Sonographic imaging of cervical scars after Cesarean section

Authors


Abstract

Objective

To investigate whether uterine contractions at the time of a Cesarean section have an impact on future presence and location of a cervical Cesarean scar.

Methods

A targeted transvaginal ultrasound examination of the fetus, uterus and cervix was done in 2973 consecutive women at 14–16 weeks' gestation. The sonographer was blinded to the women's previous obstetric histories. The presence and location of a sonographic cervical hypoechogenic line, which probably represented a Cesarean scar, was recorded.

Results

There were 180 women with a previous Cesarean section performed before the start of uterine contractions and 173 with a Cesarean section performed during contractions in labor. The cervical hypoechogenic line was more common in sections performed during contractions (75.7% vs. 52.7%; P < 0.001) and was more distally located from the internal os (17.9 ± 9.4 vs. 14.6 ± 9.1 mm; P = 0.01). A hypoechogenic line was observed in 21/2620 women without a previous Cesarean section, representing a false-positive rate of 0.8%.

Conclusion

Cesarean sections, especially those done during uterine contractions, are actually performed through cervical tissue. This finding is in agreement with the physiological process of cervical effacement during contractions. Copyright © 2004 ISUOG. Published by John Wiley & Sons, Ltd.

Introduction

Cesarean sections are usually performed by incision of the lower uterine segment. Sonographic studies have revealed various changes in the anterior uterine wall following the operation1–9. It has been suggested that uterine rupture is more common in cases with a sonographically thin uterine wall10, 11. In contrast to the literature on the uterine wall, a MEDLINE search revealed no published data on the sonographic image of the uterine cervix after Cesarean section.

Two remarkable changes occur in the uterine cervix during labor, namely effacement and dilatation. The term effacement refers to the process of shortening and softening of the cervix. It has been suggested that the effaced cervix actually incorporates into the lower uterine segment and becomes part of this segment12. However, there are no well-defined histological or sonographic data to confirm this clinical process. Since Cesarean sections are performed in the lower uterine segment, cervical tissue is probably incised during the operation.

The present study evaluated the sonographic image of the cervix in pregnant women who had a history of a previous Cesarean section. The aim of the study was to investigate whether uterine contractions at the time of the Cesarean section have an impact on future sonographic visualization and location of a cervical scar.

Methods

A targeted ultrasound examination is customary in Israel. In a pilot study we noted that in many cases of a previous Cesarean section, ultrasound examination revealed a hypoechogenic line in the cervix perpendicular to the cervical canal. We assumed that this line was a marker of a cervical Cesarean scar (Figures 1–3). The current prospective study enrolled 2973 consecutive pregnant women who were evaluated during a 1-year period. The women were Caucasian from the middle and upper socioeconomic classes, and all were under an obstetrician's care in pregnancy. Transvaginal sonographic examinations were performed at 14–16 weeks' gestation. The standard practice at our center is to do a complete survey of all fetal organs, the uterus and cervix. All examinations were performed by the same observer using a 7.5-MHz annular array vaginal transducer (Elscint Ltd, Haifa, Israel). Patients were requested not to disclose the mode of previous deliveries when their medical histories were taken prior to the ultrasound examination. During the transvaginal examination the woman's abdomen was covered and the sonographer was therefore blinded to the possibility of an abdominal scar. After completion of the ultrasound examination patients were asked to provide details on the mode of previous deliveries. Patients who had a previous Cesarean section were specifically asked about the type of operation (elective or non-elective), the timing of the operation (before the start of uterine contractions or during uterine contractions in active labor), the cervical dilatation at the time of the section, fetal presentation, birth weight, gestational age and time interval from the previous Cesarean section to the current pregnancy.

Figure 1.

Cervical images in a woman without a previous Cesarean section. (a) Image showing the demarcated cervix with all the cervical parts displayed. (b) The same cervix without demarcation. A.F, amniotic fluid; Ex/ext os, external os; In os, internal os; L., lower; U., upper.

Figure 2.

Cervical images in a woman with a previous Cesarean section. (a) Image showing the demarcated cervix. The small arrows point to the hypoechogenic line. (b) The same cervix without demarcation. Ext os, external os; Int os, internal os.

Figure 3.

Cervical images in a woman with two previous Cesarean sections. (a) Image showing the demarcated cervix. The small arrows point to two hypoechogenic lines. (b) The same cervix without demarcation. Ext os, external os; Int os, internal os.

In the present study period a search for a cervical hypoechogenic line was done in all pregnant women. In cases where a hypoechogenic line was visualized, its location was recorded and its distance from the cervical internal os was measured. Thereafter, the presence and location of the cervical hypoechogenic line was correlated to the timing of the Cesarean section, uterine activity and obstetric history.

SPSS software (SPSS Inc., Chicago, IL, USA) was used for data analysis. Categorical variables were compared with the Chi-square test, the t-test was used to compare the means of continuous variables and the Mann–Whitney U-test was used to compare ordinal variables between two groups. A value of P < 0.05 was considered statistically significant.

Results

Of the 2973 patients who were scanned, 353 (11.9%) had previous Cesarean sections. There were 180 women who were operated on before the start of uterine contractions and 173 women who were operated on during uterine contractions in labor.

The following variables had an impact on the visualization and location of the cervical hypoechogenic line: uterine contractions, cervical dilatation and gestational age. Depiction of the hypoechogenic line was significantly more common in Cesarean sections performed in labor during contractions as compared to operations performed before the initiation of contractions and labor (Table 1). Furthermore, the line was located significantly further away and distally from the internal cervical os in cases of uterine activity at the time of operation when compared to cases with no uterine activity (17.9 ± 9.4 vs. 14.6 ± 9.1 mm; P = 0.01). A significant relationship was noted between the degree of cervical dilatation at the time of operation and the ability to visualize the hypoechogenic line (Table 2). The gestational age had a significant impact only in patients who were operated on before the initiation of contractions and labor. The hypoechogenic line was more commonly observed in those patients who were operated on at a more advanced gestational age (38.6 ± 2.5 vs. 37.3 ± 3.1 weeks; P < 0.004). In contrast, there was no effect of gestational age in patients who were operated on during contractions in labor.

Table 1. Visualization of a cervical hypoechogenic line in women with previous Cesarean sections*
Visualization of hypoechogenic lineCesarean section without contractions (n = 180) (n (%))Cesarean section during contractions (n = 173) (n (%))
  • *

    Chi-square test. P < 0.001.

Yes95 (52.7)131 (75.7)
No85 (47.3) 42 (24.3)
Table 2. Relationship between visualization of the cervical hypoechogenic line and cervical dilatation at operation*
Cervical dilatation (cm)nVisualization of hypoechogenic line (n (%))
  • *

    Chi-square test for trend. P < 0.0001.

 <118096 (53.3)
1–3 5132 (62.7)
4–7 3628 (77.8)
 >8 8670 (81.4)

Fetal position (vertex or non-vertex), birth weight or interval in years from Cesarean section to present pregnancy had no significant effect on the incidence or location of the cervical hypoechogenic line in both groups of patients.

Overall, there were 43 women with two previous Cesarean sections and three women with three previous Cesarean sections. Only one hypoechogenic line was observed in these patients who had more than one Cesarean section. However, after completion of the study, we had one woman with two previous Cesarean sections in whom we saw two hypoechogenic lines (Figure 3). A hypoechogenic line was observed in only 21/2620 patients who had no history of a previous Cesarean section. This gives a false-positive diagnosis of 0.8%. No sonographic defects were observed in the uterine wall in any of the patients with previous Cesarean sections.

Discussion

The study demonstrates that the cervical sonographic hypoechogenic line that seems to be a marker of a Cesarean section scar is more common and more distally located on the cervix in cases of Cesarean sections performed in active labor. This finding is in accordance with the normal physiological process of cervical effacement and the cervix being taken up into the lower uterine segment during labor.

Evaluation of our data reveals three main issues that should be addressed, namely (1) no uterine wall defects were observed in our patients, (2) the cervical hypoechogenic line was not seen in all patients who were operated on during uterine contractions and (3) the cervical hypoechogenic line was noted in about half of the patients who had no contractions at the time of the Cesarean section.

Previous studies have focused on the sonographic changes in the uterine body but did not address the uterine cervix. Chen et al.4 evaluated both pregnant and non-pregnant women after a Cesarean section. Sonographic findings were observed in about half of the patients in both groups. Thickening of the previous incision site was the most common finding in pregnancy. They also noted thinning, ballooning and wedge defect. In the non-pregnant women, a wedge defect was the main finding. Thurmond et al.7 evaluated premenopausal women with abnormal uterine bleeding. They noted a gap in the anterior lower uterine segment in 9/310 patients who had a previous Cesarean section. Kirkinen et al.3 studied 15 women with three or more Cesarean sections. The typical findings were thinning of the myometrium and ventral ballooning of the isthmus. Monteagudo et al.8 performed saline infusion sonohysterography in 44 non-pregnant women with gynecological disorders. They noted a triangular, anechogenic, filling defect that resembled a ‘niche’ in cases of a previous Cesarean section. The authors also reported an anecdotal case of a similar finding in a pregnant woman.

We did not find sonographic evidence of uterine wall defects in our patients. However, our patients were evaluated at 14–16 weeks' gestation while other studies evaluated non-pregnant women or women in advanced gestation. It is therefore possible that the decidual changes occurring in early pregnancy obliterated small filling defects in the uterine wall. Furthermore, it is well known that dehiscence of uterine scars, in addition to spontaneous rupture of scarred uteri, usually occurs only in the second half of pregnancy when the uterine wall is stretched and over-distended.

As noted earlier, the hypoechogenic line was not detected in all cases of Cesarean section performed in labor. Three possible explanations may be proposed to explain this observation. First, the study was conducted in pregnant women, such that the normal physiological changes that occur in the cervix during pregnancy probably obscured visualization of this sonographic marker. Second, the ultrasound examination may not be sensitive enough to detect all hypoechogenic lines. Third, the lower uterine segment has quite a large area. The exact place of incision differs between different patients and different obstetricians. It is therefore possible that some incisions performed closer to the internal os will include more cervical tissue while other incisions performed higher in the uterine segment and away from the internal os will include less cervical tissue and more myometrial tissue.

The lack of multiple lines in the vast majority of women with multiple Cesarean sections is of interest. Two main explanations may be offered. First, some of the repeated Cesarean sections were elective operations performed before the initiation of labor. Second, it is possible that an acoustic barrier of one cervical scar evades the ability to observe a nearby second scar.

A cervical hypoechogenic line was detected in about half of the patients who had no contractions at the time of the Cesarean section. This is in agreement with the normal physiology of pregnancy, since cervical effacement may start prior to active labor. Our data further confirm this process by showing that visualization of the hypoechogenic line was more common with advanced gestational age, namely closer to term. In contrast, there was no significant effect of gestational age in cases of Cesarean section performed in labor. This finding is not surprising since cervical effacement normally occurs during labor independent of gestational age.

There was no effect of fetal weight or position at the time of operation on the visualization of the hypoechogenic line. This is in agreement with the fact that the size of the presenting fetal part has an effect on the configuration of the uterus but has no effect on the cervical effacement.

The clinical implications of the present study need further evaluation. For instance, it would be of interest to discover if there is any correlation between a cervical scar and the success rate or failure of trial of vaginal birth after a Cesarean section.

In summary, the present study shows that the chance of an incision through cervical tissue is higher in cases of Cesarean section performed during uterine contractions in labor. This finding is in agreement with the physiological process of cervical effacement and incorporation of the cervix into the lower uterine segment. The term ‘high cervical transverse incision’ might therefore be suggested in such cases.

Ancillary