Endometriosis is an inflammatory disease characterised by an ectopic proliferation of endometrial tissue (stroma and epithelial cells). It affects around 10% of women of reproductive age, and this prevalence has increased in the last decades [1,2]. While it is a benign disease, it has a high impact on the patient's quality of life as well as some peculiar characteristics, such as local infiltration and distant dissemination.

It presents as three different entities: superficial peritoneal lesions, deep endometriosis or ovarian endometriosis, but the ovaries are the most common site of endometriosis [3]. Ovarian endometriosis may manifest as superficial implants associated with fibrous adhesions, micro intra-ovarian endometriomas or deep implants with repeated cyclic hemorrhage, resulting in endometriotic cysts (endometriomas). The exact prevalence of endometriomas is not known, but their presence has been reported in 17–44% of women with endometriosis [3,4].

A diagnosis of endometriosis can be suspected in the presence of typical symptoms and can be reinforced by imaging. Magnetic resonance imaging (MRI) is considered a second-line technique after ultrasound (US) and is often used as an additional examination method in complex cases. Transvaginal US is highly sensitive and specific for the differential diagnosis of endometriomas, but MRI is considered by some authors as the best imaging modality [5]. In a meta-analysis, the sensitivity and specificity of MRI for the diagnosis of endometriomas were 95% and 91%, respectively [6,7].

First-line medical treatment for endometriosis can also be used empirically and includes non-steroidal anti-inflammatory drugs (NSAIDs), other analgesics, combined contraceptive pills and traditional progestins [8].

The choice of treatment for endometriomas must be balanced, considering the risk of unintentional removal or destruction of ovarian follicles with surgical treatment, which can be characterised by a reduction in serum Anti-Müllerian hormone (AMH) levels and antral follicle count (AFC) on US after conservative ovarian surgery [3]. Surgical removal of endometriomas may worsen ovarian reserve by removing the healthy ovarian cortex or compromising blood flow to the ovary, particularly in the case of cystectomy or the use of bipolar energy [9]. Moreover, after surgical treatment of endometriomas, there is a high recurrence rate, which can vary between 12 and 30% after 2–5 years of follow-up [10].

For these reasons, it is crucial to individualize endometriosis treatment and to carefully balance the risks and benefits before choosing the treatment option, considering the age of the patient, the desire of pregnancy, the disease severity and symptoms. Then, all these aspects should be discussed with the patient.

There is clear evidence of the efficacy of combined oral contraceptives after surgery in terms of reducing the recurrence of endometriomas[10], [11], [12], [13], delaying the recurrence of the disease [14] and reducing the need for repetitive surgery [15].

Conversely, there are limited data on the effect of hormonal treatment on the size of endometriomas. The majority of the available trials have analysed size modification of endometriomas in patients under treatment with a single medicament, without considering a control group [16], [17], [18], [19], [20], or have compared the effects of two different medicaments [21], [22], [23], [24].

Therefore, there are no conclusive data on the effect of hormonal therapy on endometriomas or on the superiority of a specific medicament. Moreover, no intrinsic features of endometriomas are associated with their evolution. Accordingly, the objective of our study was to evaluate the evolution of endometriomas under medical treatment or without treatment.

This retrospective observational study was performed at the Gynaecological Center, Hôpital La Conception (Assistance Publique Hopitaux de Marseille). We included patients followed between January 2015 and April 2021, with at least one endometrioma diagnosed by MRI and at least one additional MRI exam after more than three months. Inclusion was done progressively during the follow-up consultations by one of the department's physicians. Patients were divided into those receiving medical treatment in the period between the MRIs and those without medical treatment. Medical treatments were assumed to be high-dose progestins, low-dose progestins or combined contraceptives. Exclusion criteria were surgical treatment or medically assisted procreation throughout the observational period.

We gathered the following information on patients: parity, body mass index, history of endometriosis surgery, presence of symptoms potentially related with endometriosis (dysmenorrhea, dyspareunia, chronic pelvic pain, pain on defecation, rectorragia, diarrhea, constipation, pain on urination), primary or secondary infertility, hormonal treatment, amenorrhea during medical treatment and its duration and age at the latest MRI exam.

MRIs were performed in the Radiology of Hôpital La Conception or in other centres, but with a second reading by a radiologist from our center. For every MRI exam, we collected data about the number of endometriomas in each ovary and the maximal diameter of each endometrioma. Specifically, we measured the maximal diameter of the endometriomas (mm).

The primary objective of our study was to evaluate the evolution of endometriomas under hormonal treatment or without treatment based on a comparison of MRI data. The primary evaluation criterion was the diameter of the endometrioma, while the secondary evaluation criterion was the number of endometriomas for each patient observed. The secondary study objective was to determine whether different categories of hormonal treatment have different impacts on the evolution of endometriomas.

The study was declared and registered with the General Data Protection Regulation (GDPR), registered in the GDPR/AP-HM register under the number 2021–55 and validated by the Aix-Marseille University Ethics Committee (N° 2021–12–16–03)

Statistical analysis was performed using IBM SPSS Statistics 20 software (SPSS Inc., Chicago, IL, USA). Statistical significance was accepted at P < 0.05. First, a descriptive analysis of the demographic and clinical characteristics of the sample was conducted. Qualitative variables are expressed as numbers and percentages and quantitative variables as mean value +/- standard deviation.

A comparative analysis was done using Pearson Chi-2 test or Fisher exact test for qualitative variables and the Student's T-test, Mann–Whitney test, analysis of variance (ANOVA) or Kruskal–Wallis test for quantitative variables. A repeated measures variance analysis (general linear model) was also performed, with a quantitative dependent variable and a qualitative inter-subject factor. Regarding the evolution of the diameter of endometriomas, the principal objective in this study, weighted data were considered.