Adverse Perinatal Outcomes in COVID-19 Infected Pregnant Women: A Systematic Review and Meta-Analysis

The impact of COVID-19 virus infection during pregnancy is still unclear. This systematic review and meta-analysis aimed to quantitatively pool the evidence on impact of COVID-19 infection on perinatal outcomes. Databases of Medline, Embase, and Cochrane library were searched using the keywords related to COVID-19 and perinatal outcomes from December 2019 to 30 June 2021. Observational studies comparing the perinatal outcomes of COVID-19 infection in pregnancy with a non-infected comparator were included. The screening process and quality assessment of the included studies were performed independently by two reviewers. Meta-analyses were used to pool the comparative dichotomous data on perinatal outcomes. The database search yielded 4049 results, 1254 of which were duplicates. We included a total of 21 observational studies that assessed the adverse perinatal outcomes with COVID-19 infection. The odds of maternal death (pooled OR: 7.05 [2.41−20.65]), preeclampsia (pooled OR: 1.39 [1.29−1.50]), cesarean delivery (pooled OR: 1.67 [1.29−2.15]), fetal distress (pooled OR: 1.66 [1.35−2.05]), preterm birth (pooled OR: 1.86 [1.34−2.58]), low birth weight (pooled OR: 1.69 [1.35−2.11]), stillbirth (pooled OR: 1.46 [1.16−1.85]), 5th minute Apgar score of less than 7 (pooled OR: 1.44 [1.11−1.86]) and admissions to neonatal intensive care unit (pooled OR: 2.12 [1.36−3.32]) were higher among COVID-19 infected pregnant women compared to non-infected pregnant women.


Introduction
Coronavirus disease 2019 , caused by the SARS-CoV-2 virus, continues to be an alarming global public health crisis [1] with a sharply escalating number of deaths that have largely surpassed previous fatalities caused by epidemics such as Middle Eastern Respiratory Syndrome (MERS) and Severe Acute Respiratory Syndrome (SARS) [2]. At the time of writing (3 December 2021), 263,563,622 confirmed cases of COVID-19, including 5,232,562 deaths, had been reported to the World Health Organization (WHO) [3]. This situation raises concerns in vulnerable populations such as pregnant mothers, fetuses and their neonates. Pregnant women are at higher risk of developing severe illness from respiratory infections, largely due to immunodeficiency associated with physiological adaptations during pregnancy [4]. Respiratory infections could escalate rapidly to respiratory failure, leading to potentially fatal consequences for both mother and fetus [5]. A recent multinational retrospective cohort study of 388 pregnant women reported that SARS-CoV-2 infected pregnant women risk fatal consequences from compromised respiratory functions and need intensive care [6]. Healthcare systems continue to become over-burdened, risking

Materials and Methods
We conducted this systematic review and meta-analysis based on the PROSPERO protocol registered on 18 May 2021 (CRD42021254974). This review included studies focused on perinatal outcomes of COVID-19 infection in pregnancy, mainly to evaluate the reported adverse perinatal outcomes in COVID 19 infected mothers and the prevalence of adverse perinatal (maternal, fetal, and newborn) outcomes in COVID-19 infected pregnant women. This review reports adverse maternal, fetal, and newborn outcomes of COVID-19 infected pregnant women, comorbidities in COVID-19 infected pregnant women, and the incidence of COVID-19 infection among pregnant women in line with the updated PRISMA 2020 guidelines for reporting systematic reviews [12].

Eligibility Criteria, Data Sources and Search Strategy
Observational studies (cohort and case-control) investigating the perinatal outcomes in COVID-19 infected pregnant women and published as peer-reviewed articles in English were eligible for inclusion. Case reports, case series, editorials, letters to the editor, perspectives, conference papers, narrative or systematic reviews, and studies without a noninfected pregnant group as the comparator were excluded. We searched Medline, EMBASE, and Cochrane Library databases to identify the published studies from December 2019 to 30 June 2021. The search strategy included a combination of keywords for COVID-19 and perinatal outcomes (Table S1).

Study Selection
All of the identified studies from the database search were exported to EndNote reference management software (version EndNote X9.3.3.). Then, Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia) was used to manage the independent screening process at both the stages of title and abstract screening (M.L.P., B.P.P.S., T.S.D.) and full-text screening (M.L.P., B.P.P.S., T.S.D.). Reasons for full-text exclusion were documented at the full-text screening stage. Any disagreement was resolved by consensus or by consultation with a third reviewer at both stages.

Data Extraction
The data from the included studies were extracted to an Excel sheet by one author, and another author cross-checked the accuracy. The extracted data included the study characteristics (country, year of publication, study design and methodology, study period, population and setting, total number of participants, number of cases, number in control group and drop-outs), participants' socio-demographic and baseline data, comorbidities, adverse perinatal outcomes (maternal death, termination of pregnancy, miscarriage or abortion, preeclampsia, pre-labor rupture of membrane (PROM), preterm pre-labor rupture of membrane (PPROM), intrauterine death, fetal distress, preterm birth, low birth weight, stillbirth, Apgar score, admissions to neonatal intensive care unit (NICU), neonatal deaths, cesarean section deliveries, and operative vaginal births, the incidence of COVID-19 among pregnant women, and the outcome of interest of each study.

Assessment of Risk of Bias
Quality assessments of the included studies were performed using the National Institute of Health's (NIH) study quality assessment tool for observational, cohort, and cross-sectional studies and the NIH study quality assessment tool for case-control studies [13]. The quality of each study was independently assessed by two assessors (B.P.P.S. and T.S.D). Any disagreement was resolved through consensus between the two assessors.

Data Synthesis and Analysis
The characteristics of the included studies, characteristics of the COVID-19 infected pregnant women and the summary findings were tabulated. Further, the incidence of COVID-19 among pregnant women was graphically presented. Quantitative meta-analysis was carried out to pool the comparative dichotomous data of perinatal outcomes when more than one study presented the data for the relevant outcome. If individual studies reported no adverse outcome in the infected group or non-infected group, they were excluded from the meta-analysis of that particular perinatal outcome. Heterogeneity of studies was determined using the I 2 statistic, where substantial heterogeneity was defined as I 2 ≥ 30. Random effects estimates of the pooled odds of each perinatal outcome and comorbidity condition were generated using the Mantel-Haenszel method. The findings of each outcome comparison were summarized with odds ratio, 95% confidence interval, p-value, and the I 2 statistic. Funnel plots were generated to visually evaluate the presence of publication bias.

Study Selection
Two thousand seven hundred ninety-five (2795) studies were identified through the search engines for the title and abstract reviews after removing 1254 duplicates. Out of the total screened abstracts, 120 were selected for full-text screening, of which 99 studies were excluded (mainly due to lack of a comparison group or the presentation of inadequate data), and 21  studies were included in this systematic review and meta-analysis ( Figure 1).

Risk of Bias of Included Studies
Regarding the quality of the included cohort studies, 10 criteria out of 14 (71%) were satisfied by 40% of the included studies. Almost all of the studies had clearly stated research objectives, clearly defined study populations, clearly defined valid and reliable outcomes, and over ≥50% participation rate by eligible persons. In almost all the included studies, the quality assessment was unable to determine the level of exposures related to examined outcomes, exposure measures more than once over time, and follow-up after baseline. Blinding of the assessors to the exposure status was a serious concern for all the included studies. Only 40-55% of the included cohort studies were marked positively for the criteria of adjusting for potential confounding factors and having a justified sample size. With regards to the quality of included case-control studies, eight criteria out of 12 (75%) were satisfied by 60% of the included studies. All the studies satisfied the criteria related to clearly defined objective/s, clearly defined study population, selection of the control from the same population, consistent use of defined inclusion and exclusion criteria, clearly defined and differentiated case and control groups, ability to confirm the exposure occurred prior to the development of the condition, implementation of valid and reliable exposure measures, and measuring and adjusting for confounding variables. Blinding of the assessors to the exposure status was not determinable in all the studies. Less than 35% of the included case-control studies had a justified sample size (Figure 2A,B). Individual study assessments were attached as a supplementary file (Table S2).
Healthcare 2022, 10, x 6 of 21 Less than 35% of the included case-control studies had a justified sample size ( Figure  2A,B). Individual study assessments were attached as a supplementary file (Table S2).

Incidence of COVID-19 Infection in Pregnant Women
Eleven studies reported the incidence of COVID-19 among pregnant women, with rates ranging from 1.3% to 27%. Only cohort studies were used to determine the incidence of COVID-19 infection in pregnant women. Even though there were 18 cohort studies, a few did not report the total number of admissions, making it difficult to quantify the incidence. Among the 11 that reported incidence, there were six studies from the USA [15,20,24,25,29,31], three single-center [15,20,25] and three multicenter studies [24,29,31]. The reported rates in the USA ranged from 1.3% to 19%. The highest rate (27%) of COVID-19 in pregnancy was reported from a single-center study conducted in France [22], while 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Incidence of COVID-19 Infection in Pregnant Women
Eleven studies reported the incidence of COVID-19 among pregnant women, with rates ranging from 1.3% to 27%. Only cohort studies were used to determine the incidence of COVID-19 infection in pregnant women. Even though there were 18 cohort studies, a few did not report the total number of admissions, making it difficult to quantify the incidence. Among the 11 that reported incidence, there were six studies from the USA [15,20,24,25,29,31], three single-center [15,20,25] and three multicenter studies [24,29,31]. The reported rates in the USA ranged from 1.3% to 19%. The highest rate (27%) of COVID-19 in pregnancy was reported from a single-center study conducted in France [22], while the second-highest rate was noted from a multicenter study conducted in Spain [26] (Figure 3).
The reported rates in the USA ranged from 1.3% to 19%. The highest rate (27%) of COVID-19 in pregnancy was reported from a single-center study conducted in France [22], while the second-highest rate was noted from a multicenter study conducted in Spain [26] (Figure 3).

Characteristics of COVID-19 Infected Pregnant Women
In the 21 included studies, a total of 14,131 COVID-19 infected pregnant women were studied compared to 585,376 COVID-19 non-infected pregnant women. The reported mean age of infected pregnant women ranged from 24.7 to 32.6 years, while some of the studies reported median (IQR) values ranging from 25 (21-31) to 33.3 (29-37) years (Table  2).

Characteristics of COVID-19 Infected Pregnant Women
In the 21 included studies, a total of 14,131 COVID-19 infected pregnant women were studied compared to 585,376 COVID-19 non-infected pregnant women. The reported mean age of infected pregnant women ranged from 24.7 to 32.6 years, while some of the studies reported median (IQR) values ranging from 25 (21-31) to 33.3 (29-37) years (Table 2).

Summary Findings of Included Individual Studies
Out of 21 studies, six reported that COVID-19 infection during pregnancy was not associated with adverse perinatal outcome [15,18,25,[30][31][32]. A study conducted in Spain concluded that even with no difference in the overall rate of adverse perinatal outcomes among COVID -19 infected women, symptomatic status was associated with a modest increase in preterm delivery and intrapartum fetal distress [18]. All of the other studies reported one or more significant adverse perinatal outcomes associated with COVID-19 in pregnancy. Table 3 shows the summary findings of individual studies included in this systematic review.  [29] Increased risk Cesarean section delivery Ríos-Silva et al., 2020 [30] No difference Steffen et al., 2021 [31] No difference Trahan et al., 2021 [32] No difference Increased risk Cesarean section delivery Neonatal intensive care unit admission ‡ Relevant to the studied perinatal outcomes in the current systematic review, † No difference in the overall rates but the symptomatic status was associated with modest increases in preterm delivery and intrapartum fetal distress, † † Study encompassed only the asymptomatic pregnant women. One study was not included in the table as its outcome was based on disease severity [17].

Adverse Fetal Outcomes
The reported fetal outcomes included intrauterine death and fetal distress. Out of these, fetal distress was found to be statistically significant. Based on the data from 1248 newborns born to COVID-19 infected pregnant women and 7422 newborns born to COVID-19 non-infected pregnant women [14,18,21,31,33], a statistically significant increase in fetal distress was observed among the newborns of the COVID-19 infected women compared to the COVID-19 non-infected (pooled OR 1.66 [95% CI 1.35−2.05]; p < 0.05; I 2 = 26%). Four studies [16,17,22,27] reported data on intrauterine death, and of them, one study [16] was excluded from the meta-analysis as no adverse events were reported in infected and non-infected cohorts. The meta-analysis of the remaining three studies (348 COVID-19 infected pregnancies and 1376 COVID-19 non-infected pregnancies) found no statistically meaningful change in intrauterine deaths related to COVID-19 infection during pregnancy (pooled OR 1.79 [95% CI 0.51−6.23]; p = 0.36; I 2 = 68%) ( Figure 4B). ‡ Relevant to the studied perinatal outcomes in the current systematic review, † No difference in the overall rates but the symptomatic status was associated with modest increases in preterm delivery and intrapartum fetal distress, † † Study encompassed only the asymptomatic pregnant women. One study was not included in the table as its outcome was based on disease severity [17].

Adverse Fetal Outcomes
The reported fetal outcomes included intrauterine death and fetal distress. Out of these, fetal distress was found to be statistically significant. Based on the data from 1248 newborns born to COVID-19 infected pregnant women and 7422 newborns born to COVID-19 non-infected pregnant women [14,18,21,31,33], a statistically significant increase in fetal distress was observed among the newborns of the COVID-19 infected women compared to the COVID-19 non-infected (pooled OR 1.66 [95% CI 1.35−2.05]; p < 0.05; I 2 = 26%). Four studies [16,17,22,27] reported data on intrauterine death, and of them, one study [16] was excluded from the meta-analysis as no adverse events were reported in infected and non-infected cohorts. The meta-analysis of the remaining three studies (348 COVID-19 infected pregnancies and 1376 COVID-19 non-infected pregnancies) found no statistically meaningful change in intrauterine deaths related to COVID-19 infection during pregnancy (pooled OR 1.79 [95% CI 0.51−6.23]; p = 0.36; I 2 = 68%) ( Figure  4B).

Discussion
We conducted this systematic review to pool the available evidence of adverse perinatal outcomes caused by COVID-19 infection in pregnancy. We retrieved a total of 21

Discussion
We conducted this systematic review to pool the available evidence of adverse perinatal outcomes caused by COVID-19 infection in pregnancy. We retrieved a total of 21 observational studies that assessed the adverse perinatal outcomes in pregnant women with COVID-19 infection published from December 2019 to June 2021.
Overall findings of our study were, (1) the reported incidence rates of COVID-19 infection among pregnant women ranged from 1.3% to 27%, disregarding the fact that the results were based on single-center studies to multinational studies; (2) with regards to the adverse maternal outcomes, we found that there was a statistically significant increase in maternal deaths, preeclampsia, and cesarean deliveries, while miscarriages/abortions, PROMs/PPROMs, and operative vaginal births were non-significant in COVID-19 infected pregnant women compared to non-infected; (3) with regards to the adverse fetal outcomes, fetal distress was found to be statistically significant, while intrauterine death was nonsignificant in COVID-19 infected pregnancies; and (4) with regards to the adverse neonatal outcomes, all reported fetal outcomes except neonatal death, including preterm birth, low birth weight, stillbirth, fifth minute Apgar score < 7, and admissions to NICU showed significant differences in births to COVID-19 infected women compared to non-infected.
The current study findings were consistent with previously published systematic reviews relevant to maternal death [35], preeclampsia [36], preterm birth [35,36], stillbirth [36] and admissions to NICU [35]. In addition to those findings, we found increased cesarean section deliveries among COVID-19 infected women compared to non-infected, 12982, and 583619. However, the data included in the present study did not consider whether those cesarean sections were elective or emergency cases based on COVID-19 status. Pooling of comorbidity data of infected and non-infected pregnant women revealed that comorbidities during pregnancy were not significantly higher in COVID-19 infected pregnancies. This finding was inconsistent with the findings of a previous systematic review, which observed a higher risk of COVID-19 infection in pregnancy when having pre-gestational diabetes mellitus, gestational diabetes mellitus, and chronic hypertension [35]. Out of 21 studies, more than 90% of the studies in this review assessed perinatal outcomes regardless of the disease severity. Consequently, not enough information was available to assess the differences in maternal, fetal, and neonatal outcomes based on disease severity. Therefore, further studies are recommended to assess the perinatal outcomes based on disease severity in order to clear up uncertainties in this area.

Implications for Clinical Practice
Healthcare providers should be aware that women infected with COVID-19 have an elevated risk of disease severity, including maternal mortality. Pregnant women should be advised of the disease's increased severity and encouraged to take precautions to avoid infection. Primary healthcare providers will need to balance the necessity for routine multidisciplinary prenatal care and the management of women suspected of having COVID-19 infection, preferably via virtual antenatal clinics. Pregnant women who become infected with COVID-19 before reaching term may require management in a tertiary healthcare facility equipped with cesarean section and NICU facilities to manage preterm infants, infants with low Apgar scores, and infants with fetal distress.

Strengths and Limitations
This systematic review has several strengths. First, the study followed a sound methodology and was able to quantify the findings using meta-analyses. Second, a comprehensive search strategy was used to minimize the risk of missing relevant studies. Third, the screening was independently assessed by pairs of reviewers, and discrepancies solved by consensus. Fourth, excluding the publication types such as case studies, case reports, and case series left studies with a quality study design included in the final analysis. Finally, the present systematic review adhered to a rigorous quality appraisal. An important amount of evidence was summarized and critically appraised in addition to the highlighted evidence gaps.
Our systematic review also has limitations. Firstly, the method of diagnosis of COVID-19 in pregnancy was different from study to study. Secondly, without data on disease severity, perinatal outcomes based on disease severity could not be determined. Thirdly, many studies represented developed countries with only meager contributions from lowresource countries. However, the findings of this systematic review have implications for low and middle-income countries with limited resources, where the negative impacts are prominent due to region-specific management strategies and resources. Finally, asymmetry of the funnel plots was observed for the assessed variables, and the presence of publication bias was suggested. This asymmetry may be also due to some other factors such as poor methodological design, reporting bias, chance or study heterogeneity. Despite all limitations, we undertook a comprehensive literature review and meta-analysis with the most updated findings relevant to adverse perinatal outcomes in COVID-19 infected pregnant women.

Conclusions
Several adverse maternal, fetal, and neonatal effects were significantly higher in COVID-19 infected pregnant women than non-infected. These included maternal death, preeclampsia, cesarean section delivery, fetal distress, preterm birth, low birth weight, stillbirth, low Apgar score at the fifth minute, and admission to NICU. The comorbidity conditions had no added risk of being infected with COVID-19 infection during pregnancy. Therefore, a COVID-19 infected pregnant woman should be treated with special precautions to avoid and minimize the identified adverse events during perinatal care. Further studies are recommended to collect more robust data relevant to the adverse perinatal outcomes that will enable effective clinical decision-making in maternal and child health care.