Prematurity is a major public health problem. Premature births represent 75% of perinatal mortality and more than half of infantile morbidity over the long term. Between 50,000 and 60,000 infants are born prematurely each year in France (1).

There are many factors of risk of prematurity. They can be related to obstetrical elements, maternal history and environmental factors. Some of the factors of risk changed during lockdown due to the SARS-CoV-2 virus, in particular imposed inactivity, atmospheric changes, changes to daily living and an increase in hygiene. France went into lockdown as of Tuesday 17th of March 2020 up to Sunday the 10th of May 2020, included. All nurseries, schools, universities have been closed as well as restaurant and business not essential. Outdoor gatherings, family or friendly reunions were no longer allowed. Teleworking has been promoted and barrier measures have been put in place (wearing a mask, social distancing, hygine promotion) (2).

The main aim of our study was to examine the association between lockdown and the birth of premature newborns in the general population at the level III maternity of the University Hospital of Nice.

Despite substantial variation in the number of hospital stays each month, fluctuation was random, a regular seasonal cycles was not observed (Figure 2). The number of premature births <37 weeks of gestation decreased continually each year in a non significant way, giving a lower frequency in 2020 (N = 227). We found that this non significant decline was mainly related to « severely premature » newborns (N = 59 in 2020 vs. N = 92 in 2017, N = 89 in 2018, N = 76 in 2019). Significant variation in the weights of the newborns was found for the different periods (Table 1). We did not find a significant decrease in either the frequency of prematurity <37 weeks of gestation or low weights at birth during lockdown compared to the average for 2017–2019 (p = 0.579 for term pregnancies, p = 0.392 for <1,000 g, p = 0.698 for <1,500 g) or date to date p = 0.689 for term pregnancies, p = 0.197 for <1,000 g, and p = 0.504 for <1,500 g). There was no significant increase in the number of stillborns during lockdown compared to no lockdown p > 0.99) (Table 2). When we compared the profiles of the mothers and newborns according to the term in lockdown vs. no lockdown we noted a significant decrease in the frequency of bronchopulmonary dysplasia of the severely premature, born before 28 weeks of gestation p = 0.049) but only when compared to the average for 2017–2019; this significant difference was not found when compared to the date to date period p = 0.076). In addition, the profiles of the mothers and newborns were comparable for the different periods (Tables 3, 4).

This study did not reveal any evidence of a significant decrease in either prematurity or low birth weights during lockdown. There was no significant increase in the number of stillbirths. There was no significant difference in the profiles of the mothers or newborns. The collection of information is limited by the retrospective nature of the study since some of the data was missing or absent, which leads to a degree of bias regarding the information. Some of the absent information concerned maternal risk factors such as the preconception body mass index, the marital status, the economic status and level of education. In addition, the study lacked power due to the low number of individuals included, despite the collection of date over four years. Thus, the monocentric nature of the study does not allow generalisation of the results, which limits the external validity. However, the strength of the study lies in the exhaustive collection of hospitalised births admitted into the critical care neonatal unit and neonatology department of the Nice University Hospital during lockdown as well as the uniformity of the data and the collection of the number of stillborns. The number of stillborns was not different between periods and did not have an impact on prematurity.

The results are in agreement with some of the published literature and in particular with some meta-analyses concerning this subject, including the meta-analyses of Vaccaro et al. (5), Chmielewska et al. (6), and Yang et al. (7). Several studies showed that lockdown did not result in a decrease in prematurity, including in France (8), Spain (9), Sweden (10), Israel (11), the United States of America (12), the United Kingdom (13), and China (14). A meta-analysis by Vaccaro et al. evaluated the impact of lockdown on prematurity, low birth weight <2,500 g and stillbirths but did not report an association between lockdown and these issues (5). Another meta-analysis by Chmielewska et al. did not find an association between lockdown and prematurity and low birth weight <2,500 g (6). A third meta-analysis by Yang et al. did not find any association between prematurity and lockdown in studies using regional/national data (7). In contrast, the initial published results were in favour of a decrease in prematurity and a low weight at birth during lockdown, notably by studies performed in spring of 2020 in Ireland (15) and Denmark (16). Other studies evaluating the association between prematurity, low birth weight and the number of stillborns performed around the world also found similar results, including in Iran (17), Australia (18), Saudi Arabia (19), Italy (20), Netherlands (21) and Austria (22). Two meta-analyses also demonstrated a decrease of prematurity during lockdown. Calvert et al. showed small reductions in preterm birth in high income and upper middle income countries during the first, second and third months of lockdown (but not in the fourth month) (23). Yao et al. identified a reduction in preterm birth during pandemic compared with pre pandemic period, but further subgroup analysis showed that there were no difference in studies from multicenter or low and middle income countries (24). According to these studies the potential positive effects of lockdown on prematurity and low birth weight were due to several factors. These included the decrease in social interaction, the shutdown of schools, the waring of masks and an increase in hygiene, which may have decreased the risk of contact with pathogens and thus a decrease in maternal infection (15–21).

There are many factors of risk of prematurity but the ethology is sometimes not well understood. They can be maternal, obstetric, infectious or environmental. The factors of risk of prematurity that may have changed during lockdown include atmospheric pollution (due to limited journeys), rest and maternal stress (with more time at home, the set up of work from home, more family support with the partner at home and certain financial assistance from governments). However, these factors of risk remain debatable and have yet to be proven. Atmospheric pollution, which may have decreased during lockdown, is a controversial factor of risk since it depends on the type of particule studied and the trimester of exposure (25, 26). With respect to rest, a systematic review by Cochrane (27) published in 2015, did not find evidence to show that bedrest reduced prematurity, on the contrary it may have negative effects such as an increase in demineralisation of bone and deconditioning during exercise or an increased risk of deep vein thrombosis (28). With respect to maternal stress, this is a subjective element with different definitions according to the studies so it is difficult to establish an association with perinatal issues such as prematurity. While it is qualified as a risk factor some studies do not report any difference and even report a decrease in prematurity (29). There also exists a hypothesis suggesting that stressful events do not have the same impact when experienced in the first, second or third trimester of pregnancy, with a higher degree of stress at the beginning of pregnancy (30).

It should be noted that preventive measures exist to reduce prematurity (31, 32). There are three types: primary prevention that concerns all women, secondary prevention to reduce and eliminate already existing risks and tertiary prevention to improve the outcome of infants born prematurely. Tertiary prevention has been the most developed in recent years, with the set up of networks of organisations providing perinatal care and health care to mothers and their newborns in adapted maternities, of antenatal corticosteroid therapy use and administration of magnesium sulphate.

The objective of our study was to examine the consequences of lockdown in France due to the SARS-CoV-2 pandemic on prematurity before 37 weeks of gestation in the level III maternity of the Nice University Hospital. We did not find evidence of an association between lockdown and prematurity, which is in agreement with published meta-analyses. Certain factors of risk such as atmospheric pollution, rest and maternal stress that were discussed during lockdown are debatable and their involvement remains to be demonstrated. The identification of the factors of risk of prematurity and the preventive measures are still a major public health issue.