Authors: Gizem Aktemur, Betül Tokgöz Çakır, Gülşan Karabay, Can Ozan Ulusoy, Zeynep Seyhanlı, Serap Topkara Sucu, Nazan Vanlı Tonyalı, Can Tekin İskender
Categories: Maternal-Fetal Medicine, D-dimer (DD), Hypercoagulable state, Prediction of labor onset, Preterm premature rupture of membranes (PPROM)
Source: Archives of Gynecology and Obstetrics
Authors: Gizem Aktemur, Betül Tokgöz Çakır, Gülşan Karabay, Can Ozan Ulusoy, Zeynep Seyhanlı, Serap Topkara Sucu, Nazan Vanlı Tonyalı, Can Tekin İskender
Pregnancy induces a hypercoagulable state, characterized by increased coagulation factors and decreased anticoagulants, alongside ongoing fibrinolysis marked by elevated D-dimer (DD) levels. Reference values for DD in pregnancy often exceed the non-pregnant cutoff due to these changes. Elevated DD levels are common in late pregnancy and may correlate with complications such as gestational diabetes, hypertension, and preterm delivery, particularly in cases of preterm premature rupture of membranes (PPROM). This study investigates the association between DD levels, the duration from PPROM diagnosis to delivery, and neonatal outcomes.
This retrospective study was conducted at the Department of Perinatology, Etlik City Hospital, Ankara, Turkey, from October 2022 to May 2023. Eighty patients with PPROM between 24 and 36 weeks of gestation were included. Routine blood tests and coagulation parameters, including DD, were monitored every other day for 2 weeks. Patients were classified into two subgroups based on whether labor occurred within 7 days of PPROM diagnosis. Statistical analyses included the Mann–Whitney U test, Student’s t test, Chi-square test, Friedman test, Durbin–Conover test, generalized estimating equations (GEE), and ROC analysis.
Gestational age at admission was significantly lower in patients who delivered later than 7 days post-PPROM. Significant differences were observed in ultrasonographic measurements, with larger fetal parameters in the early delivery group. Higher DD levels at the third follow-up correlated with shorter durations to delivery (p = 0.021). Longitudinal analysis showed significant fluctuations in DD levels over time, particularly near delivery. The GEE analysis demonstrated a strong inverse relationship between DD levels and time to delivery (p = 0.004), supported by ROC analysis (AUROC = 0.811).
Elevated DD levels are associated with shorter durations from PPROM diagnosis to delivery, indicating their potential utility in predicting labor onset. Monitoring DD levels may help in clinical decision-making for managing PPROM, including planning neonatal care and timing of interventions.
D-dimer levels and labor This study demonstrates a relationship between elevated D-dimer levels and the time to delivery in PPROM cases. Higher D-dimer levels at the third follow-up were significantly correlated with shorter durations from diagnosis to delivery, suggesting the potential of DD as a predictive marker for labor onset.Clinical Implications for PPROM Monitoring D-dimer levels in pregnant women with PPROM could enhance clinical decision-making by providing insights into the timing of delivery. This can aid in planning neonatal care and interventions, ultimately improving outcomes for both mothers and infants.
Pregnancy is characterized by an increase in coagulation factors and a decrease in anticoagulants, which is widely recognized. This situation, characterized by reduced fibrinolysis, is in contrast to elevated levels of D-dimers (DD). Nevertheless, it is clear that fibrinolysis continues to be an ongoing process. [1, 2]. DD is the smallest part of fibrin degradation products [3]. The reference values for DD during pregnancy were derived from the reference values of healthy, non-pregnant adults. Nevertheless, because of the established knowledge of the rise in clotting factors during pregnancy, it is appropriate to regard the reference value above the usual cutoff of 0.5 mg/mL [4–6]. Multiple studies have consistently demonstrated that around 78% of pregnant women in the second trimester and close to 100% of pregnant women in the third trimester have elevated levels of DD compared to women in the first trimester [7, 8].
Research has indicated that gestational diabetes (GDM), gestational hypertension (GHT), preeclampsia, preterm delivery, and preterm premature rupture of membranes (PPROM) may be caused by alterations in the hemostatic system during pregnancy [9, 10]. Studies have shown that the increase in DD during pregnancy is significant, especially for GHT and GDM, and that it may be associated with pregnancy complications [11, 12]. Thrombin is believed to have a significant impact on the development of preterm delivery by initiating uterine contractions. Additionally, hypercoagulation may contribute to the development of PPROM [13]. Studies have demonstrated that DD serves not only as a marker for coagulation, but also as an indicator of inflammation and as a prognostic tool for significant disorders. Multiple investigations on COVID-19 patients have consistently demonstrated a correlation between elevated levels of DD and increased rates of both mortality and morbidity [14, 15].
Our study aimed to determine if there is an association between the duration from diagnosis to delivery, changes in DD, and other coagulation markers in individuals diagnosed with PPROM. We examined if there is a correlation between the variation in DD and the timing of birth.
This retrospective study was conducted at the Department of Perinatology, Etlik City Hospital in Ankara, Turkey, from October 2022 to May 2023. The study received approval from the hospital’s ethics committee and adhered to the universal standards outlined in the Declaration of Helsinki.
The study included 80 patients diagnosed with PPROM who were admitted to our clinic and subsequently delivered. Routine blood tests, including complete blood count, routine biochemistry (including ALT: alanine aminotransferase AST: aspartate aminotransferase, glucose, CRP: C-reactive protein), and coagulation parameters (PT: prothrombin time, APTT: activated partial thromboplastin time, fibrinogen, D-dimer), were performed upon admission. CRP and D-dimer levels were evaluated every other day for 2 weeks.
PROM refers to the rupture of the amniotic membranes before the onset of labor at or after 37 weeks of gestation, whereas PPROM refers to the rupture of the membranes before labor and before 37 weeks of gestation, typically occurring between 24 and 36 + 6 weeks [16]. Patients between 24 and 33 weeks of gestation with active amniotic fluid leakage or confirmed diagnosis of PPROM by a positive AmniSure (QIAGEN, Germany) test were included in the study. Exclusion criteria were complicated pregnancies (e.g., preeclampsia, gestational diabetes mellitus), conditions affecting coagulation parameters (e.g., thrombophilia, hyperhomocysteinemia), hematological diseases, and the use of antithrombotic or anticoagulant medications.
Patients were classified into two those who gave birth within 7 days and those who did not. If labor did not occur within 2 weeks, CRP and D-dimer monitoring was discontinued. Demographic and neonatal data were retrieved from patient records.
The prophylactic antibiotic regimen we use in PPROM patients admitted to our clinic is as 1 gram (g) of azithromycin orally on admission and 2 g of ampicillin intravenously every 6 h for 48 h, followed by 500 mg of amoxicillin orally every 8 h for 5 days [17].
Statistical analyses were performed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). For demographic and ultrasonographic data, group comparisons were made using median and interquartile range (IQR) values. Continuous variables were analyzed using both the Mann–Whitney U test and Student’s T test to assess differences between groups. Categorical variables were presented as numbers and percentages and analyzed using the Chi-square test. The longitudinal changes in D-dimer levels were analyzed using the Friedman test, and specific differences between results were assessed post-hoc with the Durbin-Conover test. Additionally, the impact of D-dimer levels on the duration from PPROM to delivery was investigated using generalized estimating equations (GEE). A receiver operating characteristic (ROC) analysis was conducted to evaluate the relationship between values predicted by the GEE model and actual observed values. This analysis was used to measure the accuracy and predictive power of the model. The area under the ROC curve (AUC) was reported, indicating the model’s ability to distinguish between different states. The significance level was set at p < 0.05 for all analyses, and results were reported with a 95% confidence interval (CI).
The data from Table 1 illustrates the differences in demographic, neonatal, and ultrasonographic measurements between women who delivered within 7 days following PPROM and those who delivered later. The gestational age at admission was significantly lower in the group delivering later at 28 weeks (IQR: 5.0) compared to 32 weeks (IQR: 5.5) in those who delivered within 7 days (p < 0.001). No statistically significant differences were found in maternal age, parity, number of abortions, and body mass index (BMI) between the two groups. Ultrasonographic measurements showed that biparietal diameter (BPD) was significantly larger in the group that delivered within 7 days, measuring 79 mm (IQR: 18.0) compared to 70 mm (IQR: 18.0) in the later delivery group (p = 0.008). Similarly, measurements of HC, AC, and FL were also larger in the early delivery group (p < 0.05, for all).Table 1Comparison of demographic, neonatal, and ultrasonographic data in terms of whether labor occurred within 7 daysPatient characteristics,median (IQR), n (%)Births within 7 daysn = 27Births after 7 daysn = 53p valueMaternal age, median (IQR)28.0 (7.50)28.0 (9.00)0.534Nulliparity, n (%)16 (59.5)22 (41.5)0.133Abortus, median (IQR)0 (1.0)0 (1.0)0.954BMI (kg/m^2^), median (IQR)27.0 (6.50)28.0 (7.00)0.950Gestational age at admission, median (IQR)32.0 (5.50)28.0 (5.00) < 0.001Mode of delivery, n (%)0.434Vaginal7 (63.6)16 (50.0)C-section4 (36.4)16 (50.0)IVF treatment, n (%)0 (0.0)2 (3.8)0.307Ultrasonographic characteristicsBiparietal diameter (mm), median (IQR)79 (18.0)70 (18.0)0.008Head circumference (mm), median (IQR)290 (63.0)251 (60.0)0.014Abdominal circumference (mm), median (IQR)271 (57.5)226 (61.0)0.006Femur length (mm), median (IQR)62 (12.5)51 (15.0)0.002Amnion DVP (mm), median (IQR)30 (20.0)40 (30.0)0.978Interval between PPROM and birth (days), median (IQR)4.0 (3.0)21.0 (20.0)<0.001Time of birth (weeks), median (IQR)33.0 (5.50)33.0 (5.00)0.597Neonatal characteristicsBirth weight (g), median (IQR)1720 (1153)1820 (1055)0.596APGAR score < 7 at 1st min, n (%)9 (33.3)21 (39.6)0.583APGAR score < 7 at 5th min, n (%)5 (18.5)9 (17.0)0.864NICU admission, n (%)17 (63.0)34 (64.2)0.917IQR interquartile range, BMI body mass index, IVF in vitro fertilization, DVP deep vertical pocket, PPROM preterm premature rupture of membranes, NICU neonatal intensive care unit
Hematological data presented in Table 2 indicates that white blood cell count (WBC) was higher in the group that delivered within 7 days, measuring 13.7 (10^3/mm^3) (IQR: 4.71) compared to 12.0 (10^3/mm^3) (IQR: 4.88) in the later delivery group, though this difference was not statistically significant (p = 0.066). No significant differences were observed in hemoglobin and hematocrit levels between the groups (p = 0.158 and p = 0.584, respectively). Notably, D-dimer levels at the third follow-up were significantly higher in the early delivery group, measuring 1.680 μg/mL (IQR: 1.220) compared to 0.930 μg/mL (IQR: 0.713) in the later delivery group (p = 0.021). No significant differences were found in APTT, PT, and INR levels between the two groups.Table 2Comparison of hematologic data in terms of whether labor occurred within 7 daysMedian (IQR)Births within 7 daysn = 27Births after 7 daysn = 53p valueHemoglobin (g/dL)11.6 (1.65)11.3 (1.60)0.158Hematocrit (%)35.3 (5.10)35.0 (5.00)0.584WBC (10^3^/mm^3^)13.7 (4.71)12.0 (4.88)0.066Protrombin time (s)7.88 (0.680)8.09 (0.680)0.281Activated partial thromboplastin time (s)27.5 (2.75)27.0 (2.80)0.419INR0.890 (0.040)0.900 (0.055)0.751C-reactive protein (mg/L)8.46 (21.9)7.68 (12.3)0.141Fibrinogen (mg/dL)479 (138.0)473 (88.0)0.401D-dimer (μg/mL)At admission0.870 (1.360)0.950 (0.750)0.6071st follow-up0.960 (1.050)0.890 (0.820)0.1942nd follow-up0.965 (1.160)0.860 (0.810)0.2363rd follow-up1.680 (1.220)0.930 (0.713)0.021WBC* white blood cell, INR International normalized ratio
The longitudinal analysis of D-dimer levels shown in Table 3 reveals significant changes over time (χ^2^ = 18.7, df = 6, p = 0.005). This outcome suggests that D-dimer levels fluctuate, particularly as the delivery process approaches, potentially correlating with phases of pregnancy and timing of labor. High D-dimer levels measured close to the time of delivery may indicate the imminent onset of labor.Table 3Friedman analysis of repeated measure of D-dimer levelMeasurement timeD-dimer level (μg/mL), medianChi-square (χ^2^)df**p valueAt admission0.99518.760.0051st follow-up0.9652nd follow-up0.9203rd follow-up0.9954th follow-up1.0005th follow-up1.0156th follow-up1.220
Post hoc analyses presented in Table 4 evaluates the statistical differences between D-dimer levels at various times. Results from the Durbin–Conover test indicated that levels measured closer to the time of delivery were significantly higher compared to initial measurements. The most notable difference was observed between the initial and the sixth measurement of D-dimer levels (p = 0.002).Table 4Pairwise comparisons (Durbin–Conover) of D-dimer levelsD-dimer measurement timeStatisticpAt admission1st follow-up0.30150.763At admission2nd follow-up0.42200.673At admission3rd follow-up0.09040.928At Admission4th follow-up0.78380.434At admission5th follow-up0.21100.833At admission6th follow-up3.10500.002 1st follow-up2nd follow-up0.12060.9041st follow-up3rd follow-up0.39190.6961st follow-up4th follow-up0.48230.6301st follow-up5th follow-up0.51250.6091st follow-up6th follow-up3.4065- <0.0012nd follow-up3rd follow-up0.51250.6092nd follow-up4th follow-up0.36170.7182nd follow-up5th follow-up0.63310.5272nd follow-up6th follow-up3.5271- <0.0013rd follow-up4th follow-up0.87420.3833rd follow-up5th follow-up0.12060.9043rd follow-up6th follow-up3.01460.0034th follow-up5th follow-up0.99480.3214th follow-up6th follow-up3.8888<0.0015th follow-up6th follow-up2.89400.004
The GEE analysis assessing the relationship between the time from PPROM to delivery and D-dimer levels indicated a significant negative correlation. The beta coefficient for D-dimer levels was −3.146, suggesting that higher levels of D-dimer are associated with a shorter duration to delivery (Odds Ratio: 0.043, 95% CI: 0.005–0.375, p = 0.004). This finding emphasizes a strong inverse relationship, indicating that as D-dimer levels increase, the time to delivery decreases (Table 5).Table 5Generalized estimated equations of D-dimer levels predicting time from PPROM to birthBetaORCI (95%)p valueD-dimer level−3.1460.0430.005–0.3750.004OR odds ratio, CI confidence interval
Figure 1 illustrates the distribution graph showing the relationship between D-dimer levels and the time from PPROM to delivery in days. The downward trend line indicates a negative correlation, suggesting that higher D-dimer values are associated with a shorter duration to delivery. This visual representation supports the GEE analysis findings, highlighting the potential value of D-dimer levels in predicting the timing of delivery post-PPROM.Fig. 1Correlation between D-dimer levels and time from PPROM to birth
Figure 2 displays the ROC curve comparing the predictive values from the GEE analysis to observed values. The AUC was 0.811 with a 95% CI of 0.737–0.886, indicating the model’s good discriminatory ability. The high AUC value demonstrates that the model’s predictions closely align with observed data, effectively predicting the relationship between D-dimer levels and the time to delivery in PPROM cases.Fig. 2ROC curve for the predicted values from GEE analysis compared to observed values
Currently, PPROM is a significant condition that has adverse effects on both obstetric and neonatal outcomes. It is a well-established fact that 40% of instances involving preterm labor are further aggravated by PPROM. Simultaneously, the biophysical processes that trigger PPROM also induce activation of the myometrium [18, 19]. There are studies on the correct planning of the time of birth in PPROM from both an obstetric and neonatal perspective. In their study, Baser et al. demonstrated that the duration between the diagnosis of PPROM and birth did not have any adverse effects on the outcomes of both the mother and the newborn. Furthermore, they found no rise in neonatal problems after 32 weeks of gestation [20]. Therefore, the standard procedure for these individuals is to keep delivery to 34 weeks, if possible. Nevertheless, the existing literature does not provide any substantiated information to forecast the exact date of arrival. Our study assessed the DD elevations as a means of predicting the date of delivery in patients who sought medical attention at our clinic with a diagnosis of PPROM. On the 3rd day, there was a significant difference in DD between the groups. A significant variation was noticed between the measurements on day 6 and those on all other days. The most prominent elevation occurred between the measurements taken on day 1 and day 6. Furthermore, there was a statistically significant reduction in the duration of birth as the DD level rose. Our results indicate that DD values can be used to predict prognosis in the follow-up of PPROM patients.
PPROM is linked to compromised physiological conversion of the spiral arteries, atherosclerosis, decidual vasculopathy, and thrombosis [21, 22]. One possible explanation for the development of this condition in PPROM patients is the increased buildup of thrombin resulting from the activation of the coagulation system [15]. In the literature, patients with PPROM are usually compared with patients with term birth. Keren-Politansky et al. examined all coagulation factors in patients with preterm birth. They discovered a substantial difference in the levels of PT and APTT between the group of individuals who had given birth at term and the group under study. Nevertheless, their findings demonstrated that there was no statistically significant difference in the alterations of fibrinogen and DD. Nevertheless, when comparing the patients with PPROM to those with premature uterine contractions, there were no noticeable differences in terms of PT, APTT, fibrinogen levels, and DD levels [23]. Chaiworapongsa et al. discovered that thrombin-antithrombin complexes (TAT) were increased in patients with preterm delivery and PPROM, serving as a marker for hypercoagulation [13]. Our study found a negative correlation between the time of birth and the increase in DD, which is a coagulation factor and a sign of enhanced fibrinolysis. The DD value on day 3 was markedly elevated in the group that had delivered within the initial 7 days. The DD values tested on other days exhibited no substantial difference between the two groups. Nevertheless, all changes detected between day 6 and the other days exhibited a noteworthy alteration. Additional coagulation variables were not predictive of the timing of birth. Elevation in DD values shortly before birth indicates their potential utility in predicting the commencement of labor or pregnancy problems. To our knowledge, this is the first time that this situation has been explored in literature. Erez et al. The study investigated the concentrations of tissue factor (TF) and tissue factor pathway inhibitor (TFPI) in individuals with PPROM. The study found that the presence of infection and inflammation did not significantly affect these two factors in severe PPROM patients. Nevertheless, there was a notable alteration in plasma concentrations as compared to pregnant women who had normal levels [15].
High thrombin levels have been shown to be an important marker of inflammation in some studies [24]. Some studies have found evidence supporting the association between preterm delivery and PPROM with maternal intravascular inflammation [25, 26]. This indicates that the elevated level of thrombin in the mother’s bloodstream is a result of widespread inflammation throughout the body. In addition, the presence of elevated levels of thrombin in the tissues leads to an increase in fibrinolysis. As a result, the increase in DD, which is a parameter that indicates this process, can be interpreted as a finding associated with inflammation. This condition corroborates with the rise in DD levels in pregnant patients, in accordance with our study. Patients with PPROM are monitored with a standardized antibiotic regimen [27]. There was no significant change in the levels of WBC and CRP, which are often used to measure inflammation, in these patients due to the administration of antibiotics. In this instance, we believe it would be advantageous to assess DD as an indicator of inflammation and provide insight into estimating the date of childbirth.
Studies have found that abnormal coagulation and fibrinolysis may be associated with increased maternal and fetal morbidity and mortality. Therefore, increased DD levels may be associated with increased pregnancy complications such as hypertension, preeclampsia, and detachment [28, 29]. In their study, Zeng et al. showed that there is a connection between increased DD values and preeclampsia and premature birth, which are pregnancy complications. However, when examining PPROM patients and normal pregnant women, no statistically significant difference was found in the DD values [30]. However, in contrast to this study, we found that monitoring DD values every other day in PPROM patients can help determine the timing of birth. The variability in DD level could potentially have advantages in terms of planning neonatal intensive care services for birth, determining the timing of antenatal corticosteroid administration, and deciding on the usage of neuroprotective magnesium in fetuses that are at risk of being born prematurely due to PPROM.
This study has several limitations. First, the number of patients is limited, and they were collected from a single center. Multicenter studies with a large number of participants can overcome this limitation. Additionally, while we monitored DD levels longitudinally, we did not establish specific cutoff values for each trimester, even though DD levels vary throughout pregnancy. Future studies should consider incorporating trimester-specific reference ranges for DD values to better contextualize the findings. Despite this, our primary focus was on the relative changes in D-dimer levels over time rather than absolute cutoff values. An investigation into DD levels during certain trimesters in cases of PPROM may help to address this limitation. Despite these limitations, a notable advantage of our study lies in its pioneering investigation of the correlation between the gestational age at which PPROM occurs and the duration between delivery and D-dimer levels.
In conclusion, there are studies in the literature on predicting the time of birth and prognosis of PPROM patients. Nevertheless, a reliable indicator capable of furnishing information on this matter has not yet been created. DD is a straightforward, quick, and affordable marker that can be beneficial in terms of both the time of birth and the outcome during the period immediately before and after birth. It is necessary to conduct prospective, randomized, controlled trials on this subject.