Gynecological Endocrinology
ISSN: 0951-3590 (Print) 1473-0766 (Online) Journal homepage: http://www.tandfonline.com/loi/igye20
Optimizing the first-line fertility treatment
Mohamad Irani, Stephen Chow, Derek Keating, Simone Elder, Zev
Rosenwaks & Gianpiero Palermo
To cite this article: Mohamad Irani, Stephen Chow, Derek Keating, Simone Elder, Zev
Rosenwaks & Gianpiero Palermo (2018): Optimizing the first-line fertility treatment, Gynecological
Endocrinology, DOI: 10.1080/09513590.2018.1441825
To link to this article: https://doi.org/10.1080/09513590.2018.1441825
Published online: 20 Feb 2018.
Submit your article to this journal
Article views: 21
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=igye20
GYNECOLOGICAL ENDOCRINOLOGY, 2018
https://doi.org/10.1080/09513590.2018.1441825
ORIGINAL ARTICLE
Optimizing the first-line fertility treatment
Mohamad Irania, Stephen Chowa, Derek Keatinga, Simone Elderb, Zev Rosenwaksa and Gianpiero Palermoa
a
The Ronald O. Perelman and Claudia Cohen Center for Reproductive Medicine, Weill Cornell Medicine, New York, NY, USA; bWeill Cornell
Medicine, New York, NY, USA
ABSTRACT ARTICLE HISTORY
The objective of this study was to identify sperm score thresholds to achieve satisfactory intrauterine Received 28 November 2017
insemination (IUI) success rates according to the response to stimulation with clomiphene citrate (CC). To Revised 20 January 2018
minimize the confounding effect of female age, we included only CC/IUI cycles of women 35 years old. Accepted 14 February 2018
A total of 1,194 CC/IUI cycles were included. Semen volume, concentration, and motility influenced the Published online 20 February
2018
clinical pregnancy rate (CPR). Normal morphology (4%) was associated with a comparable CPR with 3%,
2%, and 1% normal forms (15.6%, 16.1%, 18.1%, and 13.1%, respectively). A combination of the total num- KEYWORDS
ber of motile spermatozoa in the ejaculate before semen preparation (TM) at a threshold 20 106 was Clomiphene citrate;
associated with a CPR of 17.8% compared to 4.6% for a threshold <20 106 (p < .001). Interestingly, the endometrial thickness;
TM threshold to achieve satisfactory outcomes was lower (10 106) in patients who had an optimal follicle; intrauterine
response to CC (2 dominant follicles with an endometrial thickness 7 mm) compared to 40 106 for insemination; total motile
those who had a suboptimal response (one dominant follicle with an endometrial thickness <7 mm). In spermatozoa
conclusion, the response to superovulation with CC determines each patient’s TM threshold required for
satisfactory outcomes. Couples whose TM is below the threshold may benefit from a superovulation with
gonadotropins or in vitro fertilization.
Introduction ovarian follicles, and female age. With this in mind, we analyzed
cycles of clomiphene citrate (CC), a cost-effective and widely
Infertility affects approximately 10–15% of couples [1,2]. The
used medication for superovulation, in young women (35 years
major causes of infertility include diminished ovarian reserve,
old) to better delineate individualized sperm score thresholds
ovulatory dysfunction, and tubal, uterine, cervical, male, and
that depend on the female patient’s response to stimulation.
idiopathic factors [3]. Intrauterine insemination (IUI) is usually
the first-line treatment for couples whose female partners have at These thresholds would guide the management of each patient
least one patent fallopian tube [4]. Despite its overall lower suc- according to her partner’s sperm score.
cess rate compared to in vitro fertilization (IVF), IUI is used
more often than IVF because it is a simple, inexpensive, and
relatively safe technique [5,6]. Methods
Although semen analysis is an imperfect tool, it remains the Cycle inclusion criteria
mainstay of assessment for male infertility [7]. In 2010, the
World Health Organization (WHO) established new guidelines The institutional review board at Weill Cornell Medicine
for semen parameters [8]. They considered the fifth percentiles approved this study. All CC/IUI cycles performed at our center
of almost 2,000 fertile men as the lower limits of the normal between January 2004 and December 2015 were reviewed for
range: volume of 1.5 ml, concentration of 15 million/mL, motility potential inclusion. To better define the role of sperm score and
of 40%, and strict normal sperm morphology of 4% [8]. It is well minimize the effect of female confounding factors, we included
established that patients with severe male factor infertility would women 35 years old who had at least one patent fallopian tube.
benefit from IVF-intracytoplasmic sperm injection (ICSI) rather
than IUI [9]. However, there are conflicting data on the role of
sperm morphology and the total number of ejaculated motile Clinical protocols
spermatozoa (TM) threshold below which ICSI would be recom-
mended to reach satisfactory success rates [10–14]. Before initiating therapy, all couples underwent a complete infer-
Several factors are related to IUI outcomes including female tility assessment. Semen analysis was performed according to
age, infertility diagnosis, superovulation protocol, and semen 2010 WHO criteria [8]. Ovarian superovulation was performed
parameters. Previous studies comparing semen parameters often with CC at a dose of 50 or 100 mg daily for five days. The
did not control for confounding factors that can influence IUI response to stimulation and endometrial thickness were moni-
outcomes. Therefore, it is critical to evaluate whether there is any tored by serial transvaginal ultrasounds. Serum hormone assays
value of normal concentration, motility, or morphology, or a were also used to measure estradiol and LH levels. In the absence
score combining all of the above, that may affect IUI outcomes, of LH surge, ovulation was triggered with 10 000 IU human
taking into consideration superovulation protocol, number of chorionic gonadotropin (hCG) when the dominant follicle(s)
CONTACT Mohamad Irani moi9010@med.cornell.edu 1305 York Avenue, 6th floor, New York, NY 10021, USA
ß 2018 Informa UK Limited, trading as Taylor & Francis Group
2 M. IRANI ET AL.
reached 20 mm. The IUI was performed within 24 h after hCG Statistical analysis
injection.
All data analyses were conducted with STATA statistical software
(StataCorp LLC, TX, USA). Continuous variables were tested for
normality; Student’s t-test was used for parametric data, which
Semen collection and sperm preparation were expressed as mean ± standard deviation. Categorical varia-
bles were compared with the Chi-square (v2) test. Odds ratios
Semen samples were collected at the laboratory after 2–5 days of (ORs) with 95% confidence intervals (CIs) were calculated and
abstinence. Semen analysis was performed after 30 min of lique- adjusted for confounders. A p value of <.05 was considered stat-
faction. The samples were first diluted in media comprised of istically significant.
HEPES-buffered human tubal fluid (H-HTF; Irvine Scientific,
CA, USA) supplemented with human serum albumin (HSA-
solutionTM G-SeriesTM culture media; Vitrolife G€ oteborg, Results
Sweden) for centrifugation at 600 g for 10 min. For each sample,
A total of 1,194 CC/IUI cycles from 628 patients were analyzed.
the pellet was then resuspended and layered on a density gradi-
The mean semen volume was 2.6 ± 1 ml, concentration was
ent (Enhance-S Plus Cell Isolation Media, 90%; Vitrolife). It was
47.8 ± 21 106/mL, motility was 48.4 ± 9%, morphology
then centrifuged for 10 min at 300 g. The bottom layer containing
was 3.6 ± 2%, and paternal age was 33 ± 5 years. Maternal age was
motile spermatozoa was collected by aspiration with a glass 31.4 ± 3 years, parity was 0.1 ± 0.3, the number of dominant fol-
Pasteur pipette and resuspended in a medium for a final 10 min licles (18 mm) was 1.7 ± 0.8, and endometrial thickness was
centrifugation at 600 g to remove silica gel particles. The final 7.8 ± 2 mm. Infertility diagnoses were distributed as follows:
pellet was resuspended in 0.5 ml of medium and used for insem- 31.7% male factor, 17.1% idiopathic, 11.3% ovulatory dysfunc-
ination after reassessing concentration and motility. tion, 7.5% diminished ovarian reserve, and 32.4% other diagnoses
(cervical factor, uterine factor, endometriosis, and mixed).
The mean number of IUI attempts was 2.1 ± 1.3.
Variables assessed Normal morphology (4%) was associated with a comparable
CPR with 3%, 2%, and 1% normal forms (15.6%, 16.1%, 18.1%,
The primary outcome of this study is clinical pregnancy rate 13.1%, respectively) (Figure 1(A)). IUI attempts made with 0%
(CPR), which is defined as the proportion of cycles resulting in normal forms (n ¼ 13) yielded no clinical pregnancies, but results
intrauterine pregnancies with fetal heartbeat detected by ultra- were not significantly different when 1% normal forms were
sound. The impact of semen volume, concentration, motility, and used. Semen volume 1 ml was associated with a higher CPR
morphology as well as TM in the ejaculate before semen prepar- compared to a volume <1 ml (16.4% vs. 6.9%; p ¼ .03)
ation, endometrial thickness on the day of trigger, and number (Figure 1(B)). A concentration 20 106/mL also yielded a
of developing follicles on CPR, was evaluated. higher CPR compared to a concentration <20 106/mL (17%
Figure 1. (A) The clinical pregnancy rate (CPR) following CC/IUI cycles was not affected by sperm morphology except for 0% normal forms, which were associated
with 0% clinical pregnancy. (B) Semen volume 1 ml in CC/IUI cycles is associated with a higher CPR compared to a volume <1 ml (16.4% vs. 6.9%; p ¼ .03).
(C) Sperm concentration 20 106/mL in CC/IUI cycles is associated with a significantly higher CPR compared to a concentration <20 106/mL (17% vs. 5.1%;
p < .001). (D) None of the CC/IUI cycles using sperm motility of <20% (n ¼ 14) achieved a clinical pregnancy compared to 16.1% in cycles using sperm motility of
>20%. CC: Clomiphene citrate; IUI: intrauterine insemination.
GYNECOLOGICAL ENDOCRINOLOGY 3
vs. 5.1%; p < .001) (Figure 1(C)). Moreover, none of the patients Of note, the rate of multiple gestations was 8.9%, which was
with motility <20% achieved a pregnancy (n ¼ 14) (Figure 1(D)). positively correlated with the number of dominant follicles
Gravidity, parity, maternal age, paternal age, infertility diagnosis, (r ¼ 0.08; p ¼ .002).
and the number of IUI attempts did not significantly affect IUI
outcomes.
A combination of TM at a threshold 20 106 was associated Discussion
with a CPR of 17.8% compared to 4.6% at a TM threshold
To better delineate the role of semen parameters and prevent the
<20 106 (p < .001) (Figure 2(A)). Couples whose female part-
confounding effects of maternal age or superovulation protocol,
ners had more than two dominant follicles on the hCG day had
we only included young women who underwent CC/IUI cycles.
a higher CPR compared to those who had one or two follicles
Considered individually, each of the semen parameters, particu-
(20.3% vs. 14.7%; p ¼ .03). Furthermore, patients with an endo-
larly volume, concentration, and motility, can influence the CC/
metrial thickness 7 mm had a significantly higher CPR com-
IUI outcome. Indeed, the impact of concentration, motility, and
pared to those with a thickness <7 mm (17.5% vs. 12.6%;
volume has been well documented [15,16]. Concerning the role
p ¼ .03; OR ¼ 1.4; 95% CI ¼ 1.1–2.0). This ratio remained signifi-
of strict morphology, a review of 4,251 IUI cycles showed that a
cant after adjusting for the number of dominant follicles and TM
(adjusted OR ¼1.4; 95% CI ¼ 1.0–2.0).
Interestingly, the TM threshold to achieve satisfactory out-
comes was lower (10 106) in patients who had an optimal
response to CC (2 dominant follicles with an endometrial
thickness 7 mm) compared to 40 106 for those who had a
suboptimal response (one dominant follicle with an endometrial
thickness <7 mm) (Figure 2(B,C)). The TM threshold was
20 106 for women who developed one follicle with an endomet-
rial thickness 7 mm (p ¼ .007), while it was 30 106 for those
who developed 2 follicles with a thin lining (endometrial thick-
ness <7 mm).
In addition, a combination of total motile inseminated sper-
matozoa at a threshold 10 106 was associated with a CPR of
Figure 3. Among all patients undergoing CC/IUI cycles, the total motile insemi-
17.1% compared to 7.5% at a TM threshold <10 106 (p ¼ .002) nated spermatozoa threshold to achieve a satisfactory outcome was 10 106
(Figure 3). (17.1% vs. 7.5%; p ¼ .002). CC: Clomiphene citrate; IUI: intrauterine insemination.
Figure 2. (A) Among all patients undergoing CC/IUI cycles, the TM threshold to achieve a satisfactory outcome was 20 106 (17.8% vs. 4.6%; p < .001). (B) Among
patients with an optimal response to CC (two or more dominant follicles with endometrial thickness 7 mm), the TM threshold to achieve a satisfactory outcome was
10 106 (17% vs. 0%; p ¼ .01). (C) Among patients with a suboptimal response to CC (one dominant follicle with endometrial thickness <7 mm), the TM threshold to
achieve a satisfactory outcome was 40 106 (15.4% vs. 5.3%; p ¼ .03). CC: Clomiphene citrate; IUI: intrauterine insemination; TM: total number of ejaculated motile
spermatozoa before semen preparation.
4 M. IRANI ET AL.
higher percentage of morphologically normal spermatozoa <7 mm), suggesting that IUI cycles should not be canceled for
(>4%) was associated with lower pregnancy rates compared thin endometrial lining.
to 4% normal forms [11]. On the other hand, several studies Our data show that patients who have an optimal response to
have suggested that morphology does not influence pregnancy CC (two or more dominant follicles with an endometrial thick-
rates after IUI [11,12]. Another retrospective study comparing ness 7 mm) require a TM of only 10 106 to maximize IUI
the outcomes of 281 IUI cycles using normal ejaculates to 70 success rates. This threshold needs to rise to 40 106 for women
cycles with isolated abnormal morphology (4%) failed to iden- who yield only a single dominant follicle and are affected by a
tify any effect of morphology on CPRs [12]. This study was lim- suboptimal endometrial lining (<7 mm). Nonetheless, couples
ited by an extremely small sample size (n ¼ 14) for 0% and 1% who cannot meet the higher TM threshold imposed by the
normal forms. impaired response to CC may benefit from either superovulation
Our study provides important data because it compares differ- with gonadotropins or IVF.
ent categories of abnormal morphology (0%, 1%, 2%, and 3%) The TM, the number of follicles, and the endometrial lining
with normal forms (4%). We have included 84 IUI cycles of all contribute to the CC/IUI outcome. In fact, the sperm score
patients with 1% normal morphology, 204 for 2%, and 428 for threshold differs according to the female patient’s response to
3%, demonstrating that a proportion of normal forms of at least superovulation. Therefore, couples with impaired superovula-
tory response (i.e. reduced follicular development or subopti-
1% should not preclude couples from undergoing IUI because of
mal endometrial lining) may require alternative superovulation
the comparable CPRs between 1% and 2% normal morphology.
protocols that allow them to benefit from a lower sperm score
Therefore, our data suggest that a compromised morphology
threshold.
does not impair the IUI outcome except when no normal sper-
matozoa are available, indicating that a suboptimal morphology
does not represent a contraindication to IUI. Hence, recom- Disclosure statement
mending ICSI as a plausible first-line treatment for patients with
0% normal forms may be reasonable. No potential conflict of interest was reported by the authors.
On the basis of the limited clarity related to individual semen
parameters, we observed that a score combining volume, concen-
tration, and motility is able to predict IUI success rates with a
References
first-line superovulation protocol such as CC. 1. Hull MG, Glazener CM, Kelly NJ, et al. Population study of causes,
Several studies have reported a correlation between the TM treatment, and outcome of infertility. Br Med J (Clin Res Ed)
1985;291:1693–7.
used for IUI and the pregnancy rate [13,14,17]. Badawy et al. 2. Snick HK, Snick TS, Evers JL, et al. The spontaneous pregnancy prog-
[13] showed an improvement in the CPR after inseminating at nosis in untreated subfertile couples: the Walcheren primary care
least 5 106 (24.2% vs. 5.5%). Cao et al. [14] then reported a study. Human Reproduction (Oxford, England) 1997;12:1582–8.
lower threshold of TM inseminated (2 106) to reach a satisfying 3. Forti G, Krausz C. Clinical review 100: evaluation and treatment of the
infertile couple. J Clin Endocrinol Metab 1998;83:4177–88.
pregnancy rate (14.5% vs. 4.0%). Similarly, Dinelli et al. [17] 4. Macaluso M, Wright-Schnapp TJ, Chandra A, et al. A public health
showed that inseminating with >1 106 yielded a higher preg- focus on infertility prevention, detection, and management. Fertility
nancy rate compared to <1 106. In line with the literature, our and Sterility 2010;93:16 e1–e10.
data highlight that the CPR significantly increases when the TM 5. Ombelet W, Cooke I, Dyer S, et al. Infertility and the provision of
infertility medical services in developing countries. Human
is at least 20 106. Of interest, we also found that the CPR pla- Reproduction Update 2008;14:605–21.
teaus after satisfying this threshold. 6. European IVFMC, European Society of Human R, Embryology, et al.
In addition to sperm characteristics, several elements affect Assisted reproductive technology in Europe, 2011: results generated
IUI outcome in young women, such as the type of superovula- from European registers by ESHRE. Human Reproduction (Oxford,
England) 2016;31:233–48.
tion protocol and the response to the particular ovarian 7. Barratt CL. Semen analysis is the cornerstone of investigation for male
stimulation. infertility. Practitioner 2007;251:8–10, 12, 15–7.
With regard to the response to superovulation, our findings 8. World Health Organization. WHO laboratory manual for the examin-
are in agreement with the literature, which indicates that a ation and processing of human semen. 5th ed., 2010. Available from:
http://www.who.int/reproductivehealth/publications/infertility/97892415
greater number of dominant follicles is associated with a higher 47789/en
pregnancy rate [17,18]. The impact of endometrial thickness on 9. Tournaye HJ, Cohlen BJ. Management of male-factor infertility. Best
CC/IUI outcomes has also been previously investigated Pract Res Clin Obstet Gynaecol 2012;26:769–75.
[17,19–22]. For instance, Warrington et al. [21] showed that a 10. Lee RK, Hou JW, Ho HY, et al. Sperm morphology analysis using
strict criteria as a prognostic factor in intrauterine insemination. Int J
thicker endometrium is associated with a higher pregnancy rate. Androl 2002;25:277–80.
Similarly, Esmailzadeh et al. [19] reported that a thicker endo- 11. Lemmens L, Kos S, Beijer C, et al. Predictive value of sperm morph-
metrium on the day of ovulation trigger resulted in a superior ology and progressively motile sperm count for pregnancy outcomes in
IUI pregnancy rate. Furthermore, Wolff et al. [22] demonstrated intrauterine insemination. Fertil Steril 2016;105:1462–8.
12. Lockwood GM, Deveneau NE, Shridharani AN, et al. Isolated abnor-
that the pregnancy rate rose with increasing endometrial thick- mal strict morphology is not a contraindication for intrauterine insem-
ness. However, Seckin et al. [20] showed comparable pregnancy ination. Andrology 2015;3:1088–93.
rates between patients with different endometrial thicknesses. 13. Badawy A, Elnashar A, Eltotongy M. Effect of sperm morphology and
Our data indicated that an endometrial thickness 7 mm was number on success of intrauterine insemination. Fertil Steril
2009;91:777–81.
associated with 1.4–fold higher CPR compared to a thickness 14. Cao S, Zhao C, Zhang J, et al. A minimum number of motile sperma-
<7 mm (17.5% vs. 12.6%; p ¼ .03). In contrast to the majority of tozoa are required for successful fertilisation through artificial intra-
other studies, we have controlled for several variables to confirm uterine insemination with husband's spermatozoa. Andrologia
2014;46:529–34.
this association (adjusted OR ¼1.4). Yet, despite the difference in 15. Shulman A, Hauser R, Lipitz S, et al. Sperm motility is a major deter-
outcome between the two groups, the CPR was still acceptable minant of pregnancy outcome following intrauterine insemination.
for those with thin lining (12.6% for an endometrial thickness J Assist Reprod Genet 1998;15:381–5.
GYNECOLOGICAL ENDOCRINOLOGY 5
16. Dorjpurev U, Kuwahara A, Yano Y, et al. Effect of semen characteris- 20. Seckin B, Pekcan MK, Bostanci EI, et al. Comparison of pregnancy
tics on pregnancy rate following intrauterine insemination. J Med rates in PCOS patients undergoing clomiphene citrate and IUI treat-
Invest 2011;58:127–33. ment with different leading follicular sizes. Arch Gynecol Obstet
17. Dinelli L, Courbiere B, Achard V, et al. Prognosis factors of 2016;293:901–6.
pregnancy after intrauterine insemination with the husband's sperm: con- 21. Warrington C, Faraj R, Willett M. Endometrial thickness and outcome
clusions of an analysis of 2,019 cycles. Fertil Steril 2014;101:994–1000. in sub-fertile women treated with clomiphene citrate. J Obstet
18. Van Waart J, Kruger TF, Lombard CJ, et al. Predictive value of normal Gynaecol 2008;28:626–8.
sperm morphology in intrauterine insemination (IUI): a structured lit- 22. Wolff EF, Vahidi N, Alford C, et al. Influences on endometrial
erature review. Hum Reprod Update 2001;7:495–500. development during intrauterine insemination: clinical experience of
19. Esmailzadeh S, Faramarzi M. Endometrial thickness and pregnancy 2,929 patients with unexplained infertility. Fertil Steril 2013;
outcome after intrauterine insemination. Fertil Steril 2007;88:432–7. 100:194–9e1.
