Microcephaly case-control study confirms Zika virus causality

Preliminary findings from 32 cases confirm causality but the true size of the effect will only be available following the full analysis of all 200 cases and 400 controls.

The relation between Zika virus and microcephaly is widely assumed to be causal because of strong evidence of an association. However, evidence so far comes from case reports, case series, modelling studies, and preliminary reports from cohort studies – none of which have included appropriate control groups.

Today, researchers from Brazil and the UK report the preliminary findings of the first case-control study examining the association between microcephaly and in utero Zika virus infection. The study, published in The Lancet Infectious Diseases journal, was requested by the Brazilian Ministry of Health to investigate the causes of the microcephaly epidemic that was declared a Public Health Emergency of International Concern in 2016.

The study included all infants born with microcephaly delivered in eight public hospitals in Pernambuco State in North Eastern Brazil between 15 January and 2 May 2016.

For each case, two controls were selected. Controls were the first two infants born the following morning without microcephaly in one of the eight hospitals. Controls and cases were matched for region of residence and expected date of delivery.

Blood samples from cases and controls were collected and samples of cerebrospinal fluid were collected from cases with microcephaly.

Samples were tested for Zika virus and Zika virus antibodies. Blood samples were collected from mothers and analysed for Zika and dengue virus. Infants with microcephaly had their head circumference measured and most underwent brain imaging.

24 of 30 (80%) mothers of infants with microcephaly had Zika virus infection, compared with 39 of 61 (64%) mothers of controls. 13 of 32 cases (41%) tested positive for Zika virus infection in blood or cerebrospinal fluid samples, and none of the 62 controls tested positive for Zika virus infection in blood samples.

A high proportion of mothers also tested positive for dengue and other infections such as cytomegalovirus, rubella, and toxoplasma but there was no significant difference between mothers of cases and controls.

Additionally, only seven of the 27 cases with microcephaly who had a brain scan had brain abnormalities, suggesting that congenital Zika virus syndrome can be present in neonates with microcephaly and no brain abnormalities.

“A high proportion of mothers of newborns with and without microcephaly had been infected with Zika virus, reflecting the rapid spread of Zika infection in this region,” says article author Dr Thália Velho Barreto de Araújo, Federal University of Pernambuco, Recife, Brazil.

“However, when we compared laboratory confirmed Zika virus infection in newborns with and without microcephaly, we found that about half of the cases with microcephaly had laboratory confirmed Zika virus infection, compared to none of the healthy controls.

“The presence of Zika virus antibodies in the cerebrospinal fluid indicates infection in the neural system of the neonate, but interestingly not all cases of microcephaly had brain abnormalities,”

The authors warn that preliminary analyses can overestimate the strength of an association, so the true size of the effect needs to be treated with caution. The full study, which will include 200 cases and 400 controls will help quantify the risk more precisely and shed light on the role of co-factors.

The authors add that detecting the presence of Zika virus or antibodies in blood and cerebrospinal fluid is the only current method of testing for Zika virus in newborns but the reliability of this method, especially when infections occur early in pregnancy, is not fully understood.

The authors say that these limitations might partly explain why 19 (59%) of microcephaly cases were not confirmed as positive for Zika virus.

“This is the first case-control study to examine the association between Zika virus and microcephaly using molecular and serological analysis to identify Zika virus in cases and controls at the time of birth,” says Dr Thália Velho Barreto de Araújo.

“Our findings suggests that Zika virus should be officially added to the list of congenital infections alongside toxoplasmosis, syphilis, varicella-zoster, parvovirus B19, rubella, cytomegalovirus, and herpes. However, many questions still remain to be answered including the role of previous dengue infection.”


Previous Jane Halton resigns, thanked by stakeholders
Next Could popping a pill replace exercise?

NOTICE: It can sometimes take awhile for comment submissions to go through, please be patient.

1 Comment

  1. pagophilus

    It does nothing of the sort. Read the study and the write-up and analyse them and you will see why. Here is my response to the article, to be found under Responses in the BMJ:

    “The results showed that 24 (80%) of 30 mothers of babies with microcephaly had Zika virus infection……….Thirteen of the 32 babies with microcephaly (41%) had laboratory confirmed Zika virus infection, but none of the controls did.”

    The headline is slightly misleading.

    This study shows that Zika virus is involved in microcephaly but not likely the sole cause, for what caused the microcephaly in the infants of the 20% of mothers who did not have laboratory-confirmed Zika infection, and in the 59% of babies without laboratory-confirmed Zika infection?

    A preliminary report published in the NEJM (Zika Virus Disease in Colombia – Preliminary Report, DOI: 10.1056/NEJMoa1604037) seems to indicate that infection in the third trimester is not related to microcephaly. I am eagerly awaiting further results from those women who contracted it in the first and second trimesters.

    Secondly, this study had a small sample size from a small area of Brazil. The abovementioned study from Colombia included 12000 women from 21 states and 5 districts. What differences are there between Colombia and Brazil. And if a disease occurs in clusters or specific areas, an environmental cause should be considered.

Leave a reply