Impact of the introduction of chikungunya and zika viruses on the incidence of dengue in endemic zones of Mexico

Abstract

Background: With the arrival of chikungunya (CHIKV) and zika (ZIKV) viruses in Mexico, there was a decrease in diagnosed dengue virus (DENV) cases. During the first years of cocirculation (2015-2017), the algorithms established by epidemiological surveillance systems and the installed capacity limited us to one diagnostic test per sample, so there was an underestimation of cases until September 2017, when a multiplex algorithm was implemented. Therefore, the objective of this study was determine the impact of the introduction of CHIKV and ZIKV on the incidence of diagnosed DENV in endemic areas of Mexico, when performing the rediagnosis, using the multiplex algorithm, in samples from the first three years of co-circulation of these arboviruses.

Methodology and principal findings: For this, 1038 samples received by the Central Laboratory of Epidemiology between 2015 and 2017 were selected for this work. Viruses were identified by multiplex RT-qPCR, and the χ2 test was used to compare categorical variables. With the new multiplex algorithm, we identified 2.4 times the rate of arbovirosis as originally reported, evidencing an underestimation of the incidence of the three viruses. Even so, significantly less dengue was observed than in previous years. The high incidence rates of chikungunya and Zika coincided with periods of dengue decline. The endemic channel showed that the cases caused by DENV rose again after the circulation of CHIKV and ZIKV decreased. In addition, 23 cases of coinfection were identified, with combinations between all viruses.

Conclusions and significance: The results obtained in this study show for the first time the real impact on the detected incidence of dengue after the introduction of CHIKV and ZIKV in Mexico, the degree of underestimation of these arboviruses in the country, as well as the co-infections between these viruses, whose importance clinical and epidemiological are still unknown.

Fig 1. Positive cases for each Virus/Year/Algorithm.

The graphs show the positivity for each virus: (A) DENV. (B) CHIKV. (C) ZIKV. Algorithm 1: Uniplex RT-qPCR reactions for each virus, with the restriction of performing only one reaction for each sample according to the initial diagnostic suspicion (details in S1 Fig). Algorithm 2: Multiplex RT-qPCR reaction that simultaneously detected the presence of the three viruses was performed on all samples, regardless of the initial diagnostic suspicion. *Significant differences (P<0.05) between the proportion of positive cases detected with algorithms 1 and 2. Note: in the last quarter of 2017, when the multiplex algorithm was implemented for all cases suspected of some vector-borne disease (VBD), a third-party kit was used for the diagnosis, and only from the second half of 2018, the LCE started using the TaqMan Zika Virus Triplex Kit also used in the present study.

Fig 2. Coinfections detected in the study.

The different combinations of viruses identified in cases of coinfection are broken down in the bars during the entire period of study.

Fig 3. Seasonality of viruses tested.

The graph was built from the estimated annual incidents generated with the data obtained in this work for each of the viruses, distributed monthly from 2012 to 2017.

Fig 4. Contributions of DENV, CHIKV, and ZIKV to the etiology of arbovirosis 2015–2017.

The graph was built from the estimated annual incidents generated with the data obtained in this work for each of the viruses, distributed monthly from 2015 to 2017.

Fig 5. Dengue endemic channel updated for 2018.

The endemic channel was built with the estimated incidences of 2015–2017, over which the estimated monthly cases for the year 2018 were plotted.