key: cord-1038993-srv7jv8r
authors: Giesbers, Steven; Goh, Edwina; Kew, Tania; Allotey, John; Brizuela, Vanessa; Kara, Edna; Kunst, Heinke; Bonet, Mercedes; Thangaratinam, Shakila; Chatterjee, Shaunak; Gae, Andrea; Stallings, Elena; Yap, Magnus; Sheikh, Jameela; Lawson, Heidi; Coomar, Dyuti; Dixit, Anushka; Zhou, Dengyi; Balaji, Rishab; Littmoden, Megan; King, Yasmin; Debenham, Luke; Clavé Llavall, Anna; Ansari, Kehkashan; Sandu, Guirmaan; Banjoko, Adeolu; Fraser, Helen; Rajah, Tanisha; Ramkumar, Anoushka; Khashaba, Alya; Attarde, Shruit; Walker, Kate; Thornton, Jim; van Wely, Madelon; van Leeuwen, Elizabeth; Kostova, Elena; Khalil, Asma; Tiberi, Simon; Broutet, Nathalie; Rahn Kim, Caron; Thorson, Anna; Oladapo, Olufemi T; Zamora, Javier; Mofenson, Lynne
title: Treatment of COVID-19 in pregnant women: a systematic review and meta-analysis
date: 2021-10-20
journal: Eur J Obstet Gynecol Reprod Biol
DOI: 10.1016/j.ejogrb.2021.10.007
sha: 67acc2c84675ad0040b1251a99499b8990612ddc
doc_id: 1038993
cord_uid: srv7jv8r

OBJECTIVE: Clinical trials evaluating pharmacological and non-pharmacological treatment of COVID-19, either excluded pregnant women or included very few women. Unlike the numerous systematic reviews on prevalence, symptoms and adverse outcomes of COVID-19 in pregnancy, there are very few on the effects of treatment on maternal and neonatal outcomes in pregnancy. We undertook a systematic review of all published and unpublished studies on the effects of pharmacological and non-pharmacological interventions for COVID-19 on maternal and neonatal pregnancy outcomes. DATA SOURCES: We performed a systematic literature search of the following databases: Medline, Embase, Cochrane database, WHO (World Health Organization) COVID-19 database, China National Knowledge Infrastructure (CNKI), and Wanfang databases from 1 December 2019 to 1 December 2020. STUDY ELIGIBILITY CRITERIA: Studies were only included if they involved pregnant or postnatal women who were exposed to pregnancy specific interventions like the mode of delivery and type of anaesthesia, pharmacological or non-pharmacological interventions. STUDY APPRAISAL AND SYNTHESIS METHODS: We first screened the titles and abstracts of studies and then assessed the full text of the selected studies in detail for eligibility. Data on study design, population, type of screening for COVID-19, country, hospital, country status (high or low and middle income), treatment given (mode of delivery, type of anaesthesia, type of pharmacological and non-pharmacological treatment was extracted. The pre-defined maternal outcomes we collected were mode of delivery (vaginal or by caesarean section), severe or critical COVID-19 (as defined by the authors), symptomatic COVID-19, maternal death, maternal hospital admission, ICU admission, mechanical ventilation, ECMO and maternal pneumonia. The pre-defined neonatal outcomes we extracted were preterm birth (<37 weeks), stillbirth, neonatal death, NICU admission, neonatal COVID-19 positive, neonatal acidosis (pH<7.0) and Apgar scores (<8 after 5 minutes). Study quality assessment was performed. RESULTS: From a total of 342 potential eligible studies, we included 27 studies in our systematic review, including 4943 pregnant women (appendix 3). Sixteen studies had a retrospective cohort design and 11 a prospective cohort design. There were no randomised controlled trials. There was a significant association between caesarean section and admission to ICU (OR 4.99, 95% CI 1.24 to 20.12; 4 studies, 153 women, I(2)=0%), and diagnosis of maternal COVID-19 pneumonia as defined by study authors (OR 3.09, 95% CI 1.52 to 6.28; 2 studies, 228 women, I(2)=0%). Women who had a preterm birth were more likely to have the baby via caesarean section (OR 3.03, 95% CI 1.71 to 5.36, 12 studies; 314 women, I(2)=0%). For pharmacological and non-pharmacological we provided estimates of the expected rates of outcomes in women exposed to various treatment of COVID-19. Comparative data for pregnant women, in particular for treatments proven to be effective in the general population, however, is lacking to provide clinically meaningful interpretation. CONCLUSIONS: We found associations for pregnancy specific interventions, like mode of delivery and outcomes of the disease, but there were too few data on pharmacological and non-pharmacological treatments in pregnant women with COVID-19. We report the rates of complications found in the literature. We encourage researchers to include pregnant women in their trials and report the data on pregnant women separately.

Studies were only included if they involved pregnant or postnatal women who were exposed to pregnancy specific interventions like the mode of delivery and type of anaesthesia, pharmacological or non-pharmacological interventions.

We first screened the titles and abstracts of studies and then assessed the full text of the selected studies in detail for eligibility. Data on study design, population, type of screening for COVID-19, country, hospital, country status (high or low and middle income), treatment given (mode of delivery, type of anaesthesia, type of pharmacological and non-pharmacological treatment was extracted. The pre-defined maternal outcomes we collected were mode of delivery (vaginal or by caesarean section), severe or critical COVID-19 (as defined by the authors), symptomatic COVID-19, maternal death, maternal hospital admission, ICU admission, mechanical ventilation, ECMO and maternal pneumonia. The pre-defined neonatal outcomes we extracted were preterm birth (<37 weeks), stillbirth, neonatal death, NICU admission, neonatal COVID-19 positive, neonatal acidosis (pH<7.0) and Apgar scores (<8 after 5 minutes). Study quality assessment was performed.

From a total of 342 potential eligible studies, we included 27 studies in our systematic review, including 4943 pregnant women (appendix 3). Sixteen studies had a retrospective cohort design and 11 a prospective cohort design. There were no randomised controlled trials. There was a significant association between caesarean section and admission to ICU (OR 4.99, 95% CI 1.24 to 20.12; 4 studies, 153 women, I 2 =0%), and diagnosis of maternal COVID-19 pneumonia as defined by study authors (OR 3.09, 95% CI 1.52 to 6.28; 2 studies, 228 women, I 2 =0%). Women who had a preterm birth were more likely to have the baby via caesarean section (OR 3.03, 95% CI 1.71 to 5.36, 12 studies; 314 women, I 2 =0%). For pharmacological and non-pharmacological we provided estimates of the expected rates of outcomes in women exposed to various treatment of COVID-19. Comparative data for pregnant has resulted in very little information on maternal and neonatal outcomes in pregnant women who have been exposed to these interventions. Unlike the numerous systematic reviews on prevalence, symptoms and adverse outcomes of COVID-19 in pregnancy, there are very few on the effects of treatment on maternal and neonatal outcomes in pregnancy.

To fill in the evidence gap, we undertook a systematic review and meta-analysis of all published and unpublished studies on the effects of pharmacological and nonpharmacological interventions for COVID-19 on maternal and neonatal pregnancy outcomes.

This systematic review is part of an ongoing set of living systematic reviews on COVID-19 in pregnancy, using a prospectively registered protocol (PROSPERO CRD42040178076; registered 22 April 2020) published elsewhere.(4) In this paper we specifically report the effects of pharmacological and non-pharmacological interventions on pregnancy outcomes. We also report on the complication rates of pregnancy specific interventions, such as mode of delivery and type of anaesthesia.

We carried out our systematic review using the preferred reporting items for systematic reviews and meta-analysis (PRISMA) recommendations (see appendix (1)

Two reviewers (SG and EG) independently selected studies using a two-stage process: they first screened the titles and abstracts of studies and then assessed the full text of the selected studies in detail for eligibility. Studies were included if they involved pregnant or recently pregnant women including postnatal women who were exposed to pregnancy specific interventions like the mode of delivery and type of anaesthesia, pharmacological or non-pharmacological interventions.

Pharmacological intervention includes antiviral, immunotherapy, (systemic) corticosteroids, antibiotics or combinations of these interventions and nonpharmacological interventions comprised mechanical ventilation, extracorporeal membrane oxygenation (ECMO) or proning. The studies also needed to report on COVID-19 related pregnancy or neonatal outcomes. Pregnancy outcomes were severe or critical COVID-19 (as defined by the individual study authors), symptomatic COVID-19, maternal death, maternal hospital admission, ICU admission, mechanical ventilation, ECMO, maternal pneumonia, preterm birth (<37 weeks), caesarean section, stillbirth, neonatal death (up to 28 days), neonatal intensive care unit (NICU) admission, neonatal COVID-19 positive, neonatal acidosis (pH <7.0) and Apgar scores (<8 after 5 minutes). Disagreements were resolved through discussion with a third reviewer (ST or JA). We included only cohort studies or case-series that reported on more than 10 women. We excluded studies that reported on duplicate data for the outcomes of interest when this was reported by the authors or when we found that the characteristics of the studies were similar to each other.

To assess the quality of comparative cohort studies for selection, comparability and outcome ascertainment bias we used the Newcastle Ottowa Scale.(5) At any time, two reviewers independently assessed the quality of the studies (SG, EG, TK). For internal validity we considered studies to be of low risk of bias when the data was collected directly from the subjects, the outcomes of interest were clearly defined, the data was collected from medical records and done in the same manner for all the subjects, the follow-up time was long enough to report the outcome and the numerators and denominators for the outcomes reported were appropriate. For external validity a study was considered to be of low risk when the studies target population closely represented the national population, universal testing was used, instead of selected testing, there was no form of random selection used to select the sample and the response rate for the study was higher than 90%.

Two reviewers (SG, EG) independently extracted data using a predefined format. In all studies we extracted data on the study design, the population, type of screening for COVID-19, country, hospital, country status (high or low and middle income), treatment given (mode of delivery, type of anaesthesia, type of pharmacological and non-pharmacological treatment and their definition. The pre-defined maternal outcomes we collected were severe or critical COVID-19 (as defined by the authors), symptomatic COVID-19, maternal death, maternal hospital admission, ICU admission, mechanical ventilation, ECMO and maternal pneumonia. The pre-defined neonatal outcomes we extracted were preterm birth (<37 weeks), mode of delivery (vaginal or by caesarean section), stillbirth, neonatal death, NICU admission, neonatal COVID-19 positive, neonatal acidosis (pH<7.0) and Apgar scores (<8 after 5 minutes). We did a deduplication process by checking the data with other studies published by the same authors or where data was collected in the same hospitals.

We contacted study authors if there were any inconsistency in their data or where data was missing. Disagreements were discussed with a third reviewer (JA)

The comparative dichotomous data assessing the association of exposures to outcomes were pooled using random-effects meta-analysis based on the Mantel-Haenszel estimation of between-study variance. The findings were summarized as odds ratios (OR) with 95% confidence intervals (CI). Heterogeneity was measured using the I-squared statistic. When the I 2 is 80% or more its meant to have high heterogeneity, less then 50% low heterogeneity, 50-80% moderate heterogeneity.

From a total of 342 potential eligible studies, we included 27 studies in our systematic review. (Figure 1 )

Of 27 studies, six (22%) were from China (6-11); five (19%) were from Italy (12-16); three from Spain (17) (18) (19) ; two each from Chile (20, 21) , Turkey (22, 23) and the United States of America (24, 25) ; one each from Brazil (26), France (27) , India (28), Israel (29) , Mexico (30), Panama (31) and Peru (32) . Fifteen were classified as highincome countries (12-21, 24, 25, 27, 29, 31 ) and 12 as low and middle-income countries (6-11, 22, 23, 26, 28, 30, 32) . Sixteen studies had a retrospective cohort design (6-10, 14, 17-22, 24, 28, 29, 32) and 11 a prospective cohort design (11-13, 15, 16, 23, 25-27, 30, 31) . There were no randomised controlled trials. Sixteen studies reported data on admitted women with COVID-19, eight reported data on all pregnant women and three reported data on a selected group of women, such as pregnant women with hypertension. Seven studies performed universal screening and testing to assess for COVID-19, nine studies did symptom-based testing, four studies did risk-based testing on the basis of epidemiological history and clinical Detailed information can be found in appendix 3.

Evaluation of study quality using the Newcastle Ottawa scale was overall low for 26 out of 27 studies. The risk of bias for study selection was low for 26 out of 27 studies, with one study scoring medium due to assessed outcome perceived to be present at study inception. Eleven out of 27 studies had a low risk of bias for the comparability of cohorts, on the basis of both design and selection. Thirteen studies had a medium and three a high risk of bias respectively for comparability. Risk of bias of study outcome was low in 25 out of 27 studies and medium in two out of 27 studies, due to either inadequate follow-up length or incomplete accountability of outcomes for all subjects at study termination.

Eighteen studies provided data on mode of delivery and the relation with maternal and/ or neonatal outcomes (1020 women). (Figure 2 ) There was a significant association between caesarean section and admission to ICU (OR 4.99, 95% CI 1.24 to 20.12; 4 studies, 153 women, I 2 =0%), and diagnosis of maternal COVID-19 pneumonia as defined by study authors (OR 3.09, 95% CI 1.52 to 6.28; 2 studies, 228 women, I 2 =0%), although in one of the two studies reporting on pneumonia it was not clear if this was COVID-19 pneumonitis. There were no associations between mode of delivery and severe COVID-19.

There was a significant association between mode of delivery and preterm birth.

Women who had a preterm birth were more likely to have the baby via caesarean section (OR 3.03, 95% CI 1.71 to 5.36; 12 studies, 314 women, I 2 =0%). There were no associations between mode of delivery and a low Apgar score, neonatal acidosis, specified. The number of pregnant women exposed to intervention was small despite the big denominators, hence it was not possible to do a meta-analysis and make any conclusions about pharmacological interventions for treatment of pregnant women with COVID-19. (Table 1) 

Eleven studies reported on non-pharmacological interventions, of which mechanical ventilation was reported in 6 studies and oxygen administration in 8 studies. In total 1738 women were included in these studies and 240 were exposed to interventions, 28 patients had mechanical ventilation and 212 had oxygen administration. (Table 2) There were no studies reporting on proning during pregnancy and no studies were found that report on extracorporeal membrane oxygenation (ECMO), haemodialysis or inotropic treatment.

All interventions, either pregnancy specific or COVID-19 related interventions in pregnant women diagnosed with the disease were poorly reported. None of the randomised trials reported outcomes specific for pregnant women. The RECOVERY Trial included outcomes however this has yet to be published, Pregnancy-specific interventions, such as the mode of delivery or the type of anaesthesia appears to be related with severity of disease, but not with perinatal outcomes. We are unable to ascertain the temporality on all cases. There is an association between caesarean section and the increased likelihood of being admitted to ICU and increased odds of COVID-19-related maternal pneumonia.

We provide estimates of the rates of outcomes in women exposed to various pharmacological treatment for COVID-19 and non-pharmacological interventions such as oxygen administration and ventilation.Comparative data for pregnant women, in particular for treatments proven to be effective in the general population, however is lacking to provide clinically meaningful interpretation.

To our knowledge this is the first systematic review looking into management and treatment of pregnant women with COVID-19. We did this in a structured manner and included not only pharmacological treatment, but also non-pharmacological and pregnancy specific interventions, such as mode of delivery. We also assessed at the quality of the studies. We refrained conducting a meta-analyses for non-comparative cohorts. We set strict inclusion and exclusion criteria for the selection of papers. The studies we have included are clear about the included women, the treatment given, and the outcomes reported. If the relationship between treatment and outcome was not clear, the paper would be excluded. Extensive collaboration and capturing of data through different databases allowed for a big pool of studies to be reviewed.

There were no language restrictions. We were limited by the paucity of the data and the heterogeneity in the studies. We could not establish the temporality for some of the interventions such as caesarean section as it is possible that some women admitted to ICU for severe COVID-19 might have had caesarean section for maternal reasons. There were no randomised controlled trials, and it is very likely that the intervention was influenced by the characteristics of the participant, the setting and the availability of resources as most of the trials were in high income settings. There is also the issue of generalisability, and therefore the rate of outcomes may not reflect the effect of treatment, but it could be more indicative of the underlying severity of the disease of the mother. We were also restricted with performing meta-analyses for most of the data due to the small number of reported treatment and outcomes. We could only provide narrative reviews for those interventions.

Comparison with existing evidence Pregnant women are known to be more often affected by severe COVID-19 than women in the general population.(1) In our systematic review, we found an association between having a caesarean section and being admitted to ICU or having COVID-19 pneumonia. This, however, may be influenced by pregnancy. By performing a caesarean section, the concern of impaired lung capacity due to the gravid uterus is diminished. Delivery allows for more postpartum treatment options and minimises the risk of causing harm to the fetus in-utero and ventilation is also easier postpartum.

The association we found between mode of delivery and preterm birth is also likely to reflect the needs for more intensive treatment. The PregCOV-19 Living Systematic

Review estimated the risk of preterm birth at approximately 17%, of which approximately 94% were iatrogenic. (1) The Royal College of Anaesthetists (RCoA) issued recommendations for types of anaesthesia to be offered to pregnant women with COVID-19.To date however,no trials have been conducted in regards to management or outcomes for such women. The same treatment principles apply to pregnant women as to non-pregnant patients with regards to non-pharmacological interventions, such as oxygen supplementation or mechanical ventilation. A very gravid uterus can cause difficul ventilation and the need for supplemental oxygen. This is due to the increased demand of oxygen in pregnancy due to the higher metabolic rate and the increased consumption of oxygen. (39) Mechanical ventilation is more difficult when a woman is pregnant, due to the gravid uterus, lung capacity can be impaired and it can be difficult to get the required volumes to support adequate ventilation. (40, 41) In the presence of ARDS, proning has been proven to help ventilate patients. (42) This cannot be practiced when a patient has a wound of a caesarean section or when she is over 34 weeks pregnant, the heavily pregnant uterus can make this position more difficult. In summary, there is a need for more data involving pregnant women in clinical trials.

Where trials are not available, more data is needed on the outcomes when drugs are given in clinical practice. Interventions and outcomes as shown in this article do appear to be associated with the severity of the disease. There is a paucity of data with regards to pregnant and postnatal women; clinical trials needs to include pregnant women.

The Authors state that they do not have any conflict of interest. 

Protocol and registration 5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web a registration information including registration number.

Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report chara language, publication status) used as criteria for eligibility, giving rationale.

Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contac additional studies) in the search and date last searched.

Search 8 Present full electronic search strategy for at least one database, including any li repeated.

Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in sys included in the meta-analysis).

Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independen for obtaining and confirming data from investigators.

List and define all variables for which data were sought (e.g., PICOS, funding so simplifications made.

12 Describe methods used for assessing risk of bias of individual studies (including done at the study or outcome level), and how this information is to be used in an Summary measures 13 State the principal summary measures (e.g., risk ratio, difference in means).

Synthesis of results 14 Describe the methods of handling data and combining results of studies, if done (e.g., I 2 ) for each meta-analysis.

Page 1 of 2

Risk of bias across studies 15 Specify any assessment of risk of bias that may affect the cumulative evidence reporting within studies).

Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, which were pre-specified.

17 Give numbers of studies screened, assessed for eligibility, and included in the re each stage, ideally with a flow diagram.

Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., stud provide the citations.

Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) sim intervention group (b) effect estimates and confidence intervals, ideally with a fo For more information, visit: www.prisma-statement.org.

Page 2 of 2 Appendix 1: Quality assessment for risk of bias using the Newcastle Ottawa Scale  Pregnancy-specific interventions, such as the mode of delivery or the type of anaesthesia appears to be related with severity of disease, but not with perinatal outcomes.

 There is an association with caesarean section and the likelihood of being admitted to ICU or due to adverse effects of COVID-19 pneumonia.

 We encourage researchers to include pregnant women in their trials and when they did so, to report them separately. Vigil-De Gracia P (2), 2020 Panama

Clinical manifestations, risk factors, and maternal and perinatal outcomes of coronavirus disease 2019 in pregnancy: living systematic review and meta-analysis

Inclusion of pregnant women in COVID-19 treatment trials: a review and global call to action

Clinical manifestations, prevalence, risk factors, outcomes, transmission, diagnosis and treatment of COVID-19 in pregnancy and postpartum: a living systematic review protocol

editor The NewcastleOttawa Scale (NOS) for assessing the quality of nonrandomised studies in metaanalyses

Clinical analysis of ten pregnant women with COVID-19 in Wuhan, China: A retrospective study

Clinical Characteristics of Pregnant Women with Covid-19 in Wuhan, China

Safety and efficacy of different anesthetic regimens for parturients with COVID-19 undergoing Cesarean delivery: a case series of 17 patients

Clinical presentations and outcomes of SARS-CoV-2 infected pneumonia in pregnant women and health status of their neonates

Anaesthetic managment and clinical outcomes of parturients with COVID-19: a multicentre, retrospective, propensity score matched cohort study

An Observational Study of Intrauterine Vertical Transmission of SARS-CoV-2 in 20 Neonates

Clinical Findings and Disease Severity in Hospitalized Pregnant Women With Coronavirus Disease 2019 (COVID-19)

Coronavirus and birth in Italy: results of a national population-based cohort study

Vaginal delivery in SARS-CoV-2-infected pregnant women in Northern Italy: a retrospective analysis

Detection of SARS-CoV-2 in Milk From COVID-19 Positive Mothers and Follow-Up of Their Infants

Safe Perinatal Management of Neonates Born to SARS-CoV-2 Positive Mothers at the Epicenter of the Italian Epidemic

Association Between Mode of Delivery Among Pregnant Women With COVID-19 and Maternal and Neonatal Outcomes in Spain

Clinical course of coronavirus disease-2019 in pregnancy

Incidence and clinical profiles of COVID-19 pneumonia in pregnant women: A single-centre cohort study from Spain

Routine screening for SARS CoV-2 in unselected pregnant women at delivery

Premature delivery in COVID-19 patients at San Juan de Dios Hospital / Parto prematuro en pacientes COVID-19 en Hospital San Juan de Dios

The clinical findings and outcomes of symptomatic pregnant women diagnosed with or suspected of having coronavirus disease 2019 in a tertiary pandemic hospital in Istanbul, Turkey

A multicenter study on epidemiological and clinical characteristics of 125 newborns born to women infected with COVID-19 by Turkish Neonatal Society

Epidemiology of coronavirus disease 2019 in pregnancy: risk factors and associations with adverse maternal and neonatal outcomes

Characteristics and Outcomes of 241 Births to Women With Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infection at Five New York City Medical Centers

Profile of pregnant and children and adolecents with COVID-19

A snapshot of the Covid-19 pandemic among pregnant women in France

Clinical profile, viral load, management and outcome of neonates born to COVID 19 positive mothers: a tertiary care centre experience from India

Safety of vaginal delivery in women infected with COVID-19

Pregnancies recovered from SARS-CoV-2 infection in second or third trimester: obstetric evolution

Is COVID-19 a risk factor for severe preeclampsia? Hospital experience in a developing country

ICM Anaesthesia COVID-19. Management of pregnant women with known or suspected COVID-19 2020

Dexamethasone in Hospitalized Patients with Covid-19

Coronavirus (COVID-19) infection and pregnancy 2021

Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial

Chloroquine or hydroxychloroquine for prevention and treatment of COVID-19

Physiology of pregnancy

Mechanical ventilation during pregnancy: sedation, analgesia, and paralysis

Prone Positioning for Pregnant Women With Hypoxemia Due to Coronavirus Disease 2019 (COVID-19)

Prone positioning in severe acute respiratory distress syndrome

Protection by Exclusion: Another Missed Opportunity to Include Pregnant Women in Research During the Coronavirus Disease 2019 (COVID-19) Pandemic

Exclusion of Pregnant Women from Clinical Trials during the Coronavirus Disease 2019 Pandemic: A Review of International Registries