Placental histology predicted adverse outcomes in extremely premature neonates in Norway—population‐based study

We evaluated the role of placental pathology in predicting adverse outcomes for neonates born extremely preterm (EPT) before 28 weeks of gestation.


| INTRODUC TI ON
The survival rates after extremely preterm (EPT) birth have increased in high-income countries in recent decades to more than 80% for deliveries at 25 weeks of gestation. 1 However, EPT birth affects the development of most organ systems, and complications are inevitable. Neonates born EPT face a high risk of intracerebral haemorrhage, white matter lesions of the brain, early-onset neonatal sepsis (EONS), necrotising enterocolitis (NEC), retinopathy of prematurity (ROP) and chronic lung disease, namely bronchopulmonary dysplasia (BPD). 1,2 All of these complications may cause life-long morbidity and are related to gestational age (GA), at birth as well as birth weight. 3 Despite GA and birth weight, it is virtually impossible to predict, at birth, which EPT neonates will eventually develop short-term or long-term complications. Environmental, social and genetic determinants and intrauterine conditions are probably important, and neonates are most likely to be born with individual susceptibilities. However, strategies to identify those at risk have failed. 4 This is a clinical challenge and makes it difficult to tailor treatment and clinical follow-ups.
The link between placental dysfunction and diseases later in life is well known. 5 However, studies exploring the associations between placental pathology and outcomes in EPT neonates have shown conflicting results. 6,7 This could have been due to the small study samples and lack of standardised protocols in evaluating the placentas. The Amsterdam Placenta Consensus Group 8 has described a consensus on placental sampling and definitions of placental lesions. The aim of this study was to use the Amsterdam criteria to examine the placental characteristics in a cohort of EPT births in western Norway and to explore whether placental pathology could predict adverse neonatal outcomes.

| Participants and study design
This study was part of the large, prospective, population-based Project Extreme Prematurity study, also known as BabyPEP. It was conducted between 7 October 2010 and 24 September 2018 in the only two tertiary referral hospitals in western Norway: Haukeland University Hospital and Stavanger University Hospital. Women with threatened preterm delivery before 28 weeks of gestation were invited to take part, and the babies were included if they born before that gestational age. The placentas were sent for histological examination, and detailed prenatal, perinatal and neonatal histories, examinations and treatments were consecutively recorded in a database during the neonates' hospital stays. Magnetic resonance imaging (MRI) of the brain was performed at term-equivalent age.
The current study comprised singleton neonates born before 28 weeks of gestation at Haukeland University Hospital. We excluded multiple pregnancies and cases without placental examinations ( Figure 1).
Written consent was obtained from the parents, and the Regional Committees for Medical and Health Research Ethics (REK ID 2010-00496) approved the study.

| Definitions
GA was determined by ultrasound biometry at 17-19 weeks of gestation or the first trimester in 88.6% of cases and estimated according to in vitro fertilisation in the other 11.4%. Small for gestational age (SGA) was defined as a birthweight below the 10th percentile, according to Norwegian sex-specific birth weight centiles. 9 The birth NEC comprised cases verified by pre-operative ultrasound or X-ray, during surgery, or by autopsy. BPD was defined as being dependent on oxygen supplementation or non-invasive ventilatory support at 36 weeks of GA. 11 Brain MRI scans were scored by a paediatric neuroradiologist (SMA) and dichotomised into normal or any pathology.
Retinopathy of prematurity was graded, as defined by the Committee for Classification of Retinopathy of Prematurity, 12 and dichotomised into no or mild ROP (grade 1-2) and severe ROP (grade [3][4][5]. The Apgar score was dichotomised using a threshold of 4 at 5 min, and the lowest pH value measured in arterial blood gas during the first 24 h was dichotomised using a threshold of pH 7.0.

| Classifications of placental histological findings
The delivered placentas had been examined by perinatal pathologists as part of our routine practice following EPT birth. All the slides were

Key Notes
• This Norwegian prospective observational study evaluated the role of placental pathology in predicting adverse outcomes for 123 neonates born extremely preterm before 28 weeks of gestation.
• Histologic chorioamnionitis was associated with necrotising enterocolitis, and it was also associated with bronchopulmonary dysplasia and brain pathology if there was a positive foetal inflammatory response.
• The only neonatal outcome that maternal vascular malperfusion was associated with was low birthweight.

| Statistics
Continuous variables were tested for normal distribution and compared using the independent samples t-test or the Mann-Whitney U test, as appropriate. Associations between categorical variables were evaluated using the Pearson's chi-square test or Fisher's exact test.
We carried out logistic regression analyses to identify the associations between placental findings and adverse neonatal outcomes, each serving as the dependent variable. Independent variables were the GA at birth and birth weight z-score as continuous variables and HCA and MVM as categorical variables. The model was additionally adjusted for the antenatal potential confounders of sex, antenatal steroids and antenatal antibiotics. p Values of ≤0.05 were considered significant.
Results were expressed as odds ratios (OR) with 95% confidence intervals (CI). Data were analysed using SPSS Statistics for Windows, version 26.0 (IBM Corp, New York, USA).

| RE SULTS
A total of 174 EPT neonates were identified during the study period, and 129 of these were live-born singletons and met the initial inclusion criteria. However, six placentas were not sent for pathological  Table 1 presents the maternal, pregnancy and neonatal characteristics and outcomes of the study group.
During the perinatal period, 15 patients died (12.2%) and these deaths were caused by the following: NEC (n = 8), respiratory failure (n = 2), severe brain pathology (n = 2), EONS (n = 1), multi-organ failure (n = 1) and complications during delivery (n = 1). Autopsies were performed in seven cases, with a partial autopsy in one case.  Divergent results have also been reported for the association between HCA and BPD. 7,18,19 One 2019 review 19 supported an association between the two, but it did not distinguish between clinical and histologic chorioamnionitis. Kim et al. 20 found that only highgrade HCA was associated with BPD, suggesting that the association between HCA and outcomes in EPT neonates varied according to the extent of inflammation. FIR, which manifests as funisitis or chorionic vasculitis, is considered to be the histological manifestation of the foetal inflammatory response syndrome. 21 One study reported that foetal inflammatory response syndrome was associated with both BPD and intraventricular haemorrhage, which partly agreed with our results. 22 We found an association between HCA with FIR+ and brain pathology, detected by MRIs performed at term-equivalent age.
Reiman et al. 23 found no correlation between histological placental inflammation and brain lesions at term age in a cohort of very

| Strengths and limitations
The major strength of our study was that it included a homogenous cohort of EPT neonates and that the placentas were examined following a structured, pre-specified method and reported according to current international guidelines. 8 Our results may have been limited by the relatively low number of study participants, due to the fact that several of the outcome measures only affected a small proportion of the cohort. The other limitation was that this was a single-centre study.

| CON CLUS ION
Placental pathology affected 96% of EPT neonates in the study, and these fell into two main categories: acute inflammatory and maternal malperfusion. We found that HCA was associated with NEC, while HCA with FIR+ was associated with both BPD and brain pathology, as seen on MRI scans at term-equivalent age. MVM, on the contrary, was only associated with a low birth weight. Information on placental pathology may be useful for clinicians, but further studies are needed to investigate how placental changes in early GA differ to changes later in pregnancy.

ACK N OWLED G EM ENTS
We thank Professor Trond Markestad for supporting this study and the mothers who allowed their neonates to be included.

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