The importance of infertility duration and follicle size according to pregnancy success in women undergoing ovulation induction with gonadotropins and intrauterine insemination

Abstract In this study, we aimed to evaluate the effect of infertility duration and dominant follicle size measured on the day of human chorionic gonadotropin (HCG) administration on pregnancy rates in infertile women undergoing ovulation induction (OI) with gonadotropin (GND). A total of 352 patients aged 20 to 41 years who were diagnosed with unexplained infertility or polycystic ovary syndrome (PCOS) were included in this study. Patients with a history of multifollicular development or follicle stimulating hormone (FSH) value more than 12 IU/ml were excluded from the study. The demographic and clinical features of the patients were obtained from the patients’ files and hospital automation system and recorded for each woman. The demographic and clinical features of the patients were recorded. Patients were divided into two groups as live birth (group 1, n = 47) and non-live birth groups (group 2, n = 305). There were no statistically significant differences in regard to age, infertility type, follicle stimulating hormone (FSH) level, oestradiol (E2) level, antral follicle count (AFC), cycle characteristics, GND type, number of follicles, E2 level and endometrial thickness on HCG day, total GND dose, dominant follicle size (p > 0.05). Infertility duration in group 1 was 3.5 ± 2.1; in group 2, 4.7 ± 3.9 years. This difference was statistically significant (p = .014). According to this study, live birth rates after GND + IUIs(intrauterine insemination) were significantly affected by the duration of infertility. But the dominant follicle size and endometrial thickness measured on HCG day in GND and IUI cycles did not have a significant effect on pregnancy rates. IMPACT STATEMENT What is already known on the subject? It is thought that the timing of the ovarian triggering is vital for the success of intrauterine insemmination (IUI) treatment. What do the results of this study add? According to our results live birth rates after GND + IUIs were significantly affected by the duration of infertility. What are the implications of these findings for clinical practice and/or further research? It is determined that the duration of infertility is significant and patients should be encouraged to the treatment as soon as possible.


Introduction
In controlled ovarian stimulation, also called ovulation induction (OI), which is used for the treatment of infertile couples, gonadotropin (GND) treatment is used in cases when clomiphene citrate (CC) treatment which is the first option does not result in pregnancy.A variety of factors affect the success of OI by GND treatment and intrauterine insemination (IUI) treatment cycles (Ayeleke et al. 2020).One of these factors is the size of the dominant follicle size on the day of human chorionic gonadotropin (HCG) administration.Although there is no general consensus on the ideal size of the follicle on the literature, the practice is that HCG is applied when the size is 17 mm or higher (Maher et al. 2017).
There are conflicting opinions about the timing of HCG in the literature, and it is thought that the timing of the ovarian triggering is vital for the success of the treatment, that if it is done while the follicle is immature, the fertility success is reduced.One study claims that prolonged cycles of treatment increase the success rate of pregnancy, along with the cost of the treatment.On the other hand, some authors advocate triggering of HCG while the follicle size is small (Teramoto et al. 2019, Silverberg et al. 1991).In our study, we aim to assess the effects of follicle size on the day of HCG application on the success of fertility among infertile women who have gone through OI treatment with GND.

Materials and methods
This case control study has been conducted between January-October 2018 in the Reproductive Endocrinology and Infertility Clinic of the University of Health Sciences, Ankara Dr. Zekai Tahir Burak Women's Health Research Centre, with the approval of ethical committee (date: 04 May 2018, approval number: 16).A total of 352 patients with ages between 20 to 41 years who were diagnosed with unexplained infertility or polycystic ovary syndrome were included in this study.
Prior to the treatment, all the patients' serum follicle stimulating hormone (FSH), oestradiol (E2), thyroid stimulating hormone (TSH), prolactin (PRL) levels are evaluated.Basic gynecological ultrasonographic screenings were conducted.The women's hysterosalpingography (HSG) screenings and their partners' spermiogram results were evaluated.While patients with normal basic infertility examination results were diagnosed with unexplained infertility, polycystic ovary syndrome(PCOS) patients are diagnosed according to the revised Rotterdam criteria (Ghosh et al. 2003).
Patients with an FSH value of more than 12 IU/ml, patients that are diagnosed with uterine and/or tubal pathology during HSG, patients who have a history of multifollicular development or whose partners have abnormal spermiogram analysis results were excluded from the study.Patients were divided into two groups as live birth (group 1, n ¼ 47) and non-live birth groups (group 2, n ¼ 305) after GND treatment.The groups were compared statistically in terms of demographic features, infertility duration, type of infertility, basal serum FSH and E2 levels, the number of basal antral follicles in both ovaries, initial GND dose, total GND dose, number of cycles, type of GND and HCG used, endometrium thickness on the day of HCG, E2 level on the day of HCG, induction duration, IUI time after HCG, number of follicles developed, dominant follicle size and live birth ratio.
OI and IUI procedures were conducted as explained here.First of all, after screening the patients between the 2 nd and 3 rd days of the cycle with transvaginal ultrasonography, according to the low dose step-up protocol, GND 37.5-75 IU/day FSH/HMG(human menopausal gonadotropin) (Gonal-F V R ; Merck Sereno/MenogonV R ; Ferring Pharmaceuticals Istanbul, Turkey) treatment was started, and follicle sizes were recorded with a series of ultrasonographic screenings.Only one dominant follicle presented on the day of triggering.When the follicle size reached 18 mm or higher, 250 mg recombinant HCG (Ovitrelle; Merck Serono, Modugno (BA), Italy) subcutaneous or 10000 IU urinary HCG (Pregnyl; Organon, Oss, Netherlands) was applied intramuscularly.Semen samples were collected via masturbation and prepared with swim up method.IUI was conducted 24 or 36 hours after HCG was applied (Cantineau et al. 2014).We used one type of insemination cannula .IUI was administered in a single session using an elastic intrauterine cannula (TecnoCath insemination cannula, Ankara, Turkey).There is 2-3 physician with similar experience involved in the IUI procedure.There was no relation between the performers and the clinical results.As a luteal phase support, 90 mg progesterone vaginal gel (Crinone; Merck Serono, Feltham, UK) was applied for 14 days following the insemination.2 weeks after the IUI, serum ß-HCG levels were examined.Clinical pregnancy diagnosis was made after foetal heartbeat was observed at transvaginal USG during 6 th -7 th week of gestation.There weren't any multiple pregnancies.Luteal phase support was continued for these patients until 10-12 th week of the pregnancy.

Statistical Analysis
Statistical Package for the Social Sciences (SPSS), v.22 (SPSS Inc., Chicago, IL, ABD) software was used for the analysis of the results.Kolmogorov-Smirnov test and Shapiro-Wilk test was used for assessment of normality.For the variables with normal distribution, Student's t test was used for the analysis.On the other hand, Mann-Whitney U test was used to analyse the variables that do not have a normal distribution.Nominal data that exhibits normal distribution was presented in the charts as mean and standard deviation.On the other hand, categorical data was represented as n (count) and percentages (%).P value less than 0.05 was considered statistically significant.The statistical program on the website of the statistical department of the University of British Colombia was used to calculate the sample size and the power of our study (https://www.stat.ubc.ca/�rollin/stats/ssize/b1.html).To achieve a level of significance of 0.05 with a two-sided test, it was found that a minimum sample size of 350 patients was sufficient to achieve a power of at least 90%.

Results
Three hundred fifty-two women were found who met the inclusion/exclusion criteria during the study period.Patients were divided into 2 groups with respect to live birth ratio and the factors which affected their pregnancy results were evaluated.Demographical, clinical and cyclic features of the patients are summarised in Table 1.Out of the 352 patients included in this study, mean ages are 29.2 ± 5.2 for group 1 and 29.3 ± 5.0 for group 2, difference of which is not statistically significant (p > 0.05).AFCs (Antral follicle count) are observed to be 7.3 ± 3 in the right ovary and 7.4 ± 3.4 in the left ovary, and no difference was noticed in terms of live birth.Among the live birth group (group 1), 45 patients are given FSH (95.7%) and 2 patients are given HMG (4.3%).Also, 12 (25.5%)patients of group 1 are given recombinant HCG while 35 (74.5%) are given urinary HCG.There were no statistically significant differences between the two groups in terms of infertility type, FSH and E2 levels, cycle characteristics, type of GND used in treatment, the number of developed follicles, E2 level on the day of HCG, endometrium thickness on the day of HCG, total GND dose and dominant follicle size.The infertility time was 3.5 ± 2.1 in group 1 and 4.7 ± 3.9 in group 2, the difference between which was statistically significant (p ¼ .014).Live birth occurred in 47 patients, 13 (27.7%) of which were unexplained infertile, and 34 (72.3%) of which were PCOS.Clinical pregnancy rate was 17.3% (n ¼ 60); while live birth ratio was 13.4% (n ¼ 47).Among clinically pregnant patients, 9 had abortion and 3 had 20þ weeks pregnancy loss.No stillbirth was observed.We also analysed comparing outcomes in Table 2 between two groups in women aged more than 35.There were statistically significant differences between the two groups in terms of age, BMI, FSH, E2, AFS(right/left) and initial dose in women aged less than 3 5 (Table 2).Also we performed multivariate regression test on the duration of infertility (Table 3).

Discussion
There is no consensus on the ideal dominant follicle size on the day of HCG during GND and IUI cycles.In one study, pregnancy rates and follicles size are compared during GND and CC cycles; and the follicle size in CC group was 19.0 ± 1.9 mm, whereas in GND group it was found to be 18.3 ± 1.7 mm.There was no statistically significant difference between the two groups in terms of pregnancy (Fallah Tafti et al. 2017).The study concluded that HCG triggering should be earlier in GND than CC, because the follicle maturation is faster in patients with GND treatment.Similarly, in our study follicle size was measured to be 19.0 ± 1.5 mm in live birth group and 19.1 ± 1.4 mm in the other group.Another study suggests HCG triggering when follicle size is 17 mm diameter, motivated by the fact that ovulation can occur if follicle is kept waiting (Silverberg et al. 1991).Ghosh et al. grouped the follicle size into 3 as immature (<15 mm), mature (15-23 mm), and postmature (>23 mm) and evaluated the pregnancy results, and found out that the highest conception ratio is in the mature group (Ghosh et al. 2003).Buzaglo et al. (2012) compared 4 categories of follicle size (17, 18, 19 and 20 mm) and studied the pregnancy results, finding out that the highest pregnancy ratio is in 19 mm group (34.2%), however the difference between the pregnancy ratio of groups was not statistically significant.
In the literature, some studies state that infertility duration is an indicator of pregnancy success in OI and IUI treatments (Wang et al. 2019, Vargas-Tominaga et al. 2020, Immediata et al. 2020).Since there are many different factors in the aetiology of unexplained infertility, this is an expected result.Couples with shorter infertility duration are expected to be affected less from longer or less severe fertility factors than couples with longer infertility duration, therefore they are expected to get better results after the treatment.In unexplained infertility, up to three cycles of IUI with ovarian stimulation can be considered an effective treatment method, even in couples who cannot achieve pregnancy with await management for two years.Especially in young couples with an infertility period of less than three years, at least two cycles of IUI can avoid a significant amount of unnecessary IVF procedures (Ombelet et al. 2020)  best initial treatment for most couples is ovarian stimulation (typically three or four cycles) followed by oral medications and intrauterine insemination (OS-IUI) (Practice Committee of the American Society for Reproductive Medicine 2020).O' Leary et al. (2009) found out that the clinical pregnancy ratio is significantly larger in urinary GND induction group when compared to recombinant FSH group (22.2% and 10.9%, respectively).Demirol and Gurgan (2007) found out that total GND dose per cycle is significantly less in recombinant FSH group than the group with HMG treatment.Furthermore, the authors found out that the number of dominant follicles is higher in the recombinant FSH group than the HMG group.However, in terms of GND treatment duration, there were no statistically significant differences found.In our study, however, no statistically significant difference was found in terms of total GND dose between FSH and HMG treatments.Treatment length is also found to be similar between the two groups.
In one study (Palatnik et al. 2012), the pregnancy ratio is found to be higher in patients with endometrium thicker than 12 mm, while in another study (Kolibianakis et al. 2004) there was no significant difference in terms of pregnancy rate between patients with different endometrial thickness.In our study, the relationship between endometrium thickness and pregnancy results is investigated, and similar with (Kolibianakis et al. 2004) no significant difference was found between the two groups.In the live birth group, both basal E2 levels and the E2 level on the day of HCG were found to be higher than the other group.However, the difference is not statistically significant.
Although our study is not a randomised controlled study, it is a significant contribution to the literature due to the high number of patients, and strict follow-up of patients until their pregnancy results are obtained.The possible limitations of our study are, on the other hand, being a retrospective study and patient groups not being perfectly homogenous.

Conclusions
It was observed that there is no statistically significant effect of dominant follicle size and endometrial thickness on pregnancy rates among unexplained and anovulatory infertile women who undergone OI and IUI treatment via GND.But it is determined that the duration of infertility is significant and patients should be encouraged to the treatment as soon as possible.

Poster presentation
This work was presented as an oral presentation at the biannual meeting of the Turkish Society of Reproductive Medicine (TSRM) which was held in Antalya, Turkey between 8 and 11 November 2018.

Study design
Retrospective cohort study.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The author(s) reported there is no funding associated with the work featured in this article.

Table 1 .
n paralel with Osmanlıo� glu et al. study results (Osmanlıo� glu et al. 2022), The American Society for Reproductive Medicine's latest guidelines for unexplained infertility recommend that the Demographical, clinical characteristics of the patients and live birth rate.

Table 2 .
Comparing baseline characteristics between two groups patients under 35 and above.