Demographic characteristics of participants
The serological surveillance study participants were randomly selected from the community of Krong Buk District. Serum samples were collected from 801 of 65,000 residents (1.2%) in January 2017. The median age is 33 years, dominated by those in the age group between 16 – 60 years (n=663 or 82.7%). A total of 87 samples were collected from children under the age of 15 years (10.9%). The proportion of females in the study was more than 73%. In the analysis by ethnicity, 503 samples (62.8%) were from the Kinh group, which is the major ethnic group in Vietnam, while a minority ethnic group Ede made up 290 samples (36.2%), followed by 8 participants of other minor origin (1%). The number of pregnant women who participated in this study was 66 (8.24%).
Zika virus and dengue seroprevalence
Of the 801 samples tested, 83 (10.3%) were positive for anti-ZIKV IgM antibodies (Table 1) with an average P/N ratio of 3.30 ± 1.48. While the P/N ratio is not a quantitative method, the ratio was used as a means to determine antibody levels of 83 samples, for the presence or absence of anti-ZIKV IgM antibodies. All participants who tested positive were asymptomatic at the time of sample collection. The prevalence of anti-ZIKV IgM antibodies varied moderately by age (p=0.05) and was highest in the 46–60 year age group (14.2%) (Figure 1(a)). The seroprevalence of ZIKV IgM did not differ significantly according to sex or ethnicity. Only one of the 66 pregnant women (1.5%) was positive for anti-ZIKV IgM antibodies.
Table 1. Demographic characteristics of the study participants (N = 801)
Variables
|
|
n
|
ZIKV IgM positive (%)
|
Crude OR (95% CI)
|
P-value
|
DENV IgG positive (%) (n=83)
|
Anti-ZIKV NS1 IgG positive (%) (n=88)
|
PRNT50 positive to ZIKV (%)
|
Age group (years)
|
£15
|
87
|
11 (12.6)
|
1.00 (Reference)
|
0.05
|
4/11 (36.3)
|
3/11 (27.3)
|
2/87 (2.3)
|
16–30
|
295
|
19 (6.4)
|
0.48 (0.22–1.04)
|
9/19 (47.4)
|
3/21 (14.3)
|
3/295 (1.0)
|
31–45
|
241
|
31 (12.8)
|
1.02 (0.49–2.13)
|
18/31 (58.1)
|
7/32 (21.9)
|
3/241 (1.2)
|
46–60
|
127
|
18 (14.2)
|
1.14 (0.51–2.55)
|
14/18 (77.8)
|
4/18 (22.2)
|
1/127 (0.8)
|
³60
|
35
|
3 (8.5)
|
0.65 (0.17–2.48)
|
3/3 (100)
|
1/5 (20)
|
3/35 (8.6)
|
NA†
|
16
|
1 (6.3)
|
|
1/1 (100)
|
1/1 (100)
|
1/16 (6.3)
|
Sex
|
Female
|
590
|
59 (10)
|
1.00 (Reference)
|
0.57
|
32/59 (54.2)
|
10/64 (15.6)
|
7/590 (1.2)
|
Male
|
211
|
24 (11.4)
|
1.15 (0.7–1.91)
|
17/24 (70.8)
|
9/24 (37.5)
|
6/211 (2.8)
|
Ethnicity
|
Ede
|
290
|
25 (8.6)
|
1.00 (Reference)
|
0.16
|
15/25 (60.0)
|
5/26 (19.2)
|
3/290 (1.0)
|
Kinh
|
503
|
58 (11.5)
|
1.42 (0.87–2.33)
|
34/58 (58.6)
|
14/62(22.6)
|
10/503 (2.0)
|
Other‡
|
8
|
0
|
|
|
|
0/8 (0.0)
|
Pregnancy
|
|
66
|
1 (1.5)
|
|
|
0/1(0.0)
|
0/1(0.0)
|
0/66 (0.0)
|
CI, confidence interval; DENV, dengue virus; IgG, immunoglobulin G; IgM, immunoglobulin M; NA, not available; OR, odds ratio; ZIKV, Zika virus; Anti-ZIKV NS1 IgG positive, ZIKV antibodies were determined using anti-ZIKV NS1 IgG ELISA (P/N ratio ≥ 2); PRNT50 positive to ZIKV was defined as plaque-reduction neutralization test titers with a ≥50% reduction in ZIKV plaque-forming units at a serum dilution of ≥1:20
† Participants with missing information on age were excluded from the multivariate logistic regression analysis
‡ Participants whose ethnicity was categorized as “Other” were combined with the “Ede” ethnic group in the multivariate logistic regression analysis.
All 83 participants with ZIKV anti-IgM antibodies were negative for dengue IgM on ELISA; however, 49 (59.0%) ZIKV IgM positive cases were detected as positive for Dengue IgG ELISA (Table 1). DENV IgM and DENV IgG antibodies was determined to examine possible cross-reactivity to ZIKV infection. The distribution of the ZIKV IgM antibodies positive cases, as well as DENV IgG antibodies positive cases, was observed in all age groups in this study (Figure 1(a)). All testing results showed higher positive rates in the majority Kinh ethnic group than those in the minority Ede ethnic group (Figure 1(b)).
Neutralizing antibody levels to Zika virus and dengue virus serotypes 1-4
All 801 samples were first screened to detect the presence of neutralizing antibodies to ZIKV. Of the 83 ZIKV IgM-positive samples, 8 demonstrated neutralizing antibodies to ZIKV (PRNT50=1: 40 to 1: 640). Of the 8 samples, 3 (Z141a, Z153a, Z735a) exhibited ZIKV antibody titers that were >4-fold higher than the antibody titers for all 4 DENV serotypes (PRNT50 ZIKV=1: 160 to 1: 640) (Table 2). The remaining 5 samples (Z120a, Z138a, Z140a, Z420a, Z476a) were classified as probable ZIKV infections (PRNT50 ZIKV=1: 40–1: 640).
Of the 718 samples that were negative for anti-ZIKV IgM antibodies, 5 (Z147a, Z472a, Z606a, Z587a and Z591a) demonstrated neutralizing antibodies to ZIKV (PRNT50=1:320–1:5120).
Table 2. Anti-Zika virus immunoglobulin M and G levels of 13 participants with neutralizing antibodies to Zika virus epidemic) (N= 879)
|
ELISA results
|
|
Neutralizing antibody titer (PRNT50)
|
|
Neutralizing antibody titer (PRNT90)
|
Sample ID
|
ZIKV IgM (P/N ratio)
|
DENV IgM (Index Value)
|
Anti-ZIKV NS1 IgG (P/N ratio)
|
|
ZIKV
|
DENV-1
|
DENV-2
|
DENV-3
|
DENV-4
|
|
ZIKV
|
DENV-1
|
DENV-2
|
DENV-3
|
DENV-4
|
During epidemic (January 2017)
|
Z120a†
|
3.3
|
0.1
|
2.2
|
|
160
|
320
|
320
|
<20
|
<20
|
|
<20
|
320
|
80
|
<20
|
<20
|
Z140a
|
2.3
|
0.3
|
4.7†
|
|
80
|
40
|
320
|
<20
|
<20
|
|
<20
|
<20
|
40
|
<20
|
<20
|
Z141a
|
4.9
|
0.1
|
3.4
|
|
640
|
160
|
80
|
<20
|
<20
|
|
40
|
80
|
<20
|
<20
|
<20
|
Z153a‡
|
2.8
|
0.3
|
3.2
|
|
160
|
<20
|
<20
|
<20
|
<20
|
|
40
|
<20
|
<20
|
<20
|
<20
|
Z476a
|
4.7
|
0.5
|
43.9†
|
|
640
|
320
|
1280
|
20
|
<20
|
|
80
|
320
|
160
|
<20
|
<20
|
Z735a
|
2.8
|
0.6
|
8.1
|
|
320
|
<80
|
<80
|
<80
|
<80
|
|
80
|
<80
|
<80
|
<80
|
<80
|
Z147a†
|
1.6
|
ND
|
4.5
|
|
320
|
80
|
160
|
20
|
40
|
|
80
|
<20
|
80
|
<20
|
<20
|
Z472a
|
1.5
|
ND
|
46.1
|
|
5120
|
160
|
160
|
80
|
40
|
|
640
|
80
|
40
|
80
|
<20
|
Z606a
|
1.6
|
ND
|
10.6
|
|
1280
|
<20
|
<20
|
<20
|
<20
|
|
160
|
<20
|
<20
|
<20
|
<20
|
Z138a
|
4.6
|
0.6
|
0.9
|
|
40
|
320
|
80
|
<20
|
<20
|
|
<20
|
160
|
<20
|
<20
|
<20
|
Z420a‡
|
3.7
|
0.1
|
0.2
|
|
160
|
160
|
160
|
20
|
<20
|
|
20
|
20
|
40
|
<20
|
<20
|
Z587a
|
1.2
|
ND
|
0.2
|
|
320
|
640
|
640
|
1280
|
160
|
|
80
|
320
|
320
|
640
|
20
|
Z591a
|
1.2
|
ND
|
0.6
|
|
640
|
640
|
640
|
1280
|
80
|
|
80
|
320
|
160
|
640
|
<20
|
Post-epidemic (July 2018)
|
Z120b†
|
ND
|
ND
|
2.3
|
|
80
|
640
|
160
|
1280
|
20
|
|
20
|
320
|
40
|
320
|
<20
|
Z140b
|
ND
|
ND
|
5.2
|
|
640
|
<20
|
160
|
<20
|
20
|
|
20
|
<20
|
80
|
<20
|
<20
|
Z141b
|
ND
|
ND
|
2.7
|
|
640
|
160
|
20
|
160
|
<20
|
|
80
|
80
|
<20
|
40
|
<20
|
Z476b
|
ND
|
ND
|
4.7
|
|
2560
|
320
|
320
|
320
|
80
|
|
160
|
80
|
80
|
160
|
<20
|
Z735b
|
ND
|
ND
|
3.5
|
|
80
|
320
|
80
|
320
|
80
|
|
<20
|
320
|
20
|
160
|
20
|
Z138b
|
ND
|
ND
|
0.5
|
|
80
|
40
|
<20
|
40
|
<20
|
|
<20
|
20
|
<20
|
20
|
<20
|
DENV, dengue virus; DENV1-4, dengue virus serotypes 1-4; ELISA, enzyme-linked immunosorbent assay; IgG, immunoglobulin G; IgM, immunoglobulin M; N, negative; ND, not detected; P, positive; PRNT50, plaque-reduction neutralization test with neutralization defined as ≥50% reduction in challenge virus plaque-forming units; PRNT90, plaque-reduction neutralization test with neutralization defined as ≥90% reduction in challenge virus plaque-forming units; ZIKV, Zika virus.
Figures in bold indicate positive results. † ZIKV antibodies were determined using anti-ZIKV NS1 IgG ELISA (P/N ratio ≥ 2). ‡ Z153 and Z420 were not available during the post endemic collection.
Consecutive samples were collected from 78 of the 83 participants with anti-ZIKV IgM antibodies collected from the same participants 18 months after the first sample collection. Second consecutive samples were not collected from 5 participants who was not available at the time of second sample collection, including two (Z153 and Z420) of the 8 participants with ZIKV neutralizing antibodies in their initial sample. Two participants (Z120 and Z735) experienced a 2- to 4-fold decrease, and 2 participants (Z140 and Z476) experienced a 4- to 8-fold increase in their ZIKV neutralizing antibody titers (Table 2). Of the 83 participants with anti-ZIKV IgM antibodies, 5 (6.0%; Z140, Z141, Z153, Z476 and Z735) had ZIKV neutralizing antibody titers that were at least 4-fold greater than their antibody titers against the 4 DENV serotypes tested.
Of the 718 samples that were negative for anti-ZIKV IgM antibodies, 5 (Z147a, Z472a, Z606a, Z587a and Z591a) demonstrated neutralizing antibodies to ZIKV (PRNT50=1:320–1:5120). Among the 5 samples, 2 samples (Z472a and Z606a) demonstrated a 4-fold or greater level of neutralizing antibodies to all 4 DENV serotypes (Table 2). In addition, by testing all 879 samples collected in both rounds using PRNT, 13 participants (1.6%, N=801) had detectable ZIKV neutralizing antibodies. Overall, 7 participants (0.9%, Z140, Z141, Z153, Z476, Z735, Z472 and Z606) demonstrated ZIKV neutralizing antibody titers that were at least 4-fold higher than their antibody titers to all 4 DENV serotypes, whereas the other 6 participants (0.8%) demonstrated comparable levels of ZIKV and DENV neutralizing antibodies. Thus, the results suggest that the 7 participants had been exposed to ZIKV during the 2016 ZIKV epidemic.
Anti-Zika virus NS1 immunoglobulin G levels in cases of probable Zika virus infection
Levels of anti-ZIKV NS1 IgG antibodies in the 83 samples that exhibited anti-ZIKV IgM antibodies and five samples that demonstrated ZIKV neutralizing antibodies but negative for anti-ZIKV IgM antibodies (N=88) were determined using anti-Zika Virus NS1 IgG ELISA (R&D Systems). Anti-ZIKV NS1 IgG assays are useful for confirming ZIKV infection because they are highly specific and possess minimal cross-reactivity to other flaviviruses [37]. Sixteen out of 83 ZIKV IgM positive samples (19.3%) demonstrated ZIKV NS1 specific IgG antibodies, with a mean of P/N ratio of 5.7 ± 10.3 (data not shown). Among 8 of 83 ZIKV IgM positive samples that demonstrated neutralizing antibodies to ZIKV, 6 samples (Z120a, Z140a, Z141a, Z153a, Z476a and Z735a) were also positive for anti-ZIKV NS1 IgG by using ELISA. All of 3 samples (Z141a, Z153a and Z735a) that demonstrated a 4-fold or greater ratio of ZIKV neutralizing antibody titers to DENV antibody titers also demonstrated ZIKV NS1 IgG antibodies (P/N ratio=3.1–8.1). These results confirm that these 3 participants were exposed to ZIKV infection. Three of the 5 samples (Z120a, Z138a, Z140a, Z420a, Z476a) with ZIKV neutralizing antibodies were also confirmed positive for anti-ZIKV NS1 antibodies. The Anti-ZIKV NS1 IgG levels in the second samples collected in July 2018 were comparable to the levels in the first samples.
Three of 5 samples (Z147, Z472, Z587, Z591 and Z606) that demonstrated ZIKV neutralizing antibodies (N=718, ZIKV IgM negative samples) were also positive for anti-ZIKV NS1 IgG antibodies. Two samples (Z472 and Z606) were positive for both ZIKV neutralizing antibodies and ZIKV NS1 IgG (P/N ratio = 10.6 and 46.1 respectively). In addition, one sample (Z147) was also positive for anti-NS1 antibodies with a P/N ratio of 4.5. Finally, a total of 19 samples demonstrated anti-ZIKV NS1 IgG antibodies, including 16 of 83 samples that exhibited anti-ZIKV IgM antibodies positive and three samples that showed ZIKV neutralizing antibodies but negative for anti-ZIKV IgM antibodies (Table 1). However, out of the 19 samples that were positive for anti-ZIKV NS1 IgG antibodies, only 9 (Z120, Z140, Z141, Z153, Z476, Z735, Z147, Z472 and Z606) exhibited neutralizing antibodies against ZIKV with PRNT50 titers ≥1:20.
In summary, out of 801 participants tested in this study, by using two methods: the anti-NS1 IgG ELISA and PRNT, we determined that 9 participants (1.12%, Z120, Z140, Z141, Z147, Z153, Z735, Z476, Z472 and Z606) had ZIKV infection. Of these 9 participants, 5 (Z120, Z140, Z141, Z147 and Z153) lived in Cu Pong commune. An infant with microencephaly and her immediate family members has been confirmed confirmed positive for ZIKV infection in the Cu Pong commune in 2016 [25]. This study however excludes samples from our previous study [25]. The 4 remaining cases (Z472, Z476, Z606 and Z735) were found in 2 neighboring communes, two cases in Chu Kbo (1.7%, Z472 and Z476) and two other cases in Pong Drang village (0.99%) (Figure 2).