Human West Nile Virus Lineage 2 Infection: Epidemiological, Clinical, and Virological Findings.

West Nile virus (WNV) lineage 2 is expanding and causing large outbreaks in Europe. In this study, we analyzed the epidemiological, clinical, and virological features of WNV lineage 2 infection during the large outbreak that occurred in northern Italy in 2018. The study population included 86 patients with neuroinvasive disease (WNND), 307 with fever (WNF), and 34 blood donors. Phylogenetic analysis of WNV full genome sequences from patients’ samples showed that the virus belonged to the widespread central/southern European clade of WNV lineage 2 and was circulating in the area at least since 2014. The incidence of WNND and WNF progressively increased with age and was higher in males than in females. Among WNND patients, the case fatality rate was 22%. About 70% of blood donors reported symptoms during follow-up. Within the first week after symptom onset, WNV RNA was detectable in the blood or urine of 80% of patients, while 20% and 40% of WNND and WNF patients, respectively, were WNV IgM-seronegative. In CSF samples of WNND patients, WNV RNA was typically detectable when WNV IgM antibodies were absent. Blunted or no WNV IgM response and high WNV IgG levels were observed in seven patients with previous flavivirus immunity.


Introduction
West Nile virus (WNV) is a mosquito-borne flavivirus transmitted by Culex mosquitoes among a variety of bird species [1]. Humans are dead-end hosts, incidentally infected through a mosquito bite. Infection in humans is generally asymptomatic; 20-30% of cases develop influenza-like illness, defined as West Nile fever (WNF), while less than 1% of infected individuals, mainly elderly and immunocompromised patients, develop neuroinvasive disease (WNND), i.e., encephalitis, meningitis, or acute flaccid paralysis. In patients with WNND, mortality is about 10% and severe sequelae persist in 20-40% of those who survive [2].
Nine different evolutionary lineages of WNV have been described so far, but only lineages 1 and 2 have been associated with disease in humans [3]. In Europe, the most widespread WNV strains belong to the central/southern European clade of WNV lineage 2, which emerged in central Europe in 2004 [4]. Then, the virus dispersed from Hungary to western and southern European countries, where it has caused large human outbreaks, such as in Greece, since 2010 [5], and in Italy, since 2013 [6]. and detection of IgG antibodies in serum and confirmation by neutralization assays. A probable WNV case was defined as an individual with only IgM antibodies detected in serum.
According to the regional surveillance plan, collection of the following biological samples was recommended for laboratory diagnosis of WNV infection: Whole blood and urine for WNV RNA detection within 4 weeks since symptom onset; serum at baseline and after 2 to 3 weeks for WNV IgM and IgG detection; and CSF from patients with neurological symptoms for the detection of WNV RNA and WNV IgM antibodies. In this study, the different sample types were not available from all patients, while additional follow-up samples were collected from several confirmed WNV cases.
The data analyzed in this study included all the laboratory results of confirmed autochthonous WNV cases (86 with WNND, 307 with West Nile fever (WNF), and 34 blood donors) that were obtained at the regional reference laboratory. Information on symptoms and clinical diagnosis were retrieved from case report forms.
In relation to blood donors, we included in this study the 34 donors with positive WNV NAT results who were confirmed with further laboratory testing, according to the case definition. All the 34 blood donors, either with symptomatic or asymptomatic WNV infection, were submitted to a follow-up evaluation, including baseline WNV RNA testing in blood and urine and IgM and IgG antibody detection in serum within 3 days after the index donation, and weekly evaluations during the first month and at month 6 after the index donation.
Suspected and probable cases of WNV infection were not included in this study.

Entomological Surveillance
Entomological surveillance was carried out in the Veneto Region from May to October using 55 CDC traps baited with CO 2 (IMT ® -Italian Mosquito Trap) located all over the region, excluding Belluno Province. The traps were run for one night every two weeks.

Laboratory Methods
Laboratory investigation of suspected WNV infections was carried out as previously described [21]. Briefly, for viral RNA detection, total nucleic acids were purified from 200 µL of whole blood by using a MagNA Pure 96 System (Roche Applied Sciences, Basel, Switzerland) and from 1000 µL of plasma, urine, or cerebrospinal fluid (CSF) by using a NucliSens EasyMag System (BioMerieux, Marcy-l'Étoile, France). Target viral RNA sequences were amplified by in-house real-time RT-PCR methods, which allowed discrimination between WNV lineage 1 and WNV lineage 2, as described [21]. Real-time RT-PCR assays were carried out using the one-step real-time kit (Thermo Fisher Scientific, Waltham, MA, USA) and run on ABI 7900HT Sequence Detection Systems (Thermo Fisher Scientific). In addition, in about 50% of WNV RNA-positive blood or urine samples, identification of the WNV lineage was done by broad-range RT-PCR targeting the NS5 region of flaviviruses [22], followed by cycle sequencing. In four samples with a high WNV RNA load, the full viral genome was sequenced, as reported [6].
The presence of WNV IgM and IgG antibodies in serum and CSF was determined by a commercial ELISA (WNV IgM capture DxSelect e IgG DxSelect, Focus Diagnostics, CA, USA). Serum samples with positive results were further tested for confirmation by the plaque reduction neutralization test (PRNT) against WNV and the microneutralization titer assay against USUV [12].
For entomological surveillance, mosquitoes were morphologically identified, pooled (100 specimens maximum), and screened for flaviviruses by using an in-house developed one-step SYBR green-based real-time RT-PCR assay, as described [9]. All flavivirus RNA-positive mosquito pools were directly sequenced to differentiate WNV, USUV, or other flaviviruses.

Statistical Analysis
Data were presented as mean value ± standard deviation or 95% confidence interval and as a median and interquartile range (IQR). Statistical analyses were conducted using unpaired Student's Viruses 2020, 12, 458 4 of 14 t-test, factorial ANOVA, and χ 2 test. The statistical significance was defined as p < 0.05. All statistical analyses were performed by using the Statistica™ software, version 13.4.0.14 (TIBCO Software Inc., Paolo Alto, CA).

Ethics Statement
The cases reported in this study were investigated with routine procedures according to the national surveillance plan for WNV and USUV infection. Therefore, no approval was required from the ethics committee.

Demographic and Epidemiological Analyses of Human Cases of WNV Infection
In the Veneto region, northern Italy, during the surveillance period (1 June -30 November 2018), of the 1967 cases of suspected autochthonous arbovirus infection, 427 were classified as confirmed WNV infections based on the laboratory test results. These confirmed cases included 86 patients with WNND, 307 with WNF, and 34 blood donors (Table 1). Nineteen patients with WNND died, accounting for 22% of the case fatality rate. The mean age of patients with WNND was significantly higher than that of patients with WNF and the prevalence of WNV disease was significantly higher in males than in females (Table 1). Normalization of the number of WNV cases against demographic data of the population in the Veneto region showed that the incidence of WNF progressively increased with age, while WNND incidence sharply peaked in subjects older than 70 years of age ( Figure 1). Data also showed that the incidence of WNND and WNF was higher in males than in females among patients aged >60 years ( Figure 1).

Male
10 (53) -- Normalization of the number of WNV cases against demographic data of the population in the Veneto region showed that the incidence of WNF progressively increased with age, while WNND incidence sharply peaked in subjects older than 70 years of age ( Figure 1). Data also showed that the incidence of WNND and WNF was higher in males than in females among patients aged >60 years ( Figure 1).  In 2018, there was an anticipation and a longer duration of the WNV transmission period than in previous years ( Figure 2). In particular, the first human case of WNV infection developed symptoms of encephalitis in mid-June and the last case developed symptoms in mid-October. The largest number of cases of infection occurred between mid-July and mid-August, with a peak during the last week of July. The total number of West Nile cases identified in the Veneto region in 2018 was 20-fold higher than the average number of cases identified in the previous two years ( Figure 2). However, considering the incidence of WNV infection in blood donors, which was not affected by variables related to the intensity of surveillance, the increase of WNV cases in 2018 compared to 2016-2017 was 7-fold. The intense training and information to physicians and the population on WNV infection and control measures during the 2018 summer season probably contributed to the increased identification of WNV infections.  The total number of West Nile cases identified in the Veneto region in 2018 was 20-fold higher than the average number of cases identified in the previous two years ( Figure 2). However, considering the incidence of WNV infection in blood donors, which was not affected by variables related to the intensity of surveillance, the increase of WNV cases in 2018 compared to 2016-2017 was 7-fold. The intense training and information to physicians and the population on WNV infection and control measures during the 2018 summer season probably contributed to the increased identification of WNV infections.
Cases were identified in all provinces of the region except Belluno Province, characterized by a mountain territory ( Figure 3). The overall incidence of WNND and WNF in the Veneto region was 1.75 and 6.26 per 100,000 population, respectively. The highest incidence was recorded in Rovigo Province, a wetland area near the Po River, with 11.6 cases of WNND and 35 cases of WNF per 100,000 population. Among blood donors, the incidence of WNV infection was 28.9 per 100,000 in the Veneto region and 52.1 per 100,000 in the Rovigo Province.
Cases were identified in all provinces of the region except Belluno Province, characterized by a mountain territory ( Figure 3). The overall incidence of WNND and WNF in the Veneto region was 1.75 and 6.26 per 100,000 population, respectively. The highest incidence was recorded in Rovigo Province, a wetland area near the Po River, with 11.6 cases of WNND and 35 cases of WNF per 100,000 population. Among blood donors, the incidence of WNV infection was 28.9 per 100,000 in the Veneto region and 52.1 per 100,000 in the Rovigo Province.

Entomological Surveillance
Similar to the human surveillance, entomological monitoring also gave the highest rate of WNVpositive mosquito pools in 2018 (n = 155; 13%), compared to the past (e.g., 1.8% in 2017). The first Culex pipiens mosquito pool was found infected on June 7, one week before the first human case, and the last on September 18, three weeks ahead of the last human case. Of the 55 mosquito trapping sites over the Veneto region, 38 (69%) were found to be positive at least once ( Figure 3), confirming the impressive and unusual viral circulation of 2018. In agreement with the human incidence rates, Rovigo Province had the highest rate of positive mosquito pools (15.8%). In all WNV RNA-positive mosquito pools, WNV lineage 2 central/southern European clade was identified, while WNV lineage 1 was not detected.

Entomological Surveillance
Similar to the human surveillance, entomological monitoring also gave the highest rate of WNV-positive mosquito pools in 2018 (n = 155; 13%), compared to the past (e.g., 1.8% in 2017). The first Culex pipiens mosquito pool was found infected on June 7, one week before the first human case, and the last on September 18, three weeks ahead of the last human case. Of the 55 mosquito trapping sites over the Veneto region, 38 (69%) were found to be positive at least once (Figure 3), confirming the impressive and unusual viral circulation of 2018. In agreement with the human incidence rates, Rovigo Province had the highest rate of positive mosquito pools (15.8%). In all WNV RNA-positive mosquito pools, WNV lineage 2 central/southern European clade was identified, while WNV lineage 1 was not detected.

WNV Antibody Dynamics
Antibody dynamics was evaluated by stratification of the WNV IgM and IgG results according to the time since symptom onset ( Figure 4). Within the first week after the onset of symptoms, about 20% of patients with WNND and 40% of patients with WNF were seronegative, while WNV IgG antibodies were already detectable in about 20% of both WNF and WNND patients. In all cases confirmed by neutralization assays, the titer of WNV neutralizing antibodies (i.e., the reciprocal of the highest dilution of the serum that reduced by 50% the number of plaques in Vero cells, WNV PRNT50) was ≥40 (generally >80); the titer of USUV neutralizing antibodies (i.e., the reciprocal of the highest dilution of the serum that showed 100% neutralization of the cytopathic effect in the microneutralization assay) was four-fold lower than WNV PRNT50 (generally ≤10). Among WNND cases, no specific IgM and IgG antibody responses were observed in five patients for whom laboratory follow-up data were available. All these patients were immunosuppressed because of underlying malignancies or old age. As expected, the rate of seronegative subjects was higher among WNV NAT-positive blood donors (about 80% at the time of the blood donation and 50% at one week after the donation), since all cases were identified during the acute phase of infection with detectable WNV RNA in blood.

WNV Antibody Dynamics
Antibody dynamics was evaluated by stratification of the WNV IgM and IgG results according to the time since symptom onset (Figure 4). Within the first week after the onset of symptoms, about 20% of patients with WNND and 40% of patients with WNF were seronegative, while WNV IgG antibodies were already detectable in about 20% of both WNF and WNND patients. In all cases confirmed by neutralization assays, the titer of WNV neutralizing antibodies (i.e., the reciprocal of the highest dilution of the serum that reduced by 50% the number of plaques in Vero cells, WNV PRNT50) was ≥40 (generally >80); the titer of USUV neutralizing antibodies (i.e., the reciprocal of the highest dilution of the serum that showed 100% neutralization of the cytopathic effect in the microneutralization assay) was four-fold lower than WNV PRNT50 (generally ≤10). Among WNND cases, no specific IgM and IgG antibody responses were observed in five patients for whom laboratory follow-up data were available. All these patients were immunosuppressed because of underlying malignancies or old age. As expected, the rate of seronegative subjects was higher among WNV NAT-positive blood donors (about 80% at the time of the blood donation and 50% at one week after the donation), since all cases were identified during the acute phase of infection with detectable WNV RNA in blood.  Notably, seven cases (five blood donors, one cord blood donor, one WNF case) did not develop WNV IgM antibodies during follow-up or had a blunted WNV IgM response but developed WNV IgG Viruses 2020, 12, 458 9 of 14 antibodies. Five of these cases, who had previous USUV immunity, have already been described [23]. The other two cases were blood donors, including a man in his 60s who developed myalgia and who had been vaccinated against yellow fever 10 years before and a man in his 40s who developed rash, asthenia, and lymphadenopathy and already had high levels (>320) of neutralizing antibodies against WNV and USUV at 12 days after the index donation. Genomic RNA of WNV lineage 2 was detectable in the blood of the first donor and both in the blood and urine in the second donor ( Figure 5).
Notably, seven cases (five blood donors, one cord blood donor, one WNF case) did not de IgM antibodies during follow-up or had a blunted WNV IgM response but developed ntibodies. Five of these cases, who had previous USUV immunity, have already been desc The other two cases were blood donors, including a man in his 60s who developed myalgi had been vaccinated against yellow fever 10 years before and a man in his 40s who deve asthenia, and lymphadenopathy and already had high levels (>320) of neutralizing antib st WNV and USUV at 12 days after the index donation. Genomic RNA of WNV lineage table in the blood of the first donor and both in the blood and urine in the second donor (F . WNV RNA Kinetics in Blood and Urine WNV RNA was detectable in the blood of 70-80% of WNND patients and 50% of WNF pa g the first 9 days after symptom onset. WNV RNA persisted in the blood and was still dete days after onset in ca 30% of WNND patients and 20% WNF patients (Figure 4). Most W WNF patients had detectable WNV RNA in urine during the first 9 days after onset. WNV sted in the urine of WNND patients for a longer time than in WNF patients and in blood do was still detectable at 30 days after onset in over 50% of cases ( Figure 4). The levels of WNV e blood were significantly higher in WNND patients than in WNF patients and in blood d re 6). Among WNV NAT-positive blood donors, 71% and 44% had detectable WNV RN d and urine, respectively, at the baseline evaluation performed within 3 days after the

WNV RNA Kinetics in Blood and Urine
WNV RNA was detectable in the blood of 70-80% of WNND patients and 50% of WNF patients during the first 9 days after symptom onset. WNV RNA persisted in the blood and was still detectable at 30 days after onset in ca 30% of WNND patients and 20% WNF patients (Figure 4). Most WNND and WNF patients had detectable WNV RNA in urine during the first 9 days after onset. WNV RNA persisted in the urine of WNND patients for a longer time than in WNF patients and in blood donors, as it was still detectable at 30 days after onset in over 50% of cases ( Figure 4). The levels of WNV RNA in the blood were significantly higher in WNND patients than in WNF patients and in blood donors ( Figure 6). Among WNV NAT-positive blood donors, 71% and 44% had detectable WNV RNA in blood and urine, respectively, at the baseline evaluation performed within 3 days after the index donation ( Figure 4). The rate of WNV RNA-positive results progressively decreased during follow-up (Figure 4).

Phylogenetic Analysis of WNV Genome Sequences
Real-time RT-PCR analyses and sequencing of a portion of the WNV NS5 region in samples with detectable WNV RNA showed that the virus belonged to lineage 2. WNV lineage 1 was not detected in any WNV case. Phylogenetic analysis of four WNV full genomes classified the virus within the central/southern Europe clade and showed that it was genetically very similar to WNV lineage 2 strains identified in northeastern Italy since 2013 [6,8]. In particular, the virus detected in the Veneto region in 2018 diverged from the WNV lineage 2 strain present over the years 2013-2014 (Veneto subclade), while it had greater similarity with the sub-clade detected in the Lombardy region since 2014, within which it constituted a distinct evolutionary group (Figure 8). The percentage of nucleotide identity among WNV genome sequences of this sub-clade was higher than 99.8%, while the percentage identity between the two sub-clades was 99.5%.

WNV RNA and IgM in Cerebrospinal Fluid (CSF)
Laboratory diagnosis of WNND is based on the detection of WNV IgM and/or WNV RNA in CSF. During the first week after symptom onset, about 30-40% of WNND patients had negative WNV IgM testing in CSF, while WNV RNA was detectable in less than 40% of cases (Figure 7). The WNV RNA load in CSF was low (median C T value, 35; range 41-31). The two tests appeared to be complementary, since most CSF samples were either WNV IgM positive or WNV RNA positive (Figure 7).

WNV RNA and IgM in Cerebrospinal Fluid (CSF)
Laboratory diagnosis of WNND is based on the detection of WNV IgM and/or WNV RNA in CSF. During the first week after symptom onset, about 30-40% of WNND patients had negative WNV IgM testing in CSF, while WNV RNA was detectable in less than 40% of cases (Figure 7). The WNV RNA load in CSF was low (median CT value, 35; range 41-31). The two tests appeared to be complementary, since most CSF samples were either WNV IgM positive or WNV RNA positive (Figure 7).

Phylogenetic Analysis of WNV Genome Sequences
Real-time RT-PCR analyses and sequencing of a portion of the WNV NS5 region in samples with detectable WNV RNA showed that the virus belonged to lineage 2. WNV lineage 1 was not detected in any WNV case. Phylogenetic analysis of four WNV full genomes classified the virus within the central/southern Europe clade and showed that it was genetically very similar to WNV lineage 2 strains identified in northeastern Italy since 2013 [6,8]. In particular, the virus detected in the Veneto region in 2018 diverged from the WNV lineage 2 strain present over the years 2013-2014 (Veneto subclade), while it had greater similarity with the sub-clade detected in the Lombardy region since 2014, within which it constituted a distinct evolutionary group (Figure 8). The percentage of nucleotide identity among WNV genome sequences of this sub-clade was higher than 99.8%, while the percentage identity between the two sub-clades was 99.5%.

Phylogenetic Analysis of WNV Genome Sequences
Real-time RT-PCR analyses and sequencing of a portion of the WNV NS5 region in samples with detectable WNV RNA showed that the virus belonged to lineage 2. WNV lineage 1 was not detected in any WNV case. Phylogenetic analysis of four WNV full genomes classified the virus within the central/southern Europe clade and showed that it was genetically very similar to WNV lineage 2 strains identified in northeastern Italy since 2013 [6,8]. In particular, the virus detected in the Veneto region in 2018 diverged from the WNV lineage 2 strain present over the years 2013-2014 (Veneto sub-clade), while it had greater similarity with the sub-clade detected in the Lombardy region since 2014, within which it constituted a distinct evolutionary group (Figure 8). The percentage of nucleotide identity among WNV genome sequences of this sub-clade was higher than 99.8%, while the percentage identity between the two sub-clades was 99.5%.

Discussion
This study, which describes a large series of 427 patients with WNV lineage 2 infection from northern Italy, allowed the epidemiological and clinical features of the currently most widespread WNV strain in Europe to be highlighted.
This study showed that the incidence of WNF progressively increased with age, while WNND incidence sharply peaked in subjects over 70 years of age. In the most affected province, the incidence of WNND and WNF were 11.6 cases and 35 cases per 100,000 population, respectively, while the incidence of WNV infection among asymptomatic blood donors was 52 cases per 100,000. As a comparison, in EU, the overall notification rate for locally acquired cases was 0.4 per 100,000 population [19]. In areas of EU countries with the highest WNV circulation, such as in Greece, Hungary, and Romania, the incidence rates ranged between 5 and 25 cases per 100,000, respectively [19].
WNV disease, especially WNND, was significantly more frequent in males than females, in agreement with epidemiological reports, which identified male gender as a risk factor for the occurrence of WNND [24]. In fact, the prevalence of anti-WNV antibodies is similar in females and males in endemic areas and the increased exposure due to outdoor activities is not sufficient to explain sex differences in the incidence of WNND [25]. Although experimental studies on the role of gender in the development of WNV encephalitis are lacking, sex differences in innate and adaptive immune responses might account for the higher prevalence and severity of WNV disease in males [26,27]. Intriguingly, in this study, the incidence rate of WNV infection was also higher in male than in female blood donors, suggesting a gender-biased risk of infection.
In patients with WNND, the case fatality rate was 22%, which was higher than the 10% reported in 2018 in the United States of America, where WNV lineage 1 is endemic [28]. Signs and symptoms of WNV infection included fever, asthenia, rash, headache, arthralgia, and myalgia, besides neurological manifestation in patients with WNND. Notably, 71% of the blood donors with WNV

Discussion
This study, which describes a large series of 427 patients with WNV lineage 2 infection from northern Italy, allowed the epidemiological and clinical features of the currently most widespread WNV strain in Europe to be highlighted.
This study showed that the incidence of WNF progressively increased with age, while WNND incidence sharply peaked in subjects over 70 years of age. In the most affected province, the incidence of WNND and WNF were 11.6 cases and 35 cases per 100,000 population, respectively, while the incidence of WNV infection among asymptomatic blood donors was 52 cases per 100,000. As a comparison, in EU, the overall notification rate for locally acquired cases was 0.4 per 100,000 population [19]. In areas of EU countries with the highest WNV circulation, such as in Greece, Hungary, and Romania, the incidence rates ranged between 5 and 25 cases per 100,000, respectively [19].
WNV disease, especially WNND, was significantly more frequent in males than females, in agreement with epidemiological reports, which identified male gender as a risk factor for the occurrence of WNND [24]. In fact, the prevalence of anti-WNV antibodies is similar in females and males in endemic areas and the increased exposure due to outdoor activities is not sufficient to explain sex differences in the incidence of WNND [25]. Although experimental studies on the role of gender in the development of WNV encephalitis are lacking, sex differences in innate and adaptive immune responses might account for the higher prevalence and severity of WNV disease in males [26,27]. Intriguingly, in this study, the incidence rate of WNV infection was also higher in male than in female blood donors, suggesting a gender-biased risk of infection.
In patients with WNND, the case fatality rate was 22%, which was higher than the 10% reported in 2018 in the United States of America, where WNV lineage 1 is endemic [28]. Signs and symptoms of WNV infection included fever, asthenia, rash, headache, arthralgia, and myalgia, besides neurological manifestation in patients with WNND. Notably, 71% of the blood donors with WNV infection reported symptoms during follow-up visits, a proportion higher than a previous estimate of 30% in blood donors with WNV lineage 1 infection from the United States of America [29].
Laboratory diagnosis of WNV infection relies on the detection of viral RNA in blood, urine, or CSF and the demonstration of a specific antibody response in serum or CSF. Based on the results of this study and on our previous experience with the diagnosis of WNV infection [6,14,21,23], we recommend an integrated approach, combining serology and molecular testing, to increase diagnostic sensitivity and specificity. In fact, within the first week after symptom onset, WNV-specific antibodies were absent in 20-40% of cases, while WNV RNA was detectable in the blood or urine of 80% of patients. In addition, it should be kept in mind that some patients with WNND remain seronegative because of their underlying immunosuppressed condition, such as five of our patients with severe encephalitis. Moreover, subjects with previous flavivirus immunity may have a blunted or absent WNV IgM response, especially in the cases of subsequent infections with antigenically related viruses like WNV and USUV [23]. The complementarity of serology and molecular testing was observed in particular when analyzing CSF samples from patients with WNND, in which WNV RNA was generally detectable when WNV IgM antibodies were absent. The levels WNV RNA in CSF were, however, lower and detectable for a shorter time than in blood and urine, which should be considered elective biological samples for WNV molecular testing. Molecular testing provides a rapid etiological diagnosis of WNV infection, at variance with antibody detection, which needs confirmation by neutralization assays. In areas where multiple flaviviruses are endemic and co-circulate, like WNV and USUV in the Veneto region, neutralization assays against these viruses should be run in parallel [12,23]. However, in patients with previous flavivirus immunity, such as the blood donor described in this study, neutralization assays may be inconclusive.
Phylogenetic analysis of partial and full WNV genome sequences from clinical samples showed that the virus belonged to the widespread central/southern European clade of WNV lineage 2. No WNV lineage 1 was detected by either human or entomological surveillance, although we cannot exclude the presence of low-level circulation of the WNV lineage 1 in the region, unrecognized by the surveillance system. This WNV lineage 2 strain, which has been detected in the Veneto region since 2014, replaced a genetically related strain, which circulated in the same area in 2013-2014 [8,30]. A similar strain replacement was reported in the neighboring Lombardy and Emilia-Romagna regions, suggesting that this strain might have acquired improved fitness [31,32]. The anticipation of the transmission season, due to the exceptionally warm climate conditions during the 2018 spring, which increased the susceptible bird and mosquito populations and viral amplification, is considered the most important factor contributing to the dramatic increase of WNV infections in humans reported in 2018 [33]. This hypothesis was supported by the results of an epidemiological mathematical model informed by entomological data collected in the neighboring Emilia-Romagna region, northern Italy [34].
In conclusion, this study on a large series of patients with WNV infection investigated at a reference laboratory highlighted the epidemiological, clinical, and laboratory features of WNV lineage 2, which is currently spreading and causing human outbreaks in Europe, providing useful information to improve clinical and laboratory practices for WNV diagnosis.