ROLE OF BIOCHEMICAL MARKERS AND TRANSVAGINAL ULTRASOUND IN THE PREDICTION OF PRETERM LABOR

By

Farid Ibrahim Hassan, Emad Maarouf Abdel Latif,

Omar Salem Abdel Aziz Mubarak, and *Salah Ahmed El Beltagi

Departments of Obstetrics & Gynecology and *Clinical Pathology,

Faculty of Medicine, Al- Azhar University.

ABSTRACT

Background :  Evaluation of cervicovaginal fetal fibronectin was considered as one of the most predictive factor for preterm labor when it is 50 ng/ml. Also, Salivary estriol (E3) has the same predictive factor for preterm labor when it is over 2.1ng/ml. The last factor in this study was measurement of cervical length.

Objective: Assessing cervicovaginal fetal fibronectin and maternal Salivary estriol (E3) as biochemical markers, as well as transvaginal ultrasound in the prediction of preterm labor.

 Patients and Methods: The study included 150 pregnant women at gestational ages between 24 weeks and 34 weeks confirmed by sure history of last menstrual period and first trimestric ultrasound examination.The patients in this study were fulfilling the inclusion criteria, and were enlisted in a prospective clinical trial. As regard cervicovaginal fetal fibronectin, measure of it as Dacron swab was rotated in posterior fornix of the vagina or ectocervix and sent for assay of fetal fibronectin.

 Salivary estriol (E3) was collected from mouth after washing it, and stored till time of assay. Cervical length was evaluated by transvaginal ultrasound.

Results: There was a highly significant difference between women with positive and negative fibronectin regarding gestational age at delivery. Salivary estriol (E3) showed    a highly significant difference between women with positive and negative E3 regarding gestational age at delivery. Also, cervical length showed a highly significant difference between women with short and normal cervical length regarding gestational age at delivery. There was no significant effect for age, parity, presence of uterine contractions and uterine anomalies on the outcome of pregnancy .

Conclusion: Cervicovaginal fetal fibronectin, salivary estriol E3 and measurement  of cervical length were more helpful in prediction of preterm labor.

INTRODUCTION

    The diagnosis of spontaneous preterm labor and accurate prediction of preterm delivery is notoriously difficult. Identification of effective risk assessment markers can potentially improve outcomes by enabling targeted therapy while allowing efficient use of resources and avoiding unnecessary interventions. Considerable progress has been made in the development of accurate methods as fetal fibronectin (FFN) and cervical length assessment to predict preterm birth (PTB) in both symptomatic and asymptomatic high risk women (Barber et al., 2010).

     As risk assessment tools, both history and CL assessment have their limitations, but when used in conjunction with biochemical markers (such as FFN), prediction is improved. The Preterm Prediction Study demonstrated that if risk factors (previous SPB, CL less than 25 mm and positive FFN) were present, the risk of SPB before 32 weeks was 50 % (Hutcheon et al., 2012).

    Of the three major placental estrogens, estriol (E3), estradiol (E2), and estrone (El), E3 is the most abundant during late pregnancy.  The feasibility of measuring salivary E3 in detecting increased risk of preterm labor has been studied. Salivary E3 level at or above 2.1 ng/mL predicted preterm labor  which is more accurately than traditional risk assessment methods (Sotiriadis et al., 2010).

    The clinical use of fetal fibronectin sampling was significantly improved with the introduction of a rapid, semiquantitative testing protocol, the Tli System (Adeza Biomedical, Sunnyvale, California). Before introduction of the Tli, analysis required specimen transportation to a central laboratory and ELISA testing.

    Multiple studies have provided evidence that a positive fetal fibronectin is                  a redictor of preterm delivery in patients presenting with preterm uterine contractions. A negative fetal fibronectin test identifies patients at low risk for preterm delivery, although a positive test has a limited positive predictive value (Souka et al., 2011).

    Ultrasonographic evidence of increased internal cervical os diameter has been reported by various researchers as being consistent with cervical incompetence. Mclaren et al. (2015) have reported that width of the internal cervical os greater than 1.9 cm was suggestive of cervical incompetence. In another study, has observed that a mean internal os diameter of about 1.6 cm or more was consistent with cervical incompetence (Lim et al., 2009). However, these studies did not consider the effect of progressive gestational age on the diameter of the internal os. And associates stated that an internal os width greater than 1.5 cm in the first trimester or greater than 2 cm during the second trimester indicates incompetence (Peng et al., 2015).

    The present study aimed to assess cervicovaginal fetal fibronectin and maternal salivary estriol (as biochemical markers) as well as transvaginal ultrasound in the prediction of preterm labor.

PATIENTS AND METHODS

    This study was a prospective study conducted at Al Azhar University Hospital and Shibin El Khom Teaching Hospital between April 2015 and April 2017.

    The study included 150 pregnant women at gestational ages between 24 weeks and 34 weeks confirmed by a sure history of last menstrual period and first trimestric ultrasound examination.                                                                                                   

Inclusion criteria: Women in the study were selected with risk factor(s) of preterm birth including:   

    Current evidence of threatened preterm labor diagnosed by the occurrence of two or more uterine contractions during 10 minutes or less with cervical dilation of 4cm or less, and intact membranes with or without cervical effacement. 

Known cervical incompetence by early ultrasound.

● Previous preterm birth.

● Multiple pregnancies.

● Known uterine anomaly such as septate or bicornute uterus.

● Previous cervical surgery (trachelor-raphy or amputation).

Exclusion criteria: Women of ruptured membranes.

Methodology:

I. Cervicovginal fetal fibronectin:  Each subject was first examined with sterile vaginal speculum. A Dacron swab was rotated in the posterior fornix of the vagina or ectocervix, and sent for assay of fetal fibronectin. The presence or absence of fetal fibronectin was measured by qualitative test and results were expressed as positive or negative.

II. Salivary estriol (E3) : Women were instructed to wash their mouth with water and to wait for 10 minutes to allow saliva without stimulus, then  saliva was collected  into standard plastic jars  during day time hours ( 9 am – 8 pm ) to avoid diurnal variation.

    The patient was instructed to abstain from eating, drinking, smoking or chewing gum for at least one hour before specimen collection and specimens frozen within 4 hours after collection and stored at – 20º C till time of assay.

III. TVU Cervical length Measurement: It was performed using 7.5 MHZ endovaginal transducer (voluson p8).  

    A standardized measurement of cervical length was followed:

1. The patient  emptied her bladder, she then lie in the dorsal  lithotomy position.

 2.  Endovaginal probe was covered with a condom, then inserted.

3. The probe was placed in the anterior fornix of the vagina.

4. A sagittal view of the cervix was obtained showing the long axis view of echogenic endocervical mucosa along the length of the canal.

5. The probe was withdrawn until the image became blurred, and then was re-advanced  just enough  to restore the clear  image.

6. The image was enlarged until the cervix occupied at least two-thirds of the screen, and the internal and external os were well seen.

7. The cervical length was measured from the internal to the external os along the endocervical canal.

8. At least, 3 measurements were obtained and  the shortest was taken.

9. Gentle transfundal pressure was applied for 15 seconds, and  the cervical length  was measured again at least 3 times.

10. The examination  lasted at least 5 minutes. The shortest measurement was only recorded for clinical management.

Statistical analysis:Data were collected and entered to the computer using SPSS (Statistical Package for Social Science) program for statistical analysis, (version 20; Inc., Chicago. IL USA).

Chi-square test was used to measure association between qualitative variables. Student's t-test was used to compare mean and SD of 2 sets of quantitative normally distributed data.

Pearson's correlation was used to study correlation between two variables having normally distributed data. Sensitivity, specificity, +ve and –ve predictive values, and diagnostic accuracy were calculated.

P-value  ≤ 0.05 was considered significant .

RESULTS

Mean age of the studied group was 28.05±4.4, 78% of the studied group were multipara, 66% of the studied group were pregnant in single baby, 56% of the studied group had uterine contractions and 79.3% of the studied group had no uterine anomalies (Table 1).

Table (1): Descriptive data of the studied group.

     Thirty six percent of the studied group had positive fibronectin, and 37.3% of the studied group had positive E3 (Table 2).

Table(2): Cervicovaginal  fetal  fibronectin and  salivary E3 .

     62.7% of women in the study had short cervix

Table (3): Cervical length.

     There was a highly significant difference between women with positive and negative E3 regarding gestational age at delivery (Table 4).

Table (4): Relation between E3 result and gestational age at delivery.

    There was a significant relationship between outcome of pregnancy and number of fetuses as 19.6% of women with twins pregnancy completed their pregnancy, while 80.4% of women with twins pregnancy delivered preterm. This difference was statistically highly significant. On the other hand, there was no significant difference in the outcome between women with singleton pregnancy. Additionally, there was no significant effect for age, parity, presence of uterine contractions, or uterine anomalies on the outcome of pregnancy (Table 5). 

Table(5): Outcome of pregnancy and patient variables.

      There was a significant relationship between +ve salivery E3 and preterm birth as 96.4% of women with +ve salivery E3 delivered preterm in contrast to only 3.6% who delivered at term. When salivery E3 was –ve  there was no statistical difference in the outcome (Table 6).

Table (6): E3 of the studied groups and outcome of pregnancy.

            No. of cases were in between brackets.

     There was a significant relationship between fetal fibronectin and delivery within 2 weeks and more (Table 7).

Table (7): Fibronectin of the studied group and delivery within 2 weeks and more.

      Fibronectin was with sensitivity of 47%, specificity of 82%, positive predictive value(PPV) of 81%, negative predictive value(NPV) of 49%, and its accuracy of 61%, salivary estriol(E3) with sensitivity of 58%, specificity of 100, PPV of 100, NPV of 72, and its accuracy of 80, cervical length with sensitivity of 68%, specificity of 86, PPV of 89%, NPV of 62, and it's accuracy of 75% (Table 8).

Table (8): Sensitivity and specificity of fibronectin, E3 and cervical length in predicting PTD.  

    There was non-significant negative correlation between cervical length and fibronicten also there is a significant positive correlation between fibronectin and E3 (Table 9).   

Table (9): Correlation between salivary E3, fetal fibronectin and cervical length.

     There was a significant positive correlation between cervical length and E3 (Table 10).    

Table (10): Correlation between salivary E3 and Cervical length.

    There was a significant relationship between Uterine contractions and outcome of pregnancy single – twins (Table 11).

Table (11): Uterine contractions of the studied group and outcome of   pregnancy.

     There was  a significant relationship between  short cervix and preterm birth as 91.5% of women with short cervix (<25mm) delivered preterm in contrast to only 8.5% who delivered at term.Similary, 87.5% of women with normal cervical length (>25mm) delivered at term in contrast to only 12.5% who delivered preterm(Table 12).

Table (12): Cervical length of the studied group and outcome of   pregnancy.

     There was  a highly significant difference between women with positive and negative fibronectin regarding gestational age at delivery (Table 13).

Table (13) : Relation between fetal  fibronectin and gestational age at  delivery.

DISCUSSION

    Fibronectin is often found in the cervicovaginal secretions before 16-18 weeks of pregnancy, and again at the end of normal pregnancy as labor approaches. It is not normally present in cervicova-ginal secretions between 22 and 37 weeks. Fibronectin found in cervicovaginal secretions after 22 weeks is a marker of disruption of the decidual-chorionic interface. It has been associated with a six-fold increased risk of preterm birth before 35 weeks and a 14-fold increased risk of preterm birth before 28 weeks (Malamud, 2011) . 

    In the present study, the mean gestational age of delivery for cases with positive fibronectin was 32.6|± 4.62|  weeks, while the  mean gestational age of  delivery for cases with negative fibronectin was 35.5|±5.01| weeks  with a statistically  significant difference .This mean gestational age for cases with positive fibronectin give chance for adjuvants measurements.

   In the present study, it was found that positive cervicovaginal  fibronectin (level ≥ 50 ng /ml) was predictive of preterm birth  with sensitivity of 47% , specificity of 82%, positive predictive value of  81%, and negative predictive value of 49%. This was in contrast with  the study of  Owen and Melissa (2012) who showed that positive cervicovaginal  fibronectin (level ≥ 50 ng /ml) was predictive  of preterm birth with  a  sensitivity of 73% , specificity of 72% , positive   predictive value of 25 %, and negative predictive value of 95%.

   In the present study,  increase in salivary E3 was observed  2-3 weeks before delivery and  this was in  agreement with Mclaren et al. (2015) who showed that E3 had a surge  approximately 3 weeks before delivery.

    Although digital examination of the cervix is largely imprecise, ultrasound evaluation of cervical length and membrane status can provide useful and reproducible informations. Abdominal ultrasound attempts at cervical evaluation are significantly affected by both habitus and bladder filling  which, however, do not affect endovaginal imaging. Thus, an appropriate endovaginal image can be obtained in more than 95% of patients. In  the present study,  cervical lengths of greater than 25 mm at 25 weeks' gestation had a negative predictive value of 87.5% for preterm birth. Further, as cervical length decreases, the positive predictive value and sensitivity for preterm delivery increase. This was in agreement with study of Lim et al. (2009) who showed that CL greater than 30 ml at 24 weeks' gestation had a negative predictive value of 97 %.

    In the current  study, transvaginal ultrasound was more sensitive ( ≥ 60%) in  women with increased risk of preterm birth which included women with history of preterm birth or mainly uterine contractions or previous cervical conization.  This was in agreement with the study of  Craigo ( 2011) who showed that  transvaginal ultrasound is more sensitive( ≥ 50%) in  women with previous cervical conization.

    In this  study, positive fetal fibronectin (50 ng/ml or more) in patients with symptoms suggestive of preterm labor has been associated with increase in birth within 7-14 days of test,  but there was no agreement with study of  Heinne et al. (2007)  Who showed that the positive predictive value for delivery within week was 18% .

    In the present study, minimal increases in fFN(l - 54 ng/mL) were associated with a modest increase in the relative risk of sPTD at late preterm GA  ( 34 and 35 weeks). Moderate increases in fFN (55-60 ng/mL) were associated with significant increase in sPTD risk at earlier and wider GA (from through 37 weeks), and the increase  levels of fFN (60 ng/mL) were associated with the highest in prediction of PTD.  There was no agreement  with Peng et al. ( 2015) who showed that minimal increase was at 1- 49 ng/mL, moderate at  50-199 ng/mL,  and highest more than 200 ng/mL.

    In the present  study,  the mean  gestational age  at delivery for cases with positive E3 was 30.8 ±2.62 weeks, and the mean gestational age  at  delivery for cases with negative E3 was 34.50 ±4.1 weeks with a statistically significant difference.

    In current  study,  it was observed that there  was a significant positive  correla-tion between E3 and fibronectin. The increase in cervicovaginal fetal fibronectin  was associated with increase in  salivery E3 in a strong  correlation. On the other hand, there was non significant negative  correlation  between  cervical length and  fibronectin.

CONCLUSION

  Cervicovaginal fetal fibronectin, salivary estriol E3 and measurement  of cervical length were more helpful in prediction of preterm labor.

REFERENCES

1. Barber MA, Eguilu I, Plasencia W, Medina M, Valle L and Garcia JA (2010): Preterm delivery and ultrasound measurement of cervical length in Gran Canaria, Spain. International Journal of Gynecology and Obstetrics, 108;  58–60.

2. Craigo SD (2011): Indicated Preterm Birth for Fetal Anomalies Semin Perinatol. The Journal of Maternal fetal And Neonatal Medicine, 35:270-276.

3. Heinne RR,  McGregor JA  and  Dullien VK  (2007): Accuracy of salivary estriol testing compared to traditional risk factor assessment in predicting preterm birth. Am J Obstet Gynecol.,180:S214-8 .

4. Hutcheon JA,  Skoll MA,  Eastabrook GD and  Lim KI (2012):  The Case for Universal Cervical Length Screening to Prevent Preterm Birth: Is it Strong Enough to Change Practice in Canada? J Obstet Gynaecol Can., 34(12):1184–1187.

5. Lim K, Kimberly VB, Crane JM and John’s NL (2009): Ultrasound cervical measurement and prediction of spontaneous preterm birth in ICSI pregnancies: a prospective controlled study. RBM on line , 18. (2); 296-300.

6. Malamud D (2011):  Saliva as a Diagnostic Fluid. Dent Clin N   Am., 55: 159–178.

7. McLaren JS,  Hezelgrave NL, Ayubi H, Seed PT and Shennan AH (2015) : Prediction of spontaneous preterm birth using quantitative fetal fibronectin after recent sexual intercourse. Am J Obstet Gynecol., 212:89.e1-5.

8. Owen j And Melissa M (2012): Cervical Cerclage for the Prevention of Preterm Birth. Obstet Gynecol Clin N Am., 39: 25–33.

9. Peng CR,  Chen CP, Wang KG,  Wang LK , Chen C and Chen Y (2015): The reliability of transabdominal cervical length measurement in a low-risk obstetric population: Comparison with transvaginal measurement and trans-abdominal measurement. Am J Obstet Gynecol ., 208: 233.el-6.

10. Sotiriadis A, Papaeheodorou S, Kavvadies K and Makrydima G (2010): Transvaginal cervical length measurement for prediction of preterm birth in women with threatened preterm labor: a meta-analysis. Ultrasound Obstet Gynecol., 35:54–64.

11. Souka AP,  Papastefanou I,  Michalitsi V, Salambasis K, Chrelias C,  Salamalekis G and Kassanos D (2011): Cervical Length Changes From the First to Second Trimester of Pregnancy, and Prediction of Preterm Birth by First-Trimester Sonographic Cervical Measure-ment. Ultrasound Med ., 30:997–1002.