In a June 30 post, I discussed a study published in the Lancet, titled “Congenital Zika virus syndrome in Brazil: a case series of the first 1501 live births with complete investigation”. (1) I had raised a number of issues and I wrote the corresponding author for some clarification. I have received a response from Prof. Cesar G Victora.
In 80% of those cases considered definite, there was no serology eliminating congenital syphilis, congenital toxoplasmosis or congenital CMV. I cannot understand how any definitive conclusion can be reached without a full TORCHS assay.
Although you are strictly correct, the clinical and neurorradiological picture of the affected babies is very different from CMV, toxoplasmosis or syphilis. CT scans reveal scattered brain calcifications which are very different from other congenital infections where these calcifications are periventricular. The cortical area usually are very thin and again much different from the other congenital infections. The clinical exam shows also deformities in the skull secondary to the brain abnormalities which were not seen before in other congenital infections. Moreover these children present cranio-facial disproportion, which is not characteristic of other congenital infections. It is extremely unlikely that in the presence of a ZIKV epidemic with a massive increase in the number of malformations, newborns with positive lab results for ZIKV would be also affected by other teratogens.
In the United States, the headlines are concluding that defects have been demonstrated in infants whose mothers had a Zika rash in the third trimester. The study cites just two sources for the conclusions in that respect, and both were in lab work using neural progenitors. If I understand correctly, that cell type disappears at around a gestational age of 17 weeks. I am not clear on how Petri dish work on early gestational cell types translates into “Malformations associated with late-pregnancy rashes” as the study states.
Our findings on late-pregnancy rashes are primarily epidemiological. In addition to our retrospective study, a small prospective cohort published in the NEJM (first author is Patricia Brasil) (2) also showed the late rashes were associated with malformation – albeit many of these children do not present with microcephaly because the skull had already grown. Smaller case series in Brazil confirm that malformations were found in women with late-pregnancy rashes. The laboratory studies showing the massive effect of ZIKV on neural cells (which also include in vivo animal studies and not only “Petri dish work”) provides a pathway explaining the epidemiological findings.
The discussion of neonatal mortality in Brazil, especially in NE Brazil, has confused me. If I read the study correctly, the normal neonatal mortalities for Brazil and for NE Brazil are compared to the neonatal mortality rates for the patients in the study, including the discarded cases. How can “normal” rates be compared to rates in a cohort that is born with birth defects?
We referred to the current mortality rates in the Northeaset – and avoided the use of the word “normal”. Infant, neonatal, and postneonatal mortality rates are commonly compared across groups, countries, and were indicators included in the Millennium Development Goals. We mention the current neonatal mortality rate in the Northeast to make the point that this is slightly smaller (but in the same ball park range) as our “discarded” cases, who did not have malformations but many of whom are likely to have suffered from intrauterine growth retardation associated with poverty and poor maternal nutrition. This information also provides greater relevance to the four fold increase we observed in the ZIKV definite/probable cases.
It is still an open question and Zika virus infection during pregnancy results in higher fetal and neonatal mortality, and our paper was the first to quantify the increased risk. We believe that this comparison is valid from a scientific point of view.
From 2010 to 2014, Brazil reported an average of 156 cases of microcephaly a year. That appears to be the basis of the statement that “fewer than 15 newborns with microcephaly were reported monthly in Brazil”. There exists a number of studies that suggest that the actual pre-Zika numbers may be 100 times greater or more. The lack of data on the discarded cases, admittedly containing patients with microcephaly not due to an infection, is worrisome given the existing under-reporting issues. As an example of the oddity of the reported Brazilian numbers, the U.S. state of Texas in the period 2006 to 2010 averaged 372 cases yearly in an average of 323,000 live births.
There is no question that microcephaly (even using the strict cut-off of – 3 standard deviations favored by geneticists and recommended by the Brazilan birth registration system) was historically underestimated in Brazil. We are not sure about the cutoffs used by the Texas study, but the most comprehensive estimate we located from the USA … “Practice parameter: Evaluation of the child with microcephaly” … (3) found a prevalence of around 500 cases per 100,000 newborns using the – 2 standard deviation cutoff. European and high-quality Latin America surveillance studies using the -3 cutoff typically report about 20 cases per 100,000.
What really strikes us is not only the numbers of cases, but that the vast majority of them show strikingly different phenotypes when compared to previous microcephalic newborns diagnosed in Brazil.
There are studies which demonstrate the rates for Down’s syndrome, fetal alcohol syndrome, congenital toxoplasmosis and congenital CMV in Brazil. In a paper currently being considered by the WHO Bulletin, I have suggested that these four causes of microcephaly alone may be responsible for over 4,600 cases of microcephaly in Brazil in 2015. The incidence of these four combined is 15.816 per 10,000.
Because our paper was based on routinely reported data, some newborns with other causes of microcephaly might have been included, particularly in the “somewhat probable” category. However, smaller studies led by the Brazilian Association of Medical Genetics (whose president is one of our co-authors, Dr Schuler-Faccini) leave no doubt that the marked increase in numbers is largely due to a completely new teratogen – definitely not Down’s syndrome or fetal alcohol syndrome, nor other infections.
In our series (which is obviously based on less than perfect, routine data), other causes of microcephaly would be either discarded if there was sufficient information on their non-Zika etiology, or included in the “somewhat probable” or the “moderately probable” categories.
We do agree with you that microcephaly was under-reported prior to the current Zika epidemic. Our difficulty in sorting out Zika-related cases from other causes of microcephaly, and from normal newborns with small heads (given the high cutoffs adopted) was what motivated us to publish the two papers that came out in the Lancet.
- The CT findings in this study, in the expert opinion of the authors, are quite different than that seen with other causes of microcephaly. While further studies should be conducted, it appears that a distinctive brain abnormality may be defining for congenital Zika virus syndrome.
- The under reporting of microcephaly may be due to the Brazilian reporting system. Prof. Lavinia Schuler-Faccini, a co-author of this study, notes in email correspondence that when the cause of microcephaly is known, that cause is normally the condition reported. “if a children has microcephaly AND Down Syndrome, this children in registered as DS and NOT as microcephaly.” The studies finding large numbers of unreported cases of microcephaly (4) (5) are likely valid but do not reveal the reported diagnoses of the patients. “Children with known syndromes are not included in our study.”
References and Citations
- França, G V A, et. al. Congenital Zika virus syndrome in Brazil: a case series of the first 1501 live births with complete investigation. Lancet. doi: http://dx.doi.org/10.1016/S0140-6736(16)30902-3
- Brasil, P., Pereira, Jr, J. P., Raja Gabaglia, C., Damasceno, L., Wakimoto, M., Ribeiro Nogueira, R. M., … & Calvet, G. A. (2016). Zika virus infection in pregnant women in Rio de Janeiro—preliminary report. New England Journal of Medicine.
- Ashwal, S., Michelson, D., Plawner, L., & Dobyns, W. B. (2009). Practice Parameter: Evaluation of the child with microcephaly (an evidence-based review) Report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology, 73(11), 887-897. doi: http:/?/?dx.?doi.?org/?10.?1212/?WNL.?0b013e3181b783f7
- de Araújo, J. S. S., Regis, C. T., Gomes, R. G. S., & Tavares, T. R. (2016). Microcephaly in northeastern Brazil: a review of 16 208 births between 2012 and 2015. Bull World Health Organ, 4.
- Campos, J. S., Cunha, A. J. L. A. D., Machado, M. M. T., Rocha, S. G. M. O., Silva, A. C., Rocha, H. A. L., … & Correia, L. L. (2016). Microcephaly: normality parameters and its determinants in northeastern Brazil: a multicentre prospective cohort study. doi: http://dx.doi.org/10.2471/BLT.16.171215