Zika virus displacement by a chikungunya outbreak in Recife, Brazil

Abstract

Background: Several arboviruses, including dengue virus (DENV), Zika virus (ZIKV) and chikungunya virus (CHIKV), transmitted by Aedes mosquitoes, circulate in northeast Brazil. Diseases caused by these viruses are of great public health relevance, however, their epidemiological features in areas where the three viruses co-circulate are scarce. Here, we present analyses of molecular and serological diagnostics in a prospective study of acute febrile patients recruited from May 2015 to May 2016 in Recife, Brazil.

Methods: Two hundred sixty-three acute febrile patients with symptoms suggestive of an arboviral disease who attended an urgent heath care clinic in the Recife Metropolitan Region in northeast Brazil were enrolled. Acute and convalescent blood samples were collected and tested using molecular and serological assays for infection with DENV, ZIKV and CHIKV.

Results: Quantitative real-time reverse-transcriptase polymerase chain reactions (qRTPCR) performed on acute phase sera detected no patients positive for DENV, but 26 (9.9%) positive for ZIKV and 132 (50.2%) positive for CHIKV. There were a few suspected and only one confirmed dengue case. Specific serological assays for ZIKV and CHIKV confirmed the qRTPCR data. Analyses of DENV IgM and IgG ELISAs in the context of qRTPCR results suggested high levels of cross reactive antibodies in ZIKV-positive samples. Results from neutralization assays highly corroborated those from qRTPCR and ZIKV ELISA, indicating very few positive DENV cases. ZIKV infections were temporally clustered in the first months of the study and started to decrease concomitantly with an increase in CHIKV infections in August 2015. The proportion of CHIKV infections increased significantly in September 2015 and remained high until the end of the study period, with an average of 84.7% of recruited patients being diagnosed from August 2015 to May 2016. ZIKV infections exhibited a female bias and the cases were spread over the study site, while CHIKV cases had a male bias and were spatially clustered in each month.

Conclusions: In 2015-2016 in the Recife Metropolitan Region, we detected the tail end of a Zika epidemic, which was displaced by a chikungunya epidemic. Few dengue cases were identified despite a high number of official dengue notifications in the area during this period. We show here important epidemiological features of these cases.

Fig 1. Females and males in different age groups.

(A) Number of females and males in different age groups among all recruited patients. (B) Number of females and males in different age groups among patients infected with Zika virus (ZIKV). (C) Number of females and males in different age groups among patients infected with chikungunya virus (CHIKV).

Fig 2. Spatial distribution of recruited patients in the study area (Recife Metropolitan Region), from May 2015 to May 2016.

(A) Patients that completed the study. (B) Patients who were considered as loss. The geodetic reference system SIRGAS2000 (Geocentric Reference System for the Americas) was the coordinated system used to represent geometric or physical terrestrial characteristics (http://www.ibge.gov.br/english/geociencias/geodesia/pmrg/faq.shtm#1).

Fig 3. Dengue virus (DENV) serological assays.

(A) Amount of IgM reactive to DENV in the acute and convalescent samples of participants infected with Zika virus (ZIKV). (B) Amount of IgM reactive to DENV in the acute and convalescent samples of participants infected with chikungunya virus (CHIKV). (C) Amount of IgG reactive to DENV in the acute and convalescent samples of participants infected with Zika virus (ZIKV). (D) Amount of IgG reactive to DENV in the acute and convalescent samples of participants infected with chikungunya virus (CHIKV). ELISA optical density was converted in Panbio units. Asterisks reflect the level of significance between groups after paired T-test was performed: **p <0.05, ****p <0.0001; NS = non-significant.

Fig 4. Zika and chikungunya infections.

Absolute number of participants infected with Zika virus (ZIKV) and chikungunya virus (CHIKV), and those not infected with either virus (Non-ZIKV, non-CHIKV), per month.

Fig 5. Spatial distribution of participants infected with Zika virus (ZIKV) or chikungunya virus (CHIKV) in the study area, from May 2015 to May 2016.

The distribution of patients was superimposed with a satellite photograph of the region. Confirmed and suspected cases of Zika virus (ZIKV) or chikungunya virus (CHIKV) infection were used in the analysis. Red star: Zika case; green square: chikungunya case. The geodetic reference system SIRGAS2000 (Geocentric Reference System for the Americas) was the coordinated system used to represent geometric or physical terrestrial characteristics (http://www.ibge.gov.br/english/geociencias/geodesia/pmrg/faq.shtm#1).

Fig 6. Spatiotemporal distribution of patients infected with Zika virus (ZIKV) or chikungunya virus (CHIKV), from May 2015 to May 2016, in the study area.

Spatiotemporal distribution of Zika virus (ZIKV) or chikungunya virus (CHIKV) infection cases in periods of three to four months, from May 2015 to May 2016. Confirmed and suspected cases were used in the analysis. Red star: Zika case; green square: chikungunya case. The geodetic reference system SIRGAS2000 (Geocentric Reference System for the Americas) was the coordinated system used to represent geometric or physical terrestrial characteristics (http://www.ibge.gov.br/english/geociencias/geodesia/pmrg/faq.shtm#1).

Fig 7. Hotspots of the distribution of participants infected with Zika virus (ZIKV) or chikungunya virus (CHIKV), from May 2015 to May 2016, in the study area.

Kernel density map of the distribution of patients with a confirmed or suspected infection with Zika virus (ZIKV) of chikungunya virus (CHIKV). The geodetic reference system SIRGAS2000 (Geocentric Reference System for the Americas) was the coordinated system used to represent geometric or physical terrestrial characteristics (http://www.ibge.gov.br/english/geociencias/geodesia/pmrg/faq.shtm#1).