Indirect impact of SARS‐CoV‐2 pandemic on pregnancy and childbirth outcomes: A nine‐month long experience from a university center in Lombardy

Abstract Objective To determine the impact on perinatal health of changes in social policies and obstetric care implemented to curb SARS‐CoV‐2 transmission. However, robust data on the topic are lacking since most of the studies has examined only the first few months of the outbreak. Methods A retrospective analysis of prospectively collected data on uninfected and asymptomatically infected women giving birth between March and November 2020 and in the same time frame of 2019 at our tertiary care center in Lombardy, northern Italy. Perinatal outcomes were compared according to the year (2019 versus 2020) and to the trimester (March–May, June–August, September–November) of childbirth, corresponding to the three phases of the pandemic (first wave, deceleration, second wave) and covering a 9‐month period. Results We identified increased rates of gestational diabetes mellitus, spontaneous preterm birth, and neuraxial analgesia in 2020 versus 2019, with different temporal distributions: gestational diabetes mellitus and spontaneous preterm birth were more prevalent during the deceleration and the second wave phase, whereas epidural analgesia was more prevalent during the first wave. Conclusion By assessing a prolonged time frame of the pandemic, we show that pandemic‐related control measures, as applied in Lombardy, impacted relevant perinatal outcomes of women giving birth at our center.


| 467
ORNAGHI et Al. specific recommendations for maternity services were rapidly issued. 4,7,9 A few tertiary-care hospitals, including the research site, were chosen as referral centers for SARS-CoV-2-infected obstetric patients. 4 In-person visits were maintained only if deemed necessary, and remotely delivered visits and childbirth preparation classes were implemented. 12,13 Skin-to-skin and birth companions were not allowed in SARS-CoV-2-positive women till May and October 2020, respectively. Breastfeeding was permitted with the use of facemask and gloves in the case of confirmed infection. One support person was allowed during the postpartum stay only in SARS-CoV-2-negative women for 2 h a day, whereas in-hospital visits by family members were suspended. 14 Although drafted to provide guidance and safety in the context of a constantly evolving situation and knowledge, these recommendations influenced crucial aspects of obstetric care. 1, 10,15,16 Direct effects of the SARS-CoV-2 infection on pregnant women and their babies have been extensively investigated, with findings unanimously suggesting that moderate-to-severe symptomatic infection associates with adverse obstetric outcomes, such as preterm birth (PTB) and low birth weight. [17][18][19][20][21][22][23][24][25][26][27] In turn, the literature examining whether the changes in social and obstetric care policies due to the pandemic have affected perinatal outcomes in uninfected and asymptomatically infected women is more controversial. Some studies have suggested that stay-at-home orders may have contributed to decreased rates of PTB, [28][29][30][31][32][33][34][35][36][37][38] although available evidence is discordant. [39][40][41][42] Similarly, contrasting reports have been published regarding the risk of stillbirth. 38,40,41,[43][44][45][46] Of note, most of the research on the topic has focused only on the first few months of the pandemic, so has probably been unable to capture all the relevant outcomes and potential temporal trend differences.
The aim of this study was to investigate perinatal outcomes among uninfected and asymptomatically infected women who gave birth during a 9-month time frame of the 2020 pandemic in a referral center for SARS-CoV-2 infection in obstetric patients in Lombardy, northern Italy, and to compare them to those of women in 2019.
We hypothesized that the modifications in women's life-style, maternity services, and labor and birth practices due to the pandemic, as well as the protracted increased workload of midwives and obstetricians working in a referral center for SARS-CoV-2 infection, ultimately affected perinatal outcomes.

| MATERIAL S AND ME THODS
This was a retrospective analysis of prospectively collected data including women giving birth between March 1 and November 30, 2020 and in the same time frame of 2019, at our tertiary-care center.
Starting in March 2020, a comprehensive questionnaire was administered to all women at hospital admission. 47 A targeted SARS-CoV-2 screening approach triggered by a positive questionnaire and based on reverse transcription polymerase chain reaction testing of nasopharyngeal swabs was used until April 8, 2020, when a universal viral screening policy was implemented. Healthcare personnel caring for suspected and confirmed SARS-CoV-2-positive women in labor wore full personal protective equipment, including N95 facemask, gloves, goggles, and gown.
Data regarding general maternal characteristics, obstetric history, pregnancy course, and perinatal outcomes of women giving birth at our center are routinely recorded in a dedicated log book, which is periodically audited as part of our institutional quality improvement program aimed at safely reducing rates of obstetric interventions, including primary cesarean section (CS) 48 and episiotomy. 49 PTB was defined as any birth occurring before 37 weeks of pregnancy. It was categorized as spontaneous and medically indicated.  Women in 2020 were older, mostly more than 40 years old, and more frequently multiparas with a previous CS in their obstetric history compared with the women in 2019 (Table 1) Approximately one-third of women used neuraxial analgesia in labor during the 2 year-study period, with higher rates in 2020 versus 2019. An even more substantial increase was identified in 2020 when analysis was restricted to only those women with no previous vaginal birth.

Globally, frequencies of abdominal and vaginal operative birth
were not affected by the pandemic; however, we observed a reduction in CS rate from 2019 to 2020 among women in Robson class 1 (8.2% versus 6.6%, P = 0.04). Similarly, episiotomy rates displayed a substantial decline (9.7% versus 6.7%, P = 0.001).
All findings of the univariate analysis, except for polyhydramnios, were confirmed by the logistic regression model adjusted for potential confounding factors (Table 3).
Analyses according to the trimester of birth identified temporal differences in the assessed perinatal outcomes, with increased rates of GDM and spontaneous PTB during the deceleration and the second wave phase of the pandemic, and of epidural analgesia during the first pandemic wave (Figure 2).
A sensitivity analysis including only SARS-CoV-2-negative women showed results similar to the analyses of the overall study population (see Tables S1-S3).

| DISCUSS ION
Our data support the hypothesis that modifications in social policies and in labor and birth practices abruptly introduced to contain viral spread at the beginning of the outbreak influenced perinatal outcomes of uninfected and asymptomatically infected women giving birth in a referral center for SARS-CoV-2 infection in Lombardy, northern Italy, over a prolonged period of time.
We observed increased rates of GDM, spontaneous PTB, and use of neuraxial analgesia in labor in 2020 versus 2019. These outcomes displayed a different temporal distribution, with GDM and spontaneous PTB being more prevalent during the deceleration and the second wave phase and epidural analgesia being more prevalent during the first wave phase. In turn, we identified a substantial reduction in the 2020 rates of obstetric interventions, including CS among women in Robson class 1 and episiotomy. Also, high levels of prenatal stress and anxiety, which have been reported in pregnant women exposed to the pandemic-related lock- Withdrawal of epidural services has been reported in some of the most affected Italian regions. 52 Although being a referral center, we were able not only to guarantee neuraxial analgesia in labor as we did in 2019 but also to face an increased request for this during the first wave. Substantial levels of stress and anxiety, due to the limited knowledge of the effects of the virus on the   This study is not without limitations. First, the retrospective nature of the design prohibited us from establishing the causality of the association. Second, it is possible that this study was underpowered to assess differences in less common but severe adverse outcomes, including stillbirth. Third, our findings are from a single SARS-CoV-2 referral center, and may not be generalizable to different settings.

TA B L E 3 Logistic regression model of perinatal outcomes of interest
Fourth, it is possible that some asymptomatically infected women might have been missed before universal SARS-CoV-2 screening was implemented on April 8, 2020; however, this is unlikely to have biased our findings because of the short time frame of universal testing unavailability compared with the entire study period, as well as of the high accuracy of the admission questionnaire guiding the targeted viral screening. 47 Finally, we did not measure levels of stress and anxiety or of physical exercise among our cohort of pregnant women.
In conclusion, we continue to live through this pandemic. Our research work, encompassing a 9-month time frame of the outbreak, provides a unique insight into its indirect effects on perinatal health.
Importantly, our data suggest that these effects may vary according to the trimester of pregnancy in which women are when strict stayat-home orders and changes in labor and birth care pathways were implemented.
Ongoing review of maternity statistics is warranted to remain vigilant for newly developing trends, in order to provide up-to-date evidence to optimally guide our service organization. In addition,

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
SO contributed to the design, planning, and conduct of the study, performed the data analysis, wrote the manuscript, and contributed to its review. SF, AN, and PV contributed to the design, planning, and conduct of the study, and to manuscript review. CKGM, GB and FI contributed to the conduct of the study and to manuscript review.
AN: design, planning, and conduct of the study, manuscript reviewing.