Study Design
Critical appraisal involves a sequential, balanced evaluation of an article to evaluate whether it is appropriate, trust-worthy and relevant (Steen, 2011). A detailed critical-review of the case-control study on ‘Association of infection due to Zika virus (ZIKV) with neonatal-microcephaly from Jan ’16 to May ’16 in Brazil’ was done. This appraisal clearly judges the merits as well as demerits of the study (Marriam-Webster, 2015). Such type of critical evaluation has to be done by the health professionals to evaluate its applicability in patient-care.
The study purpose is to determine the association between ZIKV-infections at pregnancy time with neonates born with microcephaly in Brazil. The case-control research-design along with prospective method of recruiting newborn cases with concurrent controls was employed in this study. This design is most appropriate for the study purpose as it is a non-experimental method that involves comparing the cases (i.e. individuals with some condition that requires analysis) with a matched-control group (i.e. similar individuals without condition) (Polit, 2016). Basically, it involves defining groups based on the outcome (presence/absence of disease) and the study factors (presence/absence of exposure) (THS, 2015). Prospective method was employed that begins with examining presumed causes (ZIKV) and then moves forward (in-time) to observe effects (neonatal-microcephaly). They have selected this design as they cannot be able to manipulate the cause (ZIKV) in pregnant-mothers to test its relationship with microcephaly. Additionally, they used prospective-design to find the existing association between 2 groups (cases & controls) so as to generalize the result.
The target-population is defined as the total aggregate of population in which the study-researcher is interested and is willing to generalize the results (Polit, 2016). They have selected the neonates who were gave birth by mothers living in the Recife of Pernambuco, Brazil and were delivered in 8 of the Government-maternity centers from 15th Jan to 2nd May‘16. They have selected neonates as their target population from Pernambuco as it has highest incidence of microcephaly. A sample is nothing but a subset of population, which is selected by any sampling-techniques to participate in the research-study. In this study, the samples for case-group with neonates born with microcephaly (with 2-SD head-circumference less than the mean for gender as well as gestational-age based on ‘Fenton’s Growth-chart’) (2013) and for control-group with live neonates with no evidence of microcephaly, brain-abnormalities (confirmed by trans-fontanellar scan) with brain-defects (physical-assessment) were selected.
Matching means the pairing of samples in 1 (case) group with that of the samples in other (control) group based on the similarity-features in one/more aspects to promote the comparability between groups (THS, 2015). They have appropriately matched the cases with controls by selecting 2 controls for 1 case from the 1st neonate that is born in the 8 maternity study-centers. They have matched both the area of residence and the EDD (expected date of delivery of neonate) of the participant mothers to determine that the mothers in case-group with control-group have conceived at a similar epidemic stage. Their criteria for pairing the EDD of cases and controls were specific for case’s gestational-age.
Target Population and Sample Selection
The study factors involve variables for indicating presence and/or absence of any exposure to viral infections (THS, 2015). In this study, they have analyzed the maternal- serum of both cases and controls for identifying presence of exposure to ZIKV and dengue virus by plaque-reduction neutralization’s assay and estimated gestational-age by ultrasonography or EDD to rule-out any evidence of exposure in antenatal period. Outcome factors involve variables that measure the outcome of an exposure with presence and/or absence of a disease (THS, 2015). In this study, they have identified the presence of microcephaly by measuring the neonate’s head-circumference, taking maternal and neonate’s blood samples (cases and controls), collecting CSF-samples from microcephaly-neonates and collecting umbilical-cord blood from both neonatal-groups. The radiologist has taken CT-imaging (without contrast) in cases and trans-fontanellar-scan in controls. They have sent blood specimens to Virology- & experimental- therapy’s department, Pernambuco to rule-out presence of ZIKV.
They have identified the confounding variables that obscures the association between dependent and independent variables (Polit, 2016). They have tried to identify and control the confounding variables as region of residence as well as the EDD. They have mentioned about their method of controlling the confounding variable (residential area) by matching one case with two controls which are born in the next morning to case’s birth; in 8 selected-centers. They have controlled ‘EDD’ variable appropriately by selecting the cases and control; who has conceived at same epidemic-stage. They have also mentioned that they have avoided the confounding biases by matching the term/ post term neonates, preterm and 34- 36 wk neonates of cases with similar term controls.
Their use of Fenton’s-chart rather than WHO’s chart (2016a) for ruling-out neonatal-microcephaly could cause bias in classifying neonates yet it seems to be adequate. In analysis, two cases were linked with one matched control and the reduced participation of control (76%) as compared to cases (100%) may cause bias in association.
They have calculated the crude-OR with 95%-Confidence Interval in-order to associate microcephaly with lab-ZIKV confirmation by accounting the findings of serum and/or CSF and with serum alone; both total and individualized for neonatal-cases with/ without radiological-evaluation for brain-defects. As the tests of controls showed negative for ZIKV, OR was estimated through a median-unbiased analyzer for analyzing binary-data in a logistical model of regression. This analysis is adequate for zero-cells and hence it is appropriate to this study with smaller sample-size and with sparse structure of data. Moreover, for the controls with negative ZIKV, conditioned-matched estimation wasn’t performed as either matched/un-matched estimation may produce similar result. They have estimated OR-adjusted in mother’s age and educational-status for overall study’s association. They have also evaluated whether the results of serum and CSF for ZIKV antigen-IgM was similar and have compared the ELISA finding in serum and CSF. They have utilized Stata-version (14.1) software in analysis.
The strength of this study includes recording a multi-typic (higher-frequency) flavi-viral infection that includes ZIKV, DENV-3 with DENV-4 viral-serotypes and dengue viral-serotype (Castanha, 2016). Their use of diagnostic-test (ZIKV-IgM capture-ELISA) that is sanctioned by ‘Food & Drug-Administration’ (2016) for testing antibodies adds to its strength. Their confirmation of positive-cases for ZIKV-IgM by testing its neutralizing effects to find out any false-positivity results is appropriate (Rabe, 2016). The limitation of this study lies in the preliminary analysis. Though they have fasten the study to rule out association in urgency, their interim-analysis may over-estimate the association’s strength and hence its magnitude should be carefully treated. They have not collected CSF samples from controls for ethical reasons (with no clinical indication) and hence the lab confirmation in this association is completely fair. The lab tests were not validated which could cause problems in ZIKV-confirmation (WHO, 2016b). Its weakness involves absence of generalization of findings in larger group. It is carried-out in one geographical area and hence could not be applied in other areas.
Matching and Study Factors
They have used confidence-interval to rule-out inferences by selecting 32 mothers-of-cases and 62 mothers-of-controls in which 24 cases (80%) out of 30 showed ZIKV-infection as compared to 39 controls (64%) out of 61 (p=0.12) with no ZIKV-infection (Newcombe, 2012). Among 32-cases, 13 (41%) and in 62-controls 0 (100%) had lab-ZIKV confirmation. The crude overall-OR of 55.5 (CI-95%: 8.6-∞); OR of 113.3 (CI-95%: 14.5-∞) for 7 cases having brain-defects and OR of 24.7 (CI-95%: 2.9–∞) for 4 cases with-out brain-defects. It suggests that microcephaly epidemic has resulted in the neonatal ZIKV-infection.
By considering strengths and delimitations, the study-findings seem to be crucial in this epidemic-stage in Brazil and other countries. It gives a clear picture that the ZIKV-infection at pregnancy-time affects the neonate’s brain-growth adversely resulting in microcephaly. It is highly surprising to find-out that an epidemic viral infection has affected the fetal brain. But this study has created awareness about the link between ZIKV-infection and microcephaly.
The authors have concluded that the neonatal-microcephaly epidemic has resulted from the ZIKV-infection. They recommended renaming the TORCH with congenital-infections as toxoplasmosis, other (syphilis, varicellar-zoster & parvo-viral-B1), rubella, cytomegalo-viral and herpes as TORCHZ for ZIKV. They have advised the countries to prepare themselves to face microcephaly epidemic with features of ZIKV suggesting the study findings are valid.
This study will create awareness among people about ZIKV causing congenital microcephaly. This will motivate them to control its transmission through pregnancy by family-planning/abortion to avoid microcephaly (Cauchemez, 2016, Brasil, 2016). It will help Governments to prevent vertical-transmission of ZIKV that occurs before/during pregnancy (Besnard, 2014). These findings will help to reduce neonatal-microcephaly that has progressed from 0.6/10,000 live-births (2010) to 4.2-8 (2012-15) in Brazil to take preventive measures (de-Araujo, 2016). It will definitely help health-professionals to manage microcephaly-neonates. Neonate with mild-microcephaly may not require special treatment whereas for severe type, early interventions as speech, physical with hearing therapies are needed (Cragan, 2016).
This is the first study to associate ZIKV with neonatal-microcephaly. The study finding is of greater public-health significance as it identifies the risk of ZIKV epidemic. The study purpose, design, population, samples, exclusion and inclusion criteria, procedure of collecting data, statistical analysis technique and reporting results are appropriate and adequate. Overall, the findings are clear and concise and will definitely help to take preventive measures to control microcepahly in the future.
Reference
Besnard, M et al. (2014). Evidence of perinatal transmission of Zika virus, French Polynesia, December 2013 and February 2014: Euro Surveill. 19(13): pii=20751.
Brasil, P et al. (2016). Zika Virus Infection in Pregnant Women in Rio de Janeiro: N Eng J Med. 375(24): 2321-2334.
Castanha, P.M.S et al. (2016). Placental transfer of dengue virus (DENV)-specific antibodies and kinetics of DENV infection-enhancing activity in Brazilian infants: J Infect Dis. 214: 265–72.
Cauchemez, S et al. (2016). Association between Zika virus and microcephaly in French Polynesia, 2013-15: a retrospective study: Lancet. 387(10033): 2125-2132.
Cragan, J.D. (2016). Surveillance for Microcephaly: National Center on Birth Defects and Developmental Disabilities. Retrieved from https://www.cdc.gov/ncbddd /birthdefects/ documents/surveillance-microcephaly-webinar.pdf
de-Araújo, J.S et al. (2016). Microcephaly in north-east Brazil: a retrospective study on neonates born between 2012 and 2015: Bull World Health Organ. 94(11): 835-840.
Marraim-Webster. (2015). Critique. Retrieved from https: www//learners dictionary.com/search/ Critique
Newcombe, R.G. (2012). Confidence Intervals for Proportions and Related Measures of Effect Size. Retrieved from https://books.google.co.in/books?isbn=1439812780
Polit, D.F & Beck, C.T. (2016). Nursing Research: Generating and assessing evidence for nursing practice. Lippincott Williams & Wilkins: New Delhi.
Rabe, I.B et al. (2016). Interim guidance for interpretation of Zika virus antibody test results: MMWR Morb Mortal Wkly Rep. 65: 543–46.
Steen, M. & Roberts, T. (2011). The handbook of midwifery research. Wiley-Blackwell: West Sussex.
THS. (2015). Epidemiology- Glossary of Epidemiological and Statistical Terminology: Tropical Health Solutions. Retrieved from https://www.tropicalhealthsolutions.com/statsglossary
US Food and Drug Administration. (2016). Zika virus emergency use authorization. Retrieved from https://www.fda.gov/MedicalDevices/Safety/EmergencySituations /ucm161496.htm
WHO. (2016a). Screening, assessment and management of neonates and infants with complications associated with Zika virus exposure in utero: Interim guidance update-June 30, 2016. Retrieved from https://apps.who.int/iris/bitstream/10665/204475/1/WHO_ ZIKV_MOC_ 16.3_eng.pdf?ua=1
WHO. (2016b). Screening, assessment and management of neonates and infants with complications associated with Zika virus exposure in utero: Rapid Advice Guideline. WHO/ZIKV/MOC/16.3/Rev3WHO-Aug 30, 2016. Retrieved from https://www.paho.org/hq/index.php?option=com_docman&task=doc_download& Itemid=270&gid=33551&lang=en