Background
In France, while most babies are delivered at hospital, emergency medical services (EMS) weekly manage calls for unplanned out-of-hospital births [
1‐
4]. Despite health system organizations variations, unplanned out-of-hospital birth can be defined as birth without midwife and medical care, or optimal health care conditions [
5]. This specific context must be discriminated from planned out-of-hospital births, home births or freestanding birthing centers, where midwife management is performed [
6]. Globally, unplanned out-of-hospital births prevalence is estimated to be 0.19 to 0.61% of all deliveries [
7,
8].
Out-of-hospital delivery is associated with unfavorable perinatal outcome and increased mortality [
9‐
11]. Hypothermia is the most frequently described adverse outcome [
7,
8]. However, knowledge is limited by the small sample size of the previous studies, and most of them had insufficient power to accurately assess adverse events [
12,
13]. Thus, perinatal outcome in unplanned out-of-hospital births remains unclear and large multicentric cohorts are needed to examine perinatal morbidity and mortality and their determinants.
The aim of our study was to describe neonatal morbidity and mortality, defined as death or neonatal intensive care unit hospitalization at Day 7, in a prospective multicentric cohort of unplanned out-of-hospital births.
Discussion
To the best of our knowledge, this is the first study that assessed neonatal mortality and morbidity, and risk factors for adverse perinatal outcome in a large and multicenter cohort of unplanned out-of-hospital births. Here, we found that mortality was rare in unplanned out-of-hospital births (1.2%), and that multiparity, prematurity, maternal pathology and hypothermia were independent predictive factors for poor neonatal outcome.
All emergency medical services (EMS) manage calls for unplanned out-of-hospital births
1–4. Prevalence is estimated to be 0.61% of all deliveries in the United States [
7], 0.19 and 0.42% in two different studies in France, and 0.10% in Finland [
3,
8,
11]. Most of previous works investigated out-of-hospital births without discriminating planned or unplanned births [
16‐
21]. However, unplanned out-of-hospital births may be associated with higher unfavorable perinatal outcome and increased mortality [
22]. By focusing on the management of unplanned births in the French healthcare system, AIE delivered for the first time epidemiological data powered for a robust analysis of perinatal morbidity and mortality outcome. Three previous studies reported unplanned out-of-hospital births management in small cohorts. McLelland et al. evaluated, in a retrospective data analysis collected via the Victorian Ambulance Clinical Information System, during a one year-period, 324 out-of-hospital births including 190 before paramedics’ arrival. In line with our results, mother had a mean age around 30 years and were at term. Similarly to our study, most of the births (88.3%) were uncomplicated births in multiparous women. Obstetric complications included postpartum hemorrhage (6.5%), breech (1.3%), cord prolapse (0.6%), and prematurity (11%) [
23]. Scott and Esen reported 14 cases of unplanned out of hospital births, which occurred over a three-year period, with a reported incidence of 0.31% [
24]. All the women were multiparous, without any context of maternal or fetal pathology. As described by us and McLelland et al. [
23], most of the births (79%) occurred between the hours of 20.00 and 08.00 [
24]. Moreover, Flanagan et al. reported in a retrospective analysis of 192 unplanned out-of-hospital births, that 21% of the newborns had an Apgar score scored ≤7 out of 10, whereas in our cohort, it represented only 14% of the newborns [
25]. Contrary to previous reports, a considerable proportions of mothers in our cohort (23.6%) had a midwife consultation in the 24 previous hours. For instance, Flanagan et al. reported that only 2.4% (
n = 15) of women who birthed before arrival stated that they had been sent home from hospital within the 12 previous hours [
25].
In our study including only unplanned out-of-hospital births, neonatal morbidity and mortality, defined as death or neonatal intensive care unit (NICU) hospitalization at Day 7, was recorded in 106 newborns (6.3%). Neonatal mortality varies substantially between the different cohorts already reported. In a prospective case series of consecutive out-of-hospital deliveries in United States, Moscovitz et al. reported 9 neonatal deaths among 91 out-of-hospital deliveries (9.9%), with 86% occurring in the presence of paramedics [
20]. McLelland reported nine (2.7%) neonatal deaths, including three that were not viable being less than 24 weeks gestation [
23]. In France, Renesme et al. reported one neonatal death in a retrospective case–control study of 76 unplanned out-of-hospital births, and eleven admissions in NICU (14.5%) [
3]. In a retrospective Finish cohort with 67 out-of-hospital births recorded, Ovaskainen et al. found that out-of-hospital cases were more likely to be admitted to the neonatal care unit and to be treated for suspected infections and hypothermia, even if no neonatal death was reported [
11].
Risk factors for planned or unplanned out-of-hospital births have been widely explored, highlighting contribution of factors including ethnic group, age, parity, prenatal care, education, labour duration, smoking or distance to maternity center [
3,
11,
12,
20]. Perinatal morbidity and mortality in planned and unplanned out-of-hospital births was associated with several factors, including postpartum hemorrhage, puerperal complications, low birth weight, polycythemia, and hypothermia [
9,
10,
12,
13,
19]. Unsurprisingly, as reported in our cohort, prematurity worsens perinatal prognosis [
21]. Moreover, in line with previous studies, our multivariate logistic regression model highlighted the impact of hypothermia in unplanned out-of-hospital births. A retrospective, monocentric case–control study in Israel of women who underwent unplanned home or car births (
n = 90) versus in-hospital births (
n = 180), already reported that significantly more newborns delivered out of the hospital had hypothermia [
13]. Moscovitz et al. reported that hypothermia was common (47%) in the paramedic-attended deliveries [
20], and hypothermia was reported in 50% of the newborns in another retrospective study [
24]. Similarly, in another retrospective case-control study of unplanned out-of-hospital births (
n = 81) in France, Renesme et al reported that NICU admission rate was increased in case of unplanned out-of-hospital births and that the most frequent complication was hypothermia [
3].
Our results were in line with the literature but the strength of our study is to specially highlight the association between hypothermia and neonatal adverse outcome in a cohort of unplanned out-of-hospital births. Thus, we have here opportunities to improve the out-of-hospital management of unplanned births to limit the hypothermia of the newborn. We have to study the different tools we can use to keep the newborn warm. For instance, to warm the birthplace and keep if free from drafts, to place the newborn on the abdomen of the mother and dry him with a warm dry cloth and then leave the newborn on the abdomen of the mother (skin-to-skin contact) [
26], or to systematically use an incubator device to closely monitor temperature. Moreover, we have to compare all these available tools and deliver recommendations to prevent the loss of body heat of the newborns on the out-of-hospital field. To date, the best option to reduce loss of body heat of the newborns during out-of-hospital management is unknown.
Importantly, lack of prenatal care or history of poor prenatal care was infrequent in our cohort (2.9%), and was not associated with neonatal mortality and morbidity as found in previous studies [
20]. Multiple pregnancy, maternal pregnant pathology and prematurity were independently associated to adverse neonatal outcome. One option could be to systematically send a specialized neonatal prehospital EMS unit to care for the newborn when these risk factors are reported during the medical evaluation of the call by the emergency physician dispatcher. In France, such specialized units exist, managed by pediatric emergency physicians, and are equipped with materials dedicated for neonatal and premature care. Such specialized teams may efficiently take in charge neonatal distresses. Thus guidelines, randomized and controlled trials are required to investigate strategies to optimize newborns management in the prehospital area that could decrease neonatal morbidity and mortality [
27]. Healthcare system should take into account these risk factors to give the appropriate care during the out-of-hospital phase.