{"paper_id":"cc39fd39-e154-41c3-86fa-8bc5e9b9ed5b","body_text":"Vol.:(0123456789)\nArchives of Gynecology and Obstetrics (2025) 312:505–513 \nhttps://doi.org/10.1007/s00404-025-08008-4\nRESEARCH\nComparative analysis of hysterosalpingography \nand hysterosalpingo‑contrast sonography for assessing tubal patency \nin women with endometriosis‑related infertility: a propensity \nscore‑matched study\nWencai Tian1,4 · Baoli Xie1 · Feng Liang2 · Yingqin Huang3 · Qianwen Hu1 · Jiaxin Yu1 · Aiping Qin1\nReceived: 9 January 2025 / Accepted: 10 March 2025 / Published online: 1 May 2025 \n© The Author(s) 2025\nAbstract\nPurpose This study aimed to compare the clinical pregnancy rates of hysterosalpingography (HSG) with water-based \ncontrast and hysterosalpingo-contrast sonography (HyCoSy) to investigate their impact on fertility outcomes in women with \nendometriosis-related infertility.\nMethods This retrospective study included 296 women with endometriosis-related infertility, aged 18–39, undergoing HSG \nor HyCoSy at the First Affiliated Hospital of Guangxi Medical University (January 2020–June 2022). Propensity score \nmatching was applied to balance demographic and clinical variables. Clinical pregnancy outcomes were compared between \nthe groups.\nResults The HSG group showed a higher clinical pregnancy rates compared to the HyCoSy group. Logistic regression \ndemonstrated an odds ratio (OR) of 0.41 (95% CI 0.20–0.82, p < 0.001) for clinical pregnancy in the HyCoSy group compared \nto HSG. After propensity score matching, the OR was 0.37 (95% CI 0.18–0.75). Similar results were observed across \nmultivariable and IPTW analyses, suggesting that HSG may be associated with better clinical pregnancy outcomes than \nHyCoSy.\nConclusion Hysterosalpingography (HSG) with water-based contrast appears to be associated with higher clinical pregnancy \nrates compared to HyCoSy in women with endometriosis-related infertility. This suggests that HSG may be a more favorable \ndiagnostic method for optimizing fertility outcomes, though further research is needed to confirm these findings.\nKeywords Hysterosalpingography (HSG) · Hysterosalpingo-contrast sonography (HyCoSy) · Fertility outcomes · \nEndometriosis-related infertility · Propensity score matching (PSM)\n * Aiping Qin \n qinaiping@gxmu.edu.cn\n1 Center of Reproductive Medicine, The First Affiliated \nHospital of Guangxi Medical University, Nanning, China\n2 Gynecology Department, The Reproductive Hospital \nof Guangxi Zhuang Autonomous Region, Nanning, China\n3 Center for Reproductive Medicine, Maternal and Child \nHealth Hospital in Guangxi, Guangxi 530021, China\n4 Department of Obstetrics and Gynecology, The Affiliated \nChangsha Central Hospital, Hengyang Medical School, \nUniversity of South China, Changsha, China\nWhat does this study add to the clinical work \nThis study suggests that hysterosalpingography \n(HSG) with water-based contrast appears to \nimprove clinical pregnancy rates compared to \nhysterosalpingo-contrast sonography (HyCoSy) \nin women with endometriosis-related infertility. It \nhighlights the importance of selecting diagnostic \nmethods that not only evaluate tubal patency but \nalso help improve fertility outcomes.\n\n506 Archives of Gynecology and Obstetrics (2025) 312:505–513\nIntroduction\nTubal pathology represents a significant contributor to \nfemale infertility, with a prevalence ranging from 11 to \n30%, often attributed to factors such as infections, prior \nsurgical interventions, or conditions like endometriosis \n[1, 2]. Consequently, evaluating the fallopian tubes is a \nfundamental aspect of the fertility assessment process.\nHysterosalpingography (HSG) remains the primary \nchoice for assessing tubal patency during fertility \ninvestigations [3 ]. This technique is notable for its \npotential to improve female fertility. Initially employed \nfor infertility screening and diagnosis [3 ], the expanded \nutilization of HSG among infertility patients underscores \nits efficacy in enhancing fertility outcomes [ 4–6]. A \nretrospective cohort study found no significant difference \nin IUI outcomes between women who underwent HSG \nand those who had laparoscopy, suggesting both tests \nhave similar effects on treatment success [7 ]. However, \nthe impact of HSG on fertility may vary based on the type \nof contrast agent utilized [8 ]. Research indicates that oil-\nbased contrast agents may have an advantage over their \nwater-based counterparts in enhancing conception in \ninfertile individuals [9 ].\nHysterosalpingo-contrast sonography (HyCoSy) has \nemerged as a comparably effective alternative technique. \nIt uses transvaginal ultrasonography combined with the \ninjection of an echogenic contrast agent into the fallopian \ntubes and uterus to simultaneously visualize the ovaries. \nStudies have demonstrated that its accuracy in assessing \ntubal patency parallels that of HSG [10, 11].\nEndometriosis is a challenging and complex disorder \naffecting about 10% of women of reproductive age \nworldwide [12]. Its prevalence highlights its significance \nas a contributing factor in intractable infertility cases, \naccounting for an estimated 30–50% of such instances \n[13]. Mental health issues, such as depression and anxiety, \nare also more common among infertile women, especially \nin Asia [14]. Our team’s earlier research [15] revealed \nthat HSG using oil-based contrast agents may be more \neffective than water-based agents in increasing fertility \nin individuals struggling with endometriosis-related \ninfertility. Currently, both HSG and HyCoSy are widely \nutilized hysterotubal ultrasonography techniques in clinical \nsettings. However, the comparative efficacy of HSG with \nwater-based contrast agents and HyCoSy in improving \nfertility outcomes in patients with endometriosis-related \ninfertility remains unexplored.\nTo address this question, we conducted a retrospective \nstudy comparing the impact of HyCoSy and HSG using \niodinated water contrast agents on the success of pregnancies \nin patients with endometriosis-related infertility.\nMaterials and methods\nStudy design\nThis study included patients who, between January 2020 \nand June 2022, received either HSG or HyCoSy at the First \nAffiliated Hospital of Guangxi Medical University due to \ninfertility concurrent with endometriosis. The research \nprotocol was approved by the Ethics Committee of the First \nAffiliated Hospital of Guangxi Medical University and \nfollowed the principles outlined in the STROBE Statement. \nThe inclusion criteria were: (1) female patients aged 18–39 \nwith endometriosis; (2) regular menstrual cycles; and (3) \nhaving undergone HSG. Exclusion criteria included: (1) a \ntotal motile sperm count after sperm wash of less than 3 \nmillion sperm per milliliter in the male partner (or less than \n1 million sperm per milliliter in the absence of post-wash \nanalysis); (2) uncontrolled endocrine disorders known to \nreduce natural pregnancy chances, such as the acute phase of \nsystemic lupus erythematosus; (3) incomplete data. Baseline \ncharacteristics of the patients included age, comorbidities, \nBMI, length of infertility, smoking status, surgical history, \ntotal number of prior live births, miscarriages, and treatment \nafter hysterosalpingography procedures.\nDefinition of endometriosis\nEndometriosis was defined using one or more of the \nfollowing criteria: (1) surgical confirmation via laparoscopic \nor transabdominal operations; (2) ultrasonography \nidentification of an endometrioma using transvaginal \nultrasonography, which typically appears as a single or \nmultilocular cyst with ground glass echogenicity [16]; \n(3) clinical signs of possible endometriosis, defined by \nthe presence of three out of five characteristics: infertility, \ndysmenorrhea, dyspareunia, discomfort in the sacral \nligament, or CA125 values higher than 15 mIU/mL [17].\nHSG procedures\nWithin the HSG group, the contrast medium utilized \nwas either Iohexol 300 (General Electric Pharmaceutical \n(Shanghai) Co., Ltd., Shanghai, China), Iopromide 300 \n(Bayer Healthcare Co., Ltd., Beijing, China), or Ioversol \n320 (Jiangsu Hengrui Medicine Co., Ltd., Lianyungang, \nJiangsu Province, China). The HSG procedure was \nperformed as follows: all patients had empty bladders. A \nlidocaine injection (0.1 g) was administered to the cervix \nbefore endotracheal intubation. Via the vagina, a balloon \ncatheter (model 12B; Zhanjiang Star Enterprise Co., Ltd., \nZhanjiang, China) was placed into the cervix and secured. \nOnce the uterine cavity was completely filled or the contrast \n\n507Archives of Gynecology and Obstetrics (2025) 312:505–513 \nhad entered the pelvic cavity, the prewarmed contrast was \ngradually injected. Dynamically displayed on the TV screen, \nthe process of the contrast entering the uterus and fallopian \ntubes was documented, with the relevant photos taken. \nPressure during the injection was controlled based on the \nclinician’s experience. In case of intravasation or interstitial \nreflux, patients were instructed to discontinue the contrast \nagent injection right away, get out of bed, walk, and inhale \noxygen.\nHyCoSy performance\nThe preparation before HyCoSy was the same as that for \nthe HSG group. SonoVue® (sulfur hexafluoride; Shanghai \nBracco Sine Pharmaceutical Corp., Ltd., Shanghai, China) \nwas used to evaluate tubal patency, with a maximum of \n15 ml. The sight of a continuous flow of contrast medium \nthroughout the fallopian tube or observable contrast spill \nfrom the fimbrial end over the ovary was described as \nfallopian tube patency. Each tube was then categorized as \npatent, occluded, or uncertain.\nMain outcomes\nThe main outcome was the commencement of the last \nmenstrual cycle within a year after HSG, designated as \nthe key endpoint for clinical pregnancy. The existence of a \ngestational sac found by ultrasonography was indicative of \na clinical pregnancy.\nStatistical analysis\nFor continuous variables, descriptive statistics were used; \nresults were shown as mean/standard deviation (SD), \nmedian/interquartile range (IQR), and proportions (%) for \ncategorical variables. Logistic regression models were \nutilized to investigate the association between the HSG and \nHyCoSy groups across four models.\nModel 1 was a crude model, providing an unadjusted \nassessment of the relationship between the HSG and \nHyCoSy groups. Model 2 adjusted for sociodemographic \nfactors, including age, BMI, and smoking status. Model \n3 incorporated adjustments from Model 2 and additional \nvariables such as complications, miscarriage occurrences, \nand treatment received after hysterosalpingography \ntechniques. Model 4 further adjusted for the duration of \ninfertility, surgical history, and the total number of previous \npregnancies resulting in live births. These sequential \nadjustments aimed to comprehensively address potential \nconfounding factors and enhance the understanding of the \nassociation between the HSG and HyCoSy groups.\nTo reduce potential biases associated with confounding \nvariables, a matched propensity score (PSM) analysis was \ncarried out given the non-randomized research design. The \npropensity scores were estimated using multivariable logistic \nregression, incorporating factors such as age, complications, \nBMI, duration of infertility, smoking status, surgical history, \ntotal number of previous pregnancies resulting in live births, \nmiscarriage occurrences, and treatment received after \nHSG. Additional pertinent study parameters were added as \nindependent variables for PSM. Using a greedy-matching \nmethod, the logit-transformed propensity scores were used \nto match participants in a 1:1 ratio without replacement. \nA caliper width of 0.2 standard deviation was applied to \nensure closer matches between the participants. The balance \nof covariates between the HSG and HyCoSy groups was \nassessed using standardized differences, with a value of less \nthan 10% indicating satisfactory balance [18].\nSample size determination was based on available \ndata, with no a priori statistical power assessments \nconducted. Statistical analyses were performed using R \n3.3.2, a statistical software program developed by The R \nFoundation, Shanghai, China (accessed on 10 January \n2023), in conjunction with Free Statistics Software \n1.5 [19]. A descriptive analysis was conducted on all \nindividuals to comprehensively characterize the dataset. \nFor hypothesis testing, a two-tailed analysis was employed, \nand a significance level of 0.05 was utilized to determine \nstatistical significance. This widely accepted threshold \ndenotes a standard level of confidence in interpreting the \nresults and drawing meaningful conclusions from the \nanalyses conducted.\nResults\nStudy population\nBetween January 2020 and June 2022, 296 women \nwith endometriosis-related infertility underwent \nhysterosalpingography procedures at the First Affiliated \nHospital of Guangxi Medical University. Among them, \n173 patients received hysterosalpingo-contrast sonography \n(HyCoSy), and 123 patients received hysterosalpingography \n(HSG) with water-based contrast. Following propensity \nscore matching (PSM), 89 matched pairs were included in \nthe final analysis (Fig.  1).\nClinical characteristics\nThe demographic and baseline clinical characteristics of \npatients before and after PSM are summarized in Table  1. \nThe mean age of the patients was 31.2 years (SD 4.1) in \nthe HyCoSy group and 31.5 years (SD 3.9) in the HSG \ngroup. Other baseline characteristics such as BMI, duration \n\n508 Archives of Gynecology and Obstetrics (2025) 312:505–513\nof infertility, and smoking status were well balanced after \nPSM, with standardized differences below 10%.\nRelationship between various \nhysterosalpingography techniques and clinical \npregnancy outcomes\nTable  2 illustrates the relationships between different \nhysterosalpingography techniques and clinical pregnancy \noutcomes. Univariate analysis revealed potential \nassociations, suggesting correlations between clinical \npregnancy outcomes and variables such as miscarriage \noccurrences, treatment received after hysterosalpingography \ntechniques, and the specific technique used for \nhysterosalpingography.\nFurther research with multivariate analyses may be \nuseful to explore the potential contributions of each factor to \nclinical pregnancy outcomes. In comparison to individuals \nin the HSG group, and after adjustments for age, BMI, \nsmoking status, complications, miscarriage occurrences, \ntreatment received after hysterosalpingography techniques, \nduration of infertility, surgical history, and total number of \nprevious pregnancies resulting in live births, the adjusted \nodds ratio (OR) values for various hysterosalpingography \ntechniques with clinical pregnancy outcomes were 0.41 \n(95% CI 0.20–0.82, p = 0.012) (Table 3).\nAssociations between various \nhysterosalpingography techniques and clinical \npregnancy outcomes: crude, multivariable, \nand propensity score analyses (HyCoSy group vs. \nHSG group)\nThe HyCoSy group showed an unadjusted model Odds Ratio \n(OR) of 0.49 (95% CI 0.29–0.85, p  = 0.011) compared to \nthe HSG group. Multivariable analysis revealed an OR of \n0.41 (95% CI 0.20–0.82, p  = 0.012). The OR using inverse \nprobability of treatment weighting (IPTW) regression \nanalysis was 0.48 (95% CI 0.27–0.85, p  = 0.013). In \naddition, the OR was 0.37 (95% CI 0.18–0.75, p  = 0.006) \nwhen propensity score matching was used. These statistical \nresults indicate that the HSG group had a higher clinical \npregnancy rate compared to the HyCoSy group, although \nfurther research is needed to confirm these findings (refer \nto Table  4 for details).\nFig. 1  Flowchart of study recruitment and inclusion/exclusion criteria\n\n509Archives of Gynecology and Obstetrics (2025) 312:505–513 \nDiscussion\nIn this retrospective study, we observed a significant \ndifference in clinical pregnancy rates 1  year after \nsurgery for women with endometriosis-related infertility \nundergoing HSG with water contrast versus those who \nunderwent HyCoSy. Specifically, the clinical pregnancy \nrate was higher in the HSG group compared to the \nHyCoSy group. These findings suggest a potential benefit \nof HSG in improving fertility outcomes in patients with \nendometriosis-related infertility, but further studies are \nneeded to confirm these results.\nHyCoSy has gained popularity as a method for examining \ntubal patency due to its avoidance of radiation exposure \nand lower invasiveness, while maintaining sensitivity and \nspecificity comparable to hysterosalpingography [2 ]. In \naddition to its diagnostic role, new clinical evidence and \nearlier research suggest HyCoSy’s potential therapeutic \nuse for infertility; some infertile women undergoing the \nprocedure were able to conceive spontaneously [20]. \nSpecifically, HyCoSy has been associated with improved \nspontaneous pregnancy rates within 6 months following \nthe procedure. Emilio et al. [21] proposed that HyCoSy \ncould trigger ovulation, while Gao et al. [22] suggested that \nthe liquid flow during HyCoSy could disintegrate small \nTable 1  Patient characteristics before and after propensity score matching\nHSG hysterosalpingography, HyCoSy hysterosalpingo-contrast sonography, BMI Body Mass Index\nParameter All patients (n = 296) PS-matched pairs (n = 178)\nHSG group HyCoSy group SMD HSG group HyCoSy group SMD\nn = 123 n = 173 n = 89 n = 89\nAge (year), mean (SD) 31.24 (3.77) 32.30 (3.72) 0.284 31.54 (3.77) 31.65 (3.78) 0.030\nComplication (%) 0.255 0.695\n No 76 (61.8) 107 (61.8) 54 (60.7) 78 (87.6)\n Intrauterine adhesion 5 (4.1) 4 (2.3) 2 (2.2) 2 (2.2)\n Thyroid disease 5 (4.1) 14 (8.1) 5 (5.6) 1 (1.1)\n Chronic endometritis 5 (4.1) 3 (1.7) 5 (5.6) 2 (2.2)\n Adenomyosis 3 (2.4) 7 (4.0) 3 (3.4) 2 (2.2)\n Others 29 (23.6) 38 (22.0) 20 (22.5) 4 (4.5)\nBMI (kg/m2), mean (SD) 20.85 (2.66) 20.82 (2.33) 0.012 20.92 (2.33) 20.63 (2.37) 0.124\nDuration of infertility (year), mean (SD) 2.86 (2.29) 2.98 (2.39) 0.051 3.03 (2.43) 2.93 (2.16) 0.041\nSurgical history (%) 0.477 0.307\n No 115 (93.5) 151 (87.3) 85 (95.5) 89 (100.0)\n Myoma or polyp resection or cystectomy 1 (0.8) 13 (7.5) 1 (1.1) 0 (0.0)\n Tubal surgery 3 (2.4) 6 (3.5) 3 (3.4) 0 (0.0)\n Cesarean section 4 (3.3) 0 (0.0) 0 (0.0) 0 (0.0)\n Others 0 (0.0) 3 (1.7) 0 (0.0) 0 (0.0)\n Smoking status, n (%) 3 (2.4) 4 (2.3) 0.008 1 (1.1) 1 (1.1) < 0.001\nTotal number of previous pregnancies resulting in live births (%) 0.169 0.228\n 0 99 (80.5) 150 (86.7) 75 (84.3) 79 (88.8)\n 1 22 (17.9) 21 (12.1) 12 (13.5) 10 (11.2)\n 2 2 (1.6) 2 (1.2) 2 (2.2) 0 (0.0)\nMiscarriage times (%) 0.151 0.662\n 0 96 (78.0) 141 (81.5) 73 (82.0) 89 (100.0)\n 1 19 (15.4) 23 (13.3) 10 (11.2) 0 (0.0)\n 2 8 (6.5) 8 (4.6) 6 (6.7) 0 (0.0)\n 3 0 (0.0) 1 (0.6) 0 (0.0) 0 (0.0)\nTreatment after hysterosalpingography techniques (%) 0.891 0.251\n Expectant management 69 (56.1) 99 (57.2) 65 (73.0) 56 (62.9)\n IUI 9 (7.3) 27 (15.6) 9 (10.1) 16 (18.0)\n IVF/ICSI 15 (12.2) 47 (27.2) 15 (16.9) 17 (19.1)\n Laparoscopy and/or hysteroscopy 30 (24.4) 0 (0.0) 0 (0.0) 0 (0.0)\n\n510 Archives of Gynecology and Obstetrics (2025) 312:505–513\nadhesions in the fallopian tubes, increasing the likelihood \nof spontaneous pregnancy. However, the precise mechanism \nleading to the increased conception rates after HyCoSy \nremains unclear. Another study [23] suggested that HyCoSy \nexamination might improve endometrial perfusion and \nhave a therapeutic impact on natural conception in infertile \nwomen. Nonetheless, conflicting evidence exists, as some \nstudies have failed to confirm an enhanced pregnancy rate \nafter HyCoSy [24].\nHSG has traditionally been the preferred method for tubal \npatency testing, primarily due to its capability to visualize \nand assess tubal patency using an iodinated contrast \nmedium [25]. The efficacy of HSG in enhancing fertility \noutcomes is believed to involve the mechanical flushing \naction of the contrast medium, which aids in removing \ntubal debris, mucus plugs, and other obstructive substances \n[26]. Beyond this mechanical action, the anti-inflammatory \nand antimicrobial properties of iodine, the component used \nTable 2  Association of covariates and clinical pregnancy\nHSG hysterosalpingography, HyCoSy hysterosalpingo-contrast sonography, BMI Body Mass Index, OR odds ratio, CI confidence interval, Ref  \nreference\nVariable OR_95CI p_value Variable OR_95CI p_value\nAge (year), n (%) Total number of previous pregnancies resulting in live \nbirths\n < 35 1 (reference)  0 1 (reference)\n > 35 1.84 (0.95–3.59) 0.073  1 0.97 (0.47–2) 0.935\nBMI (kg/m2), n (%)  2 1.13 (0.12–11.02) 0.918\n < 25 1 (reference) Miscarriage times\n > 25 5.57 (0.72–43.08) 0.1  0 1 (reference)\nDuration of infertility (year)  1  0.5 (0.25–0.99) 0.045\n ≤ 3 1 (reference)  2 1.47 (0.4–5.33) 0.559\n > 3 1.17 (0.7–1.97) 0.545 Treatment after hysterosalpingography techniques (%)\nComplication (%)  Expectant management 1 (reference)\n No 1 (reference)  IUI 0.38 (0.18–0.83) 0.015\n Intrauterine adhesion 1.47 (0.29–7.28) 0.641  IVF/ICSI 0.36 (0.19–0.68) 0.002\n Thyroid disease 3.56 (0.79–15.93) 0.097  Laparoscopy and/or Hysteroscopy 0.57 (0.24–1.36) 0.205\n Chronic endometritis 1.26 (0.25–6.42) 0.784 Different techniques of hysterosalpingography\n Adenomyosis 1.67 (0.34–8.14) 0.523  HSG group 1 (reference)\n Others 1.06 (0.57–1.96) 0.859  HyCoSy group 0.49 (0.29–0.85) 0.011\nSurgical history (%) Smoking status, n (%)\n No  No 1 (reference)\n Myoma or polyp resection \nor cystectomy\n1.36 (0.37–5.02) 0.644  Yes 0.49 (0.11–2.25) 0.361\n Tubal surgery 0.74 (0.18–3.05) 0.679\n Cesarean section 1.11 (0.11–10.88) 0.926\n Others 0.19 (0.02–2.08) 0.172\nTable 3  Correlation of various hysterosalpingography techniques with clinical pregnancy outcomes\nModel 1 was a crude model\nModel 2 was adjusted for sociodemographic (age, BMI and smoking status).Model 3 was adjusted for model 2 plus complication, miscarriage \ntimes and treatment after hysterosalpingography techniques.Model 4 was adjusted for model 3 plus duration of infertility, surgical history and \ntotal number of previous pregnancies resulting in live births\nHSG hysterosalpingography, HyCoSy hysterosalpingo-contrast sonography, OR odds ratio, CI confidence interval, Ref  reference, BMI Body \nMass Index\nVariable Total OR (95% CI)\nModel 1 p_value Model 2 p_value Model 3 p_value Model 4 p_value\nHSG group 123 1 (Ref) 1 (Ref) 1 (Ref) 1 (Ref)\nHyCoSy group 173 0.49 (0.29–0.85) 0.011 0.47 (0.27–0.81) 0.007 0.43 (0.22–0.84) 0.014 0.41 (0.2–0.82) 0.012\n\n511Archives of Gynecology and Obstetrics (2025) 312:505–513 \nin the contrast medium, also contribute to its potential \ntherapeutic effects. Research suggests [27] that women with \nendometriosis often exhibit elevated levels of localized and \nsystemic inflammation. Earlier studies have shown that \nendometriosis is known to have a strong immunosuppressive \nmicroenvironment [28]. Therefore, the anti-inflammatory \neffects of iodine may play a significant role in improving \nfertility outcomes in patients with endometriosis-related \ninfertility.\nSeveral mechanisms support this notion. First, iodine is \ncrucial for normal thyroid hormone synthesis and release, \nindirectly promoting ovulation. Normal thyroid-stimulating \nhormone (TSH) levels support follicular growth in oocytes, \nas TSH acts on the follicle-stimulating hormone (FSH) \nreceptors, enhancing pre-antral follicular growth induced \nby FSH [29]. In addition, iodine has a direct effect on the \novaries [30]. Secondly, Johnson et al. [9]  demonstrated that \nLipiodol, an iodine-based contrast agent, induced alterations \nin the endometrial dendritic cell phenotype in rats. This \nimplies that iodine might be responsible for immunological \nchanges in the endometrium, thereby enhancing embryo \nimplantation. Ahmed FO et al. reported improved fertility \nin cows with unexplained infertility following uterine \ninstillation of Lugol’s iodine, suggesting that infertility \nmight be secondary to subclinical endometritis, and the \nbactericidal action of iodine could help restore the damaged \nendometrium. Moreover, some studies [31–33] suggest that \niodide can modulate the proportion of T helper 17 (Th17) \ncells, regulatory T cells (Treg), and natural killer (NK) \ncells, exerting immune activity. This suggests that iodine \nmay change the immune environment and thus affect embryo \nimplantation.\nThese findings suggest that the mechanical flushing \naction and inflammatory response elicited by the iodinated \ncontrast medium during HSG may contribute to its ability to \nenhance tubal patency and promote successful conception in \nthis patient population. In contrast, HyCoSy, which utilizes \nan echogenic contrast medium, lacks the anti-inflammatory \nproperties inherent to iodine. While HyCoSy may offer \nadvantages in terms of visualization and patient comfort, its \ninability to address the underlying inflammatory component of \nendometriosis may limit its effectiveness in improving fertility \noutcomes in this patient population. Therefore, the unique \nanti-inflammatory properties of iodine-based contrast agents \nused in HSG may play a crucial role in enhancing fertility \noutcomes, particularly in patients with endometriosis-related \ninfertility.\nIn this study, no severe allergic reactions were observed \nwith either oil-based or water-based contrast agents during \nhysterosalpingography (HSG). While iodine-based contrast \nagents used in water-based HSG can potentially cause allergic \nreactions, our results showed no such events. This aligns with \nfindings from two well-conducted randomized controlled trials \n(RCTs). The 2017 RCT [4] reported no allergic reactions in \neither the oil-based or water-based contrast groups during \nHSG. Similarly, the 2022 RCT [34] conducted in China also \nfound no allergic events in either group. These well-designed \nRCTs, together with our findings, suggest that both oil-based \nand water-based contrast agents are safe for use in HSG. \nFurthermore, the retrospective cohort study [35] also supports \nthese results, although the RCTs provide more robust evidence \nfor their safety.\nThe strengths of our study include its rigorous methodology, \nincluding propensity score matching to account for potential \nconfounders, and the inclusion of a substantial number of \npatients. However, several limitations warrant consideration. \nFirst, the retrospective design introduces the potential for \nselection bias and limits the generalizability of the results. \nSecond, reliance on medical records for data collection may \nhave introduced information bias, as the accuracy of the data \ndepends on the quality of the records. In addition, while the \nsample size is relatively large, it may not have been sufficient \nto detect small differences in pregnancy outcomes between \nthe two groups. Finally, the study did not include long-term \nfollow-up on pregnancy outcomes, which is important for \nevaluating the effectiveness of the diagnostic methods. Further \nlarge-scale, prospective studies with longer follow-up are \nneeded to confirm these findings.\nConclusion\nIn summary, this research offers valuable insights \ninto the relative efficacy of HyCoSy and HSG in \nenhancing reproductive outcomes for individuals with \nTable 4  HyCoSy versus HSG: clinical pregnancy comparison\nHSG hysterosalpingography, HyCoSy hysterosalpingo-contrast \nsonography\na Shown is the odds ratio from the multivariable logistic model, with \nadjusted for all covariates in table (age, complication, BMI, duration \nof infertility, smoking status, surgical history, total number of \nprevious, pregnancies resulting in live births, miscarriage times and \ntreatment after different hysterosalpingography techniques)\nb Shown is the primary analysis with multivariable logistic model \nwith the same covariates with inverse probability weighting \naccording propensity score\nc Shown is the odds ratio from the multivariable logistic model \nwith the same strata and covariates with matching according to the \npropensity score. The analysis included 89 patients (89 in the HSG \ngroup and 89 in the HyCoSy group)\nAnalysis OR_95CI p_value\nCrude analysis 0.49 (0.29–0.85) 0.011\nMultivariable  analysisa 0.41 (0.2–0.82) 0.012\nWeighted  IPTWb 0.48 (0.27–0.85) 0.013\nPropensity score  matchedc 0.37 (0.18–0.75) 0.006\n\n512 Archives of Gynecology and Obstetrics (2025) 312:505–513\nendometriosis-related infertility. Despite the limitations, \nour findings suggest a possible benefit of HSG with water \ncontrast in achieving clinical pregnancy in this patient \npopulation. Larger sample sizes and more prospective \nstudies are needed to validate these results and explore the \nunderlying factors causing the observed variations in fertility \noutcomes between the two approaches. These efforts will \nfurther inform clinical practice and enhance the management \nof infertility in patients with endometriosis.\nAcknowledgements Thank you to the 'Clinical Scientist' team for their \nsupport to us.\nAuthor contributions W.T.: project development, data collection, \nstatistical analyses, manuscript drafting. A.Q.: project development, \nmanuscript drafting, critical revision. Y.H., F.L., J.Y., B.X.: data \ncollection, data interpretation. Q.H.: statistical analyses, data \ninterpretation. All the authors: data interpretation, critical revision, \napproval of the final version of the manuscript.\nFunding This study received no funding.\nData availability The data underlying this article will be shared on \nreasonable request to the corresponding author.\nDeclarations \nConflict of interest The authors declare that they have no conflicts of \ninterest.\nEthics approval This study was approved by the Ethics Committee of \nthe First Affiliated Hospital of Guangxi Medical University. As this is a \nretrospective cohort study, written informed consent was not required.\nOpen Access  This article is licensed under a Creative Commons \nAttribution-NonCommercial-NoDerivatives 4.0 International \nLicense, which permits any non-commercial use, sharing, distribution \nand reproduction in any medium or format, as long as you give \nappropriate credit to the original author(s) and the source, provide a \nlink to the Creative Commons licence, and indicate if you modified \nthe licensed material. You do not have permission under this licence \nto share adapted material derived from this article or parts of it. The \nimages or other third party material in this article are included in the \narticle’s Creative Commons licence, unless indicated otherwise in a \ncredit line to the material. If material is not included in the article’s \nCreative Commons licence and your intended use is not permitted by \nstatutory regulation or exceeds the permitted use, you will need to \nobtain permission directly from the copyright holder. To view a copy \nof this licence, visit http:// creat iveco mmons. org/ licen ses/ by- nc- nd/4. 0/.\nReferences\n 1. Snick HK, Snick TS, Evers JL, Collins JA (1997) The spontaneous \npregnancy prognosis in untreated subfertile couples: the \nWalcheren primary care study. Hum Reprod 12(7):1582–1588\n 2. Farquhar CM, Bhattacharya S, Repping S, Mastenbroek S, \nKamath MS, Marjoribanks J, Boivin J (2019) Female subfertility. \nNat Rev Dis Primers 5(1):7\n 3. Infertility Workup for the Women’s Health Specialist (2019) \nACOG Committee opinion summary, number 781. Obstet \nGynecol 133(6):1294–1295\n 4. Dreyer K, van Rijswijk J, Mijatovic V, Goddijn M, Verhoeve \nHR, van Rooij IAJ, Hoek A, Bourdrez P, Nap AW, Rijnsaardt-\nLukassen HGM et al (2017) Oil-based or water-based contrast \nfor hysterosalpingography in infertile women. N Engl J Med \n376(21):2043–2052\n 5. Fang F, Bai Y, Zhang Y, Faramand A (2018) Oil-based versus \nwater-based contrast for hysterosalpingography in infertile \nwomen: a systematic review and meta-analysis of randomized \ncontrolled trials. Fertil Steril 110(1):153-160.e153\n 6. van Rijswijk J, van Welie N, Dreyer K, Pham CT, Verhoeve HR, \nHoek A, de Bruin JP, Nap AW, van Hooff MHA, Goddijn M et al \n(2020) Tubal flushing with oil-based or water-based contrast at \nhysterosalpingography for infertility: long-term reproductive \noutcomes of a randomized trial. Fertil Steril 114:155–162\n 7. Joseph T, Karuppusami R, Kunjummen AT, Kamath MS (2023) \nImpact of tubal patency test selection on the live birth rate \nfollowing intrauterine insemination in couples with unexplained \ninfertility: a retrospective cohort study. Arch Gynecol Obstet \n308(2):621–629\n 8. Wang R, van Welie N, van Rijswijk J, Johnson NP, Norman \nRJ, Dreyer K, Mijatovic V, Mol BW (2019) Effectiveness on \nfertility outcome of tubal flushing with different contrast media: \nsystematic review and network meta-analysis. Ultrasound \nObstet Gynecol 54(2):172–181\n 9. Johnson NP, Bhattu S, Wagner A, Blake DA, Chamley LW \n(2005) Lipiodol alters murine uterine dendritic cell populations: \na potential mechanism for the fertility-enhancing effect of \nlipiodol. Fertil Steril 83(6):1814–1821\n 10. Saunders RD, Shwayder JM, Nakajima ST (2011) Current \nmethods of tubal patency assessment. Fertil Steril \n95(7):2171–2179\n 11. Lim SL, Jung JJ, Yu SL, Rajesh H (2015) A comparison of \nhysterosalpingo-foam sonography (HyFoSy) and hysterosalpingo-\ncontrast sonography with saline medium (HyCoSy) in the \nassessment of tubal patency. Eur J Obstet Gynecol Reprod Biol \n195:168–172\n 12. Giudice LC (2010) Clinical practice. Endometriosis. N Engl J \nMed 362(25):2389–2398\n 13. Hodgson RM, Lee HL, Wang R, Mol BW, Johnson N (2020) \nInterventions for endometriosis-related infertility: a systematic \nreview and network meta-analysis. Fertil Steril 113(2):374-382.\ne372\n 14. Salari N, Babajani F, Hosseinian-Far A, Hasheminezhad R, Abdoli \nN, Haydarisharaf P, Mohammadi M (2024) Global prevalence \nof major depressive disorder, generalized anxiety, stress, and \ndepression among infertile women: a systematic review and meta-\nanalysis. Arch Gynecol Obstet 309(5):1833–1846\n 15. Xie B, Huang Y, Hang F, Yu J, Hu Q, Li J, Qin A (2024) Impact \nof oil-based contrast agents in hysterosalpingography on fertility \noutcomes in endometriosis: a retrospective cohort study. Reprod \nBiol Endocrinol 22(1):19\n 16. Exacoustos C, Zupi E, Piccione E (2017) Ultrasound imaging for \novarian and deep infiltrating endometriosis. Semin Reprod Med \n35(1):5–24\n 17. Nisenblat V, Bossuyt PM, Shaikh R, Farquhar C, Jordan V, \nScheffers CS, Mol BW, Johnson N, Hull ML (2016) Blood \nbiomarkers for the non-invasive diagnosis of endometriosis. \nCochrane Database Syst Rev 2016(5):CD012179\n 18. Austin PC (2011) A tutorial and case study in propensity score \nanalysis: an application to estimating the effect of in-hospital \nsmoking cessation counseling on mortality. Multivariate Behav \nRes 46(1):119–151\n 19. Yang Q, Zheng J, Chen W, Chen X, Wen D, Chen W, Xiong X, \nZhang Z (2021) Association between preadmission metformin use \nand outcomes in intensive care unit patients with sepsis and type \n2 diabetes: a cohort study. Front Med (Lausanne) 8:640785\n\n513Archives of Gynecology and Obstetrics (2025) 312:505–513 \n 20. Liu Y, Zhang N, He Y, Shi J, Zhou M, Xu J, Liu H (2020) \nSpontaneous conception outcome in infertile women after \nfour-dimensional hysterosalpingo-contrast-sonography. BMC \nPregnancy Childbirth 20(1):638\n 21. Giugliano E, Cagnazzo E, Bazzan E, Patella A, Marci R (2012) \nHysterosalpingo-contrast sonography: is possible to quantify the \ntherapeutic effect of a diagnostic test? Clin Exp Reprod Med \n39(4):161–165\n 22. Chunyan G, Bin P, Ping Y, Yue Z, Yang X, Hongju T, Li S, Xi \nX (2018) Assessment of the influence on spontaneous pregnancy \nof hysterosalpingo-contrast sonography. Biomed Res Int \n2018:4901281\n 23. Li R, Qiu X, Chen XF, He M, Wang W, Qiao J, He J, Shi Q (2022) \nEffects of hysterosalpingo-contrast sonography examination \non endometrial receptivity among women with unexplained \ninfertility. Arch Gynecol Obstet 306(3):893–900\n 24. Lindborg L, Thorburn J, Bergh C, Strandell A (2009) Influence \nof HyCoSy on spontaneous pregnancy: a randomized controlled \ntrial. Hum Reprod 24(5):1075–1079\n 25. Swart P, Mol BW, van der Veen F, van Beurden M, Redekop \nWK, Bossuyt PM (1995) The accuracy of hysterosalpingography \nin the diagnosis of tubal pathology: a meta-analysis. Fertil Steril \n64(3):486–491\n 26. Broeze KA, Opmeer BC, Van Geloven N, Coppus SF, Collins JA, \nDen Hartog JE, Van der Linden PJ, Marianowski P, Ng EH, Van \nder Steeg JW et al (2011) Are patient characteristics associated \nwith the accuracy of hysterosalpingography in diagnosing tubal \npathology? An individual patient data meta-analysis. Hum Reprod \nUpdate 17(3):293–300\n 27. Symons LK, Miller JE, Kay VR, Marks RM, Liblik K, Koti M, \nTayade C (2018) The immunopathophysiology of endometriosis. \nTrends Mol Med 24(9):748–762\n 28. Chen Y, Wang K, Xu Y, Guo P, Hong B, Cao Y, Wei Z, Xue R, \nWang C, Jiang H (2019) Alteration of Myeloid-derived suppressor \ncells, chronic inflammatory cytokines, and exosomal miRNA \ncontribute to the peritoneal immune disorder of patients with \nendometriosis. Reprod Sci 26(8):1130–1138\n 29. Kobayashi N, Orisaka M, Cao M, Kotsuji F, Leader A, Sakuragi \nN, Tsang BK (2009) Growth differentiation factor-9 mediates \nfollicle-stimulating hormone-thyroid hormone interaction in the \nregulation of rat preantral follicular development. Endocrinology \n150(12):5566–5574\n 30. Slebodzinski AB (2005) Ovarian iodide uptake and \ntriiodothyronine generation in follicular fluid. The enigma of the \nthyroid ovary interaction. Domest Anim Endocrinol 29(1):97–103\n 31. Zhao X, Jiang Y, Wang L, Li Z, Li Q, Feng X (2018) Advances \nin understanding the immune imbalance between T-lymphocyte \nsubsets and NK cells in recurrent spontaneous abortion. \nGeburtshilfe Frauenheilkd 78(7):677–683\n 32. Wang WJ, Zhang H, Chen ZQ, Zhang W, Liu XM, Fang JY, Liu \nFJ, Kwak-Kim J (2019) Endometrial TGF-beta, IL-10, IL-17 and \nautophagy are dysregulated in women with recurrent implantation \nfailure with chronic endometritis. Reprod Biol Endocrinol 17(1):2\n 33. Yang X, Gao T, Shi R, Zhou X, Qu J, Xu J, Shan Z, Teng W \n(2014) Effect of iodine excess on Th1, Th2, Th17, and Treg cell \nsubpopulations in the thyroid of NOD.H-2h4 mice. Biol Trace \nElem Res 159(1–3):288–296\n 34. Zhang J, Lan W, Wang Y, Chen K, Zhang G, Yang W, Chen \nH, Xu W, Ma J, Qin W et al (2022) Ethiodized poppyseed oil-\nbased contrast medium is superior to water-based contrast \nmedium during hysterosalpingography regarding image quality \nimprovement and fertility enhancement: A multicentric, \nrandomized and controlled trial. EClinicalMedicine 46:101363\n 35. Lu J, Qi D, Xu W (2022) Fertility-enhancing effect of oil-based \ncontrast agents during hysterosalpingography and the variation of \nthis effect within a 3-year follow-up period in infertile patients. \nFront Med (Lausanne) 9:948945\nPublisher's Note Springer Nature remains neutral with regard to \njurisdictional claims in published maps and institutional affiliations.","source_license":"CC0","license_restricted":false}