Advertisement

The effect of myomectomy on reproductive outcomes in patients with uterine fibroids: A retrospective cohort study

Open AccessPublished:June 08, 2022DOI:https://doi.org/10.1016/j.rbmo.2022.05.025

      Abstract

      Research question

      What is the effect of myomectomy in women with uterine fibroids on time to live birth and other reproductive outcomes?

      Design

      This was a monocentric retrospective cohort study of 311 women with fibroids, of whom 165 eventually received myomectomy and 146 remained on expectant management. To assess the primary outcome time to live birth, a Cox proportional hazards model with a time-varying covariate for myomectomy was used. In the sensitivity analyses, this was combined with an approach to account for confounders via a cloning/censoring/weighting that aimed to emulate a randomized controlled trial.

      Results

      Notable differences in baseline characteristics between the myomectomy and expectant management group were fibroid size (fibroid >7 cm: myomectomy 48%; expectant management 15%) and whether the patient had fibroid-related complaints (myomectomy 85%, expectant management 67%). The adjusted hazard ratio for the effect of myomectomy compared with expectant management on live birth was 1.26 (95% CI 0.87–1.81). Sensitivity analyses yielded similar results, but secondary outcomes showed that women remaining on expectant management had more often received assisted reproductive technology (63%) compared with those who eventually received myomectomy (38%).

      Conclusions

      The study did not find a significant difference in time to live birth after myomectomy compared with expectant management in women with fibroids and a wish to conceive, despite more, larger and more symptomatic fibroids in the myomectomy group. The results after myomectomy encourage the execution of a randomized controlled trial in women with large (symptomatic) fibroids not or minimally distorting the intrauterine cavity, and infertility or a desire to conceive.

      Key words

      Introduction

      Fibroids are the most common benign tumours of the uterus. They are present in up to 40% of women of reproductive age, and by the age of 50 years the estimate rises to an incidence up to 80%. Worldwide, fibroids are the most common indication for a hysterectomy and therefore a major public health burden (
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ;
      • Parker W.H.
      Etiology, symptomatology, and diagnosis of uterine myomas.
      ).
      Although most women with fibroids remain fertile, studies have shown that fibroids are associated with infertility. This is partly based on the observation of fibroids in women with unexplained infertility. Of all women evaluated with infertility, 30–40% are diagnosed with unexplained infertility. Among these patients, estimates of fibroid incidence may vary from 2% to 12.6% (
      • Bosteels J.
      • van Wessel S.
      • Weyers S.
      • Broekmans F.J.
      • D'Hooghe T.M.
      • Bongers M.Y.
      • Mol B.W.J.
      Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities.
      ;
      • Brady P.C.
      • Stanic A.K.
      • Styer A.K.
      Uterine fibroids and subfertility: an update on the role of myomectomy.
      ;
      • Carranza-Mamane B.
      • Havelock J.
      • Hemmings R.
      The management of uterine fibroids in women with otherwise unexplained infertility.
      ;
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ). Additionally, a recent large cohort study in patients with uterine fibroids showed an increased time to first and subsequent pregnancies compared with women without fibroids (
      • Karlsen K.
      • Mogensen O.
      • Humaidana P.
      • Kesmodel U.S.
      • Ravn P.
      Uterine fibroids increase time to pregnancy: a cohort study.
      ).
      Different fibroid characteristics may affect fertility in ways that are not completely understood. It is postulated that the main factor is that distortion of the uterine cavity by fibroids may alter the receptivity of the intrauterine environment. This might be additionally influenced by the changes in functionality of the myometrium. Any of these alterations can consequently disturb the implantation of the embryo and may result in adverse reproductive outcomes (
      • Behbehani S.
      • Polesello S.
      • Hasson J.
      • Silver J.
      • Son W.Y.
      • Dahan M.
      The Effect of Intramural Myomas Without an Intracavity Component on In Vitro Fertilization Outcomes in Single Fresh Blastocyst Transfer Cycles.
      ;
      • Brady P.C.
      • Stanic A.K.
      • Styer A.K.
      Uterine fibroids and subfertility: an update on the role of myomectomy.
      ;
      • Christopoulos G.
      • Vlismas A.
      • Salim R.
      • Islam R.
      • Trew G.
      • Lavery S.
      Fibroids that do not distort the uterine cavity and IVF success rates: an observational study using extensive matching criteria.
      ).
      Fibroids are heterogeneous in location, size and number. The location of fibroids in the uterus can be classified via the system of the International Federation of Gynaecology and Obstetrics (FIGO;
      • Munro M.G.
      • Critchley H.O.
      • Broder M.S.
      • Fraser I.S.
      FIGO classification system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age.
      ). The location of the fibroids appears to be responsible for the size of the influence on reproductive outcomes. Previous research has found that women with submucosal fibroids, compared with those with no fibroids, have higher miscarriage rates and lower rates of implantation, ongoing pregnancy and live birth. On the other hand, subserosal fibroids have no significant impact on reproductive outcomes, while the effect of intramural fibroids remains debatable because of conflicting study results (
      • Brady P.C.
      • Stanic A.K.
      • Styer A.K.
      Uterine fibroids and subfertility: an update on the role of myomectomy.
      ;
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ;
      • Olive D.L.
      • Pritts E.A.
      Fibroids and reproduction.
      ;
      • Pritts E.A.
      • Parker W.H.
      • Olive D.L.
      Fibroids and infertility: an updated systematic review of the evidence.
      ;
      • Rikhraj K.
      • Tan J.
      • Taskin O.
      • Albert A.Y.
      • Yong P.
      • Bedaiwy M.A.
      The Impact of Noncavity-Distorting Intramural Fibroids on Live Birth Rate in In Vitro Fertilization Cycles: A Systematic Review and Meta-Analysis.
      ;
      • Somigliana E.
      • Vercellini P.
      • Daguati R.
      • Pasin R.
      • De Giorgi O.
      • Crosignani P.G.
      Fibroids and female reproduction: a critical analysis of the evidence.
      ;
      • Styer A.K.
      • Jin S.
      • Liu D.
      • Wang B.
      • Polotsky A.J.
      • Christianson M.S.
      • Vitek W.
      • Engmann L.
      • Hansen K.
      • Wild R.
      • Legro R.S.
      • Coutifaris C.
      • Alvero R.
      • Robinson R.D.
      • Casson P.
      • Christman G.M.
      • Christy A.
      • Diamond M.P.
      • Eisenberg E.
      • Zhang H.
      • Santoro N.
      Association of uterine fibroids and pregnancy outcomes after ovarian stimulation-intrauterine insemination for unexplained infertility.
      ;
      • Sunkara S.K.
      • Khairy M.
      • El-Toukhy T.
      • Khalaf Y.
      • Coomarasamy A.
      The effect of intramural fibroids without uterine cavity involvement on the outcome of IVF treatment: a systematic review and meta-analysis.
      ;
      • Tian Y.C.
      • Wu J.H.
      • Wang H.M.
      • Dai Y.M.
      Improved Fertility Following Enucleation of Intramural Myomas in Infertile Women.
      ).
      Oliveira and colleagues reported a poorer implantation and pregnancy rate in women with intramural fibroids larger than 4 cm compared with those with intramural fibroids smaller than 4 cm, but no difference in live birth rate was found. However, two other cohort studies did not find a relationship between fibroid size and pregnancy rate (
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ;
      • Oliveira F.G.
      • Abdelmassih V.G.
      • Diamond M.P.
      • Dozortsev D.
      • Melo N.R.
      • Abdelmassih R.
      Impact of subserosal and intramural uterine fibroids that do not distort the endometrial cavity on the outcome of in vitro fertilization-intracytoplasmic sperm injection.
      ). At this point, compelling evidence of fibroid size negatively affecting fertility is lacking (
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ;
      • Oliveira F.G.
      • Abdelmassih V.G.
      • Diamond M.P.
      • Dozortsev D.
      • Melo N.R.
      • Abdelmassih R.
      Impact of subserosal and intramural uterine fibroids that do not distort the endometrial cavity on the outcome of in vitro fertilization-intracytoplasmic sperm injection.
      ). In addition the influence of fibroid number on fertility has not been adequately researched, although one systematic review reported that a lower number of fibroids was associated with improved reproductive outcomes after their removal (
      • Orlando M.
      • Kollikonda S.
      • Hackett L.
      • Kho R.
      Non-hysteroscopic myomectomy and fertility outcomes: a systematic review.
      ).
      Obviously, a hysterectomy as an ultimate solution for a diseased uterus is not an option for women who wish to preserve their fertility potential, but nevertheless surgical treatment is possible for these patients by performing a myomectomy. No consensual guideline is available concerning the best surgical techniques. However, compared with laparoscopic or abdominal myomectomy, hysteroscopic myomectomy avoids invasive surgery and is the gold standard for the surgical treatment of submucosal fibroids (
      • Thubert T.
      • Foulot H.
      • Vinchant M.
      • Santulli P.
      • Marzouk P.
      • Borghese B.
      • Chapron C.
      Surgical treatment: Myomectomy and hysterectomy; Endoscopy: A major advancement.
      ). Previous research has shown that performing myomectomy while sparing the neurovascular bundle of the fibroid pseudocapsule improves fertility outcomes (
      • Tinelli A.
      • Malvasi A.
      • Hurst B.S.
      • Tsin D.A.
      • Davila F.
      • Dominguez G.
      • Dell'edera D.
      • Cavallotti C.
      • Negro R.
      • Gustapane S.
      • Teigland C.M.
      • Mettler L.
      Surgical management of neurovascular bundle in uterine fibroid pseudocapsule.
      ;
      • Tinelli A.
      • Mynbaev O.A.
      • Sparic R.
      • Vergara D.
      • Di Tommaso S.
      • Salzet M.
      • Maffia M.
      • Malvasi A.
      Angiogenesis and Vascularization of Uterine Leiomyoma: Clinical Value of Pseudocapsule Containing Peptides and Neurotransmitters.
      ). Cold loop hysteroscopic myomectomy seems to reduce thermal injury and spares the myometrial integrity of the fibroid pseudocapsule (
      • Tinelli A.
      • Favilli A.
      • Lasmar R.B.
      • Mazzon I.
      • Gerli S.
      • Xue X.
      • Malvasi A.
      The importance of pseudocapsule preservation during hysteroscopic myomectomy.
      ).
      Studies examining the effect of myomectomy on live birth are generally inconsistent with regard to the size, number and location of fibroids within the uterus. Moreover, they show a high risk of selection bias and confounding (primarily by age), and most of them are underpowered. Identical problems occur in studies focusing on early pregnancy loss, which makes the interpretation of the results strongly doubtful (
      ASRM
      Removal of myomas in asymptomatic patients to improve fertility and/or reduce miscarriage rate: a guideline.
      ;
      • Carranza-Mamane B.
      • Havelock J.
      • Hemmings R.
      The management of uterine fibroids in women with otherwise unexplained infertility.
      ;
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ;
      • Metwally M.
      • Raybould G.
      • Cheong Y.C.
      • Horne A.W.
      Surgical treatment of fibroids for subfertility.
      ;
      • Pritts E.A.
      • Parker W.H.
      • Olive D.L.
      Fibroids and infertility: an updated systematic review of the evidence.
      ). One randomized controlled trial (RCT) has studied reproductive outcomes after hysteroscopic fibroid resection (thus not laparoscopic/abdominal myomectomy) compared with a lack of surgical intervention for submucosal fibroids, and has shown a positive effect in favour of hysteroscopic resection (
      • Casini M.L.
      • Rossi F.
      • Agostini R.
      • Unfer V.
      Effects of the position of fibroids on fertility.
      ).
      A re-analysis of these data by the American Society for Reproductive Medicine (ASRM) Practice Committee and a subsequent systematic review suggested that the differences between the groups did not meet statistical significance, although the likelihood of a favourable effect on fertility could not be excluded (
      • Bosteels J.
      • van Wessel S.
      • Weyers S.
      • Broekmans F.J.
      • D'Hooghe T.M.
      • Bongers M.Y.
      • Mol B.W.J.
      Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities.
      ;
      • Metwally M.
      • Cheong Y.C.
      • Horne A.W.
      Surgical treatment of fibroids for subfertility.
      ). The removal of subserosal and transmural fibroids was also studied in only one RCT and only in patients before IVF. The authors found a beneficial effect of surgical removal, although only participants with one fibroid over 5 cm in diameter and a maximum of five fibroids were enrolled in the study (
      • Bulletti C.
      • D D.E.Z.
      • Levi Setti P.
      • Cicinelli E.
      • Polli V.
      • Stefanetti M.
      Myomas, pregnancy outcome, and in vitro fertilization.
      ).
      Previous research has not yet sufficiently determined the short- and long-term influence of fibroids on reproductive outcomes regarding location (or FIGO classification). Hence, when a woman presents with infertility but without fibroid-related symptoms, it is a unique challenge for physicians to provide evidence-based recommendations regarding treatment. This also applies to women with (severe) complaints of fibroids and a wish to preserve fertility. The aim of this retrospective cohort study was to compare the time to live birth and other reproductive outcomes after surgical treatment or lack of surgical treatment, in women with fibroids who wished to conceive, adjusting for differences between participants regarding patient and fibroid characteristics.

      Materials and methods

      Ethical approval, registration and assessment

      The study was approved by the Medical Ethical Committee of the Amsterdam UMC, VU University Medical Centre (VUmc) on 16 April 2019 (number 2019.225) and was registered at the Netherlands National Trial Register (Trial NL7990). Furthermore, the study was conducted according to STROBE guidelines (
      • von Elm E.
      • Altman D.G.
      • Egger M.
      • Pocock S.J.
      • Gøtzsche P.C.
      • Vandenbroucke J.P.
      The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.
      ), and a Risk Of Bias In Non-randomized Studies – of Interventions (ROBINS-I) assessment tool was used to assess the bias of the study (Appendix) (
      • Sterne J.A.
      • Hernán M.A.
      • Reeves B.C.
      • Savović J.
      • Berkman N.D.
      • Viswanathan M.
      • Henry D.
      • Altman D.G.
      • Ansari M.T.
      • Boutron I.
      • Carpenter J.R.
      • Chan A.W.
      • Churchill R.
      • Deeks J.J.
      • Hróbjartsson A.
      • Kirkham J.
      • Jüni P.
      • Loke Y.K.
      • Pigott T.D.
      • Ramsay C.R.
      • Regidor D.
      • Rothstein H.R.
      • Sandhu L.
      • Santaguida P.L.
      • Schünemann H.J.
      • Shea B.
      • Shrier I.
      • Tugwell P.
      • Turner L.
      • Valentine J.C.
      • Waddington H.
      • Waters E.
      • Wells G.A.
      • Whiting P.F.
      • Higgins J.P.
      ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
      ).

      Study design and patient selection

      Women between the age of 18 and 45 years at the time of diagnosis of uterine fibroids at the VUmc Amsterdam were included. Data were obtained from the patient medical file and one survey after obtaining informed consent. Patients were identified by electronic data extraction through screening for the diagnosis in the disease problem list, the disease label or the diagnostic code (International Classification of Diseases and Health Related Problems 10th revision) of the electronic patient file.
      The search was conducted in patients who presented at the VUmc between 2004 and 2018. This time frame was selected as medical patient records were not electronic before 2004 and it was estimated that at least 2 years of follow-up after possible surgical treatment was acceptable as a time interval for women to conceive. Those with uterine defects, such as congenital uterine anomalies (excluding arcuate uterus) or severe intrauterine adhesions, were excluded. Additionally, patients with severe endometriosis (ASRM grade 4; ") or dominant adenomyosis (as the uterus is more affected by adenomyosis than by uterine fibroids) were excluded.
      In September 2019, 1804 patients were sent a letter, the survey and an informed consent form with a stamped addressed return envelope. The letter included an overview of the study, what was expected from the participants, the possible advantages and disadvantages of participation, and the contact details of the main researchers. The informed consent form asked them to sign for whether they would or would not participate in the survey, and whether they objected to the use of their known medical data. A reminder was sent when they did not respond within 4 weeks.
      The survey involved questions regarding baseline characteristics (including ethnicity), complaints of fibroids, and a diagnostic/treatment/surgery/reproductive history. Data from the patient file were collected concerning the type and number of myomectomies (hysteroscopic/laparoscopic/abdominal myomectomy) and assisted reproductive technique (ART) interventions, namely intrauterine insemination (IUI), IVF and intracytoplasmic sperm injection (ICSI). Data on fibroid characteristics such as number, size and location were also collected (the location was either immediately classified according to FIGO or, when FIGO classification information was not described but ultrasound images were stored, classified retrospectively according to FIGO). Finally, data on reproductive outcomes were stored; these included the number and outcome of pregnancies, possible pregnancy complications such as preterm birth, and possible complications of delivery, for example uterine rupture.
      During a myomectomy at the VUmc the fibroid pseudocapsule is spared by using minimal coagulation, in combination with adhesion barriers and haemostatic therapy such as vasopressin. Furthermore, in the last 5 years, blood loss during myomectomy has been limited by temporary vascular occlusion using bulldog clamps on the uterine arteries and infundibulopelvic or ovarian ligament, reducing the need for additional coagulation during surgery, which allows proper suturing and approximation of the various layers. The cold loop procedure is only used in patients with large submucosal fibroids found during hysteroscopic myomectomy.

      Definitions

      The participants either continued expectant management or received myomectomy at some point during the follow-up; reasons for a myomectomy were infertility- and/or fibroid-related symptoms (e.g. abnormal uterine bleeding or bulky symptoms). For both treatments, the follow-up started at the diagnosis of fibroids at the authors’ hospital. Follow-up for the primary outcome ended at the first live birth or the end of the study. If the woman had no wish to conceive at the time of diagnosis, follow-up started at the time she was actively trying to conceive. The primary outcome was time to live birth, defined as the time till first live birth after the start of follow-up.
      For the secondary outcomes, the follow-up did not end at the first live birth, but data were collected retrospectively for all outcomes until the end of the study. Secondary outcomes were use of ART, type of ART, mode of conception, total number of live births (twin pregnancies were counted as one live birth), ongoing pregnancies, defined as a viable intrauterine pregnancy after 12 weeks of gestation, miscarriages, defined as the spontaneous loss of a clinical pregnancy before 24 weeks’ gestational age, intrauterine death after 24 weeks, preterm birth before 37 weeks, mode of delivery and complications during delivery.

      Potential confounders

      Confounders identified a priori by expert opinion and literature were female age, duration of infertility, previous live birth, education level, smoking, body mass index (BMI), diagnosis from the fertility check-up, and physical complaints from the fibroids in addition to the number, size, location and FIGO classification of the fibroids as determined on the ultrasound at diagnosis. Non-informative confounders were removed based on the Akaike information criterion (AIC) increasing by 2 or more (
      • Akaike H.
      A New Look at the Statistical Model Identification.
      ). The AIC is a statistic that contrasts the amount of information a model comprises with the complexity of that same model. The higher the AIC, the less information it contains relative to the parameters that have been added. This study used a threshold of an increase by 2 points to indicate that the confounder was not informative.

      Missing data

      Primary and secondary outcome data were complete. Baseline characteristics data were missing for a total of 5.7% of cells and were accounted for using single imputation including the cumulative hazard for a live birth to incorporate the time in follow-up (
      • White I.R.
      • Royston P.
      Imputing missing covariate values for the Cox model.
      ).

      Statistical analyses

      The baseline characteristics of participants who eventually underwent myomectomy during follow-up were compared with those who did not, using a table.
      Because myomectomy was a procedure that occurred over time, with some participants receiving it almost immediately and some waiting years, the survival analyses had to be planned accordingly to allow for a fair comparison. To estimate the effect of myomectomy over time, a primary analysis and two sensitivity analyses using alternative methodology were chosen. For all Cox models, possible violations of the proportional hazards assumption were checked (
      • Grambsch P.
      • Therneau T.
      Proportional hazards tests and diagnostics based on weighted residuals.
      ).

      Primary analysis

      First, after separating the follow-up in periods of expectant management and myomectomy, a Cox proportional hazards model was fitted using left truncation and a time-varying covariate for myomectomy including all previously mentioned selected confounders for adjustment. This way, a woman who received myomectomy 1 year after diagnosis started out as a part of the ‘expectant management’ group, i.e. there was no myomectomy, and after that year and her procedure, her treatment status and the rest of her follow-up were changed to the ‘myomectomy’ group. To keep the model as simple as possible, informative confounders were selected based on AIC and this set of confounders was used for the other analyses. In addition, by adding an interaction between complaints of symptoms and myomectomy, the study investigated whether the effect of myomectomy was different for women with symptoms from their fibroids compared with those who had no symptoms.

      Sensitivity analyses

      The first sensitivity analysis, instead of adding confounders to the model, aimed to balance the confounders between the expectant management and myomectomy groups over time, as this is more similar to the design of an RCT with women receiving myomectomy at different points in time. To this end, an inverse probability weighting of treatment was used, in which the relationship between confounders and treatment was used to re-balance the treatment groups such that the groups were similar over time (
      • Cole S.R.
      • Hernán M.A.
      Constructing inverse probability weights for marginal structural models.
      ).
      First a Cox model with the selected confounders was fitted that predicted the probability of undergoing myomectomy over time and updated these probabilities every 2 weeks (
      • Dreyer K.
      • van Eekelen R.
      • Tjon-Kon-Fat R.I.
      • van der Steeg J.W.
      • Steures P.
      • Eijkemans M.
      • van der Veen F.
      • Hompes P.
      • Mol B.
      • van Geloven N.
      The therapeutic effect of hysterosalpingography in couples with unexplained subfertility: a post-hoc analysis of a prospective multi-centre cohort study.
      ). Next, the probability of receiving a myomectomy in the entire cohort was divided by the individual probabilities of receiving treatment to obtain weights for balance, and this process was then repeated until the weights converged and the procedure was completed, known as iterative inverse probability weighting (
      • van der Wal W.
      Causal modelling in epidemiological practice. Chapter 8 Using iterative probability weighting to improve causal effect estimates.
      ). Those weights were then applied in a Cox model with a time-varying covariate for myomectomy.
      For the second sensitivity analysis the ‘cloning’ methodological approach for a treatment that is started later in time was chosen; the principle is that if the treatment decision is made at some point after diagnosis, ‘strategies’ to eventually treat or not treat are compared rather than assigning treatment at a fixed time point as in an RCT (
      • Hernán M.A.
      How to estimate the effect of treatment duration on survival outcomes using observational data.
      ). Initially the data from all patients for both strategies were used, and then patients were followed over time and censored, i.e. removed from analysis when they did not adhere to their assigned strategy, which here involves receiving treatment in the expectant management ‘strategy’ group. For example, a patient who received myomectomy at 1 year after diagnosis provided data on expectant management for both strategy groups during this year, was followed after myomectomy in the treatment strategy group but was censored at the time of myomectomy in the expectant management group. Inverse probability weighting of treatment was then applied using a priori selected confounders in logistic regression to weight and adjust for this censoring, similar to the approach described in the previous paragraph (
      • Cole S.R.
      • Hernán M.A.
      Constructing inverse probability weights for marginal structural models.
      ;
      • Hernán M.A.
      How to estimate the effect of treatment duration on survival outcomes using observational data.
      ). Finally, adjusted weighted Kaplan–Meier curves were fitted for both groups and a log-rank test was conducted to compare the curves. What was obtained was the absolute chance of a live birth over time for both strategies, something the hazard ratio itself does not provide.

      Software

      Data were prepared in SPSS 26 (IBM, USA) and analysed in R version 3.6.0 (
      Team, R.C.
      R: A language and environment for statistical computing.
      ) using the rms, foreign, mice, dplyr and prodlim R packages.

      Results

      Descriptive statistics

      In total 526 participants were included in the study. The group of patients who had a wish to conceive at the start of follow-up consisted of 311 patients who were included in the analysis. The median time to myomectomy was 14 months, the median follow-up after the procedure was 24 months, and the median follow-up for expectant management was 18 months. Again because of the type of analysis, patients were first analysed in the expectant group and, if myomectomy was performed, from the moment of myomectomy in the group of patients whom eventually received a myomectomy. The reason for myomectomy was the presence of symptoms for 49 women (30%), a wish to conceive for 50 (30%), or both symptoms and a wish to conceive for 62 (38%), with data missing for 4 participants (2%).
      The baseline characteristics of the women who remained on expectant management and those eventually receiving myomectomy are depicted in Table 1. Additional baseline characteristics including complaints from the fibroids, smoking status, infertility diagnosis and the surgical variables of the patients who eventually received a myomectomy are shown in the Supplementary Table. The largest differences between the groups at first glance were the largest fibroid diameter being >7 cm (in 48% of women who received myomectomy versus 15% of those continuing expectant management), and receiving a fertility check-up (56% for the myomectomy group versus 76% for expectant management). In addition the mean number of fibroids (2.0 [2.5–97.5th percentile 1.0–14.9] for the myomectomy group versus 1.0 [1.0–7.4] for expectant management) and the fibroid location (fibroid deforming the uterine cavity: 54% in the myomectomy group versus 36% for expectant management) differed between the groups.
      Table 1Baseline characteristics of the women who remained on expectant management and those who eventually received myomectomy
      ParameterPatients who eventually received myomectomy (n = 165)Patients who remained on expectant management(n = 146)
      Female age at diagnosis (years)34.6 (26.0–43.9)35.7 (28.0–43.4)
      BMI (kg/m2)26.2 (19.2–40.9)25.7 (19.0–35.3)
      Infertile (yes versus no)106 (64)115 (79)
      Duration of infertility at diagnosis (years), median1.6 (0.0–10.7)2.1 (0.0–13.3)
      Primary infertility (versus secondary)147 (89)129 (88)
      Fertility check–up (yes versus no)93 (56)111 (76)
      Complaints from fibroids (versus no)141 (85)98 (67)
      Number of fibroids (median)2.0 (1.0–14.9)1.0 (1.0–7.4)
      Largest fibroid diameter
      <1 cm3 (2)2 (1)
      1–2 cm12 (7)33 (23)
      2–3 cm20 (12)26 (18)
      3–4 cm13 (8)28 (19)
      4–5 cm11 (7)13 (9)
      5–6 cm13 (8)14 (10)
      6–7 cm13 (8)8 (5)
      >7 cm80 (48)22 (15)
      Fibroid location(s)
      Only submucosal21 (13)9 (6)
      Only transmural21 (13)11 (8)
      Multiple locations: at least one submucosal or transmural fibroid in combination with another fibroid35 (21)10 (7)
      Only intramural30 (18)47 (32)
      Only subserosal24 (15)47 (32)
      Multiple locations; only a combination of intramural and subserosal fibroids34 (21)22 (15)
      FIGO classification
      The FIGO location was classified according to Munro and colleagues (Munro et al., 2011). When the classification was not described and ultrasound images were stored, fibroids were classified retrospectively. BMI, body mass index.
      At least one FIGO type 0, 1, 2, 3 and/or 2–5 (causing deformation of the intrauterine cavity)89 (54)52 (36)
      Only FIGO type 4, 5, 6, 7 and/or 8 (no deformation of the intrauterine cavity)76 (46)94 (64)
      Data are mean unless indicated as median; values in brackets are percentages or the range between the 2.5th and 97.5th percentiles.
      a The FIGO location was classified according to Munro and colleagues (
      • Munro M.G.
      • Critchley H.O.
      • Broder M.S.
      • Fraser I.S.
      FIGO classification system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age.
      ). When the classification was not described and ultrasound images were stored, fibroids were classified retrospectively.
      BMI, body mass index.

      Primary analysis

      After fitting the Cox proportional hazards model with a time-varying covariate for myomectomy, the confounders of fertility check-up, smoking and fibroid FIGO location were removed for all the reported analyses, as these led to poorer statistical fits due to an increase in AIC. An adjusted hazard ratio was estimated for the effect of myomectomy compared with expectant management of 1.26 (95% confidence interval [CI] 0.87–1.81), thus showing no statistical evidence of a benefit. There was no difference in the effect of myomectomy compared with expectant management for women with or without complaints arising from their fibroids (P for interaction = 0.65).

      Sensitivity analyses

      Applying the survival weights, a hazard ratio of 1.15 (95% CI 0.74–1.77) we estimated, which was in agreement with the primary analysis. After applying the cloning approach and obtaining the two ‘strategy’ groups, a weighted Kaplan–Meier curve was produced (Figure 1). The log-rank test for the difference yielded a value of P = 0.20, again indicating a lack of statistical evidence for a clear effect of myomectomy.
      Figure 1
      Figure 1Weighted Kaplan–Meier curves for the cohort of women adhering to a strategy of expectant management or eventually receiving myomectomy.

      Secondary outcomes

      The secondary outcomes of the women who remained on expectant management and those eventually receiving myomectomy are displayed in Table 2. The total obstetric history is included. The largest differences were seen in total ART use and type, and the mode of conception of the first live birth. Ninety-two (63%) of the couples who remained on expectant management received ART versus 63 (38%) of the couples who eventually received myomectomy. Most conceptions that led to the first live birth in the myomectomy group occurred naturally (67%), while IVF/ICSI (45%) was the most common mode of conception in the expectant management group.
      Table 2Secondary outcomes of the women who remained on expectant management and those who eventually received myomectomy
      ParameterPatients who eventually received myomectomy (n = 165)Patients who remained on expectant management (n = 146)
      Mode of conception leading to first live birth


      Total


      73


      100


      Natural conception


      49 (67)


      36 (36)


      Ovulation induction


      1 (1)


      3 (3)


      IUI (with and without stimulation)


      3 (4)


      17 (17)


      IVF/ICSI


      20 (27)


      45 (45)
      Total ART63 (38)92 (63)
      Number of ART cycles
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.


      IUI (with and without stimulation)


      156


      334


      IVF/ICSI


      82


      158
      Total live births115
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      159
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      Total ongoing pregnancies301
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      306
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      Total miscarriages under 24 weeks105
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      100
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      Total intrauterine fetal deaths after 24 weeks1
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      6
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      Total preterm birth at 24–37 weeks15
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      15
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      Total deliveries


      Total
      115
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      159
      Some participants had more than one pregnancy during the follow-up, and all pregnancies were included. ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.


      Vaginal


      42



      96



      Assisted delivery
      310


      Primary Caesarean section


      49


      22


      Secondary Caesarean section


      21


      31
      Complication of (at least one) delivery16/73 (22)14/100 (14)
      Total uterine rupture1 (1)0
      Data are n or n/N (%).
      a Some participants had more than one pregnancy during the follow-up, and all pregnancies were included.ART, assisted reproductive technology; ICSI, intracytoplasmic sperm injection; IUI, intrauterine insemination.
      Complications during delivery arose for 14 women (14% of those who delivered at least once) in the expectant management group and 16 women (22% of those who delivered at least once) in the myomectomy group. A uterine rupture during delivery occurred for one participant (1%) in the group of patients eventually receiving myomectomy. In this case the uterine rupture occurred 2 years after the myomectomy, with no history of Caesarean section. This patient received two myomectomies in 1 year, one hysteroscopic and one laparoscopic. During the latter, two subserosal and two transmural fibroids were removed. The largest fibroid was 5 cm and was removed by entering the uterine cavity; therefore an elective Caesarean section was advised. However, labour started spontaneously with contractions at 35 + 3 weeks of pregnancy, after which the uterine rupture occurred. A Caesarean section was then performed and a healthy child was delivered.

      Discussion

      No significant difference was found in the time to live birth between the myomectomy and expectant management groups (adjusted hazard ratio 1.26, 95% CI 0.87–1.81). The secondary outcomes showed no clear differences in other obstetric outcomes between the women in the myomectomy or expectant management group. When looking at ART and mode of conception, participants who eventually received myomectomy underwent ART considerably less often and, consequently, more frequently conceived naturally compared with those who continued expectant management. In addition, if they did receive ART, they were likely to skip IUI and start with IVF/ICSI. Because of the retrospective nature of the study, the characteristics of the participants differed between groups, most notably in terms of fibroid size (fibroid >7 cm: myomectomy group 48%, expectant management group 15%), fibroid number (myomectomy group 2 [1.0–14.9], expectant management 1 [1.0–7.4]) and fibroid type (fibroid type causing deformation of the intrauterine cavity: myomectomy group 54%, expectant management group 36%).
      In accordance with these results, previous studies have demonstrated that a clear effect of myomectomy has not been proven (
      • Bosteels J.
      • van Wessel S.
      • Weyers S.
      • Broekmans F.J.
      • D'Hooghe T.M.
      • Bongers M.Y.
      • Mol B.W.J.
      Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities.
      ;
      • Brady P.C.
      • Stanic A.K.
      • Styer A.K.
      Uterine fibroids and subfertility: an update on the role of myomectomy.
      ;
      • Fernandez H.
      • Sefrioui O.
      • Virelizier C.
      • Gervaise A.
      • Gomel V.
      • Frydman R.
      Hysteroscopic resection of submucosal myomas in patients with infertility.
      ;
      • Khaw S.C.
      • Anderson R.A.
      • Lui M.W.
      Systematic review of pregnancy outcomes after fertility-preserving treatment of uterine fibroids.
      ;
      • Klatsky P.C.
      • Tran N.D.
      • Caughey A.B.
      • Fujimoto V.Y.
      Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
      ;
      • Metwally M.
      • Raybould G.
      • Cheong Y.C.
      • Horne A.W.
      Surgical treatment of fibroids for subfertility.
      ;
      • Orlando M.
      • Kollikonda S.
      • Hackett L.
      • Kho R.
      Non-hysteroscopic myomectomy and fertility outcomes: a systematic review.
      ;
      • Pritts E.A.
      • Parker W.H.
      • Olive D.L.
      Fibroids and infertility: an updated systematic review of the evidence.
      ). However, these studies all show that a myomectomy does not decrease fertility rates either and tends to show (non-significant) positive effects of surgical treatment.
      The current study also included women who received myomectomy primarily because of symptom complaints, and not infertility. Although the duration of trying to conceive at diagnosis was accounted for, this could have diluted the results because some fibroids can give complaints without deforming the uterine cavity, affecting myometrial contractions or influencing endometrial angiogenesis, which are depicted as important factors negatively influencing fertility (
      • Brady P.C.
      • Stanic A.K.
      • Styer A.K.
      Uterine fibroids and subfertility: an update on the role of myomectomy.
      ).
      In addition, dyspareunia caused by fibroids, in turn leading to a low exposure to intercourse, can negatively affect fertility. Although sexual complaints were experienced in both groups (myomectomy group 22%, expectant management group 14%, as depicted in the Supplementary Table) and this was included in the complaints assigned as a confounder, it is interesting that 1 year after myomectomy only 9% (15 patients) of women were experiencing sexual complaints and natural conception was higher after myomectomy (67%, versus 36% in the expectant management group).
      This is the first retrospective study to examine the difference in reproductive outcomes between participants after myomectomy or expectant management that includes all types of fibroids and has categorized them (retrospectively) according to their FIGO location. A strenghts of the study is that both the general gynaecological population and the population after ART were included in this study. This provides an insight into which type of patient is more likely to receive a myomectomy and to assess the effect of the myomectomy on reproductive outcomes. Also, the participants provided information via a survey as an addition to the information available in the patient medical file, which resulted in low amounts of missing data. Another strenght of the study is that by using statistical techniques to adjust for confounding and triangulating this approach using three methods in total, this retrospective study aimed to make a most equitable comparison between groups (
      • Hernán M.A.
      How to estimate the effect of treatment duration on survival outcomes using observational data.
      ). Different advantages of these contemporary methods were the simplicity (the confounders in the model), the induction of balance over time (inverse probability weighting over time) and the comparison of strategies as in a trial and the estimation of the absolute chances of a live birth (cloning/censoring/weighting) (
      • Hernán M.A.
      How to estimate the effect of treatment duration on survival outcomes using observational data.
      ;
      • Hernán M.A.
      • Alonso A.
      • Logan R.
      • Grodstein F.
      • Michels K.B.
      • Willett W.C.
      • Manson J.E.
      • Robins J.M.
      Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease.
      ;
      • Hernán M.A.
      • Robins J.M.
      Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available.
      ).
      However, the study also has some limitations. Information on the date that the ART was ended was missing, such that ART usage could not be adjusted for over time. This is expected to dilute the estimated effect of myomectomy due to more ART usage in the expectant management group. In addition, due to the observational nature of the study, there could be residual confounding due to other factors that were not accounted for. A third limitation is selection bias that occurred because some specific patients could be more prone to reject participation in this study. For instance, it was seen that women who had an obstetric history with multiple adverse outcomes were less likely to participate, but this is inherent to the design. Despite this, after comparing the baseline characteristics of non-responders and responders, the study found no significant differences in total live birth (P = 0.998), ethnicity (P = 0.068) or BMI (P = 0.888). There was, however, a difference in age (38.5 versus 35.1 years; P = 0.001). Furthermore, the number of participants included was too low to stratify for specific fibroid characteristics, such as the FIGO location, in relation to reproductive outcomes.
      In addition, Figure 2 shows that the cumulative chance of a live birth is below 40% in both study groups. The probability of conception in healthy people is around 95% after 2 years, so compared with this number, the cumulative conception in this study is much lower (
      • Taylor A.
      ABC of subfertility: extent of the problem.
      ). However, a selected group of people were analysed. In the Netherlands there is a tendency to be very conservative in relation to uterine fibroids, and in most centres expectant management is proposed unless severe symptoms occur. The women who visit the authors’ uterine repair centre are mostly tertiary referrals because of complexity or visit the hospital in a very late stage after the failure of various other therapies; as a consequence they have a lower probability of a live birth. This is underlined by the baseline characteristics reported in Table 1. In total 64% in of the women in the myomectomy group and 79% in the expectant management group were infertile. Moreover, the median age of the participants was around 35 years (Table 1), indicating age as an important factor negatively influencing the conception rates in this study – it is known that the cumulative probabilities of conception decline with age because of a diminished ovarian reserve (
      • Taylor A.
      ABC of subfertility: extent of the problem.
      ). Additionally, the fertility of women with fibroids is influenced by changes in the endometrial environment, pressure complaints in the tube(s) and/or endometrium, and changes in myometrial contractility, negatively influencing conception and implantation (
      • Navarro A.
      • Bariani M.V.
      • Yang Q.
      • Al-Hendy A.
      Understanding the Impact of Uterine Fibroids on Human Endometrium Function.
      ).
      After a woman has undergone myomectomy, she has a recovery time of up to 6 months in which she is advised not to conceive. Despite this inherent negative effect on the time to live birth, no clear negative effect of myomectomy was found. There was thus more evidence in this study that any negative effect due to waiting after the procedure could have been offset by a possible positive effect of myomectomy, which, in addition to reducing symptoms, provides arguments to perform a myomectomy.

      Conclusion

      Comparing the two study groups, no significant difference was found in the time to live birth, although this comparison regarding live birth is limited due to the waiting period of patients before and after the myomectomy procedure. Given that after myomectomy the time to live birth rate is similar to that of expectant management, despite the waiting period, and considering the other advantages of myomectomy in terms of reducing symptoms, the logical next step after this study and reviewing the accumulated evidence is to conduct an RCT. This should include infertile women or those with an actual desire to conceive who have fibroids that are not, or are minimally distorting, the uterine cavity. To unravel the effect of specific fibroid characteristics on reproductive outcomes, it is important that future results are stratified by the size, number and FIGO location of the fibroids. This is essential to research the effect of myomectomy free of confounding and selection and, in addition, to further determine which specific patients, with regards to fibroid characteristics, obtain the most benefit from surgical treatment.
      Providing patient-specific care for women with fibroids and infertility remains a challenge. Additional to the evaluation of fibroid characteristics, ovarian reserve is an important factor to assess. Because this study shows no negative effect on fertility after myomectomy, it provides clinicians with support to advise myomectomy in women with fibroid-related symptoms and a wish to conceive.

      Acknowledgements

      The authors thank Claire H. Laeven and Guus Vissers for their help in contacting patients and data-entry work in Castor during their scientific internships.

      References

        • Akaike H.
        A New Look at the Statistical Model Identification.
        in: Parzen E. Tanabe K. Kitagawa G. Selected Papers of Hirotugu Akaike. Springer, New York1998: 215-222https://doi.org/10.1007/978-1-4612-1694-0_16
        • ASRM
        Removal of myomas in asymptomatic patients to improve fertility and/or reduce miscarriage rate: a guideline.
        Fertil. Steril. 2017; 108: 416-425https://doi.org/10.1016/j.fertnstert.2017.06.034
        • Behbehani S.
        • Polesello S.
        • Hasson J.
        • Silver J.
        • Son W.Y.
        • Dahan M.
        The Effect of Intramural Myomas Without an Intracavity Component on In Vitro Fertilization Outcomes in Single Fresh Blastocyst Transfer Cycles.
        J. Minim. Invasive Gynecol. 2018; 25: 1241-1248https://doi.org/10.1016/j.jmig.2018.03.005
        • Bosteels J.
        • van Wessel S.
        • Weyers S.
        • Broekmans F.J.
        • D'Hooghe T.M.
        • Bongers M.Y.
        • Mol B.W.J.
        Hysteroscopy for treating subfertility associated with suspected major uterine cavity abnormalities.
        Cochrane Database Syst. Rev. 2018; 12Cd009461https://doi.org/10.1002/14651858.CD009461.pub4
        • Brady P.C.
        • Stanic A.K.
        • Styer A.K.
        Uterine fibroids and subfertility: an update on the role of myomectomy.
        Curr. Opin. Obstet. Gynecol. 2013; 25: 255-259https://doi.org/10.1097/GCO.0b013e3283612188
        • Bulletti C.
        • D D.E.Z.
        • Levi Setti P.
        • Cicinelli E.
        • Polli V.
        • Stefanetti M.
        Myomas, pregnancy outcome, and in vitro fertilization.
        Ann. N. Y. Acad. Sci. 2004; 1034: 84-92https://doi.org/10.1196/annals.1335.010
        • Carranza-Mamane B.
        • Havelock J.
        • Hemmings R.
        The management of uterine fibroids in women with otherwise unexplained infertility.
        J. Obstet. Gynaecol. Can. 2015; 37: 277-285https://doi.org/10.1016/s1701-2163(15)30318-2
        • Casini M.L.
        • Rossi F.
        • Agostini R.
        • Unfer V.
        Effects of the position of fibroids on fertility.
        Gynecological Endocrinology. 2006; 22: 106-109https://doi.org/10.1080/09513590600604673
        • Christopoulos G.
        • Vlismas A.
        • Salim R.
        • Islam R.
        • Trew G.
        • Lavery S.
        Fibroids that do not distort the uterine cavity and IVF success rates: an observational study using extensive matching criteria.
        Bjog. 2017; 124: 615-621https://doi.org/10.1111/1471-0528.14362
        • Cole S.R.
        • Hernán M.A.
        Constructing inverse probability weights for marginal structural models.
        Am. J. Epidemiol. 2008; 168: 656-664https://doi.org/10.1093/aje/kwn164
        • Dreyer K.
        • van Eekelen R.
        • Tjon-Kon-Fat R.I.
        • van der Steeg J.W.
        • Steures P.
        • Eijkemans M.
        • van der Veen F.
        • Hompes P.
        • Mol B.
        • van Geloven N.
        The therapeutic effect of hysterosalpingography in couples with unexplained subfertility: a post-hoc analysis of a prospective multi-centre cohort study.
        Reprod. Biomed. Online. 2019; 38: 233-239https://doi.org/10.1016/j.rbmo.2018.11.005
        • Fernandez H.
        • Sefrioui O.
        • Virelizier C.
        • Gervaise A.
        • Gomel V.
        • Frydman R.
        Hysteroscopic resection of submucosal myomas in patients with infertility.
        Hum. Reprod. 2001; 16: 1489-1492https://doi.org/10.1093/humrep/16.7.1489
        • Grambsch P.
        • Therneau T.
        Proportional hazards tests and diagnostics based on weighted residuals.
        Biometrika. 1994; 81: 515-526
        • Hernán M.A.
        How to estimate the effect of treatment duration on survival outcomes using observational data.
        BMJ. 2018; 360: k182https://doi.org/10.1136/bmj.k182
        • Hernán M.A.
        • Alonso A.
        • Logan R.
        • Grodstein F.
        • Michels K.B.
        • Willett W.C.
        • Manson J.E.
        • Robins J.M.
        Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease.
        Epidemiology. 2008; 19: 766-779https://doi.org/10.1097/EDE.0b013e3181875e61
        • Hernán M.A.
        • Robins J.M.
        Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available.
        Am. J. Epidemiol. 2016; 183: 758-764https://doi.org/10.1093/aje/kwv254
        • Karlsen K.
        • Mogensen O.
        • Humaidana P.
        • Kesmodel U.S.
        • Ravn P.
        Uterine fibroids increase time to pregnancy: a cohort study.
        Eur. J. Contracept Reprod. Health Care. 2020; 25: 37-42https://doi.org/10.1080/13625187.2019.1699047
        • Khaw S.C.
        • Anderson R.A.
        • Lui M.W.
        Systematic review of pregnancy outcomes after fertility-preserving treatment of uterine fibroids.
        Reprod. Biomed. Online. 2020; 40: 429-444https://doi.org/10.1016/j.rbmo.2020.01.003
        • Klatsky P.C.
        • Tran N.D.
        • Caughey A.B.
        • Fujimoto V.Y.
        Fibroids and reproductive outcomes: a systematic literature review from conception to delivery.
        Am. J. Obstet. Gynecol. 2008; 198: 357-366https://doi.org/10.1016/j.ajog.2007.12.039
        • Metwally M.
        • Cheong Y.C.
        • Horne A.W.
        Surgical treatment of fibroids for subfertility.
        Cochrane Database Syst. Rev. 2012; 11Cd003857https://doi.org/10.1002/14651858.CD003857.pub3
        • Metwally M.
        • Raybould G.
        • Cheong Y.C.
        • Horne A.W.
        Surgical treatment of fibroids for subfertility.
        Cochrane Database Syst. Rev. 2020; 1Cd003857https://doi.org/10.1002/14651858.CD003857.pub4
        • Munro M.G.
        • Critchley H.O.
        • Broder M.S.
        • Fraser I.S.
        FIGO classification system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age.
        Int. J. Gynaecol. Obstet. 2011; 113: 3-13https://doi.org/10.1016/j.ijgo.2010.11.011
        • Navarro A.
        • Bariani M.V.
        • Yang Q.
        • Al-Hendy A.
        Understanding the Impact of Uterine Fibroids on Human Endometrium Function.
        Frontiers in cell and developmental biology. 2021; 9 (633180-633180)https://doi.org/10.3389/fcell.2021.633180
        • Olive D.L.
        • Pritts E.A.
        Fibroids and reproduction.
        Semin. Reprod. Med. 2010; 28: 218-227https://doi.org/10.1055/s-0030-1251478
        • Oliveira F.G.
        • Abdelmassih V.G.
        • Diamond M.P.
        • Dozortsev D.
        • Melo N.R.
        • Abdelmassih R.
        Impact of subserosal and intramural uterine fibroids that do not distort the endometrial cavity on the outcome of in vitro fertilization-intracytoplasmic sperm injection.
        Fertil. Steril. 2004; 81: 582-587https://doi.org/10.1016/j.fertnstert.2003.08.034
        • Orlando M.
        • Kollikonda S.
        • Hackett L.
        • Kho R.
        Non-hysteroscopic myomectomy and fertility outcomes: a systematic review.
        J. Minim. Invasive Gynecol. 2020; https://doi.org/10.1016/j.jmig.2020.10.006
        • Parker W.H.
        Etiology, symptomatology, and diagnosis of uterine myomas.
        Fertil. Steril. 2007; 87: 725-736https://doi.org/10.1016/j.fertnstert.2007.01.093
        • Pritts E.A.
        • Parker W.H.
        • Olive D.L.
        Fibroids and infertility: an updated systematic review of the evidence.
        Fertil. Steril. 2009; 91: 1215-1223https://doi.org/10.1016/j.fertnstert.2008.01.051
        • Revised American Society for Reproductive Medicine classification of endometriosis: 1996
        Fertil. Steril. 1997; 67: 817-821https://doi.org/10.1016/s0015-0282(97)81391-x
        • Rikhraj K.
        • Tan J.
        • Taskin O.
        • Albert A.Y.
        • Yong P.
        • Bedaiwy M.A.
        The Impact of Noncavity-Distorting Intramural Fibroids on Live Birth Rate in In Vitro Fertilization Cycles: A Systematic Review and Meta-Analysis.
        J. Womens Health (Larchmt). 2020; 29: 210-219https://doi.org/10.1089/jwh.2019.7813
        • Somigliana E.
        • Vercellini P.
        • Daguati R.
        • Pasin R.
        • De Giorgi O.
        • Crosignani P.G.
        Fibroids and female reproduction: a critical analysis of the evidence.
        Hum. Reprod. Update. 2007; 13: 465-476https://doi.org/10.1093/humupd/dmm013
        • Sterne J.A.
        • Hernán M.A.
        • Reeves B.C.
        • Savović J.
        • Berkman N.D.
        • Viswanathan M.
        • Henry D.
        • Altman D.G.
        • Ansari M.T.
        • Boutron I.
        • Carpenter J.R.
        • Chan A.W.
        • Churchill R.
        • Deeks J.J.
        • Hróbjartsson A.
        • Kirkham J.
        • Jüni P.
        • Loke Y.K.
        • Pigott T.D.
        • Ramsay C.R.
        • Regidor D.
        • Rothstein H.R.
        • Sandhu L.
        • Santaguida P.L.
        • Schünemann H.J.
        • Shea B.
        • Shrier I.
        • Tugwell P.
        • Turner L.
        • Valentine J.C.
        • Waddington H.
        • Waters E.
        • Wells G.A.
        • Whiting P.F.
        • Higgins J.P.
        ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
        BMJ. 2016; 355: i4919https://doi.org/10.1136/bmj.i4919
        • Styer A.K.
        • Jin S.
        • Liu D.
        • Wang B.
        • Polotsky A.J.
        • Christianson M.S.
        • Vitek W.
        • Engmann L.
        • Hansen K.
        • Wild R.
        • Legro R.S.
        • Coutifaris C.
        • Alvero R.
        • Robinson R.D.
        • Casson P.
        • Christman G.M.
        • Christy A.
        • Diamond M.P.
        • Eisenberg E.
        • Zhang H.
        • Santoro N.
        Association of uterine fibroids and pregnancy outcomes after ovarian stimulation-intrauterine insemination for unexplained infertility.
        Fertil. Steril. 2017; 107 (756-762.e753)https://doi.org/10.1016/j.fertnstert.2016.12.012
        • Sunkara S.K.
        • Khairy M.
        • El-Toukhy T.
        • Khalaf Y.
        • Coomarasamy A.
        The effect of intramural fibroids without uterine cavity involvement on the outcome of IVF treatment: a systematic review and meta-analysis.
        Hum. Reprod. 2010; 25: 418-429https://doi.org/10.1093/humrep/dep396
        • Taylor A.
        ABC of subfertility: extent of the problem.
        BMJ (Clinical research ed.). 2003; 327: 434-436https://doi.org/10.1136/bmj.327.7412.434
        • Team, R.C.
        R: A language and environment for statistical computing.
        R Foundation for Statistical Computing, Vienna, Austria2019 (URL)
        • Thubert T.
        • Foulot H.
        • Vinchant M.
        • Santulli P.
        • Marzouk P.
        • Borghese B.
        • Chapron C.
        Surgical treatment: Myomectomy and hysterectomy; Endoscopy: A major advancement.
        Best Pract. Res. Clin. Obstet. Gynaecol. 2016; 34: 104-121https://doi.org/10.1016/j.bpobgyn.2015.11.021
        • Tian Y.C.
        • Wu J.H.
        • Wang H.M.
        • Dai Y.M.
        Improved Fertility Following Enucleation of Intramural Myomas in Infertile Women.
        Chin. Med. J. (Engl.). 2017; 130: 1648-1653https://doi.org/10.4103/0366-6999.209900
        • Tinelli A.
        • Favilli A.
        • Lasmar R.B.
        • Mazzon I.
        • Gerli S.
        • Xue X.
        • Malvasi A.
        The importance of pseudocapsule preservation during hysteroscopic myomectomy.
        Eur. J. Obstet. Gynecol. Reprod. Biol. 2019; 243: 179-184https://doi.org/10.1016/j.ejogrb.2019.09.008
        • Tinelli A.
        • Malvasi A.
        • Hurst B.S.
        • Tsin D.A.
        • Davila F.
        • Dominguez G.
        • Dell'edera D.
        • Cavallotti C.
        • Negro R.
        • Gustapane S.
        • Teigland C.M.
        • Mettler L.
        Surgical management of neurovascular bundle in uterine fibroid pseudocapsule.
        JSLS: Journal of the Society of Laparoendoscopic Surgeons. 2012; 16: 119-129https://doi.org/10.4293/108680812X13291597716302
        • Tinelli A.
        • Mynbaev O.A.
        • Sparic R.
        • Vergara D.
        • Di Tommaso S.
        • Salzet M.
        • Maffia M.
        • Malvasi A.
        Angiogenesis and Vascularization of Uterine Leiomyoma: Clinical Value of Pseudocapsule Containing Peptides and Neurotransmitters.
        Curr. Protein Pept. Sci. 2017; 18: 129-139https://doi.org/10.2174/1389203717666160322150338
        • van der Wal W.
        Causal modelling in epidemiological practice. Chapter 8.
        (PhD thesis) University of Amsterdam, Amsterdam, the Netherlands2011
        • von Elm E.
        • Altman D.G.
        • Egger M.
        • Pocock S.J.
        • Gøtzsche P.C.
        • Vandenbroucke J.P.
        The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.
        J. Clin. Epidemiol. 2008; 61: 344-349https://doi.org/10.1016/j.jclinepi.2007.11.008
        • White I.R.
        • Royston P.
        Imputing missing covariate values for the Cox model.
        Statistics in medicine. 2009; 28: 1982-1998https://doi.org/10.1002/sim.3618

      Biography

      Emmy Don, MD, is a PhD candidate at Amsterdam University Medical Centre in Amsterdam, the Netherlands. Her fields of interests include reproductive medicine and benign gynaecology.
      Key Message
      Despite the recovery period and more, larger and more symptomatic fibroids in the myomectomy group, the time to live birth was not significantly different after myomectomy compared with expectant management in patients with fibroids. Additionally, fibroid-related symptoms were reduced and ART use was lower after myomectomy. This supports conducting a randomized controlled trial.