Risk factors and the possibility of predicting late premature birth

Introduction

Late preterm delivery (LPD) or premature birth (PB) includes births at 34–36⁺⁶ weeks of gestation. Currently, in the structure of PBs, the share of LPD reaches up to 70%. The growth of LPD has continued over the last 15 years, which is associated with an increase in extragenital diseases, multiple pregnancies resulting from assisted reproductive technologies, and intrauterine fetal distress [1]. The infectious morbidity of “near-term” newborns is significantly lower compared to this indicator in premature newborns born at earlier stages, and prolongation of pregnancy at this stage does not have a significant impact on the mortality rates of newborns [2].

Although late preterm deliveries are considered “close to full term”, they pose serious risks to the health of the mother and baby. Understanding risk factors and predictive capabilities is important to prevent preterm birth at 34–36⁺⁶ weeks and improve pregnancy outcomes.

In this paper, we present a review of the literature over the last 6 years, including domestic and foreign scientific studies, meta-analyses, and systematic reviews related to LPD. We analyzed risk factors, which are important in the genesis of preterm birth, as well as criteria for predicting the onset of late preterm birth.

Risk factors, which may contribute to LPD, include gestation complications such as severe preeclampsia [3], premature detachment of the normally located placenta, placenta previa/acreta, vaginal bleeding, premature rupture of membranes (PRM), fetal anomaly [4], fetal growth restriction [3], preterm birth or abortion in history, inadequate antenatal care, stress, smoking, psychoactive substance using, maternal age <18 years or >40 years, poor nutrition, low body mass index, infectious agents [3], environmental factors [3], and this list is not exhaustive. In addition, maternal risk factors for LPD include short stature (<160 cm), secondary (poor) education, social unemployment, rural residence, and smoking during pregnancy [5].

These data are consistent with the research of Russian scientists, who consider the following factors to be among the most important predictors: PB in history, irregular monitoring during pregnancy, pelvic inflammatory diseases in history, smoking, obesity, the onset of sexual activity before the age of 16, the presence of multiple pregnancies [6], age under 18 and over 40, and low family income [7]. Besides, the risk group involves women who had comorbid conditions such as cardiovascular and endocrine diseases, infections, inflammations, shortening of the cervix and placentation disorders [6], intrahepatic cholestasis [8], pregnancy complicated by diabetes, hypertension, preeclampsia, or eclampsia [7]. Other studies have found a relationship between preterm birth and autoimmune diseases such as psoriasis [9] and systemic lupus erythematosus, depending on disease activity [10].

In China (2021), the examination of more than 9.5 million women revealed that 75% of them gave birth at 34 to 36.6/7 weeks, and the fact of LPD was significantly affected by maternal education, maternal age, marital status, and antenatal visits [11].

A high risk of LPD is also associated with the effect of psychological factors on the course of pregnancy. Important reasons for the increase in prenatal stress are lifestyle changes including urbanization, healthy behavior, physical activity, employment, working conditions, and use of tobacco, alcohol, and illegal drugs [12]. High scores on the Perceived Stress Scale-10 and the level of maternal prenatal stress may be putative predictors of PB [13].

The review by Yakovleva and Glukhova [14] assessed the efficacy of predicting factors stipulating the onset of preterm birth. Musaleva et al. (2020) have developed a comprehensive prognostic scale for assessing the risk of preterm births, which makes it possible to classify a pregnant woman into a high-risk group for preterm birth with a high degree of probability (sensitivity was 81%, specificity was 84.6%) [15]. The most significant risk factors for LPD are presented in Table 1.

Based on the literature data, an approximate portrait of a patient with LPD has been compiled. This is a woman under 18 or over 40 years old, with a deficit or excess of body weight, without higher education, with PB in history, most likely associated with infection and surgical interventions on the cervix. During the pregnancy, the woman had complications such as preeclampsia, abnormal placentation, diabetes, and various infections. She did not regularly attend antenatal clinics, experienced chronic stress or depression, had a low financial income, and used various psychoactive substances.

Of interest are recent studies on the prediction of LPD.

First of all, they concern the transvaginal ultrasound examination of the cervix. A short cervix (less than 25 mm) is considered [16] a predictor of PB. Cervicometry carried out over a period of 16–24 weeks in dynamics with an interval of two weeks makes it possible to identify pregnant women at risk of premature termination of pregnancy [17]. Dynamic cervicometry performed 24–48 hours after the onset of tocolytic therapy allows identifying a group of patients with a cervical length of more than 25 mm, who can be discharged in 12–24 hours after the completion of the treatment course. This tactic eliminates the possibility of long-term hospitalization and maintains the final result; moreover, the frequency of early termination of pregnancy does not increase [18]. The probability of PB increases at a threshold length of the cervix of 15 mm compared to its length of 25 mm [19].

Conducting elastography to determine tissue stiffness and density and assess the consistency of the cervix can also be used to predict preterm birth [16][20][21]. The combination of methods of cervical length measurement and density assessment enables a better assessment of the risk of spontaneous preterm birth before 37 weeks of pregnancy compared to cases with normal cervical consistency [21].

Concurrently, there is no clear evidence on the ability to accurately predict LPD even among women with cervical shortening and/or cervical dilation in the presence of symptoms of cervical insufficiency. The study “Features of Spontaneous Late Preterm Labor and Late Preterm Birth” (2020) [22] analyzed 732 women with cervical shortening, namely less than 3 cm, and/or cervical dilation, and revealed that 58.9% of them experienced LPD. Assessment of cervical length in the third trimester in combination with maternal factors may also improve the prediction of late PB [23]. However, this is inconsistent with the results of a study based on the Cochrane Pregnancy and Childbirth Registry, as well as on ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform, which found that data on the effect of cervical length measured by transvaginal ultrasound were limited in predicting and preventing preterm birth [24].

Thus, it can be concluded that considering only the condition of the cervix does not allow for the accurate prediction of the LPD probability.

Much attention is currently paid to the investigation of the relationship between high concentrations of anti-inflammatory cytokines and polymorphism of genes that affect the development of preterm labor [14][25][26]. In particular, the investigation of the cffDNA biomarker was described as a simple and non-invasive test for predicting preterm labor [27].

Against the background of systemic inflammation, fetal cell-free DNA (Egfp) concentrations are increased at preterm birth. However, this does not apply to preterm births resulting from intra-amniotic inflammation [28]. Therefore, the value of fetal cell-free DNA in predicting preterm birth continues to be debated [29].

The “Australian protein” (a sum of cytokines) and the genes encoding it are considered a trigger for PB. They participate in the initiation of the process of decollagenization and changes in the structure of the cervix [1]. Perhaps in the future, the determination of this marker will allow predicting the onset of PB [30].

In the future, a more detailed investigation of the relationships between different indicators derived from the above methods can be used to predict precisely LPD.

At the present stage, the most developed methods of predicting are biochemical markers.

It is known that fibronectin concentrations correlate with the onset of preterm birth and are used to assess the probability of preterm labor. Actually, this test has a good negative predictive value; when its values are negative, the risk of spontaneous preterm labor is minimal within 7 days [16].

The combination of cervicometry targeted at revealing a reference point for the less than 15 mm length of the closed part of the cervical canal and a qualitative determination of fibronectin has a higher positive prognostic value for the onset of PB in the next 7 days [19] than using only one test.

A high correlation was noted between elevated plasma IL-6 levels and the onset of spontaneous preterm labor within 48 hours. In addition, IL-6 concentration correlated with the plasma C-reactive protein level [31]. Moreover, the combination of elevated IL-6 and IL-8 levels in cervicovaginal secretions and cervicometry was prognostically significant for preterm delivery within 7 days [16].

Another popular concept is the investigation of vitamin D-binding protein, which is a predictor of intra-amniotic infection. The examinations showed that at a threshold value of 1.76 μg/ml, the risk of delivery within 48 hours with intact amniotic fluid increased. It is worth noting that, in the case of PRM, the concentration of vitamin D-binding protein is always high and does not correlate with the onset of labor [32].

The same study by Kook et al. (2018) revealed the predictive value of serum relaxin and determined that in preterm birth, its level was significantly higher. However, the level of relaxin in the blood serum did not correlate with the results of cervicometry [32]; this fact confirms that the issue has not been sufficiently studied.

There are studies on placental alpha microglobulin-1 (PAMG-1) as a marker for predicting preterm birth. Testing conducted in pregnant women with a cervical length of 15–30 mm established 100% sensitivity and 94% specificity of this method for predicting late preterm birth within 7 days [33][34].

A comparison of the efficacy of the qualitative test for determining PAMG-1 and the quantitative determination of fibronectin in leucorrhoea at threshold values of 10, 50, and 200 ng/ml showed that the PAMG-1 test was more reliable with maintaining a high negative predictive value for the onset of preterm labor within 7 days [35].

A comparison of the results of using PAMG-1, fibronectin, and phosphorylated insulin-like growth factor (phIGFBP-1) tests, as well as cervicometry, made it possible to conclude that the efficacy of PAMG-1 was significantly superior to phIGFBP-1 or fibronectin in predicting the onset of preterm labor within 7 days [34,35].

The combination of the fact of detecting an increased fetal adrenal volume and PAMG-1 has a sensitivity of 82.8%, a specificity of 27.2%, and can be used to calculate the risk of preterm birth within 7 days [36]. The analyzed data are presented in Table 2.

The presented data confirm the importance of PAMG-1, fibronectin, cervicometry, and elastographic ultrasound examination of the cervix as the most informative tests-predictors of the LPD onset.

Conclusion

We believe that the correct assessment of the risks of preterm birth, including LPD, and tactics based on the use of informative and “rapid” tests for assessing the threat of PBs will improve their outcomes. Risk assessment using several methods for predicting the onset of preterm birth seems optimal. However, this issue requires further investigations and clinical testing.