Perinatal outcomes after admission with COVID-19 in pregnancy: a UK national cohort study – Nature.com

Study design and oversight

The United Kingdom Obstetric Surveillance System (UKOSS) has a source population of all women giving birth or being admitted for obstetric specialist care to one of the 194 consultant-led maternity units in England, Northern Ireland, Scotland and Wales, ~720,000 maternities annually1,19,20,21. A protocol for active surveillance of pregnant women admitted to hospital with viral infection was planned in 2012, approved by the ethical review board (HRA NRES Committee East Midlands), Nottingham 1 (Ref. Number: 12/EM/0365) and hibernated (https://doi.org/10.1186/ISRCTN40092247). The protocol was activated for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) from March 1, 2020, and concluded on March 31, 2022. Routines were in place to ensure complete reporting (online supplementary)3, and follow-up information about birth outcomes up to December 31, 2022, were retrieved from clinical records until April 24, 2023. Maternal and perinatal deaths were cross-checked with the MBRRACE-UK mortality surveillance. (https://www.npeu.ox.ac.uk/mbrrace-uk). The corresponding author vouches for the accuracy and completeness of the data and reporting. Patients and public were part of the UKOSS steering committee and involved in study design, oversight, reporting and dissemination of study findings but not in the conduct of the study. The funder played no role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; nor in the decision to submit the paper for publication.

Pregnant women were included if admitted to hospital with a positive SARS-CoV-2 reverse transcriptase polymerase chain reaction (PCR) test within 7days of admission, during admission or up to 2days after giving birth. Women were further classified according to their SARS-CoV-2 symptoms (symptomatic/asymptomatic); severity of infection (mild/moderate to severe); and the dominant viral variant in UK at the time of the PCR test (wild-type March 1 to November 30, 2020; alpha December 1, 2020, to May 15 2021; delta May 16 to December 14, 2021; omicron (BA.1 and BA.2 variant) December 15, 2021, to March 31 2022)36.

Moderate to severe maternal COVID-19 was defined according to modified WHO criteria as maternal death, maternal intensive care admission, peripheral oxygen saturation below 95% at admission, pneumonia on radiological imaging or need for respiratory support (either oxygen supplementation, non-invasive ventilation (high flow nasal oxygen or continuous positive airway pressure), mechanical ventilation or extracorporeal membrane oxygenation (ECMO))37.

The sociodemographic characteristics and medical risk factors recorded were maternal age, maternal body mass index (kg/m2), employment, ethnic background, smoking, pre-existing medical conditions (no medical conditions (reference) vs asthma, hypertension, cardiac disease or diabetes), preeclampsia, gestational diabetes, parity, plurality and gestational age at admission (categorised by completed weeks into <22, 22 to 27, 28-33, 3436 and 37weeks). Vaccination status was recorded from January 2021 after vaccination was recommended for pregnant women in risk groups in the UK from December 22, 202038,39,40.

Pregnancy outcomes examined were pregnancy loss (<24weeks gestation)41, gestational age at birth, expedited birth due to COVID-19, and mode of birth (spontaneous vaginal, operative vaginal, caesarean section prior to or during labour). The perinatal outcomes were stillbirth at 24weeks gestation41, preterm birth (<34weeks and 34+0 to 36+6 weeks gestation), neonatal unit admission, and neonatal death within 7days after birth.

Percentages and frequencies were computed by symptom group, and for symptomatic women by severity of infection, dominant variant, and vaccination status. Where data were missing, percentages are presented as the proportion of cases known.

Risk ratios (RR) with 95% confidence intervals (CI) for stillbirth, preterm birth and admission to neonatal unit for births to symptomatic women were computed using symptomatic mild infection during the wild-type period as the reference category with the lowest absolute risk. Since severity and variant were known risk factors, we included a preplanned interaction analysis to assess the combined effect of these factors2,4. Several models were run: models with dominant variant only (crude RR), disease severity only (crude RR), and severity and variant along with an interaction term for both (interaction without covariates (model 1)); adjusted for selected covariates without vaccination status (model 2); and adjusted for selected covariates and vaccination status using mild infection during the alpha period as reference (model 3). The covariates were selected based on availability of information and evidence from the literature3,42,43.

Multilevel Poisson or multinomial regression model as appropriate was used with random intercept to account for clustering effect among multiple births. The multivariable model for stillbirth was adjusted for maternal age, ethnicity, employment status, body mass index, plurality, smoking, parity, medical conditions prior to or during pregnancy and gestational age at admission. The preterm birth model included the above except gestational age at admission, and the model for neonatal unit admission was adjusted for gestational age at birth, parity, and medical conditions. Statistical analyses were performed using STATA version 18 (Statacorp, TX, USA).

Study registration number: (https://doi.org/10.1186/ISRCTN40092247)

Ethics approval: HRA NRES Committee East MidlandsNottingham 1 (Ref. Number: 12/EM/0365). Information about ethnicity was self-determined according to the UK standard census categories (List of ethnic groups)GOV.UK (ethnicity-facts-figures.service.gov.uk).

Rema Ramakrishnan, Hilde M Engjom, Nicola Vousden and Marian Knight analysed the data.

The study protocol was published at the study start and is available on the UKOSS website https://www.npeu.ox.ac.uk/ukoss/completed-surveillance/covid-19-in-pregnancy.

Further information on research design is available in theNature Portfolio Reporting Summary linked to this article.

Read more:

Perinatal outcomes after admission with COVID-19 in pregnancy: a UK national cohort study - Nature.com