Highlights
- •Asymptomatic Zika virus infections have been overestimated.
- •Surveillance data confirm the results from seroprevalence studies.
- •The use of easy-to-implement surveys to estimate ZIKV circulation during outbreaks is proposed.
Abstract
French Polynesia and the French Territories of the Americas (FTAs) have experienced outbreaks of Zika virus (ZIKV) infection. These territories used similar sentinel syndromic surveillance to follow the epidemics. However, the surveillance system only takes into account consulting patients diagnosed with ZIKV disease, while non-consulting cases, as well as asymptomatic cases, are not taken into account. In the French territories under study, the ratio of consulting to non-consulting patients was found to likely be as low as 1/3 to 1/4, and rough estimates of the ZIKV asymptomatic infections indicated a lower rate than previously reported (i.e., not more than half).
Keywords
Introduction
The first documented outbreak of Zika virus (ZIKV), a mosquito-borne flavivirus associated with neurological complications (
Cao-Lormeau et al., 2016
, de Araújo et al., 2016
), was reported in 2007. This outbreak involved 70% of the population (5300 individuals) of Yap islands, in the Federated States of Micronesia (- de Araújo Thalia Velho Barreto
- Rodrigues Laura Cunha
- de Alencar Ximenes Ricardo Arraes
- de Barros Miranda-Filho Demócrito
- Montarroyos Ulisses Ramos
- de Melo Ana Paula Lopes
- et al.
Association between Zika virus infection and microcephaly in Brazil, January to May, 2016: preliminary report of a case-control study.
Duffy et al., 2009
). In 2013–2014, French Polynesia experienced a larger Zika outbreak with an estimated 11.5% of the population (over 30 000 people) having consulted a doctor for suspected ZIKV infection (Musso and Gubler, 2016
). In 2015–2016, ZIKV spread throughout the Americas and the Caribbean region ().Based on data from the Yap outbreak, it was assumed that asymptomatic ZIKV infections account for 80% of the overall infections (
Duffy et al., 2009
). However, recently published serological data from French Polynesia and Martinique have estimated it to be less than half of the overall infections (Aubry et al., 2017
, - Aubry Maite
- Teissier Anita
- Huart Michael
- Merceron Sébastien
- Vanhomwegen Jessica
- Roche Claudine
- et al.
Zika virus seroprevalence, French Polynesia, 2014-2015.
Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549
Gallian et al., 2017
). In this study, a rough estimate of the level of asymptomatic infections was performed using the sentinel syndromic surveillance data available from three comparable French Overseas Territories that experienced Zika outbreaks: French Polynesia, Martinique, and Guadeloupe.The study
French Polynesia, Martinique, and Guadeloupe have comparable syndromic surveillance systems, which include a network of 35–55 general practitioners (GPs) or consultation sites reporting the number of patients that fit the case definition for a given syndrome on a weekly basis. During the Zika outbreaks, these sentinel networks were considered as representative of all private and public GP consultations in the three territories, as they provided 16–25% of the overall medical consultations in the community, with a regular geographic distribution (Table 1).
Table 1Sentinel surveillance data to estimate ZIKV asymptomatic infection rates.
| French Polynesia | Martinique | Guadeloupe (except North Islands) | |
|---|---|---|---|
| Number of inhabitants | 270 000 (Tahiti: 180 000) | 390 000 | 400 000 |
| Surface (km2) | ∼4200 (Tahiti: ∼1000) | ∼1100 | ∼1600 |
| Date of emergence (first confirmed case) | 2013-41 | 2015-50 | 2016-02 |
| Dates of the outbreak | Oct 2013–Apr 2014 | Jan 2016–Oct 2016 | Apr 2016–Sept 2016 |
| Length of the outbreak (weeks) | 28 | 35 | 22 |
| Peak week (from first confirmed case) | 9 | 18 | 15 |
| Number of GPS or sites participating in the sentinel network during the outbreak | 45 | 55 | 35 |
| SN representativeness (ratio of SN sites reported to total consultation sites) | 30% | 25% | 16% |
| Total estimated number of symptomatic consulting cases | 32 000 | 38 350 | 30 560 |
| Consultation rate in the population | 11.5% | 9.8% | 7.6% |
| Estimated ratio of consulting/non-consulting among symptomatic cases | 1:3 to 1:4 | ||
| Estimated attack rate of symptomatic cases in the population | 34–46% | 29–39% | 23–30% |
| ZIKV overall attack rate from seroprevalence data | 49% (general population) ( Aubry et al., 2017 )
Zika virus seroprevalence, French Polynesia, 2014-2015. Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549 | 42% (blood donors, in early June) ( Gallian et al., 2017 ) | ND |
| Highest estimate of ZIKV overall attack rate | 86% ( Kucharski et al., 2016 )
Transmission dynamics of Zika virus in Island populations: a modelling analysis of the 2013-14 French Polynesia Outbreak. PLoS Negl Trop Dis. 2016; 10https://doi.org/10.1371/journal.pntd.0004726 | ND | ND |
| Estimated rate of asymptomatic cases | 3–51.5% | 3–57% | ND |
ZIKV, Zika virus; GPS, global positioning system; SN, sentinel network; ND, not determined.
Cases were reported according to local clinically suspected case definitions (
Daudens-Vaysse et al., 2016
, - Daudens-Vaysse Elise
- Ledrans Martine
- Gay Noellie
- Ardillon Vanessa
- Cassadou Sylvie
- Najioullah Fatiha
- et al.
Zika emergence in the French Territories of America and description of first confirmed cases of Zika virus infection on Martinique, November 2015 to February 2016.
Euro Surveill. 2016; 21https://doi.org/10.2807/1560-7917.ES.2016.21.28.30285
Mallet et al., 2016
). These case definitions, using a combination of a minimum of three symptoms, met those proposed by the European Centre for Disease Prevention and Control and the Pan-American Health Organization, and have proven to be fairly specific and sensitive (Chow et al., 2017
). Consultations in emergency structures (hospitals and mobile emergency services) were accounted for in the global surveillance network in French Polynesia, in Martinique, and in Guadeloupe. The total number of clinically suspected cases for the whole country was estimated by summation of these sources and extrapolation. This surveillance system was implemented to monitor the space–time evolution of the epidemic, but in the case of ZIKV, it failed to estimate the total number of infected people. As ZIKV infection mostly causes a mild illness, a significant proportion of infected people may not consult a GP. Furthermore, many cases are asymptomatic.During the Zika outbreaks, the total estimated proportion of individuals who sought medical care with symptoms compatible with ZIKV infection in the three French Territories ranged from 7.6% to 11.5% of the overall population (Table 1). Based on the ZIKV serosurvey conducted in French Polynesia (
Aubry et al., 2017
), a blood donor study in Martinique (- Aubry Maite
- Teissier Anita
- Huart Michael
- Merceron Sébastien
- Vanhomwegen Jessica
- Roche Claudine
- et al.
Zika virus seroprevalence, French Polynesia, 2014-2015.
Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549
Gallian et al., 2017
), and data from a cross-sectional survey conducted in Guadeloupe (Table 2), the estimated ratio of consulting to non-consulting symptomatic suspected cases was found to be in the range of 1:3 to 1:4. Information from the survey in Guadeloupe, which used quota sampling, was gathered through telephone-based interviews of a sample of 501 individuals aged >15 years. The questionnaire was mainly focused on knowledge of the disease and reported infectious episodes, including specific symptoms (IPSOS Antilles, 2016
). The resulting attack rates of symptomatic ZIKV infection ranged from 23% to 46% (Table 1). Using available estimates of ZIKV overall attack rates from (1) seroprevalence studies (Aubry et al., 2017
, - Aubry Maite
- Teissier Anita
- Huart Michael
- Merceron Sébastien
- Vanhomwegen Jessica
- Roche Claudine
- et al.
Zika virus seroprevalence, French Polynesia, 2014-2015.
Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549
Gallian et al., 2017
), and (2) a modeling study in French Polynesia (Kucharski et al., 2016
), it was found that surveillance data indicated asymptomatic infection rates not exceeding half of the overall infections.- Kucharski Adam J.
- Funk Sebastian
- Eggo Rosalind M.
- Mallet Henri-Pierre
- Edmunds W. John
- Nilles Eric J.
Transmission dynamics of Zika virus in Island populations: a modelling analysis of the 2013-14 French Polynesia Outbreak.
PLoS Negl Trop Dis. 2016; 10https://doi.org/10.1371/journal.pntd.0004726
Table 2Estimate of consultation rates from a cross-sectional study based on telephone interviews with individuals aged >15 years (n = 501) recruited by quota sampling at the end of the Zika outbreak—Guadeloupe, October 2016.
| Percentage | |
|---|---|
| ZIKV reported symptoms | |
| Yes (consulted a GP) | 31 (9) |
| No | 69 |
| Consultation rate of symptomatic cases | 29 |
ZIKV, Zika virus; GP, general practitioner.
a Data courtesy of IPSOS Antilles. For details (in French): http://www.ireps.gp/data/IMG/Rapport_Impact_Comm_Zika_ARS_Mai_2016.pdf.
Discussion
The estimated ZIKV asymptomatic infection rates reported here are lower than that found during the 2007 Yap outbreak. This inconsistency with the data from Yap may be due to the limited sample representativeness of the Yap study (
Duffy et al., 2009
). Alternatively, unknown risk factors, such as environmental or individual conditions, may have affected the ZIKV asymptomatic rate.Taking into consideration the underlying assumptions and limitations, the estimated asymptomatic case rates from sentinel surveillance data reported here, which are compatible with those from seroprevalence studies (
Aubry et al., 2017
, - Aubry Maite
- Teissier Anita
- Huart Michael
- Merceron Sébastien
- Vanhomwegen Jessica
- Roche Claudine
- et al.
Zika virus seroprevalence, French Polynesia, 2014-2015.
Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549
Gallian et al., 2017
), are important in order to improve surveillance systems: they give a more realistic idea of the true number of susceptible individuals throughout the outbreak, because they consider individuals who have experienced asymptomatic infection as immunized. Together with the estimate of the outbreak-specific consultation rate, the new ZIKV asymptomatic infection rate allows a rough prediction of the overall intensity of ZIKV circulation. This is particularly important on islands, where attack rates often reach high percentages (Aubry et al., 2017
).- Aubry Maite
- Teissier Anita
- Huart Michael
- Merceron Sébastien
- Vanhomwegen Jessica
- Roche Claudine
- et al.
Zika virus seroprevalence, French Polynesia, 2014-2015.
Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549
Zika outbreak surveillance should be improved by implementing simple cost-effective surveys during outbreaks to assess the outbreak-specific consultation rate and the asymptomatic infection rate, in order to obtain a less error-prone estimate of viral circulation. The first could be estimated using telephone-based interviews and the second by analyzing blood donor samples. Ideally, to be confident that estimates are not biased, a population-based survey using probability sampling rather than quota sampling should be performed. However, such surveys are often difficult to implement during outbreaks as they are costly and time-consuming. Standardized protocols should be available to allow the rapid implementation of these surveys in any country or territory where ZIKV is likely to (re-)emerge.
In conclusion, the ZIKV asymptomatic infection rate should be taken into account to estimate the level of ZIKV circulation in a country or territory. Determining viral circulation is, in turn, essential to guide public health action. Past reference studies may have overestimated the true ZIKV asymptomatic infection rate, which in the present study settings, corresponds to, at most, half of the infected people.
Conflict of interest
None declared
Acknowledgements
We thank IPSOS Antilles for sharing data.
References
- Zika virus seroprevalence, French Polynesia, 2014-2015.Emerg Infect Dis. 2017; 23https://doi.org/10.3201/eid2304.161549
- Guillain-Barrè Syndrome outbreak associated with Zika virus infection in French Polynesia: a case-control study.Lancet Lond Engl. 2016; 387: 1531-1539https://doi.org/10.1016/S0140-6736(16)00562-6
- Assessing sensitivity and specificity of surveillance case definitions for Zika virus disease.Emerg Infect Dis. 2017; 23: 677-679https://doi.org/10.3201/eid2304.161716
- Zika emergence in the French Territories of America and description of first confirmed cases of Zika virus infection on Martinique, November 2015 to February 2016.Euro Surveill. 2016; 21https://doi.org/10.2807/1560-7917.ES.2016.21.28.30285
- Association between Zika virus infection and microcephaly in Brazil, January to May, 2016: preliminary report of a case-control study.Lancet Infect Dis. 2016; 16: 1356-1363https://doi.org/10.1016/S1473-3099(16)30318-8
- Zika virus outbreak on Yap Island, Federated States of Micronesia.N Engl J Med. 2009; 360: 2536-2543https://doi.org/10.1056/NEJMoa0805715
- Zika virus in asymptomatic blood donors, Martinique: 2016.Blood. 2017; 129: 263-266https://doi.org/10.1182/blood-2016-09-737981
- Study on the impact of prevention communication and information about Zika.2016
- Transmission dynamics of Zika virus in Island populations: a modelling analysis of the 2013-14 French Polynesia Outbreak.PLoS Negl Trop Dis. 2016; 10https://doi.org/10.1371/journal.pntd.0004726
- Bilan de l’épidémie à virus Zika survenue en Polynésie française entre octobre 2013 et mars 2014. De la description de l’épidémie aux connaissances acquises après l’évènement.Bull Epidémiol Hebd. 2016; 367–73: 20-21
- Zika virus.Clin Microbiol Rev. 2016; 29: 487-524https://doi.org/10.1128/CMR.00072-15
- Zika situation report.2017
Article info
Publication history
Published online: October 25, 2017
Accepted:
October 16,
2017
Received in revised form:
September 28,
2017
Received:
July 14,
2017
Corresponding Editor: Eskild Petersen, Aarhus, DenmarkIdentification
Copyright
© 2017 The Author(s). Published by Elsevier Ltd on behalf of International Society for Infectious Diseases.
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
