Low BMI is associated with poor IUI outcomes: a retrospective study in 13,745 cycles

doi:10.21203/rs.3.rs-1603900/v1

Download PDF

Research Article

Low BMI is associated with poor IUI outcomes: a retrospective study in 13,745 cycles

https://doi.org/10.21203/rs.3.rs-1603900/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Although the effect of body mass index (BMI) on the outcome of in vitro fertilization (IVF) has been extensively studied, its correlation with intrauterine insemination (IUI) has rarely been reported. Therefore, we aimed to examine the impact of BMI on IUI outcomes．

Methods: Six thousand four hundred and seven women undergoing 13,745 IUI cycles stratified by BMI were included in the study. Cox regression was used to analyze the association between BMI and cumulative live birth across multiple IUI cycles. Generalized estimating equation (GEE) was used to analyze the live birth rate per cycle.

Results: Compared with normal weight women (n = 4,563), underweight women (n = 990) had a lower cumulative pregnancy and live birth rate (20.71% vs 25.93%, 17.17% vs 21.61%, respectively), while overweight women (n = 854) had a higher cumulative pregnancy and live birth rate (31.97%, 26.58%). Adjusted for confounders, hazard Ratio (HR) for achieving live birth following up to a maximal of 4 IUI cycles was 0.80 (95%CI: 0.67-0.95) comparing underweight with normal weight. In GEE analyses, low BMI was also associated with a lower per-cycle live birth rate 0.77 (95%CI: 0.64-0.92) with adjustment for cycle-specific parameters, including ovarian stimulation, endometrial thickness and follicular diameter.

Conclusion: Underweight is associated with poor IUI outcomes.

Intrauterine insemination

body mass index

underweight

cumulative live birth

Infertility affects approximately 10–15% of couples or more. IUI is arguably used as first choice treatment for couples whose cause of infertility is unexplained or mild male factor infertility due to its simplicity and affordability [1]. While NICE suggested that IUI has a very limited value in infertility care [2], several recent studies supported IUI as a cost-effective choice of treatment in comparison with IVF [3]. Nevertheless, large scale dataset indicated that the delivery rates per IUI cycle using homologous sperm remained stable at rates as low as 8–9% [4, 5].

Several factors, such as age, infertility diagnosis, BMI, and semen quality have been linked to the likelihood of pregnancy following IUI cycles. Among them, BMI has been cumulatively reported as a negative predictor for assisted reproductive outcome [6]. However, most of the researches were performed in IVF, and the information regarding IUI outcomes in extreme BMI is still controversial. While some authors suggested that high BMI in IUI cycles may reduce response to ovarian stimulation, increase abnormal pregnancy [7] and comprised pregnancy rate [8], other studies suggested a neutral effect of BMI on IUI outcomes [9, 10]. Notably, several authors showed that BMI was positively associated with pregnancy following IUI [11], suggesting a detrimental effect of underweight rather than obesity. The inconsistency among studies may be due to different cut-off points for BMI, discrepancy of inclusion criteria among studies, and/or varying focus of outcomes measures. In several studies [9, 12], including a very recent one [7], underweight women were combined with normal weight women or simply excluded in analyses due to low number. Such study design may underestimate the effect of low BMI on IUI outcome, especially for Asian populations in which low BMI is more frequent. In the present study, we aim to evaluate the effect of BMI on the outcome of IUI cycles in a Chinese cohort including 13,745 cycles, focusing on the effect of underweight on both per-cycle live birth and cumulative live birth following up to four cycles.

Study population

The retrospective cohort study was performed on patients who underwent IUI treatment in the Xiamen University Affiliated Chenggong Hospital in the period from July 2015 to December 2020. Before initiating therapy, all couples underwent a standard infertility evaluation. All women had at least one patent fallopian tube documented by either hysterosalpingography or laparoscopy, and all men had prewash total motile counts over 1,000,000 in an ejaculate during the infertility workup. Exclusion criteria included women with bilateral tubal pathology or other endocrine disorders, such as thyroid disease. Patients receiving more than 4 cycles (n = 34) were also excluded to reduce bias, because these patients may have poorer prognosis than others. This retrospective study was approved by the Ethics Committee of the Medical College Xiamen University.

All IUI cycles were divided into three cohorts according to the World Health Organization (WHO) guidelines: underweight (< 18.5kg/m²), normal weight (18.5–24.9kg/m²), and overweight (25.0–29.9kg/m²). Women with BMI ≥ 30.0 kg/m² were asked to optimize their weight before considering treatments, so we did not analyze the data from patients with BMI ≥ 30.0 kg/m².

IUI Protocol

Women underwent Ovarian stimulation was performed with letrozole (2.5mg) (Jiangsu Hengrui Medicine Co. Jiangsu, China) for five days on menstrual cycle days 3–5. A number of patients were stimulated with 37.5–75.0 IU/day human menopausal gonadotropin (hMG) (Maanshan Pharmaceutical Trading Co., Anhui, China). The hMG dose was adjusted according to the ovarian response and the patient’s characteristics. The follicular development and endometrial thickness were monitored by serial transvaginal ultrasounds. When the maximum follicle reached at least 18 mm in diameter, 250 µg of recombinant human chorionic gonadotropin hormone (hCG) (Merck Serono, Spain) was administered. The IUI procedure was performed 32 to 36 hours after hCG administration. Cycles were cancelled when more than three dominant follicles of ≥ 14 mm were detected to prevent high-order multiple pregnancies.

In natural IUI cycles, detection of ovulation was performed with either urine luteinizing hormone (LH) tests or by transvaginal ultrasound. Patients were inseminated 24 hours after detection of a spontaneous LH surge, or 32 to 36 hours following the administration of hCG, where follicular growth was monitored by ultrasound.

A 2-week course of daily treatment with 20 mg of dydrogestrone (Abbott Biologicals, Netherlands) was prescribed for luteal support. The serum β-hCG level was estimated 14 days after IUI. Clinical pregnancy was identified 5 weeks after the IUI with the presence of an intrauterine gestational sac and a pulsating fetal heartbeat.

Statistical analysis

BMI was categorized according to WHO guidelines. The effect of BMI on cumulative outcomes of IUI treatment was estimated with the Kaplan–Meier method according to cycle numbers. Due to the relatively low success rate of IUI treatment, we used a “conservative” model [13] to calculate the cumulative outcomes. In this model, patients who did not return for subsequent treatment were supposed to have no chance of a live birth even if they remained in the cohort.

Cox regression was used to access the effect of covariates, including female age, basal follicle-stimulating hormone (FSH), LH and antral follicle count (AFC), etiologies (Unilateral tubal obstruction, polycystic ovary syndrome (PCOS) and mild endometriosis), parity, duration of infertility and post-wash motile sperm count.

Per-cycles outcomes were evaluated using GEE, with adjustments for aforementioned covariates, as well as important cycle parameters including ovarian stimulation, endometrial thickness and number of large follicles.

All calculations were performed with the use of SPSS 22 (IBM, Armonk, NY, USA).

A total of 6,407 patients were reviewed, contributing to 13,745 cycles, 4,769 of these in natural cycles and 8,976 in stimulated cycles. Of these, 2,220 (16.15%) of the cycles involved underweight women, and 1,756 (12.78%) involved overweight women. The majority (71.07%) of women in the study were of normal weight. Demographic characteristics are summarized in Table 1. Compared with normal-weight patients, underweight patients were characterized by younger age, whereas the overweight patient demographics were biased toward advanced age. The mean baseline FSH decreased with increasing BMI from 7.56 ± 2.43 IU/L in underweight patients to 6.32 ± 1.73 IU/L in the highest BMI categories. There was an increased incidence of PCOS in overweight group. A higher incidence of endometriosis was observed in the underweight group. Overweight women had a significantly higher infertility duration compared to underweight individuals and those with BMI < 18.5kg/m². Although there was no difference in the total motile sperm count categories among groups, parity was slightly higher in the overweight group.

Table 1

Baseline characteristics
	Normal weight	Under weight	Overweight
Variable	(18.5-24.9kg/m²; n = 4,563)	(<18.5kg/m²; n = 990)	(25.0-29.9kg/m²; n = 854)	Underweight vs. Normal weight	Overweight vs. Normal weight	Overweight vs. Underweight
Age, y	30.22 ± 3.81	29.44 ± 3.41	30.51 ± 4.15	< .001	.040	< .001
Basal FSH, IU/L	7.00 ± 2.1	7.56 ± 2.43	6.32 ± 1.73	< .001	< .001	< .001
Basal LH, IU/L	5.70 ± 3.87	6.04 ± 3.9	5.15 ± 3.44	.011	< .001	< .001
Basal AFC	13.17 ± 12.49	12.02 ± 5.65	14.15 ± 6.42	.003	.018	< .001
BMI, kg/m²	21.05 ± 1.45	17.61 ± 0.71	25.32 ± 1.74	< .001	< .001	< .001
Additional etiologies
PCOS (%)	860/4563(18.85)	102/990(10.30)	224/854(26.23)	< .001	< .001	< .001
Endometriosis (%)	350/4563(7.67)	115/990(11.62)	51/854(5.97)	< .001	.082	< .001
Unilateral tubal obstruction (%)	738/4563(16.17)	153/990(15.45)	127/854(14.87)	.576	.340	.728
Duration of infertility, y				.758	.030	.008
< 2	1044/4563(22.88)	231/990(23.33)	156/854(18.27)
≥ 2	3519/4563(77.12)	759/990(76.67)	698/854(81.73)
Parity				<.001	< .001	.001
0	4094/4563(89.72)	922/990(93.13)	734/854(85.95)
≥ 1	469/4563(10.28)	68/990(6.86)	120/854(14.05)
Postwash TMC, 10⁶				.844	.086	.094
≤ 5.0	43/4563(0.94)	10/990(1.01)	7/854(0.82)
5.1–10.0	1392/4563(30.51)	286/990(28.89)	269/854(31.50)
10.1–15.0	1404/4563(30.77)	304/990(30.71)	291/854(34.07)
15.1–20.0	1046/4563(22.92)	240/990(24.24)	187/854(21.90)
> 20.0	678/4563(14.86)	150/990(15.15)	100/854(11.71)
Note: Unless otherwise specified, results are presented as mean ± SD. FSH: follicle-stimulating hormone; LH: luteinizing hormone; AFC: antral follicle count; BMI: body mass index; PCOS: polycystic ovarian syndrome; TMC: total motile sperm count.

After four IUI cycles, the conservative cumulative pregnancy rate was 25.92% (95%CI: 24.9–27.0) and the conservative cumulative live birth rate was 21.59% (95%CI: 20.6–22.6). The cumulative outcomes calculated according to BMI categories were also shown in Table 2. The conservative cumulative live birth rate in normal weight, underweight and overweight patients was 21.61% (95%CI: 20.4–22.8), 17.17% (95%CI: 14.9–19.5) and 26.58% (95%CI: 23.7–29.6), respectively. The BMI-specific live birth rates were significantly different from one another, according to log-rank tests (P < 0.001).

According to Cox regression (Table 3), underweight is associated with lower cumulative pregnancy and live birth rate while overweight is associated with higher cumulative pregnancy and live birth rate, with adjustment of age, ovarian reserved markers, parity, etiologies and total motile sperm count categories.

Table 3

Effect of patients characteristics on cumulative live births based on Cox regression
		HR (95%CI)
Variable	Category	cumulative pregnancy	cumulative live birth
BMI	18.5-24.9	Ref	Ref
	<18.5	0.85(0.73-0.98)	0.80(0.67-0.95)
	25.0-29.9	1.19(1.04-1.36)	1.19(1.02-1.38)
Age	Per year increased	0.96(0.95-0.98)	0.95(0.93-0.96)
Basal FSH	Per mIU increased	0.93(0.90-0.96)	0.94(0.91-0.96)
Basal LH	Per mIU increased	1.02(1.01-1.03)	1.02(1.00-1.03)
Basal AFC	Per AFC increased	1.00(1.00-1.01)	1.00(1.00-1.01)
PCOS	With vs. without	1.75(1.55-1.98)	1.73(1.51-1.98)
Endometriosis	With vs. without	0.65(0.52-0.82)	0.62(0.48-0.80)
Unilateral tubal obstruction	With vs. without	0.76(0.65-0.88)	0.79(0.67-0.93)
Parity	0	Ref	Ref
	≥1	0.99(0.82-1.19)	0.97(0.79-1.19)
Duration of infertility	<2	Ref	Ref
	≥2	0.84(0.75-0.94)	0.87(0.77-0.99)
Postwash TMC	≤5.0	Ref	Ref
	5.1–10.0	1.07(0.65-1.77)	1.00(0.59-1.71)
	10.1–15.0	1.07(0.65-1.77)	1.03(0.61-1.76)
	15.1–20.0	1.00(0.61-1.66)	0.94(0.55-1.61)
	>20.0	1.06(0.64-1.76)	0.97(0.56-1.67)
Abbreviations: BMI: body mass index; FSH: follicle-stimulating hormone; LH: luteinizing hormone; AFC: antral follicle count; PCOS: polycystic ovarian syndrome; TMC: total motile sperm count.

BMI was also associated with per-cycle pregnancies and live births according to GEE analyses (Table 4). Adjusted for potential confounding factors, underweight was associated with reduced clinical pregnancy rates (OR 0.77, 95%CI: 0.64–0.92). However, the clinical pregnancy rates of overweight significantly increased (OR 1.24, 95%CI: 1.05–1.46). The other variables with an independent effect on the rates of clinical pregnancy in the multivariate models were age, basal FSH, basal LH, ovarian stimulation, duration of infertility, diameter of dominant follicles, and endometrial thickness. It was found that ovarian stimulation increased the clinical pregnancy rates 1.68-fold (95%CI: 1.43–1.97) compared with natural cycles. The probability of clinical pregnancy for an endometrial thickness < 7 mm was significantly lower compared with cases with endometrial thickness > 8mm when adjustment for potential confounding factors (P < 0.01).

Table 4

GEE analysis of per-cycle live births
Variable	Category	OR (95% CI)
BMI	18.5–24.9	Ref
	< 18.5	0.77(0.64–0.92)
	25.0-29.9	1.24(1.05–1.46)
Age	Per year increased	0.95(0.93–0.97)
Basal FSH	Per mIU increased	0.92(0.88–0.95)
Basal LH	Per mIU increased	1.04(1.02–1.05)
Basal AFC	Per AFC increased	1.01(0.98–1.04)
Parity	0	Ref
	≥ 1	1.04(0.84–1.28)
Duration of infertility	< 2	Ref
	≥ 2	0.85(0.74–0.97)
Postwash TMC	≤ 5.0	Ref
	5.1–10.0	1.56(0.74–3.29)
	10.1–15.0	1.80(0.86–3.79)
	15.1–20.0	1.60(0.76–3.38)
	> 20.0	1.64(0.77–3.48)
Type of cycles	Natural cycles	Ref
	Stimulated cycles	1.68(1.43–1.97)
Endometrial thickness	< 7	Ref
	7–8	1.21(0.99–1.48)
	8–9	1.29(1.06–1.57)
	9–10	1.53(1.26–1.87)
	10–11	1.39(1.11–1.74)
	11–12	1.74(1.35–2.24)
	≥ 12	1.67(1.30–2.14)
Follicle	post ovulation	Ref
	< 17	1.36(1.02–1.81)
	17–18	1.54(1.18–2.01)
	18–19	1.52(1.18–1.95)
	19–20	1.57(1.22–2.03)
	≥ 20	1.44(1.13–1.85)
Order of IUI cycles	1	Ref
	2	0.90(0.80–1.03)
	> 2	0.84(0.72–0.98)
Abbreviations: BMI: body mass index; FSH: follicle-stimulating hormone; LH: luteinizing hormone; AFC: antral follicle count; TMC: total motile sperm count.

In this retrospective study, we found that BMI is associated with cumulative live birth rate following a maximum of 4 IUI cycles as well as per-cycle live birth outcome in a cohort of 6,407 women contributing to 13,745 cycles. The data demonstrated that underweight women might have lower chance of success in IUI treatments in comparison with normal weight women while overweight women may have higher chance of that, suggesting a positive association between BMI and IUI outcomes.

Our results are largely in accordance with the results described by Wang et al., which demonstrated a significant increase in fecundity from underweight to obese women [11]. In a cohort of 477 women undergoing 1,189 ovulation induction IUI cycles, Souter et al. also demonstrated that the patients with higher BMI (≥ 24kg/m²) were associated with higher odds of achieving a pregnancy and live birth compared with the patients with lowest BMI (< 24 kg/m²) [14]. More recently, Immediata et al. also observed a positive association between BMI and live birth following IUI cycles in a cohort of 2,901 patients by univariate analyses. However, BMI was not included in their multivariate model, and they finally concluded that IUI outcomes “were found to be significantly correlated with female age and FSH levels”. Our study further supported the existing evidence by reporting cumulative live birth in a larger IUI cohort categorized according to WHO BMI criteria.

Several other studies did report a lack of association between BMI and IUI outcomes [9, 10]. However, as mentioned by the authors, the small number of patients with extreme BMI may makes it difficult to draw strong conclusions [10].

Our study conflicts with the work of Aydin et al. [8], in which BMI is negatively associated with pregnancies. The difference may explain by distribution of BMI in the populations. In Aydin’s work, the mean BMI of non-pregnant women was as high as 27.85 kg/m², suggesting a large portion of obese patients (BMI > 30 kg/m²) in their cohort. Due to the lack of patients with BMI > 30 kg/m² in our cohort, we are unable to estimate the effect of obesity. However, other authors including obese women in their analyses still demonstrated conflicting results [7, 14, 15], calling for further studies on the effect of obesity.

BMI may affect the IUI outcomes in several aspects. While BMI is negatively associated with hormone levels and follicular growth [16], Souter et al. suggested that BMI was positively associated with endometrial thickness, suggesting that higher BMI may be a favorable factor for endometrial receptivity as soon as “the medication and response are adjusted to overcome the weight effect” [14]. Therefore, heterogeneity in medication and the patients’ response may contribute to the difference in the effect of BMI shown by previous studies. Our per-cycle multivariate analyses showed that the effect of BMI was not confounded or mediated by either ovarian stimulation or follicular growth, supporting a possible role in endometrial receptivity.

Our study is strengthened by a relatively large sample size, but is limited by its retrospective nature. Selection bias exists because women with high BMIs maybe recommend to lose weight before treatment. This bias may skew the estimation of the higher end of BMI. On the hand, the reasons of low BMI in the population is not well documented. A low BMI may be simply due to reduced food intake, but may also suggest chronic conditions. Unreported conditions may contribute to unknown confounding factors in this retrospective study.

In conclusion, our study suggested that BMI is positively associated with cumulative IUI outcomes in women with a BMI < 30 kg/m². Especially, underweight women may have a lower chance of getting pregnancy in comparison with normal weight and overweight women. With increasing concerns of the health effect obesity, several authors just ignore underweight women [7] or pooled them with women with normal BMI [14]. However, our data argue that underweight should be considered as well as obesity, when giving consult to patients receiving IUI treatments.

IVF: in vitro fertilization; BMI: body mass index; IUI: intrauterine insemination; FSH: follicle-stimulating hormone; LH: luteinizing hormone; AFC: antral follicle count; PCOS: polycystic ovarian syndrome; TMC: total motile sperm count; GEE: generalized estimating equation;

Acknowledgements

The authors thank all of the staff, nurses, and physicians at the Reproductive Medicine Center for their support in generating this manuscript.

Authors’ contributions

The study was designed by JZ and JC; and conducted by JZ, JC, YG, JS. Data analysis was by JC; and the manuscript was written by JZ and JC. All authors contributed to the analysis and interpretation of the data, and reviewed and approved of the final version of the manuscript.

Funding

This work was funded by the Clinical Research Special Fund of Chinese Medical Association (NO.18010360765) and Xiamen Medical Advantage Subspecialty Construction Project (2018296).

Availability of data and materials

Not applicable.

Ethics approval and consent to participate

Institutional Review Board approval was obtained from the Ethics Committee of the Medical College of Xiamen University.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Cohlen B, Bijkerk A, Van der Poel S, Ombelet W. IUI: review and systematic assessment of the evidence that supports global recommendations. Hum Reprod Update. 2018;24(3):300–19.
Tjon-Kon-Fat RI, Bensdorp AJ, Bossuyt PM, Koks C, Oosterhuis GJ, Hoek A, et al. Is IVF-served two different ways-more cost-effective than IUI with controlled ovarian hyperstimulation? Hum Reprod. 2015;30(10):2331–9.
De Geyter C, Calhaz-Jorge C, Kupka MS, Wyns C, Mocanu E, Motrenko T, et al. ART in Europe, 2015: results generated from European registries by ESHRE. 2020;2020(1):hoz038.
Zhou Z, Chen L, Wu H, Zheng D, Li R, Mol BW, et al. Assisted reproductive technology in Beijing, 2013–2015. Reprod Biomed Online. 2018;37(5):521–32.
Sermondade N, Huberlant S, Bourhis-Lefebvre V, Arbo E, Gallot V, Colombani M, et al. Female obesity is negatively associated with live birth rate following IVF: a systematic review and meta-analysis. Hum Reprod Update. 2019;25(4):439–51.
Whynott RM, Summers KM, Van Voorhis BJ, Mejia RB. Effect of body mass index on intrauterine insemination cycle success. Fertil Steril. 2021;115(1):221–8.
Aydin Y, Hassa H, Oge T, Tokgoz VY. Factors predictive of clinical pregnancy in the first intrauterine insemination cycle of 306 couples with favourable female patient characteristics. Hum Fertil (Camb). 2013;16(4):286–90.
Isa AM, Abu-Rafea B, Alasiri SA, Binsaleh S, Ismail KH, Vilos GA. Age, body mass index, and number of previous trials: are they prognosticators of intra-uterine-insemination for infertility treatment? Int J Fertil Steril. 2014;8(3):255–60.
Huyghe S, Verest A, Thijssen A, Ombelet W. Influence of BMI and smoking on IUI outcome with partner and donor sperm. Facts Views Vis Obgyn. 2017;9(2):93–100.
Wang JX, Warnes GW, Davies MJ, Norman RJ. Overweight infertile patients have a higher fecundity than normal-weight women undergoing controlled ovarian hyperstimulation with intrauterine insemination. Fertil Steril. 2004;81(6):1710–2.
McKnight KK, Nodler JL, Cooper JJ, Jr., Chapman VR, Cliver SP, Bates GW, Jr. Body mass index-associated differences in response to ovulation induction with letrozole. Fertil Steril. 2011;96(5):1206–8.
Malizia BA, Hacker MR, Penzias AS. Cumulative live-birth rates after in vitro fertilization. N Engl J Med. 2009;360(3):236–43.
Souter I, Baltagi LM, Kuleta D, Meeker JD, Petrozza JC. Women, weight, and fertility: the effect of body mass index on the outcome of superovulation/intrauterine insemination cycles. Fertil Steril. 2011;95(3):1042–7.
Na L, Chen Y, Zhai H, Liao A, Huang D. Effects of maternal body mass index on pregnancy outcome after 8570 artificial insemination cycles with donor's sperm. Gynecol Endocrinol. 2018;34(12):1068–72.
Wang LT, Wang CX, Sun HL, Wang X, Li XF, Wang YL, et al. Effect of BMI on blood value of patients on HCG day with IUI treatment. BMC Womens Health. 2020;20(1):105.

Table 2 is available in the Supplementary Files section.

No competing interests reported.

Table2.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Low BMI is associated with poor IUI outcomes: a retrospective study in 13,745 cycles

Status:

Version 1

Abstract

Introduction

Materials And Methods

Study population

IUI Protocol

Statistical analysis

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Version 1