Preterm prelabor rupture of the membranes (PPROM) is defined as the rupture of fetal membranes with leakage of amniotic fluid before the onset of regular uterine activity prior to 37 weeks of gestational age (Mercer, 2003; Mercer, 2005). PPROM represents a phenotype of spontaneous preterm delivery that complicates approximately 3–4% of all pregnancies (Mercer, 2003; Mercer, 2005). Despite its predominantly non-infectious nature, PPROM might be complicated by the presence of acute inflammatory changes in the amniotic fluid and/or acute inflammatory lesions of the placenta (Cobo et al., 2012; Musilova et al., 2015).

The presence of acute inflammatory lesions of the placenta is characterized by diffuse infiltration of neutrophils in any of the structures of the placenta (fetal membranes, the placental disc, and the umbilical cord) and collectively are called acute histological chorioamnionitis (HCA) (Kim et al., 2015). Depending on the primary source of infiltrating neutrophils, HCA can be divided into the following: i) maternal inflammatory response, when extravasating maternal neutrophils infiltrate fetal membranes and/or the chorionic plate and ii) fetal inflammatory response, when extravasating fetal neutrophils invade the vessels of the chorionic plate and/or in structures of the umbilical cord (Kim et al., 2015).

The fetal inflammatory response is known to be the most severe form of HCA that is associated with adverse neonatal outcomes (Kim et al., 2015). It is also considered a histopathological counterpart of the fetal inflammatory response syndrome (Pacora et al., 2002). Apart from the fetal inflammatory response, a specific subtype of HCA, infiltration of the amnion by neutrophils has been shown to be related to very intense inflammatory responses, measured by various markers in the amniotic fluid and umbilical cord blood, regardless of the concurrent presence or absence of funisitis (Park et al., 2009). Therefore, two subtypes of HCA–with a fetal inflammatory response and acute inflammation of the amnion, should be of utmost clinical interest since they are associated with the most severe inflammatory responses.

It is obvious that placental lesions other than HCA (maternal and fetal vascular malperfusion, placental hemorrhage, and chronic villitis) mainly lead to impaired placental functions, which can be followed by an alteration of fetal growth (Salafia et al., 1989; Salafia et al., 1992; Salafia et al., 1995; Tyson and Staat, 2008; Mifsud and Sebire, 2014; Novac et al., 2018; Aviram et al., 2019). Collectively, these lesions represent underlying pathologies for conditions known as either small-for-gestational-age (SGA) or fetal growth restriction (FGR).

Nevertheless, some studies have provided evidence for the relationship between impaired fetal growth and HCA (Williams et al., 2000; Levy et al., 2021). This unexpected association is further supported by the following observations: i) lower birth weight is associated with upregulation of genes encoding proinflammatory transcription factor activator protein-1 (Ross et al., 2019); ii) the presence of HCA is found in approximately 10% of SGA pregnancies (DiGiulio et al., 2010); iii) a higher number of placental macrophages and increased placental inflammatory profile are found in pregnancies with impaired fetal growth or FGR (Street et al., 2006; Street et al., 2008; Sharps et al., 2020); iv) elevated concentrations of inflammatory markers in umbilical cord blood may be found in SGA newborns (Amarilyo et al., 2011; Lausten-Thomsen et al., 2014).

However, there is a shortage of information on whether the presence of HCA, particularly its most severe forms, with fetal inflammatory response and acute inflammation of the amnion, is related to impaired fetal growth in pregnancies complicated by PPROM. To fill this knowledge gap, a study on women with singleton pregnancies complicated by PPROM was conducted with the following goals: i) to assess birth weight, expressed as percentiles, and to compare the rates of the percentiles of birth weight that are less than or equal to the first, 10th, and 25th percentiles with respect to the presence of HCA with fetal and maternal inflammatory responses and the absence of HCA; ii) to assess birth weight, expressed as percentiles, and to compare the rates of the percentiles of birth weight that are less than or equal to the first, 10th, and 25th percentiles with respect to the presence of HCA with and without acute inflammation of the amnion and the absence of HCA; and iii) to assess birth weight, expressed as percentiles, and to compare the rates of the percentiles of birth weight that are less than or equal to the first, 10th, and 25th percentiles with respect to the severity of acute inflammation of the amnion.

This study was a retrospective cohort study conducted in pregnant women with PPROM admitted to the Department of Obstetrics and Gynecology, University Hospital Hradec Kralove in the Czech Republic between May 2008 and March 2021 who met the following criteria: i) singleton pregnancy; ii) gestational age at admission between 24 + 0 weeks and 36 + 6 weeks; iii) maternal age ≥18 years; iv) available histopathological results of the placenta. The exclusion criteria were as follows: i) pregnancy-related complications such as gestational diabetes, gestational hypertension, or preeclampsia; ii) chronic diseases such as pregestational diabetes and chronic hypertension; iii) structural or chromosomal abnormalities of the fetus.

Gestational age was determined based on the first-trimester ultrasound scan. The diagnosis of PPROM was established based on visual confirmation of amniotic fluid pooling in the posterior vaginal fornix by a sterile speculum examination. If uncertainty persisted after the clinical examination, the leakage of amniotic fluid was confirmed or ruled out using a test to determine the presence of insulin-like growth factor-binding protein in the vaginal fluid (Actim PROM test; Medix Biochemica, Kauniainen, Finland).

Women with PPROM at less than 34 weeks of gestation were treated with antibiotics and corticosteroids to accelerate lung maturation. Tocolytics were used only when regular uterine activity appeared during the course of corticosteroids, but not for longer than 48 h. Women with PPROM beyond 34 weeks of gestation were treated with antibiotics only. The women included in this study were managed using two different approaches. Between May 2008 and December 2013, women were treated actively (except those at < 28 gestational weeks). Labor was induced, or an elective cesarean section was performed after finalizing corticosteroid treatment but no later than 72 h after the rupture of the membranes, depending on the gestational age, fetal status, and maternal serum C-reactive protein concentrations. Since January 2014, the performance of transabdominal amniocentesis to assess the status of the intra-amniotic environment (microbial invasion of the amniotic cavity and intra-amniotic inflammation) has been a routine part of the clinical management of women with PPROM (Musilova et al., 2017). Thus, women admitted between January 2014 and May 2021 were managed differently. Women with intra-amniotic infection (the presence of both microbial invasion of the amniotic cavity and intra-amniotic inflammation) beyond the 28th gestational week were managed actively (labor was induced, or an elective cesarean section was performed after finalizing corticosteroid treatment within 72 h of membrane rupture for pregnancies before 34 weeks of gestation and within 24 h of membrane rupture for those beyond 34 weeks). The remaining women with PPROM were managed expectantly (Musilova et al., 2017).

After delivery, the placenta, fetal membranes, and umbilical cord were fixed in 10% neutral buffered formalin. Tissue samples were obtained from the placenta (at least two samples), fetal membranes (one sample from the free margin of membranes, one from the central part of the membranes, and one from the membranes with a marginal part of the placenta), and umbilical cord (usually one sample), which were routinely processed and embedded in paraffin. Sections of the tissue blocks were stained with hematoxylin and eosin.

The degree of neutrophil infiltration was evaluated separately in the free membranes (amnion and chorion-decidua), chorionic plate, and umbilical cord based on the criteria provided by Salafia et al. (Salafia et al., 1989). Histopathological examinations were performed by a single pathologist (HH) who was blinded to the clinical status of the women.

The collection of clinical samples and information was approved by the Ethics Committee of the University Hospital of Hradec Kralove, Czech Republic (19 March 2008; No. 200804 SO1P, which was renewed in July 2014 and January 2019, decisions No. 201407 S14P and No. 201902 S16P, respectively). Written informed consent was obtained from all the participants. Biological samples (amniotic fluid, cervical fluid, and umbilical cord blood) from the women included in this study were used in our previous studies and are presented in the publications. A total of 528 women from this cohort were included in our previous publication, where an association between birth weight and microbial invasion of the amniotic cavity and/or intra-amniotic inflammation was evaluated (Matulova et al., 2021). All methods used in this study were carried out in accordance with the relevant guidelines and regulations.

A total of 918 women with singleton pregnancies complicated by PPROM were eligible for the study, and 100 women were excluded for the following reasons: i) gestational diabetes mellitus (n = 53); ii) gestational hypertension (n = 19); iii) preeclampsia (n = 5); iv) pre-gestational diabetes mellitus (n = 12); v) chronic hypertension (n = 5); vi) combination of the above-mentioned diseases (n = 6). The remaining 818 women were included in the analysis.

In total, HCA was observed in 494 (60%) women. Among women with the presence of HCA, fetal and maternal inflammatory responses were identified in 343 (69%) and 151 (31%) women, respectively. Remaining 324 (60%) women had the absence of HCA. The demographic and clinical characteristics of the study population, as well as short-term neonatal outcomes, with respect to the presence of fetal and maternal inflammatory responses and the absence of HCA are shown in Table 1. Among women with the presence of HCA, 279 (57%) and 215 (43%) women were with and without acute inflammation of the amnion was observed in 279 (34%) women. The demographic and clinical characteristics of the study population, as well as short-term neonatal outcomes, with respect to the presence of HCA with and without inflammation of the amnion and the absence of HCA are shown in Table 2.

The principal findings of this study are as follows: i) HCA with acute inflammation of the amnion was associated with lower birth weight, expressed as percentiles derived from the estimated fetal weight standards; ii) HCA with acute inflammation of the amnion was related to the highest rate of newborns with a birth weight equal to or below the 25th percentile derived from the estimated fetal weight standards; iii) alteration of fetal growth, expressed as percentiles derived from the estimated fetal weight standards, associated with HCA with acute inflammation of the amnion was dependent on the severity of acute inflammation; and iv) when percentiles were derived from the birth weight standard, no difference in percentiles of the birth weights of newborns was observed among those with the presence of HCA with fetal and maternal inflammatory responses and those with the absence of HCA, as well as among those with the presence of HCA with and without acute inflammation of the amnion and those with the absence of HCA.

Neutrophils are not usually present in the placental tissue and fetal membranes (Kim et al., 2015). Following chemotactic stimulus and its gradient, neutrophils migrate towards the amniotic cavity i) from the intervillous space of the placenta into the chorionic plate and/or ii) from the decidua into fetal membranes (Kim et al., 2015). To develop acute inflammation of the amnion, neutrophils need a strong chemotactic stimulus because they must transmigrate through the entire chorionic layer (Park et al., 2009; Goldstein et al., 2020). This process is time-consuming and requires more than 36 h to develop acute inflammatory changes in the amnion from the first exposure to an inflammatory stimulus (Redline, 2006; Goldstein et al., 2020).

Besides the placentas from pregnancies complicated by PPROM and spontaneous preterm labor, the presence of HCA with acute inflammation of the amnion has also been reported in preterm pregnancies with impaired fetal growth (Salafia et al., 1995). Salafia et al. observed HCA with acute inflammation of the amnion in 37% of the pregnancies with appropriate-for-gestational-age newborns, 38% of the pregnancies with “asymmetric intrauterine growth restriction” and 8% of the pregnancies with “symmetric intrauterine growth restriction” (Salafia et al., 1995).

In this study, newborns from pregnancies with HCA with acute inflammation of the amnion had the lowest birth weight, expressed in percentile derived from the estimated fetal weight standard. This interesting finding was further supported by the fact that the alteration of fetal growth was dependent on the severity of acute inflammation of the amnion. The lowest percentiles and the highest rates of birth weight equal to or less than the 10th and 25th percentiles were found in the women with the most severe form of acute inflammation of the amnion (grade IV). The mechanistic explanation for this observation is unclear. It can be hypothesized that the placenta is affected by various lesions (maternal or fetal vascular malperfusion or chronic villitis), which are responsible for impaired fetal growth, might produce endogenous “danger signals” (e.g., heat shock protein 70, high mobility group box-1, or S100B) (Gazzolo et al., 2006; Zenerino et al., 2017; Lai et al., 2020). These signals lead to the production of chemotactic stimuli such as interleukin (IL)-8, which is followed by the migration of maternal neutrophils into the placenta and/or fetal membranes. When the production of chemotactic stimuli persists long enough, neutrophils can get a sufficient temporal period to reach and infiltrate the amnion. This hypothesis is supported by the following observations: i) placentas from pregnancies with FGR had a higher expression of mRNA for IL-8 (Hahn-Zoric et al., 2002); ii) maternal blood lymphocytes from pregnancies with FGR produce higher levels of IL-8 after stimulation with trophoblast (Raghupathy et al., 2012); iii) maternal serum concentrations of IL-8 are higher in pregnancies with SGA fetuses (Wang et al., 2015); and iv) malarial infection in pregnancy, usually leading to FGR, is associated with a higher expression of IL-8 mRNA in the placenta (Moormann et al., 1999). This hypothesis suggests that placental changes leading to impaired fetal growth are associated with the development of the acute inflammation of the amnion rather than the fetal inflammatory response, as shown in this study. We cannot fully exclude a contribution of fetal membranes on the production of the chemotactic stimuli along with the placenta for the following reasons: i) transporter proteins in the fetal membranes along with nutritional transport system suggests that fetal membranes play an equal role to that of the placenta in drug and nutrients transports (Ganguly et al., 2021; Kammala et al., 2022); ii) endogenous activities in the fetal membranes on cellular level can generate danger signals (Menon and Peltier, 2020; Sheller-Miller and Menon, 2020; Shahin et al., 2021; Shepherd et al., 2021; Tantengco et al., 2021); iii) fetal membranes function can be independent of the placenta and placental involvement (Menon, 2016); iv) fetal growth restriction can increase apoptosis in the chorionic trophoblast cells of fetal membranes and expression of parathyroid-related protein expression in the fetal membranes (Curtis et al., 2000; Murthi et al., 2005); and v) fetal membranes are not the mere extension of the placenta and have their own identity, function and hence, their compromise alone without the placental involvement can be detrimental (Collins et al., 1993; Menon and Moore, 2020). Therefore, functions of fetal membranes might be impaired in pregnancies complicated by the alteration of fetal growth.

The cutoff value of the 10th percentile is widely accepted by obstetricians and pediatricians as a threshold for identifying fetuses/newborns with SGA (DiGiulio et al., 2010; Figueras and Gratacos, 2014). This was the reason why this cutoff, along with the first percentile (to identify newborns with extremely impaired growth) and 25th percentile (to reveal a mild alteration of growth not fulfilling a threshold for SGA), were selected and used in this study. According to the abovementioned results, the highest rate (47%, 89/190) of newborns with having birth weight equal to or less than the 25th percentile was identified in the subset of women with HCA with acute inflammation of the amnion. This observation seems to be clinically relevant because it suggests that underlying placental pathologies associated with impaired fetal growth might be lasting enough to develop acute inflammatory lesions in the amnion but not the restriction of fetal growth that reaches the threshold of the 10th percentile due to subsequent/co-incidental development of PPROM.

Two INTERGROWTH-21st standards (for estimated fetal weight and newborn birth weight) were used in this study to derive percentiles for the birth weights of newborns. Two different growth charts were employed because it is still under debate which standard should be preferred (Marsal et al., 1996; Salomon et al., 2007; Kiserud et al., 2017; Stirnemann et al., 2017; Nicolaides et al., 2018). The main advantage of using the standard for estimated fetal weight is that reference ranges of the estimated fetal weight are representative of the whole population. On the other hand, the reference ranges of birth weight standards, particularly for newborns delivered preterm, suffer from the overrepresentation of pathological pregnancies resulting in iatrogenic or spontaneous preterm delivery in whom impaired placentation might be expected (Nicolaides et al., 2018). This limitation of birth weight standard might be seen mainly in the subset of newborns delivered before 33 weeks of gestation(Villar et al., 2014a). These methodological differences between the charts were observed in this study. For example, the subgroups of women with fetal inflammatory response and acute inflammation of the amnion, with the lowest gestational age at delivery, had lower medians of the percentiles of birth weight derived from the estimated fetal weight standard than when they were derived from the birth weight standard. Taking this into consideration, clinicians should be aware that using a growth chart based on birth weight might not reveal all cases with mildly impaired fetal growth, mainly in lower gestational ages. In light of this fact, it is not surprising that no differences among the assessed subgroups were observed when percentiles derived from the birth weight standard were used. In addition, no differences were observed in the rates of the birth weight equal to or lower than the first, 10th, and 25th percentiles among the subgroups of women with PPROM.

This study has two main strengths. First, the study was conducted on a large homogeneous cohort of Caucasian women with a thoroughly defined phenotype of spontaneous preterm delivery. Second, the assessment of HCA was performed by a single, experienced perinatal pathologist who was blinded to the clinical status of the women. However, there are certain limitations to this study. First, this study covers a 13 years-long interval. During this time, two different management strategies (active management and expectant management) of PPROM pregnancies were used. Therefore, attention was paid to this important confounding factor in this study, and the results were adjusted for the same. Second, no data regarding ultrasonographically estimated fetal weight from the time of delivery or shortly before delivery were available. This shortcoming prevented us from assessing the relationship between the estimated fetal weight, expressed in percentiles derived from the estimated fetal weight standards and the presence of acute inflammatory lesions in the placenta. Third, the histopathological assessment of placental lesions other than HCA (maternal and fetal vascular malperfusion, placental hemorrhage, and chronic villitis) were not systematically performed in the placentas from PPROM pregnancies. Last, the risk factors of the development of fetal and maternal inflammatory responses were not assessed and evaluated in this study.

In conclusion, in this large retrospective cohort study of singleton pregnancies with PPROM, acute inflammation of the amnion was associated with a lower birth weight but only when the percentiles of birth weight were derived from the standards for estimated fetal weight.