The retrospective study included women undergoing OH during the diagnostics and treatment of infertility in the years 2020–2021. Prior to data collection, the study received a positive opinion from the Bioethics Committee of the Jagiellonian University (no. 1072.6120.100.2021). The dataset gathered during the study has been shared in the Harvard Dataverse [8]. The following inclusion criteria were employed: i) age between 18 and 45 years, ii) intrauterine pathology accompanying infertility amenable to treatment through OH, iii) evaluation for CE or overlooked intrauterine pathologies in women under 35 with unsuccessful conception attempts lasting at least 12 months, or 6 months for those 35 and older, following the exclusion of known infertility factors, iv) evaluation for CE or overlooked intrauterine pathologies in cases of unsuccessful conception attempts after the use of therapeutic approach directed at the identified infertility factor. No exclusion criteria were applied. The procedure was preceded by qualification by a specialist in obstetrics and gynecology [9], which included a medical interview and a gynecological examination comprising speculum inspection of the vagina, bimanual palpation, and an ultrasound evaluation of the reproductive organs, including two- and three-dimensional imaging, and, in selected cases, saline infusion sonohysterography [10]. A valid cervical cytology, normal vaginal biocenosis, and negative tests for Chlamydia, Mycoplasma, and Ureaplasma were prerequisites for the procedure. The potential cause of infertility and indications for the procedure were determined based on data from the medical interview and additional tests conducted prior to the procedure. The preoperative assessment required evaluating ovulatory pattern, ovarian reserve, tubal patency, and partner's semen analysis. Idiopathic infertility was diagnosed after excluding obvious causes of subfertility. Endometriosis and adenomyosis were confirmed based on the surgical protocol, histopathological examination, or pelvic imaging [11, 12]. A history of two spontaneous miscarriages was considered as recurrent miscarriages [13]. The tubal factor was identified as the obstruction of at least one fallopian tube, confirmed through hysterosalpingography, hysterosalpingo-foam sonography, or laparoscopic chromopertubation [14]. The male factor was diagnosed based on at least one semen analysis [15]. In cases of evident intrauterine pathology without apparent subfertility factors, additional diagnostics were not strictly required for spontaneous conception attempts, assuming a favorable prognosis post-lesion removal [16]. To facilitate further calculations, preoperative diagnoses were classified into four groups, commencing with idiopathic infertility. Uterine polyps formed a separate group due to their high prevalence, while submucosal fibroids, uterine septum, cesarean scar defects, and intrauterine adhesions were collectively categorized as uterine factors. The final group encompassed other infertility-related conditions. The procedure was conducted in the follicular phase of the menstrual cycle, after obtaining the woman's informed written consent. Detected intrauterine lesions were removed, and if absent, endometrial biopsy was performed, ensuring that a tissue sample was collected at each procedure. The tissue material was promptly immersed in 10% neutral-buffered formalin and subsequently sent for histopathological examination at the Department of Pathomorphology, Jagiellonian University, where it underwent further processing and fixation in liquid paraffin. Following this, the paraffin sections were stained with hematoxylin and eosin. In select cases, immunostaining for CD138 on plasma cells was performed for diagnosing CE, using murine monoclonal antibodies (Cell Marque, Merck KGaA, Darmstadt, Germany). CE was diagnosed based on the presence of at least one plasma cell in ten high-power fields [17]. Confirmed CE cases were treated with empirical oral ofloxacin at a dosage of 400 mg daily for 10 days. Data on obstetric outcomes in the postoperative period, specifically biochemical and clinical pregnancies, live births, and the duration of attempts to achieve pregnancy, were collected through follow-up contact up to 24 months following the procedure. A biochemical pregnancy was defined as the detection of serum concentration of β-human chorionic gonadotropin prior to a miscarriage that occurred before the pregnancy could be visualized on ultrasound. A clinical pregnancy was defined as the loss of a pregnancy that was visible on ultrasound before 24 weeks of gestation. A live birth was defined as the delivery of a viable infant after 24 weeks of gestation.
Hysteroscopy
Hysteroscopy was conducted via a vaginoscopic approach with a Karl Storz Bettocchi (Karl Storz SE & Co. KG, Tuttlingen, Germany) rigid hysteroscope, featuring a 5 mm outer sheath, an operating channel for 5 Fr instruments, and a 2.9 mm telescope equipped with a 30° angled lens. The uterine cavity was distended with a 0.9% NaCl solution using gravity inflow, assisted by a manometer cuff inflated to a maximum pressure of 120 mmHg, ensuring comprehensive visualization of the endometrial cavity. The clinical diagnosis of a uterine polyp was made upon visualizing a small, smooth, rounded pinkish growth protruding from the uterine lining. A submucous fibroid presented as a protruding, smooth or lobulated mass within the uterine cavity. Clinical features of CE included endometrial micropolyps, diffuse or focal endometrial hyperemia, fibrin deposits, and endometrial edema [18]. Intrauterine adhesions appeared as whitish, thin fibrous bands of scar tissue connecting various parts of the uterine wall or obstructing the cavity. A uterine septum was diagnosed as a fibrous or muscular band of tissue dividing the uterine cavity. Cesarean scar defect appeared as a concave area with fluid accumulation within the lower uterine segment. Submucosal adenomyosis was diagnosed by the presence of endometrial bulging or cysts discharging brownish contents. Retained conception products were identified as brownish-gray, irregularly shaped intracavitary debris, resembling fibrin deposits, temporally correlated with a previous miscarriage. Endometrial elevation referred to the mucosal prominence on the posterior uterine wall. Abnormal proximal fallopian tubes appearance included irregular outlets obscured by adhesions or fluid accumulation.
Clinical diagnoses derived from hysteroscopic visualization of the uterine cavity were classified into four categories to facilitate subsequent calculations: uterine polyp, clinical features of CE without other macroscopic intrauterine pathologies, uterine factors (septum and intrauterine adhesions), and a collective of others. Lesion resection was conducted with the use of a bipolar needle electrode, blunt scissors, and grasping forceps, which were also used for collecting the endometrial biopsy. The procedure was commenced after the administration of either an infiltration anesthesia or a paracervical block. The infiltration anesthesia was administered by injecting 1% lidocaine solution in a total volume of 20 ml into the cervix at 3 and 6 o'clock. The paracervical block was achieved by injecting a similarly prepared solution through a 19 mm anesthetic cannula (Cerviblok, RI.MOS., Italy) at the 2, 4, 8, and 10 o'clock positions within the lateral fornix of the vagina. The procedure was performed by a specialist in obstetrics and gynecology or a resident physician under the supervision of a specialist.
Statistical analysis
The analysis of quantitative variables involved computing descriptive statistics, including the mean, standard deviation, median, quartiles, and the minimum and maximum values. For qualitative variables, absolute frequencies and percentages of all possible values were calculated. To compare qualitative variable values across groups, the chi-square test (with Yates' correction applied for 2x2 tables) or Fisher's exact test was utilized when the assumptions of the chi-square test regarding expected frequencies were not satisfied. For the comparison of quantitative variable values among groups, the Kruskal-Wallis test was employed. If statistically significant differences were identified between groups, the Dunn post-hoc test was applied to ascertain specific group differences. Multivariate analysis was conducted to explore the influence of potential predictors on the quantitative variable using linear regression, with results presented as regression parameters accompanied by 95% confidence intervals. Similarly, multivariate analysis for potential predictors affecting the dichotomous variable was conducted using logistic regression, with results reported as odds ratios and 95% confidence intervals. All analyses were carried out using R software, version 4.4.1 [19].