Discussion
In our previous study, we published a nomogram for the preoperative prediction of surgical treatment efficacy for CSD. We referred to Glavind’s studies that defined the class A healing group as having a TRM no less than 5.8 mm after vaginal repair13. Only 167 patients were included in the training cohort. Due to the limited number of cases, previous research only identified risk factors for failure to achieve class A healing17. Our study has improved the shortcomings and built the prediction model system.
Our research shows the timing of C-sections, the menstrual cycle, and the length, width, depth, and thickness of the lower uterine segment of the CSD can predict whether CSD repair surgery can achieve optimal healing. The results are similar to Liu’s research, which shows that the number of C-sections, a trial of labor (elective or urgent C-sections), C-section interval, and uterine position were independent risk factors for CSD18. Interestingly, we find that elective C-section surgery is a risk factor for CSD, but they respond better to repair surgery when CSD occurs. We think that the reason for this may be the trial process of myometrial stretching and contraction is a protective factor for wound healing; however, there may be potential biological mechanisms that lead to suboptimal healing in patients who develop CSD after the trial process such that conditions are not conducive to healing of the incision when the CSD is repaired.
We also find that when the menstrual cycle is greater than 26 days, optimal healing is more easily achieved after surgery. This means that patients with long menstrual cycles (>26 days) have better surgical outcomes, which is consistent with our previous finding that oral contraceptives or GnRHa can improve menstrual conditions. We think GnRHa can prolong the menstrual cycle or reduce the menstrual volume to a certain extent, so patients with a long menstrual cycle can heal relatively well after surgery. It is possible that chronic inflammation caused by menstrual stimulation can hurt wound healing. Additionally, our results show that a CSD width of greater than 13.8 mm will influence surgical efficacy, particularly the improvement of postoperative menstruation. Previous case studies found that CSD repair with a CSD width ≥18.85 mm had a poor prognosis15, which was consistent with the index in our prediction model.
On the other hand, we find that not all patients sought to achieve optimal healing by treatment. Some patients may only expect to improve menstrual conditions, while others will expect to increase the TRM to reduce the risk of pregnancy but do not care much about PAUB. Therefore, we also discuss the factors that can improve menstruation or increase TRM merely. The results show that the posterior uterus or TRM ≥2.79 mm is the independent correlation factor for TRM improvement. However, CSD width ≤13.27 mm or the thickness of the lower uterine segment ≤22.67 mm preoperatively can predict PAUB symptom improvement.
Many studies have confirmed that the presence of a CSD might increase the risk of subsequent cesarean scar pregnancy. One study showed that repair surgery both reduced the incidence of cesarean scar pregnancy and increased the number of live deliveries19. This finding reflects complete correction of the CSD by suturing, which strengthens the uterine wall and promotes either complete or partial healing of the uterine scar after surgery. Therefore, for patients with subsequent pregnancy requirements, the recommendation is to evaluate whether optimal healing can be achieved or whether TRM can be improved after surgery by a prediction model.
We hope to use the prediction model to enable more appropriate treatment options for patients with CSD. We recommend vaginal repair surgery as the first choice when the score is ≥ 16.5 because this approach is less damage, lower cost, and a shorter hospital stay. When the score is less than 16.5 suggests that the patient will have difficulty achieving optimal healing through vaginal repair. In this case, when the patient has no desire for a subsequent pregnancy, hysteroscopic resection surgery can be recommended. Hysteroscopic resection is considered the least invasive among the possible operations; however, hysteroscopy may have the potential risk for decreased resistance of the residual myometrial tissue at the level of the repair and may lead to uterine rupture during subsequent pregnancy20-22. Therefore, hysteroscopic surgery is not recommended for patients with pregnancy intentions. Additionally, the TRM should not be less than 3 mm, given the anticipated risk of perforation or bladder injuries23-24. For patients with a score<16.5 without pregnancy requirements, younger patients can be recommended to use oral contraceptives or an intrauterine device (IUD). The levonorgestrel intrauterine system may have a role in the safe and effective management of PAUB in patients with CSD. The study shows that after 1 year of treatment, 22 patients (78.6%) reported an improvement in symptoms with the levonorgestrel intrauterine system. The mean duration of menstruation is 18 and 5 days before and after treatment, respectively25. This is a good choice for patients who just want to improve their menstruation.
If the score is less than 16.5 and the preoperative residual muscle layer is less than 2.16 mm, but the patient has a strong desire to become pregnant, she should be fully informed of the possibility of a poor outcome before surgery. If the patient has a history of multiple operations and abdominal adhesions are considered preoperatively, pelvic adhesion decomposition can be performed by a single-port laparoscope that can be released to restore the normal anatomical structure, and then vaginal surgery can be performed to avoid excessive anterior or posterior flexion of the uterus, which may be helpful to improve postoperative recovery. Adding oral contraceptives or GnRHa to prolong menstruation after surgery may also help to improve the prognosis. We have demonstrated the effectiveness of GnRHa in another study, and the article has been submitted (data unpublished, under review).
Our study is the first prediction model performed in China to assess the surgical efficacy of CSD. The major strength of our study is determining which patients with CSD should be repaired or other treatment options according to the score. Several limitations should be considered when interpreting the results of this study. First, this is a retrospective study, which has inherent biases. Second, this study comes from a single institution, and the nature of the database is a strength. Although consistency in surgical techniques is ensured, there are limitations in patient selection. Third, alternative treatment options and their effectiveness should be further validated. In addition, our study requires an external validation cohort to make the results more convincing, and the inclusion of such information is planned in our future studies.