Clinical outcome in newborns of perinatally COVID-19 infected women

Abstract

Objective

Maternal COVID-19 infection during pregnancy has been associated with adverse neonatal outcomes, such as prematurity and neonatal morbidity. Those adverse events are mainly attributed to maternal factors, rather than to the neonatal infection itself. Our aim is to add our experience and present the neonatal outcome of neonates born to mothers with perinatal SARS-CoV-2 infection.

Methods

This is a prospective case-control study with data from two Academic Tertiary Referral Hospitals in Greece. Electronic records of all births from SARS-CoV-2 positive mothers between March 2020 and April 2021 were analyzed. Demographic data, the severity of maternal COVID-19 disease, gestational age (GA), mode of delivery, birth weight (BW), need for resuscitation and/or supplemental oxygen and duration of hospitalization were recorded. A comparison with 2:1 matched neonates according to sex, GA, and BW born to SARS-CoV-2 negative mothers during the same period was performed. Chi-square and Mann–Whitney U test were used for categorical and non-categorical variables respectively.

Results

A total of eighty-one neonates were born to SARS-CoV-2 positive mothers during this period. Forty-three percent of pregnant mothers were asymptomatic. Median GA and median BW were 38 weeks (Interquantile range (IQR): 36–39 weeks) and 2940 gr (IQR: 2560–3340 gr) respectively. Prematurity was observed in 24.7% of the cases. Only 2 (2.4%) neonates were PCR positive after delivery. SARS-CoV-2 positive women were more likely to undergo Cesarean section. APGAR score at 5 min and the need for resuscitation did not differ between the two groups. In comparison with the control group, neonates born to SARS-CoV-2 positive mothers presented with gastrointestinal symptoms (53.6% vs 5.1%, p-value= <.001) and hospitalization was longer, mostly due to maternal factors.

Conclusion

In our study neonatal positivity was limited and no vertical transmission was noted. Neonatal outcomes were comparable to the control group. However, the presence of gastrointestinal symptoms in neonates born to PCR-positive women compared to controls needs further investigation.

Keywords:

• COVID-19
• neonates
• pregnancy
• perinatal infection
• SARS-CoV-2

Introduction

Since the beginning of Coronavirus disease 2019 (COVID-19) pandemic, the association of maternal infection during pregnancy with adverse events and neonatal outcomes has been a major concern. As more evidence and experience is accumulating, a weak connection to vertical transmission is established [1,2].

However, the crucial question of whether maternal infection affects neonatal outcomes and to what extent remains still under investigation.

A large multinational cohort study that took place at the beginning of the pandemic connected maternal infection at any stage of pregnancy with prematurity and neonatal morbidity when comparing Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Real-Time Reverse-Transcriptase Polymerase Chain Reaction (RT-PCR) positive and negative women [3]. Preterm birth occurred in 29.7% and 16% of data obtained from case series and observational studies, respectively [4]. Differences are possibly due to the diversity of participants, with a high incidence of adverse events seen especially during the first wave of the pandemic. During this period, a meta-analysis of mainly Chinese studies including 236 pregnant women showed a preterm birth percentage of 23% [5]. At the beginning of the pandemic only targeted screening based on symptoms or contact history was done. Universal screening of all parturients has been gradually introduced worldwide. In terms of maternal factors that favored worse neonatal outcomes, the INTERCOVID study showed that severe maternal COVID-19 with fever and respiratory distress was a risk factor for worse neonatal outcomes [3]. However, the rate of neonatal transmission does not seem to differ between women with and without severe disease as concluded in a recent meta-analysis [4]. It is well known that the main route of transmission is horizontal via respiratory droplets. In cases of perinatal transmission, neonatal symptoms, such as mild respiratory distress and transient thrombocytopenia, seem mainly related to late prematurity or elective C-sections due to severe maternal conditions, rather than to the neonatal infection itself.

Despite the gradually accumulating body of evidence, the risk for babies born to women with perinatal SARS-CoV-2 infection is still a matter of concern. We aim to add our experience and present the neonatal outcome of neonates born to mothers with perinatal SARS-CoV-2 infection. To our knowledge, this is the first study concerning SARS-CoV-2 perinatal transmission in Greece.

Materials and methods

Study design and sample

This is a prospective case-control study with data from two Academic Tertiary Referral Hospitals in Greece (Attikon University Hospital in Athens and University General Hospital of Patras) with maternity wards and level 3 Neonatal Intensive Care Units (NICU), accounting for approximately 2100 births per year. Electronic records of all births from RT-PCR-positive mothers between March 2020 and April 2021 were analyzed.

Demographic data concerning maternal age, nationality, comorbidities, as well as information regarding SARS-CoV-2 infection such as symptomatic or asymptomatic infection, the time between diagnosis and delivery and maternal management were recorded.

For neonates, gestational age (GA), mode of delivery, birth weight (BW), need for resuscitation or supplemental oxygen, and management during hospitalization was recorded. In addition, a ratio 2:1 cases-control matched neonates according to sex, GA and BW born to SARS-CoV-2 negative mothers during the study period were recorded.

Power analysis and sample size calculations

Sample size calculation revealed a statistical power of 80% for detecting observed effect sizes (Cohen’s d) of at least 0.2 for the comparisons between study groups, for two-sided hypothesis tests at significance level a = 0.05 (STATA version 17 command: power two-means, m1 m2, n(238) n/(1,9)n!).

Diagnosis

SARS-CoV-2 RT-PCR in nasopharyngeal swabs was performed on all parturients on admission as per local guidelines and on neonates within 24 h of life and again on the 3rd, 7th and 14th day of life (DOL).

When possible, amniotic fluid, umbilical cord blood, and breast milk as well as neonatal fecal swabs were also tested for SARS-CoV-2 RNA.

For RNA Extraction and SARS-CoV-2 Detection QIAsymphony DSP virus/pathogen mini kit (QIAGEN) and One-step Reverse Transcription Real-Time PCR Viasure SARS-CoV-2 (CerTestBiotec) were used.

Definitions

Neonatal test positivity was considered as any positive molecular test after delivery from birth to 28th DOL. Vertical transmission was assumed on the ground of positive placental/amniotic fluid SARS-CoV-2 RT-PCR testing. Possible vertical transmission was also assumed on the ground of positive molecular testing of the neonate despite immediate strict isolation [6]. Perinatal transmission was defined as when maternal infection was diagnosed between 14 days before and up to 72 h after delivery.

Neonatal care

Neonates from SARS-CoV-2 positive women were born in isolated delivery rooms, with all personal protection measures, transferred to NICU and kept there in a separate room in incubators with 2 m distance apart until two negative consecutive RT-PCR tests within 48 h. Positive neonates were treated by dedicated personnel with all personal protective equipment (PPE) as defined by the hospital’s protocol and moved to transitional care when two consecutive molecular tests were negative for SARS-CoV-2. Unfortunately, rooming in was not possible since all positive post-partum women were hospitalized along with other patients in COVID-19 wards.

Statistical analysis

Descriptive results of pregnancy, delivery and neonatal features are presented using median and interquartile range (IQR) for quantitative variables (due to the skewed distributions) and percentages for categorical ones. Statistical comparisons between neonates born to positive mothers and controls (neonates born to COVID-19 negative ones) regarding the need for supplemental oxygen, use of antibiotics, duration of hospitalization and mode of delivery, as well as somatometric characteristics, APGAR score and clinical symptoms, were performed by applying Chi-square and Mann–Whitney U test, as appropriate. Furthermore, Cohen’s d-effect size measure was calculated for all the statistical tests performed. All statistical analyses were carried out using Stata 17 statistical software (StataCorp. 2017. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC).

Results

A total of 1836 deliveries were recorded between April 2020 and May 2021. Among them, 79 pregnant women were SARS-CoV-2 PCR positive at delivery and gave birth to 81 neonates (two sets of twins). The median maternal age of the RT-PCR-positive women was 30.5 years (IQR: 24–34 years); most of them (52/79, 65.8%) were of Greek origin. Thirty-six per cent had comorbidities, most common including gestational diabetes and thyroid dysfunction (36% and 31% respectively). Forty-three percent (34/79) of pregnant mothers were asymptomatic. Maternal treatment for severe COVID-19 during pregnancy was given to 15.9% of women.

In terms of the neonates born to SARS-CoV-2 infected mothers, median GA and median BW were 38 weeks (IQR: 36–39 weeks) and 2940 gr (IQR: 2560-3340 gr) respectively, while 24.7% of them were premature. Need for supplemental oxygen was recorded in 33% (27/81) of neonates and it was mainly connected to CS and prematurity.

In total 376 molecular tests for the detection of SARS-CoV-2 were performed in various neonatal specimens with an overall positivity of 2.9%. Only 2 (2.4%) neonates were RT-PCR positive after delivery. The first one was a full-term neonate born to a pregnant woman with mild symptoms, who had a positive nasopharyngeal and rectal swab after delivery with CS, both negative the 3rd DOL. The neonate needed supplemental oxygen probably due to the CS and had diarrhea for one day. It might concern a late perinatal infection since AF was RT-PCR negative for SARS-CoV-2. The second one, full term as well, born with CS to an asymptomatic mother, did not show any respiratory but only gastrointestinal symptoms. Infection was due to horizontal transmission. During a 6-month follow-up both infants had a normal physical examination, growth and neurodevelopment.

Thirty-one RT-PCR negative full-term neonates of median gestational age 39 weeks (IQR: 38–40) and birth weight 3055 grams (range: 2425 − 4200 gr) presented gastrointestinal symptoms (vomiting and predominantly diarrhea) that lasted 1 to 11 days, or feeding difficulties necessitating the use of nasogastric tube from 1 to 8 days. The median duration of hospitalization of neonates born to positive mothers was 10 days (range: 3-28), while the vast majority of them did not receive any antibiotics while hospitalized. Comparisons between COVID-19 positive and negative pregnancy and neonatal characteristics are presented in Table 1.

Table 1. Comparisons between neonates’ characteristics born to COVID-19 positive (cases) and negative (controls) mothers.

		Cases (n = 81)	Controls (n = 157)	Cohen’s d	p-Value
Type of delivery (%)	Natural birth	20.5	34.8	0.28	.024^a
	Cesarean section	79.5	65.2
Need for resuscitation (%)	Yes	36.1	23.4	0.24	.059^a
Need for supplemental Oxygen (%)	Yes	47.5	32.9	0.25	.049^a
Antibiotic use (%)	Yes	26.8	27.2	0.03	.950^a
Eating disorders (%)	Yes	29.8	15.7	0.29	.022^a
Gastrointestinal	Yes	53.6	5.1	0.49	<.001^a
Disorders (diarrhea) (%)
Head circumference	Median	34	34	0.07	.658^b
(cm, at birth)	IQR^2	33–35	33–35
Length	Median	49.0	50.0	0.32	.032^b
(cm, at birth)	IQR	46.4–51	48–52
APGAR score	Median	9	9	0.22	.444^b
(at 5 min)	IQR	9–10	9–10
Days of	Median	10	4	0.81	<.001^b
hospitalization	IQR	6.3 − 14.8	3–8
Days of	Median	7	7	0.08	.747^b
antibiotic use	IQR	2–10	4–9

Statistical tests used: ^aPearson Chi-Square, ^bMann–Whitney U test. IQR: interquartile range.

Discussion

In this prospective case-control study, we aimed to assess neonatal outcomes during the perinatal period in children born to women with perinatal COVID-19 infection. Our findings suggest that although SARS-CoV-2 infection was rarely transmitted, there is identifiable associated morbidity. More specifically we found that newborns of COVID-19-infected women were more likely to have gastrointestinal symptoms compared to the control group.

Additionally, we found a low rate of neonatal positivity even though in the vast majority of cases maternal infection occurred during the late second and third trimesters of pregnancy. Only two neonates were RT-PCR positive after delivery. For the first one vertical transmission is questioned since repeated neonatal testing was negative as well as RT-PCR in amniotic fluid. Unfortunately, serology was not available at the time. This is in agreement with the current literature. In a large multicenter US study, the risk of positive neonatal test born to positive mothers was calculated at 2.2%, a much lower rate compared to initial reports with neonatal test positivity above 10% [3,7,8]. This mainly represents perinatal transmission as most of the studies included participants with late trimester infection. However, one may also consider intrapartum infection in cases of natural delivery via infected maternal secretions or feces. Vertical transmission is rather rare as depicted in most studies, yet there are some anecdotal cases with possible vertical transmission based on repeated neonatal RT-PCR swabs despite strict isolation or presence of IgM antibodies [1,2]. As most of the studies include pregnant women with late second or third-trimester pregnancy as in our case, there is still a major concern about the consequences of the infection in early pregnancy. A recent study by Morhart et al. included a case of severe eye malformation in a neonate with intrauterine exposure to SARS-CoV-2 infection prior to gestational week 12 and with no other identifiable cause of this severe embryopathy [9].

Placental histology does not usually reveal specific findings for COVID-19 disease in cases of RT-PCR negative neonates born to SARS-CoV-2 infected mothers. Fetal inflammatory response is lacking, while only mild maternal vascular and/or inflammatory lesions such as malperfusion, intervillous maternal thrombosis, fibrinoid deposition and mild intervillositis and lymphoplasmacytic deciduitis have been described. The above findings are presented in a previous paper published by our team, in which the first 18 RT-PCR negative placentas from our cohort were histologically examined [10].

Τhe rate of preterm labor in healthy SARS-CoV-2 uninfected pregnant women has been estimated to be approximately 11% [11]. In our cohort, higher rates of prematurity are in agreement with rates so far in COVID-19 pregnancies, yet they do not significantly exceed prematurity rates in non-COVID-19 women in our centers [12]. This finding maybe due to the fact that both hospitals are referral centers for high-risk pregnancies. The rate of CS in COVID-19 pregnancies from our cohort, exceeds this of 66% in a meta-analysis of 14 studies [5] and is significantly higher compared to the already existing increased rates of CS in non-COVID pregnancies during the same period, rendering COVID-19 pregnancy a risk factor for CS in our study. This is explained by the common practice of obstetricians that unfortunately did not change with time, to perform COVID-19 positive deliveries with SC in order to limit exposure to a contagious environment, even though The American College of Obstetricians and Gynecologists recommendations state that SARS-CoV-2 infection alone is not an indication for preterm delivery or CS [13]. Cesarean birth, apart from the clinical impact may have on mothers and neonates, is also connected to reported maternal higher anxiety and trauma scores and lower well-being compared to vaginal birth [14]. Unaccompanied delivery practice and isolation of parturients as per protocol are additional factors with adverse psychological impact on women giving birth during the pandemic. Studies have shown increased depressive symptoms and generalized anxiety among pregnant and breastfeeding women during the COVID-19 outbreak [15,16].

The vast majority of our neonates did not show any respiratory symptoms, apart from cases related to prematurity or elective CS. Apgar score in min 5 was comparable to that in a systematic review and meta-analysis by Amirian et al. suggesting excellent postnatal condition in the majority of cases [17]. However, in our study more than half of the full-term neonates presented gastrointestinal symptoms despite lack of perinatal infection by repeated RT-PCR negative nasopharyngeal and fecal swabs. This is a finding described for the first time in RT-PCR negative neonates born to infected mothers and it is more common than in neonates born to SARS-CoV-2 negative mothers. A possible explanation for that could be that placental inflammation may mediate the transfer of inflammatory cytokines to the fetal circulation causing gastrointestinal symptoms to the neonate. Of note fecal or rectal swabs seem to be the specimen with the highest SARS-CoV-2 positivity even compared to NP swabs in neonates born to COVID-19 infected mothers, as shown in a recent systematic review and meta-analysis by Kotlyar et al. [2]. Neonates born to COVID19 positive mothers were kept more days within hospital compared to controls, and this was partly attributed to the initial instructions to discharge mothers and their offsprings only when PCR negative.

Limitations

Lack of serology, not available at the beginning of the study did not help us to differentiate between vertical and horizontal transmission in one of our neonates. In addition, the study concerns only two referral centers (not all Greek NICUs and maternity wards), that is why we cannot calculate national incidence of neonatal positivity.

Conclusions

In this case control study neonatal positivity was limited and no vertical transmission was noted. Neonatal outcomes were comparable to the control group. However, the presence of gastrointestinal symptoms in neonates born to RT-PCR positive women compared to controls needs further investigation.

Acknowledgments

The authors thank Dr Anastasia Konstantinidou, 1st Department of Pathology, Unit of Perinatal Pathology, National and Kapodistrian University of Athens, Greece, for her work on placental histopathology in a sub-group of our cohort and Dr George Antonogeorgos, School of Health Science & Education, Harokopio University, who provided statistical assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s)..

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.