virus (WNV)
West Nile f ever (WNF)
West Nile meningitis (WNM)
West Nile encephalitis (WNE)
West Nile Poliomyelitis (WNP)
70e80% N/A N/A
20% 50%
www.elsevierhealth.com/journals/jinf
West Nile virus: Should pediatricians care? Jennifer C. Smith a, Tim Mailman b, Noni E. MacDonald c,* a
Paediatric Infectious Diseases, Paediatrics Dalhousie University, IWK Health Center, 5850/5980 University Ave, Halifax, Nova Scotia B3K 6R8, Canada b Paediatrics and Microbiology, Dalhousie University, IWK Health Center, 5850/5980 University Ave, Halifax, Nova Scotia B3K 6R8, Canada c Paediatrics, Dalhousie University, IWK Health Center, 5850/5980 University Ave, Halifax, Nova Scotia B3K 6R8, Canada
Available online - - -
KEYWORDS Arbovirus; West Nile virus; West Nile virus epidemiology; West Nile virus fever; Encephalitis; Meningitis; West Nile poliomyelitis
Summary Given the recurrent serious outbreaks of West Nile Virus (WNV) in the United States over the past decade, the spread to Canada and South America, the recurrent outbreaks in Europe, and the potential for serious neurological disease even in children under 18 years, paediatricians in affected areas must consider WNV in the differential diagnosis of all children presenting with aseptic meningitis, encephalitis and acute flaccid paralysis. Additionally, given that WNV encephalitis can occur after WNV infection, suspicion for neurological WNV disease must remain high even after otherwise benign febrile illnesses if the child lives in or has traveled to an affected region. Under-diagnosis in the pediatric population is likely a serious problem, necessitating further educational efforts. More follow-up studies of WNV neurological disease in children and youth are needed to better understand the potential long-term sequelae during vulnerable times of neurodevelopment and neural remodeling. Similarly, more research is need on short and long-term fetal outcomes of maternal WNV infection. ª 2014 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
Introduction West Nile Virus (WNV) has gained increased attention in North America and parts of Europe over the past 15 years because of large outbreaks, predominately affecting adults, and for the virus’ propensity to cause neurological complications among symptomatic individuals. The question arises: should
paediatricians care? This overview provides evidence for why pediatricians should, indeed, care.
Virus details WNV is an enveloped, single-stranded, positive-sense RNA virus that belongs to the genus Flavivirus and the family
* Corresponding author. Tel.: þ1 902 470 8799; fax: þ1 902 470 7232. E-mail addresses: [email protected] (J.C. Smith), [email protected] (T. Mailman), [email protected] (N.E. MacDonald). http://dx.doi.org/10.1016/j.jinf.2014.07.019 0163-4453/ª 2014 The British Infection Association. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Smith JC, et al., West Nile virus: Should pediatricians care?, J Infect (2014), http://dx.doi.org/ 10.1016/j.jinf.2014.07.019
2 Flaviviridae. Classified as an arbovirus (arthropod-borne virus), WNV is transmitted by infected mosquitoes, and possibly by ticks, to vertebrates causing disease in both humans and animals. WNV is related to the Japanese encephalitis virus complex that includes the infectious agents responsible for Japanese encephalitis, Murray Valley encephalitis, and St. Louis encephalitis.1 The WNV genome measures approximately 11 kB in length and encodes a single polyprotein, that is post-translationally cleaved by host and virus proteases into three structural (capsid (C), premembrane (prM)/membrane (M), and envelope (E)) and seven nonstructural (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) proteins.2 The use of phylogenetic analysis, based on nucleotide sequence data, has identified the existence of at least two and potentially five distinct WNV genetic lineages3 (Fig. 1).
Epidemiology/history WNV likely evolved in Africa during the past millennium, separating from other members of the Japanese encephalitis virus complex and subsequently diverging into individual WNV lineages.4 WNV was first isolated from a woman with an undiagnosed febrile illness in the West Nile region of Uganda in 1937. During the 1950se1980s, WNV caused several outbreaks of relatively mild febrile illness in regions of Africa, the Middle East, India, and Australia. However, starting in the mid-1990s more frequent outbreaks, with increased severity of associated neuroinvasive disease, were observed in the Middle East and Eastern Europe.5 WNV first appeared in the United States in 1999, as an outbreak of 62 cases in New York City.6 Genomic sequencing
Figure 1 West Nile Virus phylogenetic tree.3 (Figure 3 from Brault AC. Changing patterns of West Nile virus transmission: altered vector competence and host susceptibility. Vet Res. 2009; 40: 43).
J.C. Smith et al. of the virus demonstrated genetic similarity to a WNV strain isolated from Israel in 1998.7 Subsequent to the original outbreak in 1999, human WNV cases were reported across the United States, with a peak of 9862 cases reported in 2003.8 In 2012, 5674 WNV cases were reported, the highest number since 2003.9 Since 2001, WNV has also spread northward to Canada as well as southward into Latin America and the Caribbean.10,11 In contrast to other locales with WNV, bird mortality has only been reported for WNV transmission in North America12 and Israel.13 The ability to alternate replication between vertebrate and invertebrate hosts, termed the “trade-off hypothesis”, may help explain the adaptation success of WNV upon arrival in the United States. This hypothesis proposes that WNV exchanges superior fitness in a single host for the ability to replicate in two disparate hosts.14 Although WNV, like other RNA viruses, is more prone to replication errors and higher mutation rates than DNA viruses,4 the use of a two-host system counters the rapid genetic evolution of WNV.14
Epidemiologic triangle The traditional model of infectious disease causation, known as the epidemiologic triangle, can be useful in understanding the epidemiology of WNV disease (Fig. 2). WNV is generally transmitted in a bird-mosquito cycle with mosquitoes serving as the vector and birds, through developing prolonged high-level viremia, as the amplifying host.15 While some female mosquitoes are ornithophilic (i.e. they only feed on birds), other mosquitoes feed on birds and other species transmitting WNV to humans, horses, and other mammals. The resultant viremia tends to be mild (e.g. short-duration, low titer) and insufficient to infect mosquitoes. Therefore, mammals, including humans are incidental rather than true WNV hosts.16 While mosquito-to-human transmission is the primary route of human infection, transmission has also occurred through blood transfusion,17 organ transplantation,18 transplacental transmission19 and possibly, breastfeeding.20 Transmission may also occur through dialysis21 and occupational exposure to WNV.22 Screening of blood products in many countries has served to substantially decrease the risk of WNV transmission23 while the screening of transplant organs for WNV remains controversial.24 Multiple factors, rather than a single mechanism, likely supported the recent spread of WNV in North and South America including: the availability of competent mosquito vectors and susceptible avian hosts, viral adaptation, and climatic conditions (e.g. warmer summers attributable to climate change).3,25,26 Given this degree of complexity, it is difficult to predict future changes in WNV epidemiology.26 Hence, surveillance efforts and public health measures (e.g. education of the public, timing of blood product screening) must remain highly sensitive to changes in patterns of WNV epidemiology.
Pathogenesis While the pathogenesis of WNV infections is not completely understood, a 2013 review by Suthar and colleagues
Please cite this article in press as: Smith JC, et al., West Nile virus: Should pediatricians care?, J Infect (2014), http://dx.doi.org/ 10.1016/j.jinf.2014.07.019
West Nile virus: Should pediatricians care?
3
Figure 2 West Nile Virus classic epidemiological triangle [adapted from Ref. 3].
provides a suggested pattern with host responses at different stages.2WNV is capable of evading the innate immune system by blocking toll-like receptors, disrupting type I interferon signaling and interfering with complement activation pathways.2 As well, mosquito saliva can counteract the host immune system response by making the host cytokine milieu more favorable to viral survival.27 Exactly how WNV evades the immune system to enter the brain and spinal cord is unclear but once inside the body WNV directly invades neurons. Resultant neuronal damage likely occurs from a combination of direct viral-mediated destruction and indirect host-mediated immune responses.15 Within the brain, a combination of interferon responses, alongside CD4þ and CD8þ involvement, helps clear WNV.2
WNV infection Asymptomatic infection and West Nile Fever (WNF) The clinical patterns associated with WNV infection are listed in Table 1. Like other arboviruses, 70e80% of WNV infections are asymptomatic followed by presumed lifelong immunity.15 However, unlike many arboviruses, WNV does
Table 1
West Nile viral infections [adapted from Ref.
15,29,32,35
not cause hemorrhagic fever. The most common manifestation of WNV is “flu”-like illness, West Nile Fever (WNF), which is seen in about 20% of infected individuals.15 The incubation period for WNV is 2e15 days but can be longer in immunocompromised hosts. WNF is characterized by abrupt onset of high fever, headache, myalgia and fatigue. In about 25% of affected individuals, nausea, vomiting and diarrhea may be prominent, sometimes leading to dehydration. On physical examination, a transient macular rash lasting less than a week and lymphadenopathy may be found. Acute symptoms usually last for only 3e7 days but fatigue may be prolonged. The median time for full recovery is 60 days. As with asymptomatic infection, lifelong immunity is presumed to follow WNF. Longer follow-up of patients with WNF has shown decreased quality of life at 18 months, as well as some functional impairment, but no objective cognitive sequelae.28
West Nile Neurological Disease (WNND) While less than 1% of cases of WNV infection are complicated by neurological involvement (Table 1), WNV has become the leading cause of neuroinvasive arboviral disease in the United States.29 West Nile Neurological Disease (WNND) is less common in children and youth less than 18 years of age than in adults. Adults with underlying conditions such as chronic renal disease, cancer, alcohol abuse, diabetes and/or hypertension have higher rates.30 Additionally, an association has been found between certain genetic differences in the interferon response pathway and WNND.31 Patients with WNV meningitis (WNM), which comprises about 35e40% of WNND, have typical findings of viral meningitis with headache, fever, stiff neck, and predominance of lymphocytes in the cerebrospinal fluid although neutrophils may also be present. In WNV encephalitis (WNE), which makes up roughly 55e60% of WNND, the findings are more global with marked lethargy, confusion, and decreased level of consciousness with or without focal neurologic deficits such as limb paralysis, cranial nerve palsies, tremors or movement disorders.29 Seizures however, are uncommon. WNV poliomyelitis syndrome (WNP) is very uncommon, making up an estimated 5e10% of WNND. WNP tends to
]. West Nile virus neurological disease (WNND)
% of WNV infection % WNND Estimated decreased quality of life and functional loss at 18 months post Estimated Fatality Rate Post infection presumed lifelong immunity
Asymptomatic infection West Nile virus (WNV)
West Nile f ever (WNF)
West Nile meningitis (WNM)
West Nile encephalitis (WNE)
West Nile Poliomyelitis (WNP)
70e80% N/A N/A
20% 50%
