Is paternal age associated with transfer day, developmental stage, morphology, and initial hCG-rise of the competent blastocyst leading to live birth? A multicenter cohort study.

In this study we investigated whether age of men undergoing assisted reproductive technology (ART) treatment was associated with day of transfer, stage, morphology, and initial hCG-rise of the competent blastocyst leading to a live birth? The design was a multicenter historical cohort study based on exposure (age) and outcome data (blastocyst stage and morphology and initial hCG-rise) from men whose partner underwent single blastocyst transfer resulting in singleton pregnancy/birth. The ART treatments were carried out at sixteen private and university-based public fertility clinics. We included 7246 men and women, who between 2014 and 2018 underwent controlled ovarian stimulation (COS) or Frozen-thawed Embryo Transfer (FET) with a single blastocyst transfer resulting in singleton pregnancy were identified. 4842 men with a partner giving birth were included, by linking data to the Danish Medical Birth Registry. We showed that the adjusted association between paternal age and transfer day in COS treatments was OR 1.06, 95% CI (1.00;1.13). Meaning that for every increase of one year, men had a 6% increased probability that the competent blastocyst was transferred on day 6 compared to day 5. Further we showed that the mean difference in hCG values when comparing paternal age group 30-34, 35-39 and 40-45 with the age group 25-29 in those receiving COS treatment, all showed significantly lower adjusted values for older men. In conclusion we hypothesize that the later transfer (day 6) in female partners of older men may be due to longer time spent by the oocyte to repair fragmented DNA of the sperm cells, which should be a focus of future research in men.

Introduction

Delayed parenthood is a growing trend worldwide with a higher age at conception as a consequence. Until recently, most research within the area of reproductive health has focused on women. However, there is now a global call in reproductive medicine for research on men. The probability of giving birth to a child, by natural conception or after assisted reproductive technology (ART) treatment, depends highly on the mother’s age [1,2], but also the father’s. Looking at national statistics, the mean age of fathers in Denmark has increased from 30.9 years in 1986 to 33.5 years in 2020 [3], while the mean age of fathers in the US has increased from 27.4 years in 1972 to 30.9 years in 2015 [4]. Advanced paternal age has been associated with reduced ability to conceive naturally [5] as well as by ART treatment. Following in vitro fertilization (IVF), reduced pregnancy- and live birth rates have been reported when the male partner is 40 years or older [6,7]. However, a recent review and meta-analysis based on the donor oocyte model suggests that advanced paternal age does not exert an independent effect on the outcome of ART cycles [8], while larger meta-analysis on naturally conceived pregnancies and miscarriage suggests a small independent effect of paternal age along with that of female age [9]. No association between blastocyst ploidy and paternal age has been found [10,11].

As soon as the sperm has fertilized the oocyte, the embryo development begins. This development is characterized by a collaboration between the sperm and the oocyte and the embryonic genome activation is under both male and female influence [12,13]. At blastulation, the cells differentiate into inner cell mass (ICM) and trophectoderm (TE) and the stage of a full blastocyst is reached on day 5 or 6 after fertilization. When a competent blastocyst meets a receptive endometrium and implants, a positive serum human chorionic gonadotropin (hCG) test will confirm the pregnancy, and a live birth will follow. In assisted reproductive technology (ART) treatment, cleavage stage embryo transfer at day 2 or 3 after fertilization has previously been practiced, while transfer at the blastocyst stage is now widely used, as the prevalence of aneuploidy in blastocysts is lower than in cleavage stage embryos [14,15]. To ensure selection of the best blastocyst to transfer, the timing, stage of development, and morphology are used as markers for competence. Worldwide, routine assessment involves grading of the development stage, the ICM, and TE as defined in the Istanbul consensus [16] and by Gardner and Schoolcraft [17]. The focus of the present study is the characteristics of the competent in vitro cultured blastocyst leading to a live birth in relation to the age of men.

The research in men within reproductive health is scarce. Furthermore, the age of men undergoing ART is increasing. For this reason, the potential effect of paternal age is important to investigate. We have recently shown that for every one-year increase in women’s age there was a 5% reduced probability that the competent blastocyst was in an advanced development stage at the time of transfer. Likewise, we have shown that the initial hCG-rise was associated with women’s age with the youngest women having the lowest hCG levels [18]. It is unknown whether the age of men undergoing ART treatment is associated with the timing (day 5 or 6), the development stage, morphology, and the early implantation of the competent blastocyst. The aim of this study was to explore whether paternal age is associated with day of transfer, developmental stage, morphology, and initial hCG-rise of the competent blastocyst leading to a live birth.

Materials and methods

Study design and participants

The study was a historical multicenter cohort study with an analysis of prospectively collected data from men and women undergoing ART treatment. Inclusion criteria were all controlled ovarian stimulation (COS) followed by fresh single blastocyst transfer with IVF or intracytoplasmatic sperm injection (ICSI) and Frozen-thawed Embryo Transfer (FET) cycles with Single Embryo Transfer (SET) resulting in a single implantation and singleton live birth. Exclusion criteria were treatments using oocyte and semen donation, or pre-implantation genetic testing. To ensure independence of observations in the dataset, treatments resulting in twin births were excluded. Further, for treatments resulting in the birth of two or more children (COS + FET) during the period of 2014–2018, only the first birth was included. Sixteen private and public Danish fertility clinics contributed with data on treatments resulting in pregnancy, from the database: Danish Medical Data Center (DMDC). Information on live birth was obtained from the Danish Medical Birth Register (DMBR) [19]. Using the woman’s unique personal identification number (CPR), data from DMDC and DMBR were linked. The same dataset was used in our cohort study of the association between female age and the competent blastocyst [18]. The study is reported according to; Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) [20].

Treatment regimen

COS were performed with either long agonist or short antagonist protocols with either recombinant follicle-stimulating hormone or menotropin. When the leading two follicles reached 17-18mm final maturation and ovulation were triggered with hCG or gonadotropin-releasing hormone agonist administration. Oocyte pick-up (OPU) was performed 34–36 hours later. Women with regular menstrual cycles underwent modified natural FET cycles using hCG to induce ovulation and blastocyst transfer after 7 days. Other women underwent substituted FET cycles using estrogen to ensure a suitable endometrial thickness prepared for supplementation with progesterone and blastocyst transfer after 5–6 days. Within clinics, the in vitro procedures as well as the administration of progesterone and the hCG analysis were identical across age groups.

Laboratory procedures

All centers used the blastocyst Gardner scoring system [17] as described in the Istanbul consensus, 2011 [16]. The laboratories at the fertility clinics were all following state of art techniques for IVF, ICSI, culture, evaluation and cryopreservation and are certified EU tissue directive centers. The primary reason to culture to day 6 before transfer or cryopreservation (main practice) was that preferred usable stage or morphology were not reached day 5 (i.e. if stage: <3 or TE

Study variables

Exposure

The age of every man at the date for OPU (age in whole years), was obtained from the clinical database DMDC and treated both as a continuous and a categorized variable. Six categories were used (21–24; 25–29; 30–34; 35–39; 40–45, ≥46 (maximum 70 years).

Outcomes

Information on timing, development stage, morphology and initial hCG-rise was obtained from the database DMDC. For COS, the evaluation of the blastocyst was done before transfer (day 5 or day 6 after fertilization) and for FET the evaluation was done before cryopreservation (day 5 or day 6 after fertilization). Further, the stage of the blastocyst, defined as the state of expansion and hatching at the day of blastocyst transfer was categorized as stage 1–6; 1: Early blastocyst (blastocoele less than half the size of the embryo); 2: Blastocyst (blastocoele half the size of the embryo or more); 3: Full blastocyst (blastocoele fills the entire embryo); 4: Expanded blastocyst (blastocoele larger than the full blastocyst and with a diminishing zona); 5: Hatching blastocyst (TE starts to break through the zona); 6: Fully hatched blastocyst (blastocyst has left the zona). Morphology parameters for blastocysts with a stage score between 3–6, including an assessment of the ICM and TE categorized as A, B or C. ICM was evaluated as: A: Many cells and closely packed; B: Several cells and clustered loosely; C: Limited number of cells. Looking at TE, these were assessed as: A: Many cells creating an organized epithelium; B: Few cells establishing a loose epithelium; C: Limited large cells [17]. In addition, the stage and morphological parameters were categorized as; Score Group 1 (highest): 6AA, 6BA, 5AA, 5BA, 4AA, 4BA, Score Group 2: 6AB, 6BB, 6CB, 6CA, 5AB, 5BB, 5CB, 5CA, 4AB, 4BB, 4CB, 4CA and Score Group 3 (lowest): 6AC, 6BC, 6CC, 5AC, 5BC, 5CC, 4AC, 4BC, 4CC, 3AA, 3AB, 3BA, 3AC, 3CA, 3BB, 3BC, 3CB, 3CC and early blastocyst. The grouping was based primarily upon high score in stage and TE as suggested in several studies [21,22]. To ensure uniformity regarding the evaluation of the blastocysts in the participating clinics, the embryologists receive annual internal training. The multivariable analyses were adjusted for clinic, as some variation may exist between IVF laboratories.

The hCG level at the time of implantation was measured as a continuous variable and defined as the initial serum hCG value (IU/L). As hCG was measured at different time points, all values of hCG were standardized to 11 days after SET. The standardization was performed by creating a linear regression analysis with day of measure as independent variable, and log hCG as dependent variable. Using estimates from the regression analysis on expected change of hCG values, each value was transformed to an expected hCG at day 11.

Covariates

Using information from the database DMDC and DMBR [19], the following covariates were included in the statistical models; Female age on the day of OPU (years, continuous); Female cigarette smoking categorized as: non-smoker, quit smoking during pregnancy, and smoker; Female Body Mass Index (BMI) (kg/m2, continuous); diagnosis categorized as: anovulation, tube factor, endometriosis, male factor, unexplained, and other; and clinic categorized as 1–16.

Statistical methods

Missing information

Only competent blastocysts resulting in live births were included in the cohort Incomplete cases were removed from the analyses.

Data analyses

Continuous and categorical variables were compared by t-test or one-way ANOVA, respectively. The association between paternal age and transfer day/cryopreservation day was tested with logistic regression. The association between paternal age and morphology was tested with ordinal logistic regression. For these analyses we recoded the morphology parameters of the TE and ICM and category C was used as reference. The association between paternal age and the hCG level at the time of implantation was tested with linear regression and multivariable linear regression. All analyses for paternal age were stratified by type of treatment (COS or FET). We present results from both crude and adjusted analyses, adjusting for female age, female BMI, female smoking, diagnosis and clinic [23,24]. Interaction between male age and female age (both continuous) was tested by including the interaction term male age * female age in the adjusted analyses of COS and FET combined.

The covariates chosen for adjustment were similar to those in our previous paper [18]. It could be argued that male BMI and male smoking would be obvious for adjustment, but male BMI and male smoking had a relative high number of missing values in our dataset (BMI 41%, smoking 24%). To test the sensitivity, we exchanged female BMI and female smoking for male BMI and male smoking in the adjusted analysis, since adding male BMI and male smoking to the list of covariates would potentially have given us problems with collinearity. Further sensitivity analyses were performed with inclusion of total FSH dose, fertilization method, and sex of the child for COS treatments and type of FET treatment for FET treatments. We performed other sensitivity analyzes by testing the day 5 blastocysts exclusively, we stratified the dataset further into IVF and ICSI and we tested the sensitivity by removing blastocysts in FET which were first transfers.

To investigate the assumptions for ordinal logistic regression each analysis was divided into several binary logistic regressions, representing each level of the dependent variable, including lower levels against higher levels. The odds ratios from these models and their 95% confidence intervals were compared to the odds ratios from the ordinal logistic regression model. For linear regression analyses, assumptions were assessed using residuals plots. For all multivariable analyses’, collinearity was investigated. All statistical tests were two-sided and P-values <0.05 were considered statistically significant. Data analysis was performed with the statistical software R, version 3.4.1 [25].

Ethical approval

Due to the historical design of the study, written informed consents were not possible to obtain from the participants. Accordingly, an ethical approval was given by the Danish Patient Safety Authority (Project number: 3-3013-2604/1). Further an ethical approval for this study was given by the Danish Data Protection Agency, Capital Region, Denmark (j.nr.: VD-2018-282, I-Suite nr.: 6522). The Danish Health Data Authority (FSEID-00003760) accepted and approved access to the DMBR and to the research platform where the statistics were carried out. To adhere with GDPR rules and regulations data were pseudonymous, which means that data were not directly personally identifiable (i.e. the CPR numbers were encrypted). Danish legislation allows sharing of pseudonymized data only if relevant legal permissions from e.g. the Danish Health Data Authority, and the Danish Data Protection Agency are obtained by the party who wants the data to be made available, along with separate permission to be obtained by the authors to hand over the data to a third party.

Results

This cohort included data from 7246 men and couples participating in ART treatment with a transferred blastocyst resulting in a clinical pregnancy. In total, 4842 competent blastocysts developed into a live birth and were included in the analyses (S1 Fig). A total of 2044 men participated in a COS treatment, and 2798 participated in FET. Mean age for men in COS/FET were 34.7/34.5 years (Table 1). Mean BMI for COS/FET were 26.2/26.0. Most men were non-smokers: 82.1%/85.0%. Regarding the blastocyst characteristics for COS/FET 59.3%/50.4% were transferred on development stage 4, 66.4%/56.5% were TE grade A and 68.6%/59.2% were ICM grade A. Mean hCG for COS/FET were respectively 320IU/L and 498IU/L (Table 1). 35.8% of the men were between 30–34 year, and 27.2% were between 35–39 year (S1 Table). Further stratification on paternal age can be seen in S1 Table. The squared correlation coefficient between women’s- and men’s age was R2 = 0.35 (S2 Fig).

Table 1

Men, treatment, blastocyst timing, morphology and implantation characteristics.

Variable	Men–COS1	Men–FET2
Age* (year), mean (sd)	34.7 (5.9)	34.5 (6.1)
Age* (year), range	21–68	19–68
Age* (year–category), n (%)
total	2044 (42.2)	2798 (57.8)
21–24	28 (1.4)	52 (1.8)
25–29	329 (16.1)	542 (18.7)
30–34	746 (36.5)	987 (35.3)
35–39	594 (29.1)	721 (25.8)
40–45	252 (12.3)	383 (13.7)
46–70	95 (4.6)	131 (4.7)
BMI (kg/m 2 ), mean (sd)	26.2 (3.9)	26.0 (3.8)
missing	851	1125
Smoking, cigarettes/day, n (%)
0	1260 (82.1)	1808 (85)
1–5	92 (6.0)	100 (4.7)
6–10	83 (5.4)	98 (4.6)
11–20	92 (6.0)	110 (5.2)
>20	8 (0.5)	11 (0.5)
missing	509	671
Female partners parity, n (%)
1	1364 (72.9)	1776 (71.2)
2	440 (23.5)	605 (24.3)
≥3	68 (3.6)	112 (4.5)
missing	172	305
Indication for cause of infertility, n (%)
anovulation	237 (11.8)	321 (11.7)
tube factor	154 (7.7)	251 (9.1)
endometriosis	92 (4.6)	132 (4.8)
male factor	716 (35.5)	984 (35.7)
Unexplained3	686 (34.1)	821 (29.8)
Other4	127 (6.3)	244 (8.9)
missing	32	45
Number of previous transfers, n (%)
0	1354 (66.2)	526 (18.8)
1	295 (14.5)	1035 (37.0)
2	171 (8.4)	522 (18.7)
3	109 (5.3)	289 (10.3)
≥4	115 (5.6)	426 (15.2)
Fertilization method, n (%)
IVF5	998 (48.8)	1208 (43.2)
ICSI6	1046 (51.2)	1590 (56.8)
Timing, n (%)
day 5 blastocyst7	2001 (97.9)	2181 (77.9)
day 6 blastocyst8	43 (2.1)	617 (22.1)
Stage, n (%)
19	33 (1.6)	20 (0.7)
29	32 (1.5)	12 (0.4)
3	289 (14.1)	465 (16.6)
4	1212 (59.3)	1409 (50.4)
5	461 (22.7)	738 (26.4)
6	17 (0.8)	154 (5.5)
Trophectoderm (TE), n (%)
A	1301 (66.4)	1489 (56.5)
B	593 (30.3)	1069 (40.5)
C	65 (3.3)	79 (3.0)
missing	85	161
Inner cell mass (ICM), n (%)
A	1343 (68.6)	1561 (59.2)
B	561 (28.6)	1013 (38.4)
C	55 (2.8)	63 (2.4)
missing	85	161
hCG 10 , mean (sd)	341 (194)	457 (286)
missing	320	498

*Paternal age at oocyte pick up

1 COS: Controlled Ovarian Stimulation

2 FET: Frozen-thawed Embryo Transfer

3Unexplained: Couples with unexplained infertility, 4Other: Female infertility caused by conditions in ovary, uterus, cervix or caused by other conditions (hepatitis, habitual abortion, asymptomatic HIV), 5IVF: In Vitro Fertilization, 6ICSI: Intracytoplasmic Sperm Injection

7Day 5 blastocyst: For COS defined as blastocyst transfer at day 5 and for FET defined as cryopreservation at day 5

8Day 6 blastocyst: For COS defined as blastocyst transfer at day 6 and For FET defined as cryopreservation at day 6

9Stage 1 and 2 (Group 3) are included in the group analysis only

10First measurement of serum human chorionic gonadotrophin (hCG).

Association between paternal age and stage and morphology of the competent blastocyst

In S2 and S3 Tables, men’s unadjusted mean age for COS and FET according to stage and morphology is shown. In FET, the age differed significantly for TE and Score Group (S3 Table), but the absolute differences were small. No statistically significant or clinically relevant differences were seen for COS (S2 Table).

The adjusted association between age and transfer day in COS treatments showed that for every increase of one year, a man had a 6% increased probability that the competent blastocyst was transferred at day 6 compared to day 5 (OR 1.06, 95%CI (1.00;1.13)) (Table 2). For FET, while the crude analysis of TE, ICM and Score Group showed significant associations with paternal age, no statistically significant associations were observed after adjustment (Table 2). A sensitivity analyses for COS treatments further adjusting for total FSH dose, fertilization method and sex of the child resulted in a stronger association between paternal age and transfer day (OR 1.11, 95% CI 1.02;1.20) and otherwise comparable results (S4 Table). Further, for FET treatments, when adding type of FET treatment to the adjusted analyses the results was comparable (S5 Table).

Table 2

The association of paternal age* with day of transfer, developmental stage and morphology of the competent blastocyst after COS and FET.

COS	N	OR	OR-adjusted**
Age*	2044
Transfer day	2044
5	2001	ref.	ref.
6	43	1.05 (1.01;1.10)	1.06 (1.00;1.13)
Stage (3–6)	1979	0.99 (0.98;1.01)	1.01 (0.99;1.03)
missing	65
TE3 (A-C)	1959	1.00 (0.99;1.02)	1.00 (0.98;1.02)
missing	85
ICM4 (A-C)	1959	1.00 (0.99;1.02)	0.99 (0.98;1.02)
missing	85
Group (1–3)	1959	0.99 (0.98;1.01)	0.99 (0.98;1.02)
missing	85
FET
Age *	2798
Cryopreservation day	2798
5	2181	ref.	ref.
6	617	1.02 (1.00;1.03)	1.01 (0.98;1.03)
Stage (3–6)	2766	1.00 (0.99;1.01)	0.99 (0.98;1.01)
missing	32
TE3 (A-C)	2637	1.02 (1.01;1.03)	0.99 (0.98;1.01)
missing	161
ICM4 (A-C)	2637	1.01 (1.00;1.03)	0.99 (0.98;1.01)
missing	161
Group (1–3)	2637	1.01 (1.00;1.02)	0.99 (0.98;1.01)
missing	161

Logistic regression. Multivariable logistic regression. Ordinal logistic regression. Ordinal multivariable logistic regression.

*Paternal age at oocyte pick up

**Adjusted for female age, female BMI, female smoking, diagnosis and clinic

1COS: Controlled Ovarian Stimulation

2FET: Frozen-thawed Embryo Transfer

3TE: Trophectoderm

4ICM: Inner Cell Mass.

Association between paternal age and the hCG level at implantation of the competent blastocyst

Women with men of advanced age (≥46) in both COS and FET had the highest unadjusted mean hCG values 370.3/510.8 (Table 3). However, the adjusted mean difference of hCG values when comparing age group 30–34, 35–39 and 40–45 with the age group 25–29 in those receiving COS treatment, all showed significantly lower adjusted values for older men (-51.7, 95% CI (-80.5;-23.0)), (-38.8, 95% CI (-72.6;-5.0)), (-55.2, 95% CI (-96.6; -13.8)) (Table 3). In those receiving FET, the crude mean difference was significantly higher only for men aged ≥46 compared to 25–29 (61.7, 95% CI (1.1;122.2)), but the association was not significant in the adjusted analysis (Table 3).

Table 3

The association of paternal age* with initial hCG rise at implantation, of the competent blastocyst after COS2 and FET.

COSPaternal age*	N	Missing	Mean hCG2 (sd)	Meandiff. (95%CI)	P-value	Adj. meandiff.(95%CI)**	P-adj
18–24	28	0	344.9 (156.7)	-22.3 (-97.5;52.9)	0.56	-9.2 (-86.5;68.1)	0.82
25–29	289	40	367.2 (197.6)	ref.		ref.
30–34	631	115	328.9 (184.1)	-38.3 (-65.3;-11.3)	0.01	-51.7 (-80.5;-23.0)	<0.001
35–39	492	102	341.2 (206.9)	-26.0 (-54.1;2.2)	0.07	-38.8 (-72.6;-5.0)	0.02
40–45	204	48	331.5 (172.4)	-35.7 (-70.5;-1.0)	0.04	-55.2 (-96.6;-13.8)	0.01
46–99	80	15	370.3 (228.5)	3.1 (-44.9;51.1)	0.90	-17.3 (-70.6;36.0)	0.52
Total	1724	320
P overall					0.07		0.01
FET Paternal age *
18–24	46	6	420.7 (240.7)	-28.4 (-15.0;58.2)	0.52	-18.5 (-111.4;74.5)	0.70
25–29	461	63	449.2 (277.1)	ref.		ref.
30–34	839	148	465.2 (289.9)	16.1 (-16.4;48.6)	0.33	0.04 (-35.6;35.7)	0.99
35–39	555	166	441.2 (278.7)	-8.0 (-43.2;27.3)	0.66	-11.4 (-54.5;31.6)	0.60
40–45	294	89	465.7 (287.9)	16.6 (-25.2;58.4)	0.44	-9.0 (-60.7;42.6)	0.73
46–99	105	26	510.8 (331.5)	61.7 (1.1;122.2)	0.04	24.9 (-46.6;96.4)	0.49
Total	2300	498
P overall					0.19		0.90

Linear regression. Multivariable linear regression.

*Paternal age at oocyte pick up

**Adjusted for female age, female BMI, female smoking, diagnosis and clinic

1human chorionic gonadotrophin

2COS: Controlled Ovarian Stimulation

3FET: Frozen-thawed Embryo Transfer.

For FET, 526 women had previously not had a transfer with fresh blastocyst (all blastocysts were frozen in the COS treatment) (Table 2) and hence the first transfer was a FET. We did a re-analysis without these treatments, which did not change the conclusions (S6 Table). In relation to Tables 2 and 3 we performed a sensitivity analysis with only the day 5 blastocysts, which did not change the conclusions (S7–S10 Tables).

Adjusting analyses for male BMI and male smoking instead of female BMI and female smoking in the adjusted analysis in Tables 2 and 3 did not change the conclusions (S7–S10 Tables).

Further, the interaction between women’s- and men’s age in relation to stage, morphology and the hCG level at implantation was tested in the full dataset (COS+FET), and no interactions were found (S15 Table).

Discussion

In this historical multicenter cohort study, we aimed to explore whether paternal age was associated with timing (transfer day (COS), cryopreservation day (FET)), stage, morphology, and the hCG level at implantation of the competent blastocyst. Our main findings were that in COS treatments, men’s age was associated with transfer day of the competent blastocyst. Furthermore, reduced hCG-values were seen in the pregnant partner of men ≥30 years compared to 25–29 years in COS treatments. No significant associations were seen in FET treatments.

One main strength of this study is the high number of exposure and outcome data, which is due to the nationwide coverage of the databases. This resulted in narrow confidence intervals, which support that our estimates are close to the true value, and the level of random error is low. Looking at the risk of systematic errors it is worth mentioning that due to reorganization of the database DMBR, it was not possible to have complete follow-up. From DMDC we had data from 2014–2018, and from DMBR we had data from 2014-March 2019. Missing data on live births throughout 2019 increased the risk of selection bias [26,27]. The subjective assessment of the blastocyst morphology is important to consider, as it could have introduced some degree of misclassification, i.e. information bias [28]. Because of this risk of variation between clinics in assessment of the blastocyst and potential other procedures, we adjusted for clinic in the analyses. The cohort was designed to include SET of blastocysts resulting in a single live birth. This ensured correct data as only one blastocyst was assessed and transferred. Further, by excluding twin births following SET we did not have to consider the influence of multiple implantations on hCG levels. On the other hand, having a design exclusively including blastocyst leading to a pregnancy and live birth will not present results differentiating between viable and not viable blastocysts, which could be perceived as a limitation. Studies investigating similar association as this study are limited, and hence knowledge of possible confounders is limited. Therefore, it cannot be ruled out that some residual confounding remains [23,24]. Sensitivity analyses were performed for consideration of different covariates. Finally, the exclusion of donor gametes limits the external validity to couples using their own gametes, but as this group constitutes the vast majority of men and women undergoing ART treatment this should be considered a minor issue.

For COS treatment, we found that for every increase of one year, a man had a 6% increased probability that the competent blastocyst was transferred at day 6, compared to day 5. While a 6% increase is not much per se, the 6% increased risk accumulates over a life span and may eventually add up to a considerably increased risk. This could indicate that by increasing paternal age, the competent blastocyst develops slower and hence reaches the full blastocyst stage for transfer later. A trend towards a slower development has earlier been suggested in a study on 1,023 oocyte donor cycles reporting significantly fewer embryos reaching the blastocyst stage day 5 if the male partner was above 55 years [29]. Likewise, a recent study on 3,837 cycles investigating the interaction between maternal and paternal age on morphological parameters showed that blastocyst development was negatively influenced by increasing maternal and paternal age [30]. However, a subsequent meta-analysis representing 12,538 oocyte-donation cycles failed to show any effect of paternal age on embryo development in vitro [31]. Further, our result should be interpreted with caution. First, OR’s overestimate the true risk ratio, especially in situations where the outcome occurs frequently. Second, there is always a risk of type 1 error, and the risk increases with increasing number of statistical tests performed. However, we followed a pre-planned analysis plan. In this study, we focused selectively on the competent blastocyst reaching a live birth, and hence we do not know if paternal age had a general impact on assessment scores of the usable (transferred and cryopreserved) blastocysts in the participating clinics. If the observed influence of male age reflects changes in the time to reach transferable developmental stage in general, remains unanswered.

Paternal genetic and epigenetic factors affecting embryo development and competence have been suggested [32]. The sperm methylome associates with the age of the man [33] and with increasing male age, an increased degree of sperm DNA damage has been observed [34]. As the mature spermatozoa has no/minimal DNA damage response [35], there will be an increasing task for the DNA repair capacity of the oocyte with increasing age of the man. Several studies, including autologous cycles, report a negative effect of paternal age in ART outcome when the female partner age is above 40 years, and it has been suggested that this could at least partly be explained by decreased oocyte competence to repair DNA breaks with increasing maternal age [36,37]. A new meta-analysis addressing the effect of paternal age on pregnancy loss suggests a small independent effect of paternal age along with that of female age [9]. A possible synergism seems not to be present in the reported parameters in our study, as test for interaction between male age and female age was insignificant for all outcomes. We speculate, if the present observation on delayed blastocyst development with increasing paternal age resulting in transfer day 6, reflects an increase in cell cycle length due to time spend to repair fragmented DNA [38]. Even so, in contrast to transfer day reaching a statistically significant association with paternal age, no association with stage (also reflecting developmental speed) was present. This can potentially reflect that the delay is subtle.

In Denmark, ICSI is performed primarily in couples with severe male factor. As sperm quality has been suggested to influence blastocyst formation rate [39], we stratified by fertilization methods (IVF/ICSI) and no differences were observed between the two groups in blastocyst stage, morphology nor transfer day in relation to paternal age.

It is well known that a single hCG measurement is not an ideal marker of the quality of an implantation [40]. In the present study, the initial hCG-rise (one hCG measurement 11 days after transfer) reflects an excellent implantation as it sustained until the live birth and is used as a parameter to further characterize the competent blastocyst.

A slower blastocyst development is supported by the association we found between a lower initial hCG-rise with advancing paternal age following transfers in COS cycles, potentially reflecting a subtle postponement in implantation. Excluding the day 6 data from the analysis did not alter the lack of association between paternal age and stage and morphology, nor the association showing decrease in initial hCG with increasing paternal age, supporting an overall delay in implantation after COS by increasing paternal age. Interestingly, in our previous analyses of the association between female age and parameters related to the development speed of the competent blastocyst, we found that increasing female age was associated with significantly lower chance of the competent blastocyst being in the high developmental stage, while no significant association was present between female age and transfer day (day 5 or 6 after fertilization) [18]. With respect to initial hCG-rise after COS, we found systematically lower hCG-levels in young women aged 18–24 years [18], whereas the hCG-levels were significantly lower in women with older partners aged 30–45 years. While the association in women may reflect better embryo-endometrium synchrony in young women compared to older women [18], as suggested the association in men may well reflect the potentially later implantation.

In FET treatments, as in COS treatments, no associations were found between the stage and morphology of the competent blastocyst and paternal age, neither in the full data analysis nor in the analysis excluding blastocysts cryopreserved day 6. Same analyses for initial hCG-rise showed no associations to paternal age.

In the FET dataset, 526 were first-transfers and these blastocysts may represent another cohort of blastocysts than the blastocysts cryopreserved after transfer of the top blastocyst in the COS cycle. However, the sensitivity analysis without these 526 first transfers did not change the results. The difference in the association between initial hCG-rise and paternal age between COS and FET may reflect that the potential delay involves a blastocyst-endometrial asynchrony, which is less relevant in FET cycles.

In our previous analyses of the association between female age and parameters related to the competent blastocyst, we found similar results, showing no association between female age and cryopreservation day, stage, TE, ICM or group in FET cycles [18], suggesting these parameters to be completely independent of age of both partners. With respect to initial hCG-rise, we found systematically lower hCG-levels in young women [18] but no association with paternal age in FET cycles, which may reflect better embryo-endometrium synchrony in young women compared to older women [18], which would occur independently of paternal age.

The mean paternal age (at OPU) for COS and FET was 34.7/34.5 years, and the mean maternal age (at OPU) for COS and FET was respectively 32.0/32.2 years [18]. Even so, while men were systematically older than women in our cohort, assessing the correlation between male and female age we found that age observations were evenly distributed around the regression line when plotting male against female age, suggesting that the age difference between men and women was fairly independent of young or older age.

In conclusion, our cohort data on competent blastocysts each reaching the birth of a child, demonstrate that advanced paternal age may be associated with increased time to reach the transferable stage and reduced level of hCG in the pregnant partner following COS and single blastocyst transfer. In FET treatments, none of the investigated associations reached statistical significance. We hypothesize 1) that the later transfer (day 6) in female partners of older men may be due to longer time spent by the oocyte to repair fragmented DNA of the sperm cells or an age dependent blastocyst-endometrial asynchrony, and 2) that the reduced level of hCG could reflect a delay in implantation which should be a focus of future research in men.

Flow chart.

1Danish medical data center, 2Preimplantation genetic testing, 3Personal identification number, 4Ultrasound testing for pregnancy, 5Danish medical birth register.

(DOCX)

Click here for additional data file.

The correlation between women’s age and men’s age at oocyte pick up.

(DOCX)

Click here for additional data file.

Male age, treatment, blastocyst timing, morphology and implantation characteristics.

*Paternal age at oocyte pick up, 1Unexplained: Couples with unexplained infertility, 2Other: Female infertility caused by conditions in ovary, uterus, cervix or caused by other conditions (hepatitis, habitual abortion, asymptomatic HIV), 3IVF: In Vitro Fertilization, 4ICSI: Intracytoplasmic Sperm Injection, 5Day 5 blastocyst: For COS defined as blastocyst transfer at day 5 and for FET defined as cryopreservation at day 5, 6Day 6 blastocyst: For COS defined as blastocyst transfer at day 6 and For FET defined as cryopreservation at day 6, 7First measurement of serum human chorionic gonadotrophin (hCG).

(DOCX)

Click here for additional data file.

The unadjusted association of paternal age* with day of transfer, developmental stage and morphology of the competent blastocyst after COS1 t–test.

One-way ANOVA. *Paternal age at oocyte pick up, 1COS: Controlled Ovarian Stimulation, 2TE: Trophectoderm, 3ICM: Inner Cell Mass, 4Group 1: 6AA, 6BA, 5AA, 5BA, 4AA, 4BA, Group 2: 6AB, 6BB, 6CB, 6CA, 5AB, 5BB, 5CB, 5CA, 4AB, 4BB, 4CB, 4CA, Group 3: 6AC, 6BC, 6CC, 5AC, 5BC, 5CC, 4AC, 4BC, 4CC, 3AA, 3AB, 3BA, 3AC, 3CA, 3BB, 3BC, 3CB, 3CC, 2AA, 2AB, 2BA, 2AC, 2CA, 2BB, 2BC, 2CB, 2CC, 1AA, 1AB, 1BA, 1AC, 1CA, 1BB, 1BC, 1CB, 1CC.

(DOCX)

Click here for additional data file.

The unadjusted association of paternal age* with day of transfer, developmental stage and morphology of the competent blastocyst after FET1 t–test.

One-way ANOVA. *Paternal age at oocyte pick up, 1FET: Frozen-thawed Embryo Transfer, 2TE: Trophectoderm, 3ICM: Inner Cell Mass, 4Group 1: 6AA, 6BA, 5AA, 5BA, 4AA, 4BA, Group 2: 6AB, 6BB, 6CB, 6CA, 5AB, 5BB, 5CB, 5CA, 4AB, 4BB, 4CB, 4CA, Group 3: 6AC, 6BC, 6CC, 5AC, 5BC, 5CC, 4AC, 4BC, 4CC, 3AA, 3AB, 3BA, 3AC, 3CA, 3BB, 3BC, 3CB, 3CC, 2AA, 2AB, 2BA, 2AC, 2CA, 2BB, 2BC, 2CB, 2CC, 1AA, 1AB, 1BA, 1AC, 1CA, 1BB, 1BC, 1CB, 1CC.

(DOCX)

Click here for additional data file.

The association of paternal age* with timing, stage and morphology of the competent blastocyst after COS1.

Logistic regression. Multiple logistic regression. Ordinal logistic regression. Ordinal multiple logistic regression. *Paternal age at oocyte pick up, **Adjusted for female age, female BMI, female smoking, diagnosis, clinic, FSH dose, fertilization method and sex of the child, 1COS: Controlled Ovarian Stimulation, 2TE: Trophectoderm, 3ICM: Inner Cell Mass.

(DOCX)

Click here for additional data file.

The association of paternal age* with timing, stage and morphology of the competent blastocyst after FET1.

Logistic regression. Multiple logistic regression. Ordinal logistic regression. Ordinal multiple logistic regression. *Paternal age at oocyte pick up, **Adjusted for female age, female BMI, female smoking, diagnosis, clinic and type of FET treatment, 1FET: Frozen-thawed Embryo Transfer, 2TE: Trophectoderm, 3ICM: Inner Cell Mass.

(DOCX)

Click here for additional data file.

The association of men’s age* with implantation, initial hCG1 rise, of the competent blastocyst after FET2 –without 526 first transfers.

Linear regression. Multiple linear regression. *Men’s age at oocyte pick up, **Adjusted for female age, female BMI, female smoking, diagnosis and clinic, 1human chorionic gonadotrophin, 2FET: Frozen-thawed Embryo Transfer.

(DOCX)

Click here for additional data file.

The association of men’s age* with stage and morphology of the competent blastocyst after COS1—without 43 day 6 blastocysts.

Logistic regression. Multivariable logistic regression. Ordinal logistic regression. Ordinal multivariable logistic regression. *Men’s age at oocyte pick up, **Adjusted for female age, female BMI, female smoking, diagnosis and clinic, 1COS: Controlled Ovarian Stimulation, 2TE: Trophectoderm, 3ICM: Inner Cell Mass.

(DOCX)

Click here for additional data file.

The association of men’s age* with stage and morphology of the competent blastocyst after FET1 –without 617 day 6 blastocysts.

Logistic regression. Multivariable logistic regression. Ordinal logistic regression. Ordinal multivariable logistic regression. *Men’s age at oocyte pick up, **Adjusted for male age, female BMI, female smoking, diagnosis and clinic, 1FET: Frozen-thawed Embryo Transfer, 2TE: Trophectoderm, 3ICM: Inner Cell Mass.

(DOCX)

Click here for additional data file.

The association of men’s age* with implantation, initial hCG1 rise, of the competent blastocyst after COS2—without 43 day 6 blastocysts.

Linear regression. Multivariable linear regression. *Men’s age at oocyte pick up, **Adjusted for female age, female BMI, female smoking, diagnosis and clinic, 1human chorionic gonadotrophin, 2COS: Controlled Ovarian Stimulation.

(DOCX)

Click here for additional data file.

The association of men’s age* with implantation, initial hCG1 rise, of the competent blastocyst after FET2– without 617 day 6 blastocysts.

Linear regression. Multivariable linear regression. *Men’s age at oocyte pick up, **Adjusted for female age, female BMI, female smoking, diagnosis and clinic, 1human chorionic gonadotrophin, 2FET: Frozen-thawed Embryo Transfer.

(DOCX)

Click here for additional data file.

The association of men’s age* with stage and morphology of the competent blastocyst after COS1.

Logistic regression. Multivariable logistic regression. Ordinal logistic regression. Ordinal multivariable logistic regression. *Men’s age at oocyte pick up, **Adjusted for female age, male BMI, male smoking, diagnosis and clinic, 1COS: Controlled Ovarian Stimulation 2TE: Trophectoderm, 3ICM: Inner Cell Mass.

(DOCX)

Click here for additional data file.

The association of men’s age* with stage and morphology of the competent blastocyst after FET1.

Logistic regression. Multiple logistic regression. Ordinal logistic regression. Ordinal multiple logistic regression. *Men’s age at oocyte pick up, **Adjusted for female age, male BMI, male smoking, diagnosis and clinic, 1FET: Frozen-thawed Embryo Transfer, 2TE: Trophectoderm, 3ICM: Inner Cell Mass.

(DOCX)

Click here for additional data file.

The association of men’s age* with implantation, initial hCG1 rise, of the competent blastocyst after COS2.

Linear regression. Multiple linear regression. *Men’s age at oocyte pick up, **Adjusted for female age, male BMI, male smoking, diagnosis and clinic, 1human chorionic gonadotrophin, 2COS: Controlled Ovarian Stimulation.

(DOCX)

Click here for additional data file.

The association of men’s age* with implantation, initial hCG1 rise, of the competent blastocyst after FET2.

Linear regression. Multiple linear regression. *Men’s age at oocyte pick up, **Adjusted for female age, male BMI, male smoking, diagnosis and clinic, 1human chorionic gonadotrophin, 2FET: Frozen-thawed Embryo Transfer.

(DOCX)

Click here for additional data file.

Analysis of interaction between women age (age) and male age (age_vir).

(DOCX)

Click here for additional data file.

13 Jan 2022

18 May 2022

Review

PONE-D-21-32713

Is paternal age associated with transfer day, developmental stage, morphology, and initial hCG-rise of the competent blastocyst leading to live birth? a multicenter cohort study

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Reply: Thank you for the reminder. We have now ensured that our manuscript meets the requirements.

2. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

Reply: Instead of inclusion of the phrase “data not shown” we have referred to and uploaded the results as supporting tables. Please see the uploads, the references in the section of the Results and the section of Supporting information (in the main document after the references).

3. Thank you for submitting the above manuscript to PLOS ONE. During our internal evaluation of the manuscript, we found significant text overlap between your submission and the following previously published works, some of which you are an author.

https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/s12884-020-03348-2

https://www.fertstert.org/article/S0015-0282(20)32191-9/fulltext

We would like to make you aware that copying extracts from previous publications, especially outside the methods section, word-for-word is unacceptable. In addition, the reproduction of text from published reports has implications for the copyright that may apply to the publications.

Please revise the manuscript to rephrase the duplicated text, cite your sources, and provide details as to how the current manuscript advances on previous work. Please note that further consideration is dependent on the submission of a manuscript that addresses these concerns about the overlap in text with published work.

We will carefully review your manuscript upon resubmission, so please ensure that your revision is thorough

Reply: Thank you for pointing out that you have found some text overlap in two articles.

The second article that you link to was written by our research group. The article has female age as exposure, and this manuscript that we uploaded to PLOSONE has male age as exposure. The article on female age was made parallel to the male age article. The outcomes in the two articles are the same. By examining the comparability between the two articles electronically, we can see that in the section of the Methods there are some small overlaps, which much be expected, since the same data and the same methodology was applied. For example in the sections “Treatment regimen” / ”Clinical setting”, “Outcomes” , “Covariates” and less in the section “Statical methods”. However, beyond these methodological issues, there is no overlap in the sections: “Introduction, “Results” and “Discussion”. Furthermore, we have referred to the article of female age in line, 102-104, 121, 196, 358, 363, 364, 377, 379, 381 and 385.

The first article that you link to has a very different topic of interest (Prevalence and associated factors of birth asphyxia among live births at Debre Tabor General Hospital, North Central Ethiopia). As we do not include data from the children it is difficult to see where the overlap should be, also when we read the paper.

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Reply: We have reviewed all the references and checked that none of them have been retracted.

Additional Editor Comments (if provided):

The paper has been appreciated by the two Reviewers. However, there are some minor aspects that need to be addressed:

1. The Authors has underlined the importance of the results of the fresh transfer on day 6 compared to day 5. However, the number of transfers in day 6 was only 43. How can the authors be so sure not to have a type 1 error? The Authors may discuss this aspect. The manuscript should be also modified according to this possible bias.

Reply: Unfortunately, it is by definition not possible to be sure that a type 1 error is avoided, as there will always be a risk of that type of error. The low number of day 6 transfers is reflected in the confidence interval of the adjusted OR: 1.06, CI (1.00;1.13) where the precision of the estimate is showed. We have added a sentence in the Discussion. Please see line 324-325 in the clean version.

2. The strategy to transfer a fresh embryo in day 6 is not so diffuse given the idea that in day 6 the endometrium is less receptive. As a matter of fact, the number of fresh transfers in day 6 is low. The Authors stated that they have corrected for centers but results are not shown. Is it possible that few centers adopted this strategy and they had older patients?

Reply: Thank you for that comment. We have adjusted for center by inclusion of “Clinic” in all our multivariable analyses. Please see the legends in Table 2 and Table 3.

3. Was there some kind of standardization of blastocyst morphology among centers?

Reply: Thank you for the comment. To reduce the risk of interobserver and intraobserver variation, most of the participating clinics have a standardized training set-up planned annually, which is mandatory to participate in. However, some variation may exist between the IVF laboratories, and therefore the multivariable analyses were adjusted for clinic. We added a sentence about this in the section of Materials and Methods. Please see line 166-168 in the clean version.

4. Given the small ORs, the conclusions seem overemphasized.

Reply: We highlight the adjusted association between male age and transfer day OR 1.06, CI (1.00;1.13), which means that with a one year increase in male age there is a 6% increased probability that the competent blastocyst will be transferred on day 6 compared to day 5. An assessment of whether this estimate is overemphasized is very relevant. The relevance per one year increase in male age is relatively small, but one should bear in mind that looking at the estimate over several years it may sum up to fairly high risk. Further, our result should be interpreted with caution, as OR overestimates the true risk ratio, particularly in case of frequent outcomes. We have added a sentence with that modification in the Discussion. Please see line 313-315 and line 322-324 in the clean version. Further we have modified the conclusions. Please see line 63-64 and line 391-393 in the clean version.

5. The choice of the bHCG levels is debatable. Please add some references to support this choice.

Reply: We do not entirely understand this comment. We have not chosen any beta-hCG-levels, and we do not report any cut-off values for hCG or anything similar. We merely compare the observed (standardized) hCG-levels.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

15 Jun 2022

Is paternal age associated with transfer day, developmental stage, morphology, and initial hCG-rise of the competent blastocyst leading to live birth? a multicenter cohort study

PONE-D-21-32713R1

Dear Dr. Borgstrøm,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Paola Viganò

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The manuscript is acceptable for publication.

Reviewers' comments:

30 Jun 2022

PONE-D-21-32713R1

Is paternal age associated with transfer day, developmental stage, morphology, and initial hCG-rise of the competent blastocyst leading to live birth? a multicenter cohort study

Dear Dr. Borgstrøm:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Paola Viganò

Academic Editor

PLOS ONE