Comparative analysis of natural versus ovarian stimulation cycles in intrauterine insemination by diverse infertility indications

Infertility is increasingly recognized as a significant global health issue, impacting approximately 10–15% of couples worldwide [ 1 ]. The emotional and psychological toll of infertility is profound. This burden is often worsened by the societal stigma associated with childlessness, which complicates the challenges faced by affected couples [ 2 ]. Additionally, financial factors are critical, as the high costs of infertility treatments often cause significant stress and anxiety for couples trying to conceive [ 3 , 4 ]. Over the years, the landscape of infertility treatment has evolved considerably and now includes a diverse array of interventions, from pharmacological therapies aimed at ovulation induction to advanced assisted reproductive technologies (ART). Among these strategies, intrauterine insemination (IUI) has emerged as a popular first-line treatment for certain types of infertility, particularly in cases characterized by unexplained infertility, ovulatory dysfunction, endometriosis, or mild male factor infertility [ 5 , 6 ]. Despite its widespread use, the efficacy of IUI remains inconsistent. This inconsistency has created ongoing debate within the medical community regarding the optimal utilization of ovulation induction in conjunction with IUI procedures to increase success rates [ 7 ]. Importantly, there is a critical gap in the literature concerning the identification of patient populations that may achieve higher pregnancy or live birth rates with ovarian stimulation cycles (OSCs) than with those who could achieve better outcomes through natural cycles (NCs). This gap highlights the need for further research to elucidate the factors that influence treatment success in IUI, particularly in relation to different underlying causes of infertility. The present study utilized a retrospective cohort design to analyze data derived from couples undergoing their first IUI cycle between January 2015 and July 2024. This methodology enables a thorough examination of clinical outcomes while controlling for confounding variables, ultimately facilitating a more comprehensive understanding of the effectiveness of NC IUI versus OSC IUI across various infertility indications. These findings can then provide evidence-based recommendations for clinical practice.

Among 1451 cycles, 833 patients underwent NC IUI, and 618 patients underwent OSC IUI. The baseline characteristics of the couples, including female age, female body mass index (BMI), and male age, differed significantly between the two groups. Females who underwent NC IUI were older, with an average age of 30.42 years compared with 29.44 years, and had a lower BMI (20.86 ± 2.66 vs 21.86 ± 2.92 kg/m 2 ) ( p  < 0.001, p  < 0.001). However, there were no significant differences between the two groups in terms of infertility duration or the proportion of patients with primary infertility ( Table 1 ). Baseline characteristics and clinical outcomes of 1451 patients who underwent IUI for the first cycle according to treatment protocol. IUI: intrauterine insemination; NC: natural cycle; OSC: ovarian stimulation cycle; AOR: adjusted odds ratio; CI: confidence interval; Ref: reference. Independent-samples t test. χ² test or Fisher’s exact test. Logistic regression was employed to model the preterm birth rate adjusted for the presence of twins. 95% CI were calculated by Wilson method. Bold indicates statistical significance. A comparison of reproductive outcomes indicated that the preterm birth rate was significantly higher in the OSC IUI group (14.29%, 95% CI 8.51–22.96) than in the NC group (4.81%, 95% CI 2.11–10.53) ( p  = 0.023) ( Table 1 and Figure 1 ). However, after adjusting for twin pregnancies, the difference in preterm birth rates between the two groups was no longer statistically significant (AOR 2.547, 95% CI 0.835–7.767, p  = 0.100). In addition, the clinical pregnancy rate, live birth rate, miscarriage rate, and ectopic pregnancy rate were comparable between the two groups ( Table 1 and Figure 1 ). Comparison of IUI outcomes between natural cycle and ovarian stimulated cycle. IUI: intrauterine insemination; * p  < 0.05; ns, not significant. Considering that some patients who underwent ovarian stimulation had ovulatory disorders, logistic regression analysis was employed to assess the impact of treatment protocols and infertility indications on clinical outcomes. As shown in Tables 2 and 3 ; Figure 2 , patients with ovulatory disorders had a significantly higher clinical pregnancy rate than did those with normal ovulation (21.99% vs 14.15%, AOR 1.992, 95% CI 1.207–3.288). Similarly, the live birth rate in patients with ovulation dysfunction was also greater (18.93% vs 11.51%, AOR 2.326, 95% CI 1.312–4.124). In addition, the preterm birth rate was greater in patients with ovulatory disorders (14.86, 95% CI 8.55–23.97) than in the control group (5.74, 95% CI 2.77–11.23) ( p  = 0.032), while the live birth rate of twins was also greater (6.76%, 95% CI 2.86–14.60 vs 0.82%, 95% CI 0.14–4.43) ( p  = 0.030). After adjusting for twin pregnancies, the difference in preterm birth rates between the two groups was no longer statistically significant (AOR 2.494, 95% CI 0.896–6.940, p  = 0.080). The miscarriage rate and the ectopic pregnancy rate were comparable between the two groups ( Table 3 , Figure 2 ). Overall, ovulation dysfunction was confirmed as an independent favorable factor influencing the clinical outcomes of IUI. Comparison of IUI outcomes between normal ovulation and ovulatory disorder. **** p  < 0.0001; * p  < 0.05; ns, not significant. Comparison of IUI clinical outcomes by patient characteristics and infertility indications. IUI: intrauterine insemination; BMI: body mass index; NC: natural cycle; OSC: ovarian stimulation cycle; AOR: adjusted odds ratio; CI: confidence interval; Ref: reference. The data are presented as rates (%). Logistic regression was used to model the clinical pregnancy rate and live birth rate. Bold indicates statistical significance. Comparison of IUI clinical outcomes between patients with and without ovulatory disorder. IUI: intrauterine insemination. AOR: adjusted odds ratio; CI: confidence interval; Ref: reference. χ² test or Fisher’s exact test. Logistic regression was employed to model the preterm birth rate adjusted for the presence of twins. Bold indicate statistical significance. Notably, AORs for the pregnancy rate and live birth rate in the OSC group were 0.729 (95% CI 0.462–1.148) and 0.601 (0.355–1.019), respectively, compared with those in the NC group ( Table 2 ). There were no statistically significant differences ( p  = 0.173 for pregnancy rate and p  = 0.059 for live birth rate). However, a numerical trend toward improved clinical pregnancy and live birth outcomes was observed in the natural cycle. Subsequently, we further investigated whether the treatment protocol affected IUI outcomes in patients with normal ovulation. Among 1060 patients with normal ovulation, 833 underwent NC IUI, and 227 received OSC IUI ( Table 4 ). The clinical pregnancy rate (14.89%, 95% CI 12.45–17.58 vs 11.45%, 95% CI 7.89–16.02) and live birth rate (12.48%, 95% CI 10.23–15.04 vs 7.93%, 95% CI 4.87–12.28) were higher in the NC IUI group than in the OSC group. Further comparison of IUI outcomes among different ovarian stimulation agents, with the natural cycle as the reference group, showed no statistically significant differences in clinical pregnancy and live birth rates ( Table 4 ). The IUI clinical outcomes of patients with normal ovulation ( n  = 1060) by treatment protocols. IUI: intrauterine insemination; NC: natural cycle; OSC: ovarian stimulation cycle; LE: letrozole; CC: clomiphene citrate; Gn: gonadotrophin; CI: confidence interval. The χ² test or Fisher’s exact test was used for comparison between groups. Next, we compared patients with normal ovulation across different indications. The baseline characteristics of the couples included age, female BMI, anti-Müllerian hormone (AMH) level, and basal LH/FSH ratio which were similar across the endometriosis, tubal infertility, male factor infertility, and unexplained infertility groups ( Table 5 ). The duration of infertility in the endometriosis group was shorter than that in the other three groups (2.12 ± 1.30 years vs. 2.50 ± 1.68, 2.37 ± 1.43, and 2.52 ± 1.56 years, p  = 0.033). Bonferroni correction for post-hoc analysis, p -value < 0.012 was considered statistically significant. The primary infertility rate in the tubal factor infertility group was lower than that in the other three groups (58.51% vs. 75.57%, 71.99%, and 66.67%, p  = 0.011). The clinical pregnancy rates were 11.36%, 13.83%, 14.40% and 14.02% in the endometriosis, tubal infertility, male factor infertility, and unexplained infertility groups, respectively, with no statistically significant difference among the four subgroups ( p  = 0.782). The live birth rate, preterm birth rate, miscarriage rate, and ectopic pregnancy rate were comparable among groups with different indications ( Table 5 ). Baseline characteristics of patients with normal ovulation and IUI clinical outcomes by infertility indications. BMI: body mass index; AMH: antimullerian hormone; LH: luteinizing hormone; FSH: follicle-stimulating hormone. Comparisons among multiple groups were performed using one-way ANOVA. Bonferroni correction for post-hoc analysis, p -value < 0.012 was considered statistically significant. The χ² test or Fisher’s exact test was used for comparison between groups. Bold indicate statistical significant. Further analysis revealed that the clinical pregnancy rate in the NC IUI group was greater than that in the OSC IUI group for patients with endometriosis (12.70%, 95% CI 7.68–19.85 vs. 8.00%, 95% CI 2.86–18.36), tubal infertility (16.67%, 95% CI 8.95–27.96 vs. 7.14%, 95% CI 1.52–20.02), and male factor infertility (15.92%, 95% CI 12.88–19.42 vs. 9.40%, 95% CI 5.45–15.39) ( Table 6 ). The difference in clinical pregnancy rates between natural and ovarian stimulation cycles for the male factor infertility group was statistically significant ( p  = 0.047). The live birth rate also showed a similar trend. Specifically, the live birth rate in the NC IUI group was higher than that in the OSC IUI group for patients with endometriosis (11.11%, 95% CI 6.35–18.17 vs. 6.00%, 95% CI 1.99–14.09), tubal infertility (10.61%, 95% CI 4.94–20.35 vs. 3.57%, 95% CI 0.44–12.37), and male factor infertility (13.06%, 95% CI 10.28–16.33 vs 6.71%, 95% CI 3.43–12.07). Conversely, the clinical pregnancy rate and live birth rate were higher for OSC IUI than for NC IUI for patients with unexplained infertility (20.51%, 95% CI 10.84–34.45 vs. 12.89%, 95% CI 8.84–17.99; 12.82%, 95% CI 5.29–24.76 vs. 11.11%, 95% CI 7.35–16.18, respectively). However, the differences in the clinical pregnancy rate and live birth rate between the two groups were not statistically significant ( p  = 0.205 and 0.970, respectively) ( Table 6 ). Comparison of clinical outcomes between NC IUI and OSC IUI according to infertility indications in the patients with normal ovulation. IUI: intrauterine insemination; NC: natural cycle; OSC: ovarian stimulation cycle. χ² test or Fisher’s exact test. Bold indicates statistical significance.

This research was carried out in compliance with the ethical principles of the World Medical Association Declaration of Helsinki and relevant institutional guidelines. This study was approved by the Ethics Committee of the First Affiliated Hospital of Ningbo University (No. 2025-172RS). From January 2015 to July 2024, a total of 1,451 infertile couples who underwent their first IUI cycle were included in this study. All the data were retrieved from the anonymized clinical database of the Reproductive Medicine Centre at the First Affiliated Hospital of Ningbo University. Women who underwent IUI treatment had at least one patent fallopian tube confirmed by hysterosalpingography or laparoscopy. Patients were excluded if they had bilateral tubal pathology, uncorrected endocrine disorders, or serious male factors; these included a sperm concentration less than 5 million/mL or a progressive motility rate less than 10%. Written informed consent for treatment and follow-up was obtained from all participants, whose privacy rights were upheld throughout the entire medical process. Treatment protocols (NC or OSC) were determined through joint decision-making between the couples and the physician. Females undergoing NC IUI were generally monitored with transvaginal ultrasound starting from cycle day 11 and then repeated every 2–3 days on the basis of follicle size. Ovulation was triggered by the injection of 5,000–10,000 IU human chorionic gonadotropin (hCG) as soon as the follicle reached 18 mm. The timing of insemination depended on patients’ serum LH levels measured on the day of hCG injection. The IUI procedure was performed on the day of hCG injection if the LH level was above 25 IU/L; otherwise, insemination was performed 24 h later. Three types of ovarian stimulation were used in this study: (1) letrozole (LE) at 2.5–5.0 mg/d starting from day 5 of the menstrual cycle for 5 days; (2) clomiphene citrate (CC) at 50–100 mg/d starting from day 5 of the menstrual cycle for 5 days; and (3) human menopausal gonadotrophin (HMG) at 75 IU/d starting from day 5 for a variable duration depending on the ovarian response. Follicles were monitored by transvaginal ultrasound, and hCG injection and insemination were performed similarly to the natural cycle. The IUI procedure was terminated if the number of mature follicles exceeded three. On the day of insemination, semen was collected and prepared via density gradient centrifugation after liquefaction. Approximately 0.5 ml of a washed semen sample was used in the insemination procedure. Patients were subsequently instructed to stay in bed for at least 30 min after the procedure and take 20 mg/d dydrogesterone (Duphaston, Abbott) starting the next day for 14 days. Patients were followed up until the IUI procedure was completed. If patients became pregnant, they were followed up until delivery. Clinical pregnancy was defined as the presence of a gestational sac observed on ultrasound examination. Live birth was defined as the birth of an infant after 28 weeks of gestation with postnatal evidence of life, such as breathing or a heartbeat. Twin pregnancy was defined as the observation of two intrauterine gestational sacs. In our study, continuous variables are expressed as the mean ± standard deviation (SD), and dichotomous variables are presented as n (%). The differences in the baseline characteristics between the two groups were analyzed via an independent samples t test. Comparisons among multiple groups were performed using one-way ANOVA, with Bonferroni correction for post-hoc analysis. Pearson’s chi-square test (χ 2 ) or Fisher’s exact test was used to compare the proportions between the two groups. 95% confidence interval (CI) were calculated by Wilson method. In addition, we used multivariable logistic regression analysis to assess whether the clinical pregnancy rate and/or live birth rate were associated with any of the following variables: female age, body mass index (BMI), duration of infertility, type of infertility, treatment protocol, or infertility factors. The adjusted odds ratios (AORs) and 95% CIs were calculated from the multivariable model to assess associations between potential predictors and clinical outcomes. A p -value < 0.05 was considered to indicate statistical significance. Statistical analysis was performed via the Statistical Package for Social Sciences (SPSS version 27.0).

Understanding patient characteristics may help tailor individualized treatment protocols and improve patient counseling by aligning clinical expectations with real-world outcomes. Ovulatory disorder was identified as an independent favorable factor for the clinical outcomes of IUI in our cohort. For clinical reference, NC IUI may suggest a potential trend that requires further confirmation for patients with endometriosis, tubal infertility, and male factor infertility, whereas OSC IUI may indicate a similar non-significant trend in patients with unexplained infertility, with all decisions requiring shared discussion between clinicians and patients based on individual clinical features. The findings offer valuable insights for future research and clinical practices aimed at optimizing fertility treatment outcomes in patients undergoing IUI.

This study focused on infertility and its management through IUI. We conducted a cohort study to compare the clinical outcomes of IUI under different treatment protocols. It is worth noting that luteal-phase support plays a crucial role in improving IUI pregnancy outcomes, as it can alleviate the risk of miscarriage caused by luteal insufficiency by directly supplementing progesterone and regulating endometrial receptivity. All patients in this cohort received standardized luteal-phase support with dydrogesterone, a regimen supported by high-level clinical evidence demonstrating its efficacy in enhancing luteal function and optimizing pregnancy outcomes in IUI cycles [ 8 , 9 ]. The clinical pregnancy rate and live birth rate were higher in patients who underwent OSC-IUI than in those who underwent NC-IUI (18.12% vs. 14.89%, p  = 0.099; 14.89% vs. 12.48%, p  = 0.186). However, logistic regression analysis revealed that ovulatory disorders were the only independent risk factors affecting the clinical pregnancy rate and live birth rate. Compared with patients without ovulation dysfunction, the AORs for the clinical pregnancy rate and live birth rate in patients with ovulatory disorders were 1.992 and 2.326, respectively. In contrast, the AORs for patients undergoing OSC were 0.729 for the clinical pregnancy rate and 0.601 for the live birth rate, compared with those of NCs. For patients with normal ovulation, natural cycle IUI appears to yield better outcomes compared to stimulated ovarian cycle IUI. Further analysis revealed that for patients with endometriosis, tubal infertility, or male infertility, the clinical pregnancy rate and live birth rate were higher in the NC IUI group than in the OSC IUI group. However, for patients with unexplained infertility, clinical outcomes appear more favorable in stimulated ovarian cycles compared to natural cycles. The strength of our study lies in the systematic analysis of IUI treatment protocols (natural cycle vs. ovarian stimulation cycle) across different infertility indications. Our analysis specifically focuses on the first treatment cycle. Most studies, including several well-designed randomized controlled trials (RCTs), have analyzed data from all cycles combined [ 10 , 11 ]. The therapeutic effects vary greatly among different IUI cycles [ 12–15 ]. Key findings indicate that different treatment protocols suit different populations. This tailored approach may lead to improved protocols and better patient outcomes. One limitation of this study is its non-RCT design, which may introduce selection bias due to the shared decision-making approach employed. Observed differences between NC and OSC groups may therefore reflect patient selection bias rather than a true independent treatment effect. And the single-center design and the relatively young mean age of female participants may limit the generalizability of the findings. Additionally, cost-effectiveness analysis was not included. Future analyses need to be conducted using well-designed RCTs that compare clinical outcomes between NC IUI and OSC IUI in the first cycle across diverse populations in a multi-center setting. Since the advent of artificial insemination technology, many studies have investigated the correlation between different treatment protocols and clinical outcomes [ 1 , 7 , 10 , 14 , 16–18 ]. In a high-quality RCT of 932 couples with unexplained or male factor infertility, superovulation combined with IUI was 1.7 times as likely to achieve pregnancy as natural cycle IUI (95% CI 1.2–2.6) [ 19 ]. Another meta-analysis published in 2020, which focused on patients with unexplained infertility, included 26 studies reporting on 5,316 women [ 20 ]. The relative risk (RR) for live birth or ongoing pregnancy when CC IUI was compared with NC IUI was 1.05 (95% CI 0.63–1.77), when letrozole IUI was compared with NC IUI was 1.15 (95% CI 0.63–2.08), and when gonadotrophin IUI was compared with NC IUI was 1.46 (95% CI 0.92–2.30). Most studies on unexplained infertility have concluded that OSC IUI yields better clinical outcomes than does NC IUI; consequently, some guidelines do not recommend NC IUI for treating unexplained infertility [ 1 , 18 , 20 ]. Consistent with the findings of previous studies, our study revealed that the clinical pregnancy rate following OSC IUI in unexplained infertility patients was higher than that following NC IUI (20.51% vs. 12.89%, p  = 0.205). However, the difference was not statistically significant, which might be due to the limited sample size. Notably, our findings in the unexplained infertility population are consistent with the ASRM guideline on unexplained infertility, which suggests that offering IUI without ovarian stimulation does not improve success rates compared to attempting conception without treatment [ 1 ]. Although our data showed a numerical advantage of OSC IUI over NC IUI in this subgroup, the lack of statistical significance further supports the ASRM guideline’s recommendation, as it indicates that NC IUI does not provide a meaningful benefit in improving pregnancy outcomes for patients with unexplained infertility. In a randomized crossover trial of 74 couples with male subfertility who collectively completed 308 treatment cycles, there was no statistically significant difference in the clinical pregnancy rate between controlled ovarian hyperstimulation IUI and natural cycle IUI [ 10 ]. One retrospective study of 601 IUI cycles from 307 couples with male infertility reported that the clinical pregnancy and live birth rates were comparable between the natural cycle and the ovarian stimulation cycle (12.1% vs. 13.1%, p  = 0.787; 11.1% vs. 11.9%, p  = 0.778). Moreover, multiple pregnancy rates were higher in the ovarian stimulation cycle of IUI than in the natural cycle (8.7% vs. 0%, p  = 0.091) [ 7 ]. In our study, the pregnancy rate and live birth rate in the natural cycle of IUI for couples with male infertility were significantly higher than those in the stimulated cycle (15.92% vs. 9.40%, p  = 0.047; 13.06% vs. 6.71%, p  = 0.034). We believe that the main reason for this inconsistency is that our study focused only on the first IUI cycle to reduce bias caused by varying cycle numbers, whereas other studies analyzed all cycles together. For patients with endometriosis-associated infertility, treatment selection requires individualized balancing of efficacy, safety, ovarian reserve, patient age, and disease severity. Recent systematic reviews have suggested that first-line in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) may be prioritized in infertile patients with endometriosis, particularly in moderate-to-severe disease, while surgery is generally reserved for symptom control, specific anatomical indications, or patient preference [ 21 ]. Surgery can improve spontaneous conception rates in selected patients with minima-to-mild endometriosis but carries potential risks to ovarian reserve and does not consistently improve IVF outcomes [ 22 ]. A retrospective study of 65 patients with moderate-to-severe endometriosis receiving 245 IUI treatment cycles was conducted to compare ongoing pregnancy rates with or without ovarian stimulation [ 13 ]. The cumulative number of ongoing pregnancies was greater in patients who underwent IUI with ovarian stimulation than in those without ovarian stimulation in the first three cycles (45.8% vs. 19.5%, p  = 0.016). While in our study, the pregnancy rates and live birth rates were comparable between natural cycles and ovarian stimulation cycles (12.70% vs. 8.00%, p  = 0.376; 11.11% vs. 6.00%, p  = 0.452). These inconsistent results might be due to differences in the study populations. Most patients who underwent IUI at our center had mild-to-moderate endometriosis, whereas patients with severe endometriosis preferred IVF. In our previous study involving 97 patients with stage I-II endometriosis, we reported that the administration of gonadotrophin-releasing hormone agonist (GnRHa) greatly elevated the pregnancy rate compared with that in the control group (15.29% vs. 11.82%, p  = 0.035), with an AOR of 23.190. However, there was no significant difference between natural and ovarian stimulation cycles in these patients (14.29% vs. 12.2%, p  = 0.423) [ 15 ]. Our data demonstrated that clinical pregnancy and live birth rates following IUI in the endometriosis group were comparable to those in tubal, male factor, and unexplained infertility groups, supporting the pragmatic role of IUI as a reasonable first-line step in appropriately selected patients before advancing to more invasive ART. These findings underscore the importance of a stepwise, patient-centered approach to infertility care, in which IUI remains a valuable and accessible intervention alongside expectant strategies, surgical correction, and IVF/ICSI.