Clinical Characteristics and Risk Factors of Coexisting Endometrial Lesions in Patients with Cesarean Scar Defect Undergoing Hysteroscopic Resection for Prolonged Menstruation.

Cesarean scar defect (CSD), also known as isthmocele, is a focal myometrial defect resulting from impaired healing of the uterine incision following cesarean delivery. In recent years, with the rising rate of cesarean sections, the incidence of CSD has gradually increased. It is particularly common among patients with abnormal uterine bleeding, such as prolonged menstruation or intermittent spotting. 1 CSD may lead to various clinical symptoms, including menstrual cycle irregularities, infertility, and chronic pelvic pain. 2 For symptomatic patients who do not respond to pharmacological treatment, hysteroscopic CSD resection serves as a minimally invasive and effective therapeutic option. 3 , 4 In clinical practice, it has been observed that patients with a CSD often present with coexisting endometrial lesions such as polyps and hyperplasia. These conditions may exacerbate symptoms of abnormal uterine bleeding and impact surgical outcomes. For instance, endometrial polyps can contribute to menstrual irregularities, while endometrial hyperplasia is associated with a potential risk of endometrial cancer. 5 Such coexisting pathologies may lead to more complex clinical presentations, increase therapeutic challenges, and introduce confounding factors in postoperative evaluation. However, there is a paucity of systematic research regarding how the presence of endometrial lesions influences the clinical characteristics and postoperative outcomes in patients with CSD. This study aims to retrospectively analyze patients with prolonged menstruation who underwent hysteroscopic CSD resection, comparing clinical features and surgical outcomes between those with isolated CSD and those with coexisting endometrial diseases, so as to provide evidence for precise evaluation and management of CSD.

This single-center retrospective cohort study included 382 women of reproductive age with prolonged menstruation who underwent hysteroscopic CSD resection at International Peace Maternity and Child Health Hospital, School of Medicine, Shanghai Jiao Tong University between January 2023 and December 2024. The inclusion criteria were: preoperative diagnosis of CSD was confirmed by transvaginal ultrasound or MRI, with transvaginal ultrasound used to evaluate the uterus and endometrium, detecting endometrial thickness, structural heterogeneity, or polypoid lesions. When ultrasound findings were unclear or further diagnosis was needed, hysteroscopic visualization or pelvic MRI was employed to further clarify intrauterine lesions and other structural abnormalities. Patients presented with prolonged menstruation or persistent spotting, and had complete preoperative examinations, surgical records, and follow-up data; all had at least one previous cesarean section (lower uterine segment). Exclusion criteria included: concurrent pregnancy, intraoperative discovery of uterine malformations or severe intrauterine adhesions, malignancy, history of menstrual disorders prior to cesarean section, use of intrauterine devices, systemic diseases such as coagulation disorders, and missing key clinical data. Endometrial lesions in this study refer to pathological processes involving abnormal endometrial proliferation, polypoid growth, or structural changes, specifically encompassing endometrial hyperplasia (characterized by disproportionate glandular and stromal proliferation) and endometrial polyps (local neoplastic growths). Based on the presence or absence of these coexisting pathologies, patients were divided into a simple CSD group (n = 208) and a group with coexisting endometrial diseases (n = 174). This study was approved by the Ethics Committee of the International Peace Maternity and Child Health Hospital, School of Medicine, Shanghai Jiao Tong University (No: GKLW-A-2023-026-01). All surgeries were performed during the follicular phase after the cessation of menstruation, under general anesthesia with the patient in the supine lithotomy position. Hysteroscopic procedures were conducted using continuous irrigation with 0.9% sodium chloride solution, maintaining a distension pressure of 80–100 mmHg. A hysteroscopic resectoscope (Olympus, equipped with a 3-mm deep and 5-mm wide loop electrode) was used to examine the uterine cavity and the diverticulum at the lower uterine segment incision. The location, size, and morphological characteristics of the diverticulum and any intracavitary lesions were identified. The loop electrode was then used to resect endometrial lesions and the inferior edge of the diverticulum, followed by electrocautery of the diverticular cavity and surrounding endometrium to improve drainage and promote blood flow restoration. The uterine cavity was curetted multiple times. Resected intracavitary lesions, curettage specimens, and diverticular tissues were sent for pathological examination. Repeat hysteroscopy was performed to ensure no significant bleeding around the incision site before concluding the surgery. Patient clinical data were collected from electronic medical records and telephone follow-up systems, including baseline information (age, number of pregnancies, deliveries, cesarean sections, history of miscarriage, and history of ectopic pregnancy surgery), history of assisted reproduction, and menstrual history (menstrual cycle length, duration of menstruation, number of days of spotting, dysmenorrhea, and VAS score). Imaging parameters such as the superior-inferior diameter, transverse diameter, anteroposterior diameter, residual myometrial thickness, and endometrial thickness of the cesarean scar defect were obtained from preoperative ultrasound or MRI records. The presence of fibroids, adenomyosis, endometriosis, para-defect cysts, intrauterine masses, and preoperative medication use was also recorded. Preoperative therapy included pharmacological interventions (eg, hormonal treatment, iron supplementation) aimed at symptom alleviation, anemia correction, or hemorrhage reduction. Postoperative efficacy was primarily evaluated based on menstrual improvement, according to the criteria described by Zhang et al. 6 The primary outcome was the proportion of patients with a reduction of more than 50% in the number of days of spotting compared to the preoperative baseline, assessed six months after randomization. Menstrual outcomes were obtained through outpatient reviews and telephone follow-ups. Spotting was defined as any of the following: intermenstrual bleeding lasting ≥2 days; brownish discharge lasting ≥2 days immediately after menstruation; or irregular bleeding with total bleeding days (including both menstrual and spotting days) exceeding 9 days. Based on the efficacy evaluation, patients were further divided into a menstrual improvement group and a non-response group for subsequent statistical analysis and modeling. Statistical analysis was performed using SPSS 22.0 software. For continuous variables, normality tests were conducted. Data conforming to normal distribution were expressed as mean ± standard deviation, while non-normally distributed data were presented as median (lower quartile, upper quartile). For between-group comparisons, independent samples t -test was applied when data met assumptions of normality and homogeneity of variance; the adjusted t -test was used when data were normally distributed but with heterogeneous variance; the Mann–Whitney U -test was employed for non-normally distributed data. Categorical variables were expressed as frequency (percentage). For intergroup comparisons of categorical variables, the chi-square test was used when all expected frequencies exceeded 5; the continuity-adjusted chi-square test was applied when expected frequencies were between 1 and 5; Fisher’s exact test was utilized when any expected frequency was below 1. Statistically significant indicators were incorporated into logistic regression analysis to identify independent factors influencing efficacy. The predictive performance of the model was evaluated using the area under the receiver operating characteristic curve (AUC). All tests were two-tailed, with P < 0.05 considered statistically significant. Internal validation of the predictive model was performed using 5-fold cross-validation. The dataset was partitioned into five subsets. In each iteration, four subsets were used for model training and the remaining one for validation. This process was repeated five times. The average AUC was 0.706 (95% CI: 0.674–0.740), with a sensitivity of 74.3% and a specificity of 63.9%, indicating acceptable model stability and predictive performance.

Compared with the simple CSD group, patients in the group with coexisting endometrial diseases had fewer abortions (p = 0.010), a higher proportion of anemia (p < 0.001), lower preoperative hemoglobin levels (p < 0.001), and a lower proportion of preoperative medication therapy (p = 0.002). This group also exhibited slightly larger diverticulum superoinferior and transverse diameters (p = 0.037), significantly increased endometrial thickness (p < 0.001), and higher rates of concomitant uterine fibroids (p = 0.022) and pericystic cysts (p < 0.001). The remaining variables and surgical outcomes showed no statistically significant differences between the two groups, as shown in Table 1 . Table 1 Comparison of Baseline Characteristics Between Groups Related Factors Simple CSD Group (N=208) Group with Coexisting Endometrial Diseases (N=174) p Age (years) 36.0 (33.5, 40.0) 37.0 (34.0, 40.0) 0.559 Gravidity 2.0 (2.0, 3.0) 2.0 (1.0, 3.0) 0.050 Parity (1/2/3/4) 94/108/6/0 86/84/3/1 0.503 Number of cesarean sections (1/2/3/4) 110/94/4/0 101/69/3/1 0.502 Number of abortions 1.0 (0, 1.75) 0 (0, 1) 0.010 History of ectopic pregnancy surgery (0/1/2/3) 196/11/1/0 165/8/0/1 0.757 Assisted reproductive technology (0/1/2) 190/17/1 0/164/10 0.482 Previous CSD surgery (Yes) 11 (5.3%) 5(2.9%) 0.241 Time since last cesarean section (years) 5 (3, 10) 6 (3, 10) 0.423 Interval from last cesarean section to symptom onset 1 (0.5, 3.0) 1 (0.5, 4.0) 0.629 Menstrual cycle length (days) 29.0 (28, 30) 30 (28, 30) 0.822 Menstrual period duration (days) 6 (5, 7) 7 (5, 7) 0.443 Prolonged spotting (days) 7 (6, 10) 9 (5, 9.25) 0.209 Dysmenorrhea (Yes) 58 (27.9%) 61 (35.1%) 0.132 VAS 0 (0, 1) 0 (0, 2) 0.066 Anemia (Yes) 36 (17.3%) 59 (33.9%) <0.001 Preoperative hemoglobin level (g/L) 127 (118.25, 133) 121 (109.75, 130) <0.001 Superior-inferior diameter of CSD (mm) 5.95 (4, 8) 6.2 (4.6, 8.4) 0.037 Transverse diameter of CSD (mm) 13.2 (11, 16) 14 (11.65, 17.125) 0.037 Anterior-posterior diameter of CSD (mm) 5.5 (4, 7) 5.5 (4, 7.05) 0.577 Residual myometrial thickness (mm) 3 (2, 5) 3 (2, 5) 0.867 Endometrial thickness (mm) 6.65 (4.5, 9) 8.4 (6, 11) <0.001 Previous medical treatment (None) 150 (72.1%) 148 (85.1%) 0.002 Uterine fibroids (Yes) 33 (15.9%) 44 (25.3%) 0.022 Adenomyosis (Yes) 15 (7.2%) 18 (10.3%) 0.278 Endometriosis (Yes) 4 (1.9%) 7 (4.0%) 0.222 CSD-related cyst (Yes) 33 (15.9%) 67 (38.5%) <0.001 Surgical outcome (Improved) 158 (76.0%) 137 (78.7%) 0.519 Comparison of Baseline Characteristics Between Groups Preoperative hemoglobin and anemia exhibited collinearity; therefore, only anemia was included in the regression model. Univariate analysis revealed that the number of abortions, anemia, history of previous medical treatment, longitudinal diameter of the diverticulum, transverse diameter of the diverticulum, endometrial thickness, uterine fibroids, and pericystic cysts were significantly associated with endometrial comorbidities (p < 0.05). These statistically significant variables were subsequently included in the multivariate logistic regression analysis. The results demonstrated that anemia (OR = 1.993, 95% CI: 1.186–3.350, p = 0.009), no history of previous medical treatment (OR = 2.013, 95% CI: 1.143–3.544, p = 0.015), endometrial thickness (OR = 1.078, 95% CI: 1.009–1.151, p = 0.025), and pericystic cysts (OR = 3.008, 95% CI: 1.806–5.010, p < 0.001) were independent influencing factors for endometrial comorbidities in patients with CSD, as shown in Table 2 . Table 2 Logistic Regression Analysis of Factors Associated with Coexisting Endometrial Diseases in Patients with CSD Related Factors Univariate Analysis Multivariate Analysis OR (95% CI) p OR (95% CI) p Number of abortions 0.798(0.651–0.979) 0.030 0.845(0.679–1.053) 0.134 Anemia (Yes) 2.451(1.521–3.950) <0.001 1.993(1.186–3.350) 0.009 Previous medication therapy (No) 2.201(1.315–3.685) 0.003 2.013(1.143–3.544) 0.015 Superoinferior diameter of CSD (mm) 1.076(1.008–1.148) 0.027 1.039(0.962–1.122) 0.329 Transverse diameter of CSD (mm) 1.059(1.012–1.108) 0.013 1.031(0.975–1.089) 0.284 Endometrial thickness (mm) 1.130(1.063–1.201) <0.001 1.078(1.009–1.151) 0.025 Uterine fibroids (Yes) 1.795(1.083–2.975) 0.023 1.501(0.865–2.606) 0.149 CSD-related cyst (Yes) 3.321(2.052–5.373) <0.001 3.008(1.806–5.010) <0.001 Logistic Regression Analysis of Factors Associated with Coexisting Endometrial Diseases in Patients with CSD A regression model was constructed to predict the presence of coexisting endometrial diseases in patients with CSD, incorporating factors such as anemia, previous medication history, endometrial thickness, and pericystic fluid around the diverticulum. ROC curve analysis demonstrated that the model exhibited good discriminative ability, with an AUC of 0.710, indicating moderate accuracy in predicting coexisting endometrial diseases, as shown in Figure 1 . Figure 1 ROC Curve of the Predictive Model for Coexisting Endometrial Diseases in Patients with CSD. ROC Curve of the Predictive Model for Coexisting Endometrial Diseases in Patients with CSD. A nomogram was developed to predict the risk of coexisting endometrial diseases in patients with CSD, incorporating variables such as anemia, history of previous medical treatment, endometrial thickness, and pericystic fluid around the diverticulum ( Figure 2 ). Each variable corresponds to a specific score, and a higher total score indicates a greater risk of concurrent endometrial lesions. Figure 2 Nomogram of the Predictive Model for Coexisting Endometrial Diseases in Patients with CSD. Nomogram of the Predictive Model for Coexisting Endometrial Diseases in Patients with CSD.

Patients with a cesarean scar defect (CSD) and coexisting endometrial lesions exhibit more complex intrauterine architecture and a higher risk of abnormal bleeding. Although short-term outcomes are comparable, preoperative identification of endometrial pathologies is crucial for optimizing treatment strategies and tailoring individualized interventions. However, given the single-center nature and inherent limitations of this study, future multi-center research is warranted to further validate the external applicability of our predictive model. Additionally, long-term follow-up studies will be essential to evaluate the model’s utility in predicting long-term postoperative efficacy and recurrence risk, thereby enhancing its robustness for clinical implementation.

In this study, patients with coexisting endometrial diseases exhibited multiple distinct clinical characteristics. Firstly, the prevalence of anemia was significantly higher in this group. Previous studies have reported that some patients with CSD may experience increased menstrual bleeding or prolonged duration. 7 , 8 Secondly, the proportion of patients receiving preoperative medication was lower in the group with coexisting endometrial diseases, suggesting that symptoms may be more insidious or misattributed to “CSD as the sole etiology,” leading to inadequate endometrial-specific treatment. 9 Additionally, endometrial thickness was significantly greater in this group compared to the simple CSD group. Coupled with a higher incidence of uterine fibroids and CSD-related cysts, these findings indicate a more complex uterine structure and broader lesion distribution. Previous studies have also suggested that patients with CSD may exhibit a unique uterine microenvironment, such as altered macrophage polarization. 10–12 Notably, although the group with coexisting endometrial abnormalities showed more structural and symptom-related abnormalities, the menstrual improvement rate at six months post-surgery did not differ significantly from that of the simple CSD group. This suggests that hysteroscopic resection achieves satisfactory outcomes in treating focal diverticulum lesions, with short-term efficacy maintained even in the presence of comorbid conditions. 6 , 13 On the other hand, the relatively coarse assessment criteria for menstrual improvement may fail to fully capture the potential impact of concurrent pathologies on long-term outcomes or recurrence risk. 14 Therefore, preoperative identification of coexisting endometrial diseases remains clinically valuable, as it not only facilitates optimized surgical strategy but also guides postoperative follow-up, adjuvant therapy, and long-term management. Studies have explored the risk factors for CSD and their relationship with clinical symptoms. Multivariate analysis in Liu et al’s study identified CSD width and prolonged menstrual bleeding as independent predictors of postoperative menstrual improvement. 15 Meanwhile, Shi et al reported that a cervical incision location, reduced residual myometrial thickness, and excessive uterine anteversion were associated with surgical difficulty and postoperative recovery, suggesting that anatomical features of the CSD and uterine position may underlie symptom occurrence and recurrence. 16 Unlike previous research, this study focuses specifically on the characteristics of patients with CSD complicated by endometrial diseases, highlighting the importance of endometrial status in understanding and classifying the condition, which warrants identification and intervention prior to surgery. Recent studies indicate that local endometrial defects and chronic inflammation within CSD play a significant role in abnormal uterine bleeding and fertility impairment. Wang et al found that, even after controlling for the degree of myometrial defect, patients with abnormal uterine bleeding exhibited more extensive epithelial loss, exposed blood vessels, and chronic endometritis in the CSD area, suggesting that endometrial abnormalities may be the primary basis for symptom development. 17 Furthermore, Bi et al demonstrated that elevated levels of pelvic inflammatory cytokines in CSD patients may contribute to infertility by altering the local endometrial environment. 18 The present study reveals that patients with CSD combined with endometrial diseases exhibit distinct clinical features, including thicker endometrium, higher rates of anemia, and increased incidence of cysts. It is also the first to identify endometrial thickness and CSD-related cysts formations around the diverticulum as significant predictors in this patient subgroup, underscoring the importance of systematic preoperative evaluation of endometrial status for predicting surgical outcomes and guiding individualized intervention. This study has several limitations. First, as a single-center retrospective study, it is subject to potential selection bias, and the generalizability of the findings needs to be further validated through multicenter research. Second, only hysteroscopy was used as the surgical approach in this study; combined hysteroscopic and laparoscopic procedures were not included for comparison. Previous studies have suggested that combined hysteroscopy and laparoscopy may offer advantages in removing residual tissue and correcting uterine anatomy. 19 , 20 Third, the follow-up duration was relatively short, with menstrual improvement assessed only at 6 months postoperatively. This study lacks long-term observation regarding endometrial healing and reproductive outcomes. Fourth, the predictive model, while informative, was developed and validated internally on a single-center cohort. Its generalizability and clinical utility require further external validation in diverse populations and settings. Furthermore, the definition of postoperative menstrual improvement was broad and did not account for individual variations or the influence of lesion severity in detail.