The cost-effectiveness of transvaginal hydrolaparoscopy versus hysterosalpingography in the work-up for subfertility.

of our study are not generalizable to women with high risk of tubal pathology. Furthermore, this economic analysis was based on VC The Author(s) 2022. Published by Oxford University Press on behalf of European Society of Human Reproduction and Embryology. All rights reserved. For permissions, please email: [email protected] Human Reproduction, Vol.37, No.12, pp. 2768–2776, 2022 Advance Access Publication on October 12, 2022 https://doi.org/10.1093/humrep/deac219 ORIGINAL ARTICLE Infertility ......................................... ................................................... ........... . ............. the Dutch healthcare system, and possibly our results are not generalizable to countries with different strategies or costs for fertility treatments. WIDER IMPLICATIONS OF THE FINDINGS: After 2 years of follow-up, we found a live birth rate of 58.4% in the THL group versus 55.4% in the HSG group and a lower mean cost per woman in the THL group, with a cost difference of /C0 e271. The findings of our trial suggest that a strategy starting with THL is cost-effective compared to a strategy starting with HSG in the workup for subfertile women. However, the cost difference between the two diagnostic strategies is limited compared to the total cost per woman in our study and be- fore implementing THL as a first-line strategy for tubal patency testing, more research in other fields, such as patient preference and accep- tance, is necessary. STUDY FUNDING/COMPETING INTEREST(S): The authors received no external financial support for the research. B.W.J.M. is supported by an NHMRC Investigator Grant (GNT1176437). B.W.J.M. reports consultancy for ObsEva, Merck KGaA, Guerbet. B.W.J.M. reports receiving travel support from Merck KGaA. C.T.P. reports consultancy for Guerbet, outside of this manuscript. All other authors have no conflicts to declare. TRIAL REGISTRATION NUMBER: NTR3462. Key words: hysterosalpingography / transvaginal hydrolaparoscopy / fertiloscopy / tubal patency test / fallopian tubes / cost-effectiveness

Tubal pathology is a common cause of subfertility. Depending on the population and the duration of subfertility, tubal factors can be found in 15–30% of subfertile women ( Evers, 2002; Anyalechi et al. , 2021 ); therefore, the evaluation of tubal function is one of the cornerstones of fertility examination. Hysterosalpingography (HSG) has been one of the most com- monly used outpatient first-line tubal patency tests, although its sen- sitivity is low in women with a low risk of tubal pathology ( Broeze et al. ,2 0 1 1). HSG was originally developed as a diagnostic instru- ment; however, recent studies show a possible therapeutic effect of tubal flushing with oil-based contrast ( Dreyer et al. ,2 0 1 7; van Rijswijk et al. ,2 0 2 0). The gold standard in tubal patency testing, however, is laparoscopy with chromopertubation. Compared to HSG, laparoscopy gives more information about other pelvic pathol- ogy, such as the presence of peritubal adhesions or endometriosis and tubo-ovarian contact. Transvaginal hydrolaparoscopy (THL) is a technique developed by Gordts (1998). In this procedure access to the pouch of Douglas is obtained by culdocentesis using the transvaginal route, after which hydroflotation is used for the exploration of the pelvis. The tubo- ovarian structures can be visualized with a small endoscope, in their natural position and without manipulation. This procedure can be per- formed in an outpatient department using local anesthesia ( Gordts, 1998) and is well tolerated by the patient. Considering the fact that laparoscopic techniques for tubal testing give additional information about pelvic pathology, one could suppose that using a laparoscopic technique leads to an earlier detection of pelvic abnormalities and that it may influence the choice of subsequent fertility treatment. However, there is insufficient evidence on which technique has the best diagnostic capacity for the subfertile couple and if there is a differ- ence in cost-effectiveness between the different strategies. Our recent randomized controlled trial (RCT) (THL-trial, NTR3462) showed no significant difference in live birth or time to conception between a strategy starting with THL or a strategy starting with HSG in 300 subfertile women with a low risk of tubal pathology (van Kessel et al., 2021). In this article, we assess the costs and effects of both strategies in a cost-effectiveness analysis, which was performed alongside the THL trial.

This economic evaluation was performed alongside the THL trial, a randomized clinical trial comparing THL and HSG in subfertile women. The study was performed according to Good Clinical Practice guide- lines and the study protocol was approved by the Institutional Review Board of the Amsterdam Medical Center (AMC, Amsterdam, The Netherlands) and by the board of directors of the other participating hospitals. The trial was registered in the Dutch Trial Registry (NTR3462). In each of the participating centers, patient counseling, data collection and monitoring were performed by dedicated research nurses. All women provided written informed consent prior to randomization. Study population Women were eligible if they were undergoing a fertility work-up with an indication for evaluation of tubal patency. Subfertility was defined as the non-occurrence of pregnancy after at least 1 year of unprotected intercourse. Women were not eligible if they had a contraindication for THL (i.e. an immobile uterus or a retroverted uterus, evidence of endometriosis, masses or cysts in the pouch of Douglas or ovarian cysts interfering with THL) if they had positive Chlamydia PCR, prior tubal testing or prior tubal surgery. Women with a known allergy to iodine or methylene blue were also not included. Study design Potential participants were recruited in four Dutch teaching hospitals (Amsterdam, Nieuwegein, Zwolle and Veldhoven) between 2013 and 2016. Eligible women were informed about the study by dedicated re- search nurses. After receiving their written informed consent, women were randomly allocated to a strategy starting with THL (experimental arm) or with HSG (control arm). Randomization was performed by Cost-effectiveness of two tubal patency tests 2769 .......................... ................................................... .............. the doctors or research nurses with the use of a secured online ran- domization program (ALEA, FormsVision) (Tros et al., 2019). Interventions In the THL group, the procedure was scheduled in the follicular phase of the menstrual cycle. THL was performed as described by Gordts (1998). In two of the participating hospitals, the Storz re-usable system was used (KARL STORZ, Tuttlingen, Germany) and in two hospitals, the disposable Fertiloscope (Fertility Focus, Warwick UK). The proce- dure was performed in the out-patient department, using local anes- thesia. An additional diagnostic laparoscopy was performed either if THL was inconclusive, or if in THL the pouch of Douglas was not reached. In the HSG group, the procedure was scheduled in the follicular phase of the cycle. The procedure was performed in the radiology de- partment. The contrast medium, either water-soluble contrast (Telebrix Hystero, Guerbet) or oil-soluble contrast (Lipiodol Ultra- Fluide, Guerbet), was used according to hospital-specific protocols. An additional diagnostic laparoscopy was performed when the HSG showed abnormalities or if HSG failed to show a reliable result. Furthermore, in women with a normal HSG, a diagnostic laparoscopy was scheduled if a pregnancy did not occur after 6–12 months, to rule out pelvic abnormalities that were not noticed by HSG. Additional treatment After completion of the fertility work-up and tubal assessment, the prognostic model of Hunault was used for the prediction of a natural conception within 12 months ( Hunault et al. , 2004 ). When there was no tubal pathology, in general, expectant management was advised when the probability of natural conception within 12 months was >30%. In couples with a probability <30%, IUI with mild ovarian hyperstimulation (IUI-MOH) was advised. IUI-MOH was also advised when there was mild male subfertility (total motile sperm count 3–10 million), or after a period (6 months) of expectant management with- out natural conception. When severe tubal pathology was diagnosed, or in case the couple did not conceive after 3–6 cycles of IUI, couples were counseled for IVF or ICSI. In women diagnosed with hydrosalpinges, endometriosis or severe adhesions, fertility-enhancing laparoscopic surgery was scheduled. Outcomes The primary outcome of the study was conception leading to a live birth within 24 months after randomization. Secondary outcomes were time to conception leading to live birth, miscarriage, ectopic preg- nancy, multiple pregnancy and complications. Economic analysis The economic evaluation was performed as a cost-effectiveness analy- sis from a healthcare perspective, focusing on direct medical costs. We compared the direct medical costs of THL and HSG, the cost of fertility treatments, and the medical costs related to pregnancy and de- livery to assess the costs made to achieve a live birth until 24 months after randomization. Data on resource use were collected from the individual case re- cord forms. For each woman, we registered the type and number of interventions received until pregnancy occurred or until the end of the study period of 24 months. We registered all types of pregnancies and their outcomes. The costs for each parameter were derived from previous publica- tions ( Lukassen et al. ,2 0 0 4; Eijkemans et al. ,2 0 0 5; Bouwmans et al. , 2008; Verhoeve et al. ,2 0 1 3; Mol et al. ,2 0 1 5; Tjon-Kon-Fat et al. , 2015; Lemmers et al. ,2 0 1 8). The costs for HSG, THL and fertility- enhancing surgery were calculated. Since HSG was performed with oil- based contrast in 35% and water-based contrast medium in 65%, we averaged the cost of the contrast medium. The cost parameters and unit costs included in the economic evaluation (expressed in euros) are presented in Table I . All calculations were standardized to 2019 prices using consumer price index data ( Centraal Bureau voor de Statistiek (CBS), the Netherlands, 2022). The unit costs were split into three main categories: costs of the diagnostic procedure, costs of fer- tility treatments, and the costs for pregnancy outcomes. The costs for fertility treatment were estimated from the unit cost for IUI multiplied by the mean number of IUI cycles per woman in each treatment group, and the estimated unit cost for IVF/ICSI was multiplied by the mean number of embryo transfers per woman in each treatment. Statistical analysis Data were analyzed using the intention-to-treat principle. The mean costs and outcomes for each treatment group were compared. Costs were combined with effectiveness by calculating incremental cost- effectiveness ratios (ICERs). The ICER per additional live birth was .......................................................................................................... Table I Unit costs for diagnostic techniques, fertility treatments and pregnancy outcomes. Cost item Unit cost (Euros) indexed to 2019

Diagnostic technique HSG 289 Calculation THL 425 Calculation Diagnostic laparoscopy 1272 Verhoeve et al. (2013) Treatment Ovulation induction 61 Eijkemans et al. (2005) IUI with MOH 323 Tjon-Kon-Fat et al. (2015) IVF/ICSI 2754 Bouwmans et al. (2008) Therapeutic laparoscopy 4084 Calculation Pregnancy and delivery Miscarriage (curettage) 1565 Lemmers et al. (2018) Singleton live birth 3302 Lukassen et al. (2004) Twin live birth 17 442 Lukassen et al. (2004) Ectopic pregnancy (tubectomy) 3420 Mol et al. (2015) MOH, mild ovarian hyperstimulation; HSG, hysterosalpingography; THL, transvaginal hydrolaparoscopy. 2770 van Kessel et al. ......................................... ................................................... ........... . ..................................... ................................ calculated by dividing the difference in total costs by the difference in outcome for the two strategies. We used a non-parametric bootstrap resampling of 1000 re- samples to investigate the effect of uncertainty in our estimates. We generated a cost-effectiveness plane to visualize the uncertainty about the expected difference in costs and outcomes between the two diag- nostic strategies. Furthermore, we generated cost-effectiveness accept- ability curves to indicate the probability that a diagnostic strategy with THL is cost-effective compared to a strategy with HSG, for a range of values of the maximum acceptable cost per live birth. Statistical analyses were performed using SPSS version 26 (IBM statis- tics, Armonk, NY, USA) and Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). A value of P < 0.05 was considered to be significant.

Participants and trial outcomes Between May 2013 and October 2016, we randomized 300 women, of which 149 were assigned to THL and 151 were assigned to HSG. After randomization, 5 women in the THL group and 1 woman in the HSG group withdrew their informed consent, leaving 144 and 150 women for analysis. Baseline characteristics can be found in Table II . At 24 months after randomization, two women in the THL group and two women in the HSG group were lost to follow-up ( Fig. 1). In both groups, the fertility treatment was started according to find- ings of the fertility examinations and based on the prognostic model of Hunault. During the follow-up period, the treatment modality was not fixed, but adjusted if needed. For example, if no conception occurred after 3–6 insemination treatments, the couple was counseled for IVF treatment. Between randomization and the scheduled procedure, 17 and 12 women conceived naturally in the THL group and HSG group, respectively. In the THL group, 28 women were managed ex- pectantly, 9 women underwent ovulation induction, 59 women were treated with IUI and 21 women were treated with IUI followed by IVF treatment. Two women were referred directly for IVF treatment. Based on the findings at THL, a diagnostic laparoscopy was performed in two women and fertility-enhancing surgery was performed in six women. All of the women that underwent fertility-enhancing surgery were subsequently treated with IUI and/or IVF. In the HSG group, 27 women were managed expectantly, 8 women underwent ovulation induction, 60 women were treated with IUI and 26 women were treated with IUI followed by IVF treatment. Three women were re- ferred for IVF treatment. In the HSG group, 22 women underwent subsequent laparoscopy. In 11 women, laparoscopy was scheduled because of an abnormal or inconclusive HSG (4 women underwent diagnostic laparoscopy and 7 women underwent laparoscopic fertility- enhancing surgery) and in another 11 women, laparoscopy was scheduled for women with a normal HSG that did not conceive after 6–12 months (in 6 women only diagnostic laparoscopy was performed and in 5 women laparoscopic fertility-enhancing surgery was per- formed because of the presence of pelvic pathology such as endome- triosis and adhesions with open tubes). After fertility-enhancing surgery nine women were subsequently treated with IUI and/or IVF. After 24 months of follow-up, we found a live birth rate of 58.4% in the THL group versus 55.4% in the HSG group (difference 3.0% (95% CI: /C0 8.3 to 14.4)). Multiple pregnancy and miscarriage rate were low and did not differ in both groups. Ectopic pregnancy did not occur in the THL group and it occurred in two women in the HSG group. No differences in time to conception ( P ¼ 0.199) and time to concep- tion leading to live birth ( P ¼ 0.308) were found. A comparison of the mean costs per woman by cost category for the two treatment groups is provided in Table III. The mean cost was lower in the THL group, compared to the HSG group (THL group e4991 versus e5262 in the HSG group, mean cost difference ¼/C0 e271 (95% CI /C0 e273 to /C0 e269)). Although the costs of the diagnostic procedure itself were higher in the THL group ( e425 versus e289, difference /C0 e135), the total mean cost per woman was higher in the HSG group. In the HSG group, more women underwent diagnostic and therapeutic laparoscopies and also had higher costs for fertility treatments such as IUI-MOH. The spread of the bootstrap samples across the south and north- eastern quadrants indicates that there is uncertainty regarding the ef- fectiveness of THL over HSG and less uncertainty regarding the cost difference between THL and HSG ( Fig. 2 ). The base case outcome is that a strategy starting with THL costs less and is more effective than a strategy starting with HSG, making THL the dominant strategy. The cost-effectiveness acceptability curve shows that if, for example, the maximum acceptable ratio is 5000 EUR per additional live birth, the probability that THL is cost-effective compared with HSG is 0.83 (Fig. 3 ). Alternatively, this can be interpreted as there being a 83% chance that the potential additional cost of having a THL, compared with having an HSG, is <5000 EUR per additional live birth. ....................................................................................................... Table II Baseline characteristics of women in a study comparing the cost-effectiveness of transvaginal hydrola- paroscopy to hysterosalpingography. Characteristic THL (n 5 144) HSG (n 5 150) Female age (years, mean § SD) 31.6 ( §3.9) 31.9 ( §4.0) Median BMI (kg/m2 IQR) 23.4 (21.0–26.9) 23.3 (21.2–26.2) Intoxications Smoking 27 (18.8%) 25 (16.7%) Alcohol 37 (25.7%) 44 (29.3%) Drugs 1 (0.7%) 1 (0.7%) Type of subfertility Primary 102 (71.0%) 124 (82.7%) Secondary 42 (29.0%) 26 (17.3%) Median duration of subfertility in months (IQR) 19 (16–26) 22 (17–30) Ovulatory cycles 108 (75.0%) 129 (86.0%) Median VCM (IQR) 47.5 (17.25–98.5) 51.0 (22.0–118.0) Positive Chlamydia serology 16 (11.1%) 16 (10.7%) VCM, (volume /C2 concentration /C2 percentage progressively moving spermatozoa); IQR, interquartile range; HSG, hysterosalpingography; THL, transvaginal hydrolaparoscopy; Cost-effectiveness of two tubal patency tests 2771 ................................Discussion We performed an economic evaluation alongside an RCT including subfertile women scheduled for tubal patency testing during fertility work-up, and randomized for THL or HSG in women with a low risk of tubal pathology. After 2 years of follow-up, we found a live birth rate of 58.4% in the THL group versus 55.4% in the HSG group (dif- ference 3.0% (95% CI: /C0 8.3 to 14.4)). Total mean costs per woman were also lower in the THL group (mean cost difference ¼/C0 e271 (95% CI /C0 e273 to /C0 e269)). This leads to a strategy with THL being the dominant strategy. Strengths and limitations An important strength is that this study was based on a large multicen- ter RCT comparing a strategy starting with THL with a strategy start- ing with HSG. The follow-up lasted for 24 months after randomization and few women were lost to follow-up. Although few studies have mentioned the costs of THL, a comparison of the costs and effect of HSG and THL has not been made before. During the 2-year follow-up period, the treatment of the couple was managed according to the Dutch guidelines of fertility manage- ment. This means that there was not one fixed treatment modality, CONSORT 2010 Flow Diagram Assessed for eligibility (n=542) Excluded (n=242) ♦ Not meeting inclusion criteria ♦ Declined to participate Analysed for primary outcome (n=142) ♦ Excluded from analysis: women that have withdrawn informed consent (n=5) and the women who were lost follow-up (n=2) Lost to follow-up (n=2) Allocated to THL (n=149) ♦ Received allocated intervention (n=119) ♦ Did not receive allocated intervention (n=30) - Withdrawn informed consent (n=5) - Did HSG instead of THL (n=2) - Declined tubal patency testing (n=6) - Natural pregnancy before THL (n=17) Lost to follow-up (n=2) Allocated to HSG (n=151) ♦ Received allocated intervention (n=134) ♦ Did not receive allocated intervention (n=17) - Withdrawn informed consent (n=1) - Did THL instead of HSG (n=2) - Declined tubal patency testing (n=2) - Natural pregnancy before HSG (n=12) Analysed for primary outcome (n=148) ♦ Excluded from analysis: women that have withdrawn informed consent (n=1) and the women who were lost follow-up (n=2) Allocation Analysis Follow-Up Randomized (n=300) Enrollment Figure 1. CONSORT flow diagram for a comparison of the cost-effectiveness of transvaginal hydrolaparoscopy versus hystero- salpingography in the work-up for subfertility. HSG, hysterosalpingography; THL, transvaginal hydrolaparoscopy. 2772 van Kessel et al. ... ................................................... ........ but that one couple could have different forms of treatment during the follow-up period. This also reflects daily practice in fertility treatment. One of the limitations of this study is that HSG was performed according to local hospital protocol, which means that oil-based con- trast medium was used in two hospitals (35% of participating women) and water-based contrast medium in the other two hospitals (65% of participating women). During the randomization period of our trial, there was no robust evidence yet of the benefits of tubal flushing with oil-based contrast medium. Therefore, it is not possible to evaluate if there is any additional effect of tubal flushing in this RCT. Second, in the THL group, we found more women with anovulatory cycles compared to the HSG group. However, the assignment to ei- ther the strategy with THL or HSG was by means of randomization and the costs of ovulation induction in the two groups were compara- ble, therefore, we do not think that this would lead to a meaningful difference in the costs of fertility treatments between the two groups. Moreover, this trial was conducted specifically in women with a low risk of tubal pathology. Women with a high suspicion of tubal or pelvic pathology, for example because of visible hydrosalpinges, deep- infiltrating endometriosis or a mass in the pouch of Douglas, would benefit by scheduling a laparoscopic procedure to visualize the pelvic cavity and to perform operative laparoscopy directly if needed ....................................................................................................... Table III Comparison of mean costs per woman for the two treatment groups. THL HSG Diagnostic technique HSG/THL and diagnostic laparoscopy e380 e352 Fertility treatments OI, IUI-MOH IVF/ICSI and therapeutic laparoscopy e2106 e2345 Pregnancy outcomes Miscarriage, ectopic pregnancy, singleton and multiple livebirth e2505 e2565 Total cost e4991 e5262 OI, ovulation induction; MOH, mild ovarian hyperstimulation; HSG, hysterosalpingog- raphy; THL, transvaginal hydrolaparoscopy. Figure 2. The cost-effectiveness plane representing the incremental costs and effects of a strategy with THL compared to HSG and the joint uncertainty around the incremental cost-effectiveness. The spread of the bootstrapped samples (n ¼ 1000) across the hori- zontal plane indicates that there is greater uncertainty regarding the magnitude of effect than the difference in costs. HSG, hysterosalpingograph y; THL, transvaginal hydrolaparoscopy. Cost-effectiveness of two tubal patency tests 2773 ......................................... ........................................ (den Hartog et al. ,2 0 0 8; National Institute for Health and Care Excellence, 2013 ; Practice Committee of the American Society for Reproductive Medicine, 2021). The results of our study are, therefore, not generalizable to women with high risk of tubal pathology. Third, this trial was conducted in the Netherlands, using price calcu- lations of the Dutch healthcare system and the currency in Euro’s. The cost difference in our study between a strategy starting with THL and a strategy starting with HSG was e271, with the THL group having higher costs for the diagnostic procedure and the HSG group having higher costs because of treatments and additional laparoscopies. The cost difference between the two diagnostic strategies is limited com- pared to the total cost per woman in our study. Our findings might not be generalizable to other countries, for example if other diagnostic and treatment strategies are used to manage the subfertile couple. Also a large variation in costs of fertility treatments worldwide ( Salam, 2017) has to be taken into account. Furthermore, in the Netherlands, fertility treatments including ART are reimbursed by the healthcare system, if the woman meets specific criteria. In other countries, where fertility treatments are not reim- bursed by the healthcare insurance, the medical costs that the couple needs to pay out of their pocket are important in the decision-making process and this could lead to different choices for fertility treatment. Our cost-effectiveness analysis is performed using a health care per- spective, and we focused on direct medical costs during treatment. During follow-up, we did not monitor indirect costs such as productiv- ity loss due to absence from work, transportation etc. From a societal perspective, these costs have to be taken into account. It could be expected that costs from a societal perspective would be higher for the strategy starting with HSG because in that group women under- went more treatments, more cycles and also more laparoscopies dur- ing the follow-up period. THL was introduced as a new technique for tubal patency testing over 20 years ago. The technique is used in different clinics as a first- line tubal patency test, but HSG still is the most commonly used test worldwide. In our RCT, we compared the cost and effects of a strat- egy of tubal patency testing with THL with a strategy of tubal patency testing with HSG. We found that the two diagnostic procedures have comparable performance in safety, pain and acceptability ( Tros et al. , 2019) and that live birth rates with the two diagnostic strategies were not statistically different. This cost-effectiveness analysis showed that THL is cost-effective compared to HSG in the subfertility work-up. From this point of view, THL could be proposed as a feasible first-line tubal patency test. The preference of the subfertile couple, however, still needs to be taken into account, in both clinical decision-making as well as in guideline development. In order to examine the preferences of subfertile women on different aspects of the procedures of tubal patency testing, we are currently conducting a discrete choice experiment. In conclusion, after 2 years of follow-up, we found a live birth rate of 58.4% in the THL group versus 55.4% in the HSG group (difference 3.0% (95% CI: /C0 8.3 to 14.4)). Although the cost of the diagnostic pro- cedure was higher in the THL group, the total mean costs per woman Figure 3. Cost-effectiveness acceptability curves for live birth. The figure shows the probability that THL is cost-effective compared to HSG over a range of values for the maximum acceptable value for added livebirth. For example, if the maximum acceptable ratio is 5000 EUR per ad- ditional live birth, the probability that THL is cost-effective compared with HSG is 0.83. HSG, hysterosalpingography; THL, transvaginal hydrolaparoscopy. 2774 van Kessel et al. ......................................... ................................................... ........... . ..................................... ................................. were higher in the HSG group. The main drivers for the higher cost in the HSG group were the additional laparoscopies and costs of fertility treatments. The findings of our trial suggest that a strategy starting with THL is cost-effective compared to a strategy starting with HSG in the workup for subfertile women. Data availability The data underlying this article will be shared on reasonable request to the corresponding author.

We thank the women who participated in the trial, the hospitals and their staff (particularly the research nurses and other recruiting staff) for their contributions to this trial. Authors’ roles M.A.v.K. and C.T.P. had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: B.W.J.M., C.A.M.K., M.A.v.K. and R.T. Acquisition of data: C.A.M.K., M.Y.B., G.J.E.O. and W.K.H.K. Statistical analysis and interpretation of the data: M.A.v.K. and C.T.P. Drafting of the manuscript: M.A.v.K. and C.T.P. Critical revision of the manuscript: all authors. Funding The author(s) received no financial support for the research, author- ship and/or publication of this article. Conflict of interest B.W.J.M. is supported by an NHMRC Investigator Grant (GNT1176437). B.W.J.M. reports consultancy for ObsEva, Merck KGaA, Guerbet. B.W.J.M. reports receiving travel support from Merck KGaA. C.T.P. reports consultancy for Guerbet, outside of this manu- script. All other authors have no conflicts to declare.

Anyalechi GE, Wiesenfeld HC, Kirkcaldy RD, Kissin DM, Haggerty CL, Hammond KR, Hook EW 3rd, Bernstein KT, Steinkampf MP, Geisler WM. Tubal factor infertility, in vitro fertilization, and racial disparities: a retrospective cohort in two US clinics. Sex Transm Dis 2021;48:748–753. Bouwmans CA, Lintsen BM, Eijkemans MJ, Habbema JD, Braat DD, Hakkaart L. A detailed cost analysis of in vitro fertilization and intracytoplasmic sperm injection treatment. Fertil Steril 2008;89: 331–341. Broeze KA, Opmeer BC, Van Geloven N, Coppus SF, Collins JA, Den Hartog JE, Van der Linden PJ, Marianowski P, Ng EH, Van der Steeg JW et al. Are patient characteristics associated with the ac- curacy of hysterosalpingography in diagnosing tubal pathology? An individual patient data meta-analysis. Hum Reprod Update 2011;17: 293–300. CBS. Consumer prices. Centraal Bureau voor de Statistiek (CBS), the Netherlands. 2022. https://www.cbs.nl/en-gb/figures/detail/ 83131eng (1 May 2022, date last accessed). den Hartog JE, Lardenoije CM, Severens JL, Land JA, Evers JL, Kessels AG. Screening strategies for tubal factor subfertility. Hum Reprod 2008;23:1840–1848. Dreyer K, van Rijswijk J, Mijatovic V, Goddijn M, Verhoeve HR, van Rooij IAJ, Hoek A, Bourdrez P, Nap AW, Rijnsaardt-Lukassen HGM et al. Oil-based or water-based contrast for hysterosalpin- gography in infertile women. N Engl J Med 2017;376:2043–2052. Eijkemans MJ, Polinder S, Mulders AG, Laven JS, Habbema JD, Fauser BC. Individualized cost-effective conventional ovulation in- duction treatment in normogonadotrophic anovulatory infertility (WHO group 2). Hum Reprod 2005;20:2830–2837. Evers JL. Female subfertility. Lancet 2002;360:151–159. Gordts S, Campo R, Rombauts L, Brosens I. Transvaginal hydrola- paroscopy as an outpatient procedure for infertility investigation. Hum Reprod 1998;13:99–103. Hunault CC, Habbema JD, Eijkemans MJ, Collins JA, Evers JL, te Velde ER. Two new prediction rules for spontaneous pregnancy leading to live birth among subfertile couples, based on the synthe- sis of three previous models. Hum Reprod 2004;19:2019–2026. Lemmers M, Verschoor MAC, Bossuyt PM, Huirne JAF, Spinder T, Nieboer TE, Bongers MY, Janssen IAH, Van Hooff MHA, Mol BWJ et al. ; MisoREST study group. Cost-effectiveness of curettage vs. expectant management in women with an incomplete evacuation after misoprostol treatment for first-trimester miscarriage: a ran- domized controlled trial and cohort study. Acta Obstet Gynecol Scand 2018;97:294–300. Lukassen HG, Schonbeck Y, Adang EM, Braat DD, Zielhuis GA, Kremer JA. Cost analysis of singleton versus twin pregnancies after in vitro fertilization. Fertil Steril 2004;81:1240–1246. Mol F, van Mello NM, Strandell A, Jurkovic D, Ross JA, Yalcinkaya TM, Barnhart KT, Verhoeve HR, Graziosi GC, Koks CA et al. ; European Surgery in Ectopic Pregnancy (ESEP) study group. Cost- effectiveness of salpingotomy and salpingectomy in women with tubal pregnancy (a randomized controlled trial). Hum Reprod 2015; 30:2038–2047. National Institute for Health and Care Excellence. Fertility Problems: Assessment and Treatment. [NICE Guideline No. 156]. 2013. https://www.nice.org.uk/guidance/cg156 (1 May 2022, date last accessed). Practice Committee of the American Society for Reproductive Medicine. Fertility evaluation of infertile women: a committee opin- ion. Fertil Steril 2021;116 :1255–1265. Salam M. Success of in vitro fertilization: a researched science or a performance indicator. J Clin Obstet Gynecol 2017;6:3–4. Tjon-Kon-Fat RI, Bensdorp AJ, Bossuyt PM, Koks C, Oosterhuis GJ, Hoek A, Hompes P, Broekmans FJ, Verhoeve HR, de Bruin JP et al. Is IVF-served two different ways-more costeffective than IUI with controlled ovarian hyperstimulation? Hum Reprod 2015;30: 2331–2339. Tros R, van Kessel MA, van Kuijk SMJ, Oosterhuis GJE, Kuchenbecker WKH, Kwee J, Bongers MY, Mol BWJ, Koks CAM. The capacity of transvaginal hydrolaparoscopy versus Cost-effectiveness of two tubal patency tests 2775 .............. ....... hysterosalpingography to diagnose tubal pathology in the work-up of subfertile women, a randomised clinical trial. Eur J Obstet Gynecol Reprod Biol 2019;236:127–132. van Kessel M, Tros R, van Kuijk S, Oosterhuis J, Kuchenbecker W, Bongers M, Mol BW, Koks C. Transvaginal hydrolaparoscopy versus hysterosalpingography in the work-up for subfertility: a ran- domized controlled trial. Reprod Biomed Online 2021;43:239–245. van Rijswijk J, van Welie N, Dreyer K, Pham CT, Verhoeve HR, Hoek A, de Bruin JP, Nap AW, van Hooff MHA, Goddijn M et al. Tubal flushing with oil-based or water-based contrast at hystero- salpingography for infertility: long-term reproductive outcomes of a randomized trial. Fertil Steril 2020;114:155–162. Verhoeve HR, Moolenaar LM, Hompes P, van der Veen F, Mol BW. Cost-effectiveness of tubal patency tests. BJOG 2013;120:583–593. 2776 van Kessel et al.