Evaluation of the fetal myocardial performance ındex and Epicardial fat thickness in pregnant women with preterm prelabor rupture of membranes

Abstract Objective To assess the epicardial fat thickness (EFT) and modified myocardial performance index (Mod-MPI) measurements in fetuses at pregnant women with preterm prelabor rupture of membranes (pPROM) and compare them to healthy pregnant women’s fetuses. Methods Forty patients who presented to our clinic at 24–36 + 6 gestational weeks and were diagnosed with pPROM were included in the patient group. During the same period, 40 healthy pregnant women at similar gestational weeks were randomly selected as the control group. Results The pPROM and control groups were similar in terms of demographics. In the pPROM group, EFT, Mod-MPI, and isovolumetric contraction(ICT) and relaxation times(IRT) were significantly higher and ejection time (ET) was significantly lower compared to the control group. In addition, Mod-MPI z-scores, IRT z-scores, ICT z-scores were significantly higher and ET z-scores was significantly lower in the pPROM. According to the ROC analysis, the optimal cutoff value of EFT was calculated as 1.55 mm, with 68% sensitivity and 71% specificity (AUC: 0.718, 95% CI: 0.550–0.786, p = .018) for predicting NICU requirement and the optimal cutoff value of EFT was calculated as 1.55 mm with 72% sensitivity and 73% specificity (AUC: 0.726, 95% CI: 0.556–0.896, p = .015) for predicting medication use in the pPROM group. Conclusion This study revealed differences in the EFT and Mod-MPI measurements of the fetuses of pregnant women with pPROM. Considering that EFT is an important energy source for the myocardium, as well as an endocrine structure in which inflammatory and anti-inflammatory markers are secreted, it is recommended to be evaluated in inflammatory conditions such as pPROM.


Introduction
Preterm prelabor rupture of membranes (pPROM) is the rupture of membranes earlier 37 weeks of gestation before labor begins. Its incidence is around 3-4% in all pregnancies [1]. Although the etiology of PPROM is not yet fully known, it is considered that infectious agents, connective tissue diseases, conditions causing an increase in local cytokines and the subsequent increase in matrix metalloproteinases, and thus progressive cellular apoptosis play a role [2]. PPROM is a pathological condition in which many inflammatory and antiinflammatory processes occur regardless of its etiology.
The Modified Myocardial Performance Index (Mod-MPI), calculated by Doppler ultrasonography, is a noninvasive measure of the global function of the myocardium. Many studies have shown that Mod-MPI is affected in conditions affecting fetal cardiac function [3,4]. In previous studies, Mod-MPI was measured in diseases accompanied by inflammation, such as intrauterine growth retardation, gestational and pregestational diabetes, and fetal anemia, and it was found to be significantly higher compared to controls [5][6][7][8].
Epicardial fat tissue is a tissue surrounding the heart that secretes various inflammatory substances and produces adipokines and chemokines with endocrine effects [9,10]. It has been shown that fetal epicardial fat thickness (EFT) increases in pregnancies with gestational diabetes mellitus (GDM) and preeclampsia, infants born large for gestational age, whereas it decreases in fetuses with intrauterine growth retardation (IUGR) [11,12].
In light of the above-mentioned information, in the current study, we aimed to evaluate EFT and Mod-MPI measurements in pregnant women with PPROM, which is an inflammatory process, and compare them to healthy pregnant women.

Material and methods
This prospective cohort study was conducted between June 2021 and December 2021 at the Perinatology Clinic of the Obstetrics and Gynecology Department of the Turkish Ministry of Health Ankara City Hospital, where an average of 20,000 deliveries occur annually. The study was designed in accordance with the principles of the Declaration of Helsinki. The study was first approved by the hospital's Human Ethics Committee, and written informed consent was obtained from all participants in the study. (date: 02/06/2021, number: E2-21-449).

Patient selection
Forty patients who presented to the Perinatology Clinic of Ankara City Hospital between June 2021 and December 2021 at 24-37 gestational weeks and were diagnosed with pPROM were included in the study group. A gestational age-matched control group was selected. We performed a power analysis with G-Power V R to define the minimum number of participants included in the study and 14 patients were found to be necessary with a power of 80% according to the study published by Aydin et al. [12,13]. For the same period, 40 healthy pregnant women at similar gestational weeks were selected as the control group. Patients with systemic diseases, such as diabetes, hypertension, and kidney disease, smokers, women with multiple pregnancies, and fetuses with anomalies were excluded from the study. Gestational weeks were calculated by confirming with first-trimester ultrasonography measurements. The patients' descriptive information, such as age, gravidity, parity, comorbidities, smoking status, drug use, and body mass index (BMI) were recorded. Fetal biometry (head diameter/circumference, abdominal circumference, femur length), amnion fluid index, fetal Doppler parameters (umbilical artery pulsatility index, middle cerebral aertery pulsatility index, ductus venosus pulsatility index) as recommended [14], gestational week, modified-MPI (Mod-MPI), isovolumetric contraction time (ICT), isovolumetric relaxation time (IRT), ejection time (ET), Mod-MPI z-scores, ICT z-scores, IRT z-scores, ET z-scores, EFT, EFW(estimated fetal weight), a gestational week at birth and mode of delivery, fetal birth weight, fetal birth weight percentile [15], Apgar scores (1st-5th min), neonatal intensive care unit (NICU) requirement, and fetal medication requirements were also recorded.
In patients with the complaint of leaking amniotic fluid, the diagnosis of pPROM was made based on the combination of the presence of fluid discharge in the sterile speculum examination, positive results of tests with high sensitivity and specificity in the diagnosis of the disease (nitrazine and/or placental alpha microglobulin-1 protein), and/or oligohydramnios on ultrasound [16].

Ultrasonography
All ultrasonographic measurements were performed by two perinatologists (BS, FDO) while the patient was looking. The Voluson S10 (GE Medical Systems) ultrasonography device and a 3.5 MHz convex transducer (6C1-PVT-375BT) transabdominal probe were used for the measurements. The left ventricular outflow tract view was obtained, which is ideal for viewing the space (epicardial fat) between the myocardium and epicardium along the right ventricle. EFT was measured outher surface of the myocardium to the inner surface of the epicardium, adjacent to the right ventricular wall ( Figure 1).
Since the original MPI may have intra-and interobserver variations, Hernandez et al. tried to optimize it and defined the modified Tei Index (Mod-MPI) to optimize the measurement method [17]. In this method, the start and end points can be clearly determined with the specific images (clicks) of the opening and closing movements of the mitral and aortic valves in the Doppler waveform. The three clearest valve clicks and waveforms for which the fetal heart rate was within the normal range (120-160 bpm) were measured and averaged to avoid variations. This aims to standardize measurements. In the current study, apical four-chamber sections were obtained when measuring MPI. The Doppler specimen was opened 2-6 mm, and placed on the lateral wall of the ascending aorta, which includes both the mitral and aortic valves. The gain of Doppler was lowered until the valve clicks were clearly visualized and the Doppler angle of insonation was kept below 30 [17]. A trace was obtained in which both lid clicks were seen. E-wave (premature ventricular filling) and A-wave (active atrial filling) were acquired. ICT was measured as the time from the onset of mitral valve closure to the aortic valve opening, IRT as the time from the aortic valve closure to the mitral valve opening, and ET as the time from the aortic valve opening to closure. Mod-MPI was calculated using "ICT þ IRT/ET" (Figure 2). All measurements were performed within 24 h after the admission to the hospital.

Management protocols
The follow-up of all pregnancies were performed in the perinatology clinic of Ankara City Hospital. Antenatal corticosteroid prophylaxis was given for fetal lung maturation before 34 þ 0 weeks of gestation and magnesium sulfate administration was administered for neuroprotection before 32 þ 0 gestational weeks. Antibiotic prophylaxis and hospitalization indications were provided according to current guidelines [18].

Statistical analysis
Statistical analyses were conducted using the Statistical Package for the Social Sciences (SPSS v. 22, IBM, SPSS for Windows, NY: IBM Corp.). Descriptive statistics were presented as mean and standard deviation values as they were normally distributed. Student's ttest was used to compare the mean values between the groups. Categorical variables were presented as numbers and percentages. The chi-square test was used to compare the categorical variables between the groups. Afterwards, performed receiver operating characteristic (ROC) curve analysis to assess the  performance of the Mod-MPI, ICT, IRT, ET, its z-scores and EFT value, in predicting NICU and medication use. A two-tailed p value of <.05 was regarded as statistically significant.

Results
A total of 80 patients (age range, 18-44 years) were included in the study, including 40 pregnant women with pPROM and 40 healthy pregnant women. The demographic data of the groups are given in Table 1. Six pregnant women with gestational diabetes mellitus, five twin pregnant women, and three pregnant women with gestational hypertension were not included in the study group. Due to maternal morbid obesity and fetal position, fetal Mod-MPI and the fetal epicardial fat thickness could not measured so five pregnant women excluded from the study. There was no statistically significant difference between the groups in terms of maternal age, gravidity, parity, gestational week, fetal doppler parameters, apgar score, and BMI (p > .05). EFT (p < .001) and Mod-MPI (p < .001) were significantly higher in the PPROM group compared to the control group. In addition, ICT (p ¼ .010) and IRT (p ¼ .003) were statistically significantly higher and ET (p ¼ .001) was significantly lower in the pPROM group. Also, ICT z-scores and IRT zscores were higher but ET z-scores lower in the pPROM group and were statistically significantly (p ¼ .013, p ¼ .003, p ¼ .001 respectivitely). In addition, in the statistical analysis, Mod-MP _ I z scores were higher in the patient group compared to the control group and were statistically significant. Estimated fetal weight (EFW) and fetal heart rate (FHR) were no statistically significant difference between the groups (p ¼ .222, p ¼ .606 respectivitely). Gestational week at birth and birth weight were found to be significantly lower in the pPROM group (p < .001). Table 1 presents cesarean section (CS), NICU admission, and fetal medication requirement of the groups. 17 of 33 pregnant women who had cesarean section in the PPROM group, had a history of previous cesarean section. _ In addition 10 of 25 pregnant women who had cesarean section in the control group, had a history of previous CS. Although the rate of CS was no statistically significant difference between the groups (p ¼ .092) but NICU admission (p ¼ .019) were proportionally higher in the PPROM group.
Gram-negative bacilli were grown in the blood cultures of two newborns and pseudomonas aeruginosa was grown in the blood cultures of one neonatal in the pPROM group who needed intensive care. Neonatal medication requirement was found to be statistically significantly higher in the PPROM group (p ¼ .030). The majority of the pregnancies remained uncomplicated on the other hand due to the small number of participants 4 pregnancies complicated by preterm delivery during the follow-up decreased the mean gestational week in the control group (a gestational week at birth; 28, 30, 32, 33 respectivitely).
Penicillin G and Gentamicin were started to neonatals in the control group who needed NICU.
In the patient group, Penicillin G and Gentamicin treatment was started in eight (44%) neonatals and Ampicillin and Amikacin treatment was started in ten (56%) neonatals. Vancomycin and Meropenem was started in three (16%) neonatals who did not respond to Penicillin G and Gentamicin treatment, and Vancomycin and Piperacillin tazobactam treatment was started in one (5%) neonatal. Colistin-sensitive pseudomonas aeruginosa growth was grown in the blood culture of one neonatal who was treated with Vancomycin and Meropenem, and Colistin (5%) was added to the treatment. All antibiotic treatments were started according to by Satar et al. [19]. All newborns needed intensive care unit due to respiratory distress syndrome and prematurity. In the pprom group, two newborns who needed intensive care due to respiratory distress and prematurity died. One of them grew gram-negative bacilli in the blood culture. One died due to late neonatal sepsis and the other due to septic shock. There was no newborn who died in the control group.

Discussion
In the present study, we found high EFT, MPI, ICT, IRT, its z-scores and low ET, its z-score in the pPROM group compare to the controls. Additionally, adverse neonatal outcomes including preterm delivery, NICU admission, low birth weight, medication use were increased in the pPROM group. Elevated fetal Mod-MPI and EFT may be valuable tool for identifying fetuses for adverse neonatal outcomes in pPROM pregnant women.
PPROM is an obstetric complication resulting from various events like infection, hypoxia, antenatal hemorrhage, or uterine distention. Many of these risk factors lead to excessive systemic inflammation in the maternal-fetal interface [20]. Although the exact cause of the rupture of membranes is often unknown, asymptomatic intrauterine infection and smoking are reported to be the most common etiologies, associated adverse neonatal outcome [21,22].
Epicardial fat tissue is an active endocrine organ that secretes many pro-inflammatory and antiinflammatory substances and makes an active contribution to myocardial energy production [23,24]. Epicardial fat tissue can also have a protective effect on response to local or systemic damage through many anti-inflammatory agents it secretes (e.g. Bold values are statistically significant (p < .05). Table 3. Receiver operating characteristic curve analysis to assess the performance of the value to predicting medication use in the pPROM group. adiponectin and adrenomedullin) [24,25]. In addition, fetal EFT has been shown to be associated with preeclampsia and GDM [26,27]. In the present study, we found increased EFT values in the patient group compared to the controls. In addition, we investigated the predictive value of EFT on NICU and medication use and we determined a cut off value, sensitivity, and specificity in the pPROM group. Mod-MPI is a noninvasive index that combines the overall ventricular function and systolic and diastolic time intervals. MPI is a marker of the entire cardiac function and is important in terms of showing early cardiac adaptation according to disease states. Although the normal values of MPI are known to vary according to the gestational week, available data are not consistent [28]. Tsutsumi et al. reported that fetal Mod-MPI decreased linearly as the gestational week progressed [29], while Eidem did not show a significant change in fetal Mod-MPI [30], and Barberato et al. found it to increase with the gestational week [31]. In fetuses with IUGR, MPI is affected in the early stages of hypoxemia. In the presence of ventricular dysfunction, MPI prolongs and remains abnormal in various phases of fetal risk [32]. In complicated pregnancies with fetal involvement, IRT is the main parameter affected, and ventricular loading and cardiac contractility and relaxation anomalies have an effect on Mod-MPI. In previous studies evaluating fetal cardiac involvement, IRT and ICT were shown to be significantly higher while ET was significantly lower. In the hypoxic environment, there is a significant decrease in ventricular ejection force, which explains the changes in ET, ICT, and IRT by showing the deterioration of cardiac filling [33]. An increase in IRT diastolic dysfunction and ICT represents systolic of the fetal heart [34]. In our study, we found higher Mod-MPI, ICT, and IRT values and lower ET in the pPROM group matched to the control group. Also, Mod-MPI zscores, ICT z-scores, and IRT z-scores were higher but ET z-scores lower in the pPROM group. This result seems to be associated with the effect of pPROMrelated inflammation on fetal circulation. Findings indicate that there is fetal myocardial involvement in pPROM pregnant women and are in agreement with the literature. The development of severe cardiac dysfunction in response to inflammation in adult patients with sepsis is a known pathophysiological process. An inflammatory process secondary to infection may develop in pPROM, and this causes the production of endotoxins in the amniotic fluid of pregnant women with pPROM. Fetal myocardial cytokines are produced in response to the endotoxins produced, and severe fetal cardiovascular deterioration develops, similar to adult patients [35]. Furthermore, the fetus responses to the increased inflammatory processes occurring in the maternal-fetal interface. This reaction is called "Fetal Inflammatory Response Syndrome" (FIRS) which is thought to be responsible for most of the neonatal complications in newborns with pPROM. A vicious circle of events occur due to an excessive fetal immune response resulting in altered function in various fetal organs like the brain, kidney, liver, and heart. Thus, altered cardiac function in the fetuses of pregnant women complicated by pPROM may be a reflection of these pathologic processes [36]. In our opinion, both fetal cardiac function and EFT may be affected by this response and if the physicians can detect these changes prenatally, better neonatal outcomes may be achieved by providing appropriate coordination between the perinatologist and the neonatologist. Further studies with larger participants may confirm this hypothesis. Determination of cut off values for MPI measurements in the prediction of adverse neonatal outcomes in the present study is valuable as it has the potential to guide physicians in the management of deliveries complicated by pPROM. Clinical findings and conservative inflammation markers like Creactive protein are the main determinants of the clinican's decision in the management of pregnancies with p PROM for the time being. However, researchers are working on different methods to identify complications like chorioamnionitis in the earlier stages. Although conservative methods have high sensitivity for predicting adverse maternal events, it is challenging to foresee neonatal complications because pPROM seems to have a prominent impact on the developing fetus. For this reason, studies investigating the impact of pPROM on fetal organ systems are valuable. On the other hand, the present study focused on only early neonatal outcomes. In our opinion, future studies including long-term outcomes will be beneficial to assess the importance of MPI and EFT evaluation of fetuses complicated by pPROM. Follow-up echocardiographic assessment of children delivered from pPROM pregnancies and comparing them with noncomplicated controls may reveal more valuable results. Additionally, the impact of fetal cardiac changes on the long-term prognosis of affected children can be identified.
In the current study, neonatal drug requirement was found to be statistically higher in the pPROM group. This supports the idea that infectious agents directly affect the fetus or activate the intrauterine fetal inflammatory system through indirectly produced endotoxins.
To the date of our knowledge, this is the first prospective study in which fetal EFT and Mod-MPI were evaluated together in pregnant women with pPROM. Prospective design, the inclusion of neonatal outcomes and the combination of Mod-PI with EFT are the main strengths of the present study. However, the relatively low number of cases and the lack of longterm outcomes are the main limitations. Furthermore, the mean gestational week at delivery was relatively lower and the NICU admission rate was relatively higher in the control group, most probably due to the small number of cases. Although the majority of the cases in the control group had no pregnancy complications, only a 10% preterm rate lowered the overall gestational age at delivery. Thus, further studies focused on larger populations might indicate more precisive results.
In conclusion, this study showed that there might be differences in the EFT and Mod-MPI measurements of the fetuses of pregnant women with pPROM compared to healthy pregnancies. According to our results, fetal EFT measurement seems to be a novel, noninvasive parameter for inflammatory events as well as for the metabolic state of pregnancy. Fetal EFT measurements can be made in specialized centers during the evaluation of high-risk pregnancies with inflammation such as pPROM. Considering that EFT is an important energy source of the myocardium and an endocrine organ in which many inflammatory and anti-inflammatory markers are secreted, future randomized controlled studies are needed to investigate the effect of EFT on fetal cardiac function and inflammatory systemic diseases in pPROM pregnancies. In addition, studies assessing the long-term results of fetuses with high EFT and Mod-MPI values in terms of systemic inflammatory diseases are warranted, and these fetuses can be followed up more closely. There is also still a need for larger-scale follow-up studies to evaluate prospective fetal EFT and Mod-MPI values together with long-term results in pregnancies with pPROM.