A total of 78 subjects were included. There were 20 nulliparous women (25.6%) with a mean (SD) age of 26.16 (1.33) years, 31 with one previous CS (39.7%) with a mean (SD) age of 31.46 (0.96) years, and 27 subjects (34.6%) who had undergone CS at least twice and had a mean (SD) age of 32.5 (0.99) years. A significant difference existed among the groups regarding age (P< 0.001). Mean gestational age in groups 1, 2, and 3 was respectively 38, 37.26, and 37 weeks (P= 0.12).

There was no significant difference regarding mean (±SD) birth weight among the three groups (3,400 (±327.26) gr in group 1, 3,253.35 (±379.81) in group 2, and 3,247.35 (±388.25) in group 3); P= 0.3. Mean BMI values in groups 1, 2, and 3 were respectively 29.93, 29.89, and 29.25 kg/m2 (P= 0.79).

Mean (±SD) LUS thickness in groups 1, 2, and 3 was respectively 6.05 (±2.5), 5.33 (±1.33), and 4.49 (±1.54) mm (P= 0.01). Range of LUS thickness in groups 1, 2, and 3 was 1 to 11 mm, 3 to 8.5 mm, and 0.8 to 7.3 mm.

Three patients (9.7%) in group 2 has dehiscence during CS. Mean (±SD) LUS thickness in these three patients was 4.40 (±0.36) mm. In group 3, two patients (7.4%) experienced dehiscence during CS with a mean (±SD) LUS thickness of 1.2 (±0.6) mm. There was a significant difference regarding mean LUS thickness between groups 2 and 3 who experienced dehiscence (P= 0.03).

Paper-thin LUS was documented in 4 patients (12.9%) of group 2 with mean (±SD) LUS thickness of 4 (±0.81) mm. This finding was seen in more patients of group 3 (11 cases, 40.7%) with a mean (±SD) LUS thickness of 3.44 (±0.75) mm. Uterine rupture occurred in only one patient who was in group 3 whose LUS thickness was 2.5 mm. This was not observed by ultrasound and rupture was diagnosed during CS.

Ultrasound showed dehiscence in only one patient in the second group whose LUS thickness was 3 mm. However, three more patients in group 2 were diagnosed with rupture during CS with LUS thickness values of 4, 4.5, and 4.7 mm. In group 3, two patients were diagnosed to have rupture by ultrasound. LUS thicknesses of these two patients were 0.8 and 2.5 mm. These were confirmed during CS.
Cut-off value to predict uterine dehiscence and rupture was 1.7 mm with a sensitivity of 78% and specificity of 76% (Figure 1).

Click here for Table 1: Apgar scores at minutes 1 and 5 in the three studied groups

Figure 1: Receiver operating curve for lower uterine segment thickness of 1.7 mm with sensitivity of 78% and specificity of 76% for predicting uterine dehiscence and rupture

DISCUSSION

Based on the obtained findings, those who had previous CS had significantly thinner LUS. This resulted in dehiscence and rupture in these patients. On the other hand, none of the nulliparous women with thicker LUS experienced dehiscence or rupture. The neonates’ birth weight did not show difference among groups, so it is highly likely that dehiscence and rupture occurred due to thinner LUS. The obtained results are in agreement with some previous reports. In a study involving 106 patients with previous CS and 68 without, LUS was thinner in the first group with a mean value of 4.58 mm than in the second group (4.8 mm) (16).

Ultrasound can detect dehiscence by showing a defective area where no myometrial layer is seen (17). In this study, in patients with more than one previous CS, US findings were in agreement with findings during CS. The cut-off value we obtained here (1.7 mm) is very close to the reported value by a previous study (1.8 mm) (18). However, some studies have proposed higher values at 2.5 to 3.5 mm among patients with previous CS (11).

Although we observed dehiscence and rupture in patients with LUS thickness of more than 3 mm, one patient who experienced rupture had a LUS thickness of 2.5 mm. A previous study showed that none of the patients with LUS thickness of <3 mm experienced dehiscence or rupture (9). In a former meta-analysis of about 2,700 patients, sensitivity and specificity for cut-off values for LUS thickness to predict uterine defects was 76% and 92% for values between 0.6 and 2 mm (19).

Ultrasound is a non-invasive method to measure LUS thickness and its ability to predict dehiscence and rupture has been investigated previously (9, 10). One of the limitations in this study was that we were not able to gather all details about previous CS. Although CS per se is considered a risk factor for scar formation and thinner LUS, other factors can also have a role in LUS thickness. In a previous study, maternal age of more than 35 years, single layer uterine closure, and non-elective CS were factors to be associated with LUS thickness (12). All these factors can affect healing of the LUS after CS and influence the integrity of LUS.

Limitations
We intended to determine the effect of multiple previous CS on LUS thickness, and it was found that LUS was thinner in those with multiple CSs, however as the rate of dehiscence and rupture was a secondary objective; the sample size was not large enough to achieve a conclusion in this regard. Future studies with larger sample size can answer the question of the effect of multiple CSs. Another limitation is that we were not able to perform transvaginal ultrasound as some studies have demonstrated that transvaginal ultrasound provides better information about myometrial thickness than transabdominal ultrasound (20). However, this may not be regarded as a significant limitation as there is evidence of more than 90% correlation between trans-abdominal and transvaginal ultrasonography and a cut-off value of 2.5 mm (21).

CONCLUSION

LUS thickness was significantly lower in pregnant mothers with previous CS and this led to dehiscence in such patients. In case of LUS thickness of < 1.7 mm, the risk of dehiscence and rupture increases.

Acknowledgment
This paper was taken from the thesis of Irandokht Alimohammadi as a requirement to receive PhD in Obstetrics and Gynecology from Kermanshah University of Medical Sciences