Effects of COVID-19 Infection during Pregnancy and Neonatal Prognosis: What Is the Evidence?

Background: This study’s aims are to assess the current evidence presented in the literature regarding the potential risks of COVID-19 infection among pregnant women and consequent fetal transmission. Methods: a systematic literature review assessing papers published in the most comprehensive databases in the field of health intended to answer the question, “What are the effects of COVID-19 infection during pregnancy, and what is the neonatal prognosis?” Results: 49 papers published in 2020 were eligible, presenting low levels of evidence. A total of 755 pregnant women and 598 infants were assessed; more than half of pregnant women had C-sections (379/65%). Only 493 (82%) infants were tested for SARS-CoV-2, nine (2%) of whom tested positive. There is, however, no evidence of vertical transmission based on what has been assessed so far, considering there are knowledge gaps concerning the care provided during and after delivery, as well as a lack of suitable biological samples for testing SARS-CoV-2. Conclusions: We cannot rule out potential worsening of the clinical conditions of pregnant women infected with SARS-CoV-2, whether the infection is associated with comorbidities or not, due to the occurrence of respiratory disorders, cardiac rhythm disturbances, and acid-base imbalance, among others. We recommend relentless monitoring of all pregnant women in addition to testing them before delivery or the first contact with newborns.


Introduction
On 30 January 2020, the World Health Organization (WHO) declared the outbreak of COVID-19, a respiratory disease caused by the new coronavirus SARS-CoV-2, as the sixth public health emergency of international concern [1,2] Due to its highly transmissible nature, by 9 April 2020, it had spread to five continents, and approximately 85,522 people had died [2].
Considering that transmission seems to mainly occur through contact with respiratory droplets [3] produced by an infected person, anticipating public health measures intended to control and prevent the infection, such as adherence to universal precautions, quarantine, and timely diagnosis, are options available to mitigate the transmission of COVID-19 [4].
Clinical manifestations range from asymptomatic cases and mild upper airway infection, up to severe and fatal cases with pneumonia and acute respiratory failure [5][6][7]. This variation is because A search was conducted in the following databases: US National Library of Medicine (PubMed), Scopus, Embase, ScienceDirect (Elsevier), Web of Science (WoS), Scholar Google, and preprints servers bioRxiv and medRxiv, as well as the bibliographic references of the selected papers (hand searching). These databases were selected due to their range and representativeness in the field of basic and health sciences. Terms that derived from the following expressions were used according to the databases/servers: "COVID-19" OR "SARS-CoV-2" AND "Pregnancy" AND "Perinatal". To avoid screening biases, two researchers with expertise in the method and topic under study independently and concomitantly searched all the databases on 25 and 26 May. The researchers had a discussion to reach a consensus about which papers would be included or excluded from the study, and a third reviewer mediated disagreements that prevented them from reaching a consensus.
Observational epidemiological studies and case reports addressing the clinical conditions of mother-fetus pairs and including primary data of patients over 18 years old were considered eligible. Manuscripts that contained only data from pregnant women, or only fetuses, or that did not address the period of delivery, such as puerperium, were disregarded. No restrictions regarding the period of publication or language were imposed. Review papers, opinions reports, local reports, abstracts of events, and similar works were excluded. Social, demographic, and clinical data included in the studies were collected.
The GRADE system was used to classify levels of evidence. The results are presented in terms of prevalence calculated in the study and combined results.
In Table 2, we present the characteristics of pregnant women and newborns, with social, demographic, and clinical data being collected.

Findings from Longitudinal Studies
The case-control study (2%) [60] was conducted using the medical records of pregnant women admitted to a hospital in China. The study compares the clinical features, maternal and neonatal outcomes of 16 pregnant women with COVID-19, and 18 without the disease, but suspected of being infected. The study does not clearly report criteria for including participants in the case or control groups. The description in Table 2 refers to COVID-19 infected pregnant women. All 16 women were in their third trimester of pregnancy, 5 (31%) of whom had no pregnancy-related comorbidities. The most frequent symptom was fever at admission (25%) and after birth (50%). Ten women had suggestive CT scans, and there was an increase in reactive C protein and neutrophilia in all women studied. Fourteen (87%) women had a C-section, but the study does not report why a C-section was indicated. A total of 17 babies were born, with one set of twins, with no complications. Only three newborns were tested for SARS-CoV-2 and were negative. Isolation measures after birth or analysis of the placenta were not reported.
The cohort study (2%) [61], also conducted in China, retrospectively describes 31 pregnant women and 35 non-pregnant women with COVID-19. Only the clinical findings of pregnant women are described. In total, 31 pregnant women were assessed. The majority was in the third trimester of pregnancy (71%), all had confirmation of diagnosis through RT-PCR, 21 (68%) developed mild to moderate pneumonia, and 10 had severe pneumonia. Twenty-eight did not present comorbidities during pregnancy. The most prevalent signs and symptoms were fever (55%) and cough (48%). The chest CT scan was suggestive in all cases, and abnormal laboratory tests were related to the increased number of neutrophils (32%), aspartate transaminase (26%), and interleukin 6. There were 17 deliveries, 13 (76%) of which were C-sections, though the authors do not report why C-sections were indicated. A total of 17 single, healthy fetuses were born and tested negative for SARS-CoV-2. No report is provided on the isolation measures adopted between mothers and babies after delivery and there was no evaluation of placentas.

Discussion
This review was intended to answer a question concerning the effects of COVID-19 infection during pregnancy and neonatal prognosis. Forty-nine studies were eligible and included case reports, cross-sectional, analytical cross-section, case-control, and cohort presenting low levels of evidence. The low levels of evidence are due to the novelty of the COVID-19 pandemic and the need for rapidly acquiring knowledge to support public policies. As the number of cases increases worldwide, evidence about the impact of this virus during pregnancy for both women and newborns is expected to become stronger, especially with the development of more robust comparative studies and follow-up with control groups.
The fact that most pregnant women were from China imposes limitations on interpreting the evidence, considering cultural and epidemiological differences when compared with pregnant women from other countries and cultures. However, even in the minority (120/16%), non-Chinese pregnant women were evaluated, and the characteristics (clinical and epidemiological) showed no differences.
Regarding the pregnant women's age, fertile period, and length of pregnancy, the studies analyzed showed a wide variation and a lack of evidence of infection by SARS-CoV-2 during the first and second trimester of pregnancy. It can be inferred that, according to the low prevalence of severe infection among pregnant women (57/8%), many of them could be asymptomatic and/or with mild symptoms, without the need for hospital care, corroborating findings from previous viral pandemics [62].
In view of the limited data regarding the Middle East Respiratory Syndrome (MERS), a systematic review with meta-analysis recovered seven studies that did not report spontaneous abortion. The rate of premature birth was 32.1% (3 of 11), all occurring before 34 weeks of gestation. Preeclampsia was described in 19. 1% (1 of 7); however, no cases of premature rupture of membranes or restricted fetal growth were reported. The rates of C-sections and perinatal death were 61.8% (5 of 8) and 33.2% (3 of 10), respectively, including two stillbirths and one neonatal death (4 h after the birth of an extremely premature baby). There were no reports of fetal distress, Apgar score <7 at 5 min, neonatal asphyxia, or admission to the neonatal intensive care unit (ICU). Finally, signs of vertical transmission were not found during the follow-up period in any of the newborns [63].
According to the above, it appears that the scarce data on infection with the new coronavirus in early pregnancy may be related, but not limited, to the absence of tests performed during this period, since asymptomatic cases may go unnoticed due to poorer surveillance of pregnancy because of restrictions, on the part of medical staff or the pregnant women themselves, to attend appointments within a hospital setting. Thus, we suggest that tests for COVID-19 should be routinely performed in prenatal care.

Clinical Findings in Pregnant Women
About diagnosis, this study verifies that of the 728 pregnant women evaluated, 589 (81%) and 139 (19%) were diagnosed by Reverse Transcription followed by Polymerase Chain Reaction (RT-PCR) and by clinical assessment, respectively. The gold standard for diagnosing Covid-19 is tissue culture in which the antigen is isolated, using Polymerase Chain Reaction (PCR), which detects nucleic acid. Even so, a single result not detected through RT-PCR for SARS-CoV-2 does not exclude a COVID-19 diagnosis, as there are various factors, such as inadequate sample collection, type of biological sample, the time elapsed between sample collection and onset of symptoms, and fluctuation of viral load, that may influence a test's result. For this reason, an RT-PCR test should be repeated in another sample of a patient's respiratory tract whenever there are discordances between results and epidemiological conditions, especially in populations where a false-positive may result in disastrous consequences.
The presence of comorbidities related to pregnancy does not seem to directly influence the adverse outcomes of pregnant women and their newborns, as the two neonatal deaths were of mothers without comorbidities, but who, for some reason, developed severe pneumonia. However, it is observed that gestational diabetes and fetal distress were the most prevalent comorbidities, showing that the conditions of the fetus should be carefully evaluated, especially in those asymptomatic and without comorbidities. Thus, the absence of comorbidities may directly influence the care provided and attention paid by professionals to pregnant women, exposing some of them to a worse prognosis [64].
Concerning the signs and symptoms of infection at the time of admission, 689 (91%) pregnant women were evaluated , and the main signs and symptoms presented were fever at admission (363/53%), cough (290/42%), and dyspnea (83/12%). It is remarkable that 83 (12%) of pregnant women were asymptomatic but were tested (through RT-PCR to detect SARS-CoV-2) due to exposure to people diagnosed with COVID-19, reinforcing the need to follow recommendations provided by the Centers for Disease Control and Prevention (CDCs) to testing risk groups in contact with those diagnosed with COVID-19, [65] though this approach may not be feasible in some contexts where there is a shortage of tests.

Childbirth
Overall, 587 childbirths were reported, with more than half of the pregnant women having C-sections (379/65%). When checking the indications for the C-sections, pregnancy-related comorbidities were the cause for performing C-sections in 148 pregnant women; in the case of 103, the indication was infection, and in 128 cases, indications were not reported. No details were provided when the indication for a C-section was the infection. In most cases, the time of delivery was determined by obstetric reasons, rather than the maternal diagnosis of COVID-19 [68].
The decision for the type of delivery is usually more frequently influenced by the presence of maternal and/or fetal impairment. When there were imminent risks, an emergency C-section was the alternative chosen, which has happened in the case of SARS-CoV-2 infections in which the pregnant woman's clinical condition is complex. However, in the presence of COVID-19, the threshold for C-sections became lower than usual so that infection control procedures could be more easily adhered to and the transmission of the disease to the fetus is minimized [69].
Note that most of the newborns did not present serious complications. The unfavorable outcomes refer to ten neonatal deaths, one spontaneous abortion, eight maternal deaths, one stillbirth, nine (2%) positive SARS-CoV-2 tests, and three newborns with high rates of IgG and IgM antibodies against SARS-CoV-2.
Regarding the nine newborns who tested positive for SARS-CoV-2, three tested positive [33,48] immediately after delivery, but then tested negative 24 h after delivery. Another newborn tested positive 36 h after delivery [36]. It is important to note that whether the mothers and babies in these three cases were isolated after delivery is not reported, making it difficult to establish the transmission routes, according to the authors, in addition to the absence of intrauterine tissue and amniotic fluid samples.
Two newborns tested positive for the new coronavirus after being breastfed by their mothers without wearing masks, as the maternal infection was not known in the postpartum period. However, vertical transmission could not be confirmed or ruled out, as the authors stressed that the newborns were not tested for COVID-19 immediately after birth. Moreover, in this same study, a pregnant woman with COVID-19 gave birth to a newborn by vaginal delivery, who tested positive for SARS-CoV-2 despite the mother wearing a surgical mask and the medical team wearing proper PPE throughout labor [45]. In another study [70], a possible transmission through breast milk was raised by the detection of viral RNA in the milk of one of the pregnant women and subsequent confirmation of contamination of the baby, although the mother followed safety precautions when breastfeeding the child. This reinforces the need for testing mothers before giving birth, using proper vestments (staff and mother), and avoiding breastfeeding when testing is not possible.
The other two newborns tested positive 16 h [30] and 53 h [46] after delivery, respectively. Both studies reported the followed guidelines were observed: the use of masks by the mother, the entire medical team was attired, mother and baby were isolated after delivery, and no breastfeeding. The authors, however, reported an important limitation of the study, the absence of complementary assessments (presence of viruses in amniotic fluid, umbilical cord blood, or placental tissue). Two out of the nine positive babies were intubated [30,45], and three had mild pneumonia [36,46,47], although they fully recovered within a few days.
Dong et al. [28] reported the case of a newborn who, shortly after birth, presented leukocytosis and a high rate of IgG and IgM antibodies against SARS-CoV-2 and IL-6 cytokines. The baby, however, presented no symptoms and tested negative for the virus. Although the newborn tested negative for SARS-CoV-2, the authors stated that the high rate of IgM antibodies within 2 h after birth suggested the occurrence of intrauterine infection, as there would be no transfer of these antibodies from the mother to the fetus through the placenta due to the size of this macromolecule. Additionally, in general, these take three to seven days to be produced by the body after contact with the infectious agent. The authors, however, do not rule out the possibility of infection during delivery.
Zeng et al. [35] described similar results regarding the assessment of the presence of specific antibodies against SARS-CoV-2 in the blood of newborns from mothers with confirmed COVID-19 infection. Two babies presented rates of IgG and IgM antibodies specific to the virus above the normal level, but none showed symptoms of the infection. The authors of this study stressed the possibility that the newborn developed IgM antibodies during the gestational period if the virus had crossed the placental barrier.

Limitations
The studies recovered in this review have several limitations. Among them, small sample sizes, retrospective evaluation of medical records with incomplete data, as well as the low number of babies tested for SARS-COV-2, compared to the number of births.
Among this review's limitations, we highlight the possibility that the same patients are reported more than once, as they may have been included in different studies and we addressed secondary data only. Most of the cases reported refer to Chinese pregnant women and there was a lack of a standard assessment in newborns to verify the presence of viruses in amniotic fluid, umbilical cord blood or placental tissue. There was also little information about the care provided during delivery and postpartum, and an absence of clear inclusion criteria for the control group. Nonetheless, despite these limitations, this review has strengths, such as presenting a comparison of all studies available so far, compiling scattered data and grouping them more clearly in order to make analyses and inferences, and, based on what was assessed, reporting that there is no evidence of vertical transmission thus far, as there are knowledge gaps concerning the care provided during and after delivery, and a lack of suitable biological samples for testing SARS-CoV-2.

Conclusions
The results show that the potential worsening of the women infected with SARS-CoV-2 cannot be ruled out, whether the infection is associated with comorbidities or not. As documented here, there is a risk of women developing respiratory disorders or having cardiac rhythm disturbances or acid-base imbalance, among others. The risk of postpartum hemorrhage and premature delivery is significant, which is why C-sections were widely used. Relentless monitoring is recommended for pregnant women who report signs and symptoms suggested of COVID-19, especially at a time when women in countries with radical isolation measures have difficulty attending regular prenatal care. Those who cannot be tested before delivery should avoid having contact with the baby and wear masks to decrease the spread of the disease.
Regarding the risk for babies, there is no concrete evidence of vertical transmission, though one cannot discard this possibility. Nonetheless, cases of respiratory diseases, abnormal Apgar indexes, and mild pneumonia were reported. Fortunately, all the babies who tested positive for the infection recovered fully and rapidly. A rigorous assessment of the newborns' clinical signs is recommended, as well as chest CT scans, within three days after birth.

Conflicts of Interest:
The authors no reported conflict of interest.