Protocol and registration

The protocol of this systematic review and meta-analysis was registered on the PROSPERO database (CRD42020191106; registered 11 June 2020). The review and its reporting were followed up on PRISMA (Preferred Reporting items for Systematic Review and Meta-Analysis)¹⁶. To keep the risk of bias under control during the study, the scanning, selection of articles, data extraction as well as the quality assessment of the articles included were handled independently by two researchers. Whenever a difference of opinion came up with regard to any aspect of the study, all of the researchers held a discussion session together and arrived at an agreement. At the same time, prior to the start of the study, a pilot study that included all the stages of the research was conducted with the participation of all of the authors, who concurred on a common road map.

Eligibility criteria

The research included those studies published in English between 1 December 2019 and 15 May 2020 and whose full texts could be accessed. The PICOS criteria were considered in the selection of the studies that would be suitable for this systematic review and meta-analysis (Table 1). The exclusion criteria encompassed traditional and systematic reviews and letters to the editor that did not refer to specific cases.

Searching strategy

The scanning for this systematic review and meta-analysis was carried out over the period 5–15 May 2020 independently by two of the researchers. Scanning for the international studies was carried out in the PubMed, Ovid, Science Direct, EBSCO and Web of Science, Google Scholar electronic databases with the search string: [‘COVID-19’ AND (‘pregnancy’ OR ‘pregnant’ OR ‘maternal outcomes’ OR ‘infant outcomes’ OR ‘fetal outcomes’ OR ‘birth’)]. At the same time, additional studies were independently checked by the other two authors against the included articles and the reference lists for the review studies.

Selection of studies

Through the processes of scanning, selection according to heading and/or abstract, and the elimination of repetitious articles. The four authors then met together to decide upon the studies that could be taken into the analysis on the basis of full texts. Later, some studies were eliminated from the scope of the analysis during the extraction or analysis processes because they did not contain data appropriate for the analysis. The selection process for the articles can be seen in Figure 1 with a PRISMA flow diagram.

Figure 1

PRISMA flow diagram of the search process

Data extraction

The researchers devised a data extraction instrument to be used in obtaining study data (Table 2). This tool made it possible to collect information on the studies included in the systematic review and meta-analysis in terms of author information, the country and year of publication, data collection dates, design, sample size, maternal age and other main findings reported in the studies (Table 1). In four studies of case-control design, only the data pertaining to the cases (pregnant women diagnosed with COVID-19) were extracted.

Methodological quality evaluation of the studies

The checklists drawn up by the Joanna Briggs Institute for cross-sectional studies, case studies, case series and case-control studies were used in the quality assessment of the articles¹⁷. The first and second of these tools comprise eight items, while the third and fourth comprise 10 items. Each item is assessed as Yes, No, Uncertain and Inapplicable. The results of the assessment were presented on the basis of the total number of items considered (number of Yes) as a ‘Quality Score’, given in Table 2.

Synthesis of the data

The data in this systematic review and meta-analysis were combined with a meta-analysis and a consolidated percentage calculation. The data obtained from the studies of retrospective and perspective cross-sectional, case series and case-control design were consolidated in the meta-analysis. The data from the case reports were consolidated by pooled analysis. The Comprehensive Meta-Analysis Version 3-Free Trial (https://www.meta-analysis.com/pages/demo.php) was used for the meta-analysis. The extent of heterogeneity in the studies was assessed with the Cochran Q and Higgins I² tests and it was agreed that the I² rate exceeding 50% was an important indication of heterogeneity. Accordingly, when I² was greater than 50%, the Random effect results, and if the value was less, the Fixed Effect results were considered. The study data were composed of categorical variables and combined point estimates were calculated at a confidence interval (CI) of 95% for each result variable. All of the tests were calculated on a two-tailed basis and p<0.05 was accepted as statistically significant.

Searching results

At the beginning, the results of the scan yielded 732 records from the databases and 33 from additional scans, a total of 765 records. With the exclusion of the records that were repeats, the review carried out according to headings and abstracts yielded the full texts of 60 articles. From the review of the full texts, a total of 54 articles reporting the outcomes of pregnant women with COVID-19 and their infants were selected for the analysis (Figure 1).

Study characteristics

All of the studies (n=54) had been published in English. Twenty-two of the studies were case presentations, 13 were case series, 15 were retrospective (13) and perspective (2) cross-sectional studies, and 4 were case-control. It was noted that the study data had been collected over the period 8 December 2019 – 22 April 2020 and all were published in 2020. In 10 of the studies, no indication had been given as to the date of data collection^18–27. The studies were from 15 different countries: China (34), United States (4), Iran (4), Italy (3), Peru (1), Columbia (1), Sweden (1), United Kingdom (1), Korea (1), Canada (1), Australia (1), Portugal (1), Turkey (1) and the Netherlands (1) (Table 2).

The studies contained data on 517 pregnant women with COVID-19 and 385 infants. The sample size of the studies varied between 1–118. It was seen that the age range of the pregnant women with COVID-19 included in the systematic review was generally between 21–45, that only one women was an adolescent (age 17 years)²⁸, and that seven studies did not specify the ages of the pregnant women^29–35.

Quality assessment results of the studies

The quality assessment scores in the case reports were Yes: 7/8 in 21 studies, and Yes: 6/8 in one study. The quality assessment scores in the case series were Yes: 10/10 in three studies and 9/10 in 11 studies. In all of the retrospective (13) and perspective (2) cross-sectional studies, the quality scores were Yes: 6/8, and in all of the case-control studies, the quality assessment scores were Yes: 7/10 (Table 2). A large majority of the studies reviewed met the criteria for quality assessment, representing a low risk of bias.

Results of the meta-analysis

COVID-19 symptoms in pregnant women

The pooled results of the studies reviewed in this systematic review and meta-analysis indicated that the symptoms of the pregnant women were observed to be high fever in 53% in 29 of the studies^{19,21,24,29-33,35,36,38-55} (Z=0.503, p=0.615) and cough in 40% in 28 studies^{19,21,24,29-31,33,35,36,38-43,45-56} (Z= -1.824, p=0.068). In fifteen studies, 21% of the pregnant women with COVID-19 had fatigue/tiredness/myalgia^{21,29,31,37-43,47,48,51,54,56} (Z= -9.034, p<0.001). Nineteen studies indicated that 21% of the pregnant women with COVID-19 suffered from dyspnea/shortness of breath^{19,21,29-31,38-44,46-48,51,54-56} (Z= -4.755, p<0.001) and 16% showed the symptom of sore throat in eight studies^{30,36-38,42,47,48,54} (Z= -4.613, p<0.001). In four studies, 16% of the pregnant women with COVID-19 had headache^21,29,39,40 (Z= -2.954, p=0.003), 15% had chest pain/tightness in seven studies^{29,37,38,40,45,47,54} (Z= -8.496, p<0.001), and 13% had phlegm in three studies^46,54,56 (Z=-3.028, p=0.002). In twelve studies, 8% of the pregnant women with COVID-19 complained of diarrhea^{29,33,36-39,41,46-49,55} (Z= -10.701, p<0.001), while 15% suffered from runny/congested nose in six studies^{21,42,46,52,56,57} (Z= -4.177, p<0.001). Three studies indicated that 12% of the pregnant women had chills/shivers^37,38,57 (Z= -2.366, p=0.018), one study reported that 7 of 9 pregnant women had anosmia⁴² (Z=1.562, p=0.118), and another reported that 6 of 9 pregnant women complained of lethargy⁴² (Z=0.980, p=0.327). Furthermore, according to the pooled results of four studies, about 35% of the cases had been observed to be asymptomatic, but it was seen that this result was not statistically significant^39,48,52,53 (Z= -0.568, p=0.570) (Table 3 and Supplementary file Figure 1).

Diagnostic test results for COVID-19 in pregnant women

According to the pooled results of 31 studies, 80% of the pregnant women tested positive on the RT-PCR test^{19,21,24,29,30,32-57} (Z=9.424, p<0.001), and 88% had abnormal chest CTs in 21 studies^{30,32-40,43,46-50,52-54,56,57} (Z=8.097, p<0.001). In three studies, 37% of the women had abnormal chest X-rays^24,42,44 (Z= -0.813, p=0.416). Moreover, in the pooled results of four studies^29,34,49,52, 38% of the pregnant women had been diagnosed on the basis of clinical findings (Z= -1.029, p=0.303) (Table 3 and Supplementary file Figure 2).

Treatment methods using for COVID-19 in pregnant women

Fifteen studies taken into the meta-analysis showed that 87% of the pregnant women were treated with antibiotics^{29-31,33,38,43-46,48,50,52,54,55} (Z=3.236, p=0.001), and 75% were given antiviral treatment in 14 studies^{19,21,30,31,38,43,45,46,48-50,54,55,57} (Z=2.866, p=0.004, I²=60%). In five studies, 54% of the pregnant women were treated with hydroxychloroquine^{19,21,29,43,44} (Z=0.141, p=0.888), while 31% were treated with corticosteroids in six studies^{31,36,38,50,54,57} (Z= -1.227, p=0.220). In five studies, 88% of the pregnant women were treated with a Chinese medication (lianhua-gingwen)^{45,46,48,55,57} (Z=4.085, p<0.001). In two studies, 11% of the pregnant women were given intravenous immunoglobulin^31,43 (Z= -3.985, p<0.001) and 8 of 9 pregnant women were given anticoagulant treatment in one study⁴³ (Z=1.961, p=0.050). Again, 63% of the pregnant women needed oxygen support in 9 studies^{29,33,38,41,44,45,50,54,55} (Z=0.682, p=0.495), and 11% were intubated/placed on mechanical ventilation in 4 studies^21,39,51,52 (Z= -1.450, p=0.147). According to the results of two studies, 77% of the pregnant women with COVID-19 had been treated and followed up as outpatients/by telephone^29,37 (Z=0.829, p=0.407) (Table 3 and Supplementary file Figure 3).

This meta-analysis showed that in 2 studies it was reported that in order to prevent preterm labor, 15% were treated with hydration^29,44 (Z= -1.177, p=0.239), and 6 of 7 pregnant women were given steroids in another two studies^21,44 (Z=1.123, p=0.262), and 1 of 2 pregnant women were treated with magnesium sulphate (MgSO₄) in one other study⁴⁴ (Table 3 and Supplementary file Figure 4).

Comorbidities of pregnant women with COVID-19

The results of the meta-analysis indicated that 19% of the pregnant women in 4 studies had asthma^21,24,29,42 (Z= -4.261, p<0.001), and it was reported in 3 studies that 18% of the women suffered from chronic hypertension^21,29,30 (Z= -1.765, p=0.078). It was reported in 5 studies that 10% of the pregnant women had hypothyroidism^{31,43,50,53,55} (Z= -4.962, p<0.001), and 9% of the pregnant women had diabetes mellitus in three studies^21,29,43 (Z= -4.813, p<0.001). It was found in 4 studies that 34% were obese^21,24,43,44 (Z= -1.280, p=0.201), 34% had heart disease in two studies^44,48 (Z= -0.317, p=0.751). It was reported in three studies that 9% of the pregnant women had Hepatitis B^30,31,53 (Z= -5.347, p<0.001) and 17% of the women had polycystic ovary syndrome in three studies^30,44,55 (Z= -2.803, p=0.005). In one study, 7% of the pregnant women were found to have thalassemia⁴⁸ (Z= -2.550, p=0.011), and 1 of 4 of the pregnant women had cholecystitis in another study⁴⁰ (Z= -0.951, p=0.341). It was noted in one study²¹ that 1 of 5 pregnant women had chronic kidney disease (Z= -1.240, p=0.2015), and 1 of 2 pregnant women had familial neutropenia in another study²⁴ (Table 3 and Supplementary file Figure 5).

Additional conditions related to pregnancy

The pooled results of 11 studies in the meta-analysis showed that about 15% of the pregnant women had gestational diabetes mellitus^{24,29-31,37,41-44,48,56} (Z= -7.792, p<0.001), 6 studies indicated that 9% had hypertension^{29,31,38,42,44,53} (Z= -5.922, p<0.001), and 7 studies reported that 20% had preeclampsia^{30,33,37,38,44,52,56} (Z= -3.749, p<0.001). It was reported in one study that 2% of the pregnant women had cholestasis²⁹ (Z= -3.694, p<0.001) and 16% of the pregnant women had anemia in another 2 studies^37,40 (Z= -2.181, p=0.029). Another study revealed that 10% of the pregnant women had hypothyroidism³⁷ (Z= -2.084, p=0.037) (Table 3 and Supplementary file Figure 6).

In another 2 studies, 8% of the pregnant women with COVID-19 exhibited reduced fetal movement^29,40 (Z= -1.879, p=0.060), 2% of the pregnant women were in multiple pregnancy in another study³⁹ (Z= -5.693, p<0.001), and 17% had early membrane rupture in 8 studies^{30,35-38,49,51,53} (Z= -6.260, p<0.001). It was reported that 11% of the pregnant women in 4 studies had placental abruption^30,36,37,54 (Z= -3.980, p<0.001), 14% of the pregnant women in four studies had placenta previa^36,40,48,54 (Z= -3.400, p=0.01), and 21% of the pregnant women in 2 studies had polyhydramnios/oligohydramnios^36,54 (Z= -1.992, p=0.046) (Table 3 and Supplementary file Figure 6).

The pooled results in this meta-analysis showed that 8 studies indicated that 12% of the pregnant women diagnosed with COVID-19 were admitted into the intensive care unit^{21,29,30,35,41,44,51,52} (Z= -3.039, p=0.002), and 12% of the pregnant women in 2 studies were recorded as maternal mortalities^39,43 (Z= -0.597, p=0.550). In one other study, it was noted that one of the two pregnant women had postpartum hemorrhage²⁴. While the postpartum hemorrhage and maternal death results pertaining to the pregnant women with COVID-19 were not statistically significant, data related to admittance into the intensive care unit was statistically significant (Table 3 and Supplementary file Figures 7 and 8).

Results on pregnancy term and mode of delivery in pregnant women

Thirty studies examined in this systematic review and meta-analysis provided results on the delivery mode of pregnant women with COVID-19^{19,21,24,29-44,46,48-57}. The pooled results of these studies were statistically significant in that 73% of the pregnant women with COVID-19 had delivered by caesarean section (Z=7.979, p<0.001). According to the results of 5 studies, 29% of the women undergoing caesarean section had a COVID-19 indication^{35,39,41,42,54} (Z= -1.872, p=0.061), and 43% of the pregnant women in 8 studies had undergone caesarean section due to obstetric indications^{29,33,37,39,41,42,52,54} (Z= -0.837, p=0.403). These results, however, were not statistically significant. It was found that 25% of the pregnant women in 17 studies went into preterm labor and this result was statistical significance^{21,29-33,35-39,42,45-47,51,54} (Z= -4.304, p<0.001) (Table 3 and Supplementary file Figure 9).

Perinatal results of newborns

In 15 studies examined in this systematic review and meta-analysis, 25% of the newborns of women with COVID-19 were born with low birth weight (<2500 g)^{19-21,24,30,31,35-38,42,43,46,54,55} (Z= -4.772, p=0.001). It was found that about 11% of the newborns in 26 studies had an APGAR score of <7 in the first minute^{19-21,24,29-31,33-40,42,43,45,46,48,50,51,53-55,57} (Z= -8.527, p<0.001). It was determined that 9% of the newborns had an APGAR score of <7 in the fifth minute in 27 studies^{19-21,24,29-31,33-38,40-43,45,46,48,50,51,53,54,55-57} (Z= -9.490, p<0.001). It was revealed that 23% of the newborns suffered fetal distress in 13 studies^{19,24,30,35-38,44,48-52} (Z= -4.591, p<0.001), and 4% of the newborns had fetal asphyxia in 10 studies^{31,35-39,47,48,53,55} (Z= -8.078, p<0.001) (Table 4 and Supplementary file Figure 10).

In 11 of the studies in the meta-analysis, 28% of the newborns had been admitted into the neonatal intensive care unit (NICU) and this result was statistically insignificant^{20,29,31,33-35,40-42,45,49,50} (Z= -4.993, p=0.066). According to the pooled results of 12 of the studies examined, the neonatal death was observed in 6% of newborns and this result was statistical significance^{31,35-39,43,45,46,48,51,54} (Z= -7.411, p<0.001). It was found that the intrauterine death rate was 8% in a meta-analysis based on the results of 5 studies^{31,35,38,43,54} (Z= -4.564, p<0.001) (Table 4 and Supplementary file Figures 10–12).

Data on neonatal nutrition and care

In 9 studies examined in this meta-analysis, 36% of the newborns were breastfed by their mothers, who wore masks and strictly complied with hand-washing rules^{19,20,24,29,41,42,50,56,57} (Z= -2.911, p=0.338), and all of the eight newborns were fed formula in three studies^19,20,40 (Z= 2.057, p=0.040). These results were statistically significant. Again, statistical significance was noted in 14 studies regarding the data that 88% of the newborns were isolated from their mothers^{19,30,32-34,40,42,44,45,47,49,50,54,57} (Z= 5.191, p<0.001). In 3 studies, 77% of the newborns were isolated together with their mothers, but this finding was not found to be statistically significant^20,29,41 (Z=0.779, p=0.436) (Table 4 and Supplementary file Figure 13).

COVID-19 test outcomes of newborns

The consolidated results of 26 studies in this meta-analysis showed that 8% of newborns of women with COVID-19 the newborns tested positive in the RT-PCT test in the first 24 hours after birth^{19-21,29,31-39,41-43,45-47,49-51,53,54,56,57} (Z= -10.581, p<0.001), in 2 studies, 2 of 8 newborns tested positive for IgG^19,34 (Z= -1.206, p=0.228), and 3 of 8 newborns tested positive for IgM in two studies^19,34 (Z= -0.683, p=0.495) (Table 4 and Supplementary file Figure 14).

The consolidated results of eight studies in this meta-analysis revealed that 10% of the newborns tested positive on the RT-PCR test in the 48th hour and on day 14 postpartum^{19,20,30,31,33,37,40,47} (Z= -4.285, p<0.001). It was also seen in another 5 studies^{36,40,47,49,57} that 31% recorded abnormal results on the chest X-ray/CT (Z= -1.199, p=0.231) (Table 4 and Supplementary file Figure 15).

COVID-19 test outcomes in placenta, umbilical blood, breast milk, vaginal swab and amniotic fluid samples

In 8 studies examined, one of the placenta and the umbilical blood samples taken from pregnant women with COVID-19 in the first 24 hours postpartum was statistically significantly positive^{19,33,38,45,46,50,56,57} (Z= -3.916, p<0.001). On the other hand, samples of breast milk in 6 studies^{19,38,39,49,50,57} (Z= -3.691, p<0.001), of vaginal swab in 2 studies^50,57 (Z= -1.647, p=0.100), and of amniotic fluid in 7 studies^{19,33,38,49,50,56,57} (Z= -4.158, p<0.001) did not test positive. In days 2–14 postpartum, 2 of 7 samples of breast milk were positive in 3 studies^19,20,55 (Z= -0.900, p=0.368), and both of the two vaginal swabs were negative in one study¹⁷ (Z= -1.039, p=0.299) (Table 3 and Supplementary file Figures 16 and 17).

Pooled analysis results from case reports

According to the pooled results of the 22 case reports^{18,20,22,23,25-28,58-71}, the most common symptoms among pregnant women recorded in case reports were fever (81.8%), cough (54.5%), fatigue/tiredness/ myalgia (40.9%) and dyspnea/shortness of breath (40.9%). It was seen that all of these pregnant women tested negative in the RT-PCR test and 72.2% had normal chest-CT scans; the method of treatment was mostly with antibiotics (68.2%) and antiviral drugs (59.1%). Of the pregnant women, 36.4% had underlying conditions and 36.4% had a history of preterm labor; 75% had delivered by caesarean. Eight mothers were admitted into intensive care and two mothers lost their lives. Six out of 20 newborns were admitted into the NICU, and only 2 were being breastfeed while 11 were separated and isolated from their mothers. It was seen that in 1 of the 13 infants tested in the first 24 hours after birth tested positive on the RT-PCR test and 1 of 3 babies tested positive for IgG and IgM. In the later period (48 hours –14 days), 2 of the 8 infants tested had positive RT-PCR results (Table 5).

Outcomes of pregnant women with COVID-19

This systematic review and meta-analysis revealed that pregnant women with COVID-19 most commonly complained of high fever, chills/shivers, fatigue/tiredness/myalgia, cough, dyspnea/shortness of breath, chest pain/tightness, phlegm, runny nose/congestion, diarrhea, headache and sore throat. Similar results have been reported in previous systematic reviews concerning pregnant women^5,6,72-75. In this study, it was found that COVID-19 was asymptomatic in approximately one-third of the pregnant women (35%). Lower percentages were reported in studies by Elshafeey et al.⁶ and Mullins et al.⁷⁶ (7.5% and 32%, respectively). The results are similar to those observed in non-pregnant women⁷⁷ and are valuable in terms of offering data that will be useful in the early diagnosis, follow-up and care of pregnant women.

It was found in this meta-analysis that a large percentage of the pregnant women with COVID-19 tested positive on the RT-PCR (80%), revealed an abnormal chest CT (88%) and that some women displayed abnormal chest X-rays (37%). The results of some studies show that a significant percentage of pregnant women with COVID-19 (38%) were diagnosed with clinical findings. Similarly, Smith et al.⁷⁴ also report that approximately 90% of the women in their study were administered the RT-PCR test to confirm the diagnosis, that in 79% of the cases, tests were positive and RT-PCR negative in 23%. The authors reported that 99% of the pneumonia cases were diagnosed with CT findings. De Rose et al.⁵ state that the respiratory system samples taken from the women tested positive on the RT-PCR test in almost all of the women (99%). Elshafeey et al.⁶ report, however, that most of the women (90%) were tested with RT-PCR for a diagnosis and the infection was confirmed with radiological and clinical findings in a smaller percentage of women (10%). Della Gatta et al.⁷² report similar results. These data were similar with the diagnostic tests used with non-pregnant individuals^77,78, indicating that more information is needed regarding the effects on pregnant women and infant health of diagnostic tests carried out with radiological methods.

It was noted in this systematic review and meta-analysis that pregnant women with COVID-19 were treated with antiviral drugs, antibiotics, intravenous immunoglobulin, a Chinese drug, hydroxychloroquine, corticosteroids, anticoagulants and oxygen support. Another study similarly indicated that pregnant women with COVID-19 were commonly treated with antibiotics, oxygen, antiviral drugs and corticosteroids⁷³. It was reported in another study that a significant percentage of hospitalized pregnant women (28%) were put on oxygen support⁷⁴. These reports show that pregnant women with COVID-19 were treated similarly to non-pregnant individuals⁷⁸. There is a need for more data on the outcomes of such treatments on pregnancy and the infant.

In 2 studies examined in this systematic review and meta-analysis, a large majority of the pregnant women with COVID-19 were treated as outpatients or followed up by telephone. Again, it was reported in some studies with respect to preterm labor that hydration, steroid and MgSO₄ therapy was applied to some of the pregnant women. These reported results are valuable in that they indicate that pregnant women with COVID-19 can be followed up with routine care similar to what would be implemented under normal circumstances.

This meta-analysis showed that some pregnant women with COVID-19 had underlying conditions such as diabetes mellitus, asthma, chronic hypertension, hypothyroidism, Hepatitis B, obesity, heart disease and polycystic ovary syndrome. On the other hand, Della Gatta et al.⁷² report no pre-existing comorbidities before the pregnancy. It is known that COVID-19 presents a life-threatening risk when there are comorbidities⁷⁷ and that high-risk pregnant women with a history of chronic disease must particularly be carefully protected, making follow-up a matter of vital importance.

The study revealed that some pregnant women had diabetes mellitus, hypertension, preeclampsia and anemia. In a similar study, Della Gatta et al.⁷² reported that one of the 35 pregnant women had gestational hypertension while another suffered from preeclampsia. These conditions can be life-threatening for pregnant women with COVID-19 and necessitate cautious care and follow-up.

Furthermore, the study demonstrated that some pregnant women have additional conditions during their illness such as reduced fetal movement, multiple pregnancy, early membrane rupture, placental abruption, placenta previa and polyhydramnios/oligohydramnios. A previous study⁵ revealed similar results. These conditions may mean added difficulties in treatment, follow-up and care.

Outcomes related to the gestation period in pregnant women with COVID-19 and their mode of delivery

Our study indicates that preterm labor was observed in about 18% of the pregnant women with COVID-19. While this percentage is lower in the studies of Della Gatta et al.⁷² and Elshafeey et al.⁶ (12% and 15%, respectively), and De Rose et al.⁵, Smith et al.⁷⁴ and Yang et al.⁷⁵ report higher rates (34%, 64% and 21%, respectively). These rates are also higher than what is reported for the general population (11%)^79,80. Preterm birth is closely associated with infant fatality and therefore this information must be carefully taken into consideration in the care and follow-up of infants with COVID-19.

This systematic review and meta-analysis reports that most of the pregnant women with COVID-19 (77%) were delivered by caesarean section. In some of the studies examined, caesarean childbirth was implemented mostly for obstetric indications (43%) and for some women for COVID-19 indications (29%). Elshafeey et al.⁶, De Rose et al.⁵ and Smith et al.⁷⁴ reported similar caesarean rates in their studies (69%, 77% and 80%, respectively). It can be seen in other studies that the caesarean birth rate in women with COVID-19 is much higher^72,73,75. Although these outcomes have not been reported as indications, it is apparent that the caesarean birth rate among pregnant women with COVID-19 is markedly high. This outcome should be assessed in terms of maternal and infant health.

Perinatal outcomes of the newborns of pregnant women with COVID-19

It was noted in this systematic review and meta-analysis that 19% of the newborns of women with COVID-19 was of low birth weight. A previous study revealed a similar percentage⁵. Contrary to this data, however, Elshafeey et al.⁶ and Yang et al.⁷⁵ report a lower frequency of low birth weight (8% and 5.3%, respectively) while Smith et al.⁷⁴ point to a higher rate (43%). These results can be associated with the incidence of preterm labor among pregnant women with COVID-19.

It was found in this systematic review and meta-analysis that some infants born of pregnant women with COVID-19 had low APGAR scores (<7) in the first (4.3%) and fifth minute (4.6%), suffered from fetal asphyxia and fetal distress (14%). The incidence of fetal distress was reported as lower in studies by Elshafeey et al.⁶ and Yang et al.⁷⁵ (8% and 11%, respectively). De Rose et al.⁵ reported this rate to be 22%. In Smith et al.⁷⁴, it was reported that all the newborns in their study exhibited normal APGAR scores. Again, in the study by Yang et al.⁷⁵, a similar outcome was reported with regard to neonatal asphyxia (1.2%). These results are significant in that they indicate that infants born of women with COVID-19 must be kept under close scrutiny in the perinatal period.

In some studies that were reviewed for this systematic review and meta-analysis, there were data that indicated that some newborns were being breastfed (33%). Elshafeey et al.⁶ had presented similar results in their systematic review. These data are valuable in terms of the health of breastfed infants and indicate the need for additional research and the adoption of urgent measures to develop methods by which women with COVID-19 can safely provide their infants with the breast milk they require.

It was seen in the meta-analysis that some newborns of women with COVID-19 were administered various tests in the first 24 hours and later after birth. The analysis showed that some infants tested positive in the first 24 hours and some in the period Day 2 – Day 14 on the RT-PCR test. It was also seen that some infants displayed positive IgG and IgM results in the first 24 hours. It was found in the meta-analysis that the samples taken from pregnant women with COVID-19 (placenta, umbilical blood, breast milk, vaginal swab and amniotic fluid) tested positive on the COVID-19 test in the first 24 hours in only one woman in the placenta and umbilical blood specimens and in two women in Days 2–14 in the breast milk specimens. Similar results can be seen in other studies as well^5,6,72,74. These results are not sufficient, however, in terms of making an assessment of the possibility of vertical transmission; more evidence is needed.

Outcomes on admittance of pregnant women with COVID-19 and their infants into the intensive care unit and maternal–infant death

It was noted in this systematic review and meta-analysis that a significant portion of the pregnant women were admitted into the intensive care unit and were administered intubation/mechanic ventilation. Elshafeey et al.⁶ similarly reported that some women were admitted into intensive care (4.4%) and were administered mechanical ventilation (1.6%). In the study by Smith et al.⁷⁴, it was reported that one woman (4.3%) needed mechanical ventilation and was admitted into the intensive care unit. These results indicated that COVID-19 is a serious health issue that threatens maternal/infant health during the period of pregnancy and beyond.

In all of the studies reviewed for this systematic review and meta-analysis, nine mothers were reported to have lost their lives. In their study of 108 pregnant women, Zaigham and Andersson⁷³ reported no incidents of maternal mortality. Although the percentage was higher than the general maternal mortality rate, it is lower that the mortality rate reported for SARS-CoV-2 (3.8%)⁸¹.

It was again seen in this study that a significant portion of the newborns were admitted into the NICU. Similarly, in some other previously published systematic reviews, it was reported that some infants had been admitted to the neonatal intensive care unit due to an additional symptom or for further care and follow-up^6,74. These results may be associated with the high level of premature birth and low birth weight rates.

In all of the studies reviewed for this systematic review and meta-analysis, it was reported that 3 out of 385 infants lost their lives in the intrauterine stage and 5 died as neonates. Smith et al.⁷⁴ also reported in the data they presented for 37 infants that one baby had died in the intrauterine and one in the neonatal stages. Moreover, Yang et al.⁷⁵ and Zaigham and Andersson⁷³ reported one mortality in the intrauterine and neonatal stages in their studies. In the systematic review of Elshafeey et al.⁶, the authors provided findings from 256 newborns, reporting two stillbirths and one neonatal death. Both De Rose et al.⁵ and Della Gatta et al.⁷² reported one stillbirth in each of their studies. The rates are higher than in general infant rates and are important in that they indicate that COVID-19 is responsible for increasing infant fatalities.

Strengths and limitations

The high score noted in the current quality assessment of the studies examined in this systematic review and meta-analysis and the wide range of additional resources available for scanning constituted the strengths of the study. Also, the large number of pregnant women and infants whose data were reviewed in this systematic review was another strong point that consequently strengthened the conclusions that were drawn. The study was further strengthened by the fact that the results were based on reliable methods of analysis, the subject matter was examined from different aspects, and the results attained were supported by outcomes reported in previous studies. It can be said, however, that the low extent of homogeneity in the studies reviewed may have weakened the power of the evidence. To keep this factor under control, the Random Effect model was preferred in analyses in which the extent of heterogeneity was high. Other limitations were that only studies published in English were taken into the analysis, most of the studies were based on small sample sizes and were derived from Chinese sources, China being the location where COVID-19 had originated, which also meant that studies published in Chinese and other languages could not be included.