CONCLUSIONS
Although births at 34+0 – 36+6 are usually considered as a homogeneous population under the definition of LP, this large prospective, observational study demonstrated significant differences in neonatal outcomes based on gestational age at delivery. We found that approximately one third of newborns at 34 weeks experienced neonatal morbidities, when compared to < 10% of the 36 weeks births (Table 4). Moreover, after controlling for confounding, adverse outcomes were 2-4 folds more common when delivery took place at 34 instead of 36 weeks’ gestation (Tables 5, 6). As also shown by the Consortium on Safe Labor [22], we identified a continuum of neonatal morbidities that inversely correlated with timing of delivery; therefore, when maternal, fetal or obstetric complications develop between 34+0 and 36+6 weeks’ gestation, obstetricians need to carefully balance the risks and benefits of birth at a specific gestational age with the consequences of pregnancy continuation beyond that time point [23].
As metabolic acidosis and the need for neonatal resuscitation reflect the fetus’ exposure to hypoxia and increased metabolic demands, we found an association with pregnancy complications rather than gestational age at delivery (Table 5, 6). De Almeida et al. [24] initially demonstrated that LP babies were at substantially increased risk for neonatal resuscitation when compared to term counterparts. Our multivariate analyses confirmed among LP newborns known risk factors for resuscitation at birth, such as delivery prompted by maternal or fetal disorders (i.e. indicated deliveries, Table 5), vaginal bleeding from placenta abruption/abnormal placentation, and non-reassuring fetal status (Table 6) [25]. These findings may help develop strategies to prepare for circumstances requiring advanced neonatal resuscitation skills, and to organize in utero transfer to Tertiary Care Centers.
In our cohort, indicated deliveries were associated with worse immediate neonatal outcomes when compared with spontaneous LP labor (Table 5), as also indicated in previous reports [8, 9]. The process of labor itself likely facilitates fetal lung maturation and improves clearance of pulmonary fluid, reducing the risk of neonatal respiratory morbidities and the need for resuscitation [26], while the underlying condition that prompted delivery may also account for poorer outcomes. Interestingly, expectantly managed pPROM had outcomes similar to spontaneous LP labor, suggesting that prompt induction of labor may not represent the only option available when rupture of membranes complicates LP gestations, as also stated in the PPROMEXIL [14] and PPROMEXIL2 [27] trials.
Our study also showed that an adverse intrauterine environment may significantly contribute to neonatal morbidity. The association between infection, inflammation and adverse neonatal outcomes may be explained by the ability of pro-inflammatory cytokines to produce the “fetal inflammatory response” [28]. Intrauterine inflammation has been demonstrated in pregnancies complicated by preeclampsia [29], maternal obesity [30], polyhydramnios [31], rupture of membranes, and chorioamnionitis [32]. Diabetes mellitus not only creates a pro-inflammatory intrauterine environment [33], but it also accounts for the damaging effects of fetal hyperglycaemia and hypoxia [34]. Accordingly, our multivariate analyses confirmed the correlation between pregestational diabetes and increasing maternal BMI with both the composite adverse neonatal outcome, and the need for respiratory support, while polyhydramnios was linked to both the composite adverse neonatal outcome and resuscitation at birth. We also showed how pregestational diabetes relates to neonatal metabolic complications, while pPROM was confirmed as a risk factor for the composite of neonatal morbidities (Table 6). Surprisingly, preeclampsia had a protective effect on neonatal complications (Table 6), suggesting that the increased antepartum surveillance once preeclampsia is diagnosed may counterbalance the risks associated with an adverse intrauterine environment
Placental ischemia and hypoxia are characterized by impairment of placental blood flow, which results in reduced delivery of oxygen and nutrients to the fetus [35]. According to our multivariate analyses, placental abruption or bleeding from abnormal placentation were associated with the composite adverse neonatal outcome, with metabolic acidosis and/or resuscitation at birth, and respiratory support as they may compromise fetal supply of oxygen and nutrients to the fetus (Table 6). Similarly, increased metabolic demands may lead to non-reassuring fetal status, that was also correlated with resuscitation at birth, the composite metabolic outcome, and respiratory support (Table 6). Insufficient intrauterine growth has been attributed to hypoxemia from placental under perfusion [36, 37], and it was found to have increased neonatal risks among preterm births [10, 11]. Our multivariate analyses confirmed such finding also in the LP population, since IUGR babies had more metabolic complications (Table 6).
It has been speculated that spontaneous onset of preterm labor may be a consequence of an earlier idiopathic activation of the normal labor process in an attempt to protect the fetus. Therefore, labor may enable the fetus to exit a potentially “hostile” in-utero environment [38]. Accordingly, our multivariate analysis showed a protective effect of spontaneous preterm labor on the risk of developing the composite metabolic outcome, as opposed to other maternal, fetal or obstetric complications.
Strengths of this study include the large sample size and multicenter nature of the cohort: both characteristics increase the generalizability of our findings. The prospective design of the survey, along with predefinition of standardized chart abstraction forms completed by Obstetricians and Pediatricians, and periodically audited by research associates, limit misclassification bias and assures data validity. We acknowledge also some limitations. The study did not classify neonatal complications according to specific obstetric, maternal or fetal indications for delivery. We deliberately chose to analyze together the outcomes of all indicated deliveries, as some disease processes such as hypertensive disorders, diabetes mellitus or intraamniotic infections may affect both mothers and babies. Moreover, we focused on specific disorders, rather than delivery indications, as those may be indirectly responsible for LP deliveries; for example, IUGR may account for non-reassuring fetal status that prompts emergent delivery, or also preeclampsia may cause placental abruption that then leads to delivery. Although inconsistencies in antenatal corticosteroids administration may reflect different opinion leaders’ viewpoints prior to the publication of the ALPS trial, we considered bethamethasone treatment among the potential confounders in our multivariate analyses.
In conclusion, gestational age at delivery, circumstances at parturition, and specific maternal, fetal, as well as obstetric complications have a significant impact on neonatal outcomes. Therefore, the decision to deliver or not during LP period should be based on the underlying conditions affecting the mother and/or the fetus, and a careful assessment of the risks of preterm delivery versus the potential benefits of expectant management should be performed. Our findings can be helpful when counselling mothers at risk of LP delivery, and can be used to plan interventions for their newborns.
DISCLOSURE OF INTERESTS : Each author declares to have no conflicts of interest.
CONTRIBUTION TO AUTHORSHIP: each author contributed significantly: FM and MB in writing paper and collecting data, GC in statistical analysis and writing paper, GG, DB, AC in the conception of the study and FF in scientific conception and principal investigator of the study.
DETAILS OF ETHICS APPROVAL: The study was approved from the Institution Review Board of the 9 Emilia Romagna Counties (n 24512 _ 25/9/2014)
FOUNDING: The study was financed by the Emilia Romagna County Grent (n 417149 _ 2014)
* Late Preterm Emilia Romagna Group: Vittorio Basevi (Health facilities, Emilia-Romagna Region, Bologna, Italy), Frusca Tiziana (Obstetrics and Gynaecology Unit, University of Parma, Parma, Italy ), Giuseppe Battagliarin (Health facilities, Emilia-Romagna Region, Bologna, Italy) , Marinella Lenzi (Maternal and Pediatrics Department, Maggiore Hospital, Bologna, Italy)., Gina Ancora (Neonatal Intensive Care Unit, Azienda Sanitaria Romagna, Infermi Hospital Rimini, Rimini, Italy ), Luigi Corvaglia (Pediatric unit, Ospedale S. Orsola- Malpighi di Bologna).
Acknowledgments: we are indebted to all collegues and research fellows for their contributions to data collection: