summary

350 words max/
this is the article

In 2015, after the outbreak of Zika virus (ZIKV) in Brazil, an increased incidence of congenital microcephaly was observed. Subsequently the causal relationship between ZIKV and a new clinical entity was established [1]: the congenital Zika syndrome (CZS) [2]. However, for children with possible CZS, the tests, specimens, and the ideal time window for a definitive diagnosis are not yet fully known. There are several reports of clinical cases consistent with CZS, but with negative tests at different time points [3], [4]. Therefore, we aimed to evaluate the prevalence of confirmed cases, according to the currently accepted laboratory criteria, among children with presumed CZS, considering clinical, epidemiological and neuroradiological criteria, followed in a referral rehabilitation programme at Salvador, Bahia, one of the most affected states during Brazil’s outbreak.

We performed a retrospective survey of the electronic medical records of suspected cases of microcephaly and/or congenital Zika infection who were enrolled in the rehabilitation programme between November 1st, 2015 and June 30th, 2018. The cases were selected if their initial International Classification of Disease code, version 10 (ICD-10), on the medical record was of microcephaly (defined as the head circumference at birth below 2 standard deviations for gestational age and sex, according to INTERGROWTH 21st) or Zika congenital infection (suspected if neuroradiological features typical of congenital infection and not explained by other causes were present in children born during the ZIKV outbreak in Brazil).

We performed a detailed evaluation of each selected medical record and considered confirmed those positive for immunoglobulin M (IgM) detected by ELISA or a positive RT-PCR assay on blood samples of the mother during pregnancy or the child during the first month after delivery. The cases considered ‘probable’ were those with abnormalities in neuroimaging suggestive of congenital infection (presence of cerebral calcifications and ventriculomegaly or malformation of cortical development) and exclusion of other common congenital infections (toxoplasmosis, rubella, syphilis and cytomegalovirus), but without a positive IgM or RT-PCR. This group was further subdivided into ‘probable symptomatic’, if there was a report of a rash illness during pregnancy suggestive of ZIKV infection, and ‘probable asymptomatic’ if there was no such report. The ‘doubtful’ cases were defined as those with incomplete or insufficient data to confirm or refute ZIKV as the cause of microcephaly. We categorized as ‘other causes’ those with a specific cause established after investigation, such as asphyxia, other congenital infections, or genetic diseases. The ‘exposed’ were those with possible exposure to ZIKV during pregnancy, but with normal neuroimaging, and without microcephaly or neurodevelopmental disorders.

During the study period, 538 patients were admitted with suspected microcephaly and/or CZS. Of those, only 11 cases (2%) were considered ‘confirmed’ (nine with an IgM-positive blood sample from the child, one with positive RT-PCR on the child’s blood and one with positive RT-PCR on the mother’s blood during pregnancy). We found 328 (61%) children classified as ‘probable’: 241 (45%) with report of a rash illness during pregnancy and 87 (16%) born to asymptomatic mothers. Other causes were identified in 155 children (29%), 39 cases (7%) were considered doubtful, and 5 (1%) were just exposed. Among the ‘probable’ group, the minority showed a positive or inconclusive IgG for ZIKV: 77 (32%) born to symptomatic mothers and 25 (29%) to asymptomatic mothers.

This survey shows a low prevalence of laboratory confirmation of CZS, using the currently accepted criteria, among children with this initial diagnosis who were enrolled into a rehabilitation programme in a referral hospital in Brazil. Also, it adds that 37% of the presumed CZS cases were probably misclassified at the beginning of the follow-up, which may indicate the need to re-estimate the contribution of CZS to the incidence rate of microcephaly in Bahia.

A limitation of this report is that serological confirmation by neutralization was not performed in any case, since this technique is not available in the study setting and is performed only in specialized research laboratories.

In individuals with a suspected ZIKV infection, it is considered confirmed when the virus is identified by a molecular method (RT-PCR) using blood or urine [5]. However, as the viraemia is usually transient and low, the success rate is higher when the test is performed within a week of the onset of symptoms. Therefore, a negative RT-PCR does not exclude the infection. Serological diagnosis is based on detection of specific IgM antibodies in serum or cerebrospinal fluid, from 4 to 5 days of the onset of symptoms for up to 12 weeks or more. The persistence of positive IgM for a longer period has been described. However, studies have demonstrated the low accuracy of commercial serological tests currently available [6]. One study reported intrauterine PCR negativization on a confirmed case of a fetus with symptomatic cerebral malformation whose RT-PCR, performed with 19 gestational weeks on amniotic fluid, was positive. The newborn, despite the typical clinical presentation, was not classified as a confirmed case, as both RT-PCR and IgM were negative [3].

We conclude that the currently accepted criteria for laboratory confirmation of CZS are hardly filled in the vast majority of suspected cases, which brings difficulties for clinical practice and in consulting families, as well as for public health and conduction of longitudinal follow-up studies of congenital Zika. Also, probably around 30–40% of the microcephaly cases at the beginning of the outbreak in Brazil may have been misclassified as being related to CZS. Therefore, the scientific community should discuss standardized criteria for the diagnosis of CZS, regarding clinical, epidemiological, neuroradiological and serological status in settings where the gold standard diagnosis confirmation is not possible.