The Differences in Clinical Presentation, Management, and Prognosis of Laboratory-Confirmed COVID-19 between Pregnant and Non-Pregnant Women: A Systematic Review and Meta-Analysis

Background: The coronavirus disease 2019 (COVID-19) pandemic has affected millions of people across the globe. Previous coronavirus outbreaks led to worsened symptoms amongst pregnant women, suggesting that pregnant women are at greater risk. Objectives: Our aim is to investigate the differences in clinical presentation, management, and prognosis of COVID-19 infection in pregnant and non-pregnant women. Methods: We ran a search on electronic databases and analysis of the relevant articles was done using Revie Manager 5.4. Results: The review consists of nine studies comprising 591,058 women (28,797 pregnant and 562,261 non-pregnant), with most of the data derived from two large studies. The risk of experiencing fever (RR: 0.74; 95% CI: 0.64–0.85), headache (RR: 0.77; 95% CI: 0.74–0.79), myalgia (RR: 0.92; 95% CI: 0.89–0.95), diarrhea (RR: 0.40, 95% CI: 0.39–0.43), chest tightness (RR: 0.86; 95% CI: 0.77–0.95), and expectoration (RR: 0.45; 95% CI: 0.21–0.97) were greater amongst non-pregnant COVID-19-infected women. Pregnant women with COVID-19 were less likely to be obese (RR: 0.68; 95% CI: 0.63–0.73) or have a smoking history (RR: 0.32; 95% CI: 0.26–0.39). COVID-19-infected non-pregnant women had a higher frequency of comorbidity such as chronic cardiac disease (RR: 0.58; 95% CI: 0.44–0.77), renal disease (RR: 0.45; 95% CI: 0.29–0.71), and malignancy (RR: 0.82; 95% CI: 0.68–0.98), compared to COVID-19-infected pregnant women. The risk of ICU admission (RR: 2.26; 95% CI: 1.68–3.05) and requirement of invasive mechanical ventilation (RR: 2.68; 95% CI: 2.07–3.47) were significantly higher amongst pregnant women. Conclusions: Although the frequency of risk factors and the risk of experiencing clinical symptoms of COVID-19 were higher among non-pregnant women, COVID-19-infected pregnant women had a higher requirement of ICU admission and invasive mechanical ventilation compared to non-pregnant COVID-19-infected women. More well-conducted studies from varying contexts are needed to draw conclusions. Prospero registration: CRD42020204638.


Introduction
In December 2019, the coronavirus disease 2019 (COVID-19) first emerged as a cluster of pneumonia cases of unknown origin in Wuhan, China [1]. The disease, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was soon declared a "public health emergency of international concern" and later characterized as a pandemic by the World Health Organization (WHO) [2]. As of 13 October 2020, there have been a total of more than 40.2 million confirmed cases of COVID-19, with over 1.1 million deaths worldwide [3]. Since the global outbreak, several studies have been published reporting clinical characteristics, laboratory findings, and management associated with COVID-19 in the general population, focusing mainly on non-pregnant adults [4][5][6][7].
Pregnancy is a unique state in which the maternal immune system has to overcome two main challenges: protecting the fetus against an immunological attack while maintaining adequate defense against various microbial threats. Physiological and mechanical changes associated with gestation predispose pregnant women to severe forms of respiratory infections with subsequent higher maternal and fetal mortality [8,9]. During the last two decades, coronavirus has been responsible for two major epidemics; the severe acute respiratory syndrome (SARS-CoV) and Middle East respiratory syndrome (MERS-CoV), with a case fatality of 10.5% and 34.4%, respectively [10]. It was found that these infections were associated with worsening symptoms and clinical outcomes among pregnant women ranging from severe maternal illness to spontaneous abortion, and even maternal death [8,9,11] Although SARS-CoV-2 appears to be less virulent than the aforementioned coronaviruses, its spread is far more rapid and efficient among close contacts [12]. Therefore, it has raised additional concerns in pregnant women because previous experiences with both SARS-CoV and MERS-CoV have shown severe complications in this vulnerable population.
The increased risk of viral pneumonia in the obstetric population makes it imperative to evaluate whether there is any difference in the clinical course and outcomes between pregnant and non-pregnant women infected with COVID-19. Furthermore, we found no systematic review that provides a comparison of available evidence on COVID-19 among women of reproductive age based on their pregnancy status. Therefore, in this systematic review and meta-analysis, we aim to describe the clinical characteristics, management, and prognosis of COVID-19-infected pregnant women compared to COVID-19-infected non-pregnant women. In the next section, we specify the method with which this systematic review and meta-analysis was conducted, after which we have the results section, the discussion, and finally, the conclusion. The findings of this review will facilitate healthcare workers in understanding the disease, aid in the clinical management and counseling of these patients, as well as allow policy makers to form guidelines for the general public.

Methodology
This systematic review has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) database with ID number CRD42020204638 and follows the recommendations established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [13] (Table S1).
A systematic literature search was conducted until 25 February 2021, using PubMed, Embase, the WHO COVID-19 database, and Google Scholar. Furthermore, medRxiv and bioRxiv were screened for pre-print papers. The following terms and their variants were combined and used in devising the search strategy: "Pregnant women" OR "Pregnancy" AND "Coronavirus" OR "Covid-19" OR "SARS-CoV-2". The full search strategy and terms used are available in Table S2. We did not apply any language restrictions; however, papers published since 31 December 2019, were included.
All observational studies (cohort, case-control, cross-sectional, or case-series) including consecutive patients with a comparison group of pregnant and non-pregnant women of reproductive age group with laboratory-confirmed SARS-CoV-2 infection and reporting clinical characteristics, management, and prognosis were considered eligible. A case-series was defined as a study with a sample size of less than ten participants. Only studies that compared pregnant women with COVID-19 with non-pregnant women with COVID-19 were eligible for inclusion. We excluded studies describing only either pregnant or nonpregnant women with COVID-19. Studies were checked for data overlapping by assessing their center of data collection and the time period during which the data were collected. When it was unclear, authors were contacted to ensure that they reported results from different centers or during different time periods. Identified overlapping papers were further assessed and the studies with inclusion of more variables, bigger sample size, and better quality of assessment were chosen, as shown in Table S3.
Two reviewers (D.S.A.K. and A.N.P.) independently screened the titles and abstracts for eligibility. After the initial search, full texts of relevant articles were examined for inclusion and exclusion criteria. Primary studies that fulfilled the inclusion criteria were selected for this systematic review. Any disagreement among the authors was resolved through consensus or consulting a senior reviewer (Z.S.L.).
Two authors (D.S.A.K. and A.N.P.) extracted relevant information independently from included studies. The following items were extracted from each study if available: author's name, study design, country, duration of the study, setting, total number of study participants, demographics, past medical history, presenting signs and symptoms, management, and complications. For clinical presentation, data on the number of asymptomatic and COVID-19-like symptoms were extracted. Data on management with antivirals, antibiotics, corticosteroids, or any other new medication/technique were also recorded. Complications in the two groups, be it progression to severe COVID-19 infection, requirement for ventilation, intensive care unit (ICU) admission, or death, were also considered. The quality of the studies included in this meta-analysis was assessed using the National Heart, Lung, and Blood Institute (NHLBI) quality assessment tool for observational cohort and case-control studies [14]. This tool helps evaluate the internal validity of a study, hence ensuring that the results are truly due to the exposure being evaluated.
Data were entered and analyzed using Review Manager (RevMan) version 5.4 [15]. Mean difference (MD) with 95% confidence intervals (CI) was used for continuous data and relative risk (RR) with 95% CI for dichotomous data. Random effect models were used and heterogeneity between the studies was explored using the p-value of Chi 2 and I 2 . Sensitivity analysis was performed by removing one large study.

Results
There was a total of 4347 titles identified after the initial search, 4179 articles were excluded after screening titles and abstracts. Of the 168 studies retrieved for full text review, only 9 were found eligible for inclusion [16][17][18][19][20][21][22][23][24]. We excluded 159 studies on full-text review, of which 42 were authors' perspectives or reviews, 44 were guidelines or guidance papers based on other coronavirus strains, 43 studies compared COVID-19 infected pregnant women with non-infected individuals or asymptomatic pregnant women, 26 studies did not have any outcomes of interest reported, and 4 were overlap studies conducted at the same center during the same time period, as shown in Figure 1.
The number of enrolled individuals in each study ranged from 36 to 409,462. Six studies had a sample size of less than 150 participants [16][17][18][20][21][22][23], whereas one study conducted in Mexico had a sample size of 181,088 [24] and another study conducted in the US had a total sample size of 409,462 [19], leading to the total number of participants included in this review to be 591,058. All the participants were COVID-19 positive with 28,797 pregnant women and 562,261 non-pregnant women. Characteristics of included studies are reported in Table 1 and their methodological quality in Table 2a,b. Results from the pooled analysis are presented in Table 3.
The number of enrolled individuals in each study ranged from 36 to 409,462. Six studies had a sample size of less than 150 participants [16][17][18][20][21][22][23], whereas one study conducted in Mexico had a sample size of 181,088 [24] and another study conducted in the US had a total sample size of 409,462 [19], leading to the total number of participants included in this review to be 591,058. All the participants were COVID-19 positive with 28,797 pregnant women and 562,261 non-pregnant women. Characteristics of included studies are reported in Table 1 and their methodological quality in Table 2a,b. Results from the pooled analysis are presented in Table 3.    The most common method of confirming COVID-19 infection was via reverse transcriptase-polymerase chain reaction test (RT-PCR) of a swab sample from either the nasopharynx or the oropharynx. Five studies confirmed COVID-19 infection only through RT-PCR alone [18,[21][22][23][24]. Two studies used both RT-PCR and serological markers (IgM and IgG antibodies) to confirm COVID-19 infection [16,20]. One study with a sample size of 64 reported testing via nucleic amplification and confirmed the testing with a chest computed tomography (CT) scan [17], whereas one study vaguely mentioned using molecular amplification detection test on clinical specimens of 409,462 individuals but did not specify details [19].

Findings
Our meta-analysis found that pregnant women with COVID-19 were 2.8 years younger compared to non-pregnant counterparts with COVID-19 (Table 3). Non-pregnant women were more commonly reported to be obese (RR 0.68; 95% CI 0.63 to 0.73) and have a smoking history (RR 0.32; 95% CI 0.26 to 0.39) compared to pregnant women with COVID-19 infection. Chronic cardiac disease (RR 0.58; 95% CI 0.44 to 0.77), renal disease (RR 0.45; 95% CI 0.29 to 0.71), and malignancy (RR 0.82; 95% CI 0.68 to 0.98) were more commonly present in COVID-19-infected non-pregnant women compared to pregnant women with COVID-19 infection. There was no difference in other reported co-morbidities including diabetes mellitus, chronic respiratory disease, hypothyroidism, mental sickness, and chronic hepatitis (Figure 2). 12 a smoking history (RR 0.32; 95% CI 0.26 to 0.39) compared to pregnant women with COVID-19 infection. Chronic cardiac disease (RR 0.58; 95% CI 0.44 to 0.77), renal disease (RR 0.45; 95% CI 0.29 to 0.71), and malignancy (RR 0.82; 95% CI 0.68 to 0.98) were more commonly present in COVID-19-infected non-pregnant women compared to pregnant women with COVID-19 infection. There was no difference in other reported comorbidities including diabetes mellitus, chronic respiratory disease, hypothyroidism, mental sickness, and chronic hepatitis (Figure 2). Overall, the most common symptoms were headache (four studies, 24.5%), cough (six studies, 23.1%), myalgia (four studies, 17.7%), and fever (seven studies, 17.5%). Pregnant women were at a lower risk of experiencing fever (RR 0.74; 95% CI 0.64 to 0.85; seven studies, 409,838 participants), headache (RR 0.77; 95% CI 0.74 to 0.79; four studies, 409,680 participants), myalgia (RR 0.92; 95% CI 0.89 to 0.95; four studies, 409,680 participants), diarrhea (RR 0.40, 95% CI 0.39 to 0.43; three studies, 409,569 participants), chest tightness (RR 0.86; 95% CI 0.77 to 0.95; three studies, 409,569 participants), and expectoration (RR 0.45; 95% CI 0.21 to 0.97; two studies, 107 participants) as compared to non-pregnant women. The risk of being asymptomatic (RR 3.94; 95% CI 1.69 to 9.20; three studies, 218 participants) was higher amongst pregnant women as compared to nonpregnant women. The risk of other symptoms such as cough, rhinorrhea, chills, fatigue, nausea and vomiting, rash, abdominal pain, dizziness, sore throat, shortness of breath, nasal congestion, and loss of appetite were similar across both groups (Figure 3). Overall, the most common symptoms were headache (four studies, 24.5%), cough (six studies, 23.1%), myalgia (four studies, 17.7%), and fever (seven studies, 17.5%). Pregnant women were at a lower risk of experiencing fever (RR 0.74; 95% CI 0.64 to 0.85; seven studies, 409,838 participants), headache (RR 0.77; 95% CI 0.74 to 0.79; four studies, 409,680 participants), myalgia (RR 0.92; 95% CI 0.89 to 0.95; four studies, 409,680 participants), diarrhea (RR 0.40, 95% CI 0.39 to 0.43; three studies, 409,569 participants), chest tightness (RR 0.86; 95% CI 0.77 to 0.95; three studies, 409,569 participants), and expectoration (RR 0.45; 95% CI 0.21 to 0.97; two studies, 107 participants) as compared to non-pregnant women. The risk of being asymptomatic (RR 3.94; 95% CI 1.69 to 9.20; three studies, 218 participants) was higher amongst pregnant women as compared to nonpregnant women. The risk of other symptoms such as cough, rhinorrhea, chills, fatigue, nausea and vomiting, rash, abdominal pain, dizziness, sore throat, shortness of breath, nasal congestion, and loss of appetite were similar across both groups (Figure 3). The risk of ICU admission was found to be significantly higher amongst pregnant women (RR 2.26; 95% CI 1.68 to 3.05; five studies, 424,587 participants) and they were also more likely to receive invasive mechanical ventilation (RR 2.68; 95% CI 2.07 to 3.47; three studies, 409,616). However, no difference in risk was found in the severity of COVID-19 infection amongst pregnant and non-pregnant women. Severe COVID-19 was reported The risk of ICU admission was found to be significantly higher amongst pregnant women (RR 2.26; 95% CI 1.68 to 3.05; five studies, 424,587 participants) and they were also more likely to receive invasive mechanical ventilation (RR 2.68; 95% CI 2.07 to 3.47; three studies, 409,616). However, no difference in risk was found in the severity of COVID-19 infection amongst pregnant and non-pregnant women. Severe COVID-19 was reported by two articles, one of which by Qiancheng et al. who defined it as "shortness of breath with a respiratory rate greater than 30 breaths/minute, or oxygen saturation less than 93% at rest, or alveolar oxygen partial pressure/faction of inspiration O 2 (PaO 2 /FiO 2 ) less than 300 mmHg" [16] (R Wang et al. failed to define the "severe" COVID-19 infection). The risk of maternal mortality (RR 1.08; 95% CI 0.89 to 1.31) was found to be equal amongst pregnant and non-pregnant women [18] (Figure 4). The risk of ICU admission was found to be significantly higher amongst pregnant women (RR 2.26; 95% CI 1.68 to 3.05; five studies, 424,587 participants) and they were also more likely to receive invasive mechanical ventilation (RR 2.68; 95% CI 2.07 to 3.47; three studies, 409,616). However, no difference in risk was found in the severity of COVID-19 infection amongst pregnant and non-pregnant women. Severe COVID-19 was reported by two articles, one of which by Qiancheng et al. who defined it as "shortness of breath with a respiratory rate greater than 30 breaths/minute, or oxygen saturation less than 93% at rest, or alveolar oxygen partial pressure/faction of inspiration O2 (PaO2/FiO2) less than 300 mmHg" [16] (R Wang et al. failed to define the "severe" COVID-19 infection). The risk of maternal mortality (RR 1.08; 95% CI 0.89 to 1.31) was found to be equal amongst pregnant and non-pregnant women [18] (Figure 4). Both groups were at an equal risk to be managed with oxygen therapy, antivirals, antibiotics, and Chinese medicine. However, the use of immunoglobulins (RR 0.46; 95% CI 0.26 to 0.81) was found to be lesser amongst pregnant females, whereas for corticosteroids (RR 1.61; 95% CI 1.02 to 2.55), it was higher amongst pregnant women as compared to their non-pregnant counterparts as shown in Figure 4. Both groups were at an equal risk to be managed with oxygen therapy, antivirals, antibiotics, and Chinese medicine. However, the use of immunoglobulins (RR 0.46; 95% CI 0.26 to 0.81) was found to be lesser amongst pregnant females, whereas for corticosteroids (RR 1.61; 95% CI 1.02 to 2.55), it was higher amongst pregnant women as compared to their non-pregnant counterparts as shown in Figure 4.

Sensitivity Analysis
The studies conducted by Zambrano (2020) and Martinez-Portilla (2021) were removed and a sensitivity analysis was performed [19,24]. This was because a large sample size (n= 409,462 and n = 181,088, respectively) came from these studies, conducted across 50 states in the US and Mexico alone.
After removing Zambrano et al. and Martinez-Portilla et al., the number of enrolled individuals in each study ranged from 36 to 132. All the participants were COVID-19 positive with 180 pregnant women and 328 non-pregnant women, a combined total of 508 participants.
The findings were similar and suggested that pregnant women were at a lower risk of experiencing fever (RR 0.66; 95% CI 0.53 to 0.83; six studies, 376 participants), shortness of breath (RR 0.57; 95% CI 0.33 to 0.96; five studies, 340 participants), expectoration (RR 0.45; 95% CI 0.21 to 0.97; two studies, 107 participants), and chills (RR 0.03; 95% CI 0.00 to 0.52; one study, 43 participants) compared to non-pregnant women. The chances of being asymptomatic (RR 3.94; 95% CI 1.69 to 9.20, 3 studies, 218 participants) was higher amongst COVID-19-infected pregnant women compared to COVID-19-infected nonpregnant women. The risk of requiring ICU admission or management with medications was equal for both groups. However, the use of immunoglobulins among non-pregnant women was still higher as compared to pregnant women, while the use of corticosteroids was higher among pregnant women (Table 4).

Discussion
Human coronaviruses are among the most common pathogens causing viral respiratory infections. In the past two decades, the world has experienced three coronaviruses outbreaks, and the most recent strain, SARS-CoV-2, has led to the greatest public health crisis of the century. Amid this pandemic, the increasing mortality rate has called for a better understanding and protection of the vulnerable populations infected with the disease.
This systematic review summarizes the findings of 591,058 women with laboratoryconfirmed COVID-19 infection, with 28,797 of them being pregnant. In the present metaanalysis; we found that in comparison with pregnant women, non-pregnant women are at a higher risk of experiencing symptoms such as headache, fever, expectoration, myalgia, chest tightness, wheezing, diarrhea, and anosmia, as primary symptoms of COVID-19. Non-pregnant women of reproductive age with COVID-19 had a higher frequency of comorbidities such as chronic cardiac diseases, renal diseases, and malignancy compared to pregnant COVID-19-infected women. The treatment modalities used in pregnant women were similar to the ones used in non-pregnant women, with a greater preference for corticosteroids in pregnant women. Pregnant women were more likely to be admitted to ICU and receive mechanical ventilation though there was no difference in the severity of the disease between both groups. Pregnant women, due to their immunocompromised state, are more likely to experience complications of infectious diseases such as influenza, SARS, and MERS [17,18]. During the influenza A subtype H1N1 pandemic in 2009, pregnant women accounted for 5% of all H1N1-related deaths and were at an increased risk for severe disease, including hospitalization, ICU admissions, and death compared to their non-pregnant counterparts [11,25]. Similar trends were observed during the SARS and MERS outbreaks [9,26]. Lam, Chui Miu et al. [27] reported that 40% of the pregnant women affected with SARS required mechanical ventilation and had a case fatality of 30%, compared to 13% and 11% in non-pregnant individuals. Our review revealed a higher risk of ICU admissions in pregnant women; however, it did not show worsening clinical symptoms in pregnant women compared to non-pregnant women infected with COVID-19. Through the review, we found non-pregnant women to be at a higher risk of experiencing symptoms like headache, myalgia, fever, expectoration, chest tightness, wheezing, diarrhea, and anosmia compared to their pregnant counterparts. Many other studies and systematic reviews, however, reported clinical characteristics to be similar amongst COVID-19 infected pregnant and non-pregnant women [28][29][30][31][32]. Furthermore, in our review, both groups received similar supportive treatments irrespective of their pregnancy status. In concert with other studies [29,[33][34][35], most patients received oxygen therapy in addition to antiviral and antibiotic medications. Our study demonstrated a greater likelihood of corticosteroid use in pregnant women. However, despite being the most commonly reported medication in another review too [36], the use of corticosteroids for COVID-19 has generally not been recommended in pregnant women, due to the increased risk of preterm birth, low birth weight, and pre-eclampsia associated with its use in pregnancy [37,38].
Some of the limitations identified for this review are (1) a small number of studies included, (2) smaller sample size for most studies with a smaller number of pregnant women compared to non-pregnant women, (3) two isolated studies with a large sample size having a greater impact on the overall result, (4) lack of data on other significant variables such as socioeconomic status and ethnicity, (5) studies from limited developed countries, and (6) unadjusted analysis in most studies. We also identified that apart from two studies, all the other studies had small sample sizes of less than 150. There is a need for studies with a bigger sample size and a comparable number of pregnant and non-pregnant COVID-19-infected women with adjusted analysis to reach more conclusive results for the future updates of this review. Moreover, there is a need to compare data on other variables, especially demographic variables such as socioeconomic status and ethnicity. The current manuscript reports data on ethnicity from a single study and therefore, is biased towards the data from the largest sample-sized study, leading to a lack of generalizability. Future studies in the domain should also highlight whether the ICU admission or worsening state in pregnant women was more likely to be due to COVID-19 infection or a delivery complication. Multivariable analysis to identify factors associated with clinical presentation, management, and prognosis in pregnant and non-pregnant women could not be done due to insufficient data. However, if data on individual patients are provided in the future, then individual patient data meta-analysis (IPD-MA) would be the ideal approach to providing insights into recognizing and managing COVID-19 infection in pregnant women. More studies on populations across the world need to be published to prevent the chance of bias towards a particular set of people. A future update of this systematic review may then be warranted and can, therefore, help reach conclusive findings.

Conclusions
In conclusion, the findings of this study summarize the epidemiological and clinical characteristics, along with the management and prognosis of women of reproductive age with COVID-19 based on their pregnancy status. With the disease burden increasing every day, these data equip healthcare workers to better identify and monitor the patients who are more susceptible to the disease and to make informed decisions when treating the patients.
Author Contributions: All authors contributed adequately towards this systematic review and metaanalysis and have been provided authorship on that basis. D.S.A.K., A.N.P. and Z.S.L. conducted the literature search, full text review, analysis, and wrote the manuscript. A.A. worked on analysis and figures for the study. J.K.D. and R.A.S. edited the manuscript and provided feedback. All authors have read and agreed to the published version of the manuscript.
Funding: This research did not receive any funding.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.