Value and influencing factors of preoperative MRI evaluation for previous cesarean scar defect associated abnormal uterine bleeding in patients undergoing laparoscopic surgery

Abstract

Objective

As the cesarean section rate increases year by year, the treatment of previous cesarean scar defects (PCSD) poses a significant challenge. This study aims to evaluate the clinical value of preoperative magnetic resonance imaging (MRI) technology and analyze relevant influencing factors for patients with abnormal uterine bleeding (AUB) associated with cesarean scar defects who underwent laparoscopic surgery.

Methods

A retrospective cohort analysis was performed on women who underwent laparoscopic surgery for PCSD-related AUB at the Department of Gynecology, the Third Xiangya Hospital of Central South University from 2018 to 2022. A total of 57 patients who underwent laparoscopic surgery for the treatment of AUB associated with PCSD were divided into 2 groups based on their postoperative clinical cure status: The clinically-cured group (n=28, 49.1%) and the non-clinically-cured group (n=29, 50.9%). After a postoperative follow-up period of 3 months for all participants, logistic regression analysis was conducted to explore the correlation between the clinical cure rate of AUB associated with cesarean scar defects treated by laparoscopic surgery and various factors. These factors included patient age, clinical symptoms, obstetric history, history of cesarean section, basic clinical information, preoperative MRI parameters, and postoperative menstrual conditions.

Results

There were no significant differences in many aspects, including the patient’s age at the time of previous cesarean section, number of pregnancy, time since the previous cesarean section, the uterus position assessed by preoperative T₂ signal MRI, defect length, defect width, residual muscle layer thickness, adjacent uterine muscle layer thickness, and distance from the defect to the external cervical os between the 2 groups (all P>0.05). However, the time of onset of AUB symptoms (P=0.036, OR=1.019, 95% CI 1.002 to 1.038) and the depth of the defect on the preoperative MRI (P=0.010, OR=5.793, 95% CI 1.635 to 25.210) were identified as risk factors affecting the clinical cure rate.

Conclusion

The time of onset of AUB symptoms and the depth of the defect on preoperative MRI are risk factors that influence the clinical cure rate of laparoscopic surgery for the treatment of AUB associated with PCSD, which could be helpful for evaluating the prognosis of disease.

According to the research of the World Health Organization (WHO), the cesarean section delivery rate continues to rise worldwide^[1]. During the past decade, the Chinese government has strived to decrease the cesarean section delivery rate, whereas the overall national rate still maintains a high level^[2-3]. The inappropriate cesarean section delivery is associated with various short-term and long-term complications in women, ranging from increased risks of infection, bleeding, pelvic adhesions, to abnormal placentation, uterine rupture, stillbirth, preterm birth, and hysterectomy in the subsequent pregnancy^[4].

Previous cesarean scar defect (PCSD), as one of the long-term complications, is the presence of a niche in the myometrium of the anterior lower uterine segment at the site of previous cesarean section, which has been demonstrated as a pathogenic factor of abnormal uterine bleeding (AUB), dysmenorrhea, chronic pelvic pain, and secondary infertility^[5-6]. AUB, the main clinical manifestation of PCSD, is early-cycle intermenstrual bleeding or prolonged menstrual duration, which plays a significant negative role on the quality of life in women after cesarean delivery^[6-9].

PCSD is usually evaluated by transvaginal ultrasonography (TVS) because of its efficiency and economy^{[5, 10]}. However, magnetic resonance imaging (MRI) is able to scan larger areas in the pelvis for further scrutiny, which could provide a more accurate and clear view in the preoperation evaluation.

Laparoscopic repair surgery has been usually advocated for patients with large defect or fertility demand^[11], which is also applied to the PCSD patients in our hospital. As we have previously reported^[12], the clinical efficacy of AUB with laparoscopic treatment is 92.3%. Despite the promising results, not all the patients could benefit from the surgery.

So far, studies on MRI for PCSD diagnosis and prognosis are rarely reported. This study aims to evaluate the clinical value of preoperative magnetic resonance imaging (MRI) technology and analyze relevant influencing factors for patients with AUB associated with cesarean scar defects who are undergoing laparoscopic surgery.

1. Patients and methods

1.1. Patients

All the patients (n=85) with PCSD-related AUB who underwent laparoscopic surgery in the Department of Gynecology, the Third Xiangya Hospital of Central South University from 2018 to 2022 were enrolled in this study. The inclusion criteria were as follows: 1) PCSD was diagnosed by hysteroscopy; 2) patients with symptomatic AUB; 3) normal ovarian function and ovulation. The exclusion criteria included: 1) Patients with other identified causes of AUB; 2) patients who suffered from acute general inflammation or severe postoperative infection; 3) patients with uncontrolled systemic disease; 4) patients with malignant tumors or precancerous lesions; 5) patients with lactation; 6) lost to follow-up. A total of 57 patients met the criteria and were enrolled in this study. A detailed flow chart of participant selection is presented in Figure 1.

Figure 1. Flow chart of patient selection PCSD: Previous cesarean scar defect; AUB: Abnormal uterine bleeding.

1.2. Ethical statement

This study was performed in accordance with the Declaration of Helsinki (as revised in 2013). The Ethics Committee of the Third Xiangya Hospital of Central South University approved this study protocol (No. 2020-S625). The diagnosis and treatment of all the participants were performed under standard procedures. All participants were fully informed of the surgical procedures, the benefits, potential risks, and outcomes of the treatment before the surgery. All participants provided informed consent.

1.3. MRI examination

All the participants underwent MRI (Ingenia 3.0T, Philips Co., Eindhoven, Netherlands) evaluation by T₂-weighted images before surgery. The parameters, including the defect length, defect width, defect depth, residual muscle layer thickness, adjacent myometrial thickness, and distance from the defect to cervix external os, were measured (detailed description in Figure 2) by the same radiologist. Defect length, defect depth, residual muscle layer thickness, myometrial thickness adjacent to the defect, and distance from the defect to cervix external os should be calculated in the sagittal plane, and the transverse plane was used only for width.

Figure 2. Defect measurement by T₂-weighted MRI before surgery.

A: Defect length (L), the defect depth (D), the residual muscle layer thickness (TRM), the adjacent myometrial thickness (MTA), and the distance from the defect to external cervix os (DTEO) measured on MRI at the sagittal plane. B: Defect width (W) measured on MRI at the transverse plane of the same patient.

1.4. Laparoscopic surgery procedures

All the participants underwent laparoscopic surgery within 1 week after menstruation. The detailed surgery procedures were described in our previous publication^[12]. The peritoneal fold was incised from the bladder by an ultrasonic scalpel (Ethicon Inc., Somerville, NJ, USA) and bipolar coagulation forceps (Kangji Co., Hangzhou, Zhejiang, China), which was pushed 2 cm away from the lower edge of the diverticulum. The defect location and size were identified by an assistant with hysteroscopy, which was then completely opened from one side to the other laparoscopically by the surgeon. The fibrotic tissue was carefully excised from the defect. Full-thickness suture for wound closure was performed with a bidirectional absorbable suture (Ethicon Inc., Somerville, NJ, USA). Hysteroscopy was performed to visualize the repair of the defect at the end of the surgery.

1.5. Follow-up method

All the participants were followed up for 3 months after the surgery. The data were collected from medical records and telephone interviews, including patient’s age, clinical manifestations, obstetric history, history of cesarean section, basic clinical information, preoperative MRI parameters, and postoperative menstrual conditions. The clinical efficacy of postoperative menstruation was evaluated based on clinical cure, improvement, and ineffectiveness. Clinical cure was defined as no postmenstrual spotting after the surgery. Improvement was defined as shortened postmenstrual spotting. Ineffectiveness was defined as no obvious change in menstruation after the surgery. The participants were divided into a clinicall-cured group and a non-clinicall-cured (improvement or ineffectiveness) group according to the duration of postoperative menstruation.

1.6. Statistical analysis

Statistical analysis was performed with the software Statistical Analysis System 9.4 (SAS Institute, Cary, NC, USA). Categorical data were presented as frequency and percentage in each group, and analyzed by the chi-square test. Normally distributed data were expressed as mean±standard deviation (SD), and analyzed by the Student’s t-test. Non-normally distributed data were shown as median and range and analyzed by the Mann-Whitney U test. The continuous variables of MRI parameters before the surgery were divided by median. A logistic regression analysis was applied to determine the dominant variables. P<0.05 was considered statistically significant.

2. Results

2.1. Patient characteristics

There were 28 clinically cured (49.1%) and 29 non-clinically cured (50.9%) patients in a total of 57 participants included in this study. The characteristics of the 57 participants before the surgery are described in Table 1. There were no significant differences in age, age of previous cesarean section, gravidity, and time since the previous cesarean section between the 2 groups (all P<0.05). Parity, number of cesarean section, and time of onset of AUB symptoms were shown to be significantly correlated with the clinical cure rate (P<0.05, Table 1).

Table 1.

Clinical characteristics of participants before surgery

Groups	n	Age/years	Age of the previous cesarean section/years	Gravidity/[No.(%)]			Parity/[No.(%)]
				1	2	≥3	1	≥2
Clinically-cured group	28	35.6±3.1	28.4±3.4	9(32.1)	7(25.0)	12(42.9)	16(57.1)	12(42.9)
Non-clinically-cured group	29	34.2±3.4	27.4±4.7	4(13.8)	10(34.5)	15(51.7)	8(27.6)	21(72.4)
t/χ2		1.65	0.841	2.769			2.259
P		0.052	0.202	0.25			0.024

Groups	Number of cesarean section/[No.(%)]		Time since the previous cesarean section/months	Time of onset of AUB symptoms/months
	1	≥2
Clinically-cured group	17(60.7)	11(39.3)	86.4±51.0	53.3±39.9
Non-clinically-cured group	9(31.0)	20(69.0)	77.6±31.2	69.9±31.9
t/χ2	5.058		0.772	1.712
P	0.025		0.222	0.046

AUB: Abnormal uterine bleeding.

2.2. MRI parameters

The parameters of the defect assessed by T₂-weighted MRI before the surgery are shown in Table 2. No significant differences were shown in the uterus position, the defect length, defect width, residual muscle layer thickness, adjacent myometrial thickness,or distance from the defect to the external cervical os beween the clincally-cured group and non-clinically cured group before laparoscopic surgery (all P>0.05). The defect depth was significantly related to the clinical cure rate postoperatively (P<0.05, Table 2).

Table 2.

MRI parameters of participants before surgery

Groups	n	Uterus position /[No.(%)]		Defect length/ [No.(%)]		Defect width/ [No.(%)]		Defect depth/ [No.(%)]
		Anteflexed	Retroflexed	<10 mm	≥10 mm	<20 mm	≥20 mm	<8 mm	≥8 mm
Clincally-cured group	28	10(35.7)	18(64.3)	12(42.9)	16(57.1)	13(46.4)	15(53.6)	19(67.9)	9(32.1)
Non-clincally-cured group	29	13(44.8)	16(55.2)	9(31.0)	20(69.0)	16(55.2)	13(44.8)	12(41.4)	17(58.6)
χ2		0.492		0.856		0.436		4.026
P		0.483		0.355		0.509		0.045

Groups	residual muscle layer thickness/[No.(%)]		Adjacent myometrial thickness/[No.(%)]		Distance from the defect to the cervix external os/[No.(%)]
	<2.5 mm	≥2.5 mm	<10 mm	≥10 mm	<30 mm	≥30 mm
Clinically-cured group	16(57.1)	12(42.9)	11(39.3)	17(60.7)	14(50.0)	14(50.0)
Non-clinically-cured group	17(58.6)	12(41.4)	12(41.4)	17(58.6)	13(44.8)	16(55.2)
χ2	0.013		0.026		0.153
P	0.910		0.872		0.696

2.3. Risk factors for PCSD prognosis

Univariate and multivariate logistic regression are illustrated in Table 3 and Table 4 to analyze the factors that could affect the PCSD prognosis after laparoscopic surgery. The risk factors for the clinical cure rate were analyzed by univariate logistic regression in Table 3. Compared with the clinically-cured group, patients in the non-clinically-cured group gave birth to more children (P=0.026, OR=3.500, 95% CI 1.188 to 11.030) and performed more cesarean sections (P=0.027, OR=3.343, 95% CI 1.179 to 10.640). The defect depth was greater in the non-clinically-cured group (P=0.026, OR=3.455, 95% CI 1.187 to 10.690). Other variables showed no differences between the 2 groups (all P>0.05).

Table 3.

Univariate analysis of the clinically-cured group and non-clinically-cured group

Variables	Estimate	SE	Wald	P	Odds ratio	95% confidence interval
Age	-0.138	0.086	2.566	0.109	0.871	0.727 to 1.025
Age of the previous cesarean section	-0.055	0.064	0.716	0.398	0.947	0.830 to 1.073
Gravidity
1
2	1.168	0.777	2.256	0.133	3.214	0.732 to 16.120
≥3	1.034	0.715	2.091	0.148	2.813	0.723 to 12.550
Parity
1
≥2	1.253	0.5643	4.928	0.026	3.500	1.188 to 11.030
Number of cesarean section
1
≥2	1.234	0.558	4.897	0.027	3.434	1.179 to 10.640
Time since the last cesarean section	-0.005	0.006	0.606	0.436	0.995	0.982 to 1.007
time of onset of AUB symptoms	0.013	0.008	2.779	0.096	1.013	0.998 to 1.029
Uterus position
Anteflexed
Retroflexed	-0.380	0.543	0.490	0.484	0.684	0.232 to 1.975
Defect length/mm
<10
≥10	0.511	0.554	0.850	0.357	1.667	0.566 to 5.050
Defect width/mm
<20
≥20	-0.351	0.532	0.435	0.510	0.704	0.245 to 1.994
Defect depth/mm
<8
≥8	1.240	0.557	4.955	0.026	3.455	1.187 to 10.690
Residual muscle layer thickness/mm
<2.5
≥2.5	-0.061	0.537	0.013	0.910	0.941	0.326 to 2.709

Table 4.

Logistic regression of significant variables in the clinically-cured group and non-clinically-cured group

Variables	Estimate	SE	Wald	P	Odds ratio	95% confidence interval
Intercept	-3.095	0.975	10.087	0.002	0.045	0.006 to 0.260
Parity (≥2)	1.375	1.854	0.551	0.458	3.957	0.093 to 194.400
Number of cesarean section (≥2)	0.693	1.823	0.145	0.704	2.000	0.047 to 89.240
Time of onset of AUB symptoms/months	0.019	0.009	4.423	0.036	1.019	1.002 to 1.038
Defect depth (≥8 mm)	1.757	0.685	6.569	0.010	5.793	1.635 to 25.210

AUB: Abnormal uterine bleeding.

Table 3.

Variables	Estimate	SE	Wald	P	Odds ratio	95% confidence interval
Adjacent myometrial thickness/mm
<10
≥10	-0.087	0.540	0.026	0.872	0.917	0.315 to 2.653
Distance from the defect to the cervix external os/mm
<30
≥30	-0.208	0.531	0.153	0.696	0.813	0.284 to 2.305

AUB: Abnormal uterine bleeding.

continued

A multivariate logistic regression in Table 4 was analyzed based on the meaningful variables from the univariate analysis. The data showed that the time of onset of AUB symptoms (P=0.036, OR=1.019, 95% CI 1.002 to 1.038) and the defect depth (P=0.010, OR=5.793, 95% CI 1.635 to 25.210) were risk factors for the clinical cure rate.

3. Discussion

The major clinical symptoms of PCSD, which is the long-term complication of cesarean delivery, are abnormal uterine bleeding, dysmenorrhea, chronic pelvic pain, and secondary infertility^[13]. Various risk factors for PCSD have been investigated, including increasing numbers of cesarean deliveries, single-layer myometrium closure without endometrial suture, prolonged duration of active labor, and wound infection^{[5-6, 14-15]}. Furthermore, maternal body mass index and gestational diabetes are also associated with PCSD^[13-15].

AUB mainly presents postmenstrual spotting or prolonged menstrual duration in PCSD patients^{[6, 15]}. The potential mechanism of PCSD-related AUB involves the residue of menstrual blood or mucus in the defect, resulting from the lack of muscle contractility around the scar and the poor drainage of menstrual flow through the fibrotic tissue below the niche^{[6, 9, 11, 16]}. In addition, blood production within the defect because of inflammation, neovascularization, and/or adenomyosis also contributes to the characteristic menstrual pattern^{[6, 11, 16-17]}.

Several surgical treatments have been applied in PCSD patients with AUB, ranging from hysteroscopic, laparoscopic, to transvaginal surgery^[13]. Laparoscopic surgery enables surgeons to explore the pelvis and remove adhesions, which could allow us to explore other factors of pain or sterility^{[11, 13]}. In previous published studies^{[11-12, 18-19]} involving our research, the most common complications of the laparoscopic approach were bladder injury and hemorrhage. So far, there are no clear surgical treatment criteria for patients with PCSD-related AUB. In our previous study^[12], the clinical efficacy of AUB after laparoscopic surgery is high. Although the treatment outcome is encouraging, a subset of patients still suffered from persistent AUB without improvement after operation. In this study, 49.1% of the participants (n=28) with no more postmenstrual spotting after the surgery were clinically cured. In the meantime, 50.9% of the patients (n=29) showed shortened postmenstrual spotting or no obvious change of menstruation, which was classified as a non-clinical-cured group. We wondered what on earth could play a decisive role in the prognosis of PCSD patients.

PCSD could be evaluated by TVS, MRI, and hysterosalpingopraphy^{[10, 20]}. TVS, with or without gel or saline contrast^{[10, 20]}, could measure the defect length, defect depth, defect width, residual myometrium thickness, adjacent myometrium thickness of uterine scar defect, and distance from the defect to external os in non-pregnant women, which is the first recommendation for evaluation because of its efficiency and economy^{[5, 10]}. However, because of the higher image resolution, MRI is a more reliable and accurate evaluation method than TVS^[21]. Besides, MRI is operator-independent and could provide reproducibility of measurements for gynecologists. In our study, we applied MRI to PCSD patients and evaluated the effect of factors related to the prognosis of PCSD patients who underwent laparoscopic surgery for the first time. Through the logistic regression models, we realized that time of onset of AUB symptoms and the defect depth are risk factors for clinical efficacy. The reason may be related to the fact that the longer time of onset of AUB symptoms, the more scar tissue around the scar, which leads to poor contractility of the uterine muscles around the cesarean scar over time. As the defect depth increases, the defect has a greater chance of having a larger diverticulum. Previous study^[22] has shown that a larger and deeper defect displayed on MRI is associated with longer menstrual bleeding. The key and most difficult part of laparoscopic surgery is to excise the scar tissue around the defect and re-suture the healthy tissue in the anterior lower uterine wall. The persistent AUB in patients postoperatively may be due to abundant fibrotic tissue around the large defect, which is difficult for us to distinguish from healthy tissue during the surgery.

One of the limitations of this study is retrospective in design, and a larger number of cases and more stratifications according to the basic characteristics of patients are further needed to provide more information. In addition, postoperative MRI parameters could be further studied to analyze the factors that could influence the clinical cure rate of patients in the future.

In conclusion, the time of onset of AUB symptoms and the depth of the defect on preoperative MRI are risk factors that influence the clinical cure rate of laparoscopic surgery for the treatment of AUB associated with PCSD.

Funding Statement

This work was supported by the Young Talents of Medical Science of Hunan Province, China (2022RC1218).

Conflict of Interest

The authors declare that they have no conflicts of interest to disclose.

AUTHORS’CONTRIBUTIONS

ZHANG Qi, LIN Changwei Research design, statistical analysis, thesis writing and revision; WU Jiaoyang Research design, data collection, thesis writing and revision; XU Dabao and ZHU Shujuan Research design, data analysis and interpretation, thesis writing and revision; JIANG Bin Research design, data collection and analysis, thesis writing and revision. The final version of the manuscript has been approved and read by all authors.

Note

http://xbyxb.csu.edu.cn/xbwk/fileup/PDF/2023091316.pdf