Reproductive outcomes after laparoscopic resection of symptomatic niches in uterine cesarean scars: Long-term follow-up on the prospective LAPNICHE study.

JH and HB contributed to the design of this study. JH, HB and WH performed the laparoscopic niche resections. JV AND RVE analyzed the data, and all the other authors contributed to the interpretation of the data. JV wrote the draft, and CB, WH, HB, RVE and JH subsequently made substantial contributions to it. All the authors critically revised the paper and give their approval for this version to be published.

Overall, 232 with a large niche were identified in the inclusion period, of whom 159 women underwent a laparoscopic niche resection. The 55 women who were excluded did not meet our inclusion criteria, one women received transvaginal niche repair and 17 women received hysterectomy. In total, 133 of the 159 women had a desire to become pregnant at baseline and were included in the current analyses (Figure  4 ). Of the 133 included women, 88 (66.2%) were infertile at baseline, with 58 (43.6%) of them having previously undergone failed ART. A total of 45/133 (33.8%) had a desire to become pregnant and gynecologic symptoms but were not infertile at baseline (see Table  1 ). Outcomes after laparoscopic niche resection at 2 years of follow‐up. Baseline characteristics women with a desire to become pregnant ( n  = 133). Subgroup 1 women with infertility ( n  = 88) and 2 women without infertility ( n  = 45) were compared at baseline. Note : Data are reported as mean ± SD, median (interquartile range) or n (valid %). Baseline characteristics are presented in Table  1 . The majority (85%) of the included women reported postmenstrual spotting and 30.8% of them reported dysmenorrhea. Median reported pain score during menstruation was 5 (IQR 3.5–8). Apart from parity, previously failed ART and the presence of intrauterine fluid there were no differences in baseline characteristics between women with or without infertility. The various reproductive outcomes are shown in Table  2 . At the 2‐year follow‐up, 83/133 (62.4%) of the women had an ongoing pregnancy, 53/88 (60.2%) in the group of women with infertility and 30/45 (66.7%) in the group of women without infertility. The ongoing pregnancy rate in patients with previous fertility problems was 60.2% compared with 66.7% in patients without infertility (OR 0.68, 95% CI 0.32–1.7). The median time to an ongoing pregnancy was 9 months (IQR 6–12.5) after a laparoscopic niche resection and thus a median time of 3 months after stopping contraceptive methods. Time to ongoing pregnancy was not different between the two groups. Crude ORs for ongoing pregnancy rate, clinical pregnancy rate, live birth rate or miscarriage rate were not different between the two subgroups. Age and parity were the only independent factors related to ongoing pregnancy outcome; previous failed ART or the presence of intrauterine fluid were not related. Logistic regression and multivariate regression analysis for pregnancy outcomes at 2‐year follow‐up. Abbreviations: CI, confidence interval; OR, odds ratio. Eight of the 11 women who had a miscarriage were followed by a subsequent ongoing pregnancy <1 year. One premature delivery at gestational age of 24 weeks after asymptomatic cervical dilation at gestational age of 22 weeks and 1 unexplained intrauterine fetal death at a gestational age of 29 weeks, vaginally delivery after induction of labor. The majority of women 58/133 (43.6%) did not receive any ART during the first year after surgery and tried to conceive naturally. At time of the 1‐year follow‐up, 16 women were still using hormonal contraceptive methods despite the fact that they reported at baseline an actual desire to conceive (Table  S2 ). In the subgroup with infertility at baseline, 16/88 (18.8%) conceived after ART compared five (11.1%) in the group without infertility at baseline (OR 2.51, 95% CI 0.88–7.2). Of the 83 women with an ongoing pregnancy, one women had a premature delivery at 24 weeks and one had a stillbirth of unknown cause (see Table  2 ). Three women delivered prematurely 36 weeks of gestation. One of these women had an uneventful vaginal delivery upon maternal request. All other deliveries were term CSs. There were no signs of uterine rupture or dehiscence of the scar. There was one patient with placenta previa and there were no patients with signs of placenta accreta spectrum. Ultrasound evaluations were performed on 121/133 women (91%) at the 3‐month follow‐up (Table  3 ). The presence of intrauterine fluid accumulation was observed in 66 (49.6%) women at baseline and in seven (5.8%) ( P  = 0.001) women at the 3‐month follow‐up. Median RMT increased in the total group from 1.0 mm (IQR 0.4–1.7) at baseline to 5.7 (IQR 4.2–6.8 mm) ( P  < 0.01) at the 3‐month follow‐up, and the median depth of the niche decreased from 9.9 mm (IQR 7.5–14.2 mm) to 4.2 mm (IQR 2.4–6.3 mm) ( P  < 0.01). There were no statistical differences found in ultrasound outcomes between the two subgroups. In 10 women (8.3%), the RMT was <3 mm at follow‐up, which we deemed to be an anatomic reconstruction failure. In this subgroup with a RMT <3 mm, five of the 10 women had an ongoing pregnancy with a median time to ongoing pregnancy of 9.5 (IQR 7–12) months. Uterus and niche characteristics on transvaginal ultrasound. Note : Data are reported as median (IQR, interquartile range) or n (valid). All 133 women at time of the 1‐year follow‐up reported a substantial decline of gynecologic symptoms (Table  S2 ). Median postmenstrual spotting decreased from 10 (IQR 6–14) to 0.5 (0–4) days, and the median pain score (VAS scale) decreased from 6 (4–8) to 2 (2–6) ( P  < 0.01). A major complication was reported in one patient, requiring a surgical re‐intervention (Table  S3 ).

Overall, 83 of the 133 (62.4%) women with a niche and an RMT of <3 mm with a desire to become pregnant prior to surgery had an ongoing pregnancy after a laparoscopic niche resection that was performed because of gynecologic symptoms or infertility, in combination with intrauterine fluid accumulation or difficulties during their embryo transfer. Besides one premature delivery and one stillbirth of unknown cause, all of the pregnancies and deliveries were uneventful. Postmenstrual spotting decreased and RMT improved substantially, and intrauterine fluid accumulation was no longer present in the majority of the patients after a laparoscopic niche resection. Various studies have reported improvement of gynecologic symptoms after a laparoscopic niche resection. 14 , 15 , 20 However, there is a relative dearth of large comparative studies evaluating reproductive outcomes after this intervention. Most studies are either retrospective or prospective cohort studies characterized by limited sample sizes and unclear selection criteria. Moreover, neither a minimum size of the niche nor a maximum residual myometrium were always a criterion for inclusion. 21 Furthermore, information about the method of data collection or follow‐up period was not always described. 20 Some studies also include either asymptomatic women with infertility or women with a desire to become pregnant as the primary indication of infertility for intervention. 2 , 22 The reported pregnancy rates in these studies varied from 37.5% to 71%. Information regarding fertility treatments prior to and after laparoscopic niche resection is also lacking in most studies. 21 , 22 , 23 , 24 In a previous study including 101 women we showed that a laparoscopic niche resection results in a strong reduction in both postmenstrual spotting and dysmenorrhea at a 6‐month follow‐up. 16 To evaluate reproductive outcomes we extended this study with more patients and a follow‐up of at least 2 years. Our study population differs from those in other studies evaluating reproductive outcomes because we included a large group of women who had previously undergone failed ART. Although one might expect pregnancy rates to be lower in infertile women with previous unsuccessful IVF therapy, we found no statistically significant differences in the fertility outcomes in those with and without preexisting infertility. However, to draw any solid conclusions concerning the additional value of a laparoscopic niche resection in women with or without infertility on reproductive outcomes, randomized controlled trials comparing surgery with or without treatment are needed in these populations. Our results showing positive reproductive outcomes and reduction of postmenstrual spotting with few complications lend support for carrying out such trials. Despite the positive results we do not advocate surgery in asymptomatic women and therefore we also support screening for the presence of a niche in asymptomatic women. However, we do think that it is important to perform an ultrasound to evaluate the presence of a niche in women with gynecologic symptoms or unexplained fertility problems after their CS. There may be a causal relation between a niche and impaired reproductive outcomes, and a laparoscopic niche resection might improve these outcomes. However, this needs to be determined in future comparative studies. To the best of our knowledge, this is the largest prospective cohort study evaluating long‐term (reproductive) outcomes after a laparoscopic niche resection with a structured and standardized follow‐up. There was strict adherence to the standardized protocol. Less than 10% of the data was lost to follow‐up. Women were included consecutively to reduce the risk of selection bias. However, we also recognize the limitations of our study. The observational cohort study design is not optimal for assessing the additional value of an intervention on reproductive outcomes due to the lack of a control group. However, before a randomized trial can be designed, according to the IDEAL framework, a proper cohort study should first be performed to both optimize the intervention and overcome the learning curve of the new intervention. 25 , 26 Given that we did not include women whose sole problem was infertility, caution should be exercised regarding extrapolation of our results to this population. A further limitation is that we did not follow a strict protocol for fertility therapy after the laparoscopic niche resection. After laparoscopic niche resection, the majority of the women were referred for usual care in their local hospital. Information about the exact duration of infertility, number of and difficulties associated with ART attempts both prior to and after laparoscopic niche resection is lacking in the majority of cases and therefore was not taken into account in our analysis. We included both women with and without reported secondary infertility at baseline, which may have compromised the internal validity of the study. On the other hand, this represents daily practice and allowed comparison between the two groups. Previous studies have suggested that the RMT may change over time. One could question whether 3 months is too early for ultrasound evaluation. We chose this moment because we aimed to combine the routinely performed ultrasound scan after laparoscopic niche resection normally planned after 3 months, and because we combined the ultrasound assessment with our routinely performed postoperative visit. We considered it too bothersome for the patients to come again for a 6‐month follow‐up, as we did not expect any clinical consequences. However, we cannot exclude that the presence of intrauterine fluid might have been slightly higher because some women were on oral contraceptives at the 3‐month follow‐up. In our study, 30.8% of the women with a large niche reported dysmenorrhea. However, we cannot exclude that adenomyosis may play a role in this, since we did not exclude women with the diagnosis adenomyosis. From a recent study we know that in approximately 60% of all niche patients one or more adenomyosis signs can be observed. 1 Currently, there is no consensus yet on how many signs of the Morphological Uterus Sonographic Assessment (MUSA) criteria need to be present to diagnose clinical relevant adenomyosis. 27 In addition, we started this study in 2011, at which time the MUSA criteria had not yet been formulated or defined. However, we previously reported a strong reduction in dysmenorrhea NRS score after a laparoscopic niche resection at the 6‐month follow‐up. 1 This underlines an association between dysmenorrhea and the presence of a niche. We advised our patients to use additional contraceptives during the first 6 months to stimulate uterine wound healing before desiring a pregnancy. There is no guideline on the optimal duration to wait before desiring a pregnancy in terms of optimal pregnancy outcomes and therefore this is a matter of debate. We chose 6 months consistent with the advice that we give after a myomectomy. Fortunately, we did not find any uterine rupture during pregnancy. We advised our patients to undergo a planned cesarean section in the absence of any data on safe vaginal delivery after a laparoscopic niche resection. Future studies are needed to evaluate the safety of a vaginal delivery after a successful laparoscopic niche resection that has resulted in a thick residual myometrium. Despite the high number of women with ongoing pregnancies reported after laparoscopic niche resection, not all women may benefit in the same way. Both pregnancy outcomes and the mode of conception depend on baseline characteristics, the reason for the laparoscopic niche resection and the applied fertility therapies. They may also depend on both the type and location of the niche; for example, a very low situated niche may impair optimal suturing after a niche resection and cervical tissue may be less optimal for wound healing. 28 Future studies are warranted to enable the development of prognostic models. Above all, randomized trials are needed to compare the additional value of a laparoscopic niche resection over other therapies. Based on the principles of the IDEAL framework [30], it should currently only be offered in a research setting within expertise centers, both to optimize the procedure and the outcomes and to be able to study the beneficial value of this technique for reproductive outcomes. Ideally, two different randomized controlled trials should be performed: one comparing a laparoscopic niche resection with usual care in women with infertility, and one in women without infertility but with an actual desire to become pregnant. Currently, two of these studies have been registered in trial registers (ref NTR6534.) and one study protocol has been published (Vissers et al. 2023 BMJ online accepted).

A laparoscopic niche resection in women with a niche in their uterine cesarean scar and an RMT of <3 mm with symptoms and/or problems during fertility treatment, resulted in an ongoing pregnancy rate of 62.4% at the 2‐year follow‐up. We could not find significant differences in the reproductive outcomes between women with or without infertility at baseline. But the high pregnancy rates in both groups that we studied, encourages the design of future studies to determine the additional value of this procedure, as it has not yet been compared in a randomized way to expectant management.

A niche is defined as an indentation of at least 2 mm in the myometrium at the site of the uterine cesarean scar, as assessed by a transvaginal ultrasound. 1 Gynecologic problems such as postmenstrual spotting, dysmenorrhea, chronic pelvic pain and infertility may occur in association with a niche. 2 , 3 , 4 , 5 , 6 , 7 According to a meta‐analysis published in 2013 that included 85 728 women, the reported probability of a subsequent pregnancy is lower among women who have had a previous cesarean section (CS) than for women who have previously had a vaginal delivery. 8 A prior CS reduced the probability of subsequent pregnancy by 10% (relative risk [RR] 0.91, 95% confidence interval [CI] 0.87–0.95). 8 In a recent retrospective study of women ( n  = 1317) undergoing in vitro fertilization (IVF), women with a previous CS had a significantly lower ongoing pregnancy rate (odds ratio [OR] 0.64, 95% confidence interval [CI] 0.48–0.87) and live birth rate (OR 0.63, 95% CI 0.45–0.87) compared with those with a previous vaginal delivery. 9 Another recent retrospective cohort study comprising 4879 IVF patients produced similar results, with significantly lower live birth rates after a previous CS compared with a previous vaginal delivery (adjusted OR 0.61, 95% CI 0.47–0.78), particularly in the event of niche presence on ultrasound (adjusted OR 0.42, 95% CI 0.20–0.90). 10 Therefore, we hypothesized that the niche may be an intermediate factor for lower pregnancy rates and live birth rates following a CS. 1 Various theories have been put forward to explain the detrimental effect of a niche on reproductive outcomes. These include impaired spermatozoa penetration and embryo implantation due to intrauterine fluid accumulation or disruptive subendometrial peristalsis in association with a niche. 11 , 12 A laparoscopic niche resection aims to restore uterine anatomy and to reduce intrauterine fluid accumulation, thereby diminishing gynecologic symptoms and possibly also fertility and pregnancy outcomes. 13 , 14 , 15 In extant literature, a large niche is primarily determined by the thickness of the residual myometrium with cut‐off values of 2.2–3 mm. 2 , 3 , 4 While several case series and small cohort studies 13 , 14 , 15 , 16 have reported on reproductive outcomes after niche repair, a recent systematic review concluded that further larger and preferably comparative studies are needed that include structured follow‐up and differentiation between those with or without infertility. 17 The current prospective cohort follow‐up study aimed to describe reproductive outcomes 2 years after laparoscopic niche resections that were performed on women with a desire to become pregnant and reported separately for women with or without infertility at baseline.

JH, WH and CL have received several research grants, all of which lie outside the scope of the submitted work. The authors have stated explicitly that there are no conflicts of interest in connection with this article.

This prospective cohort study was carried out at the Department of Obstetrics and Gynecology in Amsterdam UMC – location VU Medical Center (VUmc) in Amsterdam, the Netherlands (Amsterdam UMC – location VUmc medical ethics committee number: 2011.297). Women with a desire for pregnancy that underwent a laparoscopic niche resection between 2011 and 2019 were consecutively included. All women who were referred to our department because of a symptomatic niche in terms of gynecologic symptoms and/or fertility problems who met our selection criteria were consecutively asked to participate. Inclusion criteria included: women with a niche and an residual myometrium<3 mm measured with sonohysterography, and one or more of the following problems: (1) postmenstrual spotting; (2) midcycle intrauterine fluid accumulation diagnosed during the fertility workup; or (3) difficulties with a previous embryo transfer and preferring a surgical therapy who were referred to our clinic because of the inaccessibility of the uterine cavity for an embryo transfer catheter due to the large or complex niche, in particular in extreme retroflected uteri. Since a positive effect of a laparoscopic niche resection on fertility had not been proven at time of the design of the study we did not offer a laparoscopic niche resection to women with only secondary infertility. We also did not offer a niche resection to women with no future desire to become pregnant. They were offered hormonal therapy or hysterectomy instead. Overall exclusion criteria were: age <18 years, insufficient command of the Dutch language, contraindications for general anesthesia, pregnancy, (suspected) malignancy, uterine or cervical polyps, submucosal fibroids, atypical endometrial cells, cervical dysplasia, cervical or pelvic infection and/or presence of a hydrosalpinx. Women were informed of the experimental nature of the intervention prior to surgery and gave informed consent. A STROBE checklist for cohort studies is provided in Table  S4 . A subgroup of the authors previously published the gynecologic outcomes of the LAPNICHE study at a 6‐month follow‐up. 16 , 18 Details on the technical aspects of the laparoscopic niche resection have been published previously. 19 In short, a laparoscopic niche resection was performed under hysteroscopic guidance (Figure  1 ). After bladder dissection, the niche was opened at the thinnest point of the the residual myometrium. The niche was resected until all scar tissue had been removed and equally thick uterine wound surfaces were achieved. The uterine wound was closed in two layers using single multifilament resorbable intracorporal knotted sutures. Adhesion barrier (Hyalobarrier©, Nordic Pharma) was applied on the surface of the uterine wound. Shortening of the round ligaments was applied in the event that an extensive retroflected position of the uterus persisted after closure of the uterus (Figures  2 and 3 ). A schematic overview of the method of laparoscopic niche resection under hysteroscopic control of the niche. Double‐layer full thickness closuring of the uterus with single sliding knots. (A) (left) First layer of sliding knots. (B) (right) Second layer of knots and suspension of the round ligaments (Baldi anterior). Transvaginal ultrasound of a large niche before (A) and after (B) laparoscopic niche resection. The integrity of the uterine scar, the uterine position, the presence of intrauterine fluid and the niche features were evaluated both at baseline and 3 months after the laparoscopic niche resection, using a transvaginal ultrasound performed by sonographers with extensive expertise in niche evaluation. We used a standardized protocol, in line with the subsequently published consensus of experts. 1 The niche was measured in the sagittal plane where the niche was the largest (maximum depth, thinnest residual myometrium). The niche was also measured in the transverse plane at the largest width of the niche. Niche evaluation was performed at the same time as our routinely performed postoperative visit at a 3‐month follow‐up. We did not collect data on residual myometrial thickness (RMT) during the subsequent pregnancies of these women, as obstetric follow‐up was conducted in their local hospital. All women irrespective of their previous fertility status were advised to use contraception in the first 6 months following niche resection to allow for uterine healing prior to subsequent pregnancy.Post‐surgery, patients were referred to their local hospital for guidance on their fertility treatment, if needed, as well as on their pregnancy and delivery. Both the indication and type of applied fertility treatment were individualized by the local hospital. In the absence of safety data concerning having a vaginal delivery after a laparoscopic niche resection, we advised all patients to undergo an elective repeat CS at term. We followed patients at 6, 12 and 24 months after surgery by validated questionnaire related to their symptoms, fertility and pregnancy outcomes and received fertility therapies. The women also completed a menstrual score chart at the 12‐month follow‐up. Details concerning applied fertility treatment and pregnancy outcomes previously and during the follow‐up period were requested from local hospitals. The primary outcome in the current study was the ongoing pregnancy rate. Ongoing pregnancy rate is defined as an intrauterine pregnancy with a positive heartbeat during ultrasound examination between 10 and 12 weeks of pregnancy. 9 Secondary outcomes were (1) live birth rate; (2) miscarriage rate, (3) ectopic pregnancy rate after the 2‐year follow‐up, (4) surgery‐related complications, (5) ultrasound findings at 3 months and (6) gynecologic symptoms at the 1‐year follow‐up. Registered gynecologic symptoms included postmenstrual spotting and dysmenorrhea. Postmenstrual spotting was defined as either two or more days of intermenstrual spotting or two or more days of brownish discharge immediately following the end of the menstrual period when the total duration of the menstrual bleeding exceeded 7 days . 5 See Table  S1 for elaborated definitions of pregnancy outcomes. The statistical analysis was conducted using IBM SPSS Statistics v.26 (2019). Patients with incomplete data and those lost to care were omitted from the analysis and the remaining data subsequently analyzed (complete case analysis). We did not perform data‐imputation of the missing data. All tests were performed two sided and a P ‐value <0.05 was considered to be statistically significant. Baseline characteristics are presented using percentages, means with SD and 95% CI or medians with interquartile ranges (IQR), as appropriate. Data were tested for normality prior to the use of the t ‐test, otherwise non‐parametric tests were used. Student‘s t ‐tests (continuous variables) and Pearson's χ 2 test or, if necessary, Fisher's exact test (binary and categorical variables) were used to compare baseline characteristics between the two groups. Logistic regression analysis was used to test the relation between the presence of infertility (yes, no) at baseline and the various fertility outcomes. Logistic regression analysis was used to test the relation between the presence of previous existing infertility at baseline and the various fertility outcomes (ongoing pregnancy and live birth rates). Predefined confounding factors were age, body mass index, smoking, previous number of CSs, previous assisted reproductive technologies (ART), residual myometrium and presence of intrauterine fluid. These potential confounders were tested via multivariate analysis and the results were adjusted accordingly. Data on the number of days of postmenstrual spotting and intermenstrual spotting during one menstrual cycle, dysmenorrhea and discomfort due to spotting were not normally distributed and were analyzed using the Wilcoxon signed rank test for related samples to compare the baseline and follow‐up values. This study was approved by the medical ethics committee of Amsterdam UMC – location VU Medical Center (VUmc) in Amsterdam, the Netherlands (VUmc medical ethics committee number: 2011.297) on October 2, 2011. All participants provided written informed consent.

Table S1 Click here for additional data file. Table S2 Click here for additional data file. Table S3 Click here for additional data file. Table S4 Click here for additional data file.