1 BACKGROUND

Microcephaly is defined as having a head that is smaller than expected compared with infants of the same sex and gestational age (NBDPN, 2016). Microcephaly can be isolated or occur in combination with other birth defects; it can be present at birth (congenital) or acquired after birth (Auger, Quach, Healy-Profitós, Lowe, & Arbour, 2018; Hoyt et al., 2018; Leviton, Holmes, Allred, & Vargas, 2002; NBDPN, 2016; NCBDDD, 2020; von der Hagen et al., 2014).

Population-based survival of babies born with severe microcephaly are sparse and the case definitions and study inclusion criteria are varied (Auger et al., 2018; Glinianaia et al., 2022; Nembhard, Waller, Sever, & Canfield, 2001). In a recently published, large European survival study, investigators used a similar definition of severe microcephaly where they examined survival of severe microcephaly as isolated as well as in the presence of co-occurring birth defects. This study documented higher survival among isolated severe microcephaly as compared to severe microcephaly with co-occurring birth defects (Glinianaia et al., 2022). A 23-year hospital discharge-based study in Quebec found 20 times higher risk of mortality among infants with moderate (head circumference 2 standard deviations from age and sex matched neonates) or severe (head circumference 3 standard deviations different) microcephaly (Auger et al., 2018). Over 20 years ago, a survival analysis among specified birth defects was conducted that also documented lower infant survival rates for microcephaly in the presence of additional birth defects, regardless of severity (Nembhard et al., 2001).

The objective of this analysis was to describe the patterns and factors associated with survival among children up to 4 years of age with a diagnosis of severe microcephaly at birth. Our study includes delivery years from 1999 through 2015, which was prior to any known Zika virus-associated microcephaly in Texas. Only infants with congenital microcephaly that were considered severe (<3rd percentile for head circumference for sex and gestational age) were included, as this phenotype is most likely to be associated with greater disability (Ashwal et al., 2009; Leviton et al., 2002; McElrath et al., 2010; NBDPN, 2016; NCBDDD, 2020).

2 METHODS

The Texas Birth Defects Registry (TBDR) is a population-based active surveillance system that ascertains major structural malformations and chromosomal anomalies. Annually, all registry records are linked to their corresponding vital record(s) to provide or supplement sociodemographic data (e.g., maternal education, race/ethnicity, legal residence, indication of diabetes, maternal smoking, and maternal birthplace), as well as vital status. Texas participates in the State and Territorial Exchange of Vital Events system to share interstate vital events data (NAPHSIS, 2021).

Only liveborn infants with a definite diagnosis of congenital microcephaly (small head) of any severity (modified British Paediatric Association [BPA] codes 742.100 or 742.486) in the TBDR for delivery years 1999 to 2015 were eligible for inclusion in our study. Next, severity of microcephaly was determined based on percentile of head circumference at delivery for gestational age and sex, as standardized by the INTERGROWTH-21 Newborn Growth and the Very Preterm Size at Birth reference tables (NBDPN, 2016; NCBDDD, 2020; WHO, 2021). Infants with gestational ages of less than 24 weeks or greater than 42 weeks were excluded as outliers, because reference standards were not available. The final case definition for our analysis included only infants with severe congenital microcephaly whose head circumference at delivery was less than the 3rd percentile for the expected infant's head circumference for sex and gestational age. Information on any co-occurring birth defects was also used to evaluate severe microcephaly survival with and without co-occurring birth defects. Infants were stratified as having (1) isolated severe microcephaly; (2) severe microcephaly with a known co-occurring chromosomal or known syndromic condition (referred to as “chromosomal/syndromic”); or (3) severe microcephaly with co-occurring multiple birth defects of unknown chromosomal or syndromic association (referred to as ‘multiple’). Infants with co-occurring birth defects were also further evaluated for the presence of any critical congenital heart defects (CCHDs), as these are among the most serious of heart defects which require early medical intervention. The presence of CCHDs was determined using National Birth Defects Prevention Network Guidelines (NBDPN, 2017). Finally, we examined survival among infants with proportionate microcephaly; these infants were characterized as having a head circumference at delivery less than the 3rd percentile for expected head circumference, but the small head circumference was proportionate in relation to their body length, weight, sex, and gestational age, using INTERGROWTH-21 parameters. Clinically, an infant with proportionate microcephaly (i.e., a small baby overall) may not have any underlying brain abnormality and may be asymptomatic (Ashwal et al., 2009; Cragan et al., 2016; Hoyt et al., 2018; Leviton et al., 2002).

We submitted records of live-born infants in Texas with severe microcephaly delivered through 2013 and without an indication of death from our Texas vital records linkage to the National Death Index (NDI), which contains all deaths in the United States since 1979, to identify any additional deaths that BDES vital records linkage may have missed. No additional NDI deaths were ascertained for 1999–2013 beyond those found using deterministic linkage to the Texas vital records. Severe microcephaly deliveries in 2014–2015 included in our study were not submitted to the NDI; however, we are confident that our ascertainment of severe microcephaly deaths for those years was also quite comprehensive, given the NDI search results for the previous years. Death information was abstracted from linked Texas vital record death certificates identified through the most recent year of available death data in 2019. As we were only able to obtain indications of death through 2019, we decided to end the analysis of survival to 4 years of age, as that is the oldest survival age consistently available for the entire cohort. Infants with severe microcephaly having either an underlying or multiple contributing cause of death resulting from deaths due to accidents or violence were right censored, keeping data until the point where it is lost to follow-up. The only accidental death causation retained in the study was death due to aspiration asphyxiation, because feeding and swallowing difficulties are associated with severe microcephaly (Falcheck, 2019; NCBDDD, 2020). Survival probabilities by selected maternal and infant cofactors were analyzed using the Kaplan–Meier method and Cox proportional hazards models in SAS v9.4 (SAS Institute, Inc., Cary, NC, USA). The underlying cause of death (UCD), which is considered to be the condition that results in death, was also tabulated in order to describe causes of death, using the 130 Selected Causes of Infant Death (CDC & NCHS, 2020).

We stratified by a number of covariates. Maternal cofactors included: age group (<35 years, >34 years), educational level (≤high school, > high school), and race/ethnicity. Infant cofactors included: sex, gestational age at delivery (very preterm: 21–31 weeks; moderate preterm: 32–36 weeks; full-term >36 weeks), birthweight (<2,500 g [g], >2,499 g), proportionate or disproportionate microcephaly, presence versus absence of a co-occurring critical congenital heart defect (CCHD), and the classification group for the child, based overall on the presence versus absence of any co-occurring birth defects: (isolated, chromosomal/syndromic conditions, or multiple).

3 RESULTS

The TBDR included 2,704 live-born infants with severe microcephaly out of 6.5 million live-born infants under surveillance during delivery years 1999–2015. Severe microcephaly represented 34.1% of all infants with any severity of microcephaly identified during this time period. The overall birth prevalence of severe microcephaly was 4.13 per 10,000 live births (95% CI = 3.98–4.29). Overall, almost half (47.1%) of infants with severe microcephaly were firstborn, and the majority were full-term deliveries (72.6%); delivered to mothers <35 years of age (87.1%) and with a lower educational attainment (no post-high school education; 68.1%). The majority (95.5%) of infants with severe microcephaly were singletons (data not shown).

We followed deaths for this cohort through the year 2019. We identified three motor vehicle deaths and a homicide which was right censored. Nineteen infants with severe microcephaly were included in our analysis as they had an indication of death documented in the medical record, but no corresponding death certificate could be found, either through data linkage or the NDI search.

We found that 496 of the 2,704 infants with severe microcephaly died at 4 years of age or younger. Most deaths (42.9%) occurred in the neonatal period, another 39.9% died by 1 year of age, and an additional 17.1% died by 4 years of age. Among all infants with severe microcephaly, survival was 84.8% (95% CI = 83.4–86.1), and survival to 4 years of age was 81.7% (95% CI = 80.1–83.1). Over the study period timeframe (1999–2015), severe microcephaly survival increased by approximately 1.5% per year (data not shown).

Table 1 shows numbers of Texas infants with severe microcephaly, numbers of deaths, and neonatal, infant, and 4-year survival rates by selected maternal and infant cofactors. While most infants in our study were full-term, over half were of low birth weight. Six cofactors were highly associated with survival of infants with severe microcephaly: gestational age (p < .01), birthweight (p < .01), co-occurrence of a CCHD (p < .01), maternal age (p = .01), having proportionate microcephaly (p < .01), and having any co-occurring major birth defect (p < .01). The lowest infant survival rates were observed among very preterm (24–31 weeks) births (63.9%; 95% CI 54.1–72.1), for infants with co-occurring CCHDs (44.0%; 95% CI = 36.0–51.7), and for infants with chromosomal/syndromic anomalies (66.1%; 95% CI = 62.5–69.4). Trisomies comprised 329/720 (45.7%) of the co-occurring chromosomal/syndromic anomalies. In descending prevalence of occurrence: trisomy 21 (17.5%); trisomy 18 (9.6%); partial trisomies (9.3%); trisomy 13 (8.2%); other trisomies (1.1%). The highest infant survival rates were seen for infants with isolated severe microcephaly (i.e., no other major birth defects) (99.0%; 95% CI = 98.0–99.5) and infants of normal birthweight (91.7%; 95% CI = 89.8–93.2). Infant sex, maternal race/ethnicity, and maternal education were not associated with infant survival (Table 1).

Similar associations between the study factors above and survival were observed among neonates and 4 year olds, with two exceptions. Although maternal race/ethnicity was not associated with infant survival, this factor was statistically significantly associated with survival among both neonates (p = .01) and 4 year olds (p = .03). Additionally, infant sex was significantly associated only with 4-year survival (p = .03).

Figure 1a–d shows comparisons of 4-year survival curves with their 95% confidence intervals and infant survival reference line for some important subgroups of infants and children with severe microcephaly. All curves show little change in survival after infancy. Figure 1a displays how the presence or absence and type of co-occurring birth defects impacted overall survival. One of the most impactful factors for severe microcephaly survival was having a chromosomal/syndromic co-occurring anomaly; these had an infant survival rate of 66.1% (95% CI = 62.5–69.4) as compared to 99.0% (95% CI = 98.0–99.5) survival of infants without co-occurring defects (isolated severe microcephaly). Figure 1b demonstrates that children born very preterm with severe microcephaly have a markedly lower survival curve than those born pre-term or full-term. The moderately (32–36 weeks) preterm and full-term birth survival curves show close and overlapping confidence intervals from the late postneonatal time-period onward, indicating that survival estimates for these two groups were not significantly different. Figure 1c shows that children with co-occurring CCHDs had a markedly lower survival across the early childhood lifespan. While highly correlated with low survival (38.7%), there were very few (5.5%) infants with severe microcephaly that had a CCHD (Table 1). Figure 1d depicts that among severely microcephalic children with chromosomal/syndromic disorders, those who had proportionate severe microcephaly had the poorest infant survival of 50.9% (95% CI = 45.4–56.1).

Table 2 provides neonatal, infant, and 4-year-old survival estimates for severe microcephaly for each study factor, stratified by classification of co-occurring birth defects (chromosomal/syndromic, isolated, multiple). Only very preterm birth (24–31 weeks) was significantly associated across all strata. The isolated microcephaly group often had too few deaths to be able to calculate stable survival estimates. Infant survival associations for overall severe microcephaly shown in Table 1 with gestational age, birthweight, CCHD co-occurrence, maternal age, and head-to-body proportion were found to be largely limited to the subgroup of infants with chromosomal disorders and syndromes (Table 2). Although infant sex was not associated with survival among all infants overall, it was associated with infant survival among the chromosomal/syndromic subgroup (p = .01). Within the chromosomal/syndromic subgroup, the lowest infant survival estimates were seen, particularly for infants who were very preterm (40.0%; 95% CI = 19.3–60.0), had co-occurring CCHDs (35.0%; 95% CI = 25.9–44.1), were delivered to older mothers (58.5%; 95% CI = 50.1–66.0), or were of proportionately small size (50.9%; 95% CI = 45.4–56.1). Among infants in the multiple defect categories, only two cofactors were associated with infant survival: gestational age (p < .01) and co-occurring CCHDs (p < .01). Also, 4-year severe microcephaly survival was associated with low educational attainment (p = .04) among these infants with multiple defects. Cofactors that were associated with survival of infants with severe microcephaly and either chromosomal/syndromic conditions or with co-occurring multiple defects generally had similar associations among neonates and 4 year olds.

4 DISCUSSION

This study provides population-based neonatal, infant, and 4-year survival estimates for severe microcephaly in Texas. We identified associations between maternal and infant characteristics and among infants and children with severe microcephaly during delivery years 1999–2015. We found that 42.9% of severe microcephaly deaths occurred in the neonatal period, with 4 out of 5 deaths occurring by infancy. This is concordant with a prior Texas survival analysis of a spectrum of birth defects, which found 86.6% of all birth defects deaths occurred in infancy and another 9.7% occurred between 1 and 3 years of age (Nembhard, Salemi, Ethen, Fixler, & Canfield, 2010). Our results also aligned with a recent European analysis that noted most birth defect mortality occurred in infancy (Glinianaia et al., 2022).

Among infants with severe microcephaly, our study found an overall survival rate of 84.8%, compared with a 93.7% infant with microcephaly survival rate reported in a Canadian hospital discharge study (Auger et al., 2018). However, Auger et al. (2018) had a much higher percentage of infants with isolated microcephaly (62.8%) and a lower number of infants with chromosomal defects (5.9%) than in our cohort, which may account for their higher rate of infant survival. Just over a quarter (26.6%) of Texas infants with severe microcephaly had co-occurring chromosomal/syndromic anomalies, which is similar to the proportion of infants with microcephaly reported in a German hospital-based study (29%) (von der Hagen et al., 2014). Survival varied by the presence and type of co-occurring congenital anomalies. Infants with isolated severe microcephaly had the highest infant and 4-year survival rates (99.0%). A recent European study also observed a high survival rate (97.2%) among infants with isolated severe microcephaly in contrast to severe microcephaly with co-occurring birth defects (Glinianaia et al., 2022). However, Glinianaia et al. (2022) reported a higher percentage of isolated microcephaly in their cohort (45.2%) than in our study (32.1%). We found the lowest infant survival was among infants with severe microcephaly and with co-occurring chromosomal/syndromic associations (66.1%), followed by infants with co-occurring multiple defects (85.8%). Benjamin et al. (2021) also observed lower infant survival (83.8%) among infants with syndromic birth defects relative to infants with isolated birth defects. Sattolo et al. (2022) described a correlation between multiple birth defects and higher mortality. Nembhard et al. (2001) documented lower survival rates among any severity of microcephaly in the presence of additional birth defects, although unlike in our study, they excluded any infants with trisomy 13 (T13) and trisomy 18 (T18) from the Nembhard study. Children with co-occurring CCHDs had a markedly lower survival across the early childhood lifespan. Congenital malformations of the heart were listed in the top 5 underlying causes of death for babies with severe microcephaly and co-occurring defects, although only 5.5% of infants in our study had the most severe form of critical congenital heart defects.

Although most of our infants were full term deliveries, infant survival was lowest among those born very preterm (24–31 weeks of age), with only 63.9% of very preterm infants surviving. More Texas infants with severe microcephaly were born term and less were born preterm than reported by Cragan et al. (2016). Nembhard et al. determined gestational age to be strongest predictor of mortality among infants with birth defects, with death occurring among 47% of those with preterm birth (20–36 weeks), which was 2.7-fold higher than among those born at term (Nembhard et al., 2010). This is comparable to our findings of an almost three-fold increased risk of death among infants delivered very preterm with severe microcephaly and a 27% higher risk for death among infants delivered between 32 and 36 weeks gestation, relative to infants born full-term.

Proportionate severe microcephaly in the presence of co-occurring chromosomal/syndromic defects was associated with an almost three-fold elevated risk of death. This association was not seen for infants with isolated proportionate severe microcephaly, nor among infants with proportionate severe microcephaly and non-syndromic multiple defects. While our percentage of proportionate microcephaly was similar to that seen by Nembhard et al. (2010), it was lower than observed in a German hospital-based study who also noted genetic causes among their microcephaly patients (von der Hagen et al., 2014).

The underlying cause of death was rarely attributed to microcephaly. The leading underlying cause of death in our study was having a chromosomal/syndromic co-occurring defect, of which trisomy 13 and 18 were the most common specified.

5 STRENGTHS AND LIMITATIONS

This study's greatest strength is that Texas has a large population-based active surveillance birth defects registry. Texas residents delivered over 6.5 million.

live-born infants during 1999 to 2015 that were surveilled by the TBDR. The TBDR had large enough numbers to categorize infants with severe microcephaly by co-occurring birth defects and to classify infants into isolated, chromosomal/syndromic, and multiple defects categories to better explore potential factors associated with severe microcephaly survival. Another strength is the validation of interstate death data sharing efficacy by searching the National Death Index through delivery year 2013, which permitted the capture of all severe microcephaly deaths as well as the opportunity to evaluate the comprehensiveness of data sharing interstate agreements for Texas resident deaths for this specific malformation.

One study limitation may be the potential for anthropometric measurement errors. Head circumference measurements taken at birth may vary due to postpartum cranial molding and infant movement impacting those categorized as having severe microcephaly. Gestational ages were calculated from the last menstrual period (LMP) and are subject to error; if menstrual cycles are irregular or if there is spotting during pregnancy, this may result in an inaccurate LMP. Both the LMP and the clinical estimation of gestational age are especially subject to measurement error among preterm births, and our preferred use of LMP to calculate gestational age may have overestimated the number of preterm births (Qin, Hsia, & Berg, 2008). Additionally, some significant findings may be due to chance because of the issue of multiple comparisons. We stratified various cofactors over three time periods (neonatal, infant and 4 year), by chance alone we estimate 21 significant findings, but our analysis found over double the number expected by chance alone. Another limitation is that not all of the 2,704 records were clinically reviewed for syndromes, resulting in potential underascertainment of the co-occurring chromosomal/syndromic birth defects classification. Lastly, the classification of multiple defects may contain currently undiscovered chromosomal/syndromic associations, which may have resulted in some misclassification and elevated results.

6 CONCLUSIONS

The majority (81.7%) of infants with severe microcephaly survived to 4 years of age. We report high survival among infants and children with isolated severe microcephaly and lowest survival among infants and children with both severe microcephaly and co-occurring chromosomal/syndromic conditions. Co-occurring trisomies were the leading underlying cause of infant death (27.5%) among infants with severe microcephaly, whereas the underlying cause of infant death was rarely attributed to microcephaly itself (2.2%). Presence of a co-occurring chromosomal/syndromic defect strongly reduced survival across several maternal and infant co-factors. Other factors that most heavily negatively impacted survival included very preterm birth, low birth weight, having proportionate microcephaly, and the presence of critical congenital heart defects. These results may be useful to guide clinicians in their communications with families of infants affected by severe microcephaly, and for tailoring interventions to improve survival.

CONFLICTS OF INTEREST

The authors have no known conflicts of interest to disclose.