The cost-effectiveness of surgical excision of colorectal endometriosis compared to ART treatment trajectory (TOSCA study) in the management of colorectal endometriosis and subfertility

Primary outcome

The primary outcome is the cumulative ongoing pregnancy rate resulting in a live birth. In line with the COMMIT initiative, live birth is defined as the complete expulsion or extraction from a woman of a product of fertilisation after 20 weeks of gestational age; which, after such separation, breathes or shows any other evidence of life, such as heartbeat, umbilical cord pulsation or definite movement of voluntary muscles, irrespective of whether the umbilical cord has been cut or the placenta is attached. A birth weight of 350 g or more can be used if gestational age is unknown (Duffy et al. 2020).

Secondary outcomes

The secondary outcomes are listed in Table 1. A clinical pregnancy is defined as a viable intrauterine pregnancy confirmed by ultrasound with at least one fetus with a discernible heartbeat during ultrasonographic examination (Duffy et al. 2020). For the numeric rating scale (NRS) scores, the patient is asked to rate their pain on a scale from 0 (no pain) to 10 (worst pain imaginable). PROMs that will be administered include the EHP-30, EQ-5D-5L, LARS and the iPCQ questionnaire. The NRS scores, EHP-30 and the LARS questionnaire will be administered at baseline and 12, 24 and 36 months after inclusion. The EQ-5D-5L and the iPCQ questionnaires will be more frequently administered (EQ-5D-5L: at baseline and EQ-5D-5L and iPCQ at 6, 12, 18, 24, 30, 36 and 40 months after inclusion on T = 0). If a patient undergoes surgery and the NRS scores, EHP-30 and LARS questionnaire were completed more than 3 months prior to the planned surgery, these questionnaires will be re-administered 1–3 weeks before surgery.

Baseline characteristics

The baseline characteristics that will be collected during the study will include demographic data and medical and obstetric history, including the following variables: age, BMI, smoking, alcohol use, comorbidities affecting fertility (e.g. PCOS), hormonal usage, prior abdominal surgery (laparoscopic or laparotomy), prior endometriosis surgery (laparoscopic vs laparotomy and diagnostic vs therapeutic), prior ovarian surgery, duration active wish to conceive (coitus without protection, no birth control) at T = 0, primary or secondary subfertility, fallopian tube status (if available), prior fertility treatments (e.g. intrauterine insemination (IUI)), male factor subfertility, anti-Müllerian hormone (AMH) and antral follicle count (AFC) (if available and regarded representative: within prior 2 years not followed by ovarian surgery). Pre-treatment characteristics will include pain scores (chronic pelvic pain, dyschezia, dysmenorrhoea, dysuria and dyspareunia) and endometriosis classification (MRI and/or ultrasound) according to the #Enzian classification, including the presence of adenomyosis according to the MUSA criteria. The pre-surgical #Enzian classification according to MRI (or ultrasound if MRI not available) will be used to compare endometriosis classification at baseline between the surgery and IVF/ICSI-only group (Keckstein et al. 2021). Previous literature has shown that MRI-based #Enzian scores correlate well with intraoperative findings (Burla et al. 2019). Expert MRI/ultrasound is available in each participating centre.

Inclusion criteria

• Women aged between ≥21 years and ≤40 years at T = 0.

• Women in a heterosexual relationship or in a same-sex relationship.

• Patients with colorectal endometriosis defined as endometriosis involving the (colo)rectum (#Enzian classification score C1, C2, C3, FI (sigmoid)) diagnosed with ultrasound or MRI (Keckstein et al. 2021).

• Patients who have an active wish to conceive and fall under one of the following criteria:

  • – at least 1 year of non-conception (either spontaneous or after intrauterine (IUI));

  • – inability to have timed intercourse or to perform IUI because of pain (dyspareunia, dysuria, dyschezia, dyspareunia and/or chronic pelvic pain);

  • – severe complaints (expectant management is not acceptable (anymore)).

• The patient has an indication for IVF/ICSI according to Dutch guidelines (Werkgroep netwerkrichtlijn, December 2010):

  • – failed IUI;

  • – male factor subfertility (oligoasthenoterato­zoospermia defined as VCM (volume of ejaculate × concentration of spermatozoa × motility of spermatozoa in percentage) <1 million);

  • – bilateral tubal pathology (e.g. bilateral hydrosalpinx, bilateral tubal occlusion);

  • – age >38 years and (unexplained) subfertility; and

  • – severe endometriosis in case of subfertility.

• Patients who are informed by their endometriosis specialist on the choice between IVF/ICSI vs laparoscopic colorectal endometriosis excision surgery at T = 0, T = 1 or T = 2 (Fig. 1) in their treatment trajectory (indicated by blue boxes).

  

  

  View Full Size

  

  Treatment flow for patients with colorectal endometriosis and an unfulfilled wish to conceive. *The duration of the time to allow natural conception is determined based on the Endometriosis Fertility Index (EFI) (maximum 12 months). The EFI score will be determined based on surgical and historical factors. ICSI = intracytoplasmic sperm injection; IVF = in vitro fertilization.

  Citation: Reproduction and Fertility 5, 2; 10.1530/RAF-23-0048

  • Download Figure
  • Download figure as PowerPoint slide

Exclusion criteria

• Patients with deep endometriosis without colorectal involvement.

• Patients who conceive spontaneously prior to intervention.

• Patients requiring surgery on short notice and therefore unable to opt for IVF/ICSI (e.g. in case of unilateral or bilateral hydronephrosis, severe bowel stenosis and suspicion of an impending ileus).

• Patients with a contraindication for IVF/ICSI (e.g. diminished ovarian reserve (premature ovarian failure) (AMH (when available) <p10 adjusted for age), untreated congenital uterine abnormalities, maltreated/untreated systemic or malignant disease or severe risk factors for oocyte aspiration).

• Patients diagnosed with other diseases causing infertility (e.g. recurrent miscarriages, antiphospholipid syndrome).

• Not able to read and understand Dutch or English.

Patients experiencing severe pain symptoms, which may impair intercourse, will be included as this is a common encountered presentation in clinical practice. While these patients may not meet the official criteria of subfertility (defined as 12 months of timed intercourse without conception), the decision to lower pain symptoms either by surgery or with downregulation, in order to make it possible to pursue spontaneous conception or IVF/ICSI, is also of interest in this cohort. Patients aged over 40 will be excluded to ensure that women still have a reasonable chance to conceive spontaneously after surgery. Additionally, study participants must be able to undergo all three IVF/ICSI procedures (if necessary), and in the Netherlands, reimbursement of IVF/ICSI treatments is granted till the age of 43 (start of hormonal stimulation). Patients aged under 21 years will also be excluded because we believe that performing IVF/ICSI and/or surgery to enhance fertility is not common practice in younger women.

Figure 2 provides a global overview of the study timeline from a patient’s perspective. Patients who meet the inclusion and exclusion criteria will be verbally informed about the TOSCA study by their physician. All patients will receive an information leaflet about the potential treatment options for subfertility related to (colorectal) endometriosis and written information about the TOSCA study. The patient information and questionnaires will be available in Dutch or English. Recruitment will be on a voluntary basis, and withdrawal of consent is possible at any time without consequences for the patient’s treatment. If the patient is willing to participate after a consideration period of 1 week, informed consent will be digitally obtained after which the patients receive the baseline questionnaires. The total follow-up time per patient will include 40 months unless the study endpoint (live birth) is achieved earlier. The endpoint criteria of the study are: (1) live birth or (2) no live birth after 40 months of follow-up despite IVF/ICSI (maximum three cycles), colorectal resection surgery or a combination of both treatments. The choice between surgery or IVF/ICSI treatment will be determined through shared decision-making while taking the patient’s current quality of life into consideration.

IVF/ICSI will be performed according to the local protocol of each participating centre to enable implementation of the study in daily practice. The IVF/ICSI downregulation treatments and stimulation protocols will be documented. We do not expect the choice of ovarian stimulation protocol to significantly influence reproductive outcomes as studies did not show a significant difference in pregnancy or live birth rates between GnRH antagonist and agonist protocols (Becker et al. 2022). Patients may undergo a maximum of three IVF/ICSI cycles during the study period. An IVF/ICSI cycle is defined as the transfer of all embryos created after an ovum pickup.

Laparoscopic (colorectal) endometriosis resection and resection of other lesions (e.g. ovarian cystectomy, uterosacral ligament resection and ureterolysis) will be performed during the same procedure when indicated, as described by the working group of the European Society for Gynaecological Endoscopy, ESHRE and the World Endometriosis Society (Working group of ESGE et al. 2020). The objective of the surgery will be to excise all endometriosis lesions in the pelvic area, including colorectal lesion(s). If, for any reason, this is not deemed feasible (e.g. as determined by the surgeon or if the patient objects), this will be documented in the CRF. Colorectal surgery will be performed together with a specialised and experienced abdominal surgeon. Serosal shaving or superficial resection of endometriosis lesions from the bowel is performed when the endometriosis is only present within or on the serosa, without infiltrating the muscularis layer (Working group of ESGE et al. 2020). If the endometriosis has infiltrated the muscularis, a more radical approach such as full-thickness resection (discoid resection) or segmental bowel resection is necessary, depending on lesion(s) size, multifocality, and the degree of infiltration (Working group of ESGE et al. 2020). Colorectal surgery will not be subjected to further standardisation but will be conducted according to the judgment of the colorectal surgeon, enabling the study’s implementation in daily practice. During the analysis phase, sub-analysis will be performed to assess whether the surgical techniques (shave, discoid resection and segmental resection) varied across treatment centres based on the severity of colorectal endometriosis (#Enzian C1, C2, C3). The ‘Endometriosis Fertility Index’ (EFI) will be used to predict the spontaneous pregnancy chance following surgery based on patient characteristics and surgical findings (Adamson & Pasta 2010). Women with an EFI score of 6–10 will be given the opportunity to conceive naturally for a maximum of 12 months (Vesali et al. 2020). Both natural conception and ART are discussed with women having an EFI score of 4–5, and the treatment choice will be based on shared decision-making and the patients’ preferences (Vesali et al. 2020). In women with an EFI score of 0–3, IVF/ICSI will be strongly advised (Vesali et al. 2020). Castor EDC® will be used to collect questionnaires prospectively. Missing values are minimised through active monitoring. No additional study visits for the participants are necessary. Patients with colorectal endometriosis in situ during fertilisation (resulting in live birth) will be enrolled in the IVF/ICSI-only group. Those without colorectal endometriosis in situ during this period (who have undergone resection surgery) will be assigned to the surgery group. This means that patients in whom oocytes were retrieved prior to colorectal surgery and who underwent embryo transfer after surgery (resulting in live birth) will be categorised in the surgery group.

Study withdrawal criteria are: (1) patient-initiated withdrawal from the study, (2) if, during surgery, it becomes evident that the endometriosis lesion is not located in the colorectal area (but in the rectovaginal area), and (3) discontinuation of the patient’s intent to conceive.

View Full Size

Flowchart of TOSCA study. *The duration of the time to allow natural conception is determined by the Endometriosis Fertility Index (EFI) (maximal 12 months). The EFI score will be determined based on surgical and historical factors; **Maximal three attempts.

Citation: Reproduction and Fertility 5, 2; 10.1530/RAF-23-0048

• Download Figure
• Download figure as PowerPoint slide

Sample size

The sample size calculation is based on the expected live birth rate for IVF/ICSI-only treatment, estimated at 51.2% (after three cycles). This rate was calculated using a combination of previous published data on live births following IVF/ICSI-only treatment in 231 subfertile women, resulting in a total of 118 live births (Bendifallah 2017, Ballester 2012, Maignien 2021). The benefit of surgical treatment would have to be substantial to outweigh the risks of the procedure and was set at a 20% point increase. In the analysis phase, a minimum absolute difference of 10% in LBR in favour of the surgery group vs the IVF/ICSI-only group will be required to conclude that adding surgery to the treatment trajectory is superior to the IVF/ICSI-only treatment. Equal group sizes of 152 in both groups would be sufficient to achieve 90% power for the detection of a difference of 20% in live births with an α of 0.05. The proportion in group one (the surgery group) is assumed to be 0.51 under the null hypothesis and 0.71 under the alternative hypothesis. The proportion in group two (the IVF/ICSI group) is 0.51. With unequal group sizes and a group size ratio of 3, the sample size for the surgery group is 76, and the IVF/ICSI-only group is 228. We expect the ratio to be within these limits. In the event that the distribution of participants among the groups is more evenly balanced than initially expected, it may result in a reduced need to include a larger number of patients. Therefore, an interim analysis will be performed after the inclusion of 150 patients. Taking into account a 10% dropout during follow-up, we will include 339 patients (85 in the surgery group, 254 in the IVF/ICSI-only group) at either T = 0, T = 1 or T = 2 (Fig. 1), provided that a treatment decision between ART and surgery is possible (indicated by blue boxes). We do not anticipate a high dropout rate since the study does not affect treatment decisions.

Statistical analysis

Statistical analysis will be performed using SPSS. A Shapiro–Wilk test will evaluate the distribution of the data. Dichotomous or categorical data will be presented in percentages, and continuous data as means and standard deviations (normally distributed) or medians and interquartile ranges (non-parametric). The primary outcome (cumulative LBR in the surgery vs IVF/ICSI-only group) will be assessed using a two-sided Z-test with unpooled variance. Surgical treatment (potentially combined with IVF/ICSI) will be considered superior to the IVF/ICSI-only treatment in case the cumulative LBR in the surgical group vs the ART-only group exceeds with more than an absolute difference of 10%. In addition, a survival analysis will be performed (Kaplan–Meier curves) in both groups, with live birth considered as an ‘event’. Subgroup analysis will be performed in the group of patients who combined surgery and IVF/ICSI to assess the effect of the timing of surgery on the LBR. The difference in secondary outcomes between both groups will be analysed using the Student’s t-test for continuous variables, the Mann–Whitney U test for non-parametric variables, and Fisher's exact or chi-square tests for categorical data. Simple and multiple regression analysis will be used to assess the correlations between live birth and patient as well as treatment characteristics (as outlined in Table 1). This approach will also allow for correction of potential confounding factors. A two-tailed P-value of <0.05 will be considered statistically significant.

Cost-effectiveness

The economic evaluation will be performed as a trial-based cost-effectiveness analysis (CEA) and cost-utility analysis (CUA) from a societal perspective, according to the Dutch guidelines (Nederland 2016). For the CEA, the primary outcome parameter will be the cost per live birth rate. With this outcome parameter, the incremental cost-effectiveness ratio will provide data to guide the preference. For the CUA, the primary outcome will be the cost per quality-adjusted life-years based on the EQ-5D-5L scores, to provide data on the cost per QALY gained. The time horizon of the evaluation will be the period between inclusion and study endpoint (i.e. live birth or 40 months after inclusion). Discounting will be applied according to the Dutch guideline (4% for costs and 1.5% for benefits) in order to weigh future gains and losses less heavily than those that occur in the present. Multiple imputations will be used to deal with missing data. Differences in mean costs and effects (QALYs) between strategies will be compared with bootstrapping using 1000 replications. In a net-benefit analysis, costs will be related to QALYs and presented in a cost-effectiveness acceptability curve. In a cost-effectiveness acceptability curve, the probability of cost-effectiveness for surgical treatment compared to IVF will be shown for different values of the willingness to pay, including the Dutch threshold values ranging between 20,000 and 80,000 Euro per QALY (Zwaap et al. 2015). Sensitivity analyses will be performed to evaluate the robustness of our results. These will include analyses from a healthcare sector perspective and using EQ-VAS utilities instead of EQ-5D-5L utilities. Analyses will be performed using Stata and Excel. Both healthcare costs (secondary and tertiary care) and the cost of loss of productivity for paid and unpaid work will be included. Healthcare resource use (i.e. gynaecology visits, fertility specialist visits, visits an emergency room, hospitalisation, ambulance care, medication use) will be obtained from patients’ medical files. Healthcare utilisation will be evaluated using Dutch costing manual (Hakkaart-Van Roijen et al. 2016). Medication use will be valued with prices derived from www.medicijnkosten.nl. The iPCQ questionnaire will be used to monitor absenteeism and presenteeism from paid and unpaid work.

Budget impact analysis

A budget impact analysis (BIA) to estimate the financial impact of adoption and diffusion of surgical excision at the national level will be performed. The BIA will be conducted from the perspective of the different healthcare professionals (health insurers and healthcare providers) and from the societal perspective (i.e. by including productivity costs). Costs will be estimated per budget period (1 year) for a time horizon of 40 months.

Data management

All data will be collected in Castor EDC®, a web-based clinical data management system. Data processing will be done according to the EU General Data Protection Regulation and the Dutch Act on the implementation of the General Data Protection Regulation (in Dutch: Uitvoeringswet AVG). Analysis of the data will be performed with the coded data. The key that links the code to personal patient information will only be available to the local study team in each centre. Data will be stored for 15 years.

Focus group

An online focus group was conducted to assess the feasibility of patient participation and comprehensibility of patient information among six women diagnosed with colorectal endometriosis, who underwent IVF and/or laparoscopic colorectal endometriosis resection in the past 5 years due to endometriosis-related subfertility. Participants were recruited by their gynaecologists at participating centres (n = 2) and through advertisements on the social media account of the Dutch Endometriosis Patient Federation (n = 4). Prior to the focus group, participants received all patient information and an overview of the PROMs that will periodically be sent to patients participating in the TOSCA study. The focus group was directed by a moderator (RdK) and an observer (MB) monitoring the process, while the chair of the Dutch Patient Federation (BdB) was present to provide support. A semi-structured interview was carried out to allow participants to freely share their experiences. Participants provided informed consent prior to the session, and anonymity and confidentiality were ensured. The session was audio-ecorded and fully transcribed for analysis. RdK and MB analysed the transcript of the focus group individually to identify the most important themes. The results of the focus group can be found in the Supplementary Data section.

Ethical approval

This study is registered as the TOSCA study on clinicaltrials.gov (NCT05677269). The LDD Medical Research Ethics Committee has approved this study (Ref. No. N22.085).