Preterm prelabor rupture of the membranes (PPROM) is defined as rupture of the fetal membranes with leakage of amniotic fluid before the onset of regular uterine activity before 37 weeks of gestation (Mercer, 2003; Mercer, 2005). PPROM is one of the “great obstetrical syndromes,” with considerable medical and socio-economic impacts (Romero, 1996; Romero et al., 2006; Di Renzo, 2009; Brosens et al., 2011). PPROM remains under intensive debate among scientists, researchers, and clinicians given the necessity: 1) to fully unravel the underlying pathophysiological mechanisms to make the prevention of PPROM possible (Moore et al., 2020); 2) to understand the underlying mechanisms affecting the interval between PPROM and delivery to optimize the timing of induction of lung maturity (Battarbee et al., 2020); 3) to characterize the causality and consequences of microbial invasion of the amniotic cavity (the presence of microorganisms and/or their DNA in the amniotic fluid) and intra-amniotic inflammation (elevation of inflammatory mediators in the amniotic fluid) (Menon and Richardson, 2017); and 4) to identify risk factors and reliable biomarkers of microbial invasion of the amniotic cavity and intra-amniotic inflammation to enable a individualized therapeutic approach (Stranik et al., 2021).

The presence of microbial invasion of the amniotic cavity and intra-amniotic inflammation complicates approximately 23–41% and 17–58%, respectively, of pregnancies with PPROM (Romero et al., 2015; Kacerovsky et al., 2020a; Kacerovsky et al., 2021b). Based on these two intra-amniotic complications, the following scenarios can occur in PPROM pregnancies: 1) intra-amniotic infection (presence of both), 2) sterile intra-amniotic inflammation (intra-amniotic inflammation only), 3) colonization of the amniotic fluid (microbial invasion of the amniotic cavity only), and 4) negative amniotic fluid (absence of both) (Musilova et al., 2015; Romero et al., 2015).

There is plethora of evidence that Ureaplasma spp. are the most common microorganisms recovered from the amniotic fluid obtained from PPROM pregnancies. Therefore, these low-virulence bacteria with sizes comparable to those of viruses represent the most common cause of intra-amniotic infection or colonization of the amniotic cavity in PPROM (Kacerovsky et al., 2021b). Because they are commonly found in the cervical/vaginal niche of women with PPROM, the lower genital tract is considered the main source of amniotic fluid Ureaplasma spp.

The presence of Ureaplasma spp. in the cervical/vaginal niche in PPROM pregnancies is related to a higher risk of ascension and their subsequent presence in amniotic fluid (Musilova et al., 2016). Nevertheless, there is a shortage of information on whether their prevalence and loads in the cervical niche might be prone to a specific subtype of intra-amniotic complications, such as intra-amniotic infection, sterile intra-amniotic inflammation, and colonization of the amniotic cavity.

To fill this knowledge gap, this study aimed to determine the prevalence of Ureaplasma spp. DNA in the cervical fluid of the subgroups of women with PPROM with intra-amniotic infection, sterile intra-amniotic inflammation, colonization of the amniotic cavity, and negative amniotic fluid. The secondary aim was to compare the microbial load of Ureaplasma spp. DNA in the cervical fluid among subgroups of women with PPROM. The final aim was to compare the relative abundance of Ureaplasma spp. DNA in the cervical fluid among subgroups of women with PPROM.

In this retrospective study, we included pregnant women admitted to the Department of Obstetrics and Gynecology of the University Hospital Hradec Kralove, Czech Republic, between May 2015 and May 2021, who met the following criteria: 1) age ≥18 years; 2) singleton pregnancy; 3) gestational ages between 24 + 0 and 33 + 6 weeks; 4) PPROM; and 5) amniocentesis to assess the intra-amniotic environment. The exclusion criteria were as follows: 1) pregnancy-related complications (e.g., fetal growth restriction, gestational, gestational hypertension, and preeclampsia); 2) chronic medical complications (e.g., pregestational diabetes mellitus and chronic hypertension); 3) congenital or chromosomal fetal abnormalities; 4) signs of fetal hypoxia; 4) significant vaginal bleeding; and 5) preterm labor with intact membranes.

PPROM was diagnosed based on visual confirmation of amniotic fluid pooling in the posterior vaginal fornix by sterile speculum examination. If uncertainty about amniotic fluid leakage remained after the clinical examination, the presence of insulin-like growth factor-binding protein in the vaginal fluid was assessed (Actim PROM test; Medix Biochemica, Kauniainen, and Finland).

Body fluid samples (amniotic fluid first, cervical fluid second) were collected at the time of admission before the administration of corticosteroids, antibiotics, or tocolytics. Transabdominal amniocentesis to obtain an amniotic fluid sample is a standard part of the department’s clinical management of women with PPROM to assess the intra-amniotic environment. Women with PPROM received corticosteroids to accelerate lung maturation and intravenous antibiotics. Women with intra-amniotic inflammation received clarithromycin for 7 days unless delivery occurred. Those without intra-amniotic inflammation were treated with benzylpenicillin for 7 days unless delivery occurred. In the case of penicillin allergy, women were treated with clindamycin for 7 days unless delivery occurred. Once the final results regarding microbial invasion of the amniotic cavity from cultivation or PCR were known, the attending clinician modified the women’s antibiotic therapies accordingly. Tocolysis was used only in women who developed regular uterine activity during the course of corticosteroids or within 24 h after their administration. The women were managed expectantly, except those with intra-amniotic infection (the presence of both microbial invasion of the amniotic cavity and intra-amniotic inflammation) beyond the 28^th gestational week. This subset of women with PPROM was managed actively (labor was induced or an elective cesarean section was performed after finalizing corticosteroid treatment within 72 h of membrane rupture).

This study was approved by the institutional review board of the University Hospital Hradec Kralove (June 2014, No. 201408 S07P). All study participants were Caucasian, and informed consent was obtained from all participants.

Amniotic fluid samples, cervical fluid samples, and the clinical and demographic data of some women from this cohort were used in our previous studies (Musilova et al., 2017a; Musilova et al., 2017b; Hornychova et al., 2018; Kacerovsky et al., 2018; Musilova et al., 2018; Janku et al., 2019; Kacerovsky et al., 2020a; Kacerovsky et al., 2020b; Musilova et al., 2020; Soucek et al., 2020; Spacek et al., 2020; Kacerovsky et al., 2021a; Kacerovsky et al., 2021b; Matulova et al., 2021; Musilova et al., 2021; Stranik et al., 2021)

Overall, 217 women with singleton pregnancies and PPROM between gestational ages 24 + 0 and 33 + 6 weeks were included in the study. Intra-amniotic infection, sterile intra-amniotic inflammation, colonization of the amniotic cavity, and negative amniotic fluid were observed in 20% (44/217), 11% (24/217), 10% (21/217), and 59% (128/217) of women, respectively. The demographic and clinical data of the women with PPROM, as well as short-term neonatal outcomes are summarized in Table 1.

The most common microorganisms in the amniotic fluid were Ureaplasma spp., which were found in 59% (26/44) and 71% (15/21) of women with intra-amniotic infection and colonization of the amniotic cavity, respectively. All microbial findings in the amniotic fluid that were identified in the women with intra-amniotic infection are shown in Table 2.

Ureaplasma spp. DNA in the cervical fluid was identified in 61% (133/217) of the women. All of the women with Ureaplasma spp. DNA in their amniotic fluid (n = 41) also had Ureaplasma spp. DNA in their cervical fluid.

The principal findings of this study in women with PPROM before 34 weeks were as follows: 1) the total prevalence of Ureaplasma spp. DNA in the cervical fluid was 61%; 2) the presence of intra-amniotic infection and colonization was related to higher rates of Ureaplasma spp. DNA in the cervical fluid compared to those with negative amniotic fluid results; and 3) the presence of intra-amniotic infection, sterile intra-amniotic inflammation, or colonization of the amniotic fluid was associated with higher microbial loads of Ureaplasma spp. DNA in the cervical fluid compared to those with negative amniotic fluid; and 4) the presence of sterile intra-amniotic inflammation was related to a higher relative abundance of Ureaplasma spp. DNA in the cervical fluid compared to that in negative amniotic fluid.

Ureaplasma spp. are considered commensal microorganisms of the cervical/vaginal niche owing to: 1) their high prevalence in pregnant or non-pregnant women, and 2) the absence of a difference in the prevalence between women of reproductive ages with and without symptoms of infection of the urogenital tract (Marovt et al., 2015; Sweeney et al., 2017). In pregnant women, the presence of Ureaplasma spp. in the cervical/vaginal niche has been shown to be associated with adverse pregnancy outcomes (Abele-Horn et al., 2000; Kafetzis et al., 2004; Kataoka et al., 2006). Accordingly, Ureaplasma spp. in the cervical/vaginal niche is more frequently observed in pregnancies complicated by PPROM than in uncomplicated pregnancies (Larsen and Hwang, 2010; Donders et al., 2017; Sprong et al., 2020). Its prevalence in PPROM pregnancies varies between 53 and 73% (Kwak et al., 2014; Kwak et al., 2015; Musilova et al., 2016). The observations from this study (61%) are in agreement with previously published findings.

Interestingly, in this study, women with intra-amniotic infection and colonization of the amniotic cavity had higher rates of Ureaplasma spp. DNA in their cervical fluid than did women with sterile intra-amniotic inflammation and negative amniotic fluid. In other words, a higher proportion of Ureaplasma spp. DNA in the cervical fluid was found in the subgroups of women with microbial invasion of the amniotic cavity. This observation is consistent with the findings of our previous study, in which women with microbial invasion of amniotic inflammation had a higher rate of Ureaplasma spp. DNA in the cervical fluid than did those without this complication (Musilova et al., 2016). Notably, in this study, no difference in the frequency of Ureaplasma spp. in the cervical fluid between women with sterile inflammation and negative amniotic fluid was identified. This observation differs from that recently reported in women with preterm labor and intact membranes. In that study, the prevalence of Ureaplasma spp. DNA in the group of women with sterile intra-amniotic inflammation was comparable to that in women with intra-amniotic infection; however, it was 2-fold higher than that in women with negative amniotic fluid (Kacerovsky et al., 2021c). The diversity in observations between the studies on different phenotypes of spontaneous preterm delivery supports the hypothesis that pathophysiological pathways leading to the development of sterile intra-amniotic inflammation might differ between PPROM and preterm labor with intact membranes.

Abnormal microbiota in the cervical/vaginal niche have been shown to be associated not only with a higher frequency of Ureaplasma spp. but also with higher loads of these bacteria in that niche (Donders et al., 2017). Kwak et al. reported that a higher load of Ureaplasma spp. in the vaginal niche is associated with a higher frequency of histological chorioamnionitis in women with PPROM (Kwak et al., 2014). In line with their report, a difference in the microbial load of Ureaplasma spp. DNA in the cervical fluid among the subgroups with intra-amniotic infection, sterile intra-amniotic inflammation, colonization, and negative amniotic fluid (with the lowest levels found in those with negative amniotic fluid) was found in this study. However, this observation was not in concordance with our previous study, where no difference in the load of Ureaplasma spp. DNA in the cervical fluid was identified in women with and without microbial invasion of the amniotic cavity (Musilova et al., 2016). The reason for this difference is not fully clear; however, it can be explained by the differences between the studies, including: 1) the sample sizes of women with Ureaplasma spp. DNA (previous: n = 40; current: n = 133) that were studied; and 2) the gestational ages considered during sampling (previous: 24 + 0 to 36 + 6; current: 24 + 0 to 33 + 6).

The relative abundance of Ureaplasma spp. DNA in the cervical fluid might represent a more precise tool with which to characterize the association between the composition of the cervical microbiome and intra-amniotic complications, as opposed to the assessment of the absolute number of copies of Ureaplasma spp. DNA that are present. In this study, women with PPROM and sterile intra-amniotic inflammation had an approximately 2.5-fold higher relative abundance of Ureaplasma spp. DNA compared to women with negative amniotic fluid results. This observation should be considered in terms of the pathophysiology of the development of sterile intra-amniotic inflammation in PPROM.

Taken together, the results from this study suggest that 1) the subgroup of women with PPROM and negative amniotic fluid differs from those with intra-amniotic complications in terms of frequency, loads, and relative abundance of Ureaplasma spp. DNA in the cervical fluid; and 2) differences exist between PPROM and preterm labor with intact membranes with regard to the association between Ureaplasma spp. DNA in the cervical fluid and intra-amniotic complications. Collectively, these observations further support that pathophysiological pathways leading to the development of intra-amniotic complications might differ between PPROM and preterm labor with intact membranes.

This study has several strengths. First, a relatively large cohort of women with a well-defined clinical phenotype for spontaneous preterm delivery (PPROM) was assessed. Secondly, the presence of Ureaplasma spp. DNA in the cervical fluid was evaluated with a specific PCR procedure, which allowed us to identify even very low loads of Ureaplasma spp. DNA. Third, thorough assessment of microbial invasion of the amniotic cavity, with the use of a combination of non-specific PCR (16S rRNA) followed by sequencing, specific PCR for Ureaplasma spp., Mycoplasma hominis, and Chlamydia trachomatis, and aerobic/anaerobic cultivation allowed us to precisely dissect the subgroups of women with sterile intra-amniotic inflammation.

This study has limitations that are worth mentioning. First, in women with PPROM, amniotic fluid leaks from the amniotic cavity through the cervix into the posterior vaginal fornix. This means that the leaking amniotic fluid can contaminate the cervical fluid. Therefore, the fluid obtained with a Dacron swab from the endocervical canal of women with PPROM should be considered not just as cervical fluid but as a compound fluid, composed of both cervical fluid and amniotic fluid components. Its composition, as well as the ratio between amniotic and cervical fluids, might vary among women with PPROM and depend on various clinical parameters and scenarios (e.g., amount of residual amniotic fluid, location of the membrane rupture site, position of the fetus, and interval between PPROM and sampling) that occur at the time of sampling. Therefore, if the leaked amniotic fluid contains Ureaplasma spp. DNA (in cases of intra-amniotic infection and colonization of the amniotic cavity caused by Ureaplasma spp.), it can affect the total microbial load of the Ureaplasma spp. DNA and its relative abundance in the fluid obtained from the endocervical canal. Unfortunately, the presence of this phenomenon is inevitable in PPROM and could affect the assessment of microbial loads and the relative abundance of Ureaplasma spp. DNA in the cervical fluid among subgroups of women with PPROM. On the other hand, it is highly unlikely that the presence of Ureaplasma spp. DNA in leaked amniotic fluid could affect the assessment of the prevalence of Ureaplasma spp. in the cervical fluid, since the cervical/vaginal niche is considered to be a primary source of Ureaplasma spp.

Second, biovars, serovars, and other types of Ureaplasma spp. were not assessed and considered in this study. This fact should be taken as a shortcoming of this study. However, there is evidence of the possibility of horizontal gene transfer in Ureaplasma spp., allowing them to carry markers of multiple serovars (Xiao et al., 2011; Sweeney et al., 2017). Thus, the assessment of serovars could have a limited diagnostic value. Third, during the study period, the methods to assess the concentrations of IL-6 in amniotic fluid were modified. Therefore, the concentrations of IL-6 in this study were determined using two different approaches (lateral flow immunoassay point-of-care test and automated electrochemiluminescence). This fact prevents us from assessing the association between the intensity of the intra-amniotic inflammatory response, measured by IL-6 concentrations in amniotic fluid, and the loads of Ureaplasma spp. DNA in the cervical fluid.

In conclusion, in PPROM at < 34 weeks, the presence of intra-amniotic infection, sterile intra-amniotic inflammation, or colonization of the amniotic fluid was associated with a higher prevalence and/or load of Ureaplasma spp. DNA in cervical fluid than in the absence of intra-amniotic complications.